Rods and Cones Contain Antigenically Distinctive S-Antigens

T. Michael Nork* Nancy]. Mangini,] and Lyndell L. Millecchia*

Purpose. S-antigen (48 kDa protein or arrestin) is known to be present in rod photoreceptors. Its localization in cones is less clear with several conflicting reports among various species examined. Methods. This study employed three different anti-S-antigen antibodies (a48K, a polyclonal antiserum and two monoclonal antibodies, MAb A9-C6 and MAb 5c6.47) and examined their localization in rods and cones of human and cat retinas. To identify the respective cone types, an enzyme histochemical technique for carbonic anhydrase (CA) was employed to distinguish blue cones (CA-negative) from red or green cones (CA-positive). S-antigen localization was then examined by immunocytochemical staining of adjacent sections. Results. In human retinas, a similar labeling pattern was seen with both a48K and MAb A9-C6, i.e., the rods and blue-sensitive cones were strongly positive, whereas the red- or green-sensitive cones showed little immunoreactivity. All human photoreceptors showed reactivity to MAb 5c6.47. In the cat retina, only CA-positive cones could be found. As in the human retina, both rods and cones of the cat were positive for MAb 5c6.47. A difference from the iabeling pattern in human retina was noted for the other S-antigen antibodies; a48K labeled rods and all of the cones, whereas MAb A9-C6 reacted strongly with the rods but showed no cone staining. Conclusions. These results suggest that both rods and cones contain S-antigen but that they are antigenically distinctive. Invest Ophthalmol Vis Sci 1993; 34:2918-2925.

O-antigen is a retinal protein that, when injected into receptors; evidence of its presence in cone photore- animals, produces experimental autoimmune uveitis.1 ceptors is less clear, with conflicting reports among Recent studies show S-antigen to be identical to arres- various species examined. For example, several differ- tin (48 kDa protein) of the rod outer segments that ent studies examining anti-S-antigen immunolabeling binds to phosphorylated, photoexcited rhodopsin, in photoreceptors of toad,3 rat,4 squirrel,5 monkey56 thereby preventing its activation of transducing and and man7 indicate that S-antigen is found only in rods, indirectly preventing the continued activation of whereas the studies by Mirshahi et al suggest that it is 8 cGMP phosphodiesterase.2 On this basis, S-antigen is in both rods and cones. Long et al found positive thought to participate either in the inactivation of the staining for S-antigen in immature, but not adult, reti- 9 photoresponse or in the process of light adaptation, or nal cones of the dog. Still other investigators have both (for a discussion and additional references, see found positive staining only in one class of cones—for Palczewski et al2). Available data from immunolabeling example, the green-sensitive cones of the chicken and 10 studies show that S-antigen is abundant in rod photo- turtle and the blue-sensitive cones of the ground squirrel11 and tree shrew.12 These apparently inconsistent findings might be the result of antigenicaily distinctive S-antigens. From the *Department of Ophthalmology, West Virginia University Health Sciences Center, Morgantoum, West Virginia, and the f Lions of Illinois Eye Research Among species, the amino acid compositions of bo- Institute, Department of Ophthalmology and Visual Sciences, University of Illinois vine,13 mouse,14 rat,1516 and human17 S-antigens have at Chicago College of Medicine, Chicago, Illinois. Supported by Grants EY 08724 and EY 07586 from the National Eye Institute, been determined and, though similar, they are not Bethesda,. Maryland. identical. Within the retinas of single species, homolo- Submitted for publication October 16, 1992; accepted December 8, 1992. Proprietary interest category: N. gous (i.e., similar but not identical) proteins of vision Reprint requests: T. Michael Nork, M.D., Department of Ophthalmology, have been found not only for the visual pigments18 but University of Wisconsin-Madison, 600 Highland Avenue, F4/3, Madison, WI 19 2021 53792-3220. also for transducing and phosphodiesterase. Sus-

