Genetic Analyses of Visual Pigments of the Pigeon (Columba livia)

Corresponding author: Shozo Yokoyama, Department of Biology, Syracuse University, 130 College Pl., Syracuse, NY 13244., syokoyam{at}mailbox.syr.edu (E-mail)

Communicating editor: N. TAKAHATA

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

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 max. 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 max of the resulting SWS1_Cl pigment was shown to be 393 nm. In this article, using the same methods, the max 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 max value at 481 nm.

MOST vertebrates have two kinds of photoreceptor cells, rods and cones. Rods function in dim light, while cones function in bright light and are responsible for color vision. Photosensitive molecules, visual pigments, are located in the outer segments of these photoreceptors, each of which consists of a transmembrane protein, an opsin, and a chromophore, either 11-cis-retinal or 11-cis 3, 4-dehydroretinal (for review see YOKOYAMA and YOKOYAMA 1996 ). These visual pigments are characterized by their wavelengths of maximal absorption (max). Since the chromophore is universal to visual pigments, the wide range of max values from UV to infrared is generated mainly by the structural differences among various opsins. In many diurnal birds and reptiles, color vision is further modified by colored oil droplets in the inner segments of their cones (WALLS 1942 ; BOWMAKER 1991 ). Although their exact functions have not been fully elucidated, the oil droplets often contain a high concentration of carotenoids and are likely to serve as cut-off filters (BOWMAKER 1991 ).

Because of its easy access and suitability to behavioral and physiological studies, the color vision of the pigeon (Columba livia) has been studied extensively. The cone photoreceptor cells in the pigeon retina can be classified into single cones and double cones. The oil droplets in the single cones have been classified into red (R), yellow (Y), clear (C), and transparent (T), according to their cut-off wavelengths (cut) at ~560–580 nm, 510–540 nm, 440–450 nm, and with no significant absorbance throughout the spectrum, respectively (BOWMAKER et al. 1997 ). The principal member of the double cone contains a pale oil droplet with a cut at 440 nm, whereas the accessory member of the double cone rarely contains an oil droplet (BOWMAKER et al. 1997 ). Applying microspectrophotometry (MSP), BOWMAKER et al. 1997 have identified the rod pigments with a max at 506 nm and four different types of cone pigments with max at 567 nm (red), 507 nm (green), 453 nm (blue), and 409 nm (violet). Interestingly, there is a strong association between the types of visual pigments and those of photoreceptor cells. That is, the red pigments are found in R-type cones and in both members of double cones (BOWMAKER 1991 ; BOWMAKER et al. 1997 ), while the green, blue, and violet pigments are found only in the Y-, C-, and T-type cones, respectively (BOWMAKER et al. 1997 ).

UV sensitivity of the pigeon at ~325–385 nm has been detected by both behavioral experiments (BLOUGH 1957 ; WRIGHT 1972 ; KREITHEN and EISNER 1978 ; EMMERTON and DELIUS 1980 ) and electroretinogram (ERG) experiments (CHEN et al. 1984 ; CHEN and GOLDSMITH 1986 ; VOS HZN et al. 1994 ). It is possible in principle to identify cones containing UV pigments in the retina by MSP, but such analyses have shown no evidence for the existence of the UV receptor (BOWMAKER 1977 ; BOWMAKER et al. 1997 ). However, since MSP is based on a random sampling of the photoreceptors in a given retina, the analysis can miss UV pigments entirely if they are present in small numbers (JACOBS 1981 ; BOWMAKER et al. 1997 ).

Another way of identifying visual pigments in the retina is to clone all opsin genes and reconstruct the corresponding visual pigments (e.g., see YOKOYAMA 1997 ). At present, no genetic information on the visual pigments in the pigeon retina is available. Here we report the isolation and molecular characterization of the five classes of retinal opsin genes from the pigeon. The corresponding opsin cDNAs were isolated from the retinal and pineal RNAs, expressed in cultured COS1 cells, and reconstituted with 11-cis-retinal, and the max values of the resulting visual pigments were determined. The results show that, in addition to the pineal gland-specific pigment (KAWAMURA and YOKOYAMA 1996A ), the pigeon has one type of rod pigment and four types of cone pigments. No evidence for the "true" UV opsin gene in the pigeon genome was found.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Background information:
Visual pigments in the retinas of vertebrates are classified into five major groups: (1) the RH1 cluster (mostly consisting of rod-specific pigments with max values of ~500 nm); (2) the RH2 cluster (a mixture of pigments with max values of 470–510 nm); (3) the SWS1 cluster (consisting of short wavelength-sensitive pigments with max values of 360–420 nm); (4) the SWS2 cluster (consisting of SWS pigments with max values of 440–455 nm); and (5) the LWS/MWS cluster (consisting of long wavelength-sensitive or middle wavelength-sensitive pigments with max values of 510–570 nm) (YOKOYAMA 1994 , YOKOYAMA 1995 , YOKOYAMA 1997 ; YOKOYAMA and YOKOYAMA 1996 ; see also OKANO et al. 1992 ; HISATOMI et al. 1996 ). The RH1, RH2, SWS1, SWS2, and LWS or MWS (LWS/MWS) opsins are encoded by rh1, rh2, sws1, sws2, and lws or mws (lws/mws) opsin genes, respectively.

Some vertebrates such as marine lamprey (YOKOYAMA and ZHANG 1997 ), lizards (KAWAMURA and YOKOYAMA 1997 ), and birds (OKANO et al. 1994 ; MAX et al. 1995 ; KAWAMURA and YOKOYAMA 1996A ) are also known to have an additional group of pineal gland-specific P pigments that are encoded by p opsin genes. The pigeon p_Cl gene of this type has been determined by KAWAMURA and YOKOYAMA 1996A . The recently found VA (vertebrate ancient) pigment of Atlantic salmon, Salmo salaar (SONI and FOSTER 1997 ), and parapinopsin of channel catfish, Ictalurus punctatus (BLACKSHAW and SNYDER 1997 ), also seem to belong to this P cluster (S. YOKOYAMA and R. YOKOYAMA, unpublished results).

