University of Southern California Dissertations and Theses

Characterization of the physiological ligand and function of a novel opsin RGR from the retinal pigment epithelium

(USC Thesis Other)

Characterization of the physiological ligand and function of a novel opsin RGR from the retinal pigment epithelium

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zceb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHARACTERIZATION OF THE PHYSIOLOGICAL LIGAND AND FUNCTION OF A NOVEL OPSIN RGR FROM THE RETINAL PIGMENT EPITHELIUM by WENSHAN HAO A Dissertation Presented to the Faculty of the Graduate School of The University of Southern California in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Molecular Microbiology and Immunology December 1998 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI N um ber: 99319C.L UMI Microform 9931901 Copyright 1999, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 90007 This dissertation, written by J A I & & I M . M Q . ................................ under the direction of fuS. Dissertation Committee, and approved by aU its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY r-— ■ ? ^ ......... i :Dran~0 / Graduate Studies Date November 13, 1998 DISSERTATION COMMIT TEE j » - i du Chai rperson / ~ T m m Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgments This dissertation for the degree of philosophy is a result of six-year hard work. To be able to discover some genuine secrets of the nature gives me the greatest sense of satisfaction. I am deeply grateful for my advisor, Dr. Henry K W. Fong, to have given me this opportunity to work on this project and guided me through it. I also wish to express my thanks to my committee members, Drs. Michael Lai, Michael Stallcup, Stanley Tahara, Janet Blank, and Judy Gamer for the helpful critiques and suggestions that are essential to the success of the project. I am extremely fortunate to be able to complete my work with a wonderful group of fellow students and colleagues in Dr. Fong's lab. Their sharing and supporting spirits have made the project progress smoothly. Particularly, I owe my thanks to Daiwei Shen and Pu Chen for the time and the energy that they have spent in the tedious process of dissecting thousands of bovine eyes for my experiments. The mouse knockout project has been accomplished through a collaborative effort with Dr. Jeannie Chen and Lanyin Wu in Dr. Robert Maxon's lab. I have benefited enormously from their expertise and generosity throughout the project. Finally, I would like to dedicate the dissertation to my parents and my wife. I thank them for their sustained support during this journey of endeavor. Their encouragement and expectation have always been a major part of my motivation in face of challenges. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Abstract The putative retinal G protein-coupled receptor (RGR) is an abundant opsin that is expressed in the smooth endoplasmic reticulum membranes of the retinal pigment epithelium (RPE) and the Muller cells in the retina (Jiang et aL, 1993; Pandey et ai, 1994). Fractionation of bovine RPE cells by differential centrifugation led to the preparation of RPE microsomal membranes with RGR enriched at least fivefold and 95% of rhodopsin as contaminant in RPE cell collection removed. The ligand-binding property of RGR on RPE microsomal membranes was demonstrated. RGR binds both all-trans-retinal and 11- cis-retinal but with preference to all-trans isomer. The RGR-bound all-trans-retinal is stable during the efficient solubilization of RGR from the microsomal membranes by digitonin. Dighonin-solubOized RGR from bovine RPE was isolated by the means of immimoaffinity chromatography and copurified consistently with a minor 34-kDa protein. The absorption spectrum of RGR revealed endogenous pH-sensitive absorbance in the blue and near-ultraviolet regions of light. Membrane-bound RGR was incubated with exogenously added all-trans-retinal and formed two long-lived pH-dependent photopigments with absorption maxima o f469 ± 2.4 and 370 ± 7.3 nm. The effects of hydrogen ion concentration suggest that the blue and near-UV photopigments are tautomeric forms of RGR, in which an all-trans-retinal Schiffbase is protonated or unprotonated, respectively. The RPE pigment was also demonstrable by its reactivity to hydroxylamine in the dark. The retinaldehyde-RGR conjugate at neutral pH favors the near-UV pigment and is a novel light-absorbing opsin in the vertebrate eye. The endogenous retinaldehyde bound to RGR was extracted from purified RGR in the dark and identified as all-trans-retinal by HPLC analysis. Irradiation of RGR with iii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. monochromatic light at 370 am and 470 nm isomerizes stereospecifically all-trans-retinal into 11-cis-retinal. The stereospecific isomerization by RGR was abolished after denaturation of RGR by incubation with 2% SDS prior to light irradiation. Kinetics studies found that, unlike first order reaction of isomerizing 11-cis-retinal to all-trans- retinal by rhodopsin, the rate of isomerizing all-trans-retinal to 11-cis-retinal slowed gradually over time, suggesting the existence of stable RGR-11-cis-retinal conjugate and a photoreversal isomerization. The quantum efficiency of isomerizing all-trans-retinal by RGR is 0.12. The pKa of protonated all-trans-retinal SchifFbase bond in RGR is 6.5, and extinction coefficients ofR G R ^ and RGR3 7 0 are 6.28 x 104 M'1 cm*1 and 6.61 x 104 M'1 cm *1 , respectively, which were calculated from the spectroscopic data. The photosensitivity ofR G R ^ photopigment is about 1/3 that of rhodopsin, suggesting that RGR is capable of functioning as an isomerase involved in the production of 11-cis-retinal from all-trans-retinaL In order to further investigate the roles of RGR in retinoid metabolism, possible light-initiated signal transduction pathway, and the pathological effects of loss of RGR on photoreceptor degeneration and circadian rhythm disturbance, rgr gene-disrupted mouse is being generated. Mouse RGR cDNA and genomic gene were isolated. A 5.1-kb mouse genomic gene fragment extending from 0.2-kb upstream of exon 1 to 0.8-kb downstream of exon 4 was interrupted with a neo gene in the middle of exon 2 and flanked to a tk gene at 31 end to create the targeting vector. Four embryonic stem cell clones from CJ7 strain and 18 clones from R1 strain were demonstrated to have one of two rgr alleles disrupted, giving targeting frequencies of 0.7% and 3% for CJ7 and Rl, respectively. Mice heterozygous for mutated rgr alleles were generated from ES clone 1C 10. These mice are being mated to produce homozygous mutant mice for phenotypic analysis. iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table of Contents Chapter 1. Introduction to phototransduction and visual cycle.......................................1 1-1. Initiation of vision in the striated neural retina.............................................1 1-2. Rhodopsin and opsin family of proteins....................................................... 2 1-3. Visual signal transduction cascade in photoreceptor cells............................4 1-4. Retinal pigment epithelium as an essential partner of the retina in vision 5 1-5. Visual cycle in vertebrates............................................................................6 1-6. Nonvisual opsins in the eye...........................................................................9 1-7. Rationale of project design.......................................................................... 11 Chapter. 2. Binding retinal ligand by RGR on RPE microsomal membranes................20 2-1. Introduction................................................................................................. 20 2-2. Experimental procedures............................................................................ 22 2-2-1. Materials...................................................................................... 22 2-2-2. Preparation of RPE microsomes..................................................22 2-2-3. Preparation and analysis of [3 H]retinal....................................... 23 2-2-4. Binding of all-trans-[3 H] and ll-cis-[3 H]retinal to RPE microsomal protein.......................................................................24 2-2-5. Immunoprecipitation of PHjretinal-labeled bovine RGR.............25 2-2-6. Detergent solubilization of RGR from RPE microsomal membranes.................................................................................... 25 2-3. Results......................................................................................................... 26 2-3-1. Preparation of RPE microsomal membranes with enriched RGR...............................................................................26 2-3-2. Retinal binding..............................................................................28 v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-3-3. Immunoprecipitation to confirm that RGR bind to retmaldehyde..................................................................................29 2-3-4. Solubilization of RGR from microsomal membranes by detergent........................................................................................29 2-4. Discussion......................................................................................................30 Chapter. 3. Blue and ultraviolet light-absorbing opsin from the retinal pigment epithelium.......................................................................................................46 3-1. Introduction................................................................................................... 46 3-2. Experimental procedures............................................................................... 47 3-2-1. Isolation of bovine RGR.................................................................47 3-2-2. Spectroscopic measurements......................................................... 48 3-2-3. Incubation of membrane-bound RGR with all-trans-retinal...........49 3-2-4. Effect of hydroxylamine on binding of all-trans-[3 H]retinal and light absorbance....................................................................... 49 3-2-5. Illumination of RGR......................................................................50 3-3. Results............................................................................................................50 3-3-1. Purification of RGR from bovine RPE cells...................................50 3-3-2. UV-visible absorption spectra of RGR..........................................51 3-3-3. The effect of all-trans-retinal on the absorption spectrum of RGR.......................................................................... 51 3-3-4. pH-dependent absorption spectra of RGR preincubated with all-trans-retinal.......................................................................52 3-3-5. Hydroxylamine reactivity of RGR................................................. 53 3-3-6. Absorption spectrum of RGR after iHuminatio.............................. 54 3-4. Discussion..................................................................................................... 54 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 4. The endogenous chromophore of RGR-op sin, a retinal photoisomerase from the pigment epithelium........................................................................78 4-1. Introduction................................................................................................78 4-2. Experimental procedures............................................................................80 4-2-1. Materials.................................................................................... 80 4-2-2. Isolation of bovine RGR............................................................ 80 4-2-3. Preparation of rhodopsin............................................................ 81 4-2-4. Extraction of opsin-bound retinal isomers by hydroxylamine derivatization.......................................................81 4-2-5. Analysis of retinaloximes by HPLC........................................... 82 4-2-6. Irradiation of photopigments......................................................83 4-2-7. Rate of photoisomerization of retinal......................................... 83 4-2-8. Calculation of extinction coefficients and pKa of the retinylidene Schiff base of RGR..................................................85 4-2-9. Determination of the quantum efficiency of photoisomerization and photosensitivity of RGR.......................................................85 4-3. Results........................................................................................................ 86 4-3-1. The endogenous chromophore of RGR.....................................86 4-3-2. Photoisomerization of all-trans-retinal bound to RGR..............88 4-3-3. Photosensitivity of RGR............................................................88 4-4. Discussion...................................................................................................90 4-4-1. The chromophore of RGR........................................................ 90 4-4-2. Photoisomerase activity.............................................................92 Chapter 5. Targeted disruption of rgr gene in mouse..................................................107 5-1. Introduction.............................................................................................107 5-2. Experimental procedures...........................................................................109 vii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5-2-1. Embryonic stem (ES) cell lines and cell culture.........................109 5-2-2. Isolation of cDNA and genomic DNA clones........................... 110 5-2-3. Construction of targeting vector for homologous recombination............................................................................. i l l 5-2-4. Transfection and drug selection................................................. I ll 5-2-5. Screening recombinant clones and genotyping mouse progenies by Southern blot......................................................................... 112 5-2-6. Freezing heterozygous rgr-disrupted ES clones and culturing cells for blastocyst injection.........................................................113 5-2-7. Blastocyst injection.......................................................................113 5-3. Results.......................................................................................................114 5-3-1. Characterization of mouse RGR cDNA genomic DNA clones..................................................................114 5-3-2. Homologous recombination of rgr gene in ES cells....................115 5-3-3. Generation of germline-transmitted rgr gene disruption.............. 116 Chapter 6. Conclusion................................................................................................... 127 References.........................................................................................................................133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1 INTRODUCTION TO PHOTOTRANSDUCTION AND VISUAL CYCLE 1-1. Initiation of vision in the striated neural retina. The preeminent sensory information presented to a vertebrate brain comes from vision of the eye. Vision is the primary function of the neural retina in the back of the eye ball (Fig. 1-1). Developmentally, the retina originates from the neural tube and constitutes a part of the central nervous system (Duke-Elder, 1963). During early embryonic development, the neural tube evaginates from the ventrolateral forebrain neuroepithelium to form two ophc vesicles. Under the influence of the regional ectoderm, the anterior portion of the optic vesicle invaginates to give rise to the optic cup. The neuroepithelium on the inner layer of the optic cup proceeds to developing and maturing into the neural retinal, while the outer layer to differentiating into the retinal pigment epithelium. The retina examined under the microscope has a well organized mutilayer structure. This outlook results from the orderly arrangement of six types of visual information processing neurons in the retina that are organized into alternating nuclear layers and plexiform layers (Dowling, 1987). The nuclear layers are composed of packed cell bodies and nuclei of photoreceptors and the other neurons, and the plexiform layers is where synapses are formed among neurons (Fig. 1-2). Two types of photoreceptor cells, rods and cones, in the most posterior position of the retina are the spcialized photosensitive neurons that mediate dim light (scotopic) and daylight (photopic) vision, respectively. Structurally, a photoreceptor cell can be divided into an outer segment and an inner segment which are connected through a thin cilium (Fig. 1-3). The inner segment contains all organelles to maintain cellular metabolism, whereas the outer segment has 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. evolved into a specialized device of a vertical stack of tightly packed membrane disks abundant in visual pigment rhodopsin to fulfill the function of light detection. Light, focused by the lens in the front of the eye, penetrates the transparent vitreous humor and the retina and then strikes visual pigments to initiate the vision pathway. Electrical signals generated in photoreceptor cells are subsequently transmitted through the neuronal circuitry in the retina and the optic nerve to reach the brain. 1-2. Rhodopsin and opsin family of proteins. Rhodopsin is the visual pigment of rod cells that are responsible for detection of dim light. Rhodopsin is comprised of an apoprotein opsin and a 11-cis retinal chromophore attached to the e-amino group of the Lys2 9 6 via protonated Schiff base bond. The first cloning of bovine rhodopisn gene (Nathans and Hogness, 1983) has promoted the isolation of visual and nonvisual opsins from evolutionary divergent species (Hara-Nishimura et aL, 1990; Jiang et aL, 1993; Okano, 1994; Gartner and Towner, 1995; Maden, 1995; Max et al, 1995; Sun et aL, 1997; Provencio et al, 1998). The amino acid sequences of opsins have indicated that they share similar structural features and constitute a subgroup of the large family of the G-protein coupled receptors (Maden, 1995). Like other receptors in the family, opsins possess seven hydrophobic segments that traverse the cellular membrane as helical bundles and form a ligand binding pocket. However, in contrast to these receptors, opsins bind the retinal ligand covalently through the conserved lysine residue located in the middle of seventh helix (Hargrave, 1982; Rao and Oprian, 1996). Bovine rhodopsins are arranged on the disc membranes of rod outer segments with N-terminus residing in external side (intradiscal lumen) and C-terminus in cytoplasmic side (Dratz and Hargrave, 1983; Findlay and Pappin, 1986). There are highly conserved features that have been noticed among opsin family of proteins, including a pair of cysteines in the first and second extracellular loops that are proposed to form a disulfide bond for stabilizing the tertiary 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. structure, an ammo acid motif of Glu/Asp-Arg-Try or a close match at the amino terminus of the second cytosolic loop, and the multiple potential phosphorylation sites of serines and/or threonines in the carboxyl terminus (Fig. 1-4). Using bovine rhodopsin as a model system, the roles of these conserved residues in maintaining and regulating opsin functions have been well established (Khorona, 1992; Ernst et aL, 1995; Acharya and Kamik, 1996; Chen et aL, 1995). Light absorption by visual pigments is the function of the covalently bound 11-cis- retinal chromophore or its dehydro derivative (Findlay and Pappin, 1986; Gartner and Towner, 1995). As isolated compound in organic solvent, retinal or its Schiff base derivatives absorb in the near UV region with X m a x less than 400 nm. However, vertebrate visual pigments from various species have a significant bathochromic-shifted (longer wavelength shift) absorption maxima ranging from 440 nm to 580 nm. Investigation of retinal spectroscopic properties found that increased delocalization of electron distribution and reduced bond alternation of retinal can cause bathochromic shift of X m a x (Honig et al., 1976). The protonation of retinal Schiffbase bond results in red shift of the X m a x from less than 400 nm to 440 nm due to the delocalization induced by the proton. Under the condition of constant pH, the pKa of Schiffbase bond determines the equilibrium of absorptions at 440 nm and below 400 nm. hi visual pigments, the value of pKa can be altered by charged or polar amino acid residues surrounding the Schiffbase bond. Particularly, the existence of a counterion—a negatively-charged glutamate (corresponding to Glu1 1 3 in bovine rhodopsin) or a polar tyrosine—in the proximity of the Schiffbase bond stabilizes the proton and dramatically increases the Schiffbase pKa to a point that completely eliminates the deprotonation under physiological pH, thereby creating visual pigments that can absorb only visible light ranging from 440 nm to 580 nm. Mutation of the counterion in the bovine rhodopin has resulted in a pair of conjugated pigments capable of absorbing at both 380 nm and 490 nm at pH 6.5 (Sakmar et aL, 1989; 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Zhukovsky and Oprian, 1989; Nathans, 1990), clearly indicating a drop of Schiffbase pKa resulting from the loss of counterion. Besides the effect of the counterion, the other charged and polar residues in opsin form a combinatory force to further influence the delocalization that finally determines the unique absorption maxima for individual visual pigments (Sakmar et al, 1991; Merbs and Nathans, 1993; Asenjo et al, 1994). 1-3. Visual signal transduction cascade in photoreceptor ceils. The signal transduction pathway that leads to the generation of neuronal sensory signal in photoreceptor cells has been well characterized and established using bovine rhodopsin as the paradigm (Hofmann and Heck, 1996; Rao and Oprian, 1996). The visual process is initiated with the striking of photons on the retinal chromophore bound to opsin. The energy of photons promotes isomerization of 11-cis-retinal to all-trans configuration. The ligand structure change consequently induces the conformational change of retinal-bound opsin through a series of intermediates with different absorption maxima. In the end, the salt-bridge between the counterion and the protonated Schiffbase bond is broken, and the Schiffbase bond is deprotonated, leading to formation of an activated receptor that absorbs at 380 nm. The activation of receptor causes the cytoplasmic domains of receptor to interact with GDP-bound heterotrimeric transducin (G-protein) and subsequent exchange of GTP for GDP in the Ga subunit. The binding of GTP activates transducin and promotes the dissociation of Ga subunit-GTP complex from the other two subunits. The signal flow is now carried by Ga subunit-GTP complex, which in turn activates the downstream phosphodiesterase (PDE) to hydrolyze the cGMP to S'-GMP. Lowered cGMP levels trigger the closure of cGMP-gated ion channel and generates hyperpolarization across the photoreceptor cell membrane (Fig. 1-5). The excited status of neuronal photoreceptor cells are restored to the ground state and photosensitive level by turning off activation at multiple steps of the signal 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transduction pathway (Hofmann and Heck, 1996). The closure of ion channel reduces the intracellular calcium level, which in turn is sensed as the signal to activate guanylate cyclase that restores the cellular level of cGMP. In the mean time, PDE is shut off by reuniting the active a and ( 3 subunits of the enzyme with two inhibitory y subunits separated during enzyme activation. The ysubunit is freed by deactivation of Ga subunit, an autonomous process due to its intrinsic GTPase activity. Rhodopsin is deactivated through a concerted effort of several proteins. In addition to interacting with transducin, the carboxyl terminus of the activated rhodopsin also binds to rhodopsin kinase, which leads to the phosphorylation at multiple C-terminal serines and threonines. Phosphorylation reduces receptor activity and promotes the binding of arrestin to the phosphorylated C-terminus to quench the activity of the receptor. At this stage, retinol dehydrogenase begins to catalyzes the reduction of photolyzed all-trans-retinal into all- trans-retinol in the ligand binding pocket. The step of reduction is essential to release the arrestin from rhodopsin and expose the phosphorylated C-terminus to dephosphorylation by phosphatase 2A (Hofmann et aL, 1992). In the end, an opsin is renewed and is ready to bind 11-cis-retinal to regenerate photosensitive rhodopsin. 1-4. Retinal pigment epithelium as an essential partner of the retina in vision. Developmentally, the retinal pigment epithelium (RPE) is the only structure in the eye that shares the same origin of neural tube as the retina (Duke-Elder, 1963). The mature RPE is situated between the neural retina and Bruch's membrane of the choroid as a monolayer of polygonal cells. The apical membranes of RPE cells evaginate into microvilli that wrap around the outer segments of cones and rods; the basal membrane infolds and contains fibrils that stick tightly to the Bruch's membrane; and the lateral membrane forms tight junctions among RPE cells that constitute the part of blood-retina barrier (Fig. 1-6). 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The retinal pigment epithelium has multiple physiological functions that are essential for vision (Saari, 1990; Bok, 1993). Structurally, the close encounter between the apical microvilli of RPE cells and the outer segments of photoreceptor cells provides physical protection and an adhesive basis for the retina. The tight junction that contributes to the retina-blood barrier makes RPE cells the sole passage for retina to retrieve nutrients from blood and clear up metabolic waste. The dark melanin granules in RPE cells absorb scattered background light and enhances the sharpness of images. In addition to the supportive roles in vision, RPE cells are also directly involved in the process of vision by phagocytizing the aged outer segments shed at the distal ends of photoreceptor cells, regulating the supply of retinal chromophore during dark and light adaptation, and above all, producing 11-cis-retinal to regenerate of bleached visual pigments. 