Antagonistic pleiotropy and the compensatory landscapes of distinct adaptive trajectories

Adaptation rarely produces perfect solutions. Instead, it proceeds through a series of mutational steps that often produce pleiotropic side effects. Hence, a crucial component of adaptation is compensating for the antagonistic effects of preceding mutations. To investigate how different degrees of antagonistic pleiotropy shape the dynamics of adapting populations, we performed temporal sequencing of eight populations of Escherichia coli at 11 time points throughout 2,000 generations of adaptation to high temperature (42.2°C). The eight populations represent two different pathways to high-temperature adaptation typified by mutations in rpoB, which encodes a subunit of the RNA polymerase complex, and rho, which encodes a major transcriptional terminator. Previous work has shown that rpoB and rho differ in their pleiotropic effects, with single mutations in rpoB causing more gene expression shifts than single mutations in rho. We therefore predicted that rpoB populations should accumulate more mutations than rho populations due to a greater capacity for compensatory evolution. We also predict that the greater number of mutations increase the amount of clonal interference in rpoB populations relative to rho populations. By reconstructing and quantifying the mutational trajectories of our populations, we found support for both predictions: rpoB populations indeed accumulate more mutations and experience greater clonal interference than their rho counterparts.