Genomic library screening:
A EMBL3 genomic library was constructed using the high-molecular-weight DNA extracted from the blood of one pigeon (C. livia; KAWAMURA and YOKOYAMA 1996A ). The average size of the insert DNA in the genomic library was ~16 kb. Since the nuclear DNA content of the pigeon is ~50% of the human genome (MANFREDI ROMANINI 1973 ), it requires at least 4.6 x 10⁵ plaques of recombinant phages to clone a single-copy gene of ~16 kb in length (SAMBROOK et al. 1989 ). Thus, we screened a total of 6.3 x 10⁵ recombinant plaques. As hybridization probes, we used bovine RH1 opsin cDNA (NATHANS and HOGNESS 1983 ), human LWS and SWS1 opsin cDNAs (NATHANS et al. 1986A ), and a mixture of the exons from sws2 genes of Mexican cavefish, Astyanax fasciatus (YOKOYAMA and YOKOYAMA 1993 ), and American chameleon, Anolis carolinensis (KAWAMURA and YOKOYAMA 1996B ). Probe labeling and plaque hybridization were performed as described previously (KAWAMURA and YOKOYAMA 1993 ). Hybridized membranes were washed four times (30 min each) in 1x SSC (0.15 M NaCl/0.015 M Na₃, citrate)/0.1% SDS at 55°, which allows ~30% mismatch (SAMBROOK et al. 1989 ). The four probes were used sequentially by recycling the membranes. Old probes were removed from the membranes by washing them in 0.4 M NaOH at 45° for 30 min and then in 0.1x SSC/0.1% SDS/0.2 M Tris (pH 7.5) at 45° for 30 min.

A total of 62 phage clones were isolated and subjected to restriction enzyme mapping. Among these, clones CL89 (representing rh1_Cl ), CL5 (rh2_Cl), CL37 (sws1_Cl-S), CL36 (sws1_ClL), CL25 (sws2_Cl-S), CL34 (sws2_Cl-L), and CL102 (lws_Cl) contained entire coding regions of the opsin genes and were subcloned into pBluescript SK(-) vectors. The nucleotide sequences of these clones were determined in both strands by standard dideoxynucleotide-chain-termination method (SAMBROOK et al. 1989 ). The remaining 55 clones overlap with one of these 7 clones.

Southern hybridization:
About 5 µg each of pigeon genomic DNA was digested with restriction enzymes, separated by size on a 0.5% agarose gel, and transferred onto a Hybond-N⁺ nylon membrane (Amersham, Piscataway, NJ). Using the random priming method, exon 4 of bovine rh1, those of rh2, sws1, and sws2 of American chameleon, and a portion of exon 5 of lws of American chameleon were labeled with [-³²P]dATP. Different exons of sws1_Cl were also used as hybridization probe. Hybridization was carried out at 65°, using the commercial protocol for Hybrid-N⁺ membrane. The hybridization membrane was washed at 65° in 1x SSC/0.1% SDS.

cDNA cloning:
Using the acid-guanidinium extraction method (CHOMCZYNSKI and SACCHI 1987 ), ~50 and 4 µg of total RNAs were prepared from the retina and pineal gland of the second pigeon, respectively. RH1_Cl, RH2_Cl, SWS2_Cl, and LWS_Cl opsin cDNAs and P opsin cDNA were amplified from 200 ng of total retinal RNA and 70 ng total RNA from the pineal gland, respectively, by the reverse transcriptase-PCR amplification (RT-PCR) method.

Figure 1 shows RT-PCR primers used to amplify the five types of full-length opsin cDNAs. Forward and reverse primers contain EcoRI and SalI linkers, respectively, to permit cloning of the PCR products into the pMT5 expression vector (KAWAMURA and YOKOYAMA 1998 ; YOKOYAMA et al. 1998 ). To facilitate translation, each forward primer contains the Kozak consensus sequence, CCACC (KOZAK 1984 ), followed by the initiation codon. The SalI linkers in the reverse primers are followed by 21–24 nucleotides of the reverse complement sequence starting from the second nucleotide position of the last codon. The extra 6 nucleotides 5' to the linker sites are to facilitate restriction digestion and correspond to the genomic noncoding sequences in each respective location.

View larger version (38K): In this window In a new window Download PPT slide   Figure 1. Oligonucleotide primers for RT-PCR amplification of pigeon opsin mRNAs.

The first-strand cDNA synthesis was carried out at 42° for 1 hr in a total volume of 20 µl containing reaction buffer (10 mM Tris-HCl, pH 9.0, 1 mM MgCl₂, 50 mM KCl, 0.1% Triton X-100), 1 mM dNTPs, 5 µM reverse primers, 20 units of RNasin (Promega, Madison, WI), and 200 units of SuperScript II reverse transcriptase (Gibco BRL, Gaithersburg, MD). The resulting cDNA was combined with the same reaction buffer containing 200 mM dNTPs, 1 µM each forward and reverse primers, and 5 units of Taq polymerase (Promega) in a total volume of 100 µl. PCR amplification was performed by 30 cycles at 92° for 45 sec, 55° for 60 sec, and 72° for 90 sec. At each cycle, the duration of the extension reaction was progressively extended by 3 sec. After the final extension step at 72° for 10 min, the PCR products were resolved in 1.5% agarose gel electrophoresis. The opsin cDNA band of ~1.1 kb was extracted and cloned into the EcoRV-digested pBluescript plasmid vector with T-overhang attached to 3' ends (HADJEB and BERKOWITZ 1996 ). Nucleotide sequences of the entire region of the cDNA 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). With the exception of p_Cl, we selected clones that encode identical amino acid sequences to those of the corresponding sequences deduced from the genomic clones for spectral analyses of the visual pigments (see RESULTS).

Regeneration of visual pigments and spectral analysis:
The fragment of the EcoRI/SalI-digested pMT5 expression vector, ~5 kb in length, contains the sequences necessary for expression in cultured COS1 cells and the last 15 codons of the bovine rhodopsin, encoding Ser-Thr-Thr-Val-Ser-Lys-Thr-Glu-Thr-Ser-Gln-Val-Ser-Pro-Ala that are necessary for immunoaffinity purification (MOLDAY and MACKENZIE 1983 ). This EcoRI/SalI fragment was ligated with the EcoRI/SalI opsin cDNA fragments. The resulting plasmids were 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 pigments were purified by binding to the monoclonal anitbody 1D4 Sepharose in buffer W1 (50 mM HEPES, pH 6.6, 140 mM NaCl, 3 mM MgCl₂, 20% [w/v] glycerol, and 0.1% dodecyl maltoside; KAWAMURA and YOKOYAMA 1998 ; YOKOYAMA et al. 1998 ).

