This guest post is by Julien Roux on Roux et al. “Patterns of positive selection in seven ant genomes“, arXived here.

The publication of the honeybee genome in 2006 can be considered the birth date of "sociogenomics", a research field whose agenda is to understand social life in molecular terms. Recently, this field has entered a period of rapid discovery with the publication of full genome sequences of multiple Hymenoptera species. In particular, the release of seven ant genome sequences gave us the opportunity to look for the molecular origins of some of the spectacular adaptations of the ant lineage, through patterns of positive selection on amino-acid substitutions in ant genes. We used rigorous methods to detect episodic positive selection while controlling for false positives inspired by the database Selectome. All data is publicly available for people to reuse.

An original aspect of our paper is that we analyzed not only ant genomes, but also data from 10 species of bees and 12 species of flies with the same methods to permit an unbiased comparison of positive selection patterns between lineages. For example, immune genes were enriched for positive selection signal in all three lineages. This may not look surprising since these are classical hits of positive selection scans, but it was previously hypothesized that the evolution of social hygienic behaviors in ants and bees may have relaxed the selective pressure on immune genes. Our analysis indicates that this effect is either absent or relatively small.

Other hypotheses have been put forward in relation to the evolution of sociality in Hymenoptera. Notably, it was proposed that the challenges of social life in the colonies should be reflected by increased positive selection signal on neurogenesis genes. Similarly, because communication is mostly based on chemical signals in colonies of social insects, it was suggested that increased positive selection should be observed on olfactory receptors compared to non-social insects. Our results question both these hypotheses, since we observed that increased positive selection on these classes of genes does not coincide with (but predated) the evolution of sociality in Hymenoptera.

Finally, the comparison between the three lineages allowed us to pinpoint some patterns that were most likely specific to the ant lineage. We found less positive selection on genes related to metabolism in ants compared to bees and flies. We think this could be the sign of relaxed selection on these genes, possibly in relation to the important reduction on metabolic needs with the loss of flight in ant workers. By contrast, we identified a robust pattern of directional selection specific to the ant lineage on genes functioning in the mitochondria. Several pieces of evidence suggest that this pattern might be linked to the remarkable lifespan extension that evolved in the ant lineage. Queens of some ant species can indeed live up to 100 times longer than solitary insects, (that is up to 30 years!). Positive selection possibly played a role in optimizing the activity of mitochondria, where the respiratory chain is the primary source of production of Reactive Oxidative Species (ROS), an important proximal cause for aging, thus contributing to the evolution of increased lifespan in ants.

In conclusion, protein level episodic positive selection appears to have played an important role in the evolution of social insects, notably regarding strong mitochondrial adaptation in ants.