Downloaded from iovs.arvojournals.org on 09/27/2021 Antigenically Distinctive S-Antigens 2919

peering that S-antigen also exists in homologous forms ethanol, and incubated in 0.1% trypsin solutions at depending on photoreceptor type, we performed im- 37°C and at a pH of 7.8 for 20 minutes. Immunola- munocytochemical localization with three different an- beling of S-antigen was then done via the avidin-bio- tibodies to S-antigen and compared their distribution tin-complex technique with a polyclonal antiserum among the photoreceptors of human and cat retinas. prepared against toad S-antigen (a48K)23 at a dilution We also compared the sections with adjacent ones of 1:500. Incubation with the primary antibody was reacted for carbonic anhydrase (CA), which, in hu- for 16 hours at 4°C. Diaminobenzidine served as the mans, can be used to distinguish blue cones (CA-nega- chromogen. For control sections, incubation with tive) from red or green cones (CA-positive).22 a48K was omitted. This technique, using alternate sections, was chosen over a double labeling procedure because the re- MATERIALS AND METHODS agents required for the immunocytochemical staining might have interfered with the enzyme histochemical The human retinas were obtained from a deceased or- labeling and vice versa. gan donor within 1 hour of death and were placed Other nonadjacent sections of human retina were immediately into phosphate-buffered 4% paraformal- cut either tangentially or radially to the plane of the dehyde at 4°C, where they remained for 24 hours. retina and immunolabeled with one of three different They were then stored in 0.1 M phosphate buffer at antibodies to S-antigen, namely, antiserum to toad S- 4°C. Segments of retina were removed and embedded antigen (a48K), MAb A9-C6, or MAb 5c6.47. MAb in glycol methacrylate (GMA; JB-4, Polysciences, A9-C6 and MAb 5c6.47 are monoclonal antibodies Warrington, PA). To preserve enzyme activity, the tis- prepared against bovine S-antigen and are directed sue was embedded directly in GMA without the usual against different sequences of the molecule. Specifi- dehydration in graded alcohols. Instead, the speci- cally, MAb A9-C6 binds to residues 376-386 (VFEE- mens were gently agitated in graded concentrations of FARHNLK) at the carboxyl terminus of S-antigen.24 GMA monomer with benzoyl peroxide catalyst and MAb 5c6.47 binds to residues 39-47 (PVDGWLVD) distilled water, beginning with a 1:1 mixture. Polymer- at the amino terminus.25 This latter sequence is highly ization was carried out at 0°C for 16 hours. Serial conserved among S-antigens from the several species sections 2-/*m thick were then cut tangentially to the examined13"17 and in S-antigen-like proteins of other plane of the retina and floated on distilled water. At tissues.26'27 the level of the photoreceptor inner segments, alter- The specific epitopes targeted by the polyclonal nate sections were picked up on clean glass slides and a48K are not known. However, this antiserum does allowed to dry at room temperature for at least 24 cross react with /3-arrestin, the S-antigen-like protein hours for later immunocytochemical processing. The of the /3-adrenergic receptor system26 (personal com- remaining sections were reacted for CA as described munication from Dr. M. Lohse, 1990). previously.22 This involved floating the cut sections on Retinas from cat eyes were obtained and prepared the surface of solution consisting of 0.00175 M in a similar manner except that the animals were anes- CoSO4, 0.0269 M H2SO4, 0.0117 M KH2PO4, and thetized (as part of another study) and then sacrificed 0.157 M NaHCO3 for 3 to 6 minutes. The sections by perfusion of the left cardiac ventricle, first with were rinsed by floating them on three changes of dis- heparinized saline and then with a mixture of 2.5% tilled water and then floating them on a 0.5% solution buffered glutaraldehyde and 0.5% formaldehyde. of (NH4)2S for 1 minute and rinsing again on water. Some of the cats were briefly anesthetized 24 hours They were mounted on glass slides and lightly counter- before sacrifice, and 0.2 ml of a 1% aqueous solution stained with toluidine blue. For control sections, acet- of lucifer yellow (dilithium) was injected via a 30-gauge azolamide (an inhibitor of carbonic anhydrase) was needle through the pars plana into the vitreous cavity added to the incubating solution at a concentration of of one eye. 10 mmol/1. The method of quantitation of the relative popula- The alternate sections were then immunolabeled tions of photoreceptor types was as described previ- for S-antigen. A sodium ethoxide-benzene solution ously.22 was prepared for use as an etching agent by mixing a The protocols involving animals adhered to the saturated solution of NaOH and absolute ethanol and ARVO Resolution on the Use of Animals in Research adding an equal volume of benzene. The mixture was and were approved by our animal care and use com- allowed to stand in a polyethylene bottle for 2 weeks at mittee. The method of obtaining the human tissue in- room temperature. The sections were then etched by cluded proper consent (by surviving relatives), was ap- placing them in the filtered sodium ethoxide-benzene proved by our institutional human experimentation solution for 30 minutes. They were washed in absolute committee, and complied with the Declaration of Hel- ethanol 4 times for 2 minutes each time, rinsed in 70% sinki.

