Molecular evolution of color vision of zebra finch

Gene 259 (2000) 17–24 www.elsevier.com/locate/gene

Molecular evolution of color vision of zebra finch Shozo Yokoyama *, Nathan S. Blow, F. Bernhard Radlwimmer Department of Biology, Syracuse University, 130 College Place, Syracuse, NY 13244, USA Received 3 April 2000; received in revised form 19 June 2000; accepted 25 August 2000 Received by T. Gojobori

Abstract We have isolated and sequenced the RH1 , RH2 , SWS2 , and LWS opsin cDNAs from zebra finch retinas. Upon binding Tg Tg Tg Tg 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 l . Previously, SWS1 opsin cDNA was isolated from max Tg zebra finch retinal RNA, expressed in cultured COS1 cells, reconstituted with 11-cis-retinal, and the l of the resulting visual max pigment was shown to be 359 nm. Here, the l values of the RH1 , RH2 , SWS2 , and LWS pigments are determined to max Tg Tg Tg Tg be 501, 505, 440, and 560 nm, 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. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Color vision; Visual pigments; Wavelength absorption; Zebra finch

1. Introduction Zebra finches (Taeniopygia guttata) are sexually dimorphic birds, in which the male’s plumage is more ornate than that of the female. There is extensive evidence for sexual selection and the development of sexual preferences of these birds based on their color preferences (Zann, 1996). Females prefer males with bright red beaks, whereas males prefer females with yellow–orange beaks (Burley and Coppersmith, 1987; De Kogel and Prijs, 1996). The zebra finches are also attracted sexually to members of the opposite sex that are wearing specific colored leg bands (Price and Burley, 1994; Swaddle and Cuthill, 1994). The birds with ‘attractive’ colors tend to have higher reproductive success (Price and Burley, 1994), lower parental investment per offspring, and lower mortality rates (Burley, 1988). Several authors, however, have failed to replicate some of these results (for a review, see Collins and ten Cate, 1996). These conflicting results seem to occur because of the absence or presence of ultraviolet ( UV ) light under the experimental conditions Abbreviations: l , cut-off wavelength by a colored oil droplet; cut l , wavelength of maximal absorption; LWS, long wavelength-sensimax tive; MWS, middle wavelength-sensitive; RH1, rhodopsin; RH2, rhodopsin-like; SWS1, short wavelength-sensitive type 1; SWS2, short wavelength-sensitive type 2; Tg, Taeniopygia guttata. * Corresponding author. Tel.: +315-443-9166; fax: +315-443-2012. E-mail address: [email protected] (S. Yokoyama)

(Hunt et al., 1997; Bennett et al., 1996). The absorption spectra of the photoreceptors and light sensitive molecules, visual pigments, of the zebra finch have also been evaluated using microspectrophotometry (MSP) (Bowmaker et al., 1997). Despite the wealth of behavioral and physiological observations, the molecular bases of the color vision of the zebra finches are not well understood. So far, only the UV opsin cDNA clone of the zebra finch has been isolated and characterized (Yokoyama et al., 2000). By expressing it in cultured COS1 cells, reconstituting the products with 11-cis-retinal, the l value of the resulting pigments has max been determined to be 359 nm (Yokoyama et al., 2000). In this paper, we report the cloning and molecular characterization of the remaining four paralogous opsin cDNAs from the zebra finch retinas. We then evaluate the l max values of the corresponding visual pigments using the in vitro assays. Using the amino acid sequences and l max values of the pigments of the zebra finch and other avian species, we also infer the molecular bases for the l max shifts of the different evolutionary groups of pigments.

2. Materials and methods 2.1. Background information The photoreceptor cells of the zebra finches can be distinguished into three classes: rods, single cones, and

0378-1119/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 0 0 ) 0 0 43 5 - 2

