Visual pigment evolution in Characiformes: The dynamic interplay of teleost whole‐genome duplication,surviving opsins and spectral tuning

Vision represents an excellent model for studying adaptation, given the genotype‐to‐phenotype map that has been characterized in a number of taxa. Fish possess a diverse range of visual sensitivities and adaptations to underwater light, making them an excellent group to study visual system evolution. In particular, some speciose but understudied lineages can provide a unique opportunity to better understand aspects of visual system evolution such as opsin gene duplication and neofunctionalization. In this study, we showcase the visual system evolution of neotropical Characiformes and the spectral tuning mechanisms they exhibit to modulate their visual sensitivities. Such mechanisms include gene duplications and losses, gene conversion, opsin amino acid sequence and expression variation, and A₁/A₂‐chromophore shifts. The Characiforms we studied utilize three cone opsin classes (SWS2, RH2, LWS) and a rod opsin (RH1). However, the characiform's entire opsin gene repertoire is a product of dynamic evolution by opsin gene loss (SWS1, RH2) and duplication (LWS, RH1). The LWS‐ and RH1‐duplicates originated from a teleost specific whole‐genome duplication as well as characiform‐specific duplication events. Both LWS‐opsins exhibit gene conversion and, through substitutions in key tuning sites, one of the LWS‐paralogues has acquired spectral sensitivity to green light. These sequence changes suggest reversion and parallel evolution of key tuning sites. Furthermore, characiforms' colour vision is based on the expression of both LWS‐paralogues and SWS2. Finally, we found interspecific and intraspecific variation in A₁/A₂‐chromophores proportions, correlating with the light environment. These multiple mechanisms may be a result of the diverse visual environments where Characiformes have evolved.     

Why UV vision and red vision are important for damselfish (Pomacentridae): structural and expression variation in opsin genes

Coral reefs belong to the most diverse ecosystems on our planet. The diversity in coloration and lifestyles of coral reef fishes makes them a particularly promising system to study the role of visual communication and adaptation. Here, we investigated the evolution of visual pigment genes (opsins) in damselfish (Pomacentridae) and examined whether structural and expression variation of opsins can be linked to ecology. Using DNA sequence data of a phylogenetically representative set of 31 damselfish species, we show that all but one visual opsin are evolving under positive selection. In addition, selection on opsin tuning sites, including cases of divergent, parallel, convergent and reversed evolution, has been strong throughout the radiation of damselfish, emphasizing the importance of visual tuning for this group. The highest functional variation in opsin protein sequences was observed in the short‐ followed by the long‐wavelength end of the visual spectrum. Comparative gene expression analyses of a subset of the same species revealed that with SWS1, RH2B and RH2A always being expressed, damselfish use an overall short‐wavelength shifted expression profile. Interestingly, not only did all species express SWS1 – a UV‐sensitive opsin – and possess UV‐transmitting lenses, most species also feature UV‐reflective body parts. This suggests that damsels might benefit from a close‐range UV‐based ‘private’ communication channel, which is likely to be hidden from ‘UV‐blind’ predators. Finally, we found that LWS expression is highly correlated to feeding strategy in damsels with herbivorous feeders having an increased LWS expression, possibly enhancing the detection of benthic algae.     

Rapid light-induced shifts in opsin expression: finding new opsins, discerning mechanisms of change, and implications for visual sensitivity

Light-induced shifts in cone frequency and opsin expression occur in many aquatic species. Yet little is known about how quickly animals can alter opsin expression and, thereby, track their visual environments. Similarly, little is known about whether adult animals can alter opsin expression or whether shifts in opsin expression are limited to critical developmental windows. We took adult wild-caught bluefin killifish (Lucania goodei) from three different lighting environments (spring, swamp and variable), placed them under two different lighting treatments (clear vs. tea-stained water) and monitored opsin expression over 4 weeks. We measured opsin expression for five previously described opsins (SWS1, SWS2B, SWS2A, RH2-1 and LWS) as well as RH2-2 which we discovered via 454 sequencing. We used two different metrics of opsin expression. We measured expression of each opsin relative to a housekeeping gene and the proportional expression of each opsin relative to the total pool of opsins. Population and lighting environment had large effects on opsin expression which were present at the earliest time points indicating rapid shifts in expression. The two measures of expression produced radically different patterns. Proportional measures indicated large effects of light on SWS1 expression, whereas relative measures indicated no such effect. Instead, light had large effects on the relative expression of SWS2B, RH2-2, RH2-1 and LWS. We suggest that proportional measures of opsin expression are best for making inferences about colour vision, but that measures relative to a housekeeping gene are better for making conclusions about which opsins are differentially regulated.     

