Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Glucose metabolism in the outer segments of photoreceptor cells : its involvement in the phototransduction… Hsu, Shu-Chan 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_phd_hsu_shu-chan.pdf [ 6.28MB ]
JSON: 831-1.0086355.json
JSON-LD: 831-1.0086355-ld.json
RDF/XML (Pretty): 831-1.0086355-rdf.xml
RDF/JSON: 831-1.0086355-rdf.json
Turtle: 831-1.0086355-turtle.txt
N-Triples: 831-1.0086355-rdf-ntriples.txt
Original Record: 831-1.0086355-source.json
Full Text

Full Text

GLUCOSE METABOLISM IN THE OUTER SEGMENTS OFPHOTORECEPTOR CELLS: ITS INVOLVEMENT IN THEPHOTOTRANSDUCTION PROCESSbySHU-CHAN HSUB.Sc., University of British Columbia, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF BIOCHEMISTRYUNIVERSITY OF BRITISH COLUMBIAWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF^SH COLUMBIAApril, 1993© Shu-Chan Hsu,1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of^V-3 IOCA-ki\A 1 SThe University of British ColumbiaVancouver, CanadaDate ^rDE-6 (2/88)i iABSTRACTThe existence and function of glucose metabolism in the outer segments ofphotoreceptor cells were investigated using biochemical and immunocytochemicalapproaches. The presence of glycolytic enzymes in photoreceptor outer segmentswas detected by enzyme activity assays, Western blotting, and immunofluorescencemicroscopy. Activities of six glycolytic enzymes including hexokinase,phosphofructokinase, aldolase, glyceraldehyde-3-phosphate dehydrogenase,phosphoglycerate kinase, pyruvate kinase, and lactate dehydrogenase, were found tobe present in purified rod outer segment preparations in quantities similar to thatfound in human red blood cells. Immunofluorescence microscopy of bovine andchicken retina sections labeled with monoclonal antibodies against glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, and lactate dehydrogenasehave confirmed that these enzymes are present in rod and cone outer segments andare not simply contaminants from the inner segments or other retinal cells. One ofthese glycolytic enzymes, glyceraldehyde-3-phosphate dehydrogenase, was found tomake up approximately 2 % of the total rod outer segment protein and over 11 % ofthe plasma membrane protein. It has been purified by affinity chromatography on aNAD +-agarose column and shown to associate reversibly and electrostatically witha specific protein in the rod outer segment plasma membrane.The rod outer segment plasma membrane was also found to contain aGLUT-1 type glucose transporter of Mr 45K as detected by 3-O-methylglucoseuptake and exchange studies and Western blot analyses using type-specific glucosetransporter antibodies. Solid-phase radioimmune competitive inhibition studiesindicated that the rod outer segment plasma membrane contained 15 % the numberof glucose transporters found in human red blood cell membranes and had anestimated density of 400 glucose transporter per um 2 of rod outer segment plasmaiiimembrane. Immunofluorescence microscopy indicated that both rod and coneouter segments have a GLUT-1 type glucose transporter.The involvement of glucose metabolism in supporting the phototransductionprocess was studied by measuring the activities of glycolytic and hexosemonophosphate pathways in isolated rod outer segments by spectrophotometric andradiometric enzyme assays. Glucose metabolism in rod outer segments was foundto produce both ATP and NADPH required for the maintenance ofphototransduction. ATP produced by glycolysis at a rate of 35-44 nmol/min/mg rodouter segment protein can potentially support cGMP regeneration, one of the mostenergy-consuming processes in phototransduction, under dark but not lightconditions. NADPH produced by the hexose monophosphate pathway at a maximalrate of 40 nmol/min/mg ROS protein is sufficient to support the reduction of all-trans-retinal to all-trans-retinol in rod outer segments occurring at a rate of 1.2nmol/min/mg ROS protein. A high hexose monophosphate pathway capacitysuggests that the pathway may also be involved in supporting the glutathione redoxcycle to protect rod outer segments from oxidative stress.In summary, photoreceptor outer segments contain enzymes involved inglucose metabolism and a GLUT-1 type glucose transporter for glucose supply.Glucose metabolism in this organelle can potentially function to maintain a constantROS cGMP concentration in dark, to buffer against sudden changes in cytoplasmicATP concentration upon visual excitation and to supply NADPH required for visualrecovery.ivTABLE OF CONTENTSPageABSTRACT^ iiTABLE OF CONTENTS^ ivLIST OF TABLES viiiLIST OF FIGURES^ ixLIST OF ABBREVIATIONS^ xiACKNOWLEDGEMENTS xiiiCHAPTER 1INTRODUCTION1.1.^RETINA^ 11.2. PHOTORECEPTOR CELLS^ 11.3. PHOTORECEPTOR OUTER SEGMENTS^ 41.3.1.^Structure ^ 41.3.2.^Protein composition ^ 41.4. PHOTOTRANSDUCTION 51.4.1.^Dark state ^ 51.4.2.^Visual excitation 91.4.3.^Visual recovery^ 121.5. GLUCOSE TRANSPORT 161.5.1.^Facilitative glucose transporters ^ 161.5.2.^Na+ -glucose cotransporter 201.6. GLUCOSE OXIDATION^ 221.7. PHOSPHOCREATINE SHUTTLE^ 221.8. RETINAL GLUCOSE METABOLISM ANDPHOTOTRANSDUCTION^ 251.8.1.^El ectroretinogram 25^1.8.2.^Effects of glucose metabolism on retinal function^271.9. GLUCOSE METABOLISM OF RETINA^ 271.9.1.^Intact retina ^ 281.9.2.^Photoreceptor cells 291.9.3.^Histochemistry of retina sections ^ 311.9.4.^Photoreceptor outer segments 331.10. THESIS INVESTIGATION^ 35CHAPTER 2IDENTIFICATION OF A MAJOR PROTEIN ASSOCIATED WITH THE PLASMAMEMBRANE OF RETINAL PHOTORECEPTOR OUTER SEGMENTS ASGLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE2.1 MATERIALS^ 372.1.1. Animal tissues^ 372.1.2. Chemicals 372.1.3. Immunoreagents^ 372.2 METHODS^ 372.2.1. Preparation of bovine ROS^ 372.2.2. Isolation of ROS disk and plasma membranes^ 382.2.3. Preparation of bovine erythrocyte ghosts 392.2.4. Purification of G3PD^ 392.2.5. Assay of G3PD activity 402.2.6. Amino acid sequence analysis^ 402.2.7. Generation of anti-G3PD monoclonal antibody^ 402.2.8. Conditions for extraction of G3PD from ROS membranes ^ 412.2.9. Binding of G3PD to ROS disk and plasma membranes^ 412.2.10. Effect of trypsin and chymotrypsin on G3PD binding sites onROS membranes^ 42vi2.2.11.^Polyacrylamide gel electrophoresis^ 422.3. RESULTS 432.3.1. Identification of a ROS 38 kDa protein as G3PD^432.3.2. Nature of G3PD binding to ROS membranes 512.4. DISCUSSION^ 55CHAPTER 3DETECTION OF GLYCOLYTIC ENZYMES AND A GLUT-1 GLUCOSETRANSPORTER IN THE OUTER SEGMENTS OF ROD AND CONEPHOTORECEPTOR CELLS3.1 MATERIALS^ 613.1.1. Animal tissues^ 613.1.2. Chemicals 613.1.3. Immunoreagents^ 613.2. METHODS^ 613.2.1. Preparation of bovine ROS, ROS membranes and ROSlysates 613.2.2. Preparation of red blood cells, red blood cell ghosts and ratbrain microsomes^ 623.2.3. Glycolytic enzyme activity assays^ 623.2.4. Determination of glucose transport activity in intact ROS 633.2.5. Monoclonal and polyclonal antibodies^ 653.2.6. Radioimmune competitive inhibition assay 663.2.7. Immunofluorescence microscopy^ 663.2.8. Polyacrylamide gel electrophoresis and Western blotting ^ 673.3. RESULTS^ 683.3.1. Glycolytic enzymes in photoreceptor rod outer segments ^ 683.3.2. Glucose transport in photoreceptor outer segments^ 733.3.3. Cone outer segments also contain glycolytic enzymes and aGLUT-1 type glucose transporter^ 85vii3.4.^DISCUSSION^ 88CHAPTER 4GLUCOSE METABOLISM IN THE OUTER SEGMENTS OF RETINALPHOTORECEPTOR CELLS: ITS INVOLVEMENT IN THE MAINTENANCE OFTHE PHOTOTRANSDUCTION PROCESS4.1.^MATERIALS^ 914.1.1.^Materials 914.1.2.^Chemicals^ 914.2.^METHODS 914.2.1.^Preparation of bovine ROS^ 914.2.2.^Quantitation of glycolysis, hexose monophosphate pathway andretinal reduction in bovine ROS preparation^924.2.3.^Glycolytic enzyme assays^ 944.2.4.^Assessment of ROS purity 954.3.^RESULTS 954.3.1.^Glycolytic flux in ROS^ 954.3.2.^Hexose monophosphate pathway^ 1014.3.3.^Retinal reduction^ 1034.4.^DISCUSSION^ 106SUMMARY^ 109REFERENCES 115viiiLIST OF TABLESTable^ PageI. Protein composition of bovine ROS disk and plasma membranes^7II. Characteristics of mammalian glucose transporters^ 17III. Purification of glyceraldehyde-3-phosphate dehydrogenase frombovine ROS^ 47IV. Purification of glyceraldehyde-3-phosphate dehydrogenase frombovine ROS plasma membranes^ 48V. Effect of nucleotides and chelating agents on the binding of the38 kDa protein to ROS membranes 54VI. Effect of trypsin and chymotrypsin on ROS G3PD binding sites^57VII. Glycolytic enzyme activities in bovine ROS^ 69VIIb.^Comparison of bovine ROS glycolytic enzyme activities with valuesreported in the literatures^ 70VIII. Glycolytic enzyme activity ratios in bovine ROS and other tissues^72IX. Kinetic properties of ROS and erythrocyte glucose transporter^80X. Specificity of glycolytic enzymes for guanine and adenine nucleotides ^99XI.^Glucose utilization by bovine ROS^ 104ixLIST OF FIGURESFigure^ Page 1. Synaptic connections of neural retina^ 22. Schematic diagram of rod and cone photoreceptor cells^33. Differential protein composition in the plasma and disk membranes of rodphotoreceptor cells^ 64. Dark current in the rod photoreceptor cell^ 85. Visual excitation and recovery in the rod photoreceptor cell^ 106. Structural model for facilitative glucose transporters 187. Schematic diagram illustrating the alternating conformational model forglucose transport by GLUT-1 glucose transporter^ 198. Structural model for Na t -glucose co-transporter in plasma membrane^219. Glucose metabolism via glycolysis, the hexose monophosphate pathwayand the tricarboxylic acid cycle^ 2310. Phosphocreatine shuttle mediates high energy phosphate transportbetween sperm mitochondrion and flagellum^ 2411. Electroretinogram illustrating the rat retinal response to lightillumination^ 2612. Localization of glycolytic enzymes in retina^ 3213. SDS-polyacrylamide gel electrophoresis of the 38 kDa protein isolatedfrom ROS plasma membrane^ 4414. The N-terminal amino acid sequence of bovine ROS 38 kDa protein andhomology to G3PD from other mammalian tissues^ 4515. Purification of the 38 kDa protein from bovine ROS membranes^5016. Anti-G3PD monoclonal antibody gpd 2C11-labeled Western blot of ROSplasma membrane and red blood cell ghosts and their NaCl extracts^5217. Effect of NaC1 concentration on the extraction of the 38 kDa proteinfrom ROS membranes^ 5318. Specific binding of the 38 kDa protein to bovine ROS plasma membrane ^5619. Western blots of ROS and purified G3PD and PGK labeled withmonoclonal antibodies against G3PD and PGK^ 7420. Localization of glycolytic enzymes in the photoreceptor layer of bovinexretina by immunofluorescence microscopy^ 7521. Effect of cytochalasin B on 3-0-methylglucose uptake by bovine ROS^7622. Inhibition of 3-0-methylglucose efflux from bovine ROS by phloretin^7823. Concentration dependence of 3-0-methylglucose net uptake andequilibrium exchange by bovine ROS^ 7924. Western blots of red blood cell membranes, brain microsomes, and ROSmembranes labeled for the GLUT-1 type glucose transporter^8225. Solid-phase radioimmune competitive inhibition assay of the GLUT-1glucose transporter in ROS and in red blood cell ghosts^8426. Localization of the GLUT-1 type glucose transporter in bovinephotoreceptors by immunofluorescence microscopy 8627. Localization of glycolytic enzymes and a GLUT-1 glucose transporter inthe photoreceptor cell layer of chicken retina by immunofluorescencemicroscopy^ 8728. Glycolytic flux in isolated bovine ROS^ 9729. Specificity of phosphofructokinase for ATP and GTP^ 10030. Hexose monophosphate pathway in isolated bovine ROS 10231. Retinal reduction in isolated bovine ROS^ 10532. Maintenance of phototransduction by glucose metabolism in photoreceptorinner and outer segments^ 114xiLIST OF ABBREVIATIONSADP^ adenosine 5'-diphosphateALD^ aldolaseATP^ adenosine 5'-triphosphateBCA^ bicinchoninic acidcGMP^ guanosine 3':5'-cyclic monophosphateCTAB^ cetyltrimethylammonium bromideEDTA^ ethylenediamine tetraacetateFITC^ fluorescein isothiocyanateG3PD^ glyceraldehyde-3-phosphate dehydrogenaseGDP^ guanosine 5'-diphosphateGMP ^ guanosine 5'-monophosphateGTP^ guanosine 5'-triphosphateHEPES ^ (N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid])HK^ hexokinaseHMP ^ hexose monophosphate pathwayIg^ immunoglobulinLDH^ lactate dehydrogenaseMr^molecular weightNAD^ nicotinamide adenine dinucleotideNADP ^ nicotinamide adenine dinucleotide phosphatePBS^ phosphate buffered salinePFK^ phosphofructokinasePGK^ phosphoglycerate kinasePK^ pyruvate kinaseRBC^ red blood cellsROS^ rod outer segmentsxi iSDS^ sodium dodecyl sulphateTCA^ tricarboxylic acid cycleTris^ (Tris[hydroxymethyl]aminomethane)w/v^ weight per volumeACKNOWLEDGEMENTI would like to thank Dr. Robert Molday for his excellent supervision and supportthroughout my stay in his lab. I would also like to extend my gratitude to LaurieMolday, Dr. Greg Connell, Dr. Delyth Wong, Dr. Dale Laird, Dr. Jim Richards, Dr.Andy Goldberg and Sue Curtis for their patience in teaching me many biochemical,immunochemical, photographic and computer management techniques used in mystudies. Much appreciation is also extended to my graduate student advisorycommittee (Dr. Philip Bragg, Dr. Roger Brownsey and Dr. Sid Katz) for theirencouragement, helpful advice and letters of reference. I am also grateful to Drs.Ann Milam and Greg Hageman for disucussions relating to immunofluorescencemicroscopy of retina, to Dr. Gustav Lienhard for information about glucosetransporter antibodies and to Drs. Langxing Pan and Peter G. Isaacson for theirgenerous gift of Mab65 against lactate dehydrogenase, to Dr. Alison Buchan for theuse of her Zeiss Microscope and to Andrea Dose, Michelle Illing, Orson Moritz andDr. Andy William for enjoyable discussion sessions. Finally, I would like to thankmy parents and my brother Yi-Te Hsu for their support and interest in my work.1CHAPTER 1INTRODUCTION1.1. RETINAThe neural retina is a highly specialized extension of the central nervoussystem responsible for light detection. In vertebrates, the retina is composed of sixtypes of neurons: rod photoreceptors, cone photoreceptors, bipolar cells, horizontalcells, amacrine cells, and ganglion cells (Fig. 1). Light entering the eye first passesthrough the ganglion cell layer and finally reaches the photoreceptor cells where thephototransduction process, or the conversion of light energy to electrical potential,occurs. The photosignals in the form of electrical impulses are relayed from thephotoreceptor cells to the ganglion cells and finally to the visual cortex in the brain1.2. PHOTORECEPTOR CELLSIn vertebrates, rod and cone photoreceptor cells are the light-sensitive cellsof the retina. Rod photoreceptor cells mediate black-and-white vision under dimlight. In humans, they have a diameter of 1-3 urn and a length of 40-60 um and areprimarily located in the peripheral region of the retina. The cone photoreceptorcells are responsible for color vision in daylight. In humans, they are 1-1.5 um indiameter and 75 urn in length and are found in the macular (central) region of theretina (Shichi, 1983). Both types of photoreceptor cells can be divided into fourcompartments: an outer segment, an inner segment, a cell body and a synapticterminus (Fig. 2). The outer segment, which is connected to the inner segment by athin, nonmotile cilium junction, is a specialized organelle where thephototransduction process takes place. The inner segment and the adjacent cell220 N mFig. 1: Synaptic connections of neural retina.Schematic diagram (A) and light micrograph stained with hematoxylin andeosin (B) showing layers of vertebrate retina. Os, outer segments of cone (C) androd (R) photoreceptor cells; Is, photoreceptor inner segments; ONL, outer nuclearlayer containing cell bodies of photoreceptor cells; OPL, outer plexiform layercontaining synaptic connections among photoreceptor, bipolar (B) and horizontal(H)) cells; INL, inner nuclear layer containing cell bodies of bipolar, horizontal andamacrine (A) cells; IPL, inner plexiform layer containing synaptic connectionsamong bipolar, amacrine and ganglion (G) cells; GCL, ganglion cell layer. Thearrow indicates the direction of light entering the eye (diagram adapted from Shichi,1983 and light micrograph adapted from Johnson and Blanks, 1984).Outer SegmentInner SegmentCell BodySynaptic Terminal3Rod Cell^ Cone CellFig. 