FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2097322897> ?p ?o ?g. }

• W2097322897 endingPage "9466" @default.
• W2097322897 startingPage "9460" @default.
• W2097322897 abstract "Zebrafish and goldfish are both diurnal freshwater fish species belonging to the same family, Cyprinidae, but their visual ecological surroundings considerably differ. Zebrafish are surface swimmers in conditions of broad and shortwave-dominated background spectra and goldfish are generalized swimmers whose light environment extends to a depth of elevated short wavelength absorbance with turbidity. The peak absorption spectrum (λmax) of the zebrafish blue (SWS2) visual pigment is consistently shifted to short wavelength (416 nm) compared with that of the goldfish SWS2 (443 nm). Among the amino acid differences between the two pigments, only one (alanine in zebrafish and serine in goldfish at residue 94) was previously known to cause a difference in absorption spectrum (14-nm λmax shift in newt SWS2). In this study, we reconstructed the ancestral SWS2 pigment of the two species by applying likelihood-based Bayesian statistics and performing site-directed mutagenesis. The reconstituted ancestral photopigment had a λmax of 430 nm, indicating that zebrafish and goldfish achieved short wavelength (-14 nm) and long wavelength (+13 nm) spectral shifts, respectively, from the ancestor. Unexpectedly, the S94A mutation resulted in only a -3-nm spectral shift when introduced into the goldfish SWS2 pigment. Nearly half of the long wavelength shift toward the goldfish pigment was achieved instead by T116L (6 nm). The S295C mutation toward zebrafish SWS2 contributed to creating a ridge of absorbance around 400 nm and broadening its spectral sensitivity in the short wavelength direction. These results indicate that the evolutionary engineering approach is very effective in deciphering the process of functional divergence of visual pigments. Zebrafish and goldfish are both diurnal freshwater fish species belonging to the same family, Cyprinidae, but their visual ecological surroundings considerably differ. Zebrafish are surface swimmers in conditions of broad and shortwave-dominated background spectra and goldfish are generalized swimmers whose light environment extends to a depth of elevated short wavelength absorbance with turbidity. The peak absorption spectrum (λmax) of the zebrafish blue (SWS2) visual pigment is consistently shifted to short wavelength (416 nm) compared with that of the goldfish SWS2 (443 nm). Among the amino acid differences between the two pigments, only one (alanine in zebrafish and serine in goldfish at residue 94) was previously known to cause a difference in absorption spectrum (14-nm λmax shift in newt SWS2). In this study, we reconstructed the ancestral SWS2 pigment of the two species by applying likelihood-based Bayesian statistics and performing site-directed mutagenesis. The reconstituted ancestral photopigment had a λmax of 430 nm, indicating that zebrafish and goldfish achieved short wavelength (-14 nm) and long wavelength (+13 nm) spectral shifts, respectively, from the ancestor. Unexpectedly, the S94A mutation resulted in only a -3-nm spectral shift when introduced into the goldfish SWS2 pigment. Nearly half of the long wavelength shift toward the goldfish pigment was achieved instead by T116L (6 nm). The S295C mutation toward zebrafish SWS2 contributed to creating a ridge of absorbance around 400 nm and broadening its spectral sensitivity in the short wavelength direction. These results indicate that the evolutionary engineering approach is very effective in deciphering the process of functional divergence of visual pigments. Vertebrate visual pigments consist of a protein moiety (i.e. opsin), and a chromophore, either 11-cis-retinal (vitamin A1 aldehyde) or 11-cis 3,4-dehydroretinal (vitamin A2 aldehyde). These pigments reside in rod and cone photoreceptor cells in the retina. Rods are used for dim light vision, and cones are used for daylight and color vision. Vertebrate visual opsins can be classified into five phylogenetic groups: RH1 1The abbreviations used are: RH1, rod opsin; RH2, RH1-like cone opsin; SWS1, short wavelength-sensitive type 1 cone opsin; SWS2, short wavelength-sensitive type 2 cone opsin; M/LWS, middle to long wavelength-sensitive cone opsin; TM, transmembrane. (rod opsin or rhodopsin), RH2 (RH1-like, or green, cone opsin), SWS1 (short wavelength-sensitive type 1, or UV-blue, cone opsin), SWS2 (short wavelength-sensitive type 2, or blue, cone opsin), and M/LWS (middle to long wavelength-sensitive, or red-green, cone opsin) (1Yokoyama S. Prog. Retin. Eye Res. 2000; 19: 385-419Crossref PubMed Scopus (446) Google Scholar). When an opsin forms a photopigment with the A2 chromophore, its peak absorption spectrum (λmax) is located at a longer wavelength than it forms with the A1 chromophore. The spectral shift is wavelength-dependent; it can be as great as 60 nm at longer wavelengths, and as little as 5–10 nm at shorter wavelengths (2Whitmore A.V. Bowmaker J.K. J. Comp. Physiol. A. 1989; 166: 103-115Crossref Scopus (127) Google Scholar, 3Bowmaker J.K. Prog. Retin. Eye Res. 1995; 15: 1-31Crossref Scopus (171) Google Scholar). Despite having all five types of opsins in common and sharing ancestry in the same family Cyprinidae, zebrafish (Danio rerio) and goldfish (Carassius auratus) use