Evolution of new genes under intermittent selection

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

New genes can evolve by mutations that generate a new function in an existing gene 1 . However, these mutations often have a negative impact on the original function, leading to trade-offs that constrain their further evolution 2 . Genes that exhibit a strong trade-off between the original and the new function are expected to evolve through gene duplication, which can increase the expression of a weak new activity, buffer against negative effects on the original function, and provide more targets for beneficial mutations to arise 3 . The expected outcome of evolution in conditions where both functions are beneficial is a new pair of paralogs, each specialized for one function. Despite this, there are examples in nature where bi-functional generalist enzymes have evolved from a presumed specialist ancestor 4 . This study tests the hypothesis that generalist enzymes can evolve from specialist ancestral enzymes if selection for the new function is repeatedly interrupted by periods without selection ( i.e. selection for the new function is intermittent). In evolution experiments using bacteria lacking an enzyme in the tryptophan synthesis pathway, with intermittent selection for restoring tryptophan synthesis, multiple examples were found where initially specialized enzymes in two different pathways evolved towards becoming bi-functional generalist enzymes supporting both their new and original functions. Our results highlight the importance of considering selection in nature not as a constant, but as a force that may fluctuate, and that fluctuating selection can drastically change the outcome by forcing evolution along paths that are highly constrained by conflicting selection pressures. Especially for duplications, which are mechanistically unstable and often costly, intermittent selection is expected to have a huge impact: If selection for a beneficial limiting function fluctuates, duplications may be counter selected and lost at regular intervals, forcing alternative paths of evolution that do not require duplications.

Article activity feed