2: Schematic diagram of rod and cone photoreceptor cells.Both rod and cone photoreceptor cells can be divided into fourcompartments: the outer segment, the inner segment, the cell body and the synapticterminus. The outer segment is connected to the inner segment by a cilium junction(diagram provided by Dr. Robert S. Molday).4body contain the primary metabolic (mitochondria) and biosynthetic machinery(endoplasmic reticulum, ribosomes, Golgi apparatus and nucleus) of thephotoreceptor cell. The synaptic terminus modulates the secretion ofneurotransmitters to the bipolar and horizontal cells.1.3. PHOTORECEPTOR OUTER SEGMENTS1.3.1. StructureCone and rod outer segments consist of a highly ordered array of about 2000flattened disks surrounded by a plasma membrane (Fig. 2). In the cone outersegment, the disk membrane is continuous with the plasma membrane. In the rodouter segment, the disks are separate from each other and from the plasmamembrane except at the proximal end of the outer segment where new disks areformed from evagination of the plasma membrane (Steinberg et al., 1980). Newdisk and plasma membranes are constantly synthesized at the base of the outersegment to replace distal outer segment tips phagocytized by retinal pigmentepithelial cells.1.3.2. Protein compositionBecause the cilium junction between the inner and outer segments of the rodphotoreceptor cell is very fragile, rod outer segments (ROS) can be readily brokenoff from the retina and isolated by sucrose density centrifugation for biochemicalstudies (McConnell, 1965; Papermaster and Dreyer, 1974). Recently, the disk andplasma membranes of ROS have been separated from each other using a ricin-gold-dextran affinity density perturbation method (Molday and Molday, 1987) and shownto have different protein compositions. Protein compositional analysis of these twomembranes indicates that with the exception of rod visual protein, rhodopsin, whichis found in both disk and plasma membranes, ROS proteins are preferentially sorted5to either the plasma or disk membranes (Fig. 3). Known major disk and plasmamembrane proteins are listed in Table I.1.4. PHOTOTRANSDUCTIONPhototransduction is the process in which light initiates a cascade of eventsleading to the hyperpolarization of the photoreceptor cell. In rod photoreceptorcells, this process can be divided into three phases: dark state, visual excitation andvisual recovery (Reviews: Kaupp and Koch, 1986; Yau and Baylor, 1989; Lolley andLee, 1990; Pugh and Lamb, 1990; and Stryer, 1986,1991).1.4.1. Dark stateIn dark the resting membrane potential of a rod photoreceptor cell ismaintained at -30 to -40mV. A steady state current (also called dark current) of 25 -71 pA or 1.5 - 4.3 x 108 monovalent charges/sec/ROS has been detected flowingoutward from the inner segment and inward into the outer segment of the rodphotoreceptor cell (Fig. 4, refs. Hagins et al., 1970; Baylor et al., 1979a,b; Stryer,1986). The inward current is carried by Na + (80-85 %), Ca2+ (10-15 %) and Mg2+(3 %) entering the outer segment through a cGMP-gated cation channel(Yokishima and Hagins, 1970; Fesenko et al., 1985; Yau and Nakatani, 1985;Hodgkin et al., 1985; Nataki and Yau, 1988a; Cook and et al., 1987,1989). Thebalancing outward current is carried by K + from the inner segment. Na+ enteringthe outer segment is pumped out by a Na+ /IC" ATPase in the inner segment(Sillmann et al., 1969; Hagins et al., 1970; Stirling and Lee, 1980). This pumpmaintains the Na+ and K+ gradients across the photoreceptor plasma membranerequired for the perpetuation of the dark current. Ca2+ entering the outer segmentis extruded by an electrogenic Na+ /Ca 2+ -K + exchanger in the outer segment (Yauand Nataki, 1984; Schroder and Fain, 1984; Cook and Kaupp, 1988). Extrusion ofRhodopsin DiskLamellar RegionPlasmaMembraneDiskRim RegionSpectrin—likeProteinRhodopsincGIAP—gated^NaCation Channel C aPlasma MembraneGlycoproteinsCHONa/Ca—K Exchanger NaPeripherinCHOWr ,^Mitititt6Fig. 3: Differential protein composition in the plasma and disk membranes of rodphotoreceptor outer segments.A proposed model for the localization of major disk membrane-specificproteins such as peripherin and ROM-1 and plasma membrane-specific proteinssuch as the spectrin-like protein, the cGMP-gated cation channel and the Na/Ca-Kexchanger in rod photoreceptor outer segments. CHO denotes carbohydrate(diagram adapted from Dr. Robert S. Molday).7Table IMajor Proteins of Bovine ROS Disk and Plasma MembranesProtein^MW (kDa)^FunctionDisk membranerhodopsinaperipherinROM-1rim proteinPlasma membranerhodopsinG3PDdcGMP-gatedcation channelspectrin-likeproteinNa + /Ca2+ -K+exchanger^38^photon receptor'39^disk rim structure maintenanceb37^disk rim structure maintenance(interacts with peripherin)`220^unknown38^photon receptor34^glycolysis69^allows Na + and Ca2 + to enterROS240^associates with thecGMP-gated cation channel complex230^extrudes Ca2+ and K+ in exchange forNa+ entry into ROSa Rhodopsin is the only known protein found in both ROS plasma and diskmembranes. It constitutes about 50 % of total plasma membrane protein and up to90 % of total disk membrane protein.b Hypothesized function; ref Connell and Molday, 1991.Hypothesized function; ref Bascom et aL, 1992.d Associated with the ROS plasma membrane under hypotonic conditions.8Fig. 4: Dark current in the rod photoreceptor cell.The dark current consists of an outward current carried by K+ from avoltage-gated 1( 1- channel (V) in the inner segment (IS) and an inward currentcarried by Nat, Ca2+ and M,g2 + into the outer segment (OS) through a cGMP-gatedcation channel (C). Na t entering the outer segment is eventually pumped out of thephotoreceptor cell by a Nat /K+ ATPase (A) in the inner segment. Ca 2 + enteringthe outer segment is extruded by a Na+/Ca 2 +-K+ exchanger (E) in the outersegment. The extrusion mechanism of Mg 2 + is still not known.9each Ca' is coupled to the influx of four Na+ and the efflux of one K+(Schnetkamp, 1986; Cervetto et aL, 1989). The mechanism of Mg' extrusion is stillunknown.1.4.2. Visual ExcitationThe process of visual excitation is a G protein mediated signal transductionpathway analogous to that initiated by hormones epinephrine and glucagon. In thecase of visual excitation, light is the signal whose interaction with its receptorrhodopsin results in the activation of an effector enzyme (cGMP-phosphodiesterase)through a G protein (transducin). This process is illustrated in Fig. 5 and brieflydescribed below:1.Upon absorption of a photon, the 11-cis-retinyl chromophore of rhodopsinisomerizes to all-trans-retinal through a series of unstable intermediates leading tothe release of all-trans-retinal from the rhodopsin apoprotein or opsin (review.Shichi, 1983). One intermediate, metarhodopsin II, formed within 1 msec afterrhodopsin photobleaching, activates transducin (Emeis et al., 1982 and Bennett etal., 1982). This is the first amplification step in the visual cascade. Onephotoactivated rhodopsin can activate as many as 500 transducin molecules (Fung etal., 1981).2.Each transducin molecule has three subunits; a (39 kDa), (36 kDa) and11 (8 kDa). Metarhodopsin II activates transducin by catalysing the exchange ofGDP for GTP bound at the a subunit of transducin (Fung and Stryer, 1980; Kiihnet al., 1981; Fung, 1983). The a subunit with its bound GTP then dissociates fromthe O'Y subunits. The ease of a subunit dissociation from the O'Y subunits hasbeen postulated to be modulated by the type of fatty acyl chain covalently linked tothe amino terminus of the a subunit (Kokame et al., 1992; Neubert et al., 1992).GTP] ^IDTPTap T.PDE1^1Ta-POE•••••...COP^II IIITQ^ CaMca2K+GTPATPi NOMGTP/5-GMP /GI(4N140.2+K+V••• CaMCa2^eGA4P,p/01cGMP W cGMP^,GCNa 4CA 2*GOPMTarTOTNatca 2+hV1Rho4  — Rho^RhoATPho10Fig. 5: Visual excitation and recovery in the rod photoreceptor cell.This diagram illustrates the visual excitation^and recovery ( - -pathways. These processes are described in more detail in sections 1.4.1, 1.4.2, and1.4.3 (diagram adapted from Dr. R.S. Molday).In the visual excitation process:^1^Light activates rhodopsin (rho).2Activated rhodopsin activates transducin (T).^(3^The a subunit of activated transducin then activates the cGMP-phosphodiesterase (PDE).(4) Hydrolysis of cGMP to 5'GMP by PDE results in the closure of the cGMP-gated cation channel (C), preventing cations from entering the outersegment.In the visual recovery process:(i) Rhodopsin is inactivated by rhodopsin kinase (RK) and arrestin (AR).(ii) Transducin a subunit is inactivated by GTP hydrolysis and reassociates withtransducin j37 subunits.(iii) PDE is inactivated upon dissociation from transducin.(iv) A decrease in the intracellular Ca2 + concentration due to a continuousextrusion of Ca2+ ions by the Na+/Ca 2 +-K+ exchanger (E) results in theactivation of guanylate cyclase possibly by recoverin (R). Cyclic GMP isresynthesized from GMP by guanosine monophosphate kinase (GK),nucleoside diphosphate kinase (NDK) and guanylate cyclase (GC).(v)^A low intracellular Ca2 + allows a calmodulin-mediated facilitated reopeningof the cation channel.11The dissociated a subunit activates cGMP-phosphodiesterase (PDE), the effectorenzyme in this signal transduction pathway.3. Each PDE molecule consists of an a (88 kDa), a (84 kDa) and two 7(11 kDa) subunits. Transducin activates PDE by relieving the inhibitory effect ofPDE 'Y subunits on PDE a subunits. It is still not clear whether this transducina subunit-mediated relief of PDE inhibition is carried out by dissociating theinhibitory PDE 7 subunits from the catalytic PDE a/3 subunits (Fung andGriswold-Prenner, 1989; Wensel and Stryer, 1990; Yamazaki et al., 1990) or bychanging the interaction between the PDE 7 and oe13 subunits (Arshaysky andBownds, 1992). Once activated, one PDE molecule can hydrolyze approximatelycGMP to 5'-GMP per second (Wheeler and Bitensky, 1977; Yee and Liebman,1978; Woodruff and Bownds, 1979; Liebman and Pugh, 1980). This is the secondamplification step in the visual excitation process.4. In ROS, although the concentration of total cGMP is approximately 60 uM,the majority of cGMP is bound at the non-catalytic cGMP-binding site of PDE andis not readily available for binding to the cGMP-gated cation channel or forhydrolysis by PDE (Yamazaki et al., 1980; Charbonneau et al., 1990). An estimatedconcentration of 4-10 uM free cGMP maintains approximately 1-5 % of the totalROS cGMP-gated channel in their opened state in the dark (Nataki and Yau,1988b). Since the cGMP-gated cation channel requires bound cGMP to remainopen, a reduction in the free ROS cytoplasmic cGMP concentration due to cGMPhydrolysis by PDE upon illumination results in channel closure (Fesenko et al., 1985;Koch and Kaupp, 1985). Closure of the channel decreases the permeability of ROSplasma membrane to cations, preventing the influx of Nat, Ca' and Mg' intoROS (thus the inward dark current is decreased). A continual efflux of K+ from theinner segment results in the hyperpolarization of the rod cell.12Unlike other neurons which have a threshold of stimulation, rod cells undergoa spatially limited hyperpolarization in response to the absorption of even onephoton (Baylor et al., 1979b). Amplification of the visual cascade in response to onephoton can result in the hydrolysis of 105 cGMP and lead to a 0.2-1.5 pA decrease inthe dark current or an average reduction of the influx of 5.3 x 106 Na+ per second(due to closure of 300-400 channels or about 4% of the total opened cationchannels, ref. Baylor et aL, 1979b; reviews, Stryer, 1986; Lolley and Lee, 1990). Themagnitude of hyperpolarization is directly proportional to the number of rhodopsinmolecules activated, to a maximum of about 10 rhodopsin activations per rod cell(Penn and Hagins, 1972; Baylor et al., 1979a,b; Lamb et al., 1981).5. In the dark, neurotransmitters are continuously secreted from the synapticterminus of rod photoreceptor cells. Hyperpolarization of the rod cell decreases theneurotransmitter secretion through an unknown mechanism. Subsequent eventsleading to the perception of vision are not well understood. The mechanism of thevisual excitation in cone outer segments has not yet been extensively studied at themolecular level. However, it has been observed that cone photoreceptor cells aremuch less sensitive (25-100 times) and have a faster electrical response (severalfold) to light than rod photoreceptor cells (review: Yau et al., 1988).1.4.3. Visual recovery (Fig. 5): Restoration of bleached ROS to the dark staterequires the deactivation of proteins involved in the visual excitation process,reopening of the cGMP-gated cation channel to restore the dark current andregeneration of rhodopsin.Quenching of visual cascadeRhodopsin:^Activated rhodopsin is deactivated 2-3 seconds afterphotoactivation, long before the spontaneous decay of metarhodopsin II to opsin13and all-trans-retinal (103 seconds at 20 °C). This inactivation results fromphosphorylation of metarhodopsin II by rhodopsin kinase (Kuhn and Dreyer, 1972;Miller et al.,' 1986) and the binding of arrestin to phosphorylated rhodopsin (Wildenand Kuhn, 1982; Kiihn et al., 1984; Zuckerman et al., 1985). These reactions inhibitthe interaction between transducin and metarhodopsin II.Transducin: Transducin a subunit is deactivated by hydrolysis of its boundGTP to GDP due to its slow intrinsic GTPase activity (Kuhn, 1980; Kiihn et al.,1981). This slow GTPase activity is accelerated by the binding of PDE, specificallythe 7 subunit of PDE, to the transducin a subunit (Arshaysky and Bownds, 1992).Binding of cGMP to PDE at the non-catalytic cGMP-binding sites on PDE aflsubunits partially suppresses this acceleration (Arshaysky and Bownds, 1992).Following nucleotide hydrolysis, the a subunit with bound GDP is able toreassociate with transducin