18

S. Yokoyama et al. / Gene 259 (2000) 17–24

double cones. Like many other birds, color vision of the zebra finches is regulated strongly by colored oil droplets in the inner segments of their cone photoreceptors. Each single cone in the zebra finch contains one of the four types of oil droplets: red (R), yellow ( Y ), clear or colorless (C ), and transparent ( T ). These oil droplets are classified by their ‘cut-off ’ wavelengths (l ) at cut about 570 nm (R), 510 nm ( Y ), 410 nm (C ), and with no significant absorption throughout the spectrum ( T ) (Bowmaker et al., 1997). The principal member of the double cone contains a pale (P) oil droplet with a l cut at about 430 nm, whereas the accessory member of the double cone rarely contains an oil droplet (Bowmaker et al., 1997). The retinal visual pigments are classified into evolutionarily distinct RH1, RH2, SWS1, SWS2, and LWS/MWS groups ( Yokoyama, 1994, 1995, 1997, 2000a; Yokoyama and Yokoyama, 1996; see also Okano et al., 1992; Hisatomi et al., 1994). In the zebra finch, the RH1 pigments are found in the rod pigments, Tg whereas the LWS pigments are found in R-type cones Tg and in both members of double cones (Bowmaker et al., 1997). Similarly, RH2 , SWS2 , and SWS1 pigments Tg Tg Tg are found only in the Y-, C-, and T-type cones, respectively (Bowmaker et al., 1997). 2.2. Screening, cloning, and sequencing of cDNA clones Twenty zebra finches were purchased from local pet stores and total RNA was isolated from the retinas of these birds (for the procedure, see Kawamura and Yokoyama, 1998; Yokoyama et al., 1998; Yokoyama and Radlwimmer, 1998, 1999). Following the protocols supplied by the manufacturer (Stratagene, La Jolla, CA), we constructed a cDNA library in the lZAPII vector by using 5 mg of the RNA and Poly (A) Quick mRNA isolation kit, ZAP-cDNA Synthesis kit, and Gigapack III Gold extracts. About 4×105 recombinant plaques were transferred to nylon membrane (Hybond-N+, Amersham) for sequential hybridization with four 32P random-labeled opsin cDNA probes: RH1 opsin cDNA from bovine (Bos tauras) (Nathans and Hogness, 1983), LWS cDNA from human (Nathans et al., 1986), and RH2 and SWS2 cDNAs from American chameleon (Anolis carolinensis) ( Kawamura and Yokoyama, 1995, 1996). Hybridization was carried out at 55°C and hybridized membranes were washed four times (30 min each) in 1×SSC (0.15 M NaCl/0.015 M Na citrate)/0.1% SDS at 55°C. Old 3 probes were removed from the membranes by washing them in 0.4 M NaOH at 45°C for 30 min and then in 0.1×SSC/0.1% SDS/0.2 M Tris (pH 7.5) at 45°C for 30 min. From these screenings, 14 RH1 , five RH2 , six Tg Tg SWS2 , and seven LWS opsin cDNA clones have Tg Tg been isolated, among which 14, four, six, and three were

found to contain the entire coding regions, respectively. Four representative clones each from the RH1 , Tg RH2 , SWS2 , and LWS opsin cDNA clones were Tg Tg Tg subcloned into pBluescript SK(−) vectors. The nucleotide sequences of these clones were determined by cycle sequencing reactions using the Sequitherm Excel II Long-Read kits ( Epicentre Technologies, Madison, WI ) with dye-labeled M13 forward and reverse primers. Reactions were run on a LI-COR 4200LD automated DNA sequencer (LI-COR, Lincoln, NE). 2.3. RT-PCR of cDNAs, regeneration of visual pigments, and spectral analyses Using the primers ( Fig. 1), we amplified the four types of full-length opsin cDNAs from 200 ng of total RNA by reverse transcriptase-polymerase chain reaction ( RT-PCR) method. The cDNA synthesis by RT-PCR amplification was performed as described previously ( Kawamura and Yokoyama, 1998; Yokoyama et al., 1998; Yokoyama and Radlwimmer, 1999). Nucleotide sequences of the entire region of the cDNA clones were determined by cycle sequencing reactions. These cDNAs were subcloned into the expression vector pMT, transiently expressed in COS1 cells, and the transfected cells were incubated with 11-cis-retinal (Storm Eye Institute, Medical University of South Carolina) in the dark. The pMT vector contains the last 15 codons of the bovine RH1 gene that are necessary for immunoaffinity purification (for more details, see Yokoyama, 2000b). It is known that the extra 15 amino acids do not affect their spectral properties ( Yokoyama, 2000b). UV visible absorption spectra of visual pigments were recorded at 20°C in the dark (dark spectra), using a Hitachi U-3000 dual beam spectrophotometer. When the visual pigments were bleached by a 60 W room lamp with 440 nm cut-off filter, the peaks of the dark spectra are replaced by those at 380 nm, which is caused by alltrans-retinal dissociated from an opsin. Thus, the l max

Fig. 1. Oligonucleotide primers used for RT-PCR amplification of zebra finch opsin mRNAs.