A Fish Eye Out of Water: Ten Visual Opsins in the Four-Eyed Fish,Anableps anableps

The “four-eyed” fish Anableps anableps has numerous morphological adaptations that enable above and below-water vision. Here, as the first step in our efforts to identify molecular adaptations for aerial and aquatic vision in this species, we describe the A. anableps visual opsin repertoire. We used PCR, cloning, and sequencing to survey cDNA using unique primers designed to amplify eight sequences from five visual opsin gene subfamilies, SWS1, SWS2, RH1, RH2, and LWS. We also used Southern blotting to count opsin loci in genomic DNA digested with EcoR1 and BamH1. Phylogenetic analyses confirmed the identity of all opsin sequences and allowed us to map gene duplication and divergence events onto a tree of teleost fish. Each of the gene-specific primer sets produced an amplicon from cDNA, indicating that A. anableps possessed and expressed at least eight opsin genes. A second PCR-based survey of genomic and cDNA uncovered two additional LWS genes. Thus, A. anableps has at least ten visual opsins and all but one were expressed in the eyes of the single adult surveyed. Among these ten visual opsins, two have key site haplotypes not found in other fish. Of particular interest is the A. anableps-specific opsin in the LWS subfamily, S180γ, with a SHYAA five key site haplotype. Although A. anableps has a visual opsin gene repertoire similar to that found in other fishes in the suborder Cyprinodontoidei, the LWS opsin subfamily has two loci not found in close relatives, including one with a key site haplotype not found in any other fish species. A. anableps opsin sequence data will be used to design in situ probes allowing us to test the hypothesis that opsin gene expression differs in the distinct ventral and dorsal retinas found in this species.     

Conserved visual sensitivities across divergent lizard lineages that differ in an ultraviolet sexual signal

The sensory drive hypothesis predicts the correlated evolution of signaling traits and sensory perception in differing environments. For visual signals, adaptive divergence in both color signals and visual sensitivities between populations may contribute to reproductive isolation and promote speciation, but this has rarely been tested or shown in terrestrial species. We tested whether opsin protein expression differs between divergent lineages of the tawny dragon (Ctenophorus decresii) that differ in the presence/absence of an ultraviolet sexual signal. We measured the expression of four retinal cone opsin genes (SWS1, SWS2, RH2, and LWS) using droplet digital PCR. We show that gene expression between lineages does not differ significantly, including the UV wavelength sensitive SWS1. We discuss these results in the context of mounting evidence that visual sensitivities are highly conserved in terrestrial systems. Multiple competing requirements may constrain divergence of visual sensitivities in response to sexual signals. Instead, signal contrast could be increased via alternative mechanisms, such as background selection. Our results contribute to a growing understanding of the roles of visual ecology, phylogeny, and behavior on visual system evolution in reptiles.     

Molecular cloning of cDNA encoding red opsin gene in the retinas of five Antarctic notothenioid fishes

Previous evidence suggested that notothenioid fish had lost red-sensitive (LWS) visual pigment and photoreceptors, but retained ultraviolet-sensitive (SWS1), blue-sensitive (SWS2), and green-sensitive (RH2) pigments. We used RT-PCR and Southern blot to isolate the LWS opsin gene in five notothenioid species. We determined full-coding LWS opsin sequences and genomic sequences. The expected peak absorbance of the LWS opsin, based on the five-sites rule that is primarily responsible for the spectral sensitivities in vertebrates, ranged from 541 to 553 nm. In Antarctic waters, light of this wavelength penetrates to dozens of meters. Thus, we conclude that notothenioids use tetrachromatic color vision in shallower waters, at least during the Antarctic summer.     