different chromophores, A1 and A2, respectively, under normal conditions (4Nawrocki L. BreMiller R. Streisinger G. Kaplan M. Vision Res. 1985; 25: 1569-1576Crossref PubMed Scopus (73) Google Scholar, 5Palacios A.G. Varela F.J. Srivastava R. Goldsmith T.H. Vision Res. 1998; 38: 2135-2146Crossref PubMed Scopus (82) Google Scholar, 6Saszik S. Bilotta J. Vision Res. 1999; 39: 1051-1058Crossref PubMed Scopus (35) Google Scholar). This is consistent with the differences in their visual ecology: zebrafish are surface swimmers under broad and short wavelength-dominated background spectra, while goldfish are generalized swimmers whose light environment extends to a depth of elevated short wavelength absorbance with turbidity, although both are diurnal freshwater fish species (7Levine J.S. MacNichol Jr., E.F. Sens. Processes. 1979; 3: 95-131PubMed Google Scholar, 8Nicol J.A.C. The Eyes of Fishes. Oxford University Press, New York1989Google Scholar). Furthermore, the λmax values of the zebrafish visual pigments are generally shifted to short wavelengths compared with those of other species (9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar). In particular, the blue (SWS2) visual pigment is greatly shifted to short wavelength (λmax at 416 nm) compared with SWS2 pigments of other vertebrates (λmax ranging from 430 to 450 nm) under the conditions involving the A1 chromophore (9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar, 10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar). This is in sharp contrast with goldfish whose A2-based SWS2 visual pigment has a λmax around 454 nm (11Parry J.W. Bowmaker J.K. Vision Res. 2000; 40: 2241-2247Crossref PubMed Scopus (20) Google Scholar). When the goldfish SWS2 visual pigment is reconstituted in vitro with the A1 chromophore, its λmax is 443 nm (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar) and is still 27 nm longer than the λmax of the zebrafish SWS2 pigment. By site-directed mutagenesis of the cottoid fish SWS2 opsins, T118A (threonine to alanine substitution at residue 118: site numbers hereafter follow those of bovine rod opsin), T118G and T269A were shown to cause a recognizable blue shift of their SWS2 pigments (12Cowing J.A. Poopalasundaram S. Wilkie S.E. Bowmaker J.K. Hunt D.M. Biochemistry. 2002; 41: 6019-6025Crossref PubMed Scopus (52) Google Scholar). Mutagenesis to the newt SWS2 opsin resulted in spectral effects in P91S, S94A, I122M, Y261F, and A292S in a short wavelength direction (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). The spectral effect of T269A, together with that of S269A, were also verified in another mutagenesis experiment in pigeon, chicken, finch, American chameleon, bull frog, salamander, and goldfish SWS2 pigments (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar). In addition, I49A, V52I, T93V, and L207I were inferred to cause greater than a 5-nm spectral shift by multiple regression analysis (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar). However, among these amino acid substitutions, only S94A is detected among amino acid differences between goldfish and zebrafish SWS2 pigments. The S94A mutation results in a 14-nm blue shift when introduced into the newt SWS2 pigment which has an exceptionally long λmax value (474 nm) among A1-based SWS2 pigments (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). Even if the S94A mutation contributes to the spectral difference between the goldfish and zebrafish SWS2 pigments with the 14-nm magnitude, additional, and previously unforeseen, mutations are required to fill the 27-nm spectral difference between the two pigments. The objective of this study is to extract amino acid substitutions that contribute to the spectral difference between the zebrafish and goldfish SWS2 pigments. For this purpose, we employed an evolutionary engineering approach, that is, inference of the ancestral amino acid sequence of the zebrafish and goldfish SWS2 opsins by likelihood-based Bayesian statistics, reconstitution of full and partial ancestral photopigments by site-directed mutagenesis, and spectral measurements of the reconstituted pigments. Sequence Analysis—The following fish SWS2 opsin genes were studied: cottoid (Batrachocottus nikolskii) (GenBank™ accession number AJ430474) (12Cowing J.A. Poopalasundaram S. Wilkie S.E. Bowmaker J.K. Hunt D.M. Biochemistry. 2002; 41: 6019-6025Crossref PubMed Scopus (52) Google Scholar), cichlid (Dimidiochromis compressiceps) SWS2A (GenBank™ accession number AF247113), and SWS2B (GenBank™ accession number AF247117) (14Carleton K.L. Kocher T.D. Mol. Biol. Evol. 2001; 18: 1540-1550Crossref PubMed Scopus (206) Google Scholar), medaka (Oryzias latipes) (GenBank™ accession number AB001602) (15Hisatomi O. Satoh T. Tokunaga F. Vision Res. 1997; 37: 3089-3096Crossref PubMed Scopus (62) Google Scholar), Mexican cavefish (Astyanax fasciatus) (GenBank™ accession number AH007939) (16Yokoyama R. Yokoyama S. FEBS Lett. 1993; 334: 27-31Crossref PubMed Scopus (26) Google Scholar), goldfish (Carassius auratus) (GenBank™ accession number L11864) (17Johnson R.L. Grant K.B. Zankel T.C. Boehm M.F. Merbs S.L. Nathans J. Nakanishi K. Biochemistry. 