Conservation, Duplication, and Divergence of Five Opsin Genes in Insect Evolution

Abstract

Opsin proteins covalently bind to small molecular chromophores and each protein-chromophore complex is sensitive to particular wavelengths of light. Multiple opsins with different wavelength absorbance peaks are required for color vision. Comparing opsin responses is challenging at low light levels, explaining why color vision is often lost in nocturnal species. Here, we investigated opsin evolution in 27 phylogenetically diverse insect species including several transitions between photic niches (nocturnal, diurnal, and crepuscular). We find widespread conservation of five distinct opsin genes, more than commonly considered. These comprise one c-opsin plus four r-opsins (long wavelength sensitive or LWS, blue sensitive, ultra violet [UV] sensitive and the often overlooked Rh7 gene). Several recent opsin gene duplications are also detected. The diversity of opsin genes is consistent with color vision in diurnal, crepuscular, and nocturnal insects. Tests for positive selection in relation to photic niche reveal evidence for adaptive evolution in UV-sensitive opsins in day-flying insects in general, and in LWS opsins of day-flying Lepidoptera specifically.

Keywords: adaptive evolution; arctiidae; butterfly; lepidoptera; molecular evolution.

Fig. 1.—

Phylogenetic tree of 166 opsins obtained using Phylobayes and a GTR-Γ model. On each node, the three support values shown are (left to right): PP of opsin + outgroup data set, PP of opsin unrooted data set, ML bootstrap of the opsin + outgroup data set. All phylogenetic trees were performed under GTR-Γ. The lack of support for the UV + Blue opsin clade in the ML tree is caused by phylogenetic instability of a single opsin from the human ectoparasitic louse (compare supplementary figs. S1–S3, Supplementary Material online). The orphan sequence not assigned to a named clade derives from pea aphid (Acyrthosiphon pisum). The same tree with species names is given in supplementary figure S1, Supplementary Material online.

Fig. 2.—

Opsin gene repertoire in genome sequences compared with a cladogram of insect evolution accordingly to Misof et al. (2014). When the number of genes identified is greater than one, this is indicated inside the rectangles. Additional genes are only accepted if the assembly predicts six or seven transmembrane domains; assignment deduced from phylogenetic analysis. White boxes indicate a gene is not found; this may reflect gene loss or genome incompleteness.

Fig. 3.—

Pattern of positive selection inferred for the various opsin paralogues. (a) Comparison of ω for each paralogue across all insects, in relationship to lifestyle. (b) Comparison of ω for each paralogue across Lepidoptera, excluding intron-less Cal. dominula LWS sequences, assuming the common ancestor of the clade was nocturnal.