UV visible absorption spectra of visual pigments were recorded at 20°, using a Hitachi U-3000 dual beam spectrophotometer. Visual pigments were bleached by a 60-W room lamp with 440-nm cut-off filter. Recorded spectra were analyzed using SigmaPlot software (Jandel Scientific, San Rafael, CA).

Qualitative RT-PCR assay:
Total RNA was mixed with the PCR reaction mix (10 mM Tris-HCl, pH 9.0, 1.5 mM MgCl₂, 50 mM KCl, 0.1% Triton X-100, 200 µM dNTPs, and primers at 1 µM each), MMLV reverse transcriptase, and Taq polymerase in total volume of 25 µl. The primers given in Figure 1 were used for the amplification of rh2_Cl and p_Cl cDNAs. The primers used for others were: 5'-AGCCCTGGAAGTTCTCGGCT-3' (forward [F]: starting position at 128) and 5'-TTCATTGTTGATCTCCGGC-3' (reverse [R]; starting position at 633) for rh1_Cl; 5'-ACTTCCGCTTCAACTCCAAACACG-3' (F: position at 419) and 5'-GGCCGCCCGCACACCAG-3' (R: position at 959) for sws1_Cl; 5'-AGCCCCGGCGTGTTCCGC-3' (F: position at 121) and 5'-GAGGGCCAGGGGGACCCC-3' (R: position at 682) for sws2_Cl; and 5'-GTGGTGGTGGCGTCGGTGTT-3' (F: position at 170) and 5'-TGGCCAGCCAGACTTGCAG-3' (R: position at 720) for lws_Cl. The samples were placed in a thermal cycler at 50° for 8 min, followed by 35 cycles of 92° for 45 sec, 55° for 60 sec, and 72° for 90 sec. A total of 5 µl each of PCR products was electrophoresed on 2.5% agarose gel. PCR was also carried out without reverse transcriptase for each opsin gene, resulting in no amplification in both tisses.

Phylogenetic analysis:
The six types of visual pigments of the pigeon were compared to those of chicken, Gallus gallus (RH1 pigment, GenBank accession no. D00702; RH2 pigment, M92038; SWS1 pigment, M92039; SWS2 pigment, M92037; LWS pigment, M62903; and P pigment, U15762) and American chameleon (RH1 pigment, L31503; RH2 pigment, AF134189, AF134190, AF134191; SWS1 pigment, AF134192, AF134193, AF134194; SWS2 pigment, AF133907; LWS pigment, U08131; and P pigment, AF134767, AF134768, AF134769, AF134770, AF134771). To construct a rooted phylogenetic tree of these pigments, we used four different pigments of Drosophila melanogaster (Rh1 pigment, K02315; Rh2 pigment, M12896; Rh3 pigment, M17718; and Rh4 pigment, X65880) as the outgroup. The deduced amino acid sequences were aligned by CLUSTAL W program (THOMPSON et al. 1994 ) and adjusted visually. The number (K) of amino acid substitutions per site for two sequences was estimated from K = -ln(1 - p), where p is the proportion of different amino acids for a pair of sequences. The phylogenetic tree was reconstructed by applying the NJ method to the K values (SAITOU and NEI 1987 ). The reliability of the phylogenetic tree was evaluated by the bootstrap analysis with 1000 replications (FELSENSTEIN 1985 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Retinal opsin genes and visual pigments:
We cloned all five major groups of the retinal opsin genes rh1_Cl, rh2_Cl, sws1_Cl, sws2_Cl, and lws_Cl from the pigeon, where both sws1_Cl and sws2_Cl have two forms with different intron sizes (Figure 2). rh1_Cl, rh2_Cl, sws1_Cl, and sws2_Cl contain five putative exons and four introns, whereas lws_Cl contains one extra exon. The introns 1, 2, 3, and 4 of rh1_Cl, rh2_Cl, sws1_Cl, and sws2_Cl and the corresponding introns 2, 3, 4, and 5 of lws_Cl interrupt the coding sequence at exactly the same positions. These exon-intron structures have been well conserved among the retinal opsin genes in vertebrates (YOKOYAMA and YOKOYAMA 1996 ). Splice junction signals (GT/AG) are conserved in all introns and there are no nonsense mutations in the coding regions.

View larger version (21K): In this window In a new window Download PPT slide   Figure 2. The genomic structures of the retinal opsin genes of the pigeon. The coding regions are indicated by solid boxes. Introns 1, 2, 3, and 4 of rh1Cl consist of 856 bp, 99 bp, 241 bp, and 0.6 kb, respectively. Introns 1, 2, 3, and 4 of rh2Cl consist of 1.7 kb, 457 bp, 91 bp, and 173 bp, respectively. Introns 1, 2, 3, and 4 of sws1Cl-S consist of 3 kb, 420 bp, 1.4 kb, and 1.5 kb, respectively, and the corresponding introns of sws1Cl-L consist of 3.3, 0.5, 1.5, and 1.8 kb. Introns 1, 2, 3, and 4 of sws2Cl-S consist of 1.6, 1.3, 3, and 2.4 kb, respectively, and those of sws2Cl-L are given by 2, 1.3, 2.2, and 3 kb, respectively. Introns 1, 2, 3, 4, and 5 of lwsCl consist of 0.8, 2.2, 0.7, 2, and 1.5 kb, respectively. Circled nucleotides indicate polymorphisms between sws1Cl-S and sws1Cl-L and between sws2Cl-S and sws2Cl-L. B, BamHI; E, EcoRI; H, HindIII; K, KpnI; Sl, SalI; Ss, SstI. The sequences reported in this article have been deposited in the GenBank database (AF149230–AF149231 for rh1Cl; AF149232–AF149233 for rh2Cl; AF149234–AF149237 for sws1Cl; AF149238–AF149242 for sws2Cl; AF149243–AF149248 for lwsCl).