1993; 32: 208-214Crossref PubMed Scopus (148) Google Scholar) and zebrafish (GenBank™ accession number AB087809) (9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar). Additionally, the following SWS2 opsins of non-fish vertebrates were used as outgroups: newt (Cynops pyrrhogaster) (GenBank™ accession number AB040148) (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar), salamander (Ambystoma tigrinum) (GenBank™ accession number AF038946) (18Xu L. Hazard III, E.S. Lockman D.K. Crouch R.K. Ma J. Mol. Vis. 1998; 4: 10PubMed Google Scholar), bull frog (Rana catesbeiana) (GenBank™ accession number AB010085) (19Hisatomi O. Takahashi Y. Taniguchi Y. Tsukahara Y. Tokunaga F. FEBS Lett. 1999; 447: 44-48Crossref PubMed Scopus (35) Google Scholar), American chameleon (Anolis carolinensis) (GenBank™ accession number AF133907) (20Kawamura S. Yokoyama S. Vision Res. 1996; 36: 2797-2804Crossref PubMed Scopus (31) Google Scholar, 21Kawamura S. Yokoyama S. Vision Res. 1998; 38: 37-44Crossref PubMed Scopus (104) Google Scholar), zebrafinch (Taeniopygia guttata) (GenBank™ accession number AF222332) (22Yokoyama S. Blow N.S. Radlwimmer F.B. Gene (Amst.). 2000; 259: 17-24Crossref PubMed Scopus (33) Google Scholar), chicken (Gallus gallus) (GenBank™ accession number P28682) (23Okano T. Fukada Y. Artamonov I.D. Yoshizawa T. Biochemistry. 1989; 28: 8848-8856Crossref PubMed Scopus (130) Google Scholar, 24Okano T. Kojima D. Fukada Y. Shichida Y. Yoshizawa T. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5932-5936Crossref PubMed Scopus (316) Google Scholar), and pigeon (Columba livia) (GenBank™ accession number Af149238) (25Kawamura S. Blow N.S. Yokoyama S. Genetics. 1999; 153: 1839-1850Crossref PubMed Google Scholar). We aligned the deduced amino acid sequences of these opsins using CLUSTAL W (26Thompson J.D. Higgins D.G. Gibson T.J. Nucleic Acids Res. 1994; 22: 4673-4680Crossref PubMed Scopus (56002) Google Scholar). Alignment of amino acids corresponding to residues 2–8 and the last 27 of zebrafish SWS2 opsin were less reliable, requiring multiple gaps. The remaining 320 amino acids were considered for phylogenetic analysis. Using the MEGA2 program version 2.1 (27Nei M. Kumar S. Molecular Evolution and Phylogenetics. Oxford University Press, New York2000Google Scholar, 28Kumar S. Tamura K. Jacobsen I.B. Nei M. MEGA2, Molecular Evolutionary Genetics Analysis. 2.1 Ed. Arizona State University, Tempe, AZ2001Crossref Scopus (4557) Google Scholar), the number of amino acid substitutions per site for two sequences was estimated by Poisson correction and the phylogenetic tree was reconstructed using the neighbor-joining method (29Saitou N. Nei M. Mol. Biol. Evol. 1987; 4: 406-425PubMed Google Scholar) with 1000 bootstrap replications. Given the tree topology reconstructed (Fig. 1), we inferred the ancestral amino acid sequences of SWS2 opsins at every node in the tree by using the PAML computer program with a likelihood-based Bayesian method (30Yang Z. Kumar S. Nei M. Genetics. 1995; 141: 1641-1650Crossref PubMed Google Scholar, 31Yang Z. Comput. Appl. Biosci. 1997; 13: 555-556PubMed Google Scholar) (abacus.gene.ucl.ac.uk/software/paml.html). For computations, we used the empirical substitution matrix of Dayhoff (32Dayhoff M.O. Schwartz R.M. Orcutt B.C. Dayhoff M.O. Atlas of Protein Sequence and Structure. National Biomedical Research Foundation, Silver Spring, MD1978: 345-352Google Scholar) or JTT (33Jones D.T. Taylor W.R. Thornton J.M. Comput. Appl. Biosci. 1992; 8: 275-282PubMed Google Scholar) as a mathematical model of amino acid substitutions. These models have often been used in constructions of maximum likelihood phylogenetic trees for protein sequences; the JTT model has been shown to give sufficiently accurate results for most purposes (27Nei M. Kumar S. Molecular Evolution and Phylogenetics. Oxford University Press, New York2000Google Scholar). Both models are based on amino acid substitution data for conserved proteins and have been applied to inference of ancestral sequences of vertebrate opsin genes (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar, 34Yokoyama S. Radlwimmer F.B. Mol. Biol. Evol. 1998; 15: 560-567Crossref PubMed Scopus (141) Google Scholar, 35Yokoyama S. Shi Y. FEBS Lett. 2000; 486: 167-172Crossref PubMed Scopus (83) Google Scholar, 36Yokoyama S. Radlwimmer F.B. Genetics. 2001; 158: 1697-1710PubMed Google Scholar, 37Shi Y. Yokoyama S. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 8308-8313Crossref PubMed Scopus (154) Google Scholar, 38Zhang J. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 8045-8047Crossref PubMed Scopus (8) Google Scholar). The certainty of each inferred amino acid was expressed by its posterior probability. The ancestral amino acids of poorly aligned regions (i.e. residues corresponding to 2–8 and the last 27 of zebrafish SWS2) were inferred visually. Construction of the Ancestral SWS2 Opsin cDNA—The SWS2 cDNA of zebrafish (9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar) and goldfish were cloned into the pMT5 (39Khorana H.G. Knox B.E. Nasi E. Swanson R. Thompson D.A. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 7917-7921Crossref PubMed Scopus (82) Google Scholar) expression vector. We introduced point mutations into the pMT5 clones using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). All mutated cDNAs were sequenced to confirm that no spurious mutations were incorporated. Sequencing was carried out using the Thermo Sequence Cycle Sequencing kit (Amersham Biosciences) with dye-labeled primers and the LI-COR 4200L-1 automated DNA sequencer. Visual Pigment Reconstitution—The pMT5 expression vector contains the last 15 amino acids of bovine rod opsin necessary for immunoaffinity purification by the 1D4 monoclonal antibody (40Molday R.S. MacKenzie D. Biochemistry. 