Differential gene expression and coding sequence evolution play complementary roles in the adaptive diversification of cichlid sensory systems.

Phenotypic evolution may occur either through alterations to the structure of protein-coding genes or their expression. Evidence for which of these two mechanisms more commonly contribute to the evolution of a phenotype can be garnered from examples of parallel and convergent evolution. The visual system of East African cichlid fishes is an excellent system with which to address this question. Cichlid fishes from Lakes Malawi (LM) and Victoria together exhibit three diverse palettes of coexpressed opsins and several important protein-coding mutations that both shift spectral sensitivity. Here we assess both opsin expression and protein-coding diversity among cichlids from a third rift lake, Lake Tanganyika (LT). We found that Tanganyikan cichlids exhibit three palettes of coexpressed opsins that largely overlap the short-, middle-, and long-wavelength-sensitive palettes of LM cichlids. Bayesian phenotypic clustering and ancestral state reconstructions both support the parallel evolution of the short- and middle-wavelength palettes among cichlids from LT and LM. In each case, these transitions occurred from different ancestors that expressed the same long-wavelength palette. We also identified similar but distinct patterns of correlated evolution between opsin expression, diet, and lens transmittance among cichlids from LT and LM as well. In contrast to regulatory changes, we identified few functional or potentially functional mutations in the protein-coding sequences of three variable opsins, with the possible exception of the SWS1 (ultraviolet) opsin. These results underscore the important contribution that gene regulation can make to rapid phenotypic evolution and adaptation.

Sensory systems play crucial roles in survival and reproduction. Therefore, sensory plasticity has important evolutionary implications. In this study, we examined retinal plasticity in five species of cichlid fish from Lake Malawi. We compared the cone opsin expression profiles of wild-caught fish to lab-reared F(1) that had been raised in a UV minus, reduced intensity light environment. All of the opsin genes that were expressed in wild-caught fish were also expressed in lab-reared individuals. However, we found statistically significant differences in relative opsin expression among all five species. The most consistent difference was in the SWS2B (violet) opsin, which was always expressed at higher levels in lab-reared individuals. Estimates of visual pigment quantum catch suggest that this change in expression would increase retinal sensitivity in the light environment of the lab. We also found that the magnitude of plasticity varied across species. These findings have important implications for understanding the genetic regulation of opsin expression and raise many interesting questions about how the cichlid visual system develops. They also suggest that sensory plasticity may have facilitated the ecological diversification of cichlids in Lake Malawi.

Sensory systems provide crucial information about an organism's external environment and, thus, are often subject to strong natural selection. Because of the large variation in the intensity and spectral quality of light in aquatic environments, studies of sensory adaptation have focused on the visual systems of fish for over a half a century. Recently, the molecular genetic mechanisms that determine the spectral sensitivity of visual pigments have been characterized in several fishes including zebrafish, guppies, medaka, killifish, bream, and cichlids. The results of these studies suggest that teleost fish have incredibly diverse visual systems. In this paper, we review the role that opsin duplication and differential gene expression have played in the diversification of visual pigments. We compare our findings in cichlids to five other taxonomic groups and highlight the ways that their similarities and differences may provide new insights into the molecular genetic basis of sensory adaptation and diversification.

The evolution of sexual dimorphism has long been attributed to sexual selection, specifically as it would drive repeated gains of elaborate male traits. In contrast to this pattern, New World oriole species all exhibit elaborate male plumage, and the repeated gains of sexual dichromatism observed in the genus are due to losses of female elaboration. Interestingly, most sexually dichromatic orioles belong to migratory or temperate-breeding clades. Using character scoring and ancestral state reconstructions from two recent studies in Icterus, we tested a hypothesis of correlated evolution between migration and sexual dichromatism. We employed two discrete phylogenetic comparative approaches: the concentrated changes test and Pagel's discrete likelihood test. Our results show that the evolution of these traits is significantly correlated (CCT: uncorrected P < 0.05; ML: LRT = 12.470, P < 0.005). Indeed, our best model of character evolution suggests that gains of sexual dichromatism are 23 times more likely to occur in migratory taxa. This study demonstrates that a life-history trait with no direct relationship with sexual selection has a strong influence on the evolution of sexual dichromatism. We recommend that researchers further investigate the role of selection on elaborate female traits in the evolution of sexual dimorphism. K E Y W O R D S :Bird coloration, comparative method, female ornaments, sexual dimorphism, sexual selection.