O'Y subunits.PDE: Deactivation of PDE has been widely believed to be coupled to thehydrolysis of GTP bound to the transducin a subunit. Dissociation of transducina subunit from the inhibitory PDE 7 subunit allows the PDE 7 subunit toinhibit the catalytic activity of the PDE a )3 subunits. A recent report by Ericksonet al. (1992), however, suggests that PDE is inhibited by an unknown solubleinhibitor before transducin a subunit is inactivated by GTP hydrolysis. PDE isinactivated within seconds following the initiation of visual excitation (Sitaramayyaand Liebman, 1983).Reopening of the cGMP -gated cation channelIn the dark, the ROS cytosolic Ca' concentration is maintained atapproximately 0.3 uM (McNaughton et al., 1986; Ratto et al., 1988; Korenbrot andMiller, 1989) by a balanced Ca" extrusion and re-entry cycle described above in1.4.1 (Dark state). It has been hypothesized that Ca 2+ is involved in reopening the14cGMP-gated channel by modulating the activity of two calcium-binding proteins:recoverin (which promotes the resynthesis of hydrolyzed cGMP) and calmodulin(which facilitates the reopening of the cGMP-gated cation channel).Cyclic GMP resynthesisCyclic GMP hydrolyzed by PDE upon light stimulation is resynthesized from5'GMP by guanosine monophosphate kinase (synthesis of GDP from GMP),nucleoside diphosphate kinase (synthesis of GTP from GDP) and guanylate cyclase(synthesis of cGMP from GTP). Unlike guanosine monophosphate kinase andnucleoside diphosphate kinase which are constitutively active, guanylate cyclase hasbeen proposed to be activated only upon light illumination by a 26 kDa solubleactivator protein, recoverin, in a Ca" dependent manner (Koch and Stryer, 1988;Dizhoor et al., 1991; Lambrecht and Koch, 1991). In dark, guanylate cyclase isinactive because Ca' binding to recoverin prevents guanylate cyclase activation byrecoverin. In light, closure of the cGMP-gated cation channel prevents the re-entryof Ca' ions extruded by the light-insensitive Na + /Ca"-K + exchanger, thusresulting in a decrease in the cytoplasmic Ca" concentration to below 70 nM. Alow cytoplasmic Ca' concentration causes Ca' to dissociate from recoverin,allowing recoverin to activate guanylate cyclase. Restoration of the Ca" levels tothat in the dark state will again result in the binding of Ca" to recoverin and theinactivation of guanylate cyclase. A direct interaction between recoverin andguanylate cyclase, however, has not been detected. At present it is still controversialwhether recoverin is the soluble factor that regulates guanylate cyclase activity.Facilitated channel reopening: Calmodulin has been postulated toparticipate in facilitating the reopening of the cGMP-gated cation channel throughits interaction with the spectrin-like protein associated with the cation channelcomplex (Hsu and Molday, 1993). Following visual excitation, a low cytoplasmic15Ca2+ concentration due to cGMP-gated cation channel closure results in thedissociation of calmodulin from the channel/spectrin-like protein complex.Dissociation of calmodulin causes the channel complex to switch to a high affinityconformation for cGMP, allowing the channel to reopen at a lower free cGMPconcentration than that required to keep the channel open in the dark. Thisfacilitates the recovery of the photoreceptor cell after visual excitation. Reopeningof the cation channel restores the cytoplasmic Ca 2+ concentration to its dark statelevel, allowing calmodulin to bind to the channel/spectrin-like protein complex.Reassociation of calmodulin causes the channel to switch to its low affinity cGMP-binding conformation. The channel, in its low cGMP affinity form, will now againbe sensitive to small decreases in free cGMP concentration.Regeneration of rhodopsin: Upon photoactivation of rhodopsin, all-trans-retinalreleased from activated rhodopsin is reduced to all-trans-retinol by a NADPH-specific retinol dehydrogenase (Wald, 1968; Lion et al., 1975; and Ishiguro et al.,1991) and transported to the retinal pigment epithelial cells by aninterphotoreceptor retinoid binding protein (Bok and Heller, 1976; Liou et al., 1982;Pfeffer et al., 1983). In the retinal pigment epithelial cell, all-trans-retinol isconverted to 11-cis-retinal through a series of enzyme catalyzed reactions andtransported back to the ROS by the interphotoreceptor retinoid binding protein(Bok, 1985; Saari and Bredberg, 1988, 1989; Deigner et al., 1989).Upon reduction of all-trans-retinal to all-trans-retinol, arrestin is releasedfrom opsin which is then dephosphorylated by a phosphatase (Fowles et al., 1989;Palczewski et al., 1989). The newly synthesized 11-cis-retinal is transported backfrom retinal pigment epithelial cells and interacts with dephosphorylated opsin toregenerate rhodopsin.161.5. GLUCOSE TRANSPORTTransport of glucose into eukaryotic cells is mediated by two classes ofglucose transporters: the facilitative glucose transporters and the Na t -glucosecotransporter (reviews: Gould and Bell, 1990; Bell et al., 1990; Kasanicki and Pilch,1990; Thorens et al., 1990 and Silverman, 1991).1.5.1. Facilitative glucose transportersThis family of glucose transporters consists of five isoforms designated asGLUT 1, GLUT 2, GLUT 3, GLUT 4 and GLUT 5. Some physical properties ofthese isoforms are listed in Table II.Structure: The primary sequences of these isoforms have been determinedand shown to range from 492 to 524 amino acids in length. In humans, the isoformsshare 39-65% identity and 50-70% similarity among each other. Hydropathyanalyses, in conjunction with protease cleavage, antibody tagging and spectroscopicstudies of the GLUT 1 isoform, suggest that all isoforms have twelvetransmembrane helices with their N- and C- termini located on the cytoplasmic sideof the plasma membrane. There is a single N-glycosylation site at the exofacial loopbetween transmembrane helices 1 and 2 (Fig. 6).Transport mechanism: All transporter isoforms prefer D-glucose in thechair conformation as their substrate. Kinetic properties of these transporterisoforms are listed in Table II. Presently, the molecular mechanism of glucosetransport has only been studied in detail for the GLUT 1 isoform in human redblood cells (Vidaver, 1966; review: Baldwin and Lienhard, 1981; Walmsley, 1988and Silverman, 1991). According to the alternating conformation model, thetransporter oscillates between inward- and outward-facing conformations (Fig. 7).A transport cycle involves the binding of a glucose molecule to one conformation ofthe transporter through hydrogen bonding at helices seven to eleven, followed by17Table HCharacteristics of Mammalian Glucose TransportersType^Tissue^Size^Kineticdistribution (no. of amino acids)^propertiesA) Facilitative glucose transportersGLUT-2^liver, kidney,intestines,pancreatic^cells492^human erythrocyte:- asymmetric- Vmax: uptake < exchange- Km: 5-30 mM (depending onassay condition)524^liver:- symmetric- Km: 20-66 mMintestines:- asymmetric- Km: 23-48 mMGLUT-1^ubiquitous; mostabundant in fetalplacentaGLUT-3GLUT-4ubiquitous; very^496abundant in braincardiac and skeletal^509muscle; brown andwhite fatadipocyte:- symmetric- Km: 2-10 mMGLUT-5^small intestinesB) NaVglucose co-transporterSGLT1^small intestines501664^intestines:- Km:-high affinity: 0.35 mM-low affinity: 6 mMThis table summarizes some physical and biochemical properties of glucosetransporters reviewed by Kasanicki and Pilch (1990), Mueckler (1990), and Silverman(1991).18InsideFig. 6: Structural model for facilitative glucose transporters.1 - 12 denotes eleven putative transmembrane a-helices in facilitative glucosetransporters. Asn 45 denotes the single asparagine-linked glycosylation site. Cyt Bshows the proposed cytochalasin binding site. Cytochalasin B, an inhibitor ofGLUT-1 type glucose transporter, binds to the cytoplasmic side of the transporter(diagram adapted from Walmsley, 1988).Fig. 7: Schematic diagram illustrating the alternating conformation model forglucose transport by the GLUT-1 glucose transporter.According to the model, each glucose transport cycle starts with (1) thebinding of glucose to one conformation of the transporter, follows by (2) the re-orientation of the transporter shuttling glucose to the other side of the membrane,and concludes with (3) the release of glucose from the transporter. The transporter(4) can return to the original conformation with or without bound glucose (diagramadapted from Baldwin and Lienhard, 1981).2 0reorientation of the transporter to the other conformation facing the opposite sideof the membrane, and concludes with the release of glucose from the transporter. Ifthere is a glucose concentration gradient across the membrane, the emptytransporter will return to the opposite-facing conformation without glucose. Thisresults in a net movement of glucose across the membrane until an equilibrium isachieved. Since transporter reorientation is the rate-limiting step and binding ofglucose to the transporter facilitates this process, the rate of glucose transport acrossthe membrane is faster under equilibrium exchange than under net transportconditions.1.5.2. Na + -glucose cotransporterThis type of transporter is detected only in specialized epithelial cells ofsmall intestines and proximal tubules of kidney (Fig. 8).Structure: The primary sequence of this type of transporter has beendetermined in humans and shown to contain 664 amino acids (Hediger et al., 1987).Hydropathy analysis suggests that it contains 11 transmembrane helices with its N-and C- termini located at opposite sides of the membrane. There is a single N-glycosylation site at the exo-facial loop between helices 5 and 6. Presently it is notknown if there is a family of Na+-glucose cotransporters similar to that of facilitativeglucose transporters.Transport mechanism: Na t -glucose transporter actively transports glucoseagainst its concentration gradient by coupling with Na + uptake which is transporteddown its concentration gradient. The transport is electrogenic and does not involveco- or counter-transportation of other ions. It has been hypothesized that thetransport mechanism involves the binding of Na+ to the transporter, resulting in aconformational change in the transporter that increases its affinity for glucose.21PlasmaMembraneFig. 8: Structural model for Natglucose co-transporter in plasma membrane.M1 -M11 denotes eleven transmembrane helices in the transporter. CHOindicates the single glycosylation site (diagram adapted from Bell et al., 1990).2 2Once a glucose molecule is bound, the transporter reorients to translocate both Na+and glucose to the other side of the membrane (Silverman, 1991).1.6. GLUCOSE OXIDATIONOnce glucose is inside the cell, it can be metabolized either by glycolysis andthe citric acid cycle to produce energy in the form of ATP (via respiration inmitochondria) or by the hexose monophosphate pathway to produce reducing powerin the form of NADPH (Fig. 9). Transaldolase and transketolase mediate theinterconversion of sugar intermediates between glycolytic (fructose-6-phosphate andglyceraldehyde-3-phosphate) and hexose monophosphate pathways. This allows thechanneling of sugar intermediates from the hexose monophosphate pathway back tothe glycolytic pathway so that both NADPH (via the hexose monophosphatepathway) and ATP (via glycolysis) can be produced. Both glycolytic and hexosemonophosphate pathways take place in the cytosol while the tricarboxylic acid cycleoccurs within the mitochondrion.1.7. PHOSPHOCREATINE SHUTTLEIn cells with immediate high energy requirements such as muscle, sperm andphotoreceptor cells, a phosphocreatine shuttle has been hypothesized to mediateenergy transport between mitochondria and the site of energy demand (Bessmanand Geiger, 1981, Tombes and Shapiro, 1985 and Walliman et al., 1986). Accordingto this hypothesis, the phosphocreatine shuttle is driven by two isozymes of creatinekinase (Fig. 10). A mitochondrion-associated creatine kinase isozyme transfers thehigh energy phosphate group from respiration-derived ATP to creatine to producephosphocreatine. Phosphocreatine diffuses to the target site where its high energyphosphate is transferred to ADP to produce ATP by a target site-specific creatinekinase isozyme. Creatine diffuses back to the mitochondrion and the cycle repeats.L-Malate CitrateH2O -I. Aconkase2R.3S-IsocilrateFumarateNAD®`•NADH + Hes"--r, CO2FADH2 4-■FADIsocitratedehydrogemsesuccinauedehydrogemseSuccinate o-Ketogl =rateLactosede4ordevenaseOxaloacetateGlucoseHesoItimese,r ATPADPGlucore 6-phosphate0.^0-CI-tZ z6-Phosphogluc000-1,5-Lactone110ental.7 1..1Glucosedchydrogemne Gloc000lactooasci- H2OHeFructose 6-phosphateTP 6-PhosphogluconatePhospholorces-:MauADP NADP°6-PtrosphoglocassmFructose 1.,6-bisphosphate NA DPH + H ®Aldolase cotRibulose 5-phosphateHMPI(-Amy CoA24;)CoASHThose phospkareDihydroxyacetone ^iso er:se^Glyceraldehyde - phosphate 3-phosphateGlycerakiehyde[NAD. +3-phosselesedehsdrogenase NADH + H.1,3-BisphosphoglyceratePhownwhovvilantso3-Phosphoglyceratethc'sPh"*""'"."."2-Phosphoglycerate'H2OIII PhosphoenolpyruvatePyrueste e-ADPIsinase 'ATPPynzvateADPATPGlycolysist-LactateCoASHSosinyl-CoAGTF'^syntheraseGDP-Ketoglutarate&hydrogens.compkxCoASHNAD®NADH HeCO,Succinyl CoATCA0x23Fig. 9: Glucose metabolism via glycolysis, the hexose monophosphate pathway andthe tricarboxylic acid cycle.Glucose entering the glycolytic pathway can be oxidized via either the hexosemonophosphate pathway (HMP) or the tricarboxylic acid cycle (TCA). Reversiblereactions in the glycolytic pathway are indicated by double arrows. Dotted linesrepresent the interconversion of glycolytic and hexose monophosphate pathwayintermediates by transaldolase and transketolase (diagram adapted from Stryer,1988 and Rawn, 1989).FLAGELLUMMOTILITYDYNEIN( inactiveDYNEINactivePi+ ADP^ATPCK(T)24ATP ADP+PiCO2 4^ FATTY ACIDS02RESPIRATIONFig. 