1-5. Visual cycle in vertebrates. Photoreception by rhodopsin results in apoprotein opsin and photolyzed and reduced all- trans-retinol. Both elements are reudlized for the future vision process by participating in the regeneration of new rhodopsins. While opsin is readily available for regeneration by removing arrestin and dephosphorylation after photoexcitation, the production of visual chromophore 11-cis-retinal from all-trans-retinol requires a more complicated process to accomplish. The process, termed visual cycle, consists of a series of biochemical reactions and retinoid transportation (Saari, 1990) (Fig. 1-7). The visual cycle starts with the reduction of photolyzed all-trans-retinal to all- trans-retinol by all-trans retinol dehydrogenase in red outer segments (Lion et aL, 1975), a reaction which is also essential in quenching the activated photoreceptor (Hofmann et al, 1992). The enzyme is associated with disc membranes in rod outer segments (Zimmerman et aL, 1975), and has been purified in the presence ofNADP (Ishiguro et aL, 1991). The 37-kDa purified enzyme catalyzes specifically the reduction of all-trans-retinal, but not 13- 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. , 11-, and 9-cis compounds. After reduction, it is not clear how the hydrophobic all-trans- retinol is carried through cytoplasm and translocated to the extracellular matrix across the cell membrane. A complete visua1 . cycle requires retinoids to traverse twice the interphotoreceptor matrix, the extracellular matrix surrounded by photoreceptor cells and RPE cells. A I 1 - trans-retinol produced in rod outer segments needs to flow into RPE cells to be isomerized and oxidized into 11-cis-retinal, which then needs to be transported back to rod outer segments to regenerate rhodopsin. This shuttling of both all-trans-retinol and 11-cis-retinal is carried out by interphotoreceptor retinoid-binding protein (IRBP), a 140- kDa glycoprotein that is the major soluble protein in the matrix (Lai et al, 1982; Adler and Martin, 1982). IRBP is capable of binding stoichiometrically two molecules of all-trans- retinol or 11-cis-retinal per molecule (Fong et aL, 1984; Saari et aL, 1985). Upon exposing the eye to light, the amount of all-trans-retinol bound to IRBP is increased, apparently due to the upstream accumulation of all-trans-retinol in rod outer segments (Saari et aL, 1985). When tested in the cultured RPE cells or retina-peeled eye cup, the presence of IRBP promotes the release of 11-cis-retinal from RPE cells up to fivefold higher than other retinoid-binding proteins (Okajima et al, 1989; Carlson and Bok, 1992). Although IRBP can also bind to other isomeric retinoids as well as fatty acids, the affinities for 11-cis-retinal and all-trans-retinol are the highest (Chen and Noy, 1994). In RPE cells, all known enzymatic reactions of the visual cycle take place on microsomal membranes with soluble retinoid carrier proteins shuttling substrates and end products. All-trans-retinol in RPE cells is found to be the endogenous ligand of cellular retinol-binding protein (CRBP) (Saari et aL, 1982), a member of carrier protein family for retinoids and fatty acids from various tissues (Saari, 1990). Transportation of all-trans retinol from IRBP to CRBP is proposed to be driven by CRBPs 100-fold higher affinity for all-trans-retinol and by the input of all-trans-retinol from rod outer segments to 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. interphotoreceptor matrix (Edwards and Adler, 1994), but it is not clear if translocation through plasma membrane is mediated by any specific membrane receptor. The all-trans retinol bound to CRBP is a substrate for esterification to long chain fatty acids catalyzed by enzymes on RPE microsomal membranes, the only known reaction which removes the ligand from CRBP (Saari et aL, 1984). Esterification of all-trans-retinol is catalyzed either by acyl-CoA:retinol acyltransferase (ARAT) in a CoA-dependent way, or by lecithin:retinol acyltransferase (LRAT) in a CoA-independent way (MacDonald and Ong, 1988; Saari and Bredberg, 1989). Esterification not only stores the retinoid and prevents the toxic effects of free all-trans-retinol on RPE cells (Blomhoff et aL, 1990), but also produces the all-trans -retinyl ester that is the substrate for an isomerase that catalyzes the formation of 11-cis-retinol, a key step in the visual cycle (Deigner et al., 1989). Inhibition at the step of esterification completely abolishes the production of 11-cis-retinol from all- trans-retinol (Trehan et aL, 1990). 11-cis-retinol has two possible fates after formation. It could be reesterified by LRAT under the condition that there is more than enough 11-cis- retinal for regeneration of rhodopsin (Bridges, 1976), or it could be oxidized by dehydrogenase to 11-cis-retinal when regeneration of rhodopsin is in demand (Lion et al, 1975). A cellular retinaldekyde-binding protein (CRALBP) can have an influence on this substrate-routing. CRALBP is a soluble protein capable of binding either 11-cis-retinal or 11-cis-retinol with high affinity (Saari et aL, 1982), and protect 11-cis retinoids from light- induced isomerization (Saari and Bredberg, 1987). While 11-cis-retinol is complexed with CRALBP, esterification to fatty acids by LART is reduced and oxidation to retinal by 11- cis-retinol dehydrogenase is stimulated (Saari et al., 1994). The 32-kDa membrane-bound dehydrogenase catalyzes the oxidative conversion between 11-cis-retinol and -retinal, and is probably complexed with other RPE microsomal membrane protein in vivo to regulate the retinoid metabolism in the RPE (Simon et aL, 1995). Oxidized product 11-cis-retinal 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. is presumably still carried by CRALBP, and transported to rod outer segments enroute IRBP. Isomerization of all-trans-retinyl ester to 11-cis-retinol is the key step in the visual cycle. The reaction is catalyzed by isomerohydrolase that hydrolyzes and isomerizes all- trans-retinyl ester to 11-cis-retinol in one step (Rando et al., 1991). The chemical energy released from ester hydrolysis facilitates the endothermic step of isomerization (Deigner et al., 1989). The enzyme is localized to RPE microsomal membranes (Bernstein et al., 1987), and probably interacts nonspecifically with CRALBP to enhance its rate of 11-cis- retinol production under physiological conditions (Winston and Rando, 1997). 1-6. Nonvisual Opsins in the eye. Throughout evolution, opsin-retinal complex has been conserved to mediate visual sensory process in the eyes of living beings. This light-sensing device, through which the living organisms on earth interact with environmental light, is also adopted by microorganisms for multiple biological processes like phototaxis in fungi (Saranak and Foster, 1997) and green algae (Foster et aL, 1984) and proton pump and transmembrane transportation in bacteria (Findlay and Pappin, 1986). hi addition to being used by microorganisms, the device has also been employed by more advanced species for nonvisual activities as well Retinochrome, a cephalopod opsin that is expressed on the myeloid body membranes of photoreceptor inner segments, is the best characterized nonvisual pigment (Hara and Hara, 1987). Retinochrome possesses most of conserved features of the opsin family of proteins, but does not have the counterion at the corresponding position (Hara- Nishimura et aL, 1990). Another sharp contrast to visual pigments is that retinocrhome binds all-trans-retinal ligand at the conserved retinal-binding lysine residue in the seventh transmembrane domain. While being irradiated by light, all-trans-retinal is stereospecifically isomerized into 11-cis-retinal. hi an in vitro reconstituted system 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. including retinochrome, soluble cellular retinoid-binding protein (RALBP), and rhodopsin isolated from squid, 11-cis-retinal produced by retinochrome can regenerate photobleached rhodopsin (Hara and Hara, 1987). Retinochrome-catalyzed trans to cis conversion presents a different retinoid isomerizing mechanism from that catalyzed by vertebrate isomerohydrolase in using substrates and energy sources for the reaction. Interestingly, two vertebrate opsins, peropsin (Sun et aL, 1997) and RPE-retinal G protein-coupled receptor or RGR (Jiang et al, 1993), and one opsin, melanopsin, from Xenopus (Provencio et al., 1998) share closer homology with retinochrome than to vertebrate visual opsins, indicating that opsin ancestral gene might diverged at an early stage of evolution and evolved independently for visual and nonvisual functions. Noticeably, RGR has the closest homology to retinochrome and does not possess a negative charge or polar amino acid at the counterion position either, whereas peropsin and melanopsin have tyrosine, a conserved counterion in invertebrate visual opsins, in the corresponding counterion position. The carboxyl termini of RGR, peropsin, and retinochrome are relatively shorter than that of melanopsin and other visual opsins. Biological functions of RGR, peropsin, and melanopsin remain unknown. Presumably, all three opsins bind retinal chromophores at the lysine residues conserved in the seventh hydrophobic domain though none of the endogenous chromophores have been identified. Another feature shared by the three opsins is that they are all expressed in the RPE in addition to other tissues. However, peropsin is subcellularly localized to apical membrane surrounding photoreceptor outer segments whereas RGR is on the smooth ER membranes (Pandey et aL, 1994). Other than in the RPE, melanopsin is also expressed in dermal melanophores, brain, iris, and retina inner nuclear layer. Melanopsin might be involved in the photic control of circadian rhythm at these locations. Pinopsin, which is isolated from avian pinealocytes, has also been putatively proposed as the photoreceptor initiating a G-protein signal transduction pathway 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. regulating circadian rhythm (Okano et al, 1994; Max et al, 1995). This opsin has the highest identity (43-48%) with vertebrate visual opsins among all non-visual opsins, but is still distantly diverged from them in the phylogenetic tree. Pinopsin that is expressed in eucaryotic cells binds exogenously added 11-cis-retinal to give rise to a spectrum with X max ^70 nm. 1-7. Rationale of project design. The structural similarities among RGR and other opsins have suggested that RGR might be able to bind retinal chromophore at its conserved Lys2 5 5 via a Schiffbase linkage. This idea has been investigated using tritium labeled all-trans-retinal and 11-cis-retinal compounds. A retinal-binding assay was developed for RGR prepared on RPE microsomal membranes. The assay has shown that RGR can bind both forms of retinal with differential affinity. The binding of RGR to retinals in vitro prompted the purification of RGR from bovine eye tissue. It was expected that RGR isolated directly from natural tissue has spectroscopic properties characteristic of an opsin-retinal complex if RGR indeed binds to retinal chromophore in vivo. The result from RGR purification has uncovered unique spectral properties that confirmed the existence of RGR-retinal complex in vivo. We then proceed to identify the geometric structure of the retinal isomer bound to RGR. A reliable and efficient method to extract retinals from purified RGR has been established. The extraction has identified that the endogenous retinal chromophore bound to RGR is all-trans-retinal the same retinal isomer bound to retinochrome. Upon exposure to light, RGR catalyzes the stereospecific isomerization of all-trans-retinal to 11- cis-retinaL Photobiochemical studies of RGR have suggested that RGR might be involved in retinoid trans to cis isomerization in the visual cycle or photic circadian rhythm, or be a 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. photoreceptor initiating signal transduction in RPE cells. To test these various hypotheses and understand the pathological effects of disrupting abundantly expressed rgr gene on photoreceptor degeneration and circadian rhythm disturbance, we have generated rgr gene knockout mouse and are working toward phenotype analysis. 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Incident Light aqueous humor --iris cornea vitreous humor retina choroid optic nerve fovea' Fig. 1-1. Schematic structure of the eye Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Incident Light RETINA ► PHOTORECEPTORS - RPE ► CHOROID Fig. 1-2. Light photomicrograph of the mutilayer structure from bovine retina. GCL: Ganglion Cell Layer IPL: Inner Plexiform Layer INL: Inner Nuclear Layer OPL: Outer Plexiform Layer ONL: Outer Nuclear Layer RPE: Retinal Pigment Epithelium 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. J— Synaptic Terminal- lucleu inner Segm ent itochondri; Cilium O u ter Segm ent— N asma Membran Discs C one Rod Fig. 1-3. Schematic structure of rod and cone cells Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Membrane Intradiskal side Fig. 1-4. Secondary structure of rhodopsin in lipid membrane Conserved features for opsin family of proteins: Glycosylation: Asn2, Asnl5 Disulfide Bond: CysllO, Cysl87 Counterion: Glull3 Conserved Triplet: Glul34-Argl35-Tyrl36 Retinal Binding Site: Lys296 Phosphorylation Sites: Ser334, Thr335, Thr336, Ser338 Thr340, Thr342, Ser343 Palmitoylation Sites: Cys 322, Cys323 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. RPE Disc Membrane Cytoplasm Plasma Membrane y aU-tran^-rctlnol * ^ y R *p A -« ^— R *p VISUAL CYCLE OPSIN 11-cis-rctinal- -R- UGHT R K R* G-GDP CTP CDP \ Fig. 1-5. Visual phototransduction pathway R: rhodospsin; R‘: activated; R*p: phosphorylated R* A: arrestin; RK: rhodopsin kinase PDE: phosphodiesterase; PDE*: activated PDE PDE-I: PDE inhibitory subunit GC: guanylate cyclase +: activating effect; inhibiting effect G-QTP + (OVPZR- POLARJZXnON) PDK.I PDE DE* 5'GMP cGMP DEPOLARIZATION Aoical Microvilli Tight Junction Phagosom e Melanin Basal Membrane Infolding — / n — 0 ; _ ^ ~ '“ " ~ - f - ^ - - p r u c h s Membrane--“--^— : v r s s *^r - ^ r * ~ ~ Choroid Fig. 1-6. Cross section or the retinal pigment epithelium Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V T T R A I k I 1 1 \ V RHO-11 -cis-retinal RHO-all-trans-retinal all-trans RDH RHOall-trans-retinol- Opsin 11 -cis-retinal all-trans-retinol IRBP«11 -cis-retinol IRBP»all-trans-retinol CRALBP«11 -cis-retinal 11 -cis RDH CRBP*all-trans-retinol CRALBP'1 -cis-retinol LRAT or ARAT all-trans-retinyl ester 11 -cis-retinyl ester all-trans-retinol Fig. 1-7. Visual cycle rhodopsin (RHO), retinol dehydrogenase (RDH), interphotoreceptor retinoid-binding protein (IRBP), cellular retinol-binding protein (CRBP), cellular retinaldehyde-binding protein (CRALBP), lecithin:retinol acyl transferase (LR A T), acy 1-C oA : retinol acyltransferase (ARAT), isomerohydrolase (IH), covalent bond (-), carrying without covalent bond (•)- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2. BINDING RETINAL LIGAND BY RGR ON RPE MICROSOMAL MEMBRANES 2-1. Introduction Retinal pigment epithelium is a monolayer of polygonal cells that sits adjacent to the neuronal sensory retina and forms part of the blood-retina barrier through tight junctions among the cells (Miller and Steinberg, 1982). It has essential roles in maintaining the normal physiological function of photoreceptor cells, including offering the physical protection and an adhesive basis, supplying nutrients and carrying away metabolic waste, absorbing the scattered background light to enhance the sharpness of images, isomerizing all-trans retinol into 11-cis-retinal for rhodopsin regeneration, regulating the supply of the photoreceptor chromophores during the dark and light adaptation, and renewing aged photoreceptor cells through phagocytosis of sheded outer segments at the distal ends (Saari, 1990; Bok, 1993). To characterize these RPE physiological functions from a molecular level, a differential hybridization approach was taken to search for RPE specific genes. One of the genes cloned from bovine RPE cDNA library encodes a novel opsin (Jiang et aL, 1993). The deduced amino acid sequence of the putative retinal pigment epithelium G protein-coupled receptor, RGR, shows homology to G protein-coupled receptors with seven transmembrane domains. In particular RGR displays the closest sequence similarity to visual pigment rhodopsin and squid photopigment retinochrome than to the other members of the opsin family, hi addition to seven hydrophobic segments that presumably traverse the membrane, RGR shares other conserved features of the opsin 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. family. Two cysteines, Cys8 8 and Cys1 6 2 , are noted at the first and second putative extracellular domains, where they presumably form an intrachain disulfide bond to stabilize opsin tertiary structure; a short sequence motif of Gly1,2-Argn3-Tyr1 1 4 in the N- terminal side of the second cytoplasmic loop is the close match to a conserved motif of Glu/Asp-Axg-Tyr that has been found in nearly all G protein-coupled receptors and is involved in the activation of G proteins (Ernst et al, 1995); and above all, Lys2 5 5 is localized in the seventh transmembrane domain that corresponds to the retinal ligand binding site of all opsins (Gartner and Towner, 1995; Maden, 1995). Like rhodopsin, RGR is localized to the intracellular compartments rather than on the plasma membrane as are for other G protein-coupled receptors (Pandey et aL, 1994). The e-amino group of conserved lysine in the seventh transmembrane domain of opsins reacts with retinal chromophore to form a SchifFbase bond in characterized photopigments (Hargrave, 1982; Rao and Oprian, 1996). The geometric structure of retinal isomer isolated in the dark is 11-cis-retinal from visual pigments (Findlay and Pappin, 1986; Gartner and Towner, 1995) and all-trans-retinal from nonvisual pigment retinochrome (Hara and Hara, 1987). The structures of retinal isomers bound to opsins are closely associated to the functions of individual photopigments. Photoisomerization of 11-cis-retinal to all-trans-retinal in visual pigment induces the conformational change of opsin that leads to the initiation of phototransduction pathway, whereas the reversal ofphotoisomerization in retinochrome produces the 11-cis-retinal to regenerate the bleached rhodopsin. The overall structural similarities to opsins and the conservation of Lys2 5 5 in RGR have prompted us to propose that RGR, like other characterized opsins, binds retinal ligand at Lys2 5 5 via a SchifF base bond. In order to test this idea, we first fractionated RPE cells and prepared RPE microsomal membranes with enriched RGR. We then developed a retinal-binding assay to confirm that RGR on the membranes 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. indeed binds specifically both all-tans- and 11-cis-retinal with higher affinity for all-trans isomer under the experimental conditions. Finally, in preparation for purifying RGR in the future, detergents were tested in order to determine an appropriate condition to solubilize RGR from RPE microsomal membranes efficiently and functionally. 2-2. Experimental Procedures 2-2-1. Materials. Tritium-labeled vitamin A, [11, 12-3 iyretinol (40-60 Ci/mmol), was purchased from New England Nuclear Research Products (Boston, MA). Purified all-trans-retinal was purchased from Sigma (St. Louis, MO), and 11-cis-retinal was obtained from the National Eye Institute, courtesy of Dr. Rosalie K. Crouch (Medical University of South Carolina, Charleston, SC). Digjtonin was the product of Eastman Kodak (Rochester, NY), Zwittergent and dodecyl-J3-D-maltoside were made by Calbiochem (La Jolla, CA), and CHAPSO was purchased from Sigma (St. Louis, MO) 2-2-2. Preparation of RPE microsomes. The steps of preparing RPE microsomes are summarized in figure 2-1. Postmortem bovine eyes were obtained from a local abattoir. The collection of bovine RPE cells and the preparation of microsomes were carried out under dim yellow light within 2 h of enucleation. Afier excision of the anterior segment and removal of the lens, vitreous, and neural retina, RPE cells were removed by gently scraping the cell monolayer with a spatula. The cells were collected by centrifugation and homogenized in an ice-cold sucrose buffer of 0.25M sucrose, 30 mM Tris-acetate (pH 7.0), and 1 mM DTT using a Dounce glass homogenizer (type B or loose). The homogenate was centrifuged at 200 g at 4 °C to remove nuclei and unbroken cells. The pellet was resuspended, and the 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. homogenization and centrifugation steps were repeated four times. The combined supernatants from the homogenization steps were centrifuged in a Sorvall SS-34 rotor at 10,000 g, 15,000 g, or 20,000 g for 20 min at 4 °C. The supernatant was then further centrifuged to pellet microsomal membranes in a Beckman 70 Ti rotor at 150,000 g for 1 h at 4 °C. 2-2-3. Preparation and analysis of [3 H]retinal. All-trans-[l 1, te^HJRetinal was prepared by chemical oxidation of all-trans-[ 11, 12-3 Hj] retinol (40-60 Ci/mmol), as described previously (Ball et aL, 1948). The all-trans-[ 11,12-3H2]retinol (250 pCi) was oxidized in the presence of 2.4 mg of Mn02 in a hexane solution saturated with retinoic acid. The reaction was performed for 2 h in the dark under a nitrogen atmosphere, and the products of the reaction were filtered through a GF/C glass filter. The all-trans-[l l,12-3 H2 ]retinal was isolated by HPLC and used to prepare 1 l-cis-[ 11, ^^H Jretinal by photoisomerization. The irradiation of purified all- trans-[l l,12-3H2]retinal in ethanol by a fiber optic light source for 5 min resulted in a mixture containing 13-cis-, 11-cis-, 9-cis-, and all-trans-retinals. The all-trans- and 11- cis-[ 11,12-3 Ii,]retinal isomers were isolated and routinely analyzed for purity by normal phase chromatography using a LiChrosorb RT Si60 silica column (4 x 250 mm, 5 pm) (E. Merck, Darmstadt, Germany) and a Bio-Rad HPLC system. The HPLC column was precalibrated using purified isomers of retinal. The standards were eluted with 2% dioxane in hexane as the mobile phase and detected by UV absorbance at 325 nm. The isolation of the labeled retinoids was based on the elution times of the standards. The crystals of isomeric retinal standards were stored at -80 °C in a light-protected container before use. 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-2-4. Binding of all-trans-[3 H] and ll-cis-[3 H]retinal to RPE microsomal protein. RPE microsomes from 10 bovine eyes were prepared as described earlier and resuspended in 1.0 ml of cold 67 mM sodium phosphate (pH 6.5) (protein concentration: 0.85 mg/ml). The membranes were exposed for 1 h to light from a fiber optic light source, and then equal aliquots (0.25 ml) of the membrane suspension were mixed in the dark with 2.5 pi of an ethanolic solution of all-trans-[3 H] or 1 l-cis-[3 H]retinal (each isomer at 1 x 105 or 0.4 x 105 cpm, 50 Ci/mmol). The mixtures were incubated in the dark with gentle agitation for 3 h at room temperature. After incubation, the membranes were collected by centrifugation at 38,500 rpm for 25 min at 4 °C using a Beckman SW60 rotor. The pellet was washed three times and resuspended in 1.0 ml of 67 mM sodium phosphate (pH 6.5). After adjustment of the buffer pH to 8.0 with 1 M NaOH, the membrane suspension was mixed with 38 mg of sodium borohydride (1 M NaBH4 , final concentration) and then immediately irradiated for 5 min by a flood lamp light source. The membranes were recovered again by centrifugation and washed three times with phosphate buffer. The preceding experiment was repeated, except that the sodium borohydride reaction and all subsequent washing steps were carried out completely in the dark. Labeled microsomal proteins were analyzed by fluorography after SDS-PAGE. For fluorography, the 12% polyacrylamide gel was saturated with Enlightening reagent (DuPont-NEN Research Products, Boston, MA), dried, and exposed to Kodak X-omat AR film at -80 °C for a period of 5 days. The entire retinal binding assay was summarized in figure 2-2. 2-2-5. Immunoprecipitation of [3 H]retinal-labeled bovine RGR. After incubation with all-trans-[l 1, n^H Jretinal and treatment with sodium borohydride, as described earlier, the PRE microsomal proteins were solubilized in a 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. solution of 1.2% digitonin and 67 mM sodium phosphate (pH 6.5). For each condition, the digitonin extract (200 |il) was added to 600 fxl of binding buffer (500 mM NaCl and 10 mM sodium phosphate (pH 7.2)) and 150 |xl of immunoaffinity resin. The mixtures were incubated at 4 °C for 2 h in the presence or absence of excess blocking peptide. Two of the samples included a high concentration of bovine RGR peptide: either 100 p . M amino-terminal peptide (AESGTLPTGFGELEVC) or 100 (iM carboxyl-terminal peptide (CLSPQRREHSREQ). The immunobeads were washed in binding buffer containing 0.3% digitonin and recovered by centrifugation. Samples of the immunoprecipitates and original extract were analyzed by fluorography after electrophoresis in 12% SDS-polyacrylamide gel The gel was exposed to Kodak X-omat AR film at -80 °C for a period of 8 weeks. The immunoaffininty resin was conjugated with anti-bovine RGR monoclonal antibody 2F4, which was produced and purified as was described (Shen et al, 1994). The antibody-containing fractions from Mono Q chromatography were pooled and dialyzed three times in 0.1 M MOPS (pH 7.5) at 4 °C. Activated Affi-Gel 10 resin (Bio- Rad, Hercules, CA) was added to the antibody solution, and the suspension was agitated gently for 4 h at 4 °C. After the coupling reaction, the gel was incubated for 1 h in 0.1 M ethanolamine to block the remaining reactive sites. The immunoaffinity gel was then washed with water, equilibrated with binding buffer, and stored at 4 °C until use. 2-2-6. Detergent solubilization of RGR from RPE microsomal membranes. Frozen RPE microsomal membrane pellet from 20 eyes was thawed at room temperature and suspended in 1 ml of 67 mM phosphate buffer (pH 6.5) containing 150 mM NaCl, 1 mM DTT, and 0.5 mM EDTA. To test detergent effect on stability of retinal ligand bound to RGR, the membrane suspension was transfered to a glass vial containing all- trans-PH]retinal (6 x 105 cpm, 50 Ci/mmol) in 10 pi ethanoL The mixture was first 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. incubated for 3 h at room temperature with gentle shaking, and then divided into 16 aliquots. Various amounts of 67 mM phosphate buffer (pH6.5) and 10% (w/v) stock solutions of CHAPSO, digitonin, dodecyl-P-D-maltoside, and zwittergent3-14 were added to the aliquots to make detergent solutions at desired concentrations in a final volume of 60 [il (90 jig protein). Detergent solubilization was proceeded for 1 h at 4 °C with gentle shaking. The unsolubilized residue was separated by centrifugation at 40,000 rpm for IS min at 4 °C using Beckman TSL 55 rotor. The supernatants were reduced with 30 mM NaHB4 and the labeled RGR was visualized as described above. To test detergent efficiency in solubilizing RGR from membranes, the membrane suspension was divided into 16 aliquots, which were then solubilized by the four detergents under the same conditions as above. The solubilized proteins were electrophoresed on a 12% polyacrylamide gel, and RGR was probed using the monoclonal antibody 2F4. 