Downloaded from iovs.arvojournals.org on 09/27/2021 2920 Investigative Ophthalmology & Visual Science, September 1993, Vol. 34, No. 10

RESULTS

Human Retina The staining for all three S-antigen antibodies was found to be somewhat weaker with GMA-embedded retinal segments than for those embedded in paraffin, but GMA was chosen for this study because although the immunolabeling was not as strong as in paraffin, GMA allowed cutting thinner sections with better pres- ervation of morphologic features and because it is an ideal medium for performing enzyme histochemistry for CA. In the adjacent sections with alternating histochemical staining for CA and immunochemical staining with a48K, positive immunolabeling with a48K was observed for all of the rods and ~9% of the cones of the sections of peripheral retina (approximately 20° of visual arc eccentric to the fovea).22 The a48K-positive cones were distributed in a regular array. In peripheral retina, no areas were found in which there were clusters or concentrations of a48K-positive cones; no areas were found where the density of positive cones was less than 9%. A positive CA enzyme histochemical reaction product was noted for ~91% of the cones; the rods and ~9% of the cones were negative. It was possible to identify the same individual photorecep- FIGURE 1. Immunolabeling of photoreceptors in human ret- tors on adjacent sections and thereby to determine ina with a48K (antiserum prepared against S-antigen from rod outer segments of toad) versus carbonic anhydrase histo- whether a given photoreceptor was positive with a48K chemislry in adjacent tangential sections passing through or for CA. Without exception, all photoreceptor inner the rod and cone inner segments of human retina. The same segments labeled positively with either a48K or CA; individual photoreceptors can be identified in both sections none of the photoreceptors was positive for both (arrows). (Top) Positive imniunolabeling with a48K is seen (Fig. 1). for all rods (small discs) and a minority population of cones A similar staining pattern was observed in sections (large discs). (Bottom) Enzyme histochemical reaction for of peripheral retina incubated with MAb A9-C6, carbonic anhydrase (CA). The reaction product is black, and though a direct comparison with CA activity was not the counterstain (toluidine blue) appears gray. The majority performed (Fig. 2, top). All the rods and ••~9% of the of cones show CA activity, whereas the rods and a minority population of cones are negative. When single cones are cones were positive. In contrast to a48K and MAb A9- compared, it can be seen that those positive for S-antigen C6, the monoclonal antibody MAb 5c6.47 stained all are negative for CA and vice versa (bar =10 cones and rods with about equal intensity (Fig. 2, bottom). Radial sections snowed immunolabeling throughout the positively staining photoreceptors for all three antibodies (Fig. 3). CA-positive cones in the central fovea. All the macular Technical problems were encountered with the rods were negative for CA. Because of the high density macular sections. Labeling with a48K, though present, of cones and the relatively weak a48K labeling, at- was weaker than for the peripheral retina. Even so, the tempts at identifying individual photoreceptors on ad- staining pattern was such that the ratio of a48K-posi- jacent sections were unsuccessful. tive to a48K-negative cones (~9%) was roughly con- stant for decreasing foveal eccentricities even though Cat Retinas the absolute density of cones was greatly increased. Near the foveal center, this ratio decreased and only The inner segments of both the rods and the cones of rare a48K-positive cones could be seen. The few rods the cat retina were much smaller in cross section com- in the macular region showed a48K immunolabeling pared to human photoreceptors. By contrast with hu- as well. A reverse ratio of positive to negative cones man retina, all the cones in cat retina appeared to be was observed for CA enzyme activity; i.e., ~91% in the positive for CA (Fig. 4). No CA-negative cones could paramacular area and increasing to virtually 100% be definitely identified in the control retinas. How-