By comparing the five opsin genes of the pigeon to those of chicken and American chameleon, we evaluated the proportions of identical nucleotides per site (p_{I, nuc}) for all pairs. The amino acid sequences of RH1_Cl, RH2_Cl, SWS1_Cl, SWS2_Cl, LWS_Cl, and P_Cl pigments deduced from the nucleotide sequences of rh1_Cl, rh2_Cl, sws1_Cl (swsCl-S and sws1Cl-L), sws2_Cl (represented by sws2_Cl-S), lws_Cl, and p_Cl are shown in Figure 3, where amino acid residue numbers correspond to those of RH1_Cl pigment. By comparing these amino acids to those of the chicken and American chameleon, the proportions of identical amino acids per site (p_{I, aa}) were also evaluated. The comparisons of the p_{I, nuc} and p_{I, aa} show that RH1_Cl, RH2_Cl, SWS1_Cl, SWS2_Cl, and LWS_Cl pigments belong to RH1, RH2, SWS1, SWS2, and LWS/MWS groups, respectively (see MATERIALS AND METHODS). That is, the p_{I, nuc} values between orthologous opsin genes range from 0.74 to 0.92, whereas those between paralogous genes range from 0.51 to 0.75. Similarly, the p_{I, aa} values between orthologous and paralogous pigments range from 0.82 to 0.98 and from 0.41 to 0.73, respectively.

View larger version (73K): In this window In a new window Download PPT slide   Figure 3. Alignment of the deduced amino acid sequences of the retinal and nonretinal opsins of the pigeon. Gaps necessary to optimize the alignment are indicated by dashes. Seven transmembrane regions (HARGRAVE et al. 1983 ) are indicated by horizontal lines above the sequences. RH1, RH2, SWS1, SWS2, LWS, and P denote RH1Cl, RH2Cl, SWS1Cl, SWS2Cl, and LWSCl pigments, respectively. The residue number is according to that of RH1Cl.

From Figure 3, we can also identify functionally important residues that are conserved among these pigments. They include Lys296 for Schiff base linkage to the chromophore (WANG et al. 1980 ), Glu113 for Schiff base counterion (SAKMAR et al. 1989 ; ZHUKOVSKY and OPRIAN 1989 ; NATHANS 1990 ), Cys110 and Cys187 for disulfide bond (KARNIK et al. 1988 ), and multiple Ser and Thr in the C-terminal region for the targets of opsin kinase (OHGURO et al. 1994 ). RH1_Cl and RH2_Cl pigments have Asn2 and Asn15 for N-glycosylation (OKANO et al. 1992 ) and Cys322 and Cys323 for palmitoylation sites (OVCHINNIKOV et al. 1988 ) as those in the orthologous pigments of other vertebrates. Furthermore, Ser164, His181, Tyr261, Thr269, and Ala292 of the LWS_Cl pigment exhibit the LWS-specific character that is important for a red-light detection (YOKOYAMA and RADLWIMMER 1998 , YOKOYAMA and RADLWIMMER 1999 ). Note that sites 164, 181, 261, 269, and 292 correspond to 180, 197, 277, 285, and 308 in human red and green pigments.

rh1_Cl, represented by Cl89, spans 2.9 kb from the start to stop codons (Figure 2). When 1056 nucleotide sites of the entire coding region of rh1_Cl, including the stop codon, are compared to those of the orthologous chicken and American chameleon genes, the p_{I, nuc} values are given by 0.92 and 0.78, respectively. The p_{I, nuc} value for rh1 genes between chicken and American chameleon is 0.78 and is identical to the corresponding value between pigeon and American chameleon. The p_{I, aa} values for RH1 pigments between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.98, 0.85, and 0.86, respectively. These p_{I, nuc} and p_{I, aa} values between the orthologous molecules of the two bird species are higher than the corresponding values between the bird and reptile, reflecting the phylogenetic relationships of the three species.

The length of rh2_Cl, represented by Cl5, is ~3.5 kb from the start to stop codons (Figure 2). The p_{I, nuc} values for rh2 genes between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.91, 0.83, 0.84, respectively. The p_{I, aa} values for RH2 pigments between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.98, 0.92, and 0.92, respectively. Again, the two comparisons reflect the phylogenetic relationships of the three species well.

sws1_Cl-L, ~8 kb in length, represented by Cl36, is ~0.8 kb longer than sws1_Cl-S, represented by Cl37 (Figure 2). When 1044 nucleotide sites of the entire coding region of these two genes are compared, there is only one silent nucleotide difference, C in sws1_Cl-S and T in sws1_Cl-L, at nucleotide position 519 in exon 3 (Figure 2). When 898 bp in the noncoding regions of the two opsin genes are compared, there are only 5 different nucleotides (0.56% difference). This small difference and the Southern analysis (see DISCUSSION) show that sws1_Cl-S and sws1_Cl-L are different alleles rather than two genes at different loci. The p_{I, nuc} for sws1 genes between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.84, 0.87, and 0.83, respectively, while the corresponding p_{I, aa} values are given by 0.87, 0.87, and 0.84. These p_{I, nuc} and p_{I, aa} values are about the same for all pairwise combinations of the pigeon, chicken, and American chameleon pigments and, surprisingly, do not reflect the phylogenetic relationships of the three species. As we see later, this seems to be caused mainly by a slow evolution of sws1 gene in the common ancestor of the two bird species.

Cl25 and Cl34 represent two types of sws2_Cl genes, sws2_Cl-S and sws2_Cl-L, respectively. sws2_Cl-S and sws2_Cl-L span 9.4 and 9.6 kb from the start to stop codons, respectively (Figure 2). Along the 1098 bp sites of the coding regions, the two genes differ at one nucleotide site at nucleotide position 499 in exon 2, causing one amino acid difference (Ala167 and Thr167 for SWS2_Cl-S and SWS2_Cl-L pigments, respectively; Figure 2). When 688 bp sites in the noncoding regions of the two genes are compared, there are 21 different nucleotides (3.1% difference). Interestingly, both sws2_Cl-S and sws2_Cl-L are physically linked to lws_Cl (Figure 4). The distance between the stop codon for sws2_Cl-L and the initiation codon of lws_Cl is ~7 kb, while that between sws2_Cl-S and lws_Cl is ~5 kb (Figure 4). Thus, like sws1_Cl-S and sws1_Cl-L, it is most likely that sws2_Cl-S and sws2_Cl-L also represent two alleles rather than two distinct genes. The p_{I, nuc} values of sws2 genes between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.84, 0.74, and 0.74, respectively, while the corresponding p_{I, aa} values are 0.87, 0.84, and 0.82. Thus, the phylogenetic relationships of the three species are reflected in the p_{I, nuc} and p_{I, aa} values.