Intraspecific cone opsin expression variation in the cichlids of Lake Malawi

The expression of cone opsin genes is a primary determinant of the characteristics of colour vision. Interspecific variation in opsin expression is common in African cichlids. It is correlated with foraging among cichlids from Lake Malawi, and with ambient light environment among cichlids from Lake Victoria. In this study, we tested whether gene expression varied within species such that it might be important in contributing to divergence. We hypothesized that light attenuation with depth would be correlated with predictable changes in gene expression in Lake Malawi, and that this variation would tune visual sensitivities to match the ambient light environment. We observed significant differences in cone opsin expression in three different comparisons among populations of the same species. Higher LWS expression was found in shallow versus deep Copadichromis eucinostomus. In Metriaclima zebra, individuals from Zimbawe Rock expressed significantly more SWS2B than those from Thumbi West Island, although these locales have similar ambient light environments. Finally, Tropheops gracilior from deeper water had significantly more variation in expression than their shallow counterparts. These results support that gene expression varies significantly between populations of the same species. Surprisingly, these results could not be explained by predicted visual performance as models predicted that differential expression patterns did not confer sensitivity advantages at different depths. This suggested that expression variation did not confer a local sensitivity advantage. Therefore, our findings were contrary to a primary requirement of the sensory bias hypothesis. As such, other explanations for intraspecific gene expression variation need to be tested.     

Functional characterization of visual opsin repertoire in Medaka (Oryzias latipes)

A variety of visual pigment repertoires present in fish species is believed due to the great variation under the water of light environment. A complete set of visual opsin genes has been isolated and characterized for absorption spectra and expression in the retina only in zebrafish. Medaka (Oryzias latipes) is a fish species phylogenetically distant from zebrafish and has served as an important vertebrate model system in molecular and developmental genetics. We previously isolated a medaka rod opsin gene (RH1). In the present study we isolated all the cone opsin genes of medaka by genome screening of a lambda-phage and bacterial artificial chromosome (BAC) libraries. The medaka genome contains two red, LWS-A and LWS-B, three green, RH2-A, RH2-B and RH2-C, and two blue, SWS2-A and SWS2-B, subtype opsin genes as well as a single-copy of the ultraviolet, SWS1, opsin gene. Previously only one gene was believed present for each opsin type as reported in a cDNA-based study. These subtype opsin genes are closely linked and must be the products of local gene duplications but not of a genome-wide duplication. Peak absorption spectra (lambda(max)) of the reconstituted photopigments with 11-cis retinal varied greatly among the three green opsins, 452 nm for RH2-A, 516 nm for RH2-B and 492 nm for RH2-C, and between the two blue opsins, 439 nm for SWS2-A and 405 nm for SWS2-B. Zebrafish also has multiple opsin subtypes, but phylogenetic analysis revealed that medaka and zebrafish gained the subtype opsins independently. The lambda and BAC DNA clones isolated in this study could be useful for investigating the regulatory mechanisms and evolutionary diversity of fish opsin genes.     

Gene Duplication and Divergence of Long Wavelength-Sensitive Opsin Genes in the Guppy, Poecilia reticulata

Female preference for male orange coloration in the genus Poecilia suggests a role for duplicated long wavelength-sensitive (LWS) opsin genes in facilitating behaviors related to mate choice in these species. Previous work has shown that LWS gene duplication in this genus has resulted in expansion of long wavelength visual capacity as determined by microspectrophotometry (MSP). However, the relationship between LWS genomic repertoires and expression of LWS retinal cone classes within a given species is unclear. Our previous study in the related species, Xiphophorus helleri, was the first characterization of the complete LWS opsin genomic repertoire in conjunction with MSP expression data in the family Poeciliidae, and revealed the presence of four LWS loci and two distinct LWS cone classes. In this study we characterized the genomic organization of LWS opsin genes by BAC clone sequencing, and described the full range of cone cell types in the retina of the colorful Cumaná guppy, Poecilia reticulata. In contrast to X. helleri, MSP data from the Cumaná guppy revealed three LWS cone classes. Comparisons of LWS genomic organization described here for Cumaná to that of X. helleri indicate that gene divergence and not duplication was responsible for the evolution of a novel LWS haplotype in the Cumaná guppy. This lineage-specific divergence is likely responsible for a third additional retinal cone class not present in X. helleri, and may have facilitated the strong sexual selection driven by female preference for orange color patterns associated with the genus Poecilia.     