2-3. Results 2-3-1. Preparation of RPE microsomal membranes with enriched RGR. The seven hydrophobic segments of RGR suggest that RGR is an integral membrane protein. By means of electron microscopic immunocytochemistry, RGR was localized uniformly throughout the RPE cell rather than confined to certain sub cellular organelles like plasma membrane, mitochondria, rough endoplasmic reticulum (ER), or the nucleus. Instead, the pattern of RGR subcellular localization was found to be consistent with that of smooth endoplasmic reticulum (Pandey et aL, 1994). Based on this observation, we decided to use differential centrifugation to fractionate RPE cells and prepare RPE microsoms that are comprised of membranes mainly from ER and Golgi organelles. 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Postmortem bovine eyes were purchased from a local abattoir. The eyes were dissected and RPE cells were collected in the cold sucrose buffer of 0.25 M sucrose, 30 mM Tris-acetate (pH 7.0), and 1 mM DTT. RPE cells were washed and then homogenized in the same buffer using Dounce glass homogenizer as was described in the methods. The 29 ml supernatant after centrifugation of homogenate at 200 g was divided into three 9-ml aliquots with 2-ml leftover saved for Western blot. Cell debris after homogenization was pelleted and suspended in ~6 ml sucrose buffer. The first aliquot was centrifuged at 10,000 g for 20 min in a sorvall 34 rotor. The supernatant was transferred to a clean tube, and the pellet was suspended in 9 ml sucrose buffer. The second and third aliquots were centrifuged at 15,000 g and 20,000 g, respectively, for 20 min in the Sorvall 34 rotor, and their supernatants and pellets were treated the same way as for first aliquot. Samples of equal volume from all supernatants and suspended pellets after differential centrifugations were compared for their contents of RGR and rhodopsin by Western blot [Fig. 2-3(A)]. The presence of rhodopsin in the homogenate of RPE cells showed that a certain extent of contamination by rod outer segments is inevitable dining collecting RPE cells from bovine eyes. Densitometry showed that more than 95% of RGR and rhodopsin was kept in homogenate after low speed centrifugation at 200 g, indicating very little agglutination of RGR and rhodopsin to nuclei and cell debris. Centrifugation o f200 g supernatant at high speed (from 10,000 g to 20,000 g) precipitated more than 95% of the rhodopsin, while less than half of RGR was pelleted from low speed supernatant at speed of 10,000 g, around half at 15,000 g, and more than half at 20,000 g. This differential centrifugation of membranes carrying RGR and rhodopsin presented an effective way to remove the majority of ihodopsin contamination from RGR prepared from micromsomal membranes. Centrifugation of the 15,000g supernatant at 150,000g completely sediments RGR in the membrane pellet, which contains dominantly 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. microsomal membrane vesicles. Taking equal amounts of protein from homogenate and RPE microsomal membrane pellet to do Western immunoblot, it was found that RGR was enriched at least fivefold in RPE microsomal membranes while 95% rhodopsin was removed in the meantime [Fig. 2-3(B)]. 2-3-2. Retinal binding. The hypothesis that RGR is a receptor for one or more isomers of retinal was tested by analysis of the covalent binding of 3 H-labeled aQ-trans and 11-cis-retinals to bovine RPE microsomal proteins. A preparation of RPE microsomes was first exposed to light for 1 h in an attempt to photobleach RGR that may be bound endogenously to retinal. Equal aliquots of the membrane suspension were then incubated in the dark with either all- trans-PHJ-or 1 l-cis-[3 H]retinaL After the incubation, sodium borohydride was added to the microsomes, and the membranes were then either kept in the dark or immediately irradiated for 5 min. The labeling of proteins in the microsomes was analyzed by fluorography after SDS-polyacrylamide gel electrophoresis. Although added to a highly complex mixture of membrane proteins, both all- trans- [3 H]- and 1 l-cis-[3 H]retinals covalently bound and specifically labeled a single protein in the RPE microsomes, both under light and in the dark. The labeled protein was 32 kDa conforming to bovine RGR in size. The degree of labeling was consistently greater with all-trans-[3 H]retinal than with the 11-cis isomer. Exposure to light during the reduction step in the presence of sodium borohydride did not significantly afreet the intensity of the bands. Although some rhodopsin was present as a contaminant, the labeled protein in RPE microsomes differed in size from rhodopsin, as detected by Western immunoblot [Fig. 2-4(A)]. The binding of retinal isomers to the 32-kDa PRE protein is presumably through a Schiffbase bond that is heat labile. In order to stabilize the bond for enduring heat 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. treatment, the SchifF base bond was reduced to a heat stable single bond using sodium borohydride prior to boiling for SDS-PAGE analysis (Fig. 2-2). We tested for the presence of SchifFbase bond by omitting the step of sodium borohydride treatment. As predicted, the entire labeling was almost eliminated in the absence of bond reduction [ Fig. 2-4(B)]. 2-3-3. Immunoprecipitation to confirm that RGR bind to retinaldehydes. The binding of retinal to RGR was confirmed by specific immunoprecipitation of the protein from RPE microsomes (Fig. 2-5). A portion of RPE microsomes that was labeled covalently with all-trans- [3 H]retinal was solubilized in digitonin-containing phosphate buffer, and RGR in the soluble extract was immunoprecipitated using a monoclonal antibody directed against the carboxyl terminus of bovine RGR. The labeled 32-kDa protein was immunoprecipitated in a specific manner by the antibody-conjugated Affi-Gel resin. The immunoprecipitation of the protein was completely blocked when the antibody was incubated in the presence of 100 piM carboxyl-terminal peptide, but was not affected when the antibody was incubated with the same concentration of amino-terminal bovine RGR peptide. The immunoprecipitated protein comigrated with RGR, as detected by Western immunoblot. 2-3-4. Solubilization of RGR from microsomal membranes by detergents. Studies on membrane proteins rely heavily on detergents to solubilize them from membranes for subsequent purification and biochemical characterization. Various detergents have differential capabilities in sohiblizing a given membrane protein and preserving biological function. The choice of a suitable detergent for individual membrane proteins is often determined by randomly testing a number of classes of detergents. To prepare for purification of RGR, four detergents, digitonin, (3-dodecyl-D- 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. maltoside, zwittergent 3-14, and CHAPSO, were tested for their abilities to solubilize RGR from RPE microsomal membranes efficiently and functionally. Except for zwittergent 3-14, the other three have been reported to be successfully applied to rhodopsin as well as other G protein-coupled receptors (Fong et aL, 1982; Baneijee et aL, 1995). Each detergent was prepared at four different concentrations to solubilize equal aliquots of RPE microsomal membranes. RGR on the membranes was labeled with all-trans-[3 H]retinal prior to solubilization. The functional solubilization was measured as the stability of aIl-trans-PH]retinal bound to RGR during detergent solubilization, and the solubilization efficiency was measured by Western blot as the amount of RGR freed from RPE microsomal membranes (Fig. 2-6). Among four detergents used, only digitonin and P-dodecyl-D-maltoside preserved all-trans-[3 H]retinal bound to RGR during solubilization of membranes. Therefore zwittergent 3-14 and CHAPSO were excluded from the candidate list even though zwittergent at concentration of 1.5%, 1%, and 0.5% and CHAPSO at 1% were comparable to other two detergents with regard to RGR solubilization efficiency. Digitonin of 0.5% was inefficient in solubilizing RGR, and that was probably the main reason accounting for the low intensity of labeled RGR. Digitonin tested at other three concentrations all gave a high level of both labeled RGR and solubilized RGR. In contrast, while P-dodecyl-D-maltoside tested at all four concentrations gave rise to as high level of solubilized RGR as digitonin, RGR-bound all-trans-retinal was less stable during solubilization by P-dodecyl-D-maltoside, particularly at concentrations of 2% and 3%. Since purification of RGR requires certain range of detergent concentration variation in the buffer, digitonin was clearly the best detergent of choice. 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion Electron microscopic study suggested that RGR was most likely an integral membrane protein on smooth ER in the RPE cells. In a previous report, fractionation of RPE cells by differential centrifugation to prepare smooth ER microsomal membranes was used to determine the subcellular localization of 11-cis-retinol dehydrogenase (Zimmerman, 1976). Following the same protocol except that a Dounce homogenizer was used, we fractionated RPE cells to prepare microsomal membranes. The distribution pattern of RGR among supernatants and pellets after various centrifugation steps was essentially the same as that of 11-cis-retinol dehydrogenase. Based on the result from Western blot, about half of total RGR in the RPE cells was recovered in the RPE micosomal membranes after centrifugation at 15,000 g, in compensation for the loss, RGR was enriched at least fivefold. Another benefit of preparing microsomes is to remove the majority of rhodopsin contained in the collection of RPE cells obtained from dissected bovine eyes. The origin of rhodopsin could be from the contamination of rod outer segments accompanying collection of RPE cells from bovine eyes, or from phagocytizing aged photoreceptor cell fragments by RPE cells. To avoid any potential interference in our assay for RGR on the microsomal membranes, it was to our advantage to remove as much rhodopsin as possible. In summary, RGR-enriched microsomal membrane pellet was a convenient source to test directly the retinal-binding property of RGR and to solubilize RGR with detergent for protein purification and further biochemical characterization. The homology between RGR and visual pigments begat the prediction that covalent binding of all-trans- or 11-cis-retinal to RGR may be demonstrated in an assay that involves the reduction of potential Schiff base linkages by means of sodium borohydride (Bownds, 1967). This hypothesis was tested, and the results were 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consistent with the formation of a Schiif base linkage between the retinaldehyde isomers and a 32-kDa protein in the RPE microsomes and subsequent reduction of the bond to a stable secondary amine. The detection of noncovalently bound ligands would not be expected in this assay involving conventional SDS-PAGE and fluorography. Although the aldehyde group of retinal may react nonspecifically with primary amines, it reacts only very slowly in neutral solution and at low substrate concentrations (Ball et al, 1949). Therefore, the observation that a single RPE microsomal protein was strongly labeled suggests that the protein has a high-affinity reactive site for the retinaldehyde isomers. The labeled protein is similar in size to bovine RGR and comigrates with RGR detected by immunoblot analysis. The labeled protein is unlikely to be rhodopsin, which is still present as a contaminant of the RPE microsomes though majority was removed during preparation of RPE microsomes. It differed in size from rhodopsin, as detected by Western immunoblot, and did not aggregate upon boiling the protein samples prior to gel electrophoresis. Indeed, labeled rhodopsin may not have been detected because of its low amount and its susceptibility to degradation in the presence of sodium borohydride (Wang et al, 1980). The reduction of the Schiff base linkage between the retinylidene chromophore and rhodopsin by means of sodium borohydride is enhanced by light- induced isomerization of the protein-bound 11-cis-retinal to all-trans-retinal (Bownds, 1967). In contrast, the 32-kDa protein was labeled as well in the dark as it was upon illumination. Verification that RGR is the labeled protein was achieved by its specific immunoprecipitation with an antipeptide antibody directed against the carboxyl terminus of bovine RGR. The RGR protein bound both all-trans-retinal and 11-cis-retinal. Since the exclusive function of 11-cis-retinal in vertebrates is its role in vision, RGR is probably directly involved in an aspect of the visual process. The 32-kDa receptor bound all- 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. trans-retinal more extensively than it bound the 11-cis isomer. The preferential labeling of RGR with all-trans-retinal, rather than 11-cis-retinal, is surprising because although opsin binds many isomeric analogues of retinal with little stereoselectivity, it is known to bind the all-trans isomer quite poorly or at a significantly reduced rate (Liu et aL, 1984). Since RGR appears to bind all-trans-retinal preferentially, one of its functions may be to catalyze isomerization of the chromophore by a mechanism that is similar to the invertebrate mechanism of isomerization, which involves retinochrome (Hara & Hara, 1987). Like RGR, retinochrome binds both all-trans- and 11-cis-retinals. It would be improbable that the conformation of the bound retinylidene on RGR is not subject to transformation by light, since directional photoisomerization is a conserved feature of all opsin-related receptors, including bacteriorhodopsin. Further evidence is required to determine whether RGR participates in a pathway for all-trans-retinal to 11-cis-retinal The binding of retinal isomers by RGR strongly suggested the existence of RGR- retinal complex in vivo. Further confirmation and characterization of this complex from bovine eyes require purification of RGR from RPE microsomal membranes. In order to fulfill this goal, the entire purification procedure must not cause the loss of retinal chromophore from RGR. Particularly, detergent of choice needs to be tested for its ability to preserve the integrity of RGR-retinal complex during solubilization of microsomal membranes. We utilized the retinal binding assay to test the stability of RGR-bound all-trans-retinal in four types of detergent solutions. In comparison, digitonin was proven as the best choice in terms of amount of solubilized RGR and the stability of retinal ligand. This result was in contrast to the test on rhodopsin and another G protein-coupled serontonin 5-HT1 A receptor, in which CHAPSO and n- dodecyl-P-D-maltoside demonstrated a higher efficiency in solubilizing receptors than digitonin, and were the top two out of fourteen tested detergents in solubilizing functionally serotonin 5-HT1 A receptor (Fong et aL, 1982; Baneqee et al, 1995). 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Digitonin and dodecyl-maltoside have been widely used in various aspects of rhodopsin functional studies though the latter is sometimes more favorable because this detergent can be replaced or removed easier. Since different detergents can result in various solubilized lipid/protein ratios and differentially extract lipid species from membranes (Baneijee et aL, 1995), the different results from solubilizing three structurally similar integral proteins may reflect that individual proteins are associated with specific classes of lipid species for their functions or they are simply different proteins. 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. <D 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RPE cells ^ cell fractionation Microsomal membranes + [3H ]-retinal Opsin-(CH2)4-NH2 + 0=CH-R 4 Opsin-(CH2 )4 -N+H=CH-R . I NflBH4 reduction Opsia-(CH2)4-N+H2-CH2-R 4 SDS PAGE analysis 4 Autoradiography Fig. 2-2. Siunaiy of retinal binding assay. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2-3. Separation of rhodopsin and RGR during preparation of RPE microsomal membranes. (A) Distribution of rhodopsin and RGR in supernatants and pellets after differential centrifugation. RPE cells from 20 bovine eyes were homogenized using Dounce homogenizer. The pellet from centrifugation of the homogenate at 200g was suspended in 1/S volume of the supernatant, and the supernatant was divided into three equivalent aliquots that were further spun at 10,000g, 15,000g, and 20,000g, respectively. The pellets from high speed centrifugation were suspended in the buffer at the same volume as the supernatants. Equal amount of sample from all pells and supernatants were analyzed by Western blots for the presence of rhodopsin (RHO) and RGR (B) Enrichment of RGR and removal of rhodopsin in RPE microsomal membranes. Supernatant from 15,000g centrifugation was centrifuged at 100,000g to pellet microsomal membranes. Proteins (S pg) from the homogenate (H) and RPE microsomal membranes (M), respectively, were analyzed by Western blots to probe the amount of rhodopsin and RGR 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. m 0 jt. C l' m m m * r tr ;>.i ■ - ,u 'i V f 1 "! /.■> Ns X A y ' W ,'s v * M W s^lsw ^ ✓ « * M 4 f Fig. 2-3(A). Separation of rhodopsin and RGR by differential centrifugation 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RHO RGR H M H M 9 7 .4 - 66.2 - 4 2 .7 - 3 1 .0 - 2 1 . 5 - Fig. 2-3(B). Separation of rhodopsin and RGR by differential centrifugation Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2-4. Binding of all-trans-[3 H]- and 1 l-cis-[3 H]retinals to protein in RPE microsomes. Bovine RPE microsomes (0.2 mg of protein/0.25 ml) were incubated with either all- trans-[3 H]- or 1 l-cis-[3 H]retinal, with each isomer at 100,000 and 40,000 cpm and specific activity 46 Ci/mmoL The samples were subsequently mixed with 1 M sodium borohydride and either (A) irradiated immediately with a fiber optic light source or (B) kept completely in the dark. The reaction products were analyzed by SDS- polyacrylamide gel electrophoresis and fhiorography. (C) In another independent experiment, all-trans-[3 H]retinal at 100,000 cpm was used to label RGR, RPE microsomes were then either treated with (+) or without (-) 1 M sodium borohydride followed by light irradiation and gel electrophoresis. 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. retinal isomer cpm xlO” 3} A 32 kD -* } a\\-trans 11 -cis B 32 kD‘ 100 4 0 100 C. Reduction of Schiff base bond by NaBhk NaBH4 + 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2-5. Immunoprecipitation of RGR from RPE microsomes labeled with all-trans-[3 H] retinal. An extract of RPE microsomal proteins that were labeled with all-trans-[3 H]retinal was prepared by solubilization of the membranes in a solution of 1.2% digitonin. RGR was immunoprecipitated from equal aliquots (0.02 mg of protein) of the extract in the presence or abesence of amino-terminal (NHj) or carboxyl-terminal (COOH) peptide. The immunoprecipitated samples were analyzed by SDS-polyacrylamide gel electrophoresis and fhiorography. The immunoprecipitated protein comigrated with RGR, as detected by Wester blot. 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97.4- 66.2- 42.7- 31.0- 21.5- ^ Immunoprecipitation 3 (+peptide) o i zp 6 ± s g X O e S I ? v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 2-6. Solubilization of RGR by detergents. RPE microsomes in 67 mM sodium phosphate (pH 6.S) were incubated with all-trans-[ 3 H]retinal (300,000 cpm, S O Ci/mmol) in the dark for 3 h at room temperature. Equal aliquots of reaction products (90 pg protein) were mixed with four detergents at various concentrations in the volume of S O |iL. Solubilized proteins were separated from residues by centrifugation at 100,000g and treated with 1 M sodium borohydride. The samples were subject to 12% SDS-polyacrylamide gel electrophoresis. RGR in the gels are visualized by either (A) autoradiogram or (B) Western blot. Lanes 1-4: dighonin at 3%, 2%, !.2%, 0.5%; 4-8: Zwittergent3-14 at 1.5%, 1%, 0.5%, 0.2%; 9-12: dodecyl-P- D-maltoside at 3%, 2%, 1%, 0.5%; 13-16: CHAPSO at 4%, 3%, 2%, 1%. 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. all-trans-[ H]retinal binding Digitonin Zwittergent Dodecylmaltoside CHAPSO Immunoblot Digitonin Zwittergent Dodecylmaltoside CHAPSO 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER. 3 BLUE AND ULTRAVIOLET LIGHT-ABSORBING OPSIN FROM THE RETINAL PIGMENT EPITHELIUM 3-1. Introduction Retinal photoreceptors contain ihodopsin or cone pigments that consist of 11-cis-retinal (or 1 l-cis-3,4-dehydroretinal) as chromophore and a specific opsin as protein (Wald, 1968; Dratz & Hargrave, 1983; Nathans, 1987). Upon illumination all visual pigments activate G proteins to initiate the process of photosensoiy transduction (Stryer, 1988; Pak & Shortridge, 1991). Recently, identification of nonvisual opsins has revealed additional evolutionary branches in the vertebrate opsin gene family. Visual pigment homologues have been detected in the chicken pineal gland (Okano et al., 1994; Max et al., 199S) and in mammalian retinal pigment epithelium (RPE) and Muller cells (Jiang et al, 1993). The RPE opsin (RPE retinal G protein-coupled receptor, or RGR) and pineal gland opsin (pinopsin, or P-opsin) are approximately 25% and 45% identical, respectively, in amino acid sequence to the vertebrate visual pigments. RGR is also related distantly to retinochrome and shares amino acid sequence similarity with the invertebrate photoisomerase (Hara-Nishimura et al, 1990). The RPE opsin has been shown to be membrane-bound and located primarily within the cytoplasm of RPE and Muller cells (Pandey et aL, 1994). RGR contains a conserved lysine residue that is homologous with the retinaldehyde attachment site in the visual pigments and is able to bind both all-trans and 11-cis isomers of retinal in vitro (Shen et aL, 1994). The ability of RGR to bind retinal in vitro and its status as a member of the opsin family 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. support the prediction that RGR mediates its biological function through photoreception. Given the significant divergence between the amino acid sequences of RGR and the visual pigments and the absence of a conserved Schiff base counterion in RGR, it is probable that the retinal-bound RPE opsin possesses a unique absorption spectrum as a result of specific protein-chromophore interactions. In this study, we demonstrate photosensitivity of RGR, obtain an absorption spectrum for the purified protein, and analyze some of the spectral properties of bovine RGR that is bound to all-trans-retinaL 3-2. Experimental Procedures 3-2-1. Isolation of bovine RGR. After excision of anterior structures and the neural retina, RPE cells were removed from bovine eyes by gently scraping the epithelial cell monolayer with a spatula. Microsomal membranes from RPE cells were isolated under red or dim yellow light within four hours of enucleation, as described previously (Shen et aL, 1994). The subsequent purification of RGR was conducted under red light or in the dark. The membranes were twice extracted for 1 h at 4 °C with 1.2% digitonin (Eastman Kodak Co., Rochester, NY) in 10 mM sodium phosphate buffer, pH 6.5, containing 150 mM NaCl and 0.5 mM EDTA. After centrifugation of the extract at 100,000 g for 20 min, the supernatant was mixed for 2 h at 4 °C with Affi-Gel 10 resin (Bio-Rad, Hercules, CA) conjugated to anti-bovine RGR monoclonal antibody 2F4 (Shen et al, 1994). The immunoaffinity resin was transferred to a column for washing with 25 bed volumes of 10 mM sodium phosphate buffer, pH 6.5, containing 0.1% digitonin, 150 mM NaCl and 0.5 mM EDTA. The column was then loaded 10 times with 0.5 bed volumes of wash buffer containing 100 mM bovine RGR carboxyl terminal peptide (CLSPQRREHSREQ). The ehiates were pooled and concentrated approximately fourfold using a Centricon-3 concentrator (Amicon, Inc., 47 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Beverly, MA). Gel electrophoresis and immunoblot of proteins were performed, as described previously (Jiang et aL, 1993; Pandey et aL, 1994). The concentration of purified RGR was measured using the Bio-Rad Protein Assay reagents, following removal of the elution peptide from RGR by repeated dilution and ultrafiltration through Centricon-3 concentrator tubes. The measurements were corrected for background using a control volume of the elution buffer that was concentrated and treated identically as the RGR sample. Bovine serum albumin was used as the standard for protein quantitation. The abundance of RGR in the RPE cell homogenate and microsomal membranes was determined from the relative signal intensities on Western blots, which contained varying amounts of each fraction. The band intensities were quantitated using an Ultroscan XL laser densitometer (Pharmacia LKB, Bromma, Sweden). 3-2-2. Spectroscopic measurements. The UV-visible absorption spectra of purified RGR were recorded with a Hitachi U-3000 scanning spectrophotometer on samples of 1.0-cm path length at room temperature. The reference sample consisted of the elution buffer-10 mM sodium phosphate (pH 6.5), 150 mM NaCl, 0.5 mM EDTA, 0.1% digitonin, and 100 mM carboxyl terminal peptide. The reference showed no light absorbance in the visible region; however, digitonin and the peptide were capable of absorbing UV light. This absorbance by the buffer components introduced some distortion in the far-UV region of the absorption spectra of RGR; thus, the region below 300 nm was omitted from the spectra, and absorbance ratios were not used as a criterion of purity. The pH of the sample was raised by addition of 1 M Na2HP03, or lowered by addition of 1 M NaH^POj, 12 mM HC1 or 5 M trichloroacetic acid (TCA). The spectra were plotted from data files using Cricket Graph IH software. 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3-2-3. Incubation of membrane-bound RGR with all-trans-retinal. Ail-trans-retinal was purchased from Sigma (St. Louis, MO) and analyzed by normal phase chromatography using a LiChrosorb RT Si60 silica coliunn (4 x 250 mm, 5 mm) (£. Merck, Darmstadt, Germany) and Bio-Rad HPLC system Equivalent suspensions of RPE microsomes in 10 mM sodium phosphate buffer, pH 6.5, containing 150 mM NaCl and 0.5 mM EDTA, were incubated in the dark with or without 50 mM all-trans-retinal for 2 h at room temperature. The membranes were recovered by centrifugation at 100,000 g for 20 min. The microsomal proteins were then solubilized with digitonin solution, and RGR was purified by an immunoaffinity procedure, as described earlier. 