Neutral and adaptive explanations for an association between caste-biased gene expression and rate of sequence evolution

Abstract

The castes of social insects provide outstanding opportunities to address the causes and consequences of evolution of discrete phenotypes, i.e., polymorphisms. Here we focus on recently described patterns of a positive association between the degree of caste-specific gene expression and the rate of sequence evolution. We outline how neutral and adaptive evolution can cause genes that are morph-biased in their expression profiles to exhibit historical signatures of faster or slower sequence evolution compared to unbiased genes. We conclude that evaluation of different hypotheses will benefit from (i) reconstruction of the phylogenetic origin of biased expression and changes in rates of sequence evolution, and (ii) replicated data on gene expression variation within versus between morphs. Although the data are limited at present, we suggest that the observed phylogenetic and intra-population variation in gene expression lends support to the hypothesis that the association between caste-biased expression and rate of sequence evolution largely is a result of neutral processes.

Keywords: antagonistic pleiotropy; neutral evolution; phenotypic plasticity; polymorphism; social insects.

Figure 1

Predicted patterns of evolutionary history of gene expression and sequence evolution for the five scenarios outlined in the text. The left hand column shows the expected relative variation in expression within morphs and hypothetical changes in morph biased gene expression over time. White bars for morph 1 and gray bars for morph 2. The second column shows how average evolutionary rates are predicted to change over time (p, purifying; n, neutral; d, directional), and the third column shows how the patterns would be seen in a phylogeny in a group where some species show biased expression (B) for the gene in question, whereas others do not (U). Solid line, neutral rate; hatched line, purifying selection; dotted line, directional selection.

Figure 2

Why variability in expression pattern may be difficult to interpret. Boxes with solid and dotted outlines refer to expression in two different morphs. Top row: Within contexts gene expression that are consistently morph biased (dark gray boxes) or morph biased in a context-dependent manner (e.g., body size, age, past or current social experience) (white boxes) will show a morph biased pattern. In contrast, pooling samples across contexts makes context-dependent expression indistinguishable from expression variability per se (light gray boxes). Bottom row: In pooled samples genes regulated according to context (white boxes) are indistinguishable from genes with an expression pattern that is un-biased but highly variable (gray boxes).