Depth‐dependent plasticity in opsin gene expression varies between damselfish (Pomacentridae) species

Phenotypic plasticity plays an important role in adapting the visual capability of many animal species to changing sensory requirements. Such variability may be driven by developmental change or may result from environmental changes in light habitat, thereby improving performance in different photic environments. In this study, we examined inter‐ and intraspecific plasticity of visual sensitivities in seven damselfish species, part of the species‐rich and colourful fish fauna of the Great Barrier Reef in Australia. Our goal was to test whether the visual systems of damselfish were tuned to the prevailing light environment in different habitats and/or other aspects of their lifestyle. More specifically, we compared the opsin gene expression levels from individuals living in different photic habitats. We found that all species expressed rod opsin (RH1) used for dim‐light vision, and primarily three cone opsins (SWS1, RH2B and RH2A) used for colour vision. While RH1 levels changed exclusively following a diurnal cycle, cone opsin expression varied with depth in four of the seven species. Estimates of visual pigment performance imply that changes in opsin expression adjust visual sensitivities to the dominant photic regime. However, we also discovered that some species show a more stable opsin expression profile. Further, we found indication that seasonal changes, possibly linked to changes in the photic environment, might also trigger opsin expression. These findings suggest that plasticity in opsin gene expression of damselfish is highly species‐specific, possibly due to ecological differences in visual tasks or, alternatively, under phylogenetic constraints.     

Cone visual pigments in two marsupial species: the fat-tailed dunnart (Sminthopsis crassicaudata) and the honey possum (Tarsipes rostratus)

Uniquely for non-primate mammals, three classes of cone photoreceptors have been previously identified by microspectrophotometry in two marsupial species: the polyprotodont fat-tailed dunnart (Sminthopsis crassicaudata) and the diprotodont honey possum (Tarsipes rostratus). This report focuses on the genetic basis for these three pigments. Two cone pigments were amplified from retinal cDNA of both species and identified by phylogenetics as members of the short wavelength-sensitive 1 (SWS1) and long wavelength-sensitive (LWS) opsin classes. In vitro expression of the two sequences from the fat-tailed dunnart confirmed the peak absorbances at 363 nm in the UV for the SWS1 pigment and 533 nm for the LWS pigment. No additional expressed cone opsin sequences that could account for the middle wavelength cones could be amplified. However, amplification from the fat-tailed dunnart genomic DNA with RH1 (rod) opsin primer pairs identified two genes with identical coding regions but sequence differences in introns 2 and 3. Uniquely therefore for a mammal, the fat-tailed dunnart has two copies of an RH1 opsin gene. This raises the possibility that the middle wavelength cones express a rod rather than a cone pigment.     

Molecular evolution of color vision of zebra finch

We have isolated and sequenced the RH1(Tg), RH2(Tg), SWS2(Tg), and LWS(Tg) opsin cDNAs from zebra finch retinas. Upon binding to 11-cis-retinal, these opsins regenerate the corresponding photosensitive molecules, visual pigments. The absorption spectra of visual pigments have a broad bell shape, with the peak being called lambda(max). Previously, SWS1(Tg) opsin cDNA was isolated from zebra finch retinal RNA, expressed in cultured COS1 cells, reconstituted with 11-cis-retinal, and the lambda(max) of the resulting visual pigment was shown to be 359nm. Here, the lambda(max) values of the RH1(Tg), RH2(Tg), SWS2(Tg), and LWS(Tg) pigments are determined to be 501, 505, 440, and 560nm, respectively. Molecular evolutionary analyses suggest that specific amino acid replacements in the SWS1 and SWS2 pigments, resulting from accelerated evolution, must have been responsible for their functional divergences among the avian pigments.     