Downloaded from iovs.arvojournals.org on 09/27/2021 Antigenically Distinctive S-Antigens 2921 DISCUSSION This study demonstrates the presence of S-antigen in all photoreceptors of the human and cat retinas. How- ever, different antigenic determinants were identified in the rods and blue cones and in the red and green cones of the human retina, and between the rods and the dominant cone population in the cat retina. Our results are consistent with the recent findings *«« of Nir and Ransom,2829 who also used MAb A9-C6, MAb 5c6.47, and a polyclonal antiserum against bovine S-antigen to examine the localization of S-antigens in cones in mouse and monkey retinas. Nir and Ransom found positive staining for all photoreceptors with all antibodies in the mouse, but in the monkey, the red and green cones were negative for MAb A9-C6 whereas the rods and blue cones were positive (all monkey photoreceptors stained with the polyclonal antiserum and with MAb 5c6.47).

FIGURE 2. Immunolabeling of photoreceptors in human retina with monoclonal antibodies MAb A9-C6 {top panel) versus MAb 5c6.47 (bottom panel). Both panels show tangential sections passing through rod and cone inner segments. (Top) MAb A9-C6 binds to an epitope near the carboxyl terminus of S-antigen; positive reaction product is evident in the rods and a minority population of cones (arrows). (Bot- tom) MAb 5c6.47 binds to a highly conserved sequence near the amino terminus of S-antigen. Note that all photoreceptors show positive staining (bar =10

ever, in the retinas from the eyes that had been injected with lucifer yellow, an occasional cone was found to have a fluorescent yellow staining. These were irregular in distribution, and they were negative for CA (Fig. 5). Each of the three antibodies to S-antigen showed a distinctive staining pattern. By contrast with the labeling pattern in human retina, a48K labeled rods and densely labeled all cat cones rather than labeling a minority population (Fig. 6, top). MAb A9-C6 gave a strong reaction for the rods but a weak cone reaction (Fig. 6, middle). Both rods and cones had equal reactiv- FIGURE 3. Radial sections of human retina immunolabeled ity with MAb 5c6.47 (Fig. 6, bottom). with a48K (top), MAb A9-C6 (middle), and Mab 5c6.47 The results for human and cat retinas are summa- (bottom). The rods and an occasional cone (arrows) are posi- rized in Table 1. Control sections for both CA and tive for a48K and MAb A9-C6, whereas all photoreceptors S-antigen showed no positive staining. have positive staining for MAb 5c6.47 (bar =10 /im).

Downloaded from iovs.arvojournals.org on 09/27/2021 2922 Investigative Ophthalmology & Visual Science, September 1993, Vol. 34, No, 10

tence of blue-sensitive cones in the cat retina.30"39 How- ever, we are aware of only one histologic study showing a subpopulation of cones in the cat retina that take up vital dyes.34 Although we also found that some cones labeled with a vital dye, these cones were in the distinct minority (fewer than 2% of the total cone population), were often distorted, and were irregularly distributed across the retina. Double labeling with CA (Fig. 5, top) suggested that they were CA-negative, but we cannot rule out the possibility that the dye interfered with CA activity—possibly by damaging the cell. We were unable to identify CA-negative cones in retinas of cat eyes that had not been injected with vital dye. Differing from immunocytochemical reactions that may detect only certain epitopes of a protein, the enzyme histochemical technique used to localize CA is not dependent upon the amino acid sequence or ter- tiary structure of the molecule. As such, this technique should identify all the isoenzymes of CA. This would mean that, unlike the molecules of phototransduction

FIGURE 4. Enzyme histochemical staining for CA in cat retina. (Top) Tangential section at the level of the outer segments. No CA-negative cones can be identified (see text). (Bottom) Radial section showing CA activity in the cones, especially the outer segments (arrow). Reaction product is also found in the Mulier's cells (between the nuclei of the photoreceptors), the inner tapetal cells, and the endothelial cells of the choriocapillaries. Toluidine blue counterstain (bar = 10