10: Phosphocreatine shuttle mediates high energy phosphate transportbetween sperm mitochondrion and flagellum.Phosphocreatine shuttle is driven by transphosphorylation reactionscatalyzed by creatine kinase isozymes CK(Mi) in the mitochondrion and CK(T) inthe sperm flagellum. In the mitochondrion, CK(Mi) transfers the high energyphosphate group from ATP produced during respiration to creatine (Cr).Phosphocreatine (PCr) produced by this reaction diffuses to the sperm flagellumwhere its high energy phosphate group is transferred to ADP by CK(T) to produceATP required for flagellar motility. Creatine diffuses back to the mitochondrionand the cycle repeats again (diagram adapted from Tombes and Shapiro, 1985).2 51.8. RETINAL GLUCOSE METABOLISM AND PHOTOTRANSDUCTIONThe function of retina has been assessed either directly by measuring retinalelectrical activities (ERG) or indirectly by quantitating retinal ATP content.1.8.1. ElectroretinogramElectroretinogram or ERG was first recorded by Holmgren (1865) instudying retinal function by electrophysiology. It can be obtained either fromtransretinal recording of whole retina which measures the sum of electrical activitiesof all retinal cells or from intraretinal recording which selectively measureselectrical activities of cells from a single layer of retina. The pattern of ERGrecordings vary among different species and under different assay conditions(reviews: Brown, 1968 and Winkler, 1972, 1981b). ERG recorded in vitro and invivo, however, has been shown to be essentially identical to each other in a numberof animals and can therefore be used to evaluate retinal phototransduction inresponse to various metabolic stress in vitro. A typical ERG of neural retina beginswith a hyperpolarizing a-wave followed by a depolarizing b-wave withsuperimposing oscillatory wavelets (Fig. 11). The a-wave or Fast PIII starts about2.5 cosec after light stimulation and reaches its peak amplitude after about 12 msec.It represents the hyperpolarization of photoreceptor cells due to a decrease in theirmembrane permeability to sodium ions (Brown, 1968). The b-wave or PII peaksapproximately 50 msec after a light flash and and is believed to be due to thedepolarization of bipolar or Muller cells (Miller and Dowling, 1970). The lowamplitude oscillatory wavelets superimposed on the b-wave have been suggested tooriginate from bipolar and amacrine cells (Ogden, 1973). Postphotoreceptorpotentials such as b-wave and oscillatory wavelets can be eliminated from ERG ofrat retina by including sodium aspartate in the retina perfusion medium to blocksynaptic transmission from the photoreceptor cells to the postsynaptic neurons.2 6Fig. 11:^Electroretinogram illustrating the rat retinal response to lightillumination.A typical electroretinogram in response to light consists of an a-wave and ab-wave with superimposing wavelets (diagram adapted from Winkler, 1981b).2 7Changes in the a-waves or the photoreceptor potential in response to variousmetabolic poisons can thus be utilized to assess the importance of glucosemetabolism in supporting phototransduction.1.8.2. Effects of glucose metabolism on retinal functionParallel studies of ERG and retinal ATP content indicate that while the rateof retinal glucose metabolism is not stimulated by visual excitation, maintenance ofthe phototransduction process in photoreceptor cells is highly dependent on glucosemetabolism (Cohen and Noell, 1960; and Winkler, 1981a,b) Inhibition of retinalglucose metabolism by removing the extracellular glucose or by adding themetabolic poison iodoacetate (which inhibits triose phosphate isomerase andglyceraldehyde-3-phosphate dehydrogenase in glycolysis) abolishes thephotoreceptor potential. Glycolysis alone in the absence of aerobic glucosemetabolism supports only 40 % of the optimal photoreceptor potential in thepresence of physiological concentrations of glucose. Aerobic glucose metabolismalone in the presence of 5 mM pyruvate can maintain up to 95 % of retinal ATPcontent and photoreceptor potential in the presence of 5 mM pyruvate.Maintenance of a 100 % photoreceptor potential and retinal ATP content, however,requires the participation of both anaerobic (glycolysis) and aerobic (tricarboxylicacid cycle) glucose metabolism (Winkler, 1981b).1.9. GLUCOSE METABOLISM OF RETINAGlucose metabolism in the retina has been investigated in vitro using intactretinas and freeze-dried tangential retina sections. These studies showed thatalthough vertebrate retinas are similar to each other in cellular organization, retinalglucose metabolism varies among different species. These differences may be2 8attributed to variations in the physical properties of the retinas such as cell densityand thickness of various retinal layers, ratio of cones to rods, morphology ofphotoreceptor cells and blood (therefore nutrient and oxygen) supply to the retinas.In addition, the observed in vitro rate of retinal metabolism is approximately twofold higher than that expected in vivo and is affected by the concentration of ionssuch as bicarbonate, chloride, potassium, sodium, and calcium in the retinaincubation media.1.9.1. Intact retinaStudies with isolated whole retinas showed that retinal metabolism has threedistinctive features compared to other tissues (Futterman and Kinoshita, 1959;Graymore, 1959, 1970; Cohen and Noell, 1960, 1965; Riley and Voaden, 1970 andWinkler, 1981a,b):1. The retina has one of the highest respiratory rates compared to othertissues (average rate is approximately 1.5 umol 02/mg dry wt/hr). The respiratoryrate of the brain cortex, for example, is only 50 % that of retina even though theretina is an extension of the central nervous system.2. The retina has one of the highest glycolytic capacities compared to othertissues, converting glucose to lactate at a rate of 1.5 umol lactate/mg dry wt. ofretina/hr even in the presence of oxygen.3. Glycolysis and respiration in retina are not as tightly coupled to eachother, as demonstrated by comparatively small Pasteur and Crabtree effects in adultretinas. The Pasteur effect describes the stimulation of respiration and depressionof glycolysis under aerobic conditions. In the retina, aerobic conditions depress theglycolytic lactate production by a maximum of only 50 %. The Crabtree effect ischaracterized by the depression of glycolysis and stimulation of respiration by theremoval of glucose from the extracellular medium. Removal of glucose from the2 9retina incubation medium resulted in only a 10 % increase in retinal respiration.Both Pasteur and Crabtree effects are believed to be, in part, due to bothcompetition for substrates and cofactors by aerobic and anaerobic glucose metabolicpathways and allosteric inhibition of glycolytic enzymes (especiallyphosphofructokinase) by intermediates and products from the citric acid cycle. Alack of these two effects suggests that aerobic and anaerobic glucose metabolicpathways may be separately compartmentalized in the retina.The retina is also very rich in glucose-6-phosphate dehydrogenase, the firstenzyme in the hexose monophosphate pathway (Shimke, 1959). Glucose taken upby the retina can be utilized by the hexose monophosphate pathway to produceNADPH required for retinal reduction and for protection of the retina againstoxidative damage. When the retina is not under oxidative stress, the hexosemonophosphate pathway produces approximately 23 % of the total carbon dioxideformed by the retina. This rate is increased at least three-fold when the retina isunder oxidative stress. There is no appreciable recycling of glucose in the hexosemonophosphate pathway (via the gluconeogenic pathway) under both aerobic andanaerobic conditions since only the first carbon of glucose is oxidized to CO 2 by thepathway when the citric acid cycle is inhibited (Futterman and Kinoshita, 1959;Cohen and Noell, 1960; Rahman and Kerly, 1961; Graymore and Towlson, 1965;review, Winkler, 1981b). Glycogen storage in the retina is very low and is primarilyrestricted to the Muller (glial) cells (Crane and Ball, 1951; Kuwabara and Cogan,1961; Mizuno and Sata, 1975).1.9.2. Photoreceptor cellsGlucose metabolism in photoreceptor cells was indirectly studied bycomparing glucose metabolism in retinas lacking photoreceptor cells with that innormal retinas in vitro (Cohen and Noell, 1960; Graymore, 1959,1960; Graymore3 0and Tansley, 1959; Graymore, Tansley and Kerly, 1959). Retinas lacking fullydeveloped photoreceptor cells can be obtained from either 7-day postnatal animalslacking fully developed photoreceptor cells, rats with degenerate visual cells due toretinitis pigmentosa or animals having resorbed photoreceptor cells due tointravenous iodoacetate administration.The respiratory rate in retinas lacking photoreceptor cells is 12 to 60 % lowerthan that of normal retinas. Glucose oxidation accounts for 80 % of respiration invisual cells compared to only 55 % in non-visual cells indicating that glucose is themain substrate used by the photoreceptor cells for energy production. This mayexplain, in part, why photoreceptor cells are more sensitive to poisons which affectglucose metabolism (such as iodoacetate) than non-visual cells. A significantamount of energy (approximately 50 %) derived from glucose oxidation is used bythe Na+ /K+ ATPase in the photoreceptor inner segment for Na + and K + gradientmaintenance (Winkler, 1981a). Glucose is not completely oxidized by thetricarboxylic acid cycle to CO 2. Approximately 20 % of carbon derived from glucoseis used for the synthesis of glutamate, a neurotransmitter of the photoreceptor cells(Cohen and Noell, 1960; Morjaria and Voaden, 1979).Glycolytic rates in retinas without visual cells was found to range from 50 -100 % of values in normal retinas, depending on the species and conditions oftested animals. Assuming that the metabolism of other retinal cells is not affectedby the absence of the photoreceptor cells from the retina, these observations suggestthat the rate of respiration and glycolysis is higher, if not similar, in visual cellscompared to other cells in the retina. The Pasteur and Crabtree effects are alsomore apparent in retinas lacking a photoreceptor cell layer, suggesting that thereare very little or no Pasteur and Crabtree effects in the visual cells. Theseobservations suggest that glycolysis and respiration are not tightly coupled to eachother in the photoreceptor cells. Studies using retinas from postnatal animals31suggest that photoreceptor cells also have higher hexose monophosphate pathwayactivity than other non-visual cells in the retina.1.9.3. Histochemistry of retina sectionsDistribution of enzymes involved in glucose metabolism in retinas has beenstudied by quantitative histochemistry of freeze-dried retina sections. Activities ofsix glycolytic enzymes, hexokinase, phosphoglucoisomerase, phosphofnictokinase,aldolase, glyceraldehyde-3-phosphate dehydrogenase, phosphoglyceromutase andlactate dehydrogenase, have been quantitated in the retina by Lowry et al. (1956,1961, Fig. 12). Activity of the first glycolytic enzyme, hexokinase, was found to peakin the inner segments of photoreceptor cells near the choroidal blood supply.Activities of other glycolytic enzymes, however, were found to peak near thesynaptic end of photoreceptor cells. This differential distribution of glycolyticenzymes may suggest that glycolysis is most active at the photoreceptor synapticterminus. Alternatively, this may also reflect a mechanism by which glucoseentering the inner segment from the blood supply can be drawn to the synapticterminus located farther away from the blood supply by the formation of a glucose-6-phosphate gradient. Lowest glycolytic enzyme specific activities were found inphotoreceptor outer segments and retinal pigmented epithelial cells. Activities ofthese enzymes in the outer segments were found to be 15 % or lower of thatobserved for brain.Activities of tricarboxylic acid cycle enzymes malate and succinatedehydrogenases, as in the case of hexokinase, were found to peak at the ellipsoidregion of the photoreceptor inner segment where all the mitochondria of thephotoreceptor cell are located (Lowry, 1961; Wislocki and Sidman, 1954). Malatedehydrogenase showed a reciprocal retinal distribution pattern with respect to thatof lactate dehydrogenase, an anaerobic glycolytic enzyme found in highest3 2Fig. 12: Localization of glycolytic enzymes in retina.Distribution profiles of hexokinase and phosphofructokinase activities showthat while hexokinase is most abundant in photoreceptor inner segments,phosphofructokinase activity peak in photoreceptor synaptic termini. EnzymesInvolved in aerobic glucose metabolism such as succinate and malatedehydrogenases show retinal distribution patterns similar to that of hexokinase.Glycolytic enzymes (including lactate dehydrogenase) other than hexokinase exhibitretinal distribution patterns similar to that of phosphofructokinase. Glucose-6-phosphate dehydrogenase, the first enzyme in the hexose monophosphate pathway,is found in highest quantity in both photoreceptor inner segment and synapticterminus 1, retinal pigment epithelial cells; 2a, photoreceptor outer segments; 2b,photoreceptor inner segments; 4, outer nuclear layer; 5, outer plexiform layer; 6,inner nuclear layer; 7, inner plexiform layer; 8, ganglion cell body; and 9, ganglioncell fibers (diagram adapted from Lowry et al., 1961).3 3concentration near the synaptic terminus which is devoid of mitochondria. This is inagreement with intact retina studies suggesting that high rates of glycolysis underaerobic conditions in the retina may result from compartmentalization of enzymesinvolved in oxidative and anaerobic glucose metabolism in different parts of theretina.Glucose-6-phosphate dehydrogenase and 6-phosphogluconatedehydrogenase, the first and third enzymes in the hexose monophosphate pathway,were found to be exceedingly rich in the photoreceptor inner segment and synapticterminus (peak specific activity of glucose-6-phosphate dehydrogenase exceeds thatof brain by a factor of 25). Activities of these two enzymes were found to be muchlower in photoreceptor outer segment and other retinal cells.Thus, results from both biochemical and histochemical studies suggest:1. Photoreceptor cells have higher respiratory, glycolytic and hexosemonophosphate pathway capacity than other non-visual cells in the retina.2. Aerobic and anaerobic glucose metabolism in photoreceptor cells are nottightly coupled to each other due to differential compartmentalization of enzymesinvolved in the two pathways.1.9.4. Photoreceptor outer segmentsAlthough it is known that the maintenance of the phototransduction processin photoreceptor outer segments requires a large amount of energy and nucleotidessuch as GTP, ATP and NADPH, the source of its energy and nucleotide supplyremains unclear.Phosphocreatine shuttle: The inner segment contains all the mitochondriaof the photoreceptor cell and exhibits much higher glycolytic enzyme activities thanthe outer segment (Lowry, 1956,1961; Lolley and Hess, 1969). Thiscompartmentalization separating the phototransduction process in the outer3 4segment and glucose metabolism in the inner segment has led to a general view thatenergy required by the phototransduction process is supplied entirely by glucosemetabolism in the inner segment. Localization of mitochondrial and brain-typecreatine kinase isozymes in the inner and outer segments of photoreceptor cell,respectively, further suggests that a phosphocreatine shuttle analogous to that foundin muscle, electric organ of Torpedo marmorata and sperm cells channels energy inthe form of phosphocreatine from inner segment mitochondria to the outer segment(Wallimann, 1986). This hypothesis is supported by the detection of enzymeactivities which rapidly transfer high-energy phosphate groups among adenine andguanine nucleotides and phosphocreatine (Dontsov et al., 1978; Schnetkamp andDaemen, 1981).Glucose metabolism in photoreceptor outer segments: McConnell et al.