S. Yokoyama et al. / Gene 259 (2000) 17–24

19

values of visual pigments can also be estimated by subtracting the new spectra from the dark spectra (dark– light spectra). Recorded spectra were analyzed using SigmaPlot software (Jandel Scientific, San Rafael, CA).

branch lengths were calculated from { p/[n(1−p)]}1/2, where n is the number of amino acids compared (Nei, 1987).

2.4. Sequence data analysis

3. Results

The RH1 , RH2 , SWS2 , SWS1 , and LWS Tg Tg Tg Tg Tg pigments of the zebra finch have been compared to the corresponding pigments of the chicken (Gallus gallus; GenBank accession nos. D00702, M92038, M92037, M92039, and M62903), pigeon (Columba livia; AF149230, AF149232, AF149238, AF149234, and AF149243), and chameleon (Anolis carolinensis; L31503, AF134189, AF133907, AF134192, and U08131). For the SWS1 group, the orthologous pigments of parakeet (Melopsithacus undulatus; Y11787) and canary (Serinus canaria; Das et al., 1999) are also included. The numbers (K ) of amino acid substitutions per site for all pairwise sequences were estimated by a Poisson correction. The phylogenetic tree was reconstructed by applying the NJ method (Saitou and Nei, 1987) to the K values. The reliability of the phylogenetic tree was evaluated by the bootstrap analysis with 1000 replications (Felsenstein, 1985). Standard errors associated with the

3.1. Molecular characterization and phylogenetic relationship of the zebra finch pigments Previously, the SWS1 opsin cDNA clone was isoTg lated from a retinal cDNA library of the zebra finch ( Yokoyama et al., 2000). Here, screening the same cDNA library, we have obtained four additional different types of opsin cDNA clones. RH1 (GenBank Tg accession no. AF222329), RH2 (AF222330), Tg SWS2 (AF222332), and LWS (AF222333) opsins, Tg Tg deduced from the nucleotide sequences of these four cDNA clones, consist of 351, 355, 362, and 365 amino acids, respectively (Fig. 2). The length of SWS1 opsin Tg is somewhat shorter and consists of 346 amino acids ( Yokoyama et al., 2000). Phylogenetic analysis shows that the visual pigments of the zebra finch, chicken, pigeon, parakeet, canary, and chameleon are distinguished into RH1, RH2, SWS1,

Fig. 2. Alignment of the amino acid sequences of the zebra finch pigments. RH1, RH2, SWS1, SWS2, and LWS denote RH1 , RH2 , SWS1 , Tg Tg Tg SWS2 , and LWS pigments, respectively. Dots indicate the identity of the amino acids with those of the RH1 pigment. Gaps necessary to Tg Tg increase the sequence similarity are indicated by dashes (–). Seven putative transmembrane regions (Hargrave et al., 1983) are indicated by horizontal lines. Five LWS pigment-specific amino acids are boxed.

20

S. Yokoyama et al. / Gene 259 (2000) 17–24

Fig. 3. The phylogenetic tree for RH1, RH2, SWS1, SWS2, and LWS pigments of birds and chameleon. The bootstrap supports are indicated besides branch nodes. Values after P indicate l values. max

SWS2, and LWS/MWS pigment groups, which are generated by four gene duplication events (Fig. 3). The bootstrap support values for the five groups of pigments range from 0.95 to 1.0 and are highly reproducible. The comparative analyses of the amino acid sequences of more than 100 vertebrate pigments have shown that these gene duplication events occurred prior to the divergence of various vertebrate species and the vertebrate ancestor already possessed all of these pigments ( Yokoyama and Yokoyama, 1996; Yokoyama, 1997, 2000a). Within each evolutionary group, the avian pigments form one cluster, to which the chameleon pigment is the most distantly related (Fig. 3). The exact phylogenetic positions of the avian pigments, however, cannot be established (see also Yokoyama et al., 2000). 3.2. Light absorption spectra The l values of pigments have been measured max from the dark spectra (Fig. 4) and from the dark–light spectra (Fig. 4, insets). When the regenerated pigments were exposed to light, new absorption peaks at ~380 nm were observed (result not shown), indicating that 11-cisretinal in the pigment was isomerized by light and alltrans-retinal was released. When measured in the dark, RH1 , RH2 , SWS2 , and LWS pigments have Tg Tg Tg Tg l values at 501±1, 505±1, 440±1, and 560±3 nm, max respectively. The respective l values estimated from max