1983; 22: 653-660Crossref PubMed Scopus (355) Google Scholar). We transfected each pMT5-cDNA clone into cultured COS-1 cells (RIKEN Cell Bank, Tsukuba, Japan), incubated the resuspended cells with 5 μm 11-cis-retinal (Storm Eye Institute, Medical University of South Carolina, Charleston, SC), solubilized the cells with 1% dodecyl maltoside, and purified the resulting pigments using immobilized 1D4 (Cell Culture Center, Minneapolis, MN) as described previously (21Kawamura S. Yokoyama S. Vision Res. 1998; 38: 37-44Crossref PubMed Scopus (104) Google Scholar). The UV-visible absorption spectrum of each pigment was recorded from 250 to 650 nm at 1-nm intervals using the Hitachi U3010 dual beam spectrometer at 20 °C. Five replicates were performed in the dark and five more after 3 min of light exposure using a 366 nm UV illuminator (model UVL-65 Blak-Ray Lamp, 100 V 0.16 A; UVP, Inc.) as described previously (21Kawamura S. Yokoyama S. Vision Res. 1998; 38: 37-44Crossref PubMed Scopus (104) Google Scholar). Savitzky-Golay's least squares smoothing method (41Gorry P.A. Anal. Chem. 1990; 62: 570-573Crossref Scopus (589) Google Scholar) was carried out for each absorbance curve using 100 repetitions to eliminate spurious spikes. All λmax values in this study were taken from spectra generated in dark conditions. We evaluated the time course of 11-cis-retinal dissociation from the zebrafish SWS2 pigment at 20 °C by adding hydroxylamine (pH 6.7) to the purified zebrafish SWS2 pigment at a final concentration of 20 mm. The pigment was exposed to light for 20 min after hydroxylamine bleaching. The light source was a 60-watt room lamp, and cut-off wavelengths below 440 nm were filtered with the Kodak Wratten Gelatin Filter No.3. Ancestral Pigment of Zebrafish and Goldfish SWS2—The ancestral amino acid sequence of the zebrafish and goldfish SWS2 opsins was estimated by using the PAML program with either the JTT or Dayhoff model of amino acid substitutions. Under both models, the average posterior probability over the entire amino acid sequence subjected to PAML analysis was 0.98. Amino acids with posterior probabilities less than 0.9 by either model are listed in Table I. There were two amino acid differences between the sequences inferred by the two models: glycine (JTT) and alanine (Dayhoff) at site 50 (designated G50A) and L85F. Fig. 2 shows the ancestral sequence inferred by the JTT model. From the ancestral SWS2 opsin, there were 24 amino acid differences in the goldfish SWS2 opsin and 42 differences plus one indel (in the amino terminus) in the zebrafish SWS2. We therefore used the goldfish SWS2 opsin cDNA as a template to create the ancestral opsin cDNA by substituting the 24 amino acids. We then reconstituted the ancestral photopigment (pigment ZG) with the A1 chromophore 11-cis-retinal, and measured the λmax to be 430 nm (Table II and Fig. 3). The pigment was shown to be photoreactive, exhibiting a λmax of 380 nm of all-trans-retinal upon light exposure (data not shown). The Dayhoff-predicted versions of the ancestral pigments were also reconstituted by introducing the G50A and L85F substitutions into the pigment ZG (ZG_G50A and ZG_L85F, respectively). Neither substitution affected λmax (430 nm, see Table II). These results indicate that a 14-nm blue shift and a 13-nm red shift occurred in the λmax of SWS2 pigments in zebrafish and goldfish lineages, respectively, after the separation of the two fish species.Table IAmino acid residues having less than 0.9 posterior probabilities by either JTT or Dayhoff model in ancestral sequence reconstruction of the SWS2 visual pigments of zebrafish and goldfishSiteJTTDayhoff3Phe (0.899)Phe (0.886)4His (0.760)His (0.875)15Ile (0.681)Ile (0.572)33Asn (0.769)Asn (0.882)34Ser (0.886)Ser (0.910)42Val (0.850)Val (0.903)50Gly (0.550)Ala (0.499)82Ala (0.852)Ala (0.823)85Leu (0.481)Phe (0.542)88Ile (0.664)Ile (0.564)92Ser (0.683)Ser (0.641)93Val (0.513)Val (0.550)99Phe (0.794)Phe (0.777)101Arg (0.735)Arg (0.817)162Ile (0.805)Ile (0.994)163Phe (0.521)Phe (0.429)217Ser (0.732)Ser (0.723)266Ala (0.773)Ala (0.827)281Glu (0.628)Glu (0.674) Open table in a new tab Table IIλmax values of current and ancestral SWS2 visual pigments of zebrafish and goldfish reconstituted in vitroPigmentλmax ± S.E.nmCurrentZebrafish416 ± 1.0aMeasured as described in Ref. 9Goldfish443 ± 0.0AncestralZG430 ± 0.3ZG_G50A430 ± 0.7ZG_L85F430 ± 0.0a Measured as described in Ref. 9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar Open table in a new tab Fig. 3Absorption spectra of zebrafish, goldfish, and their ancestral SWS2 visual pigments reconstituted in vitro. Spectra are normalized for their peak height at λmax to be 1. The spectrum of the zebrafish pigment was taken from Ref. 9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar, and those of the others were measured in this study.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Mutations in Zebrafish SWS2—Amino acid substitutions known to exert spectral effects in SWS2 pigments are I49A, V52I, P91S, T93V, S94A, T118G, T118A, I122M, L207I, Y261F, T269A, S269A, and A292S (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar, 12Cowing J.A. Poopalasundaram S. Wilkie S.E. Bowmaker J.K. Hunt D.M. Biochemistry. 