(1969), Futterman et al. (1970) and Schnetkamp (1981), on the other hand, haveshown that isolated ROS contain glycolytic and hexose monophosphate pathwayenzyme activities. Lactate and CO2 production by isolated ROS in the presence ofglucose indicate that both glycolytic and hexose monophosphate pathways are activein this organelle. Localization of a glycolytic enzyme, triose phosphate isomerase, inthe outer segment layer of frozen bovine retina sections by enzyme histochemistryindicates that the observed enzyme activities are not likely due to contaminationfrom other retinal cells (McConnell et al., 1969). The detection of aphosphocreatine shuttle operating between the photoreceptor inner and outersegments and the presence of very low glycolytic enzyme activities in photoreceptorouter segments compared to other retinal layers, however, still favor the belief thatphototransduction in the outer segment is maintained entirely by glucosemetabolism in the inner segment.3 51.10. THESIS INVESTIGATIONThe principal objective of this thesis is to study the contribution ofphotoreceptor outer segment anaerobic glucose metabolism to the maintenance ofthe phototransduction process. In order to investigate the existence and function ofphotoreceptor outer segment glucose metabolism, glucose uptake and metabolicprocesses in this organelle were studied by activity assays, Western blot analyses andimmunocytochemistry.The first section of the thesis (Chapter 2) focuses on the identification andcharacterization of glyceraldehyde-3-phosphate dehydrogenase (G3PD), a glycolyticenzyme, in ROS. A major 38 kDa protein associated with ROS plasma, but notdisk, membranes was identified to be G3PD by N-terminal sequence, specificactivity and Western blotting analyses. Enzyme activity measurements indicate thatG3PD makes up approximately 2 % of the total ROS protein and over 11 % of theROS plasma membrane protein. Protease digestion and G3PD binding studiessuggest that this enzyme reversibly interacts with a protease-sensitive plasmamembrane-specific protein of ROS. The finding that G3PD is present in largequantities in ROS suggests that glycolysis may take place within this organelle.Chapter 3 is devoted to the detection and characterization of glycolyticenzymes and a GLUT-1 type glucose transporter in photoreceptor outer segments.Enzyme activities of six glycolytic enzymes including the first and last enzymes in theglycolytic pathway were detected in purified ROS preparations. Localization ofthree glycolytic enzymes in the outer segment layer of both bovine and chickenretina sections indicate that glycolytic enzymes are present in both rod and conephotoreceptor outer segments and are not simply contaminants from the innersegments or other retinal cells. Glucose transport activity, Western blot andimmunofluorescence microscopy analyses showed that both rod and cone outer3 6segment plasma membranes have a GLUT-1 type glucose transporter found inerythrocytes and brain cells.Chapter 4 summarizes studies carried out to investigate the existence andfunction of glucose metabolism in rod photoreceptor outer segments. Glycolytic,hexose monophosphate and retinal reduction pathways were quantitated in purifiedROS preparations. Glucose metabolism in ROS produces both ATP and NADPHrequired by phototransduction. A rate of 44 nmol ATP/min/mg ROS proteinproduced by glycolysis in ROS is sufficient to support cGMP regeneration, one ofthe major energy-consuming processes in phototransduction, under dark but notlight conditions. The hexose monophosphate pathway in ROS can produce 40 nmolNADPH/min/mg ROS protein to support retinal reduction occurring at a rate of1.2 nmol/min/mg ROS protein and the glutathione redox cycle to protect ROSfrom oxidative stress.37CHAPTER 2IDENTIFICATION OF A MAJOR PROTEIN ASSOCIATED WITH THE PLASMAMEMBRANE OF RETINAL PHOTORECEPTOR OUTER SEGMENTS ASGLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE2.1. MATERIALS2.1.1. Animal tissues: Fresh bovine eyes and blood were obtained fromIntercontinental Packers (Vancouver, British Columbia).2.1.2. Chemicals: NAD+-0-hexane-agarose was purchased from Pharmacia LKBBiotechnology Inc. (Uppsala, Sweden), and Immobilon membranes were obtainedfrom the Millipore Corp. (Bedford, MA). Arthrobacter ureafaciens neuraminidasewas a product of Boehringer Mannheim Biochemicals (Indianapolis, IN).Bicinchoninic acid (BCA) protein assay reagents were purchased from Pierce(Rockford, IL). All other chemicals were from Sigma Chemical Co. (St. Louis, MO)or British Drug House Chemical Co. (Montreal, Quebec). Glyceraldehyde-3-ph osph at e was prepared from glyceraldehyde-3-phosphate diethylacetalmonobarium salt (Sigma) using the Dowex-50 resin and stored at -70 °C.2.1.3. Immunoreagents: Freund's complete adjuvant was purchased from SigmaChemical Co.. Goat anti-mouse Ig antibodies were from Boehringer MannheimBiochemicals. Na'I was obtained from New England Nuclear (Lachine, Quebec).2.2. METHODS2.2.1. Preparation of bovine ROS3 8ROS were prepared under dim red light from 100 freshly dissected bovineretinas as described by Molday and Molday (1987). Briefly, dissected retinas weregently agitated in a homogenization solution containing 20 % w/v sucrose, 10 mMf3-D-glucose, 10 mM taurine, 0.25 mM MgC1 2 and 20 mM Tris-acetate, pH 7.4. Thesuspension was filtered through a Teflon 300 urn mesh screen and approximately 5-8ml filtrate was layered on a 28 - 60 % (w/v) sucrose gradient containing 10 mMtaurine, 10 mM /3-D-glucose, 0.25 mM MgC1 2 and 20 mM Tris-acetate, pH 7.4.After centrifugation in a SW 27 rotor (Beckman Instruments; Palo Alto, CA) at25,000 rpm for 50 min at 4 °C, intact ROS were collected as a band near the top ofthe gradients. The collected ROS were washed once with 5-10 volumes ofhomogenization solution by centrifugation in a Sorvall SS-34 rotor (DuPont Co.,Newton, CT) at 12,000 rpm for 10 min The washed ROS were resuspended in thehomogenization buffer at a protein concentration of 8-12 mg/ml and stored at -70°C.2.2.2. Isolation of ROS disk and plasma membranesROS plasma membrane was separated from disk membranes using a ricin-gold-dextran affinity density perturbation method previously described by Moldayand Molday (1987). Briefly, ROS (20-80 mg) in 3-8 ml of homogenization solutionwere treated with 0.1 units of neuraminidase for 1 h, labeled with ricin-gold-dextranparticles (approximate diameter, 15 nm), hypotonically lysed in 20 mM Tris-acetatebuffer, pH 7.2, and treated with 0.4 ug/ml trypsin for 30 min to dissociate the disksfrom the plasma membrane. After trypsin digestion was stopped with excesssoybean trypsin inhibitor, the disk membranes were separated from the ricin-gold-dextran-labeled plasma membrane by sucrose gradient centrifugation. The diskmembrane band and the plasma membrane pellet were collected, resuspended in 103 9% sucrose, 0.1 mM EDTA, 1 mM Na2HPO4, and 10 mM Tris-acetate, pH 7.2, andused immediately or stored at -70 °C until needed.2.2.3. Preparation of bovine erythrocyte ghostsRed blood cells and hemoglobin-free red blood cell ghosts were preparedaccording to the method of Fairbanks et al. (1971).2.2.4. Purification of G3PDAll procedures were carried out at 4 °C, and all centrifugations wereperformed in a Sorvall SS-34 rotor at 15,000 rpm for 15 min. Bleached orunbleached ROS (90 mg) were lysed in 15 ml of 10 mM Tris-acetate buffer, pH 7.4,and washed twice by centrifugation with 15 ml of the same buffer containing 100 uMGTP to remove transducin and other peripheral proteins of ROS (Kuhn, 1980).The ROS membranes were further extracted with 5 ml of NaC1 extraction buffer(150 mM NaCI, 1 mM EDTA, 5 mM Na2HPO4, and 10 mM Tris-acetate at pH 7.4).After removal of the stripped ROS membranes by centrifugation, the NaC1 extractcontaining G3PD was loaded onto a NAW-C 6-hexane-agarose column (0.5 mg -1.0mg of NaCI extract per ml of beads) equilibrated with the NaC1 extraction buffer.After the column was washed with 10 column volumes of 0.5 M NaC1, 1 mM EDTA,5 mM Na2HPO4, and 10 mM Tris-acetate at pH 7.4, G3PD was eluted with the NaC1extraction buffer containing 10 mM NAD+. A flow rate of 0.25 ml min -1 was usedthroughout the chromatography. Fractions containing the dehydrogenase activitywere pooled and protein concentration was determined after dialysis against 0.1 mMEDTA and 5 mM Na2HPO4, pH 7.4, at 4 °C. G3PD was extracted from isolatedROS plasma membranes and purified by affinity chromatography by the sameprocedure. Bovine erythrocyte G3PD was also extracted by isotonic washes asdescribed above for ROS membranes. The ROS membranes and purified ROS4 0plasma membranes after extraction were further washed with the NaCl extractionbuffer by centrifugation until no significant G3PD activity (less than 1 % of the totalG3PD activity associated with the membranes before NaC1 extraction) wasdetected. These G3PD-stripped ROS membranes and purified plasma membranewere used in the G3PD binding assays.2.2.5. Assay of G3PD activityG3PD activity in ROS membranes, NaCl extracts of ROS membranes andpurified G3PD preparations was monitored by following the reduction of NAD toNADH at 340 nm according to the method of Steck (1974). All assays were carriedout in 30 mM sodium pyrophosphate buffer, pH 8.4, in a final volume of 0.5 ml,containing 12 mM sodium arsenate, 1 mM NAD, and 1.5 mM glyceraldehyde-3-phosphate. One unit of enzyme activity is defined as the production of 1 umol ofNADH at 25 °C/min/mg of protein using 6.22 x 10 3 M-1 cm-1 as the extinctioncoefficient for NADH.2.2.6. Amino acid sequence analysisThe N-terminal sequence of NaCl-extracted 38 kDa protein from ROSplasma membrane and of the NAD + -C6-hexane-agarose column-purified G3PD wasdetermined by the Protein Microchemistry Centre at the University of Victoriausing an Applied Biosystems Gas-Phase Microsequenator. A protein search of thesequence was performed using a Swiss-Prot protein data base.22.7. Generation of anti-G3PD monoclonal antibodyHybridoma cell line gpd 2C11 was generated by fusion of NS-1 mousemyeloma cells with spleen cells from a BALB/c mouse immunized with NAD+affinity-purified bovine ROS G3PD as described by Lane et al., 1986. Briefly, a41female Balb/c mouse was immunized with subcutaneous injections of 50 ug ofG3PD emulsified in 0.1 nil of Freund's complete adjuvant at 3 week intervals. Anintraperitoneal booster injection of 50 ug G3PD was given 14 days after the thirdimmunization and cell fusion was carried out 5 days later. Approximately 1 x 10 8spleen cells were fused with 5 x 10 7 mouse NS-1 cells in 1 ml 50 % polyethyleneglycol 1500. Hybridoma cells secreting antibodies against G3PD were detected bysolid phase ELISA 10-14 days following cell fusion. Culture supernatant from thecloned antibody secreting hybridoma cell line was used for Western blot analyses.2.2.8. Conditions for extraction of G3PD from ROS membranesThe effect of salt, nucleotides and chelating agents on the extraction ofG3PD from ROS membranes was studied as follows: 1 mg of ROS was lysed in 1 mlof 10 mM Tris-acetate buffer, pH 7.4, for 1 h at 4 °C. The ROS membranes weresedimented by centrifugation at 15,000 rpm for 15 min in a Sorvall SS-34 rotor. TheROS membrane pellet was resuspended in 50 ul of 10 mM Tris-acetate buffer, pH7.4, containing 25-200 mM NaC1, 1 mM nucleotide, or 2 mM EDTA or EGTA.After incubation for 15 min at 4 °C, the supernatants were collected followingcentrifugation as above, and 25 ul was subjected to SDS-polyacrylamide gelelectrophoresis. The relative quantity of extracted G3PD protein was determinedby densitometry using a LKB Ultra Scan XL laser densitometer.2.2.9. Binding of G3PD to ROS disk and plasma membranesNAD+ affinity-purified G3PD (75 ug) was added to a mixture of purified diskand plasma membranes prepared as described above (3 mg of ROS disk membranesand 0.15 mg G3PD-stripped ROS plasma membranes in a total volume of 500 ul)and dialyzed overnight against 0.1 mM EDTA, 5 % (w/v) sucrose, and 5 mMNa2HPO4, pH 7.4, at 4 °C. The unbound G3PD was separated from the membranes4 2by centrifugation in a Sorvall SS-34 rotor at 17,000 rpm for 30 min. The disk andplasma membranes were then separated on a 25 -60 % (w/v) sucrose gradient in 20mM Tris-acetate, pH 7.4, by centrifugation at 45,000 rpm for 20 min in a TLS-55rotor (Beckman Instruments). The association of G3PD with the membranes wasassayed by both SDS-polyacrylamide gel electrophoresis and G3PD enzymaticactivity.2.2.10. Effect of trypsin and chymotrypsin on the G3PD binding sites on ROSmembranesAll centrifugations were carried out in a SS-34 rotor at 15,000 rpm for 15 minand proteolytic digestion was carried out at 25 °C. Aliquots of 1 mg G3PD-strippedROS membranes prepared as described above were treated with 10 ug/ml of eithertrypsin or chymotrypsin in 100 ul of 1 % sucrose, 0.1 mM EDTA and 5 mMNa2HPO4, pH 7.4 (buffer A). The digestion was stopped by transferring 25 ulaliqouts of digest to Eppendorf tubes containing appropriate protease inhibitors at 0(before the addition of protease), 10 and 50 min intervals. Trypsin andchymotrypsin digestion were stopped with 20 ul of 1 mg/ml of soybean trypsininhibitor and PMSF, respectively. The digested ROS membranes were then washedthree times in 0.2 ml of buffer A before incubation for 30 min with 40 ul of 0.1mg/ml NAD + affinity-purified G3PD previously dialyzed against 0.1 mM EDTAand 5 mM Na2HPO4, pH 7.4. The extent of G3PD binding to the membranes wasmonitored by measuring G3PD activity in both supernatants and membrane pelletsfollowing centrifugation.2.2.11. Polyacrylamide gel electrophoresisSDS-polyacrylamide slab gels were prepared and run as previously described(Molday and Molday, 1987), and gel slices were stained with Coomassie Blue or4 3electroblotted onto Immobilon membranes for Western blotting or N-terminalsequence analysis. Anti-G3PD monoclonal antibody gpd 2C11 and 125I-labeled goatanti-mouse Ig (specific activity 1-2 x 10 6 dpm/ug) were used as primary andsecondary antibodies, respectively, for Western blotting as previously described(Molday and Molday, 1987).2.3. RESULTS2.3.1. Identification of ROS 38 kDa protein as G3PDN-terminal sequence analysis: Previous studies have indicated that the ROSplasma membrane isolated by the ricin-gold-dextran affinity density perturbationmethod under hypotonic conditions contains a major protein which migrates justabove rhodopsin on SDS-polyacrylamide gels (Molday and Molday, 1987). Thisprotein having an apparent molecular weight of 38,000 was absent in isolated ROSdisk membranes (Fig. 13, lanes a and b). When the ROS plasma membrane wasextracted with 0.15 M NaC1, the major M r = 38,000 protein and minor proteins ofMr = 40,000 and 60,000 were released from the membrane (Fig. 13, lane d). Theextracted ROS plasma membrane contained significantly reduced amounts of the38-kDa protein (Fig. 13, lane c).The N-terminal sequence of the extracted 38-kDa protein was determinedafter electrophoretic transfer onto Immobilon membranes. As shown in Fig. 14, thesequence of the N-terminal 22 amino acids is essentially identical to the N-terminalsequence of glyceraldehyde-3-phosphate dehydrogenase (G3PD) from bovine,porcine, rat, and human liver and muscle (Kulbe et al., 1975; Nowak et al., 1981;Arcari et al., 1984; Fort et al., 1985).On the basis of this sequence identity, NAD + -agarose affinitychromatography was used to further purify the 38-kDa protein from the NaC1Fig. 13. SDS-polyacrylamide gel electrophoresis of the 38 kDa protein isolated fromROS plasma membrane.Bovine ROS plasma membrane was separated from disk membranes by aricin-gold-dextran affinity density perturbation method. The ROS plasmamembrane was then extracted with 0.15 M NaCl and this extract was subjected toNADtaffinity chromatography. Samples (20-30 ug) were subjected to SDS-polyacrylamide gel electrophoresis on a 9% slab gel and stained with CoomassieBlue. Lane a, ROS disk membranes; lane b, ROS plasma membrane; lane c, ROSplasma membrane after extraction with 0.15 M NaCl; lane d, NaCI extract of theROS plasma membrane; lane e, the 38 kDa protein purified from the NaC1 extractby NAD tagarose affinity chromatography.45Bovine ROS 1- V-K-V-G-V-N-G-F -G-R-I-G-R-L-V-T-R-A-A-F -N-8-Bovine and 1- V-K-V-G -V-N -G -F -G -R-I-G-R -L-V-T -R-A-A-F -N-S-porcine liverHuman liver 3- V-K-V-G-V-N-G-F-G-R-I-G-R-L-V-T-R-A-A-F-N-S-Human muscle 3- V-K-V-G -V Q -F-G-R-I-G-R-L-V-T-R-A-A-F-N-S-Rat muscle 1- V-K-V-G-V-N-G-F-G-R-I-G-R-L-V-T-R-A-A- POFig. 