the dark–light difference are given by 504±1, 506±1, 458±1, and 560±2 nm. The corresponding dark and dark–light spectra of the RH1 , RH2 , and LWS Tg Tg Tg pigments are close to each other. For the SWS2 Tg pigment, however, the dark–light spectrum is 18 nm higher than the dark spectrum. Curiously, the dark–light spectra for the SWS2 pigments of chameleon ( Kawamura and Yokoyama, 1998) and pigeon ( Kawamura et al., 1999) are also 10–15 nm higher than the corresponding dark spectra. At present, the cause for the difference between the dark and dark– light spectra for the SWS2 pigments is not clear. From Fig. 4, we can see that the ratio between the protein absorption peak and pigment absorption peak of the SWS2 pigment is 2.2, showing a highly reliable level Tg of the pigment regeneration. Thus, the direct estimate of the dark spectrum seems to be more reliable than the indirect estimate of the dark–light measure. Indeed, the dark spectra of the RH1 (501 nm), RH2 (505 nm), Tg Tg SWS2 (440 nm), and LWS (560 nm) pigments are Tg Tg very close to the corresponding estimates 507±1, 503±3, 443±5, and 567±3 nm obtained by MSP (Bowmaker et al., 1997) (Table 1). So far, the l max value for the SWS1 pigment has been evaluated only Tg by using the in vitro assay and is given by 359 nm ( Yokoyama et al., 2000). As noted earlier, the RH2 , SWS1 , SWS2 , and Tg Tg Tg LWS pigments are expressed specifically in Y-, C-, T-, Tg and R-type cones, respectively. In the zebra finch, the l values of the visual pigments and those (effective max l values) of the cones are not necessarily the same. max It turns out that the ‘effective l ’ values of the C- and max T-cones are close to those of the visual pigments, whereas those of the Y- and R-type cones are more than 30 nm red-shifted than the l values of the max corresponding visual pigments ( Table 1). 3.3. Evolutionary rates of amino acid replacement The comparable branch lengths of the visual pigments from the birds and chameleon in each of the RH1, RH2, Table 1 Absorption spectra of the zebra finch pigments Pigments

RH1 RH2 SWS1 SWS2 LWS

l values (nm) max In vitro

MSPa

Oil droplet (l )b cut

501±1 505±1 359±1c 440±1 560±1

507±1 503±3 NDd 430±5 567±3

– Y (~510) T (no absorption) C (~410) R (~570)

a Bowmaker et al. (1997). b Bowmaker and Knowles (1977). c Yokoyama et al. (1998). d ND, not determined.

Effective l (nm)b max – 530–550 ~360 ~450 600–620

21

S. Yokoyama et al. / Gene 259 (2000) 17–24

Fig. 4. Absorption spectra of regenerated zebra finch pigments measured in the dark and the dark–light difference spectra (insets).

SWS1, SWS2, and LWS/MWS groups do not differ significantly (results not shown; see also Fig. 3). Thus, by computing the weighted averages of the K values (see Materials and methods), we can evaluate the evolutionary rates (k) of amino acid replacement for the avian and chameleon pigments separately. It has been suggested that bird and chameleon diverged about 220 million years ago (McLaughlin and Dayhoff, 1972; Kumar and Hedges, 1998). Using this divergence time, the k values for the avian SWS1 and SWS2 pigments are given by 0.34–0.40×10−9/ site/year, whereas those for the paralogous RH1, RH2, and LWS pigments are given by 0.21–0.26×10−9/ site/year ( Table 2). Among the avian pigments, SWS1 pigment is evolving significantly faster than RH2 and LWS pigments and SWS2 pigment than RH1, RH2,

Table 2 Evolutionary rates of amino acid substitution per site per year (×109)* Pigments

Birds

Chameleon

RH1 RH2 SWS1 SWS2 LWS

0.26±0.029a,k 0.21±0.026b,c 0.34±0.034c,e 0.40±0.038a,b,d 0.22±0.027d,e

0.42±0.039f,g,k 0.19±0.025f,h,i 0.35±0.035i,j 0.32±0.033h 0.25±0.029g,j

* Standard errors were computed from { p/[n(1−p)]}1/2, where p is the proportion of different amino acids per site and n (=340) is the number of amino acid sites compared. a,b,c,d,e,f,g,h,i,k Difference in branch lengths is significant at 1% level. j Difference in branch lengths is significant at 5% level.

