Taxonomically restricted genes are associated with the evolution of sociality in the honey bee

Abstract

Background: Studies have shown that taxonomically restricted genes are significant in number and important for the evolution of lineage specific traits. Social insects have gained many novel morphological and behavioral traits relative to their solitary ancestors. The task repertoire of an advanced social insect, for example, can be 40-50 tasks, about twice that of a solitary wasp or bee. The genetic basis of this expansion in behavioral repertoire is still poorly understood, and a role for taxonomically restricted genes has not been explored at the whole genome level.

Results: Here we present comparative genomics results suggesting that taxonomically restricted genes may have played an important role in generating the expansion of behavioral repertoire associated with the evolution of eusociality. First, we show that the current honey bee official gene set contains about 700 taxonomically restricted genes. These are split between orphans, genes found only in the Hymenoptera, and genes found only in insects. Few of the orphans or genes restricted to the Hymenoptera have been the focus of experimental work, but several of those that have are associated with novel eusocial traits or traits thought to have changed radically as a consequence of eusociality. Second, we predicted that if taxonomically restricted genes are important for generating novel eusocial traits, then they should be expressed with greater frequency in workers relative to the queen, as the workers exhibit most of the novel behavior of the honey bee relative to their solitary ancestors. We found support for this prediction. Twice as many taxonomically restricted genes were found amongst the genes with higher expression in workers compared to those with higher expression in queens. Finally, we compiled an extensive list of candidate taxonomically restricted genes involved in eusocial evolution by analyzing several caste specific gene expression data sets.

Conclusions: This work identifies a large number of candidate taxonomically restricted genes that may have played a role in eusocial evolution. This work thus lays the foundation for future functional genomics work on the evolution of novelty in the context of social behavior. We also present preliminary evidence, based on biased patterns of gene expression, that taxonomically restricted genes may have played a role in the evolution of caste systems, a characteristic lineage specific social trait.

Figure 1

Number of genes in honey bee OGS2 without homologs in other genomes. Although all 12 Drosophila genomes were used, only Drosophila melanogaster is shown because all Drosophila species had very similar numbers of homologs with Apis mellifera. Number of genes without homologs roughly followed taxonomic relatedness with the closely related ants having homologs for the largest number of Apis mellifera genes.

Figure 2

Percentages of Apis mellifera genes in taxonomically and non-taxonomically restricted categories.

Figure 3

Percentages of Apis mellifera genes in taxonomically restricted categories with evidence for origin via gene duplication. A small percentage of genes in the most taxonomically restricted categories have paralogs that show homology to genes outside their taxonomic class. Hence, there is little support for most of these genes having formed via gene duplication followed by rapid sequence divergence. A much higher percentage of insect specific genes could, however, have evolved in this fashion, as a high percentage of such genes have paralogs that are widely conserved.