14. The N-terminal amino acid sequence of bovine ROS 38 kDa protein andhomology to G3PD from other mammalian tissues.The sequence of the first 22 amino acid residues of bovine ROS 38 kDaprotein isolated by NaC1 extraction of ROS plasma membrane followed by NAD+-affinity chromatography was determined by partial N-terminal amino acidsequencing. Sequences for other G3PD were obtained through the Swiss-Protprotein data base (Kulbe et al., 1975; Nowak et al., 1981; Arcari et al., 1984; Fort etal., 1985). Amino acid residues in the listed G3PD sequences which are differentfrom ROS G3PD are circled. The position of the first amino acid residue of eachdisplayed sequence in G3PD is indicated by the number preceding the sequence. Anumber "1" means the first amino acid residue of the sequence shown is the N-terminal amino acid residue of G3PD.4 6extract. As shown by SDS-polyacrylamide gel electrophoresis (Fig. 13, lane e), the38 kDa protein was purified to homogeneity by this procedure.Specific activity of G3PD purified from ROS and ROS plasma membranes:To further confirm the identity of the 38-kDa protein as G3PD and to obtaininformation on its abundance, the dehydrogenase activity was determined duringpurification of this protein from both ROS and isolated ROS plasma membranes.Tables III and IV show typical results obtained from the purification of G3PD fromROS and ROS plasma membranes.Hypotonic lysis of ROS resulted in the recovery of 70-78 % of the G3PDactivity in the ROS membrane fraction. Over 50 % of the lost G3PD activity wasrecovered in the supernatant fraction. Greater than 99 % of the G3PD activityassociated with the ROS membrane fraction could be extracted from the membranewith 0.15 M NaCl. As shown in Table IV, the G3PD activity was associated with theplasma membrane fraction when the plasma membrane was separated from diskmembranes by the ricin-gold-dextran affinity density perturbation method (Moldayand Molday, 1987). The mild trypsin treatment of ROS membranes used todissociate the plasma membrane from disk membranes did not result in significantloss in G3PD activity from the ROS membranes (Table IV). Approximately 60-75% of the ROS membrane protein content was recovered in the disk and plasmamembrane fractions, whereas approximately 30-40 % of the ROS membrane-associated G3PD activity was recovered in these fractions. It would appear that upto half of the G3PD is inactivated and/or released from the ROS plasmamembranes during sucrose gradient centrifugation and subsequent washingprocedures. Essentially all the membrane-bound G3PD activity is associated withthe plasma membrane fraction (Table IV). In most isolated disk membranepreparations, no G3PD activity was detectable. In disk preparations which didcontain G3PD activity (Table IV, for example), the specific activity was over 200-4 7Table IIIPurification of Glyceraldehyde-3-phosphate Dehydrogenase From Bovine ROSProtein^Specific^Total^% Activity(mg) Activity^Activity Recovered(Units/mg)^(units)Intact ROS 91 1.8 164 100Lysed ROS 73 1.8 128 78NaCl Extract 1.2 95 114 70NAD + -agarose 1.0 95 95 584 8Table IVPurification of Glyceraldehyde-3-phosphate Dehydrogenase From Purified Bovine ROSPlasma MembranesProtein^Specific^Total^% Activity(mg)^Activity Activity Recovered(Units/mg)^(units)Ricin-gold ROS 19 1.9 37 100Lysed ROS 16 1.6 26 70Trypsinized ROS 16 1.5 24 65Disk membranes' 12 0.04 0.5 1Plasma membranes 0.67 11.0 7.4 20NaC1 Extract 0.08 92.5 7.4 20NAD + -agarose 0.07 91.4 6.4 17a Activity quoted in this table represents an upper limit in the amount of G3PD activityassociated with the disk membranes; most preparations did not have detectableG3PD activity associated with the disk membranes.49fold lower than that for the plasma membrane fraction. This residual activitypresent in some isolated disk preparations is most likely derived from contaminatingplasma membrane fragments which are not completely dissociated from the disksduring trypsin treatment (Molday and Molday, 1987).The specific activity of G3PD isolated from total ROS membranes (TableIII) or from plasma membranes (Table IV) by NaCl extraction and NAD -agaroseaffinity chromatography was 90-100 units/mg. This is within the range of specificactivities reported for G3PD isolated from other mammalian systems (Heinz andKulbe, 1970; Dagher and Deal, 1977; Kulbe et al., 1982; Wang and Alaupovic,1980). Comparison of the specific activities of G3PD in ROS and ROS plasmamembranes with that of the purified enzyme indicates that G3PD constitutes about2 % of the total ROS protein and over 11 % of the plasma membrane protein(Tables III and IV).Purification of G3PD from ROS was also monitored by SDS-gelelectrophoresis (Fig. 15). On 9 % SDS-polyacrylamide gels, G3PD co-migrates withthe a and f3 subunits of ROS transducin. On gradient gels, however, these threecomponents were resolved with G3PD migrating between the a - and ri -transducin subunits (data not shown). After removal of transducin with GTP, G3PDwas extracted with 150 mM NaCI and purified to homogeneity as a 38 kDa proteinby NAD +-agarose chromatography. As in the case of G3PD from other systems,bovine ROS G3PD was found to have a molecular weight of approximately 140,000by gel filtration chromatography under non-denaturing conditions. Accordingly,ROS G3PD, like G3PD from various muscle sources (Harris and Perham, 1965 andHarrington and Karr, 1965), appears to be a tetrameric protein composed ofidentical subunits.Monoclonal antibody to G3PD: Monoclonal antibody gpd 2C11 raisedagainst NAD + affinity-purified ROS G3PD was found to bind to G3PD in both5 0Fig. 15. Purification of the 38 kDa protein from bovine ROS membranes.ROS which were hypotonically lysed and washed with GTP were furtherextracted with 0.15 M NaCl. The 38 kDa protein was isolated from this extract byNAD+ affinity chromatography. Samples (20-30 ug) were subjected to SDS gelelectrophoresis on a 9% polyacrylamide slab gel and stained with Coomassie Blue.Lane a, ROS isolated from a sucrose gradient; lane b, hypotonically lysed ROSmembranes washed with 0.1 mM G ; lane c, ROS membranes after extractionwith 0.15 M NaCl; lane d, NaC1 extract of ROS membranes; lane e, the 38 kDaprotein purified from the NaC1 extract by NADtagarose affinity chromatography.51bovine ROS and bovine red blood cells by Western blotting (Fig. 16, lanes a and b).Extraction of erythrocyte ghosts with 150 mM NaC1 resulted in a complete elutionof G3PD from red blood cell ghosts (Fig. 16, lanes d and f). Two faint protein bandsin the molecular mass range of 38-39 kDa were visible on the SDS-polyacrylamidegel following the NaC1 extraction of ROS plasma membrane (Fig. 16, lanes c). Theresidual G3PD associated with the ROS plasma membrane was likely due tononspecific adsorption of G3PD to ricin-gold-dextran particles since G3PD could becompletely extracted from lysed ROS membranes. A small amount of G3PD wasdegraded by trypsin during the plasma membrane purification process, generatingfragments of G3PD which still bind to the plasma membrane (Fig. 16, lanes a).Monoclonal antibody gpd 2C11 was found to cross-react with G3PD from humanred blood cells and rabbit skeletal muscle (data not shown).2.3.2. Nature of G3PD binding to ROS membranesConditions for G3PD elution from ROS membranes: The effect of NaClconcentration on the elution of G3PD from hypotonically lysed ROS membraneswas studied. As shown in Fig. 17, a sigmoidal relation was apparent for the releaseof G3PD as a function of NaC1 concentration. Half-maximum release was obtainedat a concentration of 120 mM NaC1, and 95 % of maximum release was obtained at150 mM NaCl. Other salts such as KC1 and Na 2HPO4 were also effective inreleasing G3PD from ROS membranes indicating that ionic strength was thecontributing factor. Extraction of G3PD was found to be independent of the state ofbleaching of the ROS. In agreement with human erythrocyte G3PD (Kant andSteck, 1973; Shin and Carraway, 1973), 1 mM ATP and NADH partially releasedG3PD from ROS membranes (Table V). Other nucleotides including NAD, cGMPand cAMP, and chelating agents including EDTA and EGTA resulted in the releaseof less than 10 % of G3PD activity from ROS membranes. The basis for this52Fig. 16. Anti-G3PD monoclonal antibody gpd 2C11 labelled Western blot of ROSplasma membrane and red blood cell ghosts and their NaC1 extracts.Coomassie Blue-stained gel (CB) and immunoblot (G3PD) of themembranes and their NaC1 extracts is shown for comparison. Lanes a, ROS plasmamembrane; lanes b, red blood cell ghosts; lanes c, NaCl extracted ROS plasmamembrane; lanes d, NaCl extracted red blood cell ghosts; lanes e, NaCl extract ofROS plasma membrane; lanes f, NaC1 extract of red blood cell ghosts. Each lanecontains 20-30 ug of membranes or 5 ug of NaC1 extract.50 200150100100806040200 ^053NaC1 (mM)Fig. 17. Effect of NaC1 concentration on the extraction of the 38 kDa protein fromROS membranes.ROS membranes (1 mg protein) were incubated in 10 mM Tris buffer, pH7.4, containing 20 - 200 mM NaC1 for 15 min at 4 °C. The ROS membranes weresedimented by centrifugation and the supernatant extract was subjected to SDS-polyacrylamide gel electrophoresis. The amount of 38 kDa protein in thesupernatant was quantified by scanning densitometry of Coomassie Blue stainedgels. The assay was carried out in duplicate.54Table VEffect of Nucleotides and Chelating Agents on the Binding of the 38 kDa Protein toROS MembranesEluting Agent % 38 kDa protein Eluted FromROS Membranes150mM NaCl. 1001mM ATP 491mM NAD 91mM NADH 341mM cGMP 41mM cAMP 62mM EDTA 92mM EGTA 9* The amount of 38 kDa protein eluted by each eluting agent was assessed bycomparison with its elution from the ROS membranes by 150 mM NaCl. The assaywas carried out in triplicate.55nucleotide-specific elution of G3PD from ROS membranes and from erythrocytemembranes is not known. However, in the erythrocyte system, it has beenpostulated that ATP induces conformational changes in membrane proteinsresulting in the dissociation of G3PD from the membrane (Shin and Carraway,1973). The ionic strength of the buffer used to extract G3PD from ROS is similar tothat used by Kuhn to extract a 35-kDa protein from hypotonically lysed ROS (Kuhn,1980).G3PD specifically binds to ROS plasma membrane: The binding of affinity-purified G3PD to ROS disk and G3PD-stripped ROS plasma membranes wasstudied by SDS-gel electrophoresis and enzyme activity measurements. Asillustrated in Fig. 18, G3PD specifically bound to ROS plasma membranes, but notto disk membranes in low ionic strength buffer (0.1 mM EDTA and 5 mMNa2HPO4). No G3PD activity was detected in either disk membranes or in NaC1-extracted plasma membranes before the addition of exogenous G3PD. Whenpurified G3PD was reassociated with disk membranes and G3PD-stripped ROSplasma membrane at a disk to plasma membrane protein ratio of 20:1, a G3PDspecific activity of 5 units/mg of protein was measured for plasma membranes, butno activity was detected for disk membranes. When G3PD-stripped ROSmembranes were treated with 10 ug/ml trypsin or chymotrypsin prior toreassociation with G3PD, binding of G3PD to these membranes was reduced by upto 75 % (Table VI). This indicates that G3PD preferentially binds to a protease-sensitive protein in ROS plasma membrane.2.4. DISCUSSIONIn this study, partial N-terminal sequence analysis and specific enzymeactivity measurements have confirmed that the major 38-kDa protein associated56Fig. 18. Specific binding of the 38 kDa protein to bovine ROS plasma membrane.Purified ROS disk (3 mg) and NaC1 extracted plasma membranes (0.15 mg)were mixed with NAD+-agarose purified G3PD (75 ug) in a total volume of 0.5 ml.After dialysis against 0.1 mM EDTA, 5 % sucrose and 5 mM sodium phosphatebuffer, pH 7.4, the membranes were separated from the unbound 38 kDa protein bycentrifugation and subjected to SDS-polyacrylamide gel electrophoresis. Lane a,ROS disk membranes before the addition of 38 kDa protein; lane b, ROS diskmembranes incubated with 38 kDa protein and washed by centrifugation; lane c,NaC1 extracted ROS plasma membranes before the addition of 38 kDa protein; laned, ROS plasma membrane incubated with 38 kDa protein and washed bycentrifugation; lane e, NAD tagarose purified 38 kDa protein.57Table VIEffect of Trypsin and Chymotrypsin on ROS G3PD Binding SitesG3PD activity.TrypsindigestChymotrypsindigestSample control 10 50 control 10 50min min min minROS membranepellet280 60 55 280 105 50Supernatant 34 200 220 34 150 190. Aliquots of G3PD-stripped ROS membranes were treated with ttypsin orchymottypsin at 25 °C for 0, 10 or 50 minutes before incubation with NAD + -affinitypurified G3PD. The extent of G3PD association with digested membranes wasassessed by assaying G3PD activity associated with the membrane pellet andsupernatant of each membrane aliquot. Activities are expressed as nanomoles ofNADH produced per min for each membrane aliquot. The assay was carried out intriplicate.58with purified ROS plasma membrane is G3PD. This enzyme makes up as much as 2% of the total ROS protein and can be readily isolated from ROS membranes orpurified ROS plasma membrane by extraction with 0.15 M NaCI and NAD+-affinitychromatography. Activity measurements suggest that G3PD makes up at least 11 %of the protein in ROS plasma membrane prepared by hypotonic lysis of ROS anddensity gradient centrifugation in low ionic strength buffers. Coomassie Bluestaining of ROS plasma membrane proteins separated by SDS-gel electrophoresis,however, indicates that up to 17 % is the 38 kDa protein (Molday and Molday,1987). This difference may be explained, in part, by the presence of anotherprotein(s) with similar mobility on the SDS-polyacrylamide gel. This protein may bethe 39-kDa protein reported by Matesic and Liebman (1987) to have cGMP channelactivity. Earlier, Kiihn (1980) reported that a 35 kDa protein could be specificallyeluted from ROS membranes with 100 mM