22

S. Yokoyama et al. / Gene 259 (2000) 17–24

and LWS pigments ( Table 2). Similarly, among the chameleon pigments, SWS1 and SWS2 pigments are evolving faster than RH2 and LWS pigments, but RH1 pigment is evolving at the fastest rate. When the orthologous RH2, SWS1, SWS2, and LWS pigments of the birds and chameleon are compared separately, their k values are similar, but the chameleon RH1 pigment is evolving significantly faster than the avian counterpart ( Table 2).

4. Discussion The avian RH1, RH2, SWS1, SWS2, and LWS pigments have l values at 501–503, 503–508, 359– max 415, 440–455, and 559–571 nm, respectively ( Fig. 3). Thus, the difference between the two extreme l values max of the SWS1 pigments is the largest (56 nm), followed by those of the SWS2 pigments (19 nm), LWS pigments (13 nm), RH2 pigments (6 nm), and RH1 pigments (3 nm), in that order. We have just seen that, among the avian pigments, SWS1 and SWS2 pigments are evolving faster than RH1, RH2, and LWS pigments. Thus, there exists a positive correlation between the evolutionary rate of amino acid replacement and the level of the divergence among the l values of max the orthologous pigments. Where does this correlation come from? Generally, RH1 pigments are required for the dim vision and are characterized by their l values at around 500 nm. max Because of this rather strict functional requirement, most amino acid changes occurring in the avian RH1 pigments must have been eliminated by purifying selection. Amino acid replacements D83N, E122Q, and A292S in the RH1 pigments are known to cause blue shifts in the l (e.g. see Yokoyama, 1999, 2000c). The max majority of the currently known RH1 pigments have amino acids D83, E122, and A292, strongly suggesting that the ancestral RH1 pigment also had them ( Yokoyama, 1999, 2000c). Certainly, these amino acids have not changed during the evolution of the avian RH1 pigments ( Fig. 2). Among the RH1 pigments, the chameleon pigment is evolving with the fastest rate. It turns out that the chameleon is unique among the terrestrial vertebrates, having a pure-cone retina (Crescitelli, 1972). The accelerated evolutionary rate of the chameleon RH1 pigment is probably due to the evolutionary adaptation of the rod-specific RH1 pigment to the cone photoreceptor ( Yokoyama, 1997). Six chameleon RH1 pigment-specific amino acid replacements S22N, M155I, F159C, N199H, E232A, and T319M have been identified ( Kawamura and Yokoyama, 1994), some of which might have been important for specific functional changes required for a cone RH1 pigment. In the zebra finch, the Y- and R-type oil droplets function as cut-off filters (Bowmaker et al., 1997). In

the Y-cone, the RH2 pigment has a l value at Tg max 505 nm, whereas the Y-type oil droplet has a l value cut at 510 nm. Combining these two absorption spectra, the Y-cone achieves the ‘effective l ’ value at about max 540 nm ( Table 1). Amino acid replacements D83N, E122Q, A164S, and M207L in the RH2 pigments of vertebrates cause blue shifts in the l ( Yokoyama, max 1999, 2000c; Yokoyama et al., 1999). Since all extant RH2 pigments of birds and reptiles have amino acids D83, Q122, A164, and M207, the ancestral RH2 pigment must also have had these amino acids. Again, these functionally important amino acids have not changed during the avian evolution. Similarly, using the l value of 560 nm of LWS max Tg pigments and the l value of 570 nm of the R-type oil cut droplet, the zebra finch R-cone achieves the ‘effective l ’ at about 610 nm ( Table 1). The differences in the max l values of the vertebrate LWS/MWS pigments are max determined mostly by amino acid differences at the five critical sites ( Yokoyama and Radlwimmer, 1998, 1999). As suspected, like other LWS pigments, LWS pigment Tg has LWS pigment-specific S164, H181, Y261, T269, and A292 (Fig. 2). Clearly, in the zebra finch, the Y- and R-cones achieve specific ‘effective l ’ values through max the interaction between the absorption spectra of the visual pigments and oil droplets. Thus, once such interactive relationships were established, most mutations in the RH2 and LWS pigments must have been eliminated by purifying selection. Compared with the function of the Y- and R-type oil droplets, the role of C-type oil droplet is less apparent. However, the low optical densities of these droplets suggest that the cones containing this oil droplet have much wider ranges of spectral sensitivity (Bowmaker et al., 1997). Furthermore, the T-type oil droplet has no significant light absorption throughout the spectrum. Consequently, compared with the RH1, RH2, and LWS pigments, the SWS1 and SWS2 pigments regulate the ‘effective l ’ of the photoreceptor cells more directly, max suggesting that SWS1 and SWS2 pigments have much stronger impacts on the divergence of the ‘effective l ’ values of the cones. In order to acquire their new max functions, the SWS1 and SWS2 pigments must have experienced certain periods of relaxation from the pressure of the purifying selection, causing the accelerated evolution. Then, animals must have utilized, if necessary, certain amino acid changes that cause the l shifts, max providing an opportunity for the adaptive changes of the visual pigments. Following the residue number of the SWS1 pigment, Tg the UV pigments of zebra finch, canary, and parakeet have C84, whereas all other orthologous SWS1 pigments, including the violet pigments of chicken and pigeon, have S84. In fact, using mutagenesis experiments and comparative amino acid sequences, we can show that the UV pigments of birds have evolved from the