3-2-4. Effect of hydroxylamine on binding of aU-trans-[3H]retinal and light absorbance. All-trans[l 1,12-3H]-retinal was prepared and analyzed, as described previously (Shen et aL, 1994). Equal aliquots of a suspension of RPE microsomes in 67 mM sodium phosphate (pH 6.5) were incubated in the dark for 2.5 h at room temperature with purified all-trans-PH]retinal (1 x 105 cpm, 50 Ci/mmol). Hydroxylamine (1 M NHjOH in H^O, pH 6.5) was then added to one sample to a final concentration of 0.25 M, and the incubation was continued for 30 min. After incubation the membranes were collected by centrifugation at 40,000 rpm for 25 min at 4 °C using a Beckman SW55 rotor. The pellet was washed three times and resuspended in 1.0 ml of 67 mM sodium phosphate, pH 6.5. After adjustment of the buffer pH to 8.0 with 1 M NaOH, the membrane suspension was mixed with 38 mg sodium borohydride (1 MNaBH4, final concentration), and subsequent steps were carried out in the light. The membranes were recovered again by centrifugation, and labeled microsomal proteins were analyzed by fluorography after SDS- PAGE. For fluorography, the 12% polyacrylamide gel was saturated with Enlightning 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reagent (Dupont NEN Research Products, Boston, MA), dried, and exposed to Kodak X- omat AR film at -80 °C. Hydroxylamine reactivity was also determined by measuring the absorption spectra of RGR. RPE microsomes were incubated with S O mM nonradioactive all-trans-retinal, as described earlier, and RGR was purified at pH 6.5. The pH of the sample was adjusted to 4.2, and the absorption spectrum was obtained. Hydroxylamine was then added to a final concentration of 80 mM, and the absorption spectrum of the sample was obtained at various time intervals from 0.S to 40 min at room temperature. No further change in the absorption spectrum occurred after 30 min. Smoothed difference spectra were computed by applying a 7-point moving average. 3-2-5. Illumination of RGR. Membrane-bound RGR was incubated with all-trans-retinal and isolated in the dark, as described earlier. The purified protein was then irradiated at room temperature by a 30- watt fiber optic light source for 5 min at a distance of 10 cm. UV-visible absorption spectra were determined before and immediately after illumination. 3-3. Results 3-3-1. Purification of RGR from bovine RPE cells. Bovine RGR was isolated from the microsomal membranes of RPE cells under red or dim yellow light. The RPE microsomes were enriched with RGR, but remained highly contaminated with pigmented material. The membranous proteins were extracted in 1.2% digitonin solution at pH 6.5, and solubilized RGR was purified by means of inmnmoafSnity chromatography. The purified fraction contained a major protein band that was approximately 31 kDa and reacted specifically to a bovine RGR antibody (Fig. 3- 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I). Minor amounts of a 34-kDa protein and two other proteins between 70 and 76 kDa were also isolated. In a typical purification, approximately 26 mg RGR was isolated from RPE cells from 20 fresh bovine eyes (Table 3-1). The data shown in table 3-1 indicate that RGR represents about 3% of the initial RPE cell homogenate. 3-3-2. UV-visible absorption spectra of RGR. The absorption spectrum of purified RGR showed endogenous pH-sensitive absorbance in the blue and near-ultraviolet (UV) regions of light [Fig. 3-2(A)]. When the pH was raised from 6.5 to 8.0, light absorbance by the native protein between 460 to 490 nm was nearly abolished. When the pH was lowered from 6.5 to 4.2, the absorbance in this region of blue light increased noticeably. The large peak in the UV region « 330 nm) was relatively unaffected by the pH. Nevertheless, some acid-sensitive changes in extinction could be detected within the near-UV region. Generally, increases in the hydrogen ion concentration reduced light absorbance in the near-UV region and vice versa. The pH effect in the near-UV region can be illustrated by the difference spectrum that was obtained by subtraction of the spectrum at pH 8.0 from that at pH 4.2 [Fig. 3-2(B)]. The difference spectrum shows a pH-sensitive absorption peak at approximately 364 nm, as well as a peak at 466 nm 3-3-3. The effect of all-trans-retinal on the absorption spectrum of RGR. hi an effort to improve the recovery and analysis of a photoreceptive RGR, the microsomes were incubated in the dark with exogenously added all-trans-retinal prior to solubilization and purification of the protein. RGR from the treated microsomes displayed a conspicuous and reproducible absorption peak between the wavelengths o f460 to 480 nm that was substantially enhanced in comparison to that of control RGR from RPE microsomes that had not been preincubated with all-trans-retinal [Fig. 3-3(A)]. The 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. difference spectrum, which is dependent on exogenously added all-trans-retinal, showed two absorption peaks, one in the visible region with 466 nm and another in the near- UV region with A ^ 375 nm [Fig. 3-3(B)]. The broad absorption peak in the blue visible region resembles that of all-trans- retinal (A^g = 389 nm in aqueous digitonin); however, its absorption maximum is shifted to a longer wavelength, 466 nm [Fig. 3-3(B)]. The peak at 375 nm is also shifted slightly from the A ^ of all-trans-retinal, and in contrast to the free retinal, it was pH-sensitive. The all-trans-retinal difference spectrum was determined in three separate experiments, which indicated absorption maxima o f469 ± 2.4 nm and 370 ± 7.3 nm in the blue and near-UV regions, respectively. The retinal-dependent difference spectra did not exhibit the peak at 330 nm. 3-3-4. pH-dependent absorption spectra of RGR preincubated with all-trans-retinal. The A ^ at 466 nm in the difference spectra shown in figure 3-3(B) is somewhat greater than that expected for a protonated all-trans-retinal Schiflf base salt, which absorbs maximally at about 440 nm (Pitt et al, 1955). It is possible that the shift in maximal absorbance to 466 nm is produced through chromophore-protein interactions of a protonated all-trans-retinylidene bound to RGR. This hypothesis was tested by measuring the effect of pH and acid denaturation on light absorbance by RGR that was isolated from microsomes treated with all-trans-retinal At pH 8.0 the absorbance near 466 nm decreased greatly and appeared as a relatively small shoulder [Fig. 3-4(A)]. At pH 5.2 and 4.2 the absorbance increased with each addition of acid Concomitantly, opposite changes in extinction were observed with respect to pH for the apparent absorbance in the near- UV region. The pH-dependent curves cross at an isosbestic point and indicate a pKa close to 6.5 for an acid-binding group. 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Subtraction of the spectrum at pH 8.0 from that at pH 4.2 gives a pH difference spectrum that suggests an interconversion of two pigments with absorption peaks at approximately 466 nm and 363 nm [Fig. 3-4(B)]. The pH difference spectra of retinal- treated and untreated RGR show virtually identical absorption peaks [Figures 3-2(B) & 3- 4(B)], except that the absorbance by retinal-treated RGR is about twofold higher for the same amount o f protein. Acid denaturing conditions greatly affected the difference spectrum of RGR that results from the presence or absence of all-trans-retinal (Fig. 3-5). RGR was purified, and its absorption spectrum was measured in 0.1 M TCA. Apparently, RGR in digitonin micelle is not immediately precipitated in 0.1 M TCA at room temperature, and the spectra can be obtained. The shape of the difference spectrum in figure 3-5(B) varied considerably from that seen in figure 3-3(B). The high acidity resulted in elimination of the two absorption peaks at 466 nm and 375 nm and the appearance of a new absorption peak with 450 nm Purified RGR was also denatured with SDS, and the all-trans-retinal difference spectrum was determined at pH 6.5 and pH 12. In the absence of SDS, the two photopigments at approximately 468 and 375 nm were identified [Fig. 3-6(A)]. Treatment of RGR with 2% SDS resulted in loss of the 468-nm absorption peak and the emergence of a new peak at 442 nm [Fig. 3-6(B)]. The UV peak also was reduced at pH 6.5, but at pH 12 it was the predominant absorption peak (X ^ « 372 nm), following the addition of NaOH 3-3-5. Hydroxylamine reactivity of RGR. The stability of RGR against hydroxylamine was tested by binding of all-trans-PH] retinal and by absorbance. A complex mixture of RPE microsomal proteins was incubated in the dark with all-trans-PH]retinal before the addition of 0.25 M hydroxylamine, or a control 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. buffer. Sodium borohydride was added last to the microsomes, and the labeling of proteins was analyzed by gel electrophoresis and fluorography. Hydroxylamine abolished completely the specific binding of all-trans-PHJretinal to a 31-kDa protein [Fig. 3-7(A)]. The single labeled protein was identified previously as RGR by specific immunoprecipitation (Shen et al, 1994). Decomposition of RGR at pH 4.2 is shown via absorbance measurements [Fig. 3- 7(B)], that were made at various times after the addition of hydroxylamine. The decay of the RPE pigment was about 50% complete after 10 minutes and unchanged after 30 minutes. The absorption spectrum at 36 min was subtracted from each spectrum obtained at other time points and before the addition of hydroxylamine. The difference spectra indicate the loss of a pigment (A ^ » 470 nm) and formation of a new pigment « 364 nm), which is consistent with retinaldehyde oxime. 3-3-6. Absorption spectrum of RGR after illumination. The absorption spectrum of RGR was determined before and after illumination (Fig. 3-8). Irradiation did not lead to substantial bleaching of the blue-absorbing pigment; however, it did cause a decrease in the absorbance in the near-UV region. The difference spectrum showed an absorption peak at approximately 368 nm. The effect of light in the near-UV region was highly reproducible and occurred without treatment of RGR with exogenous all-trans-retinal (result not shown). 3-4. Discussion We have isolated and characterized some of the absorbance properties of a novel pigment from RPE cells. The shape of the absorption peaks and biochemical properties of this pigment are consistent with those of a retinylidene Schiflf base chromophore and reveal the 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. existence of two pH-dependent species with absorption maxima at 469 ± 2.4 and 370 ± 7.3 nm. The results provide strong evidence that RGR is conjugated in vivo to an endogenous retinaldehyde chromophore. The RPE opsin differs from the visual pigments in its facile ability to form a stable photopigment by recombination with the all-trans isomer. Its absorption maxima and pH sensitivity are quite distinct from that of rhodopsin. A few minor proteins were co-purified with RGR In particular, a distinct 34-kDa protein was consistently visible by protein staining. None of the minor protein bands were detected on immunoblots with an RGR antibody, even when the immunoblots were overdeveloped; nor were these microsomal proteins labeled covalently with [3 H]retinals. It is possible that one or more of the co-isolated proteins forms a complex or associates functionally with RGR, although the identity of these proteins is not yet known. Interestingly, the 34-kDa protein approximates the size of cellular retinaldehyde binding protein (Crabb et aL, 1988) and G protein a and 3 subunits. Purified RGR displayed a pH indicator property, not unlike that of retinochrome (Hara & Hara, 1965), squid metarhodopsins (Hubbard & St. George, 1958), or counterion mutants of bovine rhodopsin (Zhukovsky & Oprian, 1989; Sakmar et al, 1989; Nathans, 1990; Sakmar et aL, 1991). High pH favors the formation of a UV-absorbing pigment; low pH, its conversion into a blue-absorbing pigment. This pH dependence is the inverse of that reported for bovine metarhodopsins, Mil (X ^ = 380 nm) and MI (X ^ = 480 nm), which are favored by lower and higher pH, respectively (Matthews et al, 1963), and it indicates that the RPE pigment is not a rhodopsin contaminant, which would be the most likely contaminating visual pigment. Although the structure of the endogenous chromophore is unknown, RGR binds all-trans-retinal readily. The binding of all-trans-retinal generates a pair of long-lived pigments comprised of blue and UV light-absorbing conjugates. The two absorption 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. maxima in the all-trans-retinal difference spectrum [Fig. 3-3(B)] agree closely with those calculated from the pH difference spectra [Figures 3-2(B) & 3-4(B)]. The similarity in measurements strongly supports the notion that the blue and UV light-absorbing pigments are tautomeric forms of RGR, in which the retinylidene Schiff base is protonated or unprotonated, respectively. It is known that protonation of a retinylidene Schiff base leads to an increase in the absorption maximum of the pigment from about 370 nm to a wavelength of440 nm or greater (Pitt et al, 1955; Loppnow et aL, 1989). The pH-dependent equilibrium between two forms of RGR may result from the absence of an endogenous counterion for the protonated Schiff base. The protonated group on RGR has a pKa at about 6.5, which is far lower than the pKa of the retinylidene Schiffbase in rhodopsin and suggests that the Schiffbase nitrogen in RGR is not located near a stabilizing counterion. Glutamate is not conserved in RGR at the position corresponding to the counterion residue of rhodopsin (Zhukovsky & Oprian, 1989; Sakmar et aL, 1989; Nathans, 1990; Sakmar et aL, 1991); instead, histidine is found in the homologous position in bovine and human RGR (Jiang et al, 1993; Shen et aL, 1994). Upon acid denaturation, RGR absorbed maximally at 450 nm. This result supports the proposition that the all-trans-retinal is bound to the protein. The opsin shift in blue- absorbing RGR is about 19 nm (900 cm-1). Similar to the treatment of RGR with acid or exogenous all-trans-retinal, the incubation of RGR with hydroxylamine also demonstrated a blue-absorbing pigment (\nax ~ 470 nm) in the difference spectrum. The hydroxylamine-dependent loss of PH]retinal binding to a specific 31-kDa protein corresponded directly to the disappearance of the 470-nm absorption peak. RGR was unstable in the dark against hydroxylamine at concentrations as low as 80 mM. The high reactivity with hydroxylamine suggests that the Schiffbase bond is readily accessible to small molecules. Since the opsin shift of blue-absorbing RGR is weak, the chromophore itself may be 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bound near the surface of RGR, and consequently, it may be particularly accessible to other retinaldehyde-bindiag proteins in the RPE or Muller cells. Hydroxylamine did not affect the absorption peak at 330 nm, which is present in the absolute spectrum of purified RGR. As indicated by difference spectra, the 330-nm peak was not pH-sensitive and was not generated by incubation with exogenous all-trans- retinal. The 330-nm peak may be derived from another RPE pigment that is co-purified with RGR or is a residual impurity. The identity and biochemical properties of the 330-nm component remain to be determined. There is an absence of electrophysiological or microspectrophotometrical data on the absorption spectrum of RGR, and the biological function of the RPE opsin is not yet understood. Thus, it is unknown whether the absorption spectrum of RGR that we observe corresponds to the native protein structure or not. The presence of retinal on the protein is unlikely to be a simple artifact of the preparative procedure. During its isolation, RGR was not exposed to an extreme pH condition, which would enhance the production of an indicator yellow compound (Morton & Pitt, 1955). When crude RPE microsomes were incubated with [3 H]retinal, RGR was the only detectable protein that bound the radiolabeled isomer. This observation argues that the chromophore is bound specifically to RGR, insofar as it is unlikely that random Schiffbase linkages could form exclusively in only one of the numerous microsomal proteins. Similar conditions were used for incubation of RGR with unlabeled all-trans-retinal. Even without treatment of RGR with all-trans-retinal in excess, absorbance by the pH-sensitive photopigment was about 50% of the absorbance of retinal-treated RGR, and the absorption maxima from the pH difference spectra were not materially altered by incubation with exogenous retinal Clearly, additional work must be done to affirm and elucidate the biophysical basis for the absorption spectrum of RGR. Preliminary evidence indicates that the purified RGR apoprotein retains the ability to bind all-trans-retinal and regenerate the observed pigments 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in vitro. Other tests for RGR with a native structure may include circular dichroism spectroscopy, an efficient photoresponse with distinctive dark reactions, and a functional assay. Since there are two forms of RGR at neutral pH, analysis of the effects of irradiation may be complex. Illumination did not lead to bleaching of the blue-adsorbing pigment. As for many invertebrate visual pigments, such as squid and Drosophila ihodopsins (Hubbard & St. George, 19S8; Hillman et aL, 1983), retinal may not dissociate readily from RGR in the presence of light. In the near-UV region, a decrease in extinction was always observed after irradiation. This change in absorption may indicate the formation of a new species of photopigment with a lower extinction coefficient. With purified RGR it will be possible to investigate the light reactions of the RPE opsin in greater detail. From the results of this paper, we conclude that RGR binds retinal in vivo and propose that its biological function in the eye entails photoisomerization of the chromophore by blue or UV light. At neutral pH, the naturally occurring UV form would predominate, and thus, RGR may have a physiological role in UV light reception. 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table I: Purification of RGR from bovine RPE cells Step Protein RGR/protein Purification fold mg mg • mg -1 Homogenate 41 0.030 fl Microsomes 3.3 0.11* 3.7 Immunoaffinity 0.026 ~1 33 a The am ount of RGR in crude extracts was determ ined by m eans of im m unoblots using a purified RGR standard for calibration. The RPE cell hom ogenate and m icrosom al m em branes w ere prepared from twenty eyes, as described previously (Shen et al., 1994). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-1. Isolation of RGR from the bovine retinal pigment epithelium RGR was isolated from a digitonin extract of microsomal proteins by an immunoaffinity procedure and electrophoresed in a 12% SDS-polyaciylamide gel The protein extract (14 mg) in lane 1 and purified RGR (0.7 mg) in lane 2 were analyzed by (A) protein silver staining and (B) immunoblot analysis using monoclonal antibody 2F4 directed against the carboxyl terminus of bovine RGR (Shen et aL, 1994). Alkaline phosphatase-conjugated anti-mouse IgG and colorimetric substrates were used to detect the bound antibody. The arrows point to proteins that are approximately 31 and 34 kDa. 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-2. Absoiption spectra o f RGR. RGR from bovine RPE was purified in 10 mM sodium phosphate buffer, pH 6.5, containing 150 mM NaCl, 0.5 mM EDTA and 0.1 % digitonin. (A) The spectrum of RGR was determined at pH 6.5, 5.2, 4.2 and 8.0. The absolute curves were superimposed at 660 to 680 nm to equalize the baselines. (B) The spectrum at pH 8.0 was subtracted from the spectrum at pH 4.2. The maximum and minimum for the absorption peaks lie at approximately 466 and 364 nm, respectively. 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0.06 0.04 0.02 5.2 6.5 8.0 300 400 500 600 W avelength (nm) 700 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 466 nm p H 4 . 2 - p H 8 . 0 0.01 364 nm. - 0.01 400 500 600 700 300 W avelength (nm) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-3. The effect o f all-trans-retinal on the spectrum of RGR. (A) RPE microsomal membranes were incubated with or without S O mM all-trans-retinal prior to purification in the dark and scanning the sample at pH 6.5 and at 24 °C. (B) Difference spectrum of RGR at pH 6.5. The spectrum of untreated RGR was subtracted from the spectrum of RGR that was incubated with all-trans-retinal The maximum for the two prominent absorption peaks lie at 466 and 375 nm. After three separate eperiments, mean wavelengths o f469 ± 2.4 and 370 ± 7.3 nm (+ standard deviation) were obtained for the absorption maxima Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65 Absorbance 0.06 0.04- + all-trans-retinal 0.02- — all-trans-retinal 0. 01“ 400 300 500 600 700 W avelength (nm) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-4. Absorption spectra and pH indicator property of RGR after incubation with all-trans-retinal. (A) The absorption spectrum of purified RGR at each pH was determined after incubation of RPE microsomal membranes with S O pM all-trans-retinal. The absolute curves were superimposed at 660—680 nm to equalize the base lines. (B) pH difference spectrum for RGR incubated with all-tm as-retinal. The spectrum at pH 8.0 was subtracted from the spectrum at pH 4.2. The absorption maximum and minimum for the interconverted peaks lie at approximately 466 and 363 nm, respectively. 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0.06 0.04 5.2 • 6.5 0.02 8.0 300 400 500 600 700 W avelength (nm) 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0.03 466 nm 0.02- pH4.2-pH8.0 0.01- 0 - - 0.01- - 0.02- 363 nm -0.03 300 400 500 600 700 W avelength (nm) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-5. Effect of all-trans-retinal on the spectrum o f RGR in 0.1 M TCA RGR was purified in the dark at pH 6.5, and the solution was adjusted to 0.1 M TCA immediately before scanning. (A) Spectra of acid-denatured RGR from microsomal membranes that were incubated with or without 50 |iM all-trans-retinal. (B) Difference absorption spectrum of RGR in 0.1 M TCA. The spectrum of untreated RGR was subtracted from the spectrum of RGR that was preincubated with all-trans-retinal. Instead of two main absorption peaks at 466 and 375 nm, as shown in Fig. 3-3(B), the absorption maximum appears at aprroximately 450 nm 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0.08 0.06- + all-trans-retinal 0.04- 0.02- — all-trans-retinal 0.04 0.02- 600 700 500 400 300 W avelength (nm) 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-6. Effect of SDS on the all-trans-retinal difference spectrum of RGR. RGR was pruified in the dark at pH 6.5 and then incubated at 20 C for 30 min in the absence or presence of 2% SDS immediately before scanning. (A) all-trans-Retm sl difference spectrum of control RGR at pH 6.5. The spectrum of untreated RGR was subtracted from the spectrum of RGR that was preincubated with all-trans-retinal, as in Fig. 3-3(B). (B) all-trans-Retinal difference spectrum of RGR in 2% SDS. The difference absorption spectra were determined from the SDS-treated samples, first at pH 6.5 and then at pH 12, afier the addition of 5 M NaOH to a final concentration of 0.1 M. 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0 .0 2 0.01- 0.02 - pH 6.5 0.01- pH 12 400 500 300 600 700 W avelength (nm) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-7. Instability of RGR toward hydroxylamine. (A) The binding of all-trans-[3H]retinal to RGR was abolished by 0.25 M hydroxylamine. Each lane contained approximately 0.2 mg protein from RPE microsomes. The X-ray film was exposed for a period of 5 days. (B) Hydroxylamine (80 mM) was added to purified RGR at pH 4.2. Decay of absorbance is shown by smoothed difference spectra, which were obtained by subtraction of the spectrum at 36 min from the spectra at the indicated time points. 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A 1 2 130 _ 75 “ 50 - -0.02-1------- 1 -------1 -------1 -------1 ------ 1 -- i -------1 -------1 300 400 500 600 700 Wavelength (nm) 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3-8. The effect of light on the absorption spectrum of RGR. (A) Absolute spectrum of RGR before and after illumination. (B) Difference between dark and post-Ohimination spectra shown in panel A. The difference spectrum showed change in absorption at approximately 368 nm. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Absorbance 0.06 0.04 0.02 0 0.01 0.005 0 300 400 500 600 700 W avelength (nm) 77 [ S E E K S Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4. THE ENDOGENOUS CHROMOPHORE OF RGR-OPSIN, A RETINAL PHOTOISOMERASE FROM THE PIGMENT EPITHELIUM 4-1. Introduction A number of visual pigment homologues have been identified outside of photoreceptor cells in vertebrates. Nonvisual opsins reside in the pineal gland, melanophores, Muller cells, and the retinal pigment epithelium (RPE) (Jiang et aL, 1993; Okano et aL, 1994; Max et aL 1995; Sun et aL, 1997; Provencio et aL, 1998). The RPE and Muller cell opsin, a putative RPE retinal G protein-coupled receptor (RGR, or RGR-opsin), is most similar in amino acid sequence to retinochrome, a photoisomerase that catalyzes the conversion of all-trans- to 11-cis-retinal in squid photoreceptors (Hara and Hara, 1987; Hara-Nishimura et aL, 1990). RGR has been isolated from bovine RPE microsomal membranes under dark conditions, and its absorption spectrum reveals two pH-dependent species with absorption maxima in the blue (A ^ - -466 nm) and near-ultraviolet (A ^ = -364 nm) regions of light (Hao and Fong, 1996). The shape of the absorption peaks and biochemical properties of the photopigment are consistent with those of a retinylidene SchifF base chromophore, the pKa of which is markedly different from those of the visual pigments. RGR is localized to intracellular membranes in RPE and Muller cells (Pandey et aL, 1994) and is able to bind exogenously added all-trans-retinal more efficiently than 11-cis-retinal (Shen et aL, 1994), however the structure of its endogenous chromophore is unknown. In addition to RGR, the RPE contains the visual pigment homologues, peropsin and melanopsin (Sun et aL, 1997; Provencio et aL, 1998). Peropsin is another 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. retinochrome-like opsin, and melanopsin most closely resembles cephalopod visual pigments. The presence of multiple opsins in the RPE signals that the RPE may consist of primary photoreceptive cells. This monolayer of highly differentiated epithelial cells is essential for the normal function of adjoining photoreceptors. Its diverse and unique roles in the visual process include the removal by phagocytosis of the discarded tips of photoreceptor outer segments (Young and Bok, 1969), storage of retinoids (Zimmerman, 1974; Bridges, 1976) and the isomerization of all-trans to 11-cis retinoids for regeneration of visual pigments (Saari, 1990; Rando et aL, 1991). To process the flow of retinoids through the visual cycle, the RPE contains an abundance of specialized proteins, such as cellular retinaldehyde-binding protein (CRALBP) (Futterman et aL, 1977), lecithin retinol acyl transferase (Barry et aL, 1989; Saari and Bredberg, 1989), isomerohydrolase (Deigner et aL, 1989; Trehan et aL, 1990), 11-cis retinyl ester hydrolase (Berman et aL, 1985; Blaner et aL, 1987; Mata et aL, 1998), 11-cis-retinol dehydrogenase (Zimmerman, 1976; Driessen et aL, 1995; Simon et aL, 1995), and possibly other preferentially expressed proteins. Regulation of the multi-step retinoid pathways in RPE is highly coordinated with visual pigment status and lighting conditions through complex and unknown mechanisms (Dowling, 1960). On the basis of its subcellular localization in RPE microsomes, ability to form a stable photopigment with bound all-trans-retinal, and amino acid sequence homology with retinochrome, RGR may be involved in a retinochrome-like mechanism of the vertebrate visual cycle or in a novel form of phototransduction. All retinaldehyde-based opsins involve stereo specific cis-trans photoisomerization of the bound chromophore as a central step in their biological function. In this paper, we describe studies to directly identify the endogenous chromophore of RGR from the RPE and to investigate the photochemistry of RGR by analyzing the effect of light on the isomeric configuration of the bound retinaL 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4-2. Experimental Procedures 4-2-1. Materials. Digitonin was from Acros Organics (GeeL, Belgium). Hydroxylamine and all-trans-retinal were purchased from Sigma (St. Louis, MO). 