The role of mutational spectrum in the selection against mutator alleles

Abstract

Rapidly adapting microbe and cancer cell populations often evolve high mutation rates. Yet, once adaptive opportunity declines, antimutator alleles are expected to take over as a result of indirect selection against deleterious mutations. Theory indicates that the most important determinant of antimutator invasions is the extent of mutation rate reduction. However, inconsistent results from evolution experiments suggest that additional factors may also play a major role in antimutator dynamics. Here we show that the idiosyncratic mutation bias exhibited by different mutators - a previously unrecognized factor - can greatly alter mutator susceptibility to antimutator invasions. Using a simulation model calibrated to mimic a well-known long-term evolution experiment with bacteria, we show that differences in average deleterious load can account for order-of-magnitude changes in antimutator fitness for a realistic range of parameters. Since these parameters are known to vary with the environment, our results reveal an unanticipated source of variability in antimutator dynamics. Finally, we estimated the genome-wide average disruptive effect on proteins of mutations caused by different mutators, and found marked and systematic differences emerging across mutators and species with different genomic GC compositions. Taken together, our results suggest that antimutator dynamics may be highly dependent on the specific genetic, ecological and evolutionary history of a given population. Such dependence reveals a more complex picture than anticipated, being relevant for understanding mutators in clinical settings, as well as how hypermutability shapes the evolution of bacterial genome size and composition.