Unravelling the genetic basis for the rapid diversification of male genitalia betweenDrosophilaspecies

Abstract

Corresponding authors: msantos-nunes@brookes.ac.uk (MDSN) and amcgregor@brookes.ac.uk (APM). AbstractIn the last 240,000 years, males of the Drosophila simulans species clade have evolved striking differences in the morphology of their epandrial posterior lobes and claspers (surstyli). These changes have most likely been driven by sexual selection and mapping studies indicate a highly polygenic and generally additive genetic basis. However, we have limited understanding of the gene regulatory networks that control the development of genital structures and how they evolved to result in this rapid phenotypic diversification. Here, we used new D. simulans / D. mauritiana introgression lines on chromosome 3L to generate higher resolution maps of posterior lobe and clasper differences between these species. We then carried out RNA-seq on the developing genitalia of both species to identify the genes expressed during this process and those that are differentially expressed between the two species. This allowed us to test the function of expressed positional candidates during genital development in D. melanogaster. We identified several new genes involved in the development and possibly the evolution of these genital structures, including the transcription factors Hairy and Grunge. Furthermore, we discovered that during clasper development Hairy negatively regulates tartan, a gene known to contribute to divergence in clasper morphology. Taken together our results provide new insights into the regulation of genital development and how this evolves between species.

Sox21bunderlies the rapid diversification of a novel male genital structure betweenDrosophilaspecies

“…Only RNAi against CKIIalpha-i1, knrl, MED10 (all cercus area) and Sox21b (posterior lobe area) altered structures consistent with the direction of the difference in their expression and phenotype in D. mauritiana D1 compared to D. simulans w 501 (Table 1; Figure S1A and S1C). From our RNAi knockdown experiments in D. melanogaster, Sox21b was the only differentially expressed gene between D. mauritiana and D. simulans within a mapped introgressed region to significantly affect the genitalia morphology consistent with the direction of the gene's expression and phenotypic difference between these two species (Table 1) 27 . The posterior lobes were significantly larger when Sox21b expression was reduced by RNAi in D. melanogaster (with and without accounting for body size) evidencing a possible repressive function consistent with the expression and difference in size of this structure between D. mauritiana and D. simulans (Figure 1B, 1C and 1D; Table S1).…”

“…The posterior lobes evolved from co-option of a Hox regulated gene network from the posterior spiracles 10 and have since diversified in shape and size in the D. simulans clade in particular over the last 240,000 years driven by sexual selection [18][19][20][21] . The genetic basis of this diversification is highly polygenic but to the best of our knowledge none of the causative genes have yet been identified despite extensive mapping [22][23][24][25][26][27][28][29][30] . Identifying the genes underlying the diversification of these secondary sexual structures is essential to understanding the basis of changes in their morphology and the evolutionary impact on copulation and species recognition.…”

“…Our previous analysis of RNA-Seq data from the developing male genitalia of D. simulans and D. mauritiana revealed 49 differentially expressed TF encoding genes 27 , 24 on the chromosome 3L arm, where we have previously generated highresolution introgression maps of regions contributing to divergence in genital structures between these two species 26,27,37 . We tested the function of these 24 genes during genital development using RNAi knockdown in D. melanogaster.…”

“…To better understand how the male genitalia develops and have evolved between D. simulans and D. mauritiana (Figure 1A), we used a candidate gene approach to interrogate RNA-seq data and genomic regions identified by introgression mapping 27 . We focussed on genes encoding transcription factors (TFs) because in many previous studies they have been shown to occupy key nodes in gene regulatory networks and contribute to morphological evolution [31][32][33][34][35][36] .…”

“…Since Sox21b regulates posterior lobe development in D. melanogaster, is located in a genomic region 27 that contributes to differences in posterior lobe size between D. mauritiana and D. simulans, and is expressed differently between these two species, we then directly tested if this gene has contributed to the evolution of their posterior lobe size difference. To do this, we carried out a reciprocal hemizygosity test 37,[45][46][47] .…”