11-cis-Retinal was a gift from Dr. Rosalie Crouch (Medical University of South Carolina, Charleston, SC). The all-trans- and 11- cis-rednal isomers were analyzed for purity by high-performance liquid chromatography (HPLC) before use, as described previously (Shen et aL, 1994). All organic solvents used in this study were HPLC grade. Dichloromethane and diethyl ether were from J. T. Baker (Phillipsburg, NJ), and hexane was from Fisher Chemical (Fair Lawn, NJ). 4-2-2. Isolation of bovine RGR. RGR was isolated as described previously, except for slight modifications in some experiments. Fresh bovine eyes were obtained from a local abattoir and kept in darkness at ambient temperature for -1.5 hour before dissection. After excision of anterior segments, the RPE cells were scraped gently from the eyecups under dim red light. The isolation of RPE microsomal membranes and purification of RGR were performed under darkness or dim red light, as described previously (Hao and Fong, 1996). The membranes were extracted thrice for 1 hour at 4 °C with 1.2% digitonin in 10 mM sodium phosphate buffer, pH 6.5, containing 150 mM NaCl and 0.5 mM EDTA After centrifugation of the extract at 100,000 g for 20 min, the supernatant was incubated overnight at 4 °C with Affi-Gel 10 resin (Bio-Rad, Hercules, CA) conjugated to anti-bovine RGR monoclonal antibody 2F4 (Shen et aL, 1994). The immunoaffinity resin was transferred to a column for washing with 25 bed volumes of 10 mM sodium phosphate buffer, pH 6.5, containing 0.1% digitonin, 150 mM NaCl and 0.5 mM EDTA. The column was then loaded 10 times with 0.5 bed volumes of wash buffer containing 50 mM bovine RGR carboxyl terminal 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. peptide (CLSPQRREHSREQ). The eluates were pooled and concentrated approximately fourfold using a Centricon-3 concentrator (Amicou, Inc., Beverly, MA). Gel electrophoresis and immunoblot of proteins were performed, as described previously (Jiang et aL, 1993). 4-2-3. Preparation of rhodopsin. Bovine rod outer segment (ROS) membranes were prepared in dim red light, as described before (Hong et aL, 1982), and stored at -80 °C. For measurement of the rate of photoisomerization of 11-cis-retinal, rhodopsin was solubilized by incubation of the ROS membranes in 1.2% digitonin, 10 mM sodium phosphate, pH 6.5, 150 mM NaCl, and 0.5 mM EDTA for 1 hr at 4 °C. After centrifugation at 100,000 g, the pellet was extracted twice more. The supernatants were pooled, and photoisomerization of the rhodopsin chromophore was investigated immediately. 4-2-4. Extraction of opsin-bound retinal isomers by hydroxylamine derivatization. The retinal chromophore of purified bovine RGR was extracted under dim red light and analyzed by the method of hydroxylamine derivatization as described (Groenendijk et aL, 1979). This method has been used successfully to extract retinal from ROS membranes and has given quantitative recovery and complete retention of the geometric structure of retinal isomers (Groenendijk et aL, 1980). hi a typical extraction procedure, 100 to 300 ml of purified RGR, ROS membrane suspension, or purified retinal isomers was supplemented with 2 M hydroxylamine, pH 6.5 (0.1 voL/voL), followed by 300 ml methanol and 300 ml dichloromethane. Sodium phosphate buffer, pH 6.5, was added to bring the sample volume to 900 ml. The extraction with dichloromethane (aqueous buffer/methanol/dichloromethane, 1:1:1 by voL) was performed by vortexing the mixture for 30 sec, followed by centrifugation at 12,000 g for 1 mm The lower organic phase was 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. removed carefully, and the upper phase was extracted twice more with 300 ml dichloromethane. The organic layers were pooled and dried down under a nitrogen stream. The extracted retinaloximes were then solubilized in hexane, filtered through glass wool held in a pipet tip, and dried again. The samples were either stored in darkness at - 80 °C or analyzed immediately by HPLC. Rhodopsin and purified retinals were extracted as controls, and the resultant distribution of retinaloxime isomers was analyzed by HPLC to monitor the occurrence of any nonspecific isomerization during extraction. The all- trans-retinal and 11-cis-retinal standards were first solubilized in 0.1% digitonin in 10 mM sodium phosphate buffer, pH 6.5, 150 mM NaCl and 0.5 mM EDTA. 4-2-5. Analysis of retinaloximes by HPLC. The isomers of retinaloximes were analyzed by HPLC, as described previously (Groenendijk et aL, 1979; Ozaki et aL, 1986). The extracted retinaloximes were dissolved in hexane and applied to a LiChrosorb RT Si60 silica column (4 x 250 mm, 5 mm) (E. Merck, Darmstadt, Germany). The HPLC system was equipped with a Beckman 126 solvent module and 166 Ultraviolet (UV)/Visible detector (Beckman Instruments, Fullerton, CA). The samples were injected in a volume of 20 ml and separated by an eluent consisting of hexane supplemented with 8% diethyl ether and 0.33% ethanol (Ozaki et aL, 1986). The HPLC column was pre-calibrated using the reaction products of hydroxylamine and purified all-trans-retinal or 11-cis-retinal standards. Identification of the retinaloxime isomers was based on the retention times of the known retinaloxime products and was in agreement with the results of previous chromatograms (Groenendijk et aL, 1979; Groenendijk et aL, 1980; Ozaki et aL, 1986). Absorbance was measured at 360 nm, and the absorbance peaks from the chromatograph were analyzed with the Gold Nouveau Software (Beckman Instruments). The proportion of each isomer in the loading sample was determined from the total peak area of both its syn- and anti-retinaloxime and 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. calculated according to the following extinction coefficients (£3 5 0 , in hexane): all-trans syn = 54,900, all-trans anti = 51,600, 11-cis syn = 35,000, 11-cis anti = 29,600, 13-cis syn = 49,000 and 13-cis anti = 52,100 (Groenendijk et aL, 1979; Ozaki et aL, 1986). 4-2-6. Irradiation of photopigments. RGR or ROS membranes were irradiated with an Oriel light source (Oriel Corporation, Stratford, CT) equipped with a 150 W xenon arc lamp. The lamp produces uniform irradiance from 300 to 800 nm. Monochromatic light beams at 370 or 470 nm were formed by passing the light through a 370-nm interference filter (Oriel #53415), or both a 470-nm interference filter (Oriel #53845) and 455-nm long pass filter (Oriel #51284), respectively. The protein samples were held at room temperature in a quartz cuvette positioned 60 cm from the lamp. After delivery of the intended amount of light, the retinal chromophores were extracted by hydroxylamine derivatization and analyzed by HPLC, as described previously. 4-2-7. Rate of photoisomerization of retinal. RGR was purified, as described previously (Hao and Fong, 1996) with the following modifications. A preparation of RPE microsomes from 20 bovine eyes was incubated in the dark with 50 mM all-trans-retinal to saturate the retinal binding site of RGR. Prior to elution of RGR from the immunoaffinity column, the beads were resuspended and washed with 15 volumes of 1% bovine serum albumin (BSA) in wash buffer consisting of 0.1% digitonin, 10 mM sodium phosphate, pH 6.5, 150 mM NaCl, and 0.5 mM EDTA, followed by washing with 15 volumes of wash buffer without BSA. RGR was then eluted from a column in 7 ml peptide-containing wash buffer. The eluate was concentrated to ~0.6 ml using a Centricon-3 concentrator and then increased to 1.5 ml with 10 mM sodium phosphate, pH 6.5, 150 mM NaCl, and 0.5 mM EDTA. The pH of the sample 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was lowered to 4.2 by the addition of ISO ml 1 M sodium citrate buffer, pH 3.8. Equal aliquots of RGR were irradiated with 470-nm monochromatic light at illuminance of 410 lux for various periods of time. Each sample was mixed with 0.2 M hydroxylamine immediately upon the end of irradiation and kept in darkness on ice until the retinal chromophores were extracted from all samples and analyzed by HPLC, as described. The results were plotted as -In a/ag versus time, where a ,} is the initial amount of all-trans- retinal in nonirradiated RGR and a is the amount of the all-trans isomer in the irradiated sample. The first order rate constant was calculated using the data from time points 0 to 30 sec. Rhodopsin (ROS) was illuminated in solution containing 1.2% digitonin, 10 mM sodium phosphate, pH 6.5, ISO mM NaCl, 0.5 mM EDTA, and 20 mM hydroxylamine. Equal aliquots were irradiated with 470-nm monochromatic light for various times, and the absorption spectrum of each sample was determined immediately after irradiation. There were no further changes in the absorption spectrum by 20 min of exposure to light. Difference spectra were obtained by subtraction of the absorption spectrum at 20 min from the spectrum at each time point. The rate of photoisomerization of 11-cis-retinal was measured from the difference spectra by the decrease in absorbance at 500 nm The photoisomerization of 11 -cis-retinal in rhodopsin followed a typical first order reaction, and a rate constant was derived from the plot of -In a/a0 versus time, where a0 and a are the absorbance of nonirradiated and irradiated rhodopsin at 500 nm 4-2-8. Calculation of extinction coefficients and pKa of the retinylidene Schiff base of RGR. The pKa of the retinylidene Schiff base of RGR and extinction coefficients of blue and UV light-absorbing RGR (RG R^ and RGR3 7 0 , respectively) were calculated from previous data (Hao and Fong, 1996). From difference spectra which indicate the absorbance of 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RGR and its retinaloxixne bleaching product (Fig. 7 in Ref 8), it was determined that the molar extinction coefficient of RGR^g is -62,800 cm* 1 M'1 . From the difference in absorption spectrum of RGR at pH 8.0 and 4.2 (Fig. 4 in Ref 8) it was determined that the extinction coefficient of RGR3 70 is -66,100 cm* 1 M *1 . 4-2-9. Determination of the quantum efficiency of photoisomerization and photosensitivity of RGR. The photosensitivity of visual pigments is equal to the product of its extinction coefficient (e) and its quantum efficiency of photoisomerization (y). The quantum efficiency of rhodopsin (yA o d o p sjJ is the number of all-trans-retinal formed per number of photons absorbed and has been calculated previously (Dartnall, 1972). To determine the quantum efficiency of photoisomerization for RGR, we followed a method analogous to that used to evaluate the photosensitivity of CRALBP (Saari and Bredberg, 1987). In this experiment, RGR was converted essentially to RGR4 5 9 by lowering the pH of the sample solution to 4.2. The first order rate constants for photoisomerization of all-trans-retinal bound to RGR^g ( k RGR469) and photoisomerization of 11-cis-retinal bound to rhodopsin ( ^rhodopsin)were obtained torn the experiments on kinetics of photoisomerization, as described above. It is assumed that the ratio K R O R W ^diodcpsin is proportional to the ratio of their photosensitivity, hence KRGR469 e RGR469 YRGR469 = Eq. 1 Kthodpsn Sihodopsm yifaodcpsm where eR G R 4 6 9 = 62,800, eA o d o p s in =32,900, and yA o d o p s in = 0.67. The extinction coefficient of rhodopsin (40,600 at (Hubbard et aL, 1971) was corrected for the wavelength of 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the light used (X . = 470 nm), according to Beer's law (A = e 1 c). From equation 1, the value for Y R G r 4 6 9 (0.12) was solved, and the photosensitivity ofRGR4 6 9 (7540 cnr1 M'1 ) was then determined. 4-3. Results 4-3-1. The endogenous chromophore of RGR. RGR was purified in darkness, and the geometric configuration of its retinal chromophore was determined. The opsin-bound retinal was convened into retinaloxime by incubation of RGR with hydroxylamine. RGR is highly sensitive to hydroxylamine in the dark and reacts completely (Hao aid Fong, 1996). For each retinal isomer, conjugation with hydroxylamine results in retinaloxime products of syn and anti configuration. The syn and anti configurations refer, respectively, to the cis and trans positions of the hydroxyl group with respect to the hydrogen at C-15 (Groenendijk ct aL, 1979). Three isomers of retinal were identified by HPLC and consisted of 11-cis (6%), 13-cis (9%), and all-trans (85%) forms (Fig. 4-1). The syn/anti ratios for the 11-cis and all-trans retinaloximes were 3.9 and 11.4, respectively. In a parallel control experiment, extraction of retinal fiom rhodpsin yielded 11-cis (91%), 13-cis (2%), and all-trans (7%) isomers and the expected predominant representation of 11-cis-retinal (Fig. 4-1). The syn/anti ratios of the 11-cis and all-trans retinaloxime derivatives from rhodopsin were 1.8 and 2.0, respectively, which are consistent with results described previously (Groenendijk et aL, 1980). When pure 11- cis-retinal or all-trans-retinal standards were extracted under similar conditions, no isomerization of the retinals occurred in the dark (results not shown). The all-trans isomer of retinal was the predominant chromophore extracted from RGR. The possibility that all-trans-retinal arose from other isomers by nonspecific isomerization during the purification of RGR was investigated. To monitor changes in the 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. levels of individual isomers during purification of RGR, the percent quantity of retinal isomers was measured at various steps throughout the procedure. The experiments are outlined in a flowchart, and the results from three separate purifications are summarised in figure 4-2. In the first experiment, the retinal isomers in RPE microsomal membranes (M) and in the combined retinoid extracts of the 100,000 g pellet (P), flowthrough (F) and purified RGR (E) were analyzed. The results showed that the composition of retinal isomers in the starting material and in the terminal steps of RGR purification did not vary substantially; only 3% of the retinal appeared to be converted from 11-cis- to all-trans- retinal. The absolute amount of 11-cis-retinal that declined could not account for the total amount of all-trans-retinal in RGR. hi a second experiment, retinals were analyzed from RPE microsomal membranes (M) and from the combined retinoid extracts of the 100,000 g pellet (P) and supernatant (S). The composition of retinal isomers in the starting material and in the steps after solubilization in 1.2% digitonin varied again just slightly; approximately 3% of the retinal appeared to be converted from the 11-cis isomer to all- trans-retinaL In a third experiment, retinals were analyzed from the supernatant (S) and from the combined retinoid extracts of the flowthrough (F) and purified RGR (E). The results showed that little or no change occurred in the relative levels of retinal isomers during the step of immuno affinity chromatography. The data for the distribution of retinal isomers in these experiments were reproducible and suggest that solubilization of the membranes in 1.2% digitonin causes only a small fraction of 11 -cis-retinal to be converted to all-trans-retinal. 4-3-2. Photoisomerization of all-trans-retinal bound to RGR. The effect of light on the chromophore of RGR was investigated after solubilization and purification of the protein. RGR was irradiated or not with 470-nm monochromatic light for 3 min at pH 6.5, and retinal was extracted by hydroxylamine derivatization. The 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. retinal isomers in nonirradiated RGR consisted of 11-cis (8%), 13-cis (7%), and all-trans (85%) forms (Fig. 4-3). The configuration of retinal isomers shifted in irradiated RGR, which contained 11-cis (40%), 13-cis (8%), and all-trans (52%) forms. The data showed that irradiation of RGR results in stereospecific photoisomerization of all-trans-retinal into 11-cis-retinal without a significant alteration in the relative amount of 13-cis and other isomers. In other experiments, the irradiation of RGR with 370-nm monochromatic light also produced the stereospecific photoisomerization of all-trans- to 11-cis-retinal (Fig. 4-4). To test protein-dependent stereospecificity of photoisomerization, the effect of light on the chromophore of RGR was analyzed after treatment of the protein with SDS. Denaturation of RGR in 2% SDS is accompanied by an alteration of its absorption spectrum (Hao and Fong, 1996). The retinal isomers in nonirradiated SDS-treated RGR consisted of the 11-cis (9%), 13-cis (14%), and all-trans (77%) forms (Fig. 4-5). The resulting syn/anti ratio of the all-trans-retinaloxime was 1.3. After irradiation of the denatured protein, the distribution of isomers was 11-cis (27%), 13-cis (26%), and all- trans (47%) retinaloximes. Photoisomerization of the chromophore in denatured RGR occurred with a loss of stereospecificity, since the relative level of both 11-cis- and 13-cis- retinal rose with the decline in all-trans-retinal. 4-3-3. Photosensitivity of RGR. Photosensitivity is defined as the product of the extinction coefficient (s) of a photoreceptive molecule and its quantum efficiency (y) (Dartnall, 1972). To characterize the photoisomerase activity of RGR, its photosensitivity was compared to that of the visual pigment rhodopsin. RPE microsomal membranes were incubated in the dark with exogenously added all-trans-retinal to increase the yield of photoreceptive RGR, and the pH was lowered from 6.5 to 4.2 to convert RGR into its blue light-absorbing form, 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RGR469. Equal aliquots of RGR were then exposed to 470-nm monochromatic light for various lengths of time, and the isomerization of all-trans-retinal was measured. Steady illumination of RGR resulted in increasing conversion of all-trans-retinal to 11-cis-retinal over a period of 5 min (Fig. 4-6). The proportion of 13-cis-retinal extracted from the irradiated samples remained constant at all time points. A relatively fast initial rate of decay of all-trans-retinal was maintained for about 30 sec, after which the decay of all-trans-retinal occurred at a slower rate. For comparison, the kinetics of photoisomerization of 11-cis-retinal in rhodopsin followed a first order reaction rate. Under our experimental conditions, the first order rate constants for RGR from 0-30 sec and for rhodopsin from 0-60 sec were kR G R 4 6 9 = 8.2 x 10-3 S '1 and = 24.2 x 10-3 s-1 . Calculation of the photosensitivity of RGR4 6 9 yielded the value 7540 cm* 1 M‘l, and its quantum efficiency, yR G R 4 6 5 , equaled 0.12. The 11-cis isomer was a consistent component of the retinaloxime extracts of purified RGR, except when RPE microsomal membranes were treated in the dark with exogenously added all-trans-retinal. After incubation with excess all-trans-retinal to saturate the chromophore binding sites, the 11 -cis isomer was not detected in purified RGR. The distribution of retinal isomers from retinaldehyde-treated RGR before irradiation (t = 0 sec) was 11-cis (0%), 13-cis (13%), and all-trans (87%) forms (Fig. 4-6). After 5 min of irradiation (t = 300 sec), the distribution of retinal isomers from RGR was the 11-cis (50%), 13-cis (11%), and all-trans (39%) forms (Fig. 4-6). 4-4. Discussion The spectral properties of RGR purified from bovine RPE suggest that RGR is conjugated in vivo to a retinal chromophore by means of a covalent Schiff base bond. In this study, the endogenous chromophore of RGR has been extracted and identified following 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hydroxylamine derivatization. Our findings indicate that the predominant chromophore of RGR is all-trans-retinaL Irradiation of RGR results in stereospecific conversion of the bound all-trans isomer to 11-cis-retinal. These results support the notion that RGR functions as a stereospecific photoisomerase in the RPE and Muller cells. 4-4-1. The chromophore of RGR. In addition to all-trans-retinal, a minor constant amount of 13-cis-retinal was observed routinely in the retinoid extracts from RGR. The origin of 13-cis-retinal is unclear. Persistent small amounts of this isomer have been observed likewise in retinoid extracts of retinochrome (Ozaki et al., 1986). A small amount of the 13-cis isomer was also present in our retinoid extract of rhodopsin following the extraction conditions. As a nonphysiological constituent, the 13-cis isomer may be formed by thermal isomerization during the protein denaturation. 11-cis-Retinal also was extracted in the dark from RGR Since RGR photoisomerizes bound all-trans- to 11 -cis-retinal, it is conceivable that the 11- cis-retinaloxime originated from a population of physiologically relevant 11-cis-retinal- RGR complex. In relation to this assumption, when the RPE microsomal membranes were preincubated with exogenous all-trans-retinal, 11-cis-retinal was completely absent in purified RGR although 13-cis-retinal was still present. It is probable that a small fraction of RGR is bound in situ to 11-cis-retinal which can be quantitatively displaced from the protein by an excess of the all-trans isomer. In a control experiment, the endogenous chromophore of rhodopsin was extracted and analyzed in an similar manner As expected, the physiologically relevant 11-cis and all-trans retinal isomers were identified in rhodopsin, along with the small amount of 13-cis-retinal The procedure for purification of RGR extends over 20 hours, and during this time, there is a potential for spontaneous thermal isomerization of the retinal chromophore in RGR from a hypothetical 11-cis to all-trans configuration. The actual extent of thermal 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. isomerization of 11-cis-retinal was investigated in three separate purifications of RGR. The total composition of retinal isomers at various steps of the purification was analyzed to detect any change in retinal isomers from step to step. Overall, there were only slight differences in the distribution of retinal isomers between that in microsomes and the final steps of purification. A small -3% decrease in 11-cis-retinal and a slight increase in all- trans-retinal was associated with the step of solubilization of RPE microsomal membranes in the presence of digitonin. The actual amount of 11-cis-retinal that underwent artifactual isomerization was significantly less than the total amount of all-trans-retinal in RGR. Thus, the results exclude the possibility that the major endogenous chromophore of RGR from microsomes was 11-cis-retinal originally, which converted to the all-trans configuration during purification of RGR in the dark. Interestingly, the reaction of the chromophore of RGR with hydroxylamine produced syn- and anti-isomers of all-trans-retinaloxime in a syn/anti ratio of about 9, which is significantly higher than the syn/anti retinaloxime ratios of about 2 and 3.4 observed in reactions with illuminated rhodopsin and free all-trans-retinal, respectively (Groenendijk et aL, 1980). The singular syn/anti oxime ratio for RGR suggests that the protein retains a folded conformation. The interaction between chromophore and hydroxylamine may be constrained by the structure of the binding site for all-trans-retinal, such that the production of the syn-isomer of all-trans-retinaloxime is strongly favored in the microenvironment of the reaction. The denaturation of RGR in the presence of SDS then lowers the syn/anti all-trans-retinaloxime product ratio to about 1.3 (Fig. 4-5). The universal chromophore of the visual pigments is 11-cis-retinal or 1 l-cis-3,4- dehydroretinal. The identification of all-trans-retinal as the principal endogenous chromophore of RGR in the dark is consonant with a few exceptional characteristics of the RPE opsin. RGR in microsomal membranes binds all-trans-retinal in the dark preferentially, rather than the 11-cis isomer (Shen et aL, 1994). The amino acid sequence 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of RGR is most closely related to that of retinochrome (Hara-Nishimura et aL, 1990), an all-trans-retinal photoisomerase found in mollusks. In addition, irradiation of the all-trans retin al-RGR complex modifies the relative amounts of the two physiologically relevant isomers only, Le. all-trans- and 11-cis-retinal. 