Testing sensory drive speciation in cichlid fish: Linking light conditions to opsin expression,opsin genotype and female mate preference

Ecological speciation is facilitated when divergent adaptation has direct effects on selective mating. Divergent sensory adaptation could generate such direct effects, by mediating both ecological performance and mate selection. In aquatic environments, light attenuation creates distinct photic environments, generating divergent selection on visual systems. Consequently, divergent sensory drive has been implicated in the diversification of several fish species. Here, we experimentally test whether divergent visual adaptation explains the divergence of mate preferences in Haplochromine cichlids. Blue and red Pundamilia co‐occur across south‐eastern Lake Victoria. They inhabit different photic conditions and have distinct visual system properties. Previously, we documented that rearing fish under different light conditions influences female preference for blue versus red males. Here, we examine to what extent variation in female mate preference can be explained by variation in visual system properties, testing the causal link between visual perception and preference. We find that our experimental light manipulations influence opsin expression, suggesting a potential role for phenotypic plasticity in optimizing visual performance. However, variation in opsin expression does not explain species differences in female preference. Instead, female preference covaries with allelic variation in the long‐wavelength‐sensitive opsin gene (LWS), when assessed under broad‐spectrum light. Taken together, our study presents evidence for environmental plasticity in opsin expression and confirms the important role of colour perception in shaping female mate preferences in Pundamilia. However, it does not constitute unequivocal evidence for the direct effects of visual adaptation on assortative mating.     

Mix and match color vision: tuning spectral sensitivity by differential opsin gene expression in Lake Malawi cichlids

Cichlid fish of the East African Rift Lakes are renowned for their diversity and offer a unique opportunity to study adaptive changes in the visual system in rapidly evolving species flocks. Since color plays a significant role in mate choice, differences in visual sensitivities could greatly influence and even drive speciation of cichlids. Lake Malawi cichlids inhabiting rock and sand habitats have significantly different cone spectral sensitivities. By combining microspectrophotometry (MSP) of isolated cones, sequencing of opsin genes, and spectral analysis of recombinant pigments, we have established the cone complements of four species of Malawi cichlids. MSP demonstrated that each of these species predominately expresses three cone pigments, although these differ between species to give three spectrally different cone complements. In addition, rare populations of spectrally distinct cones were found. In total, seven spectral classes were identified. This was confirmed by opsin gene sequencing, expression, and in vitro reconstitution. The genes represent the four major classes of cone opsin genes that diverged early in vertebrate evolution. All four species possess a long-wave-sensitive (LWS), three spectrally distinct green-sensitive (RH2), a blue-sensitive (SWS2A), a violet-sensitive (SWS2B), and an ultraviolet-sensitive (SWS1) opsin. However, African cichlids determine their spectral sensitivity by differential expression of primarily only three of the seven available cone opsin genes. Phylogenetic analysis suggests that all percomorph fish have similar potential.     

Genetic analyses of visual pigments of the pigeon (Columba livia)

We isolated five classes of retinal opsin genes rh1(Cl), rh2(Cl), sws1(Cl), sws2(Cl), and lws(Cl) from the pigeon; these encode RH1(Cl), RH2(Cl), SWS1(Cl), SWS2(Cl), and LWS(Cl) opsins, respectively. Upon binding to 11-cis-retinal, these opsins regenerate the corresponding photosensitive molecules, visual pigments. The absorbance spectra of visual pigments have a broad bell shape with the peak, being called lambdamax. Previously, the SWS1(Cl) opsin cDNA was isolated from the pigeon retinal RNA, expressed in cultured COS1 cells, reconstituted with 11-cis-retinal, and the lambdamax of the resulting SWS1(Cl) pigment was shown to be 393 nm. In this article, using the same methods, the lambdamax values of RH1(Cl), RH2(Cl), SWS2(Cl), and LWS(Cl) pigments were determined to be 502, 503, 448, and 559 nm, respectively. The pigeon is also known for its UV vision, detecting light at 320-380 nm. Being the only pigments that absorb light below 400 nm, the SWS1(Cl) pigments must mediate its UV vision. We also determined that a nonretinal P(Cl) pigment in the pineal gland of the pigeon has a lambdamax value at 481 nm.     