S. Yokoyama et al. / Gene 259 (2000) 17–24

violet pigments by a single amino acid replacement, S84C ( Yokoyama et al., 2000). Thus, the accelerated evolution of the SWS1 pigments must have produced the opportunity for the zebra finch, canary, and parakeet, or for their common ancestor, to utilize this critical mutation required for the development of UV pigments. Among the currently known avian SWS2 pigments, the chicken pigment has the most red-shifted l max (Fig. 3). We already have a partial answer for the cause of this change. It is known that A269T in bovine RH1 pigment shifts the l by 14 nm toward red (Chan max et al., 1992), whereas the reverse amino acid change at the corresponding site of the human LWS pigment causes a 16 nm blue shift (Asenjo et al., 1994). At the corresponding sites, the orthologous zebra finch, pigeon, chicken, and chameleon pigments have C, S, T, and A, respectively. Thus, a significant part of the red shift in the l value of the chicken SWS2 pigment must have max been caused by the introduction of T at this site. This amino acid change must have been another byproduct of the accelerated evolution of the bird SWS2 pigments. Given the importance of vision, it is not surprising to see that animals have optimized their visual capabilities to adapt to their specific photic environments. As we saw already, gene duplication was the major force to allow such functional differentiation ( Fig. 3). Namely, the existence of two copies of the same gene enables one of the copies to accumulate mutations and eventually to emerge as a new gene, while the other copy retains the old function (Ohno, 1970). Thus, the five extant evolutionary groups of retinal pigments have been accomplished first by the creation of the five types of visual pigments by four gene duplication events prior to the vertebrate radiation and second by amino acid replacements. Nathans and colleagues have cloned the RH1 opsin genes of the bovine and human (Nathans and Hogness, 1983, 1984) and the human SWS1, MWS, and LWS opsin genes (Nathans et al., 1986). Using these and subsequently derived cDNA clones, the opsin genes from various species have been cloned. Today, we know more than 100 vertebrate visual pigments of which both amino acid sequence and l value are characterized max ( Yokoyama, 1999, 2000c). Comparative analyses of these sequence data have proven to be a powerful approach in identifying the potentially important amino acid changes that may be responsible for the spectral tuning of visual pigments. Based upon the amino acid changes identified in this way, site-directed mutagenesis experiments have been conducted. From these analyses, the molecular bases of the blue shifts in the l max ( Yokoyama, 1995; Yokoyama et al., 1999, 2000; Lin et al., 1998; Fasick and Robinson, 1998; Fasick et al., 1999) and red–green color vision (Chan et al., 1992; Asenjo et al., 1994; Sun et al., 1997) have been elucidated.

23

Thus, comparative data analyses can be used as a convenient tool in designing mutagenesis experiments. These observations clearly show that molecular evolution can have a much more practical use than had been imagined before ( Yokoyama, 1995, 1997, 2000c; Yokoyama and Yokoyama, 1996). The theoretical basis for this evolutionary approach can be traced back to Ohno (1970), who greatly improved our understanding of the creative role of gene duplication in the emergence of a gene having a new function.

Acknowledgement Comments by Ruth Yokoyama were greatly appreciated. This work was supported by NIH grant GM-42379.