4-4-2. Photoisomerase activity. The stereospecific cis-trans photoisomerization of bound retinal isomers is a basic motif of evolutionary diverse opsins, and particular isomers are involved at distinct steps in the photocycle of individual photopigments. Upon illumination, all-trans-retinal in retinochrome and bacteriorhodopsin isomerizes to 11-cis- and 13-cis-retinal, respectively (Ozaki et aL, 1983; Stoeckenius and Bogomolni, 1982). When irradiated with monochromatic light at 470 nm or 370 nm, all-trans-retinal in RGR, as in retinochrome, isomerizes stereospecifically to 11-cis-retinal. Only 11-cis-retinal is newly formed in RGR, and relative levels of the 13-cis and other isomers of retinal are unchanged. The photoisomerization of all-trans-retinal in RGR was dependent on a folded protein structure, since denaturation of RGR in SDS abolished the cis-trans stereospecificity. Although the chromophore in RGR was converted specifically to 11-cis-retinal, the highest level to which the 11-cis isomer increased was -50% of total retinals under our experimental conditions. By comparison, all-trans-retinal comprised 93% of total retinals after we irradiated rhodopsin with orange light (>530 nm) (results not shown). The formation of 11-cis-retinal in RGR contrasts with photoisomerization of the chromophore in retinochrome, in which the 11-cis isomer may increase to as much as 80% of total retinals after illumination (Ozaki et aL, 1983). The extent to which 11-cis-retinal is formed in RGR may be reduced by stereospecific photoreversal of the chromophore back to the all-trans isomer. Irradiation of RGR does not lead to bleaching or a large shift in the of the blue light-absorbing 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. form of RGR (Hao and Fong, 1996). We conclude that 11-cis-retinal does not dissociate freely from the purified protein in the presence of light. Therefore, continuous light exposure may result in photoisomerization of the 11-cis-retinal bound to irradiated RGR and, eventually, a photoequilibrium of RGR bound to both 11-cis and all-trans isomers. The larger extent to which 11-cis-retinal can be formed in retinochrome upon illumination with orange light may be attributed to the difference of ~26 nm between the of retinochrome (496 nm) and that of metaretinochrome (470 nm). Because of the lower X m a x of metaretinochrome, photoreversal of the chromophore in retinochrome would be greater in the presence of light at shorter wavelengths than under orange light (>560 nm). Indeed, when retinochrome is exhaustively irradiated with light of shorter wavelength (420 nm or near-UV light), the amount of 11-cis-retinal reaches only 42-50% of total retinals (Ozaki et aL, 1983). The extinction coefficients ofR G R ^ (62,800) and RGR3 7 0 (66,100) are slightly higher than that of retinochrome (60,800) (Hara and Hara, 1982) and 1.5- and 1.6-fold greater than that of rhodopsin (40,600), respectively. Although RGR has a higher extinction coefficient, its photosensitivity appears to be lower than that of rhodopsin. Under our experimental conditions, R G R ^ is at least a third as efficient as rhodopsin in using the energy of photons. The photosensitivity ofRGR4 6 9 is approximately 7-fold greater than that of CRALBP (1078 cm:1 M *1 ) (Saari and Bredberg, 1987). The data support the argument that photoisomerization of the bound chromophore is physiologically relevant for involvement in the function of RGR It is possible that RGR is a still more efficient photoisomerase under in vivo conditions or within an optimal lipid membrane environment. Retinal-binding proteins also may serve to transfer retinals to and from RGR and accelerate the net production of 11-cis-retinal. Recently, CRALBP has been shown to increase the rate of 11-cis-retinol synthesis by RPE isomerohydrolase (Winston and Rando, 1998). In squid, retinaldehyde-binding protein (RALBP) downloads 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11-cis-retinal from metaretinochrome to metarhodopsin through an exchange of isomers (Terakita et aL, 1989). A mechanism for rapid synthesis of 11-cis-retinal must exist for the regeneration of visual pigments under photopic conditions. The synthesis of 11-cis-retinol by the RPE isomerohydrolase provides one such mechanism that also can act during dark adaptation. Although it can be demonstrated that RGR is a stereospecific photoisomerase that favors conversion all-trans-retinal to the 11-cis isomer, there is yet no direct evidence that the 11- cis-retinal from RGR dissociates and enters the pathway for regeneration of visual pigments. The RPE appears to have divergent pools of 11-cis-retinal, the function of which may be to participate in a novel phototransduction system of the RPE, to supply a chromophore for peropsin or melanopsin, or to regenerate rhodopsin and the cone pigments under scotopic and intense photopic lighting conditions. Further characterization of RPE opsins and the precise flow of retinoids through the RPE, including analysis of retinoid metabolism in mouse mutants that lack RGR, will test the hypothesis that RGR functions in a retinochrome-like arm of the vertebrate visual cycle. 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-1. The endogenous chromophore o f RGR and rhodopsin. RGR and ROS membranes were isolated from fresh bovine eyes under dim red light. The ROS membranes were stored at -8O 0C before use, and RGR was used immediately after preparation. The bound chromophores of RGR (left) and ihodopsin (right) were reacted with hydroxylamine and analyzed by HPLC. The syn-isomers of the retinaloxime reaction products were eluted from the HPLC column first and followed by the group of anti isomers. Absorbance was measured at 360 nm, and the amounts of both syn- and anti isomers of retinaloxime were added to determine the percent quantity of each retinal isomer, all: all-trans-; 13: 13-cis-; and 11: 11-cis-retinaloxime. 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Os L O n C O o o oo o C V J C\J T— o so in os 00 TJ CO cn -> O O GO O C O O doueqjosqv 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Minutes Fig. 4-2. Distribution o f retinal isomers at various steps in the isolation of RGR. (A) Diagram of the procedure for isolation of RGR The boxed letters represent the various fractions resulting from individual purification steps. M: bovine RPE microsomal membranes; S: 100,000 g supernatant of digitonin-solubilized extract; P: 100,000 g pellet o f digitonin-solubilized extract; F: combined flowthrough and washes from immunoaffinity column; E: purified RGR eluted with bovine RGR carboxyl terminal peptide. (B) Summary of the results of three independent experiments in determining the distribution of retinal isomers during isolation of RGR The retinal isomers were extracted separately from the saved fractions and then pooled together for HPLC analysis, as indicated, hi experiment L , retinal isomers were analyzed from the microsomal membranes (M) and from the pooled retinoid extracts of pellet (P), flowthrough (F) and purified RGR (E). hi experiment H, retinals were analyzed from microsomal membranes (M) and from the pooled retinoid extracts of the pellet (P) and supernatant (S). In experiment E Q , retinals were analyzed from the supernatant (S) and from the pooled retinoid extracts of the flowthrough (F) and purified RGR (E). The results are expressed as the percent quantity of individual retinal isomers. 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RPE CELLS 1 RPE MICROSOMES Solubilization in Centrifuge 1.2% digitonin A 100,000 g S] SUPERNATANT PELLET G O Immunoaffinity purification FLOWTHROUGH ELUATE [e ] AND WASH B Experiment Fraction(s) 11-cis all-trans 13-cis (%) I M F + E + P 45 43 48 51 6.5 5.9 TT M 45 51 4.0 1 L S + P 41 55 4.4 m S 37 56 6.9 F + E 38 57 5.7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-3. Photoisomerization o f the retinal chromophore of RGR. The chromophore of purified RGR was analyzed after incubation in darkness (left) or irradiation with 470-nm monochromatic light (right). The protein at pH 6.S was irradiated for 3 min at room temperature in a quartz cuvette positioned 60 cm from the lamp 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C M O C M 00 -d * in IT) CD O CO O O O O o o o C M C O o CO O O o o o eoueqjosqv 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Minutes Fig. 4-4. Irradiation o f RGR by near-UV light. The chromophore of RGR was analyzed after irradiation of the protein at pH 6.5 for 3 mfn at room temperature with near-UV light passing through the Oriel #53415 370-nm interference filter. 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o CM C / 3 C es w * i in c a 2 t CM O CO o o o o eoueqjosqv 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Minutes Fig. 4-5. Photoisomerization o f the retinal chromophore o f SDS-denatured RGR. RGR at pH 6.5 was incubated in the dark at room temperature for 25 min in the presence of 2% SDS to denature the protein. The protein was then irradiated or not immediately before extraction of the retinal isomers. (Top) SDS-denatured RGR irradiated with light from a 150 W xenon arc lamp for 2 min. (Bottom) SDS-denatured RGR kept in the dark. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 Absorbance RGR + SDS - light 11-cis 27% all-trans 47 13-cis 26 syn 0 .001- anti 0.002- RGR + SDS - dark 11-cis all-trans 77 13-cis 14 9% 0 . 001- 20 15 Minutes Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4-6. Kinetics of photoisomerization of the retinal chromophore of RGR and rhodopsin. Before purification of RGR, RPE microsomal membranes were incubated in the dark with S O mM all-trans-retinal to maximize the yield of photopigment. The pH of the purified protein was lowered from 6.5 to 4.2 to convert RGR into its blue light-absorbing form, RGR469, and equal aliquots were then exposed to 470-nm monochromatic light for various lengths of time. The same light source was used to irradiate digitonin-solubilized ROS at various times in the presence of 20 mM hydroxylamine. The initial amount of all- trans-retinal in nonirradiated RGR, or the absorbance of nonirradiated rhodopsin, at t = 0 is ag. The amount of all-trans-retinal in RGR, or the absorbance of rhodopsin, after a given period of illumination is a. The results were plotted as -In a/ag versus time for RGR (a) and rhodopsin (s). Under our experimental conditions, the first order rate constants for RGR from 0 - 30 sec and for rhodopsin from 0 - 60 sec are: kR G R 4 6 9 = 8.2 x 10-3 s-1 and K fc o d o p a ,! = 24.2 x 10-3 s-1. At t = 0 sec, RGR contained 11-cis (0%), 13-cis (13%), and all-trans (87%) isomers. At t = 300 sec, RGR contained 11-cis (50%), 13-cis (11%), and all-trans (39%) isomers. 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.5 H RGR Rhodopsin 0 100 200 300 Seconds 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTERS TARGETED DISRUPTION OF rgr GENE IN MOUSE 5-1. Introduction Tlie retinal pigment epithelium (RPE) is a specialized monolayer of cells that lies adjacently to photoreceptor cells of the retina and forms part of the blood-retina barrier through tight junctions among the cells (Miller and Steinberg, 1982). It plays multiple roles that are essential in maintaining the normal physiological function of photoreceptor cells, including offering the physical protection and an adhesive basis, supplying nutrients and carrying away metabolic waste, absorbing the scattered background light to enhance the sharpness of images, isomerizing all-trans retinol into 11-cis-retinal for rhodopsin regeneration, regulating the supply of the photoreceptor chromophore during the dark and light adaptation, and renewing the aging photoreceptor cells through phagocytizing the shed outer segments in the distal ends (Saari, 1990; Bok, 1993). The intimate relationship between the RPE and photoreceptor cells foresees the involvement of the RPE in retinal dystrophies, which usually results in photoreceptor degeneration in the retina and subsequent loss of vision. Characterization of underlying pathological causes in the RPE at molecular level has indicated that retinal dystrophies can result from disruption of various aspects of RPE functions. An early study on RCS rat has pointed out that mutated rdy gene, with which rats fail to phagocydze rod outer segments, can lead to the overt degeneration of photoreceptor cells, a similar phenotype to a class of retinitis pigmentosa diseases in humans (Mullen and LaVail, 1976). Recently, mutations in a RPE specific gene encoding RPE6S have been linked to 5% cases of autosomal recessive 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chfldhood-onset severe retinal dystrophy, a heterogeneous group of disorders affecting both rods and cones (Gu, S. et aL, 1997). RPE65 expresses at high level in the RPE (Hamel et aL, 1993a; Hamel et aL, 1993b; B&vik et aL, 1993), and has been found in association with serum retinol-binding protein and RPE specific 11-cis-retinol dehydrogenase in vivo (Bivik et aL, 1991; B&vik et aL, 1992; Simon et aL, 1995), suggesting that it might be involved in retinoid metabolism in the RPE. The idea that disruption of retinoid metabolism causes inherited diseases in human s has been further affirmed by linking the mutations in the gene encoding cellular retinaldehyde-binding protein (CRALBP) with nonsyndromic autosomal recessive retinitis pigmentosa (Maw, M. A. et aL, 1997). CRALBP has been known to be a carrier protein for 11-cis-retinal and 11-cis-retinol in the RPE (Saari et aL, 1982), protects 11-cis retinoids from light-induced isomerization (Saari and Bredberg, 1987), and regulate retinoids routing by promoting the activity of 11-cis-retinol dehydrogenase (Saari et aL, 1994). Since retinoid metabolism in the RPE needs a concerted effort of a group of carrier proteins and enzymes other than CRALBP, it is seemingly safe to predict that more retinoid metabolism-related genes may be discovered as the causes of diseases in the future. Unlike genes encoding CRALBP and RPE65, another RPE specific gene, bestrophin, has been isolated through genetic linkage analysis and gene cloning with few clues of its functions. Mutations in the gene are responsible for Best macular dystrophy (Stone et aL, 1992; Forsman et aL, 1992; Petrukhin et aL, 1998), an inherited disease sharing similar diagnostic features to age- related macular degeneration that is the leading cause of blindness among people above 60 (Klein et aL, 1992). The gene that encodes a putative retinal pigment epithelium G protein-coupled receptor, or RGR, was isolated by a differential hybridization approach as an effort to identify RPE specific genes (Jiang et aL, 1993). Northern blot and immunohistochemical staining showed that RGR expressed only in the RPE cells and the muller cells of the 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. retina. The RGR sequence has classified the protein as a member of the opsin family of proteins that is a subgroup of G protein-coupled receptors binding retinal ligand at the highly conserved lysine residue in the seventh transmembrane domain (Maden, 1995; Hargrave, 1982;). Particularly RGR displayed the closest sequence identity to cephalopod photopigment retinochrome. The identity between the two proteins even extends to the retinal ligand they bind. Like retinochrome, RGR binds all-trans-retinal ligand (Hara and Hara, 1987; Hao and Fong, in submission). RGR purified directly from bovine eyes in the dark displays two pH-dependent absorption maxima at ~370 nm and ~470 nm that are derived from interaction between RGR opsin and all-trans-retinal ligand (Hao and Fong, 1996). Upon exposure to light irradiation, the RGR-bound all-trans-retinal is isomerized stereospecifically to 11-cis-retinal (Hao and Fong, in submission), suggesting that RGR could function as an isomerase in the light to produce 11-cis-retinal that could be utilized to regenerate bleached photoreceptors involved in vision and/or circadian rhythm. In this study, we used gene-targeting technique to generate rgr gene disrupted mouse. We intend to further investigate the isomerase hypothesis in these RGRless mice and the potential pathological effects of disrupting rgr gene on the functions of photoreceptor cells and circadian rhythm. 5-2. Experimental Procedures 5-2-1. Embryonic stem (ES) cell lines and cell culture. CJ7 strain, passage 11 from Dr. Jeannie Chen, Caltech. R1 strain, passage 10, from Dr. Andras Nagy at Samuel Lunenfeld Research Institute, Mount Sinai Hospital. The cells were first expanded and cells at passage 12 and 13 were frozen in aliquots. 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Both CJ7 and R1 cells were cultured on a monolayer of feeder cells on 0.2% gelatin-coated plates. Feeder cells were derived from mouse embryonic fibroblasts expressing the transgenic neo gene and were therefore G418 resistant. They were mitototically inactivated with y-irradiation at the dosage o f3000 rad prior to culturing ES cells. ES cells were grown in the DMEM (Dulbecco's Modified Eagle Medium) with high glucose and pyridoxine hydrochloride (Gibco, Gaithersburg, MD) supplemented with 15% fetal bovine serum (Lot AGD6411, HyClone, Logan, UT), 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, 0.1 mM (3-mercaptoethanol, 1000 unit/mL LIF (Gibco, Gaithersburg, MD), and 2 mM glutamine, 100 unit/ml penicillin, 100 ng/rnl streptomycin (Irvine Scientific, Santa Ana, CA). ES cells were changed with prewarmed fresh medium every day during culturing. When subconfluence was reached, cells were fed with fresh medium and passaged 4 hours later. To passage cells, medium was removed, followed by washing twice with PBS (Ca2 + and Mg2 + free, Gibco). Then 0.25% trypsin with 1 mM EDTA (Gibco, Gaithersburg, MD) was added to dissociate cells at 37 C, 5% COs for 5 min. Trypsin was inactivated with equal volume of medium, and cells were further disassociated from each other by pipetting cells up and down for 20-30 times. Cells were spun down at 1000 rpm for 6 min and resuspended in fresh medium. ES cells were passaged at a split of from 1/5 to 1/7. 5-2-2. Isolation of cDNA and genomic DNA clones. Four mouse RGR cDNA clones were isolated from a XZAPII retina cDNA library afier hybridization to a radiolabeled human RGR cDNA probe. One of these clones, MRGR7- 5, contained a 1.5-kb cDNA insert, which was subcloned and sequenced completely on both strands. DNA sequencing was carried out using single and double strand phagemid DNA, sequence-specific primers, and Sequenase (U.S. Biochemical Corp., Cleveland, OH), according to the manufacturers' protocol DNA clones containing the mouse rgr 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gene were isolated from a 129SV mouse genomic library in the XZAPII vector (Stratagene, Inc., La Jolla, CA). Five genomic clones (designated \mrgr9, Xmrgrl 1, X mrgrl2, Amrgrl3 and Amrgrl4) were identified by hybridization to the radiolabeled MRGR7-5 cDNA A map of the mouse rgr gene was determined by complete and partial cleavage of Notl-digested genomic DNA using BamHI, EcoRI, Hindm and SacI restriction enzymes. The locations of the exons were mapped by oligonucleotide hybridization and amplification by the polymerase chain reaction. 5-2-3. Construction of targeting vector for homologous recombination. The 5 kb SacI fragment that contains exons 1-4 was excised from mouse genomic clone, X mrgr9, and subcloned into the pBluscript (Stratagene, Inc., La Jolla, CA) to create pMrgr9-S5, from which a 500 bp Ncol-excised fragment encompassing exon 2 was bhmt- ended with Klenow polymerase and subcloned into Smal site on pBluescript to generate pNCO500. Exon 2 from pNCO500 was then cleaved at Bgin site and partially filled in. The Xhol fragment containing neo gene from pPGKneobpA cassette was also partially filled in and ligated into the partially filled Bgin site to create pNCOneo. The 500 bp Ncol fragment in pMrgr9-S5 was then replaced with the Ncol fragment from pNCOneo to create pMrgr9-S5-neo. The tk gene was excised from plasmid XHO PKS MCI TK with Xhol and Clal and inserted into pMrgr9-S5-neo to generate the final targeting construct. 5-2-4. Transfection and drug selection. ES cells at passage 13 were trypsinized from two 10-cm plates and collected by centrifugation at 1000 rpm for 6 min at room temperature. Cells were washed with PBS (Ca2 + and Mg2 + free) and finally resuspended in PBS at a density of 1.2 x 107 cells/ml. The Scal-linearized targeting vector (25 pg) was mixed with 0.8 ml of cells. The mixture, after 5-min incubation at room temperature, was subjected to electroporation at 230 V 111 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and 500 pF, and again incubated for 10 more minutes after electroporation. The cells were then plated equally onto five feeder plates to culture at 37 C, 5% COi. G418 of 300 |ig/ml (Gibco, Gaithersburg, MD) was added to the culture medium at 48 hours after plating. FIAU (2'-deoxy-2'-fluoro-J3-D-arabinofuranosyl-5-iodouraciL Moravek Biochemicals, Brea, CA) at 0.2 pM was applied to CJ7 cells at 48 hours and to R1 cells at 72 hours. Cells were then allowed to grow with a change of fiesh medium every day untill the9th day after electroporation. 5-2-5. Screening of recombinant clones and genotyping mouse progenies by Southern blot. Double-drug-resisting clones were picked and expanded in six 96-well plates as described previously (Ramirez-Solis et aL, 1993). Following expansion, cells were trypsinized, and one-half content of the cells from each well was frozen as stocks while the remainder was grown in a replica plate for Southern blot analysis. Cells reaching confluence were lysed in situ, and genomic DNA was extracted and digested with BamHI according to the protocol described before (Ramirez-Solis et aL, 1993). The digested DNA was electrophoresed in 8% agarose gel and then transferred to GeneScreen membrane (NEN, Boston, Massachusetts). The membrane was prehybridized for 4 h at 42 C in a buffer containing 50% formamide, 6 x SSC, 5 x Denhardt's solution, 0.5% SDS, 100 pg/ml of denatured salmon sperm DNA. A 3-kb Scal-fragment containing exon 5 from mouse RGR genomic gene or pPGKneobpA plasmid DNA was labeled with [a-3 2 P]dCTP using the nick translation kit from Gibco. The labeled DNA probe was hybridized to the blotted membrane in the same buffer for 20-24 h at 42 C. The membrane was then washed first by 100 ml of 2 x SSC, 0.1% SDS for 10 mm at room temperature for four times, followed by a final wash with 200 ml of 0.1 x SSC, 0.1% SDS for 30 min at 50 C. The membrane 112 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was finally exposed to Kodak X-omat AR film (Eastman Kodak, Rochester, NY) at -80 C using an intensifying screen for autoradiography. To genotype mouse progenies, mouse tails were cut at three-week old, and genomic DNA was extracted according to the described (Hogan et aL, 1994 ). The genomic DNA was digested with BamHT and analyzed by Southern blot as described above. 5-2-6. Freezing heterozygous /gr-disrupted ES clones and culturing cells for blastocyst injection. Heterozygous rgr-disrupted ES clones were confirmed by Southern blot analysis. Frozen 96-well stock plates holding these clones were thawed in a 37 C incubator, and the entire content of cells from one confirmed clone was transferred to a well of a 24-well plate to expand. Selection drugs were not contained in the culturing medium in the subsequent cell culture. Clones from CJ7 strain were then passaged to a 10-cm plate, while clones from R1 strain were to a 3.5-cm plate, hi average, 15 aliquots from each CJ7 clone and 8 aliquots from each R1 clone were frozen down in DMEM with 10% fetal bovine serum and 10% DMSO (Sigma, St. Louis, MO). To culture cells for blastocyst injection, one aliquot was thawed in 37 C water bath followed by spinning cells down to remove DMSO. For CJ7 strain, cells were seeded in a 3.5-cm plate and cultured through two passages in 3.5-cm plates for 4 to 5 days before collecting cells for blastocyst injection. Cells from R1 strain were cultured similarly except that cells were first seeded in 24-well plate and then passaged to 3.5-cm plates for subsequent culturing. 5-2-7. Blastocyst injection. C57BL/6J blastocysts at 3.5-day old from both sexually mature adult females and 3-week old hormone-primed mice were used for injection. ES cells were trypsinized, 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disassociated by pipetting, and suspended in culturing medium. Cells were suspended in the medium in a 10-cm plate for 30 min in incubator to precipitate larger feeder cells. The cell suspension was then transferred to a clean tube and centriiuged to pellet ES cells. The ES cells were finally suspended in the culturing medium for blastocyst injection. During injection, a small portion of cells was loaded into the injection chamber containing DMEM medium and 10% fetal bovine serum buffered with 20 mM HEPES (Ultrol grade, Calbiochem, La Jolla, CA), the remaining cells were kept on ice. IS to 20 cells were injected per blastocyst. Blastocysts before and after injection were cultured in M16 medium (Catalog No. M-1285, Sigma, St. Louis, MO) in the 5% CO2 incubator at 37 C. After injection, 5 to 8 blastocysts were transferred to the uterine hom at each side of a 2.5-day pseudopregnant CD-I female. Starting from August 21, 1998, part of the blastocyst injection procedure was modified. ES cells intended for injection were trypsinized, and trypsin was inactivated with culturing medium. The cells were then spun down and washed once by suspending and spinning down cells in 25mM-HEPES-buffered-DMEM (Gibco, Gaithersburg, MD), supplemented with 10% fetal bovine serum (Lot AGD6411, HyClone, Logan, UT), and 2 mM glutamine, 100 unit/ml penicillin, 100 ng/ml streptomycin (Irvine Scientific, Santa Ana, CA). ES cells were then separated from feeder cells and finally suspended in the HEPES-buffered DMEM for injection. 5-3. Results 5-3-1. Characterization of mouse RGR cDNA and genomic DNA clones. The MRGR7-5 cDNA clone was isolated from a mouse retina cDNA library and was shown to contain the entire protein-coding region of RGR. The sequence of MRGR7-5 cDNA is 1493 nucleotides in length (deposited under GenBank accession number 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AF076930). Translation of the cDNA sequence from its 5'-most ATG codon to the in- phase stop codon yields an open reading frame of 291 amino acids with a calculated molecular weight of 32,124 (Fig. 5-1). The deduced amino acid sequence of mouse RGR is 81% and 78% identical to that of human and bovine RGR, respectively. Lys255, homologous with the retinaldehyde attachment site in visual pigments, is conserved in the seventh transmembrane domain of RGR from all species. Two overlapping genomic DNA clones, kmrgr9 and kmrgrl 1, were used to obtain a restriction map of the entire 129SV mouse RGR gene (Fig. 5-2). The gene is split into seven exons and spans approximately 11 kb. The exon-intron structure of the mouse RGR gene conforms to that of the human gene. 5-3-2. Homologous recombination of rgr gene in ES cells. The targeting vector used to disrupt the rgr gene contains a 5.1-kb mouse genomic gene fragment extending from 0.2-kb upstream of exon 1 to 0.8-kb downstream of exon 4. A neo expression cassette was inserted in the middle of exon 2 with the same transcriptional orientation as rgr gene, creating a targeting construct with 2.4-kb homology at 5' flank and 2.7-kb homology at 3' flank of neo gene. This arrangement utilizes not only the neo gene as a drug selection marker but also its polyadenylation sequence at the 3' end to truncate transcription initiated from rgr gene, ensuring the disruption of rgr gene function. To increase the homologous recombinant frequency after drug selection, tk gene was added to the 3' end to fulfill a double-drug selection strategy (Fig. 5-2). 