Molecular Evidence that Only Two Opsin Subfamilies,the Blue Light- (SWS2) and Green Light-Sensitive (RH2), Drive Color Vision in Atlantic Cod (Gadus morhua)

Teleosts show a great variety in visual opsin complement, due to both gene duplication and gene loss. The repertoire ranges from one subfamily of visual opsins (scotopic vision) including rod opsin only retinas seen in many deep-sea species to multiple subfamilies of visual opsins in some pelagic species. We have investigated the opsin repertoire of Atlantic cod (Gadus morhua) using information in the recently sequenced cod genome and found that despite cod not being a deep sea species it lacks visual subfamilies sensitive towards the most extreme parts of the light spectra representing UV and red light. Furthermore, we find that Atlantic cod has duplicated paralogs of both blue-sensitive SWS2 and green-sensitive RH2 subfamilies, with members belonging to each subfamily linked in tandem within the genome (two SWS2-, and three RH2A genes, respectively). The presence of multiple cone opsin genes indicates that there have been duplication events in the cod ancestor SWS2 and RH2 opsins producing paralogs that have been retained in Atlantic. Our results are supported by expressional analysis of cone opsins, which further revealed an ontogenetic change in the array of cone opsins expressed. These findings suggest life stage specific programs for opsin regulation which could be linked to habitat changes and available light as the larvae is transformed into an early juvenile. Altogether we provide the first molecular evidence for color vision driven by only two families of cone opsins due to gene loss in a teleost.     

Salmonid Opsin Sequences Undergo Positive Selection and Indicate an Alternate Evolutionary Relationship in Oncorhynchus

Positive selection can be demonstrated by statistical analysis when non-synonymous nucleotide substitutions occur more frequently than synonymous substitutions (dN>dS). This pattern of sequence evolution has been observed in the rhodopsin gene of cichlids. Mutations in opsin genes resulting in amino acid (AA) replacement appear to be associated with the evolution of specific color patterns and the evolution of courtship behaviors. Within fish, AA replacements in opsin proteins have improved vision at great depths and have occurred in deep-sea species. Salmonids experience diverse photic environments during their life history. Furthermore, sexual selection has resulted in species-specific male and female coloration during spawning. To look for evidence of positive selection in salmonid opsins, we sequenced the RH1, RH2, LWS, SWS1, and SWS2 genes from six Pacific salmon species as well as the Atlantic salmon. These salmonids include landlocked and migratory species and species that vary in their coloration during spawning. In each opsin gene comparison from all species sampled, traditional dN:dS analysis did not indicate positive selection. However, the more sensitive Creevey–McInerney statistical analysis indicates that RH1 and RH2 experienced positive selection early in the evolution and speciation of salmonids.     

Multiple origins of the green-sensitive opsin genes in fish

Vertebrate opsins are divided into four major groups: RH1 (rhodopsins), RH2 (rhodopsinlike with various absorption sensitivities), SWS (short-wavelength sensitive), and LWS/MWS (long and middle-wavelength sensitive) groups. The green opsin genes (g101 _Af and g101 _Af ) in a Mexican characin Astyanax fasciatus belong to the LWS/MWS group, whereas those in goldfish belong to the RH2 group (Yokoyama 1994, Mol Biol Evol 11:32–39). A newly isolated opsin gene (rh11 _Af ) from A. fasciatus contains five exons and four introns, spanning 4.2 kilobases from start to stop codons. This gene is most closely related to the two green opsin genes of goldfish and belongs to the RH2 group. In the LWS/MWS group, gene duplication of the ancestral red and green opsin genes predates the speciation between A. fasciatus and goldfish, suggesting that goldfish also has an additional gene which is orthologous to g101 _Af and g103 _Af .Correspondence to: S. Yokoyama