References Asenjo, A.B., Rim, J., Oprian, D.D., 1994. Molecular determination of human red/green color discrimination. Neuron 12, 1131–1138. Bennett, A.T.D., Cuthill, I.C., Partridge, J.C., 1996. Ultraviolet vision and mate choice in zebra finches. Nature 380, 433–435. Bowmaker, J.K., Knowles, A., 1977. The visual pigments and oil droplets of the chicken retina. Vision Res. 17, 755–764. Bowmaker, J.K., Heath, L.A., Wilkie, S.E., Hunt, D.M., 1997. Visual pigments and oil droplets from six classes of photoreceptor in the retinas of birds. Vision Res. 37, 2183–2194. Burley, N.T., 1988. The differential allocation hypothesis: an experimental test. Am. Nat. 132, 611–628. Burley, N.T., Coppersmith, C.B., 1987. Bill color preferences of zebra finches. Ethology 76, 133–151. Chan, T., Lee, M., Sakmar, T.P., 1992. Introduction of hydroxylbearing amino acids causes bathochromic spectral shifts responsible for red–green color pigment spectral tuning. J. Biol. Chem. 267, 9478–9480. Collins, S.A., ten Cate, C., 1996. Does beak colour affect female preference in zebra finches? Anim. Behav. 52, 105–112. Crescitelli, F., 1972. The visual cells and visual pigments of the vetebrate eye. In: Dartnall, H.J.A. ( Ed.), Handbook of Sensory Physiology vol. VII/1. Springer, Berlin, pp. 245–363. Das, D., Wilkie, S.E., Hunt, D.M., Bowmaker, J.K., 1999. Visual pigments and oil droplets in the retina of a passerine bird, the canary Serinus canaria: microspectrophotometry and opsin sequences. Vision Res. 39, 2801–2805. De Kogel, K., Prijs, H.J., 1996. Effects of brood size manipulations on sexual attractiveness of offspring in the zebra finch. Anim. Behav. 51, 699–708. Fasick, J.I., Robinson, P.R., 1998. Mechanism of spectral tuning in the dolphin visual pigments. Biochemistry 37, 433–438. Fasick, J.I., Lee, N., Oprian, D.D., 1999. Spectral tuning in the human blue cone pigment. Biochemistry 38, 11 593–11 596. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 597–600. Hargrave, P.A., McDowell, J.H., Curtis, D.R., Wang, J.K., Juszczak, E., Fong, S.L., Mohanna Rao, J.K., Argos, P., 1983. The structure of bovine rhodopsin. Biophys. Struct. Mech. 9, 235–244. Hisatomi, O., Kayada, S., Aoki, Y., Iwasa, T., Tokunaga, F., 1994. Phylogenetic relationship among vertebrate visual pigments. Vision Res. 34, 3097–3102.

24

S. Yokoyama et al. / Gene 259 (2000) 17–24

Hunt, S., Cuthill, I.C., Swaddle, J.P., Bennett, A.T.D., 1997. Ultraviolet vision and band-colour preferences in female zebra finches, Taeniopygia guttata. Anim. Behav. 54, 1383–1392. Kawamura, S., Blow, N.S., Yokoyama, S., 1999. Genetic analyses of visual pigments of the pigeon (Columba livia). Genetics 153, 1839–1850. Kawamura, S., Yokoyama, S., 1994. Cloning of the rhodopsin-encoding gene from the rod-less lizard Anolis carolinensis. Gene 149, 267–270. Kawamura, S., Yokoyama, S., 1995. Paralogous origin of the rhodopsin like opsin genes in lizards. J. Mol. Evol. 40, 594–600. Kawamura, S., Yokoyama, S., 1996. Phylogenetic relationships among short wavelength-sensitive opsins of American chameleon (Anolis carolinensis) and other vertebrates. Vision Res. 36, 2797–2804. Kawamura, S., Yokoyama, S., 1998. Functional characterization of visual and nonvisual pigments of American chameleon (Anolis carolinensis). Vision Res. 38, 37–44. Kumar, S., Hedges, S.B., 1998. A molecular time scale for vertebrate evolution. Nature 392, 917–920. Lin, S.W., Kochendoerfer, G.G., Carroll, K.S., Wang, D., Mathies, R.A., Sakmar, T.P., 1998. Mechanisms of spectral tuning in blue cone pigments: visible and Raman spectrophotometry of blueshifted rhodopsin mutants. J. Biol. Chem. 273, 24 583–24 591. McLaughlin, P.J., Dayhoff, M.O., 1972. Evolution of species and proteins: a time scale. In: Dayhoff, M.O. ( Ed.), Atlas of Protein Sequence and Structure vol. 5. National Biomedical Research Foundation, Silver Springs, MD, pp. 47–66. Nathans, J., Hogness, D.S., 1983. Isolation, sequence analysis, and intron–exon arrangement of the gene encoding bovine rhodopsin. Cell 34, 807–814. Nathans, J., Hogness, D.S., 1984. Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc. Natl. Acad. Sci. USA 81, 4851–4855. Nathans, J., Thomas, D., Hogness, D.S., 1986. Molecular genetics of human color vision: the genes encoding blue, green, and red pigments. Science 232, 193–202. Nei, M., 1987. Molecular Evolutionary Genetics. Columbia University Press, New York. Ohno, S., 1970. Evolution by Gene Duplication. Springer, Berlin. Okano, T., Kojima, D., Fukada, Y., Shichida, Y., Yoshizawa, T., 1992. Primary structures of chicken cone visual pigments: vertebrate rhodopsins have evolved out of cone visual pigments. Proc. Natl. Acad. Sci. USA 89, 5932–5936. Price, D.K., Burley, N.T., 1994. Constraints on the evolution of attrac-