2002; 41: 6019-6025Crossref PubMed Scopus (52) Google Scholar, 13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). None of these mutations was detected between the zebrafish and ancestral SWS2 pigments (Fig. 2). To explore amino acid substitutions that can explain the 14-nm blue shift in zebrafish SWS2 pigment from ancestral pigment ZG, we first constructed three chimeric opsins: A(70)Z, Z(70)A(128)Z, and Z(128)A, using the restriction enzymes, BglII and BsaI (see Fig. 2), where amino acids 1–70, 71–128, and 129–347 of the zebrafish SWS2 opsin, respectively, were replaced with the corresponding segments of the ancestral opsin. The chimeric photopigments were reconstituted with 11-cis-retinal and absorption spectra were measured (see Table III for λmax values). Among the three chimeric pigments, pigments Z(70)A(128)Z and Z(128)A showed recognizable spectral shifts from the intact zebrafish pigment of +7 and +3 nm, respectively (Fig. 4, A and B). The spectral shift by the pigment A(70)Z was +1 nm. When segment 1–128 of the zebrafish SWS2 opsin was replaced by the corresponding ancestral region [A(128)Z], the magnitude of the spectral shift was no different from that of Z(70)A(128)Z (+7 nm) (Table III), confirming that the spectral effect of segment 1–70 is very small. When segment 71–347 was replaced [Z(70)A], the λmax shift was +12 nm (Table III and Fig. 4C), slightly larger than the sum of the individual effects of the two segments 71–128 and 129–347, +10 nm, and closer to the net difference of 14 nm between the zebrafish and ancestral pigments. These results suggest that the majority of amino acid substitutions responsible for the blue-shifted λmax of the zebrafish SWS2 pigment were in the 71–347 region.Table IIISpectral effects of mutations introduced into the zebrafish SWS2 pigmentPigmentλmax ± S.E.Δλ from zebrafish SWS2nmChimeraA(70)Z417 ± 0.4+1Z(70)A(128)Z423 ± 0.8+7Z(128)A419 ± 0.4+3A(128)Z423 ± 0.4+7Z(70)A428 ± 0.7+12Mutant 71–128RBPaAmino acids located in the retinal binding pocketZF SWS2_S117A418 ± 0.4+2Non-RBPZF SWS2_S82A416 ± 0.80ZF SWS2_W85L417 ± 0.0+1ZF SWS2_V88I416 ± 0.00ZF SWS2_A97S417 ± 0.3+1ZF SWS2_Y99F418 ± 0.5+2ZF SWS2_K100N417 ± 0.3+1ZF SWS2_I108T417 ± 0.7+1ZF SWS2_G109A417 ± 0.5+1ZF SWS2_I119L416 ± 0.30Mutant 129–347RBPaAmino acids located in the retinal binding pocketZF SWS2_C295S419 ± 0.3+3a Amino acids located in the retinal binding pocket Open table in a new tab There are a total of 12 and 15 amino acid differences in segments 71–128 and 129–347, respectively, between the zebrafish and ancestral SWS2 opsins (Fig. 2). A study of the x-ray crystal structure of bovine rod opsin identified 27 amino acids as comprising the retinal binding pocket, located within 4.5 Å from the retinal (42Palczewski K. Kumasaka T. Hori T. Behnke C.A. Motoshima H. Fox B.A. Le Trong I. Teller D.C. Okada T. Stenkamp R.E. Yamamoto M. Miyano M. Science. 2000; 289: 739-745Crossref PubMed Scopus (5056) Google Scholar). Subsequent studies identified a total of 38 residues, including the above-mentioned 27, as surrounding the retinal (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar, 43Menon S.T. Han M. Sakmar T.P. Physiol. Rev. 2001; 81: 1659-1688Crossref PubMed Scopus (283) Google Scholar) (Fig. 2). Among these, only residues 117 and 295, in the 71–128 and 129–347 segments, respectively, were different between the zebrafish and ancestral SWS2 opsins: serine in zebrafish and alanine in ancestral opsins at residue 117 (S117A) and C295S. Mutations S117A and C295S, introduced into the zebrafish SWS2 opsin, resulted in only +2- and +3-nm spectral shifts, respectively (Table III). However, the latter effect is of the same magnitude as that of the segmental replacement of residues 129–347 (Fig. 4, B and D). We therefore focused on the segment 71–128 and explored amino acid substitutions accounting for the 7-nm difference in which the segmental replacement resulted. All the remaining amino acid substitutions in transmembrane regions 2 and 3 (TM2 and TM3) in the 71–128 segment were introduced into the zebrafish SWS2 pigment one by one: S82A, W85L, V88I, A97S, Y99F, K100N, I108T, G109A, and I119L (Fig. 2). Individual spectral effects were all minor, ranging from 0 to +2 nm (Table III). However, the simple sum of these effects, +9 nm, was comparable with the segmental effect of +7 nm. Furthermore, the sum of all the substitutions in segments 71–128 and 129–347 was +12 nm, concordant with the effect of segmental replacement of 71–347 (Z(70)A in Table III). Although not all combinations of these mutations have been tested for the additivity of their spectral effects, we examined a double mutation, I108T/G109A, a triple mutation, I108T/G109A/S117A, and a quadruple mutation, I108T/G109A/S117A/C295S, and confirmed their linear additivity; their spectral shifts were +2, +5, and +7 nm, respectively. These results suggest that the blue shift of the spectral sensitivity from the ancestral to the current zebrafish SWS2 pigments was achieved by the accumulation of amino acid substitutions with individual minor spectral effects. Mutations in Goldfish SWS2—Among the known amino acid substitutions to exert spectral effects in SWS2 pigments, only S94A was observed between the goldfish and ancestral SWS2 pigments (Fig. 2). When introduced