View larger version (14K): In this window In a new window Download PPT slide   Figure 4. Genomic structure of the sws2Cl and lwsCl. The coding exons are indicated by solid boxes. B, BamHI; K, KpnI; Sl, SalI; Ss, SstI.

lws_Cl, represented by Cl102, has six exons and five introns spanning 8.3 kb from the start to stop codons (Figure 2). The p_{I, nuc} values for lws genes between pigeon and chicken, between pigeon and American chameleon, and between chicken and American chameleon are given by 0.91, 0.78, and 0.76, respectively, while the corresponding p_{I, aa} values for LWS opsins are given by 0.96, 0.90, and 0.91. Like RH1, RH2, and SWS2 pigments, these values also reflect the phylogenetic relationships of the three species reasonably well.

Evolution of the pigeon pigments:
All RH1, RH2, SWS1, SWS2, LWS, and P pigments have been characterized in chicken (OKANO et al. 1992 , OKANO et al. 1994 ) and in American chameleon (see MATERIALS AND METHODS). The phylogenetic relationships of the five different types of retinal pigments and P pigments from pigeon, chicken, and American chameleon are given by (((((RH1, RH2) SWS2) SWS1) P) LWS) (Figure 5). However, since the bootstrap support for the cluster of RH1, RH2, and SWS2 pigments is only 72%, the phylogenetic position of SWS2 pigments is not as clear-cut as the tree topology in Figure 5 may indicate. Similarly, the bootstrap value for the cluster of RH1, RH2, SWS2, SWS1, and P pigments is 80% and the phylogenetic position of P pigments in Figure 5 is also not reliable. Thus, the six groups of pigments may be distinguished roughly into three groups: (i) RH1, RH2, SWS1, and SWS2 clusters; (ii) P cluster; and (iii) LWS/MWS cluster.

View larger version (30K): In this window In a new window Download PPT slide   Figure 5. The phylogenetic tree for retinal (RH1, RH2, SWS1, SWS2, and LWS) and pineal gland-specific (P) pigments of pigeon, chicken, and American chameleon. The bootstrap supports are indicated next to branch nodes. Values after P indicate max values.

With the exception of SWS1 pigments, the bootstrap supports for the sets of the paired orthologous bird pigments are 100% and are highly reliable. Figure 5 shows that the uncertainty of the pigeon and chicken pigment cluster is due to a very slow rate of amino acid replacement in the common bird ancestor (see also DISCUSSION).

Light absorption profiles:
For spectral analyses, with the exception of p_Cl, we used cDNAs that encode identical amino acids to those of the corresponding pigments deduced from genomic clones. At codon position 77, P_Cl opsin cDNA clone contains one nonsynonymous nucleotide difference from its genomic DNA sequence that encodes Val instead of Met (Figure 3). Since the P pigments of American chameleon and chicken both have Val at the corresponding sites, the difference between the cDNA sequence from one pigeon and the corresponding genomic DNA sequence obtained from another is interpreted as a naturally occurring DNA polymorphism rather than a cloning artifact.

The max values of pigments can be measured directly from the dark spectra (Figure 6) and from the dark-light difference (Figure 6, insets). The former measurements for RH1_Cl, RH2_Cl, SWS2_Cl-S, LWS_Cl, and P_Cl pigments are given by 502 ± 3, 503 ± 2, 448 ± 2, 559 ± 2, and 481 ± 2 nm, respectively. The respective max values estimated from the dark-light difference are given by 505, 506, 459, 559, and 486 nm. With the exception of SWS2_Cl-S, the corresponding two values for each pigment are close. Using the same method, we previously obtained the max value of SWS1_Cl pigment, 393 ± 2 nm (YOKOYAMA et al. 1998 ). Out of these five types of retinal pigments, the max values of RH1_Cl, RH2_Cl, SWS2_Cl, and LWS_Cl pigments generally agree with the corresponding values estimated either by MSP or by ERG (Table 1). The max value of 393 nm for SWS1_Cl-S pigment is close to the MSP estimate of 409 nm (BOWMAKER et al. 1997 ) and the max values of about 400 nm observed by REMY and EMMERTON 1989 , GRAF and VAN NORREN 1974 , VAN NORREN 1975 , and WORTEL et al. 1984 . However, it differs considerably from the corresponding max values of 370 nm estimated by ERG (Table 1). The cause of this discrepancy remains to be elucidated (see also DISCUSSION).

View larger version (29K): In this window In a new window Download PPT slide   Figure 6. Absorption spectra of regenerated pigeon pigments measured in the dark and the dark-light difference spectra (insets).

View this table: In this window In a new window   Table 1. max values (nm) of the pigeon visual pigments

Expression of retinal and P opsins:
Using opsin gene-specific primers, we have examined the expressions of rh1_Cl, rh2_Cl, sws1_Cl, sws2_Cl, lws_Cl, and p_Cl in the retina and pineal gland of the pigeon by RT-PCR assay. In the retina, all five visual opsin genes are expressed, but p_Cl is not, whereas in the pineal gland, only p_Cl is expressed (Figure 7). Similar analyses in the chicken show that the five visual opsins and P opsin are expressed in the retina and in the pineal gland, respectively (OKANO et al. 1994 ; MAX et al. 1995 ). One study reports a low level of the lws gene expression in the pineal gland (OKANO et al. 1994 , OKANO et al. 1997 ). However, this cannot be confirmed in another study (MAX et al. 1995 ). Thus, the expression of the lws opsin in the pineal gland of the chicken is controversial. To settle the issue in the chicken, additional analyses of the lws gene expression are needed.

View larger version (80K): In this window In a new window Download PPT slide   Figure 7. RT-PCR assay for the visual and P opsin gene expression in the retina and pineal gland of the pigeon. The 100-bp DNA ladder marker is shown at the left margin.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Opsin genes and visual pigments of the pigeon:
During the molecular characterization of the pigeon opsin genes, we found physical linkage between sws2_Cl and lws_Cl. Previously, we reported a similar linkage relationship between sws2 and lws genes in Mexican cavefish, Astyanax fasciatus (YOKOYAMA and YOKOYAMA 1993 ). The same linkage relationships in fish and bird species suggest that the structure of 5'-sws2-lws-3' was established in the vertebrate ancestor, some 450 mya. So far, neither sws2 gene nor rh2 gene has been isolated from any mammals and they appear to have been nonfunctionalized in the early stage of mammalian evolution. Incidentally, in human, sws1 gene is located on chromosome 7 (NATHANS et al. 1986B ), whereas lws and relatively recently duplicated mws gene(s) are located tandemly on the X chromosome (NATHANS et al. 1986A , NATHANS et al. 1986B ).

