PLoS ONE (Jan 2014)

Insights into the development and evolution of exaggerated traits using de novo transcriptomes of two species of horned scarab beetles.

  • Ian A Warren,
  • J Cristobal Vera,
  • Annika Johns,
  • Robert Zinna,
  • James H Marden,
  • Douglas J Emlen,
  • Ian Dworkin,
  • Laura C Lavine

DOI
https://doi.org/10.1371/journal.pone.0088364
Journal volume & issue
Vol. 9, no. 2
p. e88364

Abstract

Read online

Scarab beetles exhibit an astonishing variety of rigid exo-skeletal outgrowths, known as "horns". These traits are often sexually dimorphic and vary dramatically across species in size, shape, location, and allometry with body size. In many species, the horn exhibits disproportionate growth resulting in an exaggerated allometric relationship with body size, as compared to other traits, such as wings, that grow proportionately with body size. Depending on the species, the smallest males either do not produce a horn at all, or they produce a disproportionately small horn for their body size. While the diversity of horn shapes and their behavioural ecology have been reasonably well studied, we know far less about the proximate mechanisms that regulate horn growth. Thus, using 454 pyrosequencing, we generated transcriptome profiles, during horn growth and development, in two different scarab beetle species: the Asian rhinoceros beetle, Trypoxylus dichotomus, and the dung beetle, Onthophagus nigriventris. We obtained over half a million reads for each species that were assembled into over 6,000 and 16,000 contigs respectively. We combined these data with previously published studies to look for signatures of molecular evolution. We found a small subset of genes with horn-biased expression showing evidence for recent positive selection, as is expected with sexual selection on horn size. We also found evidence of relaxed selection present in genes that demonstrated biased expression between horned and horn-less morphs, consistent with the theory of developmental decoupling of phenotypically plastic traits.