Abstract
The origin, diversification, and secondary loss of sexually dimorphic characters are common in animal evolution. In some cases, structurally and functionally similar traits have evolved independently in multiple lineages. Prominent examples of such traits include the male-specific grasping structures that develop on the front legs of many dipteran insects. In this report, we describe the evolution and development of one of these structures, the male-specific “sex brush”. The sex brush is composed of densely packed, irregularly arranged modified bristles and is found in several distantly related lineages in the family Drosophilidae. Phylogenetic analysis using 250 genes from over 200 species provides modest support for a single origin of the sex brush followed by many secondary losses; however, independent origins of the sex brush cannot be ruled out completely. We show that sex brushes develop in very similar ways in all brush-bearing lineages. The dense packing of brush hairs is explained by the specification of bristle precursor cells at a near-maximum density permitted by the lateral inhibition mechanism, as well as by the reduced size of the surrounding epithelial cells. In contrast to the female and the ancestral male condition, where bristles are arranged in stereotypical, precisely spaced rows, cell migration does not contribute appreciably to the formation of the sex brush. The complex phylogenetic history of the sex brush can make it a valuable model for investigating coevolution of sex-specific morphology and mating behavior.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
1) The current version of the manuscript contains additional phylogenetic analyses for which we assembled considerably larger dataset (250 genes from 207 Drosophila species). The results of these analyses are less definitive about the evolutionary origins of our focal trait than the smaller phylogenetic study in the previous version of the manuscript. Whereas the earlier analysis supported the convergent origin, the new analyses favor a single origin followed by multiple losses of that trait; however, support for this scenario is not very strong, and convergent evolution cannot be ruled out with any confidence either. We have revised the results and discussion sections of the manuscript accordingly. 2) The title and the abstract have been changed to reflect new conclusions. 3) The manuscript has three additional authors. 4) The result section contains the phylogenetic analyses with the larger dataset as well as the ancestral state reconstruction. 5) We also conducted more thorough quantitative analyses of cell dynamics during development. 6) The previous version of the manuscript should not be cited since its conclusions are superseded by the new analyses in the current version.