Search

Males and females have different reproductive roles and are often subject to contrasting selection pressures. This sexual antagonism can lead, at a given locus, to different alleles being favoured in each sex and, consequently, to genetic variation being maintained in a population. Although the presence of sexually antagonistic (SA) polymorphisms has been documented across a range of species, their evolutionary dynamics remain poorly understood. Here, we study SA selection on gene expression, which is fundamental to sexual dimorphism, via the evolution of regulatory binding sites. We show that for sites longer than 1 nucleotide, expression polymorphism is maintained only when intermediate expression levels are deleterious to both sexes. We then show that, in a regulatory cascade, expression polymorphism tends to become displaced over evolutionary time from the target of SA selection to upstream regulators. Our results have consequences for understanding the evolution of sexual dimorphism, and provide specific empirical predictions for the regulatory architecture of genes under SA selection.

Transcription factor binding sites vary in their specificity, both within and between species. Binding specificity has a strong impact on the evolution of gene expression, because it determines how easily regulatory interactions are gained and lost. Nevertheless, we have a relatively poor understanding of what evolutionary forces determine the specificity of binding sites. Here we address this question by studying regulatory modules composed of multiple binding sites. Using a population-genetic model, we show that more complex regulatory modules, composed of a greater number of binding sites, must employ binding sites that are individually less specific, compared to less complex regulatory modules. This effect is extremely general, and it hold regardless of the regulatory logic of a module. We attribute this phenomenon to the inability of stabilising selection to maintain highly specific sites in large regulatory modules. Our analysis helps to explain broad empirical trends in the yeast regulatory network: those genes with a greater number of transcriptional regulators feature by less specific binding sites, and there is less variance in their specificity, compared to genes with fewer regulators. Likewise, our results also help to explain the well-known trend towards lower specificity in the transcription factor binding sites of higher eukaryotes, which perform complex regulatory tasks, compared to prokaryotes.