25 ng of vector DNA was linearized and electroporated into -lO? of CJ7 and Ri cells. Drug selection began at 48 hours after electroporation. Both 300 (tg/mL G418 and 0.2 (iM FIAU were added to the culturing medium for CJ7 cells at 48 hours. For Rl cells, G418 selection was started 48 hours but FIAU was applied at 72 hours. About 600-700 drug resistant colonies were observed on 5, 10-cm plates 9 days after electroporation. 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S45 resistant colonies from CJ7 cells and 552 from R1 cells were picked and expanded. Genomic DNAs were extracted from all resistant clones and digested with BamHI. Southern blot was performed using a 3-kb mouse genomic Seal fragment as the probe, which hybridized to one 7.0-kb fragment and one 2.2-kb fragment resulting from BamHI- digested wildtype rgr gene. Homologous recombination in one of two rgr alleles is predicted to result in the insertion of 1.7-kb neo gene into exon 2 that is contained in the 7.0-kb fragment, thus creating a longer fragment of 8.7 kb in addition to the wildtype 7.0- kb and 2.2-kb fragments (Fig. 5-2). Four clones from CJ7 strain and 18 clones from R1 strain were demonstrated to have a predicted band shift to 8.7 kb (Fig. 5-3), giving targeting frequencies of 0.7% and 3% for CJ7 and Rl, respectively. These values fall in the normal frequency range of 1/10 to 1/1000 for gene-targeting using double-drug selection strategy (Ranrirez-Solis et aL, 1993). 5-3-3. Generation of germline-transmitted rgr gene disruption. So far, four heterozygously recombinant clones from CJ7 strain, and four from Rl stain had been tested for their ability to incorporate into blastocysts and contribute to germline transmission. The result is summarized in the table 5-1. Among all eight clones tested, there had been a difficulty in having the implanted foster mothers give birth to a litter. The best result came with 1C 10 clone of CJ7 strain, with which 21 out of 96 injected blastocysts, or 22%, were bom. Compared with data in the literature, this was at the lower end for a successful blastocyst injection (McMahon and Bradley, 1990; Soriano et aL, 1991). For the rest of clones, the percentage ranged from 0 to 20%. To raise the survival of the injected blastocysts in foster mothers, we are overhauling the injection procedures and trying to make some modifications, using medium without (3 - mercaptoethanol to suspend ES cells for injection and accelerating the injection pace to shorten the time for ES cells to stand on ice. 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. We generated 9 chimeric mice with 7 from 1C 10 clone and 1 from 6F3 clone of CJ7 strain, and 1 from 8D6 clone of Rl strain. We failed to observe the sex bias toward high percentage of male mice that could result from sex conversion of XX blastocysts by injected XY ES cells. However, most chimeric mice had a high level of agouti pigmentation in their coats (from 60% to mostly 90-100%), only one was estimated with 30% agouti contribution. All male chimeric mice were generated with 1C 10 clone. One of them produced progeny (no less than 35) whose coats were black, indicating that this male chimera did not carry ES-cell-derived germ cells. The other two had been proven to be infertile, a phenomenon that is not uncommon in highly chimeric mice (Patek, C. E. et aL, 1991). Of all six female chimeric mice, one generated from 8D6 clone is currently being mated with C57BL/6J male; one from 6F3 usually gave birth to a litter of 1-2 pups that died one or two days later; three from 1C 10 rarely gave birth; the last one, named 1C10-HF2, from 1C 10 was proven to be fertile and carry ES-cell-originated germ cells that can be transmitted to progenies. 1C10-HF2 has been mated with a C57BL/6J male mouse. From 16 surviving offspring out of 5 litters, four mice had agouti coats. Among them, two females, named 1F1 (DOB: 03/20/98) and 1F3 (DOB: 05/28/98), and one male, named 1M1 (DOB: 05/28/98), were confirmed to be heterozygous mice mutated at rgr alleles. Efforts are being taken to mate these heterozygous mice to generate homozygous rgr mutant mice for phenotype analysis. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 5-1. Amino acid sequence of mouse RGR aligned with that of human and bovine RGR. For comparison, only amino acid (one-letter code) sequence differences in the human and bovine proteins are shown, and identities to mouse RGR are indicated by dashes. A conserved lysine residue, which is homologous to the retinal attachment site of visual pigments, is marked by the triangle. The nucleotide sequence of the mouse RGR cDNA is deposited under GenBank accession number AF076930. 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. MAATRALPAGLGELEVLAVGTVLLMEALSGISLNGLTIFSFCKTPDLRTPSNLLVLSLAL --E-S T-F----------M---V----- L---T----------- E----CH--------- --ESGT--T-F---------------V-----L---I---L-------E----- H--------- ADTGISLNALVAAVSSLLRRWPHGSEGCQVHGFQGFATALASICGSAAVAWGRYHHYCTR __S-----------T---------Y--D---A-------V--------S---1------------- - -S----------- T---------Y------ A-------V--------S------------- F-- RQLAWDTAIPLVLFVWMSSAFWASLPLMGWGHYDYEPVGTCCTLDYSRGDRNFISFLFTM S NS-VS------L-------A---L---------- L----------K-----T------ SR-D-N- -VS- -F--L-------A---L---------- L-----------------T------ AFFNFLVPLFITHTSYRFMEQKFSRSGHLPVNTTLPGRMLLLGWGPYALLYLYAAIADVS S AM-----1---SL----LGK----Q-------A-T----------1-----V T ------L----- VV---L----LGKTSRP----V--A-T----------------- T AT T FISPKLQMVPALIAKTMPTINAINYALHREMVCRGTWQCLSPQKSKKDRTQ* mouse s----------------Mv----------- gn-------1---------RE----K* human s----------------AV__V__M GS H — i--------RREHS-E-* bovine VO 60 120 180 240 291 Fig. 5-2. Targeted disruption of rgr gene Top: Genomic gene structure and restriction enzyme map of wildtype rgr allele. A SacI fragment containing exon 5 was used as probe to determine allele specific recombination by Southern blot analysis. Middle: rgr gene targeting vector. The vector contains 5.1-kb rgr genomic fragment interrupted by insertion of neo expression cassette in exon 2, creating a targeting construct with 2.4 kb of homology S' and 2.7 kb 3' of neo gene. The 3' end of genomic DNA was flanked with tk gene to create the targeting vector for double drug selection. Bottom: The predicted gene structure of mutated rgr allele after homologous recombination. Homologous recombination at rgr allele results in the insertion of L.7-kb neo gene into exon 2, which can be picked up in Southern blot as a larger 8.7-kb BamHI- cut fragment than the wildtype 7.0-kb one. 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 kb co - co - co LU CO CO h CO - CD LU CO -a o O 4— > o 0 > O) c 0 O) v - 0 I - -a © K c o co — CO CO — CO r c o co CO L CO - CO - LU - CO »a C S CO — LU — LU CO CO- L *a C N _ CO c o - c o co LU CO CO CO CO CO CO CO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 .7 kb Fig. 5-3. Southern blot analysis of drag resistant ES clones targeted with RGR knockout vector. DNA simples from G418- and FIAU-resistant ES clones were extracted, digested with Bamffl, and probed with a exon 5-containing SacI genomic DNA fragment. Results from 38 clones of R l strain were shown here. Two bands at 7.0 kb and 2.2 kb were detected in 36 samples, indicating that no homologous recombination occured at rgr allele. Two clones, 10BS and 10F1, showed a predicted band shift at 8.7 kb in addition to 7.0-kb and 2.2-kb bands, suggesting that one of two rgr alleles was mutated through homologous recombination. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122 ♦ * ♦ ♦ n a n a • M M . T T 1 0 m o f f i < o ‘ < \ i < \ 1 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1. Summary of blastocyst injection with hemizygous rgr-disrupted GS cell clones ES cell strain Clone Blastocyst injected Bom Chimeric males Chimeric Average agouti Germline transmission females contribution from male chimeras Germline transmission from female chimeras 1C10 96 21 3 4 95% 0 1 CJ7 2D4 26 5 0 0 0 0 5E12 39 0 0 0 0 0 6F3 32 3 0 1 80% 0 0 7A2 12 0 0 0 0 0 Rl 8D6 34 5 0 1 80% 0 0 10BS 20 1 0 0 0 0 1 1 El 1 30 6 2 0 95% being tested 0 N > 4 ^ Fig. 5-4. Genotype identification of offspring derived from the rgr-disrupted hemizygotic mice. Mice that were disrupted at one of rgr alleles were mated. Genomic DNAs from eight pups of one litter were prepared and digested completely with BamHI. Southern blot was performed as described to genotype the offspring. Three of them, No. 1, 4, and 6, had homozygous disruption at rgr allele, four of them, No. 2, 3, S, and 8, had hemizygous disruption at rgr allele, and one, No. 7, had two wildtype allele. 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 6. CONCLUSION. The putative RPE-retinal G protein coupled receptor (RGR) is a novel opsin specifically expressed at high levels in the retinal pigment epithelium (RPE) and Muller cells of the neural retina (Jiang et aL, 1993). RGR is most akin in sequence to the nonvisual pigments of the opsin subfamily that include retinochrome, peropsin, melanopsin, and pineal opsin. A lysine residue, corresponding to retinal-binding site in visual pigments, is conserved in the seventh transmembrane segment and proposed to bind retinal chromophore. This research project has been focused on identifying retinal ligand that is bound to RGR at presumably the conserved lysine site and, based on the knowledge of ligand identity, further exploring RGR biological function. The identity of the ligand bound to RGR was determined through three steps. First, it was demonstrated that RGR prepared on the RPE microsomal membranes was capable of binding both all-trans- and 11-cis-retinal isomers. The binding was sensitive to the heat treatment unless the linkage between retinal and protein was reduced by sodium borohydride prior to heat boiling. This was consistent with the proposition that retinal was attached to a lysine residue in RGR via SchifFbase bond. Second, to confirm that the capability for RGR to bind retinal isomers in vitro reflects the existence of retinal-bound RGR in vivo, RGR was directly purified from fresh bovine eyes by immunoaffinity chromatography. The highly purified RGR gave rise to an unique pH-dependent absorption spectrum indicative of retinal-opsin complex. Finally, retinal isomer was isolated from the purified RGR in the dark and confirmed to be all-trans-retinal by HPLC analysis. 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Photochemical properties of RGR-all-trans-retinal were further investigated. Irradiation of all-trans-retinal-bound RGR with monochromatic lights at 370 nm and 470 nm, the absorption maxima c f RGR3 7 0 and R G R ^ respectively, resulted in specific isomerization of all-trans-retinal to 11-cis-retinaL The specific isomerization required the intact stereo structure of RGR since denaturation of the protein completely abolished the specificity of isomerization. The photosensitivity ofRGR4 6 9 is about one-third the level of that of rhodopsin but is significantly higher than that of retinal compounds and CRALBP, suggesting that RGR is a photopigment and the isomerization catalyzed by RGR in the presence of light could be closely related to its physiological function in the RPE. The specific isomerization of retinal ligands catalyzed by opsins in the presence of light is associated with the biological functions of the bound opsins in two ways. All visual pigments that function as photoreceptors bind 11-cis-retinal or its 3,4-dehydro derivative in the dark. Light-induced isomerization of 11-cis-retinal to all-trans configuration is served to trigger protein conformation change that consequently activates the photoreceptors and initiates phototransduction pathway (Findlay and Pappin, 1986; Gartner and Towner, 1995). Two nonvisual pigments bacteriorhodopsin and retinochrome bind all-trans-retinal in the dark. Bacteriorhodopsin resembles visual pigments in utilizing retinal ligand for its function. Light-induced isomerization between all-trans-retinal and 13-cis-retinal results in the conformation change of the protein that carries out the transportation of protons across membranes during the process (Findlay and Pappin, 1986). In contrast, the specific isomerization of all-trans-retinal into 11-cis- retinal catalyzed by retinochrome is proposed not as the trigger for protein activation but as the 11-cis-retinal-producing system for regeneration of bleached rhodopsin (Hara and Hara, 1987). RGR function may follow one of the two modes or a combination of the two. Isomerization of retinal ligand by RGR in the presence of light may be utilized to activate 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RGR that either initiates a signal transduction pathway or carries out non-receptor functions in the RPE, or to produce 11-cis-retinal for the visual cycle to regenerate bleached ihodopin in photoreceptor cells. Or RGR could carry out photoreceptor or nonphotorecptor functions while producing 11-cis-retinal for the visual cycle in the meantime. RGR-catalyzed isomerization of all-trans-retinoid to 11-cis-retinoid is the second protein-mediated traps to cis converting mechanism characterized in the RPE. Previously, an isomerahydrolase activity was detected in the RPE microsomal membranes that catalyzed hydrolysis and isomerization of all-trans-retinylester to 11-cis-retinol in one step by using chemical energy stored in membrane lipids (Rando et aL, 1991). RGR is different from isomerahydrolase in terms of the substrates and energy sources it uses. How the two isomerases contribute to the visual cycle in the RPE remains to be studied. To further characterize RGR function, it is important to find the protein(s) that functionally associated with RGR in the RPE. hi an in vitro reconstituted system consisting of squid rhodopisn, retinochrome, and RALBP, it was demonstrated that 11- cis-retinal generated by retinochrome could be downloaded to RALBP for regeneration of rhodopsin (Terakita et aL, 1989). A vertebrate counterpart of RALBP, CRALBP, has already been intensively studied. Investigation of the functional interaction between RGR and CRALBP might generate insight into the functions of both proteins and retinoid metabolism in the RPE. Another breakthrough may come from the identification of specific proteins that copurify with RGR during immunoaffinity chromatography. Confirmation of physical and functional interaction between RGR and any of these proteins will certainly expand understanding of RGR function. RGR-deleted knockout mouse will be another source of information about RGR function. The consequences of the loss of RGR in murine eyes will be investigated in several ways: the morphology of the eyes wQl be examined histologically, the composition of retinal isomers will be analyzed by HPLC, and the dark and light adaptation processes 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. will be studied electrophysilogically. Besides generating information about the function, the rgr knockout mouse will also be an important tool to study possible pathological effect of the loss of RGR on neural retinal degeneration and disturbance of circadian rhythm. 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig 6-1. Mammalian visual cycles, (modified from Saari, 1990) Events of the visual cycle that transport and metabolize retinoids in the mammalian neural retina and the retinal pigment epithelium are summarized by Saari The shaded area dipicts a possible novel pathway for RPE to generate 11-cis-retinaL The convertion from all-trans-retinal to 11-cis-reitnal is catalyzed by RGR in the presence of light. 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NEURAL RETINA RHODOPSIN LIGHT PROTEIN" OSPIN all-trans RETTNALDEHYDE 11-els RE TINALDEHYDE NADPH- NADP all-trans RETINOL RPE 11-els RETTNALDEHYDE NADH aH-trans RETINOL ■ N A D 11-cis RETINOL all-trans RETINYL ESTER 11-cls RETINYL ESTER all-trans RETINOL BLOOD 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES Acharya, S. and Kamik, S. S. (1996) J. Biol. Chem. 271: 25406-25411. Adler, A. J. and Martin, K. J. (1982) Biochem. Biophys. Res. 108: 1601-1608. Asenjo, A. B., Rim, J., and Oprian, D. D. (1994) Neuron 12: 1131-1138. Ball, S., Goodwin, T. W., and Morton, R. A. (1948) Biochem. J. 42: 516-523. Banerjee, P., Joo, J. B., Buse, J. T., Dawson, G. (1995) Chem. Phys. Lipids 77: 65-78. Barry, R. J., Canada, F. J., and Rando, R. R. (1989) J. Biol. Chem. 264: 9231-9238. Bavik, C.-O., Eriksson, U., Allen, R. A., and Peterson, P. A. (1991) J. Biol. Chem. 266: 14978-14985. Bavik, C.-O., Busch, C., and Eriksson, U. (1992) J. Biol. Chem. 267: 23035-23042. Bavik, C.-O., Levy, F., Heilman, U., Wemstedt, C. and Eriksson, U. (1993) J. Biol. Chem. 268: 20540-20546. Berman, E. R., Segal, N., Rothman, H., and Weiner, A. (1985) Curr. Eye Res. 4: 867-876. Bernstein, P. S., Law, W., C., and Rando, R. R. (1987) Proc. Natl. Acad. Sci. 84: 1849-1853. Blaner, W. S., Das, S. R., Gouras, P., and Flood, M. T. (1987) J. Biol. Chem. 262: 53-58. Blomhoffj R., Green, M H., Berg, T., and Norum, K. R. (1990) Science 250: 399-404. Bok, D. (1993) J. Cell Biol. Supplement 17: 189-195. Bownds, D. (1967) Nature 216: 1178-1181. Bridges, C. D. B. (1976) Exp. Eye Res. 22: 435-455. Carlson, A. and Bok, D. (1992) Biochemistry 31: 9056-9032. Chen, J., Makino, C. L., Peachey, N. S., Baylor, D. A., and Simon, M. I. (1995) Science 267: 374-377. 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chen, Y. andNoy, N. (1994) Biochemistry 33: 10658-10665. Crabb, J. W., Johnson, C. M., Carr, S. A., Armes, L. G., and Saari, J. C. (1988) J. Biol. Chem. 263: 18678-18687. Dartnall, H. J. A. (1972) in Handbook O f Sensory Physiology VoL VII, pp 122-145. Springer-Verlag, New York. Deigner, P. S., Law, W. C., Canada, F. J., Rando, R. R. (1989) Science 244: 968-971. Dowling, J. E. (1960) Nature 188: 114-118. Dowling, J. E. (1987) In The retina, an approachable part o f the brain, pp 13-41. The Belknap Press of Harvard University Press. Dratz, E. A. and Hagrave, P. A. (1983) Trends Biochem. Sci. 8: 128-131. Driessen, C. A. G. G., Janssen, B. P. M., Winkens, H. J., van Vugt, A. H. M., de Leeuw, T. L. M., and Janssen, J. J. M. (1995) Invest. Ophthalmol. Vis. Sci. 36: 1988-1996. Duke-Elder, S. (1963) System of Ophthalmology: Normal and Abnormal Development. Vol. Ill, The C. V. Mosby Co., St. Louis. Edwards, R. B. and Adler, A. J. (1994) Exp. Eye. Res. 59: 161-170. Ernst, O. P., Hofmann, K. P., and Sakmar, T. P. (1995) J. Biol. Chem. 270: 10580-10586. Findlay, J. B. C. and Pappin, D. J. C. (1986) Biochem. J. 238: 625-642. Fong, S. L., Tsin, A. T. C., Bridges, C. D. B., and Liou, G. I. (1982) Meth. Enzymol. 81: 133-140. Fong, S. L., Liou, G. I., Landers, R. A., Alvarez, R. A., and Bridges, C. D. (1984) J. Biol. Chem. 259: 6534-6542. Forsman, K. (1992) Clin. Genet. 42: 156-159. Futterman, S., Saari, J. C., and Blair, S. (1977) J. Biol. Chem. 252: 3267-3271. Gartner, W. and Towner, P. (1995) Photobiochem. Photobiol. 62: 1-16. 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Groenendijk, G. W. T., De Grip, W. J., and Daemen, F. J. M. (1979) Anal. Biochem. 99: 304-310. Groenendijk, G. W. T., De Grip, and Daemen, F. J. M. (1980) Biochim. Biophys. Acta 617: 430-438. Gu, S., Thompson, D. A., Srisailapathy Srikumari, C. R _, Lorenz, B., Finckh, U., Nicoletti, A., Murthy, K. R., Rathmann, M., Kumaramanickavel, G., Denton, M. J., Gal, A. (1997) Nature Genet. 17: 194-197. Hao, W. and Fong H. K. W. (1996) Biochemistry 35: 6251-6256. Hao, W. and Fong H. K. W. (1998) In submission. Hara, T. and Hara, R. (1965) Nature 206: 1331-1334. Hara, T. and Hara R. (1982) mMeth. Enzymol. 81: 190-197. Hara, T. and Hara, R. (1987) in Retinal Proteins (Ovchinnikov, Y. A., Ed.) pp 457-466, VNU Science Press, Utrecht, The Netherlands. Hara-Nishimura. I, Matsumoto, T., Mori. H., Nishimura, M., Hara, R. and Hara, T. (1990) FEBS Lett. 271: 106-110. Hamel, C. P., Tsilou, E., Harris, E., Pfeffer, B. A., Hooks, J. J., Detrick, B., and Redmond, T. M. (1993a) J. Neurosci. Res. 34: 414-425. Hamel, C. P., Tsilou, E., Pfeffer, B. A., Hooks, J. J., Detrick, B., and Redmond, T. M. (1993b) J. Biol. Chem. 268: 15751-15757. Hillman p., Hochstein, S., and Minke, B. (1983) Physiol. Rev. 63: 668-760. Hogan, B., Beddington, R., Costantini, F., and Lacy, E. (1994) Manipulating the Mouse Embryo: A Laboratory Manual (2nd edn). Cold Spring Harbor Laboratory: Cold Spring Harbor, U. S. A. Hubbard, R. and St. George, R. C. C. (1958) J. Gen. Physiol. 41: 501-528. Hubbard, R., Brown, P. K., and Bownds, D. (1971) Meth Enzymol. 18: 615-653. Hofmann, K. P., Pulvermuller, A., Buczylko, J., Hooser, P. V., and Palczewski, K (1992) J. Biol. Chem. 267: 15701-15706. Hofmann, K. P. and Heck, M. (1996) Biomembranes Vol. 2A: 141-198. 135 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hong, K L , Knudsen, P. J. and Hubell, W. L. (1982) Meth. Enzymol. 81: 144-150. Honig, B., Greenberg, A. D., Dinur, U., and Ebrey, T. G. (1976) Biochemistry 21: 4593-4599. Hubbard, R., Brown, P. K., and Bownds, D. (1971) in . Meth. Enzymol. 18: 615-613. Ishiguro, S., Suzuki, Y., Tamai, M., and Mizuno, K. (1991) J. Biol. Chem. 266: 15520-15524. Jiang, M., Pandey, S., and Fong, H. K W. (1993) Invest. Ophthalmol. Visual Sci. 34: 3669-3678. Khorana, H. G. (1992) J. Biol. Chem. 267: 1-4. Klein, R., Klein, B. E. K., and Linton, K. L. P. (1992) Ophthalmology 99: 933-943. Lai, Y. L., Wiggert, B., Liu, Y. P., and Chader, G. J. (1982) Nature 298: 848-849. Lion, F., Rotmans, J. P., Daemen, F. J. M., and Bonting, S. L. (1975) Biochim. Biophys. Acta. 384: 283-292. Liu, R. S. H., Matsumoto, H., Kini, A., Asato, A. E., Denny, M., Krop£ A., and DeGrip, W. J. (1984) Tetrahedron 40: 473-482. Loppnow, G. R., Barry, B. A., and Mathies, R. A. (1989) Proc. Natl. Acad. Sci. 86: 1515-1518. MacDonald, P. N. and Ong, D. E. (1988) J. Biol. Chem. 263: 12478-12482. Mata, N. L., VQlazana, E. T., and Tsin, A. T. C. (1998) Invest. Ophthal. Vis. Sci. 39: 1312-1319. Matthews, R. G., Hubbard, R., Brown, P. K., and Wald, G. (1963) J. Gen. Physiol. AT. 215-240. Maden, B. E. H. (1995) Essay Biochem. 29: 87-110. Maw, M. A., Kennedy, B., Knight, A., Bridges, R., Roth, K. E., Mani, E. J., Mukkadan, J. K., Nancarrow, D., Crabb, J. C., and Denton, M. J. (1997) Nature Genet. 17: 198-200. 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Max, M., McKinnon, P. J., Seidenmen, K. J., Barrett, R. K., Applebury, M. L., Takahashi, J. S., and Margolskee, R. F. (1995) Science 267: 1502-1506. McMahon, A. P. and Bradley, A. (1990) Cell 62: 1073-1085. Merbs, S. L. and Nathans, J. (1993) Photobiochem. Photobiol. 58: 706-710. Miller, S. S. and Steinberg, R. H. (1982) J. Membr. Biol. 67: 199-209. Morton, R. A. and Pitt, G. A. J. (1954) Biochem. J. 59: 128-134. Mullen, R. J. and LaVail, M. M (1976) Science 192: 799-801. Nathans, J. and Hogness, D. S. (1983) Cell 34: 807-814. Nathans, J. (1987) Anrtu. Rev. Neurosci. 10: 163-194. Nathans, J. (1990) Biochemistry 29: 9746-9752. Okano, T., Yoshizawa, T., and Fukada, Y. (1994) Nature 372: 94-97. Okajima, T. L., Pepperberg, D. R., Ripps, H., Wiggert, B., and Chader, G. J. (1989) Exp. Eye Res. 49: 629-644. Ozaki, K., Hara, R., Hara, T., and Kakitani, T. (1983) Biophys. J. 44: 127-137. Ozaki, K., Terakita, A., Hara, R., and Hara, T. (1986) Vision Res. 26: 691-705. Pak, W. L. and Shortridge, R. D. (1991) Photochem. Photobiol. 53: 871-875. Pandey, S., Blanks, J. C., Spee, C., Jiang, M , and Fong, H. K. W. (1994) Exp. Eye Res. 58: 605-614. Patek, C. E., Kerr, J. B., Goslen, R. G., Jones, K. W., Hardy, K., Muggleton-Harris, A. L., Handyside, A. H., Wittmgham, D. G., and Hooper, M L. (1991) Development 113:311-325 Petrukhin, K Koisti, M. J., Bakall, B., Li, W., Xie, G., Marknell, T., Sandgren, O., Forsman, K., Holmgren, G., Andreasson, S., Vujic, M., Bergen, A. A. B., McGarty-Dugan, V., Figueroa, D., Austin, C. P., Metzker, M. L., Caskey, C. T., ahid Wadehus, C. (1998) Nature Genet. 19: 241-247. Pitt, G. A. J., Collins, F. D., Morton, R. A., and Stok, P. (1955) Biochem. J. 59: 122-128. 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Provencio, I., Jiang, G., De Grip, W. J., Hayes, W. P., and rollag, M. D. (1998) Proc. Natl. Acad. Sci. 95: 340-345. Ramirez-solis, R., Davis, A. C., and Bradley, A. (1993) in Meth. Enzymol. 225: 855-878. Rando, R. R., Bernstein, P. S., and Barry, R. J. (1991) in Progress in Retinal Research (Osborne, N. N. and Chader, G. J., Eds.) VoL 10, pp 161-178, Pergamon Press, New York. Rao, V. R. and Oprian, D, D. (1996) Anrtu. Rev. Biophys. Biomol. Struct. 25: 287-341. Saari, J. C., Bredberg, L., and Garwin, G. G. (1982) J. Biol. Chem. 257: 13329-13333. Saari, J. C., Bunt-Milam, A. H., Bredberg, D. L., and Garwin, G. G. (1984) Vision Res. 24: 1595-1603. Saari, J. C., Teller, D. C., Crabb, J. W., and Bredberg, L. (1985) J. Biol. Chem. 260: 195-201. Saari, J. C. and Bredberg, D. L. (1987) J. Biol. Chem. 262: 7618-7622. Saari, J. C. and Bredberg, D. L. (1989) J. Biol. Chem. 264: 8636-8640. Saari, J. C. (1990) in Progress in Retinal Research (Osborne, N. N., and Chader, F. J., Eds.) VoL9, pp 363-381, Pergamon Press, New York. Saari, J. C., Bredberg, D. L., and Noy, N. (1994) Biochemistry 33: 3106-3112. Sarkmar, T. P., Franke, R. R., and Khorana, H. G. (1989) Proc. Natl. Acad Sci. 86: 8309-8313. Sarkmar, T. P., Franke, R. R., and Khorana, H. G. (1991) Proc. Natl. Acad Sci. 88: 3079-3083. Saranak, J. and Foster, K. W. (1997) Nature 387: 465-466. Shen, D., Jiang, M., Hao, W., Tao, L., Salazar, M., and Fong, H. K. W. (1994) Biochemsitry 33: 13117-13125. Simon, A., Heilman, U., Wemstedt, C., and Eriksson, U. (1995) J. Biol. Chem. 270: 1107-1112. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. (1991) Cell 64: 693-702. 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stoeckenius, W. and Bogomolni, R. A. (1982) Annu. Rev. Biochem. 51: 587-616. Stone, E. M., Nichols, B. E., Streb, L. M., Kimura, A. E., and Sheffield, V. C. (1992) Nature Genet. 1: 246-250. Stryer, L. (1988) Cold Spring Harbor Symp. Quant. Biol. 53: 283-294. Sun. R , Gilbert, D., Copeland. N. G., Jenkins, N. A., and Nathans, J. (1997) Proc. Natl. Acad. Sci. 94:9893-9898. Terakita, A , Hara, R., Hara, T. (1989) Vision Res. 29: 639-652. Trehan, A , Canada, F. J., and Rando, R. R (1990) Biochemistry 29: 309-12. Wald, G. (1968) Nature 219: 800-807. Wang, J. K., McDowell, R J., and Hagrave, P. A (1980) Biochemsitry 19: 5111-5117. Winston, A and Rando, R. R. (1998) Biochemistry 37: 2044-2050. Young, R. W. and Bok, D. (1969) J. Cell Biol. 42: 392-403. Zhukovsky, E. A and Oprian, D. D. (1989) Science 246: 928-930. Zimmerman, W. F. (1974) Vision Res. 14: 795-802. Zimmerman, W. F., Lion, F., Daemen, F. J. M., and Bonting, S. L. (1975) Exp. Eye Res. 21: 325-332. Zimmerman, W. F. (1976) Exp. Eye Res. 23: 159-164. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IMAftF FVAI I IATIHM .................. • W M % I I w t 1 T E ST T A R G E T (QA-3) / 1 . 0 K e . ---------- Ui 1m l e i e i .i 1 “ 1 liU lk m 1 18 ■ ■ ■ II 1 2 | u | 1.6 150mm IIVMGE. Inc 1653 East Main Street Rochester, NY 14609 USA Phone: 716/482-0300 Fax: 716/288-5989 0 1993. Applied Image. Inc.. A ll Rights Reserved Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Expression of the RGR opsin and its function in the photic visual cycle