We have seen that the genetic distances of SWS1_Cl pigments of the pigeon, chicken, and American chameleon do not necessarily reflect the phylogenetic relationships of the three species. To evaluate the evolutionary patterns of amino acid replacements in the pigments in more detail, we estimated the numbers of amino acid replacements per site (K) for the pigments of pigeon and chicken separately (Table 2). The results show that the K values for the orthologous retinal pigments of the two bird species are similar to each other. However, the branch lengths between the two P pigments are significantly different. In the pigeon, SWS1 pigment has the highest K value, followed by P, SWS2, LWS, RH1, and RH2 pigments, in that order. In the chicken, the K values for P, SWS1, and SWS2 pigments are about one order of magnitude higher than those for RH1 and RH2 pigments. These trends can also be seen in the phylogenetic tree in Figure 5. A close inspection of Figure 5 reveals that the common ancestor of the bird species experienced an unusually small number of amino acid replacements in its SWS1 pigments, followed by equally accelerated amino acid replacements in the pigeon and chicken pigments.

The "near-UV" pigment of the pigeon:
We have determined the max values of RH1_Cl, RH2_Cl, SWS2_Cl, LWS_Cl, and P_Cl pigments. This functional assay has shown no evidence for the existence of a true UV opsin gene in the pigeon genome. The only pigment that absorbs light at around the UV to violet range is a near-UV SWS1_Cl pigment with a max at 393 nm (YOKOYAMA et al. 1998 ). Using MSP, BOWMAKER et al. 1997 also identified only the "violet" pigment that absorbs light at 409 nm. Thus, although there is a small difference in the two max values, the near-UV and violet pigments appear to be the same pigment. As already noted, pigment regeneration is the most direct method for evaluating absorption spectrum of the visual pigment. Pigments were synthesized abundantly in vitro and highly purified to eliminate background absorbance of other cellular materials. Consequently, the method provides very high signal/noise ratio and small standard errors. Thus, we may conclude that the near-UV or violet pigment has a max at 393 nm. It should be noted that, being expressed in the T-type cones (see Introduction), the max values of SWS1_Cl pigments are not modified further by the oil droplet.

BOWMAKER et al. 1997 suggested that the true UV opsin gene in the pigeon might have arisen from sws1_Cl by gene duplication. This is reasonable because all known UV and violet opsin genes in vertebrates belong to the SWS1 cluster (YOKOYAMA 1997 ). To test the possible existence of such duplicated sws1 genes in the pigeon genome, we have isolated genomic DNA from the second pigeon (see MATERIALS AND METHODS) and conducted Southern hybridization analysis. When the exons 3 and 4 of sws1_Cl-S are used as the probe, the double-digested DNAs with BamHI/SstI (Figure 8, lane A) and BamHI/KpnI (lane B) show single hybridizing bands of 7.4 and 8.6 kb in size, respectively. They correspond to sws1_Cl-S (Figure 2), indicating that this particular pigeon is homozygous for the sws1_Cl-S allele. When exons 1–3 of the same gene were used as the probe, the identical hybridizing bands were detected (results not shown). These results strongly suggest that there is not an additional sws1_Cl gene in the pigeon genome and that SWS1_Cl pigment is the sole pigment that detects light below 400 nm. The existence of a sws1_Cl-S/sws1_Cl-S homozygous individual also demonstrates that sws1_Cl-S and sws1_Cl-L are two alleles rather than two genes at separate loci. We have also tested the possible existence of duplicated loci of rh1_Cl, rh2_Cl, sws1_Cl, sws2_Cl, and lws_Cl in the pigeon genome, but hybridizing bands are consistent with the restriction maps of the genomic clones in Figure 2 (results not shown). It should also be noted that no extra genes were included in the 62 clones isolated (MATERIALS AND METHODS). All these results strongly suggest that the pigeon contains only five retinal opsin genes rh1_Cl, rh2_Cl, sws1_Cl, sws2_Cl, and lws_Cl in its genome. Thus, pigeon, chicken, and American chameleon seem to have five different types of retinal visual pigments (Figure 5).

View larger version (30K): In this window In a new window Download PPT slide   Figure 8. Southern hybridization of BamHI/SstI- (lane A) and BamHI/KpnI- (B) digested genomic DNA of the pigeon with exons 3 and 4 of sws1Cl-S. The 1-kb DNA ladder marker is shown at the left margin. The hybridizing bands are indicated by the arrowheads.

UV vision of the pigeon:
As already noted, both behavioral experiments and ERG analyses demonstrated the pigeon's ability to detect UV (see Introduction). However, there exist some conflicting data on the absorption spectra of the pigeon in the range of UV–violet. For example, from behavioral experiments, REMY and EMMERTON 1989 and ROMESKII and YAGER 1976A , ROMESKII and YAGER 1976B detected only one absorption maximum at 400–415 nm. Some ERG studies also show single max values at 400–413 nm (GRAF and VAN NORREN 1974 ; VAN NORREN 1975 ; GOVARDOVSKII and ZUEVA 1977 ; WORTEL et al. 1984 ). On the other hand, using behavioral experiments, EMMERTON and REMY 1983 detected double peaks of spectral sensitivities at 360 and 400–420 nm. Using ERG, VOS HZN et al. 1994 also proposed that pigeon uses two pigments with max at 366 and 415 nm (Table 1).

These conflicting results may mean that pigeon's UV sensitivity varies from one individual to another. If this turns out to be the case, then the structural difference of the pigeon's retinas in different individuals may become an important factor. The dorsal red field of the pigeon retina contains mostly single cones and encompasses much of its binocular field of view, while the remaining ventral yellow field contains a higher proportion of rods and double cones (JACOBS 1981 ). The cut values of the oil droplets in the yellow field are often lower than the corresponding droplets in the red field (BOWMAKER 1977 ; MARTIN and MUNTZ 1978 ; WORTEL et al. 1984 ; BOWMAKER et al. 1997 ). The sensitivities of the red and yellow fields of the pigeon's retina in different individuals differ for both "visible" and UV light (REMY and EMMERTON 1989 ). Due to this variation, different numbers of UV sensitive receptors in the two fields may result in different max values among different individuals (REMY and EMMERTON 1989 ).

