Extensive exchange of transposable elements in the Drosophila pseudoobscura group

Abstract

Background: As species diverge, so does their transposable element (TE) content. Within a genome, TE families may eventually become dormant due to host-silencing mechanisms, natural selection and the accumulation of inactive copies. The transmission of active copies from a TE families, both vertically and horizontally between species, can allow TEs to escape inactivation if it occurs often enough, as it may allow TEs to temporarily escape silencing in a new host. Thus, the contribution of horizontal exchange to TE persistence has been of increasing interest.

Results: Here, we annotated TEs in five species with sequenced genomes from the D. pseudoobscura species group, and curated a set of TE families found in these species. We found that, compared to host genes, many TE families showed lower neutral divergence between species, consistent with recent transmission of TEs between species. Despite these transfers, there are differences in the TE content between species in the group.

Conclusions: The TE content is highly dynamic in the D. pseudoobscura species group, frequently transferring between species, keeping TEs active. This result highlights how frequently transposable elements are transmitted between sympatric species and, despite these transfers, how rapidly species TE content can diverge.

Fig. 1

a Phylogeny of the D. pseudoobscura group species studies here, with labels showing the estimated time of divergence (from [45, 62]). b A Venn diagram showing the number of TE families, including putative novel families, shared between the D. pseudoobscura group species in a

Fig. 2

a Comparison of TE content of D. persimilis and D. pseudoobscura for each TE family. There is a significant correlation between copy number, but most TE families show more coverage in D. persimilis (GLM t-value = 23.532, p-value < 2e-16). b Density of insertions per MB for D. persimilis and D. pseudoobscura by TE order

Fig. 3

a Pairwise comparison of silent site diversity (d_S) for nuclear genes (solid line) and shared TEs (dashed lines) between D. pseudoobscura, D. pseudoobscura bogotana, D. persimilis and other species. The lower 2.5% quartile for nuclear d_S is shown as the dotted vertical line. These distributions are consistent between all species pair comparisons (t-test p-value > 0.13), so only comparisons to D. pseudoobscura are shown. b The number of transfer events for transposable elements based on d_S and confirmed with VHICA. The number in brackets shows events that can be seen in the assembled phylogenies. Note that many events could be occurring between species vertically as well as horizontally. c effective number of codons (ENC) for genes (in grey) and TEs (black) versus d_S between species pairs. Each shape represents a species pair. The dotted line represents the lower 2.5th percentile per 5 EHC window for D.pse/D.per and D.pse/D.mir (due to high similarity). These distributions are consistent between all species pair comparisons shown in Fig. 2a. (t-test p-value > 0.05), so only comparisons to D. pseudoobscura are shown. Again, only D. affinis shows no evidence of exchange between species. d Comparison of Tajimas D across species for frequently exchanged TEs and rarely exchanged TEs shows no difference, suggesting no population expansion. e Proportion of shared nucleotide polymorphism sites between TE sequences in species, out of total nucleotide polymorphism sites, divided by TE families with low d_S relative to nuclear genes and TEs with higher d_S