Universal genomic constraints in the evolvability of thermal physiology

Abstract

Thermal physiological traits such as body temperature often show surprisingly slow evolutionary rates over macroevolutionary time, despite apparent lability at microevolutionary time scales. While long-term stabilizing selection may slow rates of thermal evolution, we propose an alternative hypothesis from a bottom-up, population genomic perspective: the nature of body temperature (Tb) as an organism-level trait that must accommodate diverse protein thermal performances leads to evolutionary constraints at the organismal level. We support this hypothesis using a simulation framework in which we modeled and compared the rates of evolution for Tb alongside one or more proteins. Protein performances and organismal Tb were modeled as evolving, QTL-encoded traits, and organismal fitness was determined based on Tb given the performance curves of each protein. As predicted, a greater number of proteins led to drastic decrease in the rate of Tb evolution. When a shift in environmental temperature was simulated, Tb evolved with an initial rapid pulse toward the new optimum, followed by a phase of gradual evolution as the cumulative fitness costs of mismatching Tb and protein optima constrained thermal adaptation. That is, lability and stasis are predictable features of body temperature evolution: rapid, yet bounded microevolutionary bursts followed by long phases of sluggish evolution are both expected outcomes of directional selection operating on hierarchically structured traits like Tb. We suggest that protein thermal coordination might contribute to intrinsic, universal macroevolutionary patterns of stasis in organismal physiology across endotherms and ectotherms.