PDF

Functional analysis of MSX2 and its role in skull patterning

PDF

A new model for hepatitis delta virus transcription and replication

PDF

Analysis of mesodermal determination genes in culture

PDF

Immunogenicity of acute lymphoblastic leukemia: In vivo and in vitro studies of Bcr -Abl specific immune responses

PDF

Cloning and characterization of Grb4: A novel adapter protein interacting with BCR -Abl

PDF

Analysis of multistep mammary tumorigenesis in Wnt1 transgenic mice: The role of fibroblast growth factor-8 in oncogenesis and apoptosis

PDF

Characterization of signal input and signal output domains of phosphorus 160 coactivators

PDF

A POU gene expressed in the early sea urchin embryo: Structure, evolution and requirement for embryonic development

PDF

Characterization of target cell entry by murine leukemia viruses

PDF

Biochemical and genetic studies on Escherichia coli DNA polymerase II, a member of group B ``alpha-like'' DNA polymerases

PDF

Cytokine regulation of retinal pigment epithelial cell function and behavior

PDF

Response Of Xanthosine 5'-Phosphate Aminase To Adenine Nucleosides: Enzymology And Function

PDF

Isolation, Separation, And Structural Identification Of Streptomyces Quinones

PDF

Cellular factors involved in mouse hepatitis virus replication

PDF

Cloning, characterization, anti -apoptotic molecular mechanism, transgenic and protein binding studies of mouse TOSO gene

PDF

Reactivation Of Phage-P2 Damaged By Ultraviolet Light

PDF

A novel reaction of mismatched cytosine-cytosine pairs associated with Fragile X

PDF

A Study Of The Mechanism Of Lead Chromate Induced Neoplastic Transformation Of C3H/10T1/2 Mouse Embryo Cells

PDF

Characterization of the interactions between cellular proteins and hepatitis C virus core protein.

Asset Metadata

Creator Hao, Wenshan (author)

Core Title Characterization of the physiological ligand and function of a novel opsin RGR from the retinal pigment epithelium

Degree Doctor of Philosophy

Degree Program Molecular Microbiology and Immunology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, genetics,chemistry, biochemistry,health Sciences, ophthalmology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Fong, Henry K.W. (committee chair), Lai, Michael (committee member), Stallcup, Michael R. (committee member), Tahara, Stanly M. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-426130

Unique identifier UC11350929

Identifier 9931901.pdf (filename),usctheses-c17-426130 (legacy record id)

Legacy Identifier 9931901.pdf

Dmrecord 426130

Document Type Dissertation

Rights Hao, Wenshan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA