Mechanisms and dynamics of orphan gene emergence in insect genomes

Abstract

Orphan genes are defined as genes that lack detectable similarity to genes in other species and therefore no clear signals of common descent (i.e., homology) can be inferred. Orphans are an enigmatic portion of the genome because their origin and function are mostly unknown and they typically make up 10% to 30% of all genes in a genome. Several case studies demonstrated that orphans can contribute to lineage-specific adaptation. Here, we study orphan genes by comparing 30 arthropod genomes, focusing in particular on seven recently sequenced ant genomes. This setup allows analyzing a major metazoan taxon and a comparison between social Hymenoptera (ants and bees) and nonsocial Diptera (flies and mosquitoes). First, we find that recently split lineages undergo accelerated genomic reorganization, including the rapid gain of many orphan genes. Second, between the two insect orders Hymenoptera and Diptera, orphan genes are more abundant and emerge more rapidly in Hymenoptera, in particular, in leaf-cutter ants. With respect to intragenomic localization, we find that ant orphan genes show little clustering, which suggests that orphan genes in ants are scattered uniformly over the genome and between nonorphan genes. Finally, our results indicate that the genetic mechanisms creating orphan genes-such as gene duplication, frame-shift fixation, creation of overlapping genes, horizontal gene transfer, and exaptation of transposable elements-act at different rates in insects, primates, and plants. In Formicidae, the majority of orphan genes has their origin in intergenic regions, pointing to a high rate of de novo gene formation or generalized gene loss, and support a recently proposed dynamic model of frequent gene birth and death.

Fig. 1.—

Abundance of SSOGs and their dependence on the distance to the MRCA (table 1). SSOGs per species were plotted against the distance to the MRCA node in the phylogenetic tree. A linear regression (solid black line) was constructed to fit the observed SSOG counts from 12 Drosophilidae, 3 Culicidae, Bombyx, Tribolium, and Nasonia (circles, R² = 0.85). The confidence interval and the prediction interval of the linear model are shown in dark and light gray, respectively. The ant SSOG data points were added after fitting the linear model and are shown as triangles.

Fig. 2.—

Contrasting the abundance and rate of emergence of orphan genes between partially overlapping taxonomic groups of Hymenoptera and Diptera. Each tested group is highlighted by a rectangle and the associated group data, including number of species, distance to the MRCA, total orphan gene count, and rate of orphan gene emergence, are shown on the right side. Branch lengths in the phylogenetic tree are approximate values and were obtained from the timetree.org database (Hedges et al. 2006).