Tris buffer. It appears likely that thisprotein is G3PD as reported here.Glyceraldehyde-3-phosphate dehydrogenase binds to ROS plasmamembranes but not disk membranes. Activity measurements and SDS-gelelectrophoresis of ROS plasma and disk membranes isolated from trypsin-treatedROS membranes indicate that G3PD is only present in the plasma membranefraction. The small amount of G3PD activity measured in the disk membranefraction in some preparations is most likely due to contamination by plasmamembrane fragments since G3PD specific activity of these disk fractions is at leasttwo-hundredfold less than that of plasma membranes from the same preparation.Furthermore, G3PD reassociates with G3PD-stripped plasma membrane, but notdisk membranes, even when disk membranes are present in significantly higherconcentrations. Trypsin treatment used in the separation of disks from plasmamembranes is not responsible for the absence of G3PD binding to disk membranessince it has been previously shown that disk membranes prepared in the absence of59trypsin do not contain the 38 kDa band characteristic of G3PD (Molday andMolday, 1987).Like G3PD of red blood cell membranes, G3PD of ROS specifically andreversibly associates with the plasma membrane under conditions of low ionicstrength. Although there has been some controversy concerning the physiologicalrelevance of low ionic strength binding of G3PD to cell membranes (Keokitichaiand Wrigglesworth, 1980; Kliman and Steck, 1980; Solti et al., 1981; Rich et al., 1984,1985; Ballas et al. , 1985), recently Rogalski et aL (1989) have shown byimmunofluorescence microscopy that G3PD is preferentially associated with theplasma membrane in intact and lysed human red blood cells. It is, therefore,possible that G3PD binding to ROS plasma membrane may also occur in situ. Therelease of 20-30 % of G3PD activity into the supernatant upon hypotonic lysis ofROS, however, may suggest that a portion of G3PD is not tightly associated withROS membranes.The site to which ROS G3PD binds is not yet known. Studies reported hereindicate that G3PD binds to a protease-sensitive ROS plasma membrane-specificprotein. In the case of red blood cell membranes, it has been shown that G3PDspecifically binds to the negatively charged N-terminus of Band 3, the anion channelprotein of red blood cell membranes (Yu and Steck, 1975). Since the binding ofG3PD to ROS plasma membranes exhibits similar properties, it is likely that anegatively charged segment of a plasma membrane protein serves as the binding sitefor ROS G3PD.Glyceraldehyde-3-phosphate dehydrogenase is an important enzyme in theglycolytic pathway. It is found in exceedingly large quantities in many cells andtissues. For example, G3PD constitutes about 5-7 % of the human red blood cellplasma membrane proteins (Kant and Steck, 1973), over 5 % of the total yeastprotein (Krebs et al., 1953), and up to 7 % of total rabbit skeletal muscle nonstromal6 0protein (Czok and Bilcher, 1960). As yet, the need for such a large excess of G3PD(up to 3 orders of magnitude in human red blood cells) compared to some glycolyticenzymes such as hexokinase and pyruvate kinase is not known. Several functionsoutside its role in glycolysis, however, have been demonstrated for G3PD in vitro.These functions include ATP-modulated bundling of brain microtubules (Huitoreland Pantaloni, 1985), catalysis of triad junction formation from rabbit muscletransverse tubules and terminal cisternae (Caswell and Corbett, 1985) and ATP-dependent protein kinase activities in rabbit muscle (Kawamoto and Caswell, 1986).The role of G3PD in ROS is not known. The ROS, however, is a highly activeorganelle requiring large quantities of GTP, cGMP and ATP for thephototransduction process. If glycolysis does occur in ROS, other glycolytic enzymesshould also be present in this organelle. The band at 40 kDa which elutes fromROS membranes with G3PD may in fact be aldolase. This enzyme is known to beassociated with Band 3 of red blood cell membranes under conditions of low ionicstrength and can be released with physiological buffer (Strapazon and Steck, 1976,1977; and Murthy et aL, 1981). Detection of other glycolytic enzymes and of glucosetransport activity in ROS is discussed in Chapter 3.61CHAPTER 3DETECTION OF GLYCOLYTIC ENZYMES AND A GLUT-1 GLUCOSETRANSPORTER IN THE OUTER SEGMENTS OF ROD AND CONEPHOTORECEPTOR CELLS3.1. MATERIALS3.1.1. Animal tissues: Fresh bovine eyes were obtained from IntercontinentalPackers (Vancouver, British Columbia) or J & L Meats (Surrey, British Columbia).Fresh chicken eyes were from Hallmark Poultry Processors Ltd. (Vancouver, BritishColumbia). Fresh human blood was obtained from healthy donors.3.1.2. Chemicals: 3-04 14C]methyl-glucose was a product of New England Nuclear(55 mCi/mmol) and the Ready Protein scintillation cocktail was from Beckman.Immobilon and nitrocellulose membranes were obtained from the Millipore Corp.and Schleicher & Schuell (Keene, NH), respectively and Sepharose 6B was fromPharmacia. All other chemicals were from Sigma or British Drug House ChemicalCo. Glyceraldehyde-3-phosphate was prepared from glyceraldehyde-3-phosphatediethylacetal monobarium salt (Sigma) using the Dowex-50 resin and stored at -70oC.3.1.3. Immunoreagents: Reagents used here are described in Chapter 2.3.2. METHODS3.2.1. Preparation of bovine ROS, ROS membranes and ROS lysatesROS were prepared under dim red light from freshly dissected bovine retinatissue by sucrose gradient centrifugation (Molday et al. 1987). ROS membranes andROS lysates (soluble fraction) were prepared as follows: 15 mg (protein) of isolated62ROS in 20 mM Tris-acetate buffer, pH 7.4 containing 20% (w/v) sucrose and 0.2mM MgC1 2 were centrifuged in a Sorvall SS-34 rotor at 10,000 rpm for 10 min. TheROS were resuspended in 1 ml of hypotonic extraction buffer (10 mM Tris-acetatebuffer, pH 7.4) for 10 min at 4 °C to initiate lysis, and were then vigorously vortexedfor 30 s. The ROS membranes and ROS lysate were separated by centrifugation in aSorvall SS-34 rotor at 15,000 rpm for 15 minutes. The ROS membrane pellet wasextracted two more times with 1 ml hypotonic extraction buffer. The ROSmembrane pellet was resuspended in the same buffer at a final concentration of 15mg protein/ml; the lysate from three hypotonic extractions were pooled forglycolytic enzyme assays. In some experiments, ROS membranes prepared by thehypotonic extraction of ROS were further extracted 3 times with 1 ml each ofisotonic extraction buffer (10 mM Tris-acetate, pH 7.4 containing 150 mM NaCl) asdescribed above. The isotonic extracts were pooled for enzyme assays.ROS disk and plasma membranes were prepared using a ricin-gold dextranperturbation method in the presence of trypsin as previously described (Molday andMolday, 1987). Protein concentrations were determined by bicinchoninic acidmethod as described by the manufacturer, Pierce, Rockford, IL.3.2.2. Preparation of red blood cells, red blood cell ghosts, and rat brainmicrosomesBovine and human red blood cells and hemoglobin-free red blood cell ghostswere prepared according to the method of Fairbanks et al. (1971). Rat brainmicrosomes were prepared as described by Wang (1987).3.2.3. Glycolytic enzyme activity assaysGlycolytic enzyme activities in intact ROS, lysed ROS membranes and ROSlysates were monitored by following the consumption or production of NADH (or63NADPH in the case of hexokinase) at 340 nm either directly or through coupledenzyme reactions. Assay of hexokinase was carried out according to the method ofEasterby and Qadri (1982); aldolase by the method of Yeltman and Harris (1982);glyceraldehyde-3-phosphate dehydrogenase (G3PD) by the method of Steck (1974);pyruvate kinase by the method of Kahn and Marie (1982) and lactatedehydrogenase (LDH) by the method of Lee et a/. (1982). The phosphofructokinaseassay was carried out according to the method of Harris et a/. (1982) except 1 mMATP was used. Phosphoglycerate kinase (PGK) activity was measured using themethod of Kulbe and Bojanovski (1982) except 100 mM Tris-HC1 buffer, instead of100 mM triethanolamine buffer, was used. All assays were carried out in a finalvolume of 0.5 ml at 23 °C. One unit of enzyme activity is defined as the productionof 1 umol of NAD(P)H/min/mg of protein using 6.22 x 10 3 M-1 cm-1 as theextinction coefficient for NADH and NADPH. In all enzyme assays, the substratefor the enzyme to be measured was added to initiate the reaction. NegligibleNADH production or consumption was detected before the addition of thesubstrate.3.2.4. Determination of glucose transport activity in intact ROSROS were isolated by sucrose density centrifugation in the absence ofglucose. The isolated ROS were washed once with transport assay buffer containing150 mM NaCl, 1 mM MgC1 2 and 20 mM 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pH 7.2 (300 ml per 50 mg ROS), bycentrifugation in a Sorvall SS-34 rotor at 9,000 rpm for 8 min. The ROS pellet wasresuspended to a final concentration of 7 - 10 mg protein/ml in the assay buffer. Astock solution of 3-0-[ 14C]methylglucose in ethanol was dried down under vacuumin Eppendorf tubes and rehydrated to the original volume in the assay buffer.Transport assay stop buffer containing 1 mM phloretin was prepared shortly before64use by dilution of a 100 mM phloretin/methanol stock solution with the assaybuffer. All transport assays were carried out in triplicate.Net influx assay:^Net uptake of 3-04 14C]methylglucose by intact ROSwas carried out at 23 °C based on modified methods of Toyoda et al. (1986) andLowe and Walmsley (1985). One ul of assay buffer containing 3-0-[14C]methylglucose (0.1 uCi) was placed at the bottom of a round-bottom glassculture tube (13 x 100 mm) placed on an orbit shaker swirling at 150 rpm. Thetransporter assay was initiated by the addition of 40 ul of ROS suspension andterminated at the desired time by the addition of 1 ml of stop buffer containing 1mM phloretin. The ROS suspension was immediately centrifuged in a BeckmanMicrofuge E for 20 s. The supernatant was carefully removed, and the ROS pelletwas washed by resuspension in another 1 ml of stop buffer and centrifugation. TheROS pellet was directly dissolved in the Ready Protein scintillation cocktail andcounted in a liquid scintillation counter. No detectable amount of ROS was lostduring the two centrifugation steps. To study cytochalasin B inhibition of 3-0-[14C]methylglucose uptake, the ROS suspension was pre-incubated with 0.1 mMcytochalasin B for at least 30 min prior to the addition of 3-04 14C]methylglucose toinitiate the transport assay. Zero-time controls were carried out by adding the stopbuffer to the ROS before initiating the transport assay. Counts obtained from zero-time controls accounted for less than 1 % of radioactivity observed at uptakeequilibrium. Net uptake of 3-0-[ 14C]methylglucose by red blood cells was carriedout as described above except the washed red blood cell pellets were lysed in waterand the RBC proteins were precipitated with 10 % trichloroacetic acid. Thesupernatants were collected and counted for radioactivity.Equilibrium exchange assay:^Equilibrium exchange uptake of 3-0-[14C]methylglucose was also carried out using the same procedure described aboveexcept that ROS were pre-equilibrated for at least 30 min with the indicated65concentration of non-radioactive 3-0-methylglucose. Increases in osmotic pressuredue to the inclusion of 3-0-methylglucose in the assay buffer was compensated bydecreasing the concentration of NaC1 in the buffer. Exchange velocity at each 3-0-methylglucose concentration was determined graphically from the initial rate of 3-0414C]methylglucose tracer uptake into ROS.Net efflux assay: The rate of 3-04 14C]methylglucose efflux from ROSwas measured at 23 °C. The ROS suspension was first incubated with 0.1 mM 3-0-[14C]methylglucose for at least 30 min and the efflux assay was initiated by theaddition of 400 ul of 3-0-rqmethylglucose loaded ROS suspension into 4.6 ml ofassay buffer with or without 1 mM phloretin in a glass culture tube swirling on anorbital shaker as described above. To terminate the assay, 0.5 ml aliquots of theROS suspension were removed at pre-determined time intervals and added into 1ml of stop buffer containing 1 mM phloretin. The ROS pellets were washed bycentrifugation and counted for radioactivity as described for the net uptake assay.3.2.5. Monoclonal and polyclonal antibodiesMonoclonal antibodies gpd 2F4, 3D4 and 3E11 and pgk 1C9 were obtainedfrom the corresponding hybridoma cell lines generated by fusion of NS-1 mousemyeloma cells with spleen cells from BALB/c mice immunized with purified humanred blood cell G3PD and Baker's yeast PGK, respectively as described in Chapter 2.Culture supernatants from gpd 3E11 and pgk 1C9 cell lines were used in Westernblots and immunofluorescence labeling. Hybridoma cell culture fluid containingmonoclonal antibody MAb65 against H-type lactate dehydrogenase (Pan et aL,1989) was generously provided by Dr. P.G. Isaacson (University College, London).Monoclonal antibody 3D6 generated against bovine rhodopsin has been shown tocross-react with bovine and frog red- and green-sensitive cone opsin (Hicks andMolday, 1986). Chicken red-sensitive cone pigment, iodopsin, also contains the rho6 63D6 binding site (Hodges et al., 1988) on the basis of sequence analysis of chickeniodopsin by Kuwata et aL (1990).Antisera against the C-terminus (synthetic peptide 480-492) of the rat brainglucose transporter (anti-GLUT-1) and against rat liver glucose transporter (anti-GLUT-2) were obtained from East-Acres Biologicals, Southbridge, MA; antiserumagainst insulin regulatable glucose transporter (anti-GLUT-4) was fromCalbiochem, San Diego, CA. Goat anti-mouse Ig was coupled to Sepharose 6B bythe cyanogen bromide method (March et al. 1974).3.2.6. Radioimmune competitive inhibition assaySolid-phase radioimmune competitive inhibition assays were carried out aspreviously described (Molday and MacKenzie, 1983). Briefly, ROS membranes orbovine or human red blood cell ghosts were solubilized in 1% Triton X-100 anddiluted in PBS to obtain a Triton X-100 concentration of 0.1%. Fifty ul of serially-diluted samples in 0.1% Triton X-100/PBS were incubated with 25 ul of 1000 timesdiluted rabbit anti-GLUT-1 glucose transporter antiserum for 1 h at 23 °C. Theunbound antibody was then determined by a solid phase radioimmune assay usingTriton X-100 solubilized human red blood cell ghosts (7.5 ug per well) asimmobilized antigen. 