into the newt SWS2 pigment, S94A is known to exert a -14-nm spectral shift (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar) and promised to explain the 13-nm difference between the goldfish (λmax 443 nm) and the ancestral (430 nm) SWS2 pigments. However, when introduced into the goldfish SWS2 pigment, S94A resulted in only a -3-nm shift (Fig. 5A and Table IV). We then tested all other amino acid substitutions in TM regions that alter either their electric charge or polarity: A52T, S80A, A87S, P92S, L116T, G124S, S163F, G217S, and S272A (Fig. 2). Among them, L116T caused a -6-nm shift (Fig. 5B and Table IV). Effects of other mutations were all minor (0–2 nm) (Table IV). However, the sum of all mutations, -12 nm, was comparable with the 13-nm difference between the goldfish and ancestral SWS2 pigments, suggesting that a red shift of the goldfish SWS2 pigment from the ancestral pigment was achieved in half by T116L and in half by accumulation of amino acid mutations, including A94S, with individual minor spectral effects.Table IVSpectral effects of mutations introduced into the goldfish SWS2 pigmentPigmentλmax ± S.E.Δλ from goldfish SWS2nmGF SWS2_A52T443 ± 0.00GF SWS2_S80A443 ± 0.00GF SWS2_A87S442 ± 0.0-1GF SWS2_P92SNo peak valueaNo recognizable absorption peak was observedGF SWS2_S94A440 ± 0.0-3GF SWS2_L116T437 ± 0.4-6GF SWS2_G124S441 ± 0.4-2GF SWS2_S163F444 ± 1.2+1GF SWS2_G217S443 ± 1.20GF SWS2_S272A442 ± 0.4-1a No recognizable absorption peak was observed Open table in a new tab Unique Shape of Absorbance Curve in Zebrafish SWS2—We have noticed that the absorbance curve of the zebrafish SWS2 pigment has a shoulder around 400 nm (Fig. 3). As far as we know, such a shoulder closely located to the λmax position has not been reported in other opsin-based vertebrate photopigments. To confirm whether this was due to spurious Schiff base linkages of 11-cis-retinal to proteins (including opsin) or contamination of non-retinal-based pigments having a 400-nm absorbance peak, we monitored the process of 11-cis-retinal dissociation from the zebrafish SWS2 pigment by hydroxylamine over time. Hydroxylamine cleaves Schiff base linkages of 11-cis-retinal and results in a free 11-cis-retinal oxime in solution and shows an absorption peak around 365 nm. As shown in Fig. 6, all absorbance curves, taken at different time points, crossed at a single point. This indicates that the retinal based pigment in this solution was homogenous and not a mixture of pigments having different rates of retinal dissociation. After completion of the dissociation by hydroxylamine followed by light exposure, we failed to see a shoulder at 400 nm (see curve 17 of Fig. 6) making it unlikely that non-retinal-based pigments (with 400 nm absorbance) were contaminated. In addition, only the SWS2 but no other visual pigments of zebrafish showed the 400-nm shoulder in our previous study (9Chinen A. Hamaoka T. Yamada Y. Kawamura S. Genetics. 2003; 163: 663-675Crossref PubMed Google Scholar). These results strongly suggest that the 400-nm absorbance ridge is a characteristic of the zebrafish SWS2 pigment. In the process of the mutagenesis experiments, we noticed that a single amino acid substitution (C295S) to the zebrafish SWS2, and a segmental replacement including this residue, abolished the absorbance ridge (Fig. 4D), but other point mutations did not. Therefore, the aberrant absorbance shape of the zebrafish SWS2 pigment was considered a result of the amino acid substitution, S295C. The peak absorption spectrum of the zebrafish blue-sensitive cone visual pigment, SWS2, is ∼30 nm shorter than that of the goldfish pigment, reflecting their different visual ecologies. These two species are phylogenetically close and are suitable for studying the evolutionary process of spectral differentiation. To study this evolutionary process, we inferred their ancestral amino acid sequence rather than just comparing the two sequences to clarify the identity of the mutations involved in spectral differentiation. Reconstitution of the ancestral pigment revealed that the spectral differences between the two current species was due in half to the short wavelength shift in the zebrafish pigment and in half to the long wavelength shift in the goldfish pigment. This was unexpected because a previous study suggested that the goldfish SWS2 retains the ancestral state of vertebrate SWS2 pigments (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar). However, in the study, the goldfish pigment was the only fish pigment and other pigments studied were those of amphibians, a reptile, and birds (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar). Therefore, the present study, using seven SWS2 sequences of various fish species, expanded our view on the evolution of fish SWS2 pigments and suggests that fish SWS2 pigments, including goldfish, have experienced repeated changes of absorption spectra, possibly reflecting the great variability of their aquatic light environments (44Levine J.S. MacNichol Jr., E.F. Sci. Am. 1982; 246: 140-149Crossref Scopus (116) Google Scholar). Mutagenesis carried out in this study showed that S94A caused -3-nm spectral shift when introduced into the goldfish SWS2 pigment. This conflicts with a much larger (-14 nm) effect of the same mutation in the newt SWS2 pigment (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). This suggests that the effect of this mutation is heavily dependent on the background amino acid sequence. In the three-dimensional structure, residue 94 is located close to the counterion and Schiff base, and substitution at this location has been suspected to perturb the environment around the Schiff base or chromophore by two or more unidentified components (13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). Dependence of the spectral effects of amino acid substitutions on background amino acid sequences has also been observed in SWS1 pigments (37Shi Y. Yokoyama S. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 8308-8313Crossref PubMed Scopus (154) Google Scholar). Prediction of λmax from amino acid sequences requires caution for the SWS2 system as well as for SWS1. When introduced into the goldfish SWS2, L116T exerts -6-nm effect and accounts for nearly half of the spectral difference from the ancestor. This location is one of the spectral tuning sites identified for SWS1 pigments, which exerts unrecognizable changes individually, but when combined with other mutations causes a significant synergistic spectral change (37Shi Y. Yokoyama S. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 8308-8313Crossref PubMed Scopus (154) Google Scholar). Perhaps because of the individual minor effects in SWS1 pigments or its locality outside the region surrounding the retinal, the residue 116 was not previously considered for mutagenesis experiments in SWS2 pigments (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar, 13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). Residue 116 is located near the counterion site 113 (glutamic acid) and is varied among the SWS2 pigments of vertebrates. Pigeon, chicken, and newt SWS2 pigments contain alanine at this location while many others contain threonine, and their relatively red-shifted λmax values (see Fig. 1) may also be associated with the residue at this site. In zebrafish, two sites in the retinal-binding pocket were found mutated, 117 and 295. Site 117 is highly conserved among vertebrate opsins: alanine in RH1, RH2, and SWS2 opsins, glycine in SWS1 opsins, and valine in M/LWS opsins. The A117G mutation is known to cause a -4-nm spectral shift when introduced into bovine RH1 (45Lin S.W. Kochendoerfer G.G. Carroll K.S. Wang D. Mathies R.A. Sakmar T.P. J. Biol. Chem. 1998; 273: 24583-24591Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). A serine residue at this site is unique to the zebrafish SWS2 opsin among all vertebrate visual opsins sequenced to date. Despite this uniqueness, the spectral effect of S117A was small (+2 nm) (Table III). The 295 site neighbors the chromophore binding site at 296 (lysine) and is also highly conserved among vertebrate opsins: serine in RH2, SWS2, and SWS1 opsins and alanine in RH1 and M/LWS opsins. The A295S mutation causes a -5-nm spectral shift when introduced into bovine RH1 (45Lin S.W. Kochendoerfer G.G. Carroll K.S. Wang D. Mathies R.A. Sakmar T.P. J. Biol. Chem. 1998; 273: 24583-24591Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). The cysteine residue at this site is also unique to the zebrafish SWS2 opsin. The S295C substitution in the zebrafish SWS2 not only shifts its λmax by 3 nm toward short wavelengths but also creates an absorbance ridge at ∼400 nm. This absorbance ridge broadens its spectral sensitivity toward shorter wavelengths. As far as we know, this is the first report of spectral shifts by changing the shape of absorbance curves. Although cysteine may affect the resonance by decreasing available protonation in the Schiff base region, thus providing a higher energy, it is not clear why the shoulder appears at 400 nm. The role of cysteine in this process and further details of its mechanism require additional studies. The shoulder is, however, unrecognizable by microspectrophotometry measurements of native photoreceptor absorbance (46Cameron D.A. Vis. Neurosci. 2002; 19: 365-372Crossref PubMed Scopus (50) Google Scholar). Microspectrophotometry is generally more susceptible to noise than in vitro measurements, and the shoulder we found in this study may be too subtle for microspectrophotometry to detect. In both goldfish and zebrafish, most of the mutations resulted in minor spectral shifts, the simple sum of which resembled the net spectral differences between the ancestral and current pigments. Although we have not rigorously tested the additivity of these effects by introducing all mutations simultaneously, the nearly additive nature of the individual spectral effects have been demonstrated for the SWS2 pigments (10Yokoyama S. Tada T. Gene (Amst.). 2003; 306: 91-98Crossref PubMed Scopus (39) Google Scholar, 13Takahashi Y. Ebrey T.G. Biochemistry. 2003; 42: 6025-6034Crossref PubMed Scopus (83) Google Scholar). Results from the present study suggest that the λmax distribution of SWS2 pigments could be continuous rather than discrete, unlike M/LWS pigments. Investigation of visual pigments in other fish species inhabiting various light environments, together with the evolutionary engineering approach and ancestral pigment reconstitution, would greatly promote our understanding of the evolution of the visual system and the interaction of vision-environment relationships. We thank Drs. Y. Fukada and T. Okano for instructive advice and BioMed Proofreading for their services." @default.
• W2097322897 created "2016-06-24" @default.
• W2097322897 creator A5014321570 @default.
• W2097322897 creator A5019366369 @default.
• W2097322897 creator A5077085670 @default.
• W2097322897 date "2005-03-01" @default.
• W2097322897 modified "2023-09-30" @default.
• W2097322897 title "Spectral Differentiation of Blue Opsins between Phylogenetically Close but Ecologically Distant Goldfish and Zebrafish" @default.
• W2097322897 cites W1909840042 @default.
• W2097322897 cites W1966055777 @default.
• W2097322897 cites W1973803699 @default.
• W2097322897 cites W1976567526 @default.
• W2097322897 cites W1979887680 @default.
• W2097322897 cites W1982846458 @default.
• W2097322897 cites W1990932315 @default.
• W2097322897 cites W1998597611 @default.
• W2097322897 cites W2006735809 @default.
• W2097322897 cites W2009542300 @default.
• W2097322897 cites W2015875853 @default.
• W2097322897 cites W2016045532 @default.
• W2097322897 cites W2022604487 @default.
• W2097322897 cites W2038764407 @default.
• W2097322897 cites W2050032235 @default.
• W2097322897 cites W2054186827 @default.
• W2097322897 cites W2062742713 @default.
• W2097322897 cites W2063036438 @default.
• W2097322897 cites W2070487314 @default.
• W2097322897 cites W2071027979 @default.
• W2097322897 cites W2078401813 @default.
• W2097322897 cites W2080005671 @default.
• W2097322897 cites W2081236714 @default.
• W2097322897 cites W2083355330 @default.
• W2097322897 cites W2085478592 @default.
• W2097322897 cites W2089380788 @default.
• W2097322897 cites W2091673574 @default.
• W2097322897 cites W2095266606 @default.
• W2097322897 cites W2098371208 @default.
• W2097322897 cites W2106882534 @default.
• W2097322897 cites W2111762123 @default.
• W2097322897 cites W2116486508 @default.
• W2097322897 cites W2116835232 @default.
• W2097322897 cites W2149963584 @default.
• W2097322897 cites W2153687216 @default.
• W2097322897 cites W2160674212 @default.
• W2097322897 cites W2187466185 @default.
• W2097322897 doi "https://doi.org/10.1074/jbc.m413001200" @default.
• W2097322897 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15623516" @default.
• W2097322897 hasPublicationYear "2005" @default.
• W2097322897 type Work @default.
• W2097322897 sameAs 2097322897 @default.
• W2097322897 citedByCount "35" @default.
• W2097322897 countsByYear W20973228972012 @default.
• W2097322897 countsByYear W20973228972013 @default.
• W2097322897 countsByYear W20973228972014 @default.
• W2097322897 countsByYear W20973228972015 @default.
• W2097322897 countsByYear W20973228972016 @default.
• W2097322897 countsByYear W20973228972017 @default.
• W2097322897 countsByYear W20973228972018 @default.
• W2097322897 countsByYear W20973228972019 @default.
• W2097322897 countsByYear W20973228972020 @default.
• W2097322897 countsByYear W20973228972021 @default.
• W2097322897 countsByYear W20973228972022 @default.
• W2097322897 crossrefType "journal-article" @default.
• W2097322897 hasAuthorship W2097322897A5014321570 @default.
• W2097322897 hasAuthorship W2097322897A5019366369 @default.
• W2097322897 hasAuthorship W2097322897A5077085670 @default.
• W2097322897 hasBestOaLocation W20973228971 @default.
• W2097322897 hasConcept C104317684 @default.
• W2097322897 hasConcept C202033177 @default.
• W2097322897 hasConcept C2776878037 @default.
• W2097322897 hasConcept C2780827179 @default.
• W2097322897 hasConcept C2909208804 @default.
• W2097322897 hasConcept C30028475 @default.
• W2097322897 hasConcept C505870484 @default.
• W2097322897 hasConcept C54355233 @default.
• W2097322897 hasConcept C59822182 @default.
• W2097322897 hasConcept C78458016 @default.
• W2097322897 hasConcept C86803240 @default.
• W2097322897 hasConcept C90856448 @default.
• W2097322897 hasConceptScore W2097322897C104317684 @default.
• W2097322897 hasConceptScore W2097322897C202033177 @default.
• W2097322897 hasConceptScore W2097322897C2776878037 @default.
• W2097322897 hasConceptScore W2097322897C2780827179 @default.
• W2097322897 hasConceptScore W2097322897C2909208804 @default.
• W2097322897 hasConceptScore W2097322897C30028475 @default.
• W2097322897 hasConceptScore W2097322897C505870484 @default.
• W2097322897 hasConceptScore W2097322897C54355233 @default.
• W2097322897 hasConceptScore W2097322897C59822182 @default.
• W2097322897 hasConceptScore W2097322897C78458016 @default.
• W2097322897 hasConceptScore W2097322897C86803240 @default.
• W2097322897 hasConceptScore W2097322897C90856448 @default.
• W2097322897 hasIssue "10" @default.
• W2097322897 hasLocation W20973228971 @default.
• W2097322897 hasOpenAccess W2097322897 @default.
• W2097322897 hasPrimaryLocation W20973228971 @default.
• W2097322897 hasRelatedWork W1970793856 @default.
• W2097322897 hasRelatedWork W2048157190 @default.
• W2097322897 hasRelatedWork W2053573702 @default.

• next