tive traits: selection in male and female zebra finches. Am. Nat. 144, 908–934. Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for reconstructing phylogenetic tree. Mol. Biol. Evol. 4, 406–425. Swaddle, J.P., Cuthill, I.C., 1994. Preference for symmetric males by female zebra finches. Nature 367, 165–166. Sun, H., Macke, J.P., Nathans, J., 1997. Mechanisms of spectral tuning in the mouse green cone pigment. Proc. Natl. Acad. Sci. USA 94, 8860–8865. Yokoyama, S., 1994. Gene duplications and evolution of the short wavelength-sensitive visual pigments in vertebrates. Mol. Biol. Evol. 11, 32–39. Yokoyama, S., 1995. Amino acid replacements and wavelength absorption of visual pigments in vertebrates. Mol. Biol. Evol. 12, 53–61. Yokoyama, S., 1997. Molecular genetic basis of adaptive selection: examples from color vision in vertebrates. Annu. Rev. Genet. 31, 311–332. Yokoyama, S., 1999. Molecular bases of color vision in vertebrates. Genes Genet. Syst. 74, 189–199. Yokoyama, S., 2000a. Molecular evolution of vertebrate visual pigments. Prog. Ret. Eye Res. 19, 385–419. Yokoyama, S., 2000b. Phylogenetic analysis and experimental approaches to study color vision in vertebrates. Vertebrate Phototransduction and the Visual Cycle, Palczewski, K. ( Ed.), Methods Enzymol. 315, 312–325. Yokoyama, S., 2000c. Color vision of the coelacanth (Latimeria chalumnae) and adaptive evolution of rhodopsin (RH1) and rhodopsin-like (RH2) pigments. J. Hered. 91, 215–220. Yokoyama, S., Radlwimmer, F.B., 1998. The ‘five-sites’ rule and the evolution of red and green color vision in mammals. Mol. Biol. Evol. 15, 560–567. Yokoyama, S., Radlwimmer, F.B., 1999. The molecular genetics of red and green color vision in mammals. Genetics 153, 919–932. Yokoyama, S., Yokoyama, R., 1996. Adaptive evolution of photoreceptors and visual pigments in vertebrates. Annu. Rev. Ecol. Syst. 27, 543–567. Yokoyama, S., Radlwimmer, F.B., Kawamura, S., 1998. Regeneration of ultraviolet pigments of vertebrates. FEBS Lett. 423, 155–158. Yokoyama, S., Zhang, H., Radlwimmer, F.B., Blow, N.S., 1999. Adaptive evolution of color vision of the Comoran coelacanth (Latimeria chalumnae). Proc. Natl. Acad. Sci. USA 96, 6279–6284. Yokoyama, S., Radlwimmer, F.B., Blow, N.S., 2000. Ultraviolet pigments in birds evolved from violet pigments by one amino change. Proc. Natl. Acad. Sci. USA. in press. Zann, R.A., 1996. The Zebra Finch. A Synthesis of Field and Laboratory Studies. Oxford University Press, Oxford.