Where do these UV-sensitive pigments come from? Humans are normally blind to UV light because it is strongly absorbed by the yellow-pigmented lens (SAID and WEALE 1959 ). However, if the lens is surgically removed, then we can detect UV light (WALD 1945 ). The UV vision in humans in this unusual circumstance must be mediated by SWS1 pigments that absorb wavelengths of 370–530 nm with max values at 420 nm (BOYNTON 1979 ). The pigeon's cornea, lens, and vitrous body transmit both visible and UV light that can reach and excite the retina (EMMERTON et al. 1980 ). Thus, it is most likely that the pigeon can detect UV light using SWS1_Cl pigments whose max values are much lower than those of the orthologous pigments in humans.

	ACKNOWLEDGMENTS

Comments by Ruth Yokoyama and anonymous reviewers were greatly appreciated. This work was supported by National Institutes of Health grant GM-42379.

Manuscript received May 25, 1999; Accepted for publication August 6, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BLACKSHAW, B. and S. H. SNYDER, 1997  Parapinopsin, a novel catfish opsin localized to the parapineal organ, defines a new gene family. J. Neurosci. 17:8083-8092[Abstract/Free Full Text].

BLOUGH, D. S., 1957  Spectral sensitivity in the pigeon. J. Opt. Soc. Am. 47:827-833[Medline].

BOWMAKER, J. K., 1977  The visual pigments, oil droplets and spectral sensitivity of the pigeon (Columba livia). Vision Res. 17:1129-1138[Medline].

BOWMAKER, J. K., 1991 The evolution of vertebrate visual pigments and photoreceptors, pp. 63–81 in Evolution of the Eye and Visual System, edited by J. R. CRONLY-DILLON and R. L. GREGORY. CRC Press, Boca Raton, FL.

BOWMAKER, J. K., L. A. HEATH, S. E. WILKIE, and D. M. HUNT, 1997  Visual pigments and oil droplets from six classes of photoreceptor in the retinas of birds. Vision Res. 37:2183-2194[Medline].

BOYNTON, R. M., 1979 Human Color Vision. Holt, Rinehart & Winston, New York.

CHEN, D.-M. and T. H. GOLDSMITH, 1986  Four spectral classes of cones in the retinas of birds. J. Comp. Physiol. A 159:473-479[Medline].

CHEN, D.-M., J. S. COLLINS, and T. H. GOLDSMITH, 1984  The ultraviolet receptor of bird retinas. Science 225:337-340[Abstract/Free Full Text].

CHOMCZYNSKI, P. and N. SACCHI, 1987  Single-step method of RNA isolation by acid guanidium thiocyanate phenol-chloroform extraction. Anal. Biochem. 162:156-159[Medline].

EMMERTON, J. and J. DELIUS, 1980  Wavelength discrimination in the "visible" and ultraviolet spectrum by pigeons. Comp. Physiol. 141:47-52.

EMMERTON, J. and M. REMY, 1983  The pigeon's sensitivity to ultraviolet and "visible" light. Experientia 39:1161-1163.

EMMERTON, J., J. SCHWEMER, I. MUTH, and P. SCHLECHT, 1980  Spectral transmission of the ocular media of the pigeon (Columba livia). Invest. Ophthalmol. Vis. Sci. 19:1382-1387[Abstract/Free Full Text].

FELSENSTEIN, J., 1985  Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:597-600.

GOVARDOVSKII, V. I. and L. V. ZUEVA, 1977  Visual pigments of chicken and pigeon. Vision Res. 17:537-543[Medline].

GRAF, V. A. and D. VAN NORREN, 1974  A blue sensitive mechanism in the pigeon retina: max 400 nm. Vision Res. 14:1203-1209[Medline].

HADJEB, N. and G. A. BERKOWITZ, 1996  Preparation of T-overhang vectors with high PCR product cloning efficiency. BioTechniques 20:20-22[Medline].

HARGRAVE, P. A., J. H. MCDOWELL, D. R. CURTIS, J. K. WANG, and E. JUSZCZAK et al., 1983  The structure of bovine rhodopsin. Biophys. Struct. Mech. 9:235-244[Medline].

HISATOMI, O., T. SATOH, L. K. BARTHEL, D. L. STENKAMP, and P. A. RAYMOND et al., 1996  Molecular cloning and characterization of the putative ultraviolet-sensitive visual pigment of goldfish. Vision Res. 36:933-939[Medline].

JACOBS, G. H., 1981 Comparative Color Vision. Academic Press, New York.

KARNIK, S. S., T. P. SAKMAR, H.-B. CHEN, and H. G. KHORANA, 1988  Cysteine residues 110 and 187 are essential for the formation of correct structure in bovine rhodopsin. Proc. Natl. Acad. Sci. USA 85:8459-8463[Abstract/Free Full Text].

KAWAMURA, S. and S. YOKOYAMA, 1993  Molecular characterization of the red visual pigment gene of the American chameleon (Anolis carolinensis). FEBS Lett. 323:247-251[Medline].

KAWAMURA, S. and S. YOKOYAMA, 1996a  Molecular characterization of the pigeon P-opsin gene. Gene 182:213-214[Medline].

KAWAMURA, S. and S. YOKOYAMA, 1996b  Phylogenetic relationships among short wavelength-sensitive opsins of American chameleon (Anolis carolinensis) and other vertebrates. Vision Res. 36:2797-2804[Medline].

KAWAMURA, S. and S. YOKOYAMA, 1997  Expression of visual and nonvisual opsins in American chameleon. Vision Res. 37:1867-1871[Medline].

KAWAMURA, S. and S. YOKOYAMA, 1998  Functional characterization of visual and nonvisual pigments of American chameleon (Anolis carolinensis). Vision Res. 38:37-44[Medline].

KOZAK, M., 1984  Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Nucleic Acids Res. 12:857-872[Abstract/Free Full Text].

KREITHEN, M. L. and T. EISNER, 1978  Ultraviolet light detection by the homing pigeon. Nature 272:347-348[Medline].

MANFREDI ROMANINI, M. G., 1973 The DNA nuclear content and the evolution of vertebrates, pp. 39–81 in Cytotaxonomy and Vertebrate Evolution, edited by A. B. CHIARELLI and E. Capanna. Academic Press, London.

MARTIN, G. R. and W. R. A. MUNTZ, 1978  Spectral sensitivity of the red and yellow oil droplet fields of the pigeon (Columba livia). Nature 274:620-621[Medline].

MAX, M., P. J. MCKINNON, K. J. SEIDENMAN, R. K. BARRETT, and M. APPLEBURY et al., 1995  Pineal opsin: a nonvisual opsin expressed in chick pineal. Science 267:1502-1506[Abstract/Free Full Text].

MOLDAY, R. S. and D. MACKENZIE, 1983  Monoclonal antibodies to rhodopsin: characterization, cross-reactivity, and application as structural probes. Biochemistry 22:653-660[Medline].

NATHANS, J., 1990  Determinants of visual pigment absorbance: identification of the retinylidene Schiff's base counter ion in bovine rhodopsin. Biochemistry 29:9746-9752[Medline].

NATHANS, J. and D. S. HOGNESS, 1983  Isolation, sequence analysis, and intron-exon arrangement of the gene encoding bovine rhodopsin. Cell 34:807-814[Medline].

NATHANS, J., D. THOMAS, and D. S. HOGNESS, 1986a  Molecular genetics of human color vision: the genes encoding blue, green, and red pigments. Science 232:193-202[Abstract/Free Full Text].

NATHANS, J., T. P. PIANTANIDA, R. L. EDDY, T. B. SHOWS, and D. S. HOGNESS, 1986b  Molecular genetics of inherited variation in human color vision. Science 232:203-210[Abstract/Free Full Text].

OHGURO, H., R. S. JOHNSON, L. H. ERICSSON, K. A. WALSH, and K. PALCZEWSKI, 1994  Control of rhodopsin multiple phosphorylation. Biochemistry 33:1023-1028[Medline].

OKANO, T., D. KOJIMA, Y. FUKADA, Y. SHICHIDA, and T. YOSHIZAWA, 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[Abstract/Free Full Text].

OKANO, T., T. YOSHIZAWA, and Y. FUKADA, 1994  Pinopsin is a chicken pineal photoreceptive molecule. Nature 372:94-97[Medline].

OKANO, T., Y. TAKANAKA, A. NAKAMURA, K. HIRUNAGI, and A. ADACHI et al., 1997  Immunocytochemical identification of pinopsin in pineal glands of chicken and pigeon. Mol. Brain Res. 50:190-196[Medline].

OVCHINNIKOV, Y. A., N. G. ABDULAEV, and A. S. BOGACHUK, 1988  Two adjacent cysteine residues in the C-terminal cytoplasmic fragment of bovine rhodopsin are palmitoylated. FEBS Lett. 230:1-5[Medline].

REMY, M. and J. EMMERTON, 1989  Behavioral spectral sensitivities of different retinal areas in pigeons. Behav. Neurosci. 103:170-177[Medline].

ROMESKII, M. and D. YAGER, 1976a  Psychological studies of pigeon color vision. I. Photopic spectral sensitivity. Vision Res. 16:501-505[Medline].

ROMESKII, M. and D. YAGER, 1976b  Psychological studies of pigeon color vision. II. The spectral photochromatic interval function. Vision Res. 16:507-512[Medline].

SAID, F. S. and R. A. WEALE, 1959  The variation with age of the spectral transmissibility of the living human crystalline lens. Gerontologica 3:213-223.

SAITOU, N. and M. NEI, 1987  The neighbor-joining method: a new method for reconstructing phylogenetic tree. Mol. Biol. Evol. 4:406-425[Abstract].

SAKMAR, T. P., R. R. FRANKE, and H. G. KHORANA, 1989  Glutamic acid-113 serves as the retinylidene Schiff base counterion in bovine rhodopsin. Proc. Natl. Acad. Sci. USA 86:8309-8313[Abstract/Free Full Text].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SONI, B. G. and R. G. FOSTER, 1997  A novel and ancient vertebrate opsin. FEBS Lett. 406:279-283[Medline].

THOMPSON, J. D., D. G. HIGGINS, and T. J. GIBSON, 1994  CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680[Abstract/Free Full Text].

VAN NORREN, D. V., 1975  Two short wavelength sensitive cone systems in pigeon, chicken and daw. Vision Res. 15:1164-1166[Medline].

VOS HZN, J. J., M. A. J. M. COEMANS, and J. F. W. NUBOER, 1994  The photopic sensitivity of the yellow field of the pigeon's retina to ultraviolet light. Vision Res. 34:1419-1425[Medline].

WALD, G., 1945  Human vision and the spectrum. Science 101:653-658[Free Full Text].

WALLS, G. L., 1942 The Vertebrate Eye and Its Adaptive Radiation. Hafner, New York.

WANG, J. K., J. H. MCDOWELL, and P. A. HARGRAVE, 1980  Site of attachment of 11-cis-retinal in bovine rhodopsin. Biochemistry 19:5111-5117[Medline].

WORTEL, J. F., R. J. WUBBELS, and J. F. W. NUBOER, 1984  Photopic spectral sensitivities of the red and the yellow field of the pigeon retina. Vision Res. 24:1107-1113[Medline].

WRIGHT, A. A., 1972  The influence of ultraviolet radiation on the pigeon's color discrimination. J. Exp. Anal. Behav. 17:325-337[Medline].

YOKOYAMA, R. and S. YOKOYAMA, 1993  Molecular characterization of a blue visual pigment gene in the fish Astyanax fasciatus.. FEBS Lett. 334:27-31[Medline].

YOKOYAMA, S., 1994  Gene duplications and evolution of the short wavelength-sensitive visual pigments in vertebrates. Mol. Biol. Evol. 11:32-39[Abstract].

YOKOYAMA, S., 1995  Amino acid replacements and wavelength absorption of visual pigments in vertebrates. Mol. Biol. Evol. 12:53-61[Abstract].

YOKOYAMA, S., 1997  Molecular genetic basis of adaptive selection: examples from color vision in vertebrates. Annu. Rev. Genet. 31:311-332.

YOKOYAMA, S. and F. B. RADLWIMMER, 1998  The "five-sites" rule and the evolution of red and green color vision in mammals. Mol. Biol. Evol. 15:560-567[Abstract].

YOKOYAMA, S. and F. B. RADLWIMMER, 1999  The molecular genetics of red and green color vision in mammals. Genetics 153:919-932[Abstract/Free Full Text].

YOKOYAMA, S. and R. YOKOYAMA, 1996  Adaptive evolution of photoreceptors and visual pigments in vertebrates. Annu. Rev. Ecol. Syst. 27:543-567