Even though the labeling pattern was similar in human retina for a48K and MAb A9-O3 (i.e., both stained rods and blue-sensitive cones but not red- or green-sensitive cones), they had a very different distribution in the cat retina, with the MAb A9-C6 staining only rods and the a48K labeling rods and densely labeling all cones, not just a minority population. This sug- gests that a48K and MAb A9-C6 bind preferentially to different regions of the S-antigen protein. Further- more, the positive labeling by MAb 5c6.47 of all photoreceptors of all species examined by us and Nir and Ransom28-29 implies yet another binding site—one that is highly conserved. FIGURE 5. Vital dye (lucifer yellow) staining of cone in cat To our knowledge, this is the first report showing retina. (Top) Enzyme histochemical reaction for CA (black the presence of CA in the cones of cat retina. As such, dots) in an eye that was previously injected with lucifer yel- it is similar to the distribution of CA in humans low. Only one CA-negative cone is found (arrow), which wherein the dominant cone population also contains stains with lucifer yellow. (Bottom) Radial section showing CA and the rods are negative for this enzyme. Previous lucifer yellow uptake in one cone (not reacted for CA) (bar = authors have reported indirect evidence for the exis- 10

Downloaded from iovs.arvojournals.org on 09/27/2021 Antigenically Distinctive S-Antigens 2923

(including S-antigen) that exist in different homolo- TABLE l. Summary of Enzyme Histochemical gous forms in rods and cones,18"21 CA seems to be ab- Staining for Carbonic Anhydrase (CA), and sent from the rods and blue cones. Using a nonspecific Immunocytochemical Staining for Three inhibitor of CA, we previously demonstrated changes 35 36 Antibodies to S-antigen (see text) in Human in the dark-adapted electroretinogram ' and subtle and Cat Photoreceptors changes in color vision37 that could be at least partly explained by this heterogeneous distribution of CA. MAb MAb The physiologic role of S-antigen is not fully un- a48K A9-C6 5c6.47 CA derstood and questions remain as to the precise nature Human Rods * + + - Blue cones -f + + — Red/green cones — — + + Cat Rods + + + - Cones + — + +

of its biochemical interactions within the rod outer segment. However, available data indicate that S-antigen binds preferentially to bleached, phosphorylated rhodopsin,2-38 that light promotes a net movement of S-antigen from the rod inner segment to the rod outer segment (see references 3 and 23 and references therein), and that S-antigen plays a role in deactivating the enzymatic cascade of phototransduction.2'38 Based on these findings, S-antigen is thought to participate in processes underlying inactivation, light adaptation, or both of the photoresponse.2-38 The present immunolabeling data showing the occurrence of S-antigen in cones suggest that, as in rods, the inactivation of phototransduction may depend upon reactions involving S-antigen (arrestin). Previotis psychophysical and electrophysiologic studies show that blue-sensitive cones have response properties more in common with rods than with red- and green-sensitive cones.3940 For example, rods and blue cones are more sensitive to light, have low spatial and temporal resolution, and become desensitized at dimmer background intensities than do red and green cones. The basis for different response properties among the respective photoreceptor types is not known. However, if S-antigen is a major factor in inac- tivating the phototransduction cascade, the differences in structure of the S-antigens in rods and blue cones compared to red and green cones may be im- portant in conferring different quenching characteristics in the two groups. On the other hand, the highly conserved epitope recognized by MAb 5c6.47 could represent an essential sequence either to function or to some essential aspect of cellular homeostasis, such FIGURE 6. Immunolabeling of photoreceptors in cat retina as transport or degradation. It will be of interest in with a48K (top pane!), MAb A9-C6 (middle panel), and MAb 5c6.47 (bottom panel). All panels show tangential sections future studies to determine whether the different S- of cat retina near the junction between the inner and outer antigens in rods and blue cones versus red and green segments. (Top) a48K labels rods and densely labels all cones may be correlated with the expression of func- cones. (Middle) MAb A9-C6 labels only in rods. (Bottom) tional differences among these different photorecep- MAb 5c6.47 labels all photoreceptors (bar - tor types.

Downloaded from iovs.arvojournals.org on 09/27/2021 2924 Investigative Ophthalmology 8c Visual Science, September 1993, Vol. 34, No. 10

Key Words 14. Tsuda M, Syed M, Bugra K, et al. Structural analysis of mouse S-antigen. Gene. 1988; 73:11-20. S-antigen, 48 kDa protein, arrestin, carbonic anhydrase, 15. Abe T, Shinohara T. S-antigen from the rat retina and photoreceptors, blue cones pineal have identical sequences. Exp Eye Res. Acknowledgments 16. Craft CM; Whitmore DM, Donoso LA. Differential The authors thank Dr. Larry A. Donoso (Wills Eye Hospital expression of mRNA and protein encoding retinal Research Division, Philadelphia, PA) for providing monoclo- and pineal S-antigen during the light/dark cycle. / nal antibody MAb A9-C6; Dr. Laura Smith-Lang (Alcon Lab- Neurochem. 1990;55:1461-1463. oratories, Inc., Fort Worth, TX) for providing monoclonal 17. Yamaki K, Tsuda M, Shinohara T. The sequence of antibody MAb 5c6.47; Dr. Paul Brown for providing the cat human retinal S-antigen reveals similarities with al- eyes; and the generous family of the human donor. pha-trahsducm. FEBSLett. 1988; 234:39-43. 18. Nathans J, Thomas D, Hogness DS. Molecular genet- References ics of human color vision: The genes encoding blue, 1. Wacker WB, Donoso LA, Kaslow CM, Yankeelov JA green,.and red pigments. Science. 1986; 232:193-202. Jr, Organisciak DT. Experimental allergic uveitis: Iso- 19. Lerea CL, Somers DE, Hurley JB, Klock IB, Bunt-Mi- lation, characterization, and localization of a soluble lam AH. Identification of specific transducin alpha- uveitopathogenic antigen from bovine retina. J Im- subunit in retinal rod and cone photoreceptors. munol. 1977; 119:1949-1958. Science. 1986; 234:77-80. 2. Palczewski K, Rispoli G, Detwiler PB. The influence of 20. Hurwitz RL, Bunt-Milam AH, Chang ML, Beavo JA. arrestin (48K protein) and rhodopsin kinase on visual cGMP phosphodiesterase in rod and cone outer seg- transduction. Neuron. 1992; 8:117-126. ments of the retina./ Biol Chem. 1985:260:568-573. 3. Mangini NJ, Pepperberg DR. Localization of retinal 21. Gillespie PG, Beavo JA. Characterization of a bovine '48K' (S-antigen) by electron microscopy. JpnJ Oph- cone photoreceptor phosphodiesterase purified by thalmol. 1987; 31:207-217. cyclic GMP-sepharose chromatography. J Biol Chem. 4. Broekhuyse RM, Leunissen JLM, Verkley AJ. Ultra- 1988;263:8133-8144. structural localization of S-antigen in retinal struc- 22. Nork TM, McCormick SA, Chao C-M, Odom JV. Dis- tures. Curr Eye Res. 1985;4:73-76. tribution of carbonic anhydrase among human pho- 5. Broekhuyse RM, Winkens HJ. Photoreceptor cell-spe- toreceptors. Invest Ophthalmol Vis Sci. 1990; 31:1451- cific localization of S-antigen in retina. Curr Eye Res. 1458.. 1985;4:703-706. 23. Mangini NJ, Pepperberg DR. Immunolocalization of 6. Rodrigues M, Hackett J, Wiggert B, et al. Immuno- 48K in rod photoreceptors: Light and ATP increase electron microscopic localization of photoreceptor OS labeling. Invest Ophthalmol Vis Sci. 1988; 29:1221- specific markers in the monkey retina. Curr Eye Res. 1234! 1987; 6:369-380. 24. Knopse V, Donoso LA, Banga JP, et al. Epitope map- 7. McKechnie NM, Al-Mahdawi S, Dutton G, Forrester ping of bovine retinal S-antigen with monoclonal anti- JV. Ultrastructural localization of retinal S-antigen in bodies. Curr Eye Res. 1988; 7:1137-1147. the human retina. Exp Eye Res. 1986;42:479-487. 25. Donoso LA, Gregerson DS, Smith L, et al. S-antigen: 8. Mirshahi M, Faure J-P, Brisson P, et al. S-antigen im- Preparation and characterization of site-specific munoreactivity in retinal rods and cones and pineal monoclonal antibodies. Curr Eye Res. 1990;9:343- photosensitive cells. Biol Cell 1984; 52:195-198. 353. 9. Long K, Philp N, Gery I, Aguirre G. S-antigen in a 26. Lohse JM, Benovic JL, Codina J, Caron MG, Lefko- heredity visual disease: Immunocytochemical and im- witz Rj. /3-Arrestin: A protein that regulates /3-adren- munological studies. Invest Ophthalmol Vis Sci. ergic receptor function. Science. 1990;248:1547- 1988;29:i594-1607. 1550. 10. Van Veen T, Vigh-Teichmann I, Vigh B, Hartwig H-G. 27. Jeansonne NE, Jazwinski SM, Donoso LA. A 48-kDa, Light and electron microscopy of S-antigen-and op- S-antigen-iike phosphoprotein in yeast-replicative sin-immunoreactive photoreceptors in the retina of complex preparations./ Biol Chem. 1991;226:14675- turtle, chicken, and hedgehog. Exp Biol. 1986; 45:1- 14680. 14. 28. Nir I. Immunocytochemical detection of arrestin in 11. Long KO, Aguirre GC. Retinal S-antigen in mamma- cones in mouse and monkey retinas. ARVO Abstracts. lian cone photoreceptors. ARVO Abstracts. Invest Invest Ophthalmol Vis Sci. 1991; 32,1150. Ophthalmol Vis Sci. i987;28:339. 29. Nir I, Ransom N. S-antigen in rods and cones of the 12. Miiller B, Peichl L, DeGrip WJ, Gery I, Korf HW. primate retina: Different labeling patterns are re- Opsin- and S-antigen immunoreactions in photore- vealed with antibodies directed against specific do- ceptors of the tree shrew retina. Invest Ophthalmol Vis mains in the molecule. J Histochem Cytochem. Sci. 1989;30:530-535. 1992;40:343-352. 13. Shinohara T, Dietzschold B, Craft CM, et al. Primary 30. Daw NW, Pearlman AL. Cat colour vision: Evidence and secondary structure of bovine retinal S antigen for more than one cone process. / Physiol. (48-kDa protein). Proc Nat Acad Sci USA. 1987; 84: 1970:211:125-136. 6975-6979. 31. Rabin AR, Mehaffey L III, Berson EL. Blue cone

Downloaded from iovs.arvojournals.org on 09/27/2021 Antigenically Distinctive S-Antigens 2925

function in the retina of the cat. Vision Res. 37. Leys M, Van Slycken S, Nork TM, Odom JV. Acetazol- 1976; 16:799-801. amide affects human medium-and long-wavelength 31. Zrenner E, Gouras P. Blue-sensitive cones of the cat sensitive cones. ARVO Abstracts. Invest Ophthalmol produce a rodlike electroretinogram. Invest Ophthal- VisSci. 1991; 32:1193. mol Vis Sci. 1979; 18:1076-1081. 38. Kiihn H, Hall SW, Wilden U. Light-induced binding 33. Loop MS, Millican CL, Thomas SR. Photopic spectral of 48-kDa protein to photoreceptor membranes is sensitivity of the cat. J Physiol. 1987;382:537-553. highly enhanced by phosphorylation of rhodopsin. 34. de Monasterio FM, Schein SJ, McCrane EP. Staining FEBS Lett. 1984; 176:473-478. of blue-sensitive cones of the macaque retina by a fluorescent dye. Science. 1981; 213:1278-1281. 39. Zrenner E. Electrophysiological characteristics of the 35. Odom JV, Weinstein GW, Nork TM, etal. The effects blue sensitive mechanism: Test of a model of cone in- of acetazolamide on the electroretinogram B-wave teraction under physiological and pathological condi- amplitude in albino rabbits. Documenta Opthalmol. tions. Documenta Ophthalmol Proc Ser. 1982; 33:103- 1989; 72:55-59. 125. 36. Schroeder B, Nork TM, Weinstein GW, Salucci S, 40. Craig AJ, Perry RJ. Difference in absolute sensitivity Odom JV. Effects of acetazolamide: Pigmented rab- and response kinetics between red-blue-and ultravio- bits and humans. ARVO Abstracts. Invest Ophthalmol let-sensitive isolated salamander cones. J Physiol VisSci. 1989; 30:125. (Lond). 1988;407:83P.

Downloaded from iovs.arvojournals.org on 09/27/2021