125I-goat anti-rabbit Ig (specific activity 2 x 10 6 dpm/ug) wasused as the secondary antibody and the level of 'I was counted in a Beckman 8000counter.3.2.7. Immunofluorescence microscopyImmunofluorescence microscopy was carried out according to the method ofJohnson and Blanks (1984) with a few modifications. Bovine eyes for cryostatsectioning were enucleated and fixed by immersion in 3 % (w/v) paraformaldehyde-PBS, pH 7.3 at 23 °C for 3 h. Some eyes were fixed in the presence of 30% (w/v)67sucrose. The inclusion of sucrose in the fixation buffer did not affect subsequentlabeling of retina sections by the antibodies. Paraformaldehyde fixed pieces of theeyecup were infiltrated with 8 % acrylamide for 8 to 12 h at 4 °C before initiatingacrylamide polymerization with ammonium persulfate. The acrylamide embeddedretina was frozen in Tissue Tek over the surface of liquid nitrogen. The embeddedretinas were sectioned and used for immunolabeling within 72 h. Prolonged storageof Tissue Tek embedded retina or retina sections at -20 °C diminished the intensityof antibody labeling and increased the auto-fluorescence of retina sections. Six umsections were first blocked with 5 % goat serum in PBS containing 0.1 % Triton X-100 for 30 min at 23 °C. The blocked sections were then incubated at 23 °C for 1-2h with primary antibodies: anti-G3PD, anti-PGK and anti-LDH monoclonalantibody culture supernatants or anti-GLUT-1 antiserum diluted 2-50x in PBScontaining 0.1 % Triton X-100 and 5 % goat serum. For controls, monoclonalantibody supernatants used in labeling were immunoprecipitated with goat anti-mouse Ig-Sepharose 6B. Anti-GLUT-1 rabbit polyclonal antiserum was pre-adsorbed with bovine red blood cell ghosts which exhibited very low glucosetransport activity (Hoos et al., 1972). For controls, the anti-GLUT-1 antiserum waspre-adsorbed with human red blood cell ghosts which contain relatively largeamounts of the glucose transporter. The labeled sections were washed four timeswith PBS and incubated for 1 h at 23 °C with affinity-purified FITC conjugated goatanti-mouse Ig or goat anti-rabbit Ig (10 ugiml) antibodies in blocking buffer.Sections were washed four times with PBS. Immunofluorescence microscopy wascarried out with a Zeiss Axiophot photomicroscope equipped with a verticalilluminator for epifluorescence. Photographs were taken using Kodak Tri-X 400film.3.2.8. Polyacrylamide gel electrophoresis and Western blotting68SDS-polyacrylamide gel electrophoresis was carried out on slab gels usingthe buffer system of Laemmli (1970). Gel slices were either stained with CoomassieBlue or electroblotted onto Immobilon or nitrocellulose membranes for Westernblots. Blots labelled with primary antibodies were detected with 125I-goat anti-mouse Ig or 125I-goat anti-rabbit Ig (1-2 X 106 dpm/ug) as previously described(Molday et al. 1987). Immunoblots were subjected to autoradiography overnight.3.3. RESULTS3.3.1. Glycolytic enzymes in photoreceptor rod outer segmentsGlycolytic enzyme activities in bovine ROS preparations: Glyceraldehyde -3 -phosphate dehydrogenase (G3PD), a key enzyme in glycolysis, has been shown to bea major protein associated with the plasma membrane of ROS (Hsu and Molday,1990; see Chapter 2). To determine if other glycolytic enzymes are also present inisolated ROS preparations, established spectrophotometric methods were used tomeasure the activities of other glycolytic enzymes. As shown in Table VII, ROSpreparations were found to contain significant activities of six other glycolyticenzymes including hexokinase and LDH, the first and last enzymes in the anaerobicglycolytic pathway. With the exception of G3PD, the enzyme activities were foundto be present in the soluble fraction after extraction of ROS with hypotonic buffer(Table VII). Isotonic conditions were required to elute most of the G3PD activityas previously shown (Chapter 2); isotonic buffer was also required to extract thealdolase activity which had not been extracted by hypotonic buffer. Glycolyticenzyme activities have also been detected by McConnell et al. (1969) and Lopez-Escalera et al. (1991), but differ from values reported here by 3 to 140 fold (TableVIIB) Large discrepancies in these three sets of data are probably due to variationin the intactness of prepared ROS and the condition of enzyme assays. Lowerglycolytic enzyme activities (2-140 fold lower compared to that found in the present6 9TABLE VIIGlycolytic Enzyme Activities in Bovine ROSEnzyme Specific Activities'^% Enzyme Activity in Extractb(Units/mg ROS)Hypotonic IsotonicG3PD 1.8 + 0.4 31.2 + 9.3 % 68.7 + 9.3 %LDH 3.0 + 0.4 94.4 + 1.8 % 5.6 + 1.8 %PK 0.7 + 0.1 92.5 + 4.9 % 7.5 + 4.9 %PGK 0.6 + 0.1 88.2 + 6.0 % 11.8 + 6.0 %PFK 0.1 + 0.01 93.0 + 4.2 % 7.0 + 4.2 %ALD 0.08 + 0.02 75.3 + 9.3 % 24.6 + 9.3 %HK 0.02 + 0.004 98.0 + 2.8 % 2.0 + 2.8 %a average data (units / mg ROS protein) from seven ROS preparations.b determined by assigning the total activity of each enzyme detected in hypotonicand isotonic extracts as 100 %. No enzyme activity was detectable in ROSmembranes after isotonic extraction.70TABLE VIIbComparison of Bovine ROS Glycolytic Enzyme Activities with Values Reported in theLiteratureEnzyme Activity (umol/min/mg ROS protein)Enzyme PresentStudyMcConnellet. al. (1969)Lopez-Escaleraet. al. (1991)G3PD 1.8 0.11LDH 3.0 0.06 4.6aPK 0.7 0.005 5 . 5bPGK 0.6 0.14PFK 0.1 -ALD 0.08 0.02HK 0.02 0.008Activities were calculated from 12.6 mM/sa and 15.0 mM/sb respectively assumingthat concentration of rhodopsin in ROS is 3 mM (Lopez-Escalera et al., 1991),rhodopsin constitutes 70 % of total ROS protein and rhodopsin has a molecular weightof 38,000 g/mol.71study) observed by McConnell et al. (1969) may be, in part, due to loss of glycolyticenzymes from ROS during a lengthy ROS isolation procedure which requiredrepeated sedimentation and resuspension of ROS. Glycolytic enzyme activities havebeen found to slowly leak out from isolated ROS even when ROS were resuspendedat high protein concentrations in 20 % sucrose and 5 % Ficoll (data not shown).For the present study, ROS were isolated from freshly dissected retinas by onesucrose density gradient centrifugation step within one and half hours after retinadissection. Glycolytic enzymes were immediately extracted from ROS by hypotonicand isotonic extraction after ROS isolation. Lopez-Escalera et al. (1991) alsoisolated ROS using a single sucrose gradient centrifugation procedure similar to thatused in the present study. Higher glycolytic enzyme activities (1.5 to 8 fold highercompared to that found in present study) observed by Lopez-Escalera et aL (1991)could be due to differences in enzyme assay procedures. Nevertheless, despitedifferences in these three sets of data, they all suggest that ROS contain higherglycolytic enzyme activities than erythrocytes, muscle and liver cells.The ratio of activity of each of the six measured glycolytic enzymes to theactivity of G3PD was compared in ROS, human red blood cells (Maretski et al.,1989) and rabbit muscle (Scopes and Stoter, 1982). As shown in Table VIII, theactivity ratios are within the same order of magnitude suggesting that theseglycolytic enzymes are present in similar proportions within these cells (Table VIII).A G3PD specific activity of 11 units per mg ROS plasma membrane protein ishigher than the specific activity of G3PD of 1.83 units per mg red blood cell ghostmembrane protein for human red blood cells (Kant and Steck, 1973), suggestingthat glycolytic enzymes are present in higher quantities in ROS compared toerythrocytes. In this regard, the rate of glycolytic flux in ROS was found to beapproximately three fold of that in human erythrocytes (section 4.4).7 2TABLE VIIIGlycolytic Enzyme Activity Ratios' in Bovine ROS and Other TissuesEnzyme ROS Erythrocyte" MuscleG3PD 1.0 1.0 1.0LDH 1.5 1.0 1.0PK 0.4 0.1 0.6PGK 0.3 1.5 0.7/1.4PFK 0.06 0.05 0.1ALD 0.04 0.02 0.1HK 0.009 0.007a ratios of each glycolytic enzyme activity to the activity of G3PD.b calculated using data from Maretski et al., 1989.C calculated using data from Scopes and Stoter, 1982.7 3Localization of glycolytic enzymes by immunofluorescence microscopy: Twomonoclonal antibodies, gpd 3E11 and pgk 1C9, were generated against human redblood cell G3PD and yeast phosphoglycerate kinase (PGK), respectively, for use asimmunochemical probes. The reactivity and specificity of these two antibodies areshown by Western blot analysis in Figure 19. Monoclonal antibody gpd 3E11labeled the G3PD protein (Mr = 38 K) in ROS (lane a) and G3PD purified fromred blood cells (lane b), but not yeast PGK (lane c). Monoclonal antibody pgk 1C9labeled the PGK protein (Mr 43 K) in ROS (lane a) and purified yeast PGK (lanec), but not purified G3PD (lane b).In order to determine whether glycolytic enzymes found in ROS preparationsare true ROS components or contaminants from other other retinal layers, thedistribution of three glycolytic enzymes, PGK, G3PD and LDH, in retinacryosections was visualized by immunofluorescence microscopy. As shown in Fig. 20(a, d, and f), monoclonal antibodies to these glycolytic enzymes labeled the bovineouter segment layer, as well as the inner segment and cell body layers. The labelingwas most intense for G3PD, the most abundant glycolytic enzyme, and least intensefor PGK. Labeling was essentially eliminated in controls in which the antibodies inthe hybridoma cell culture fluid were removed by immunoprecipitation with goatanti-mouse Ig-Sepharose (c, e and h) . The specificity of the immunochemicalreagents was also confirmed using anti-rhodopsin rho 1D4 monoclonal antibodywhich intensely labeled only the outer segment layer (g).3.3.2. Glucose transport in photoreceptor rod outer segmentsDetection of ROS glucose transport activity: Since glucose is required forglycolysis, glucose uptake activity was studied by measuring the uptake of 3-0-[14C]methylglucose, a non-metabolizable analog of glucose, by isolated bovine ROS.Fig. 21 shows the time course for the uptake of 3-0-methylglucose by bovine ROS.74CBkDa206 —97—67—43— ■111•■••111111G3PD^PGK4WD29a^b c^a b^c^a b cFig. 19. Western blots of ROS and purified G3PD and PGK labeled withmonoclonal antibodies against G3PD and PGK.Bovine ROS (lanes a), G3PD purified from human red blood cells (lanes b),and PGK purified from yeast (lanes c) were subjected to SDS gel electrophoresisand either stained with Coomassie Blue (CB) or transferred to Inunobilonmembranes and labeled with anti-G3PD monoclonal antibody gpd 3E11 (G3PD) oranti-PGK monoclonal antibody pgk 1C9 (PGK) and 1 I-labeled goat anti-mouse Igfor autoradiography. Lanes contained either 20 ug of ROS or 5 ug of purifiedenzyme.7 5Fig. 20. Localization of glycolytic enzymes in the photoreceptor layer of bovineretina by immunofluorescence microscopy.Cryosections of bovine retina were labeled with the indicated monoclonalantibodies and FITC-labeled goat anti-mouse Ig. Fluorescent micrographs ofphotoreceptors labeled with (a) pkg 1C9 hybridoma culture fluid containing anti-PGK monoclonal antibody; (c) the same culture fluid pre-adsorbed with goat anti-mouse Ig-Sepharose as a control; (d) gpd 3E11 hybridoma culture fluid containinganti-G3PD monoclonal antibody; (e) the same culture fluid pre-adsorbed with goatanti-mouse Ig-Sepharose as a control; (f) anti-LDH monoclonal antibody MAb65;(h) same antibody pre-adsorbed with goat anti-mouse Ig-Sepharose as a control.Phase contrast micrograph of photoreceptor layer showing the outer segment (os),the inner segment (is) and the cell body (cb) is shown in (b) and a fluorescentmicrograph of the outer segment labeling with anti-rhodopsin monoclonal antibodyrho 1D4 is shown in (g).376A//004 , 0 --O0^50^100^150^200Time (seconds)Fig. 21. Effect of cytochalasin B on 3-0-methylglucose uptake by bovine ROS.The uptake 3-0-[ 14C]methylglucose (45 uM) into isolated ROS wasmeasured in the absence (0) and the presence (0) of 0.05 mM cytochalasin B. Theassay was carried out in triplicate.77Uptake was inhibited by 0.05 mM cytochalasin B, a known inhibitor of glucosetransport (Bloch, 1973). The rate of 3-0-methylglucose uptake was not affected bythe state of bleaching of ROS and was not stimulated by insulin (data not shown).The efflux of 3-0-rqmethylglucose from bovine ROS was also inhibited byphloretin (LeFevre and Marshall, 1959; Whitesell and Gliemann, 1979), anotherinhibitor of the glucose transporter (Fig. 22). Approximately 60 % of loaded 3-0-methylglucose was released from ROS into 3-0-methylglucose-free medium within1 min in the absence of external phloretin. In the presence of 1 mM externalphloretin, more than 70 % of loaded 3-0-methylglucose still remained within theROS 10 mM after the ROS were diluted in 3-0-methylglucose-free buffer. Loss of3-0-methylglucose from the intracellular space of phloretin-treated ROS wasapproximately 3 % per minute. This is somewhat higher than a loss of 0 and 1 %per minute observed for phloretin-treated intact human red blood cells and ratadipocytes, respectively (Eilam and Stein, 1972; Whitesell and Gliemann, 1979), andis probably due to higher leakage from the ROS.Chacterization of ROS glucose transport activity: The kinetic parameters ofthe ROS glucose transporter and the erythrocyte GLUT-1 type glucose transportervary with the transport assay conditions (Table IX). Under net uptake (zero-transentry) condition, the ROS glucose transport exhibited Michaelis-Menten kineticswith a Km of 6.3 mM and a Vmax of 0.15 nmol/s/mg ROS membrane protein (Fig.23A). In a parallel study, human red blood cell glucose transport gave a similar Kmof 4.7 mM and a Vmax of 17.4 nmol/s/mg ghost membrane protein. The Km valuesof the ROS and red blood cell glucose transporter are within the range of Km valuesof 1.6 and 6.6 mM previously reported for GLUT-1 type glucose transporters inhuman red blood cells by Stein (1986) and Lowe and Walmsley (1985), respectively.Under the equilibrium exchange condition, ROS glucose transport had ahigher Km of 29 mM and a higher Vmax of 1.06 nmol/s/mg ROS membrane protein0•^• •78o 100a) 80cC60O01) 40200^4^8^12Time (minutes)Fig. 22. Inhibition of 3-0-methylglucose efflux from bovine ROS by phloretin.Isolated bovine ROS were loaded with 3-0-[ 14C]methylglucose for 30 min.At time zero, ROS were diluted in 3-O-methylglucose-free buffer in the absence (0)or in the presence (0) of 1 mM phloretin. The assay was carried out in triplicate.44^8S OW)1340200790^20^40S (MM)Fig. 23. Concentration dependence of 3-0-methylglucose equilibrium exchange bybovine ROS.Isolated bovine ROS were allowed to equilibrate with indicatedconcentrations of unlabeled 3-0-methylglucose for 30 min. The exchange velocityof 3-0-methylglucose was measured by determining the initial uptake rate ofexternally added tracer amounts of 3-04 14C]methylglucose into ROS. Michaelis-Menten graph shows the equilibrium exchange velocity (V) as a function of 3-0-methylglucose concentrations (S). Insert shows Hane's plot analysis of theMichaelis-Menten graph.8 0TABLE IXKinetic Properties of ROS and Erythrocyte Glucose TransporterNet Uptake^Equilibrium ExchangeKm^Vmax Km^Vmax(mM) (nmol/s/mg ROS)^(mM) (nmol/s/mg ROS)Bovine ROSa 6.3 0.15 29 1.06Bovine ROSb 10 0.90 38 3.08Human RBCci 4.7 17.4 26 144a data obtained from present study; the assays were carried out in triplicate.b data reported by Lopez-Escalera et al., 1991.81(Fig. 23B). In a parallel study, human red blood cells also showed an increased Kmof 26 mM and a higher Vmax of 144 nmol/s/mg ghost membrane protein. Similarincreases in Km and Vmax under the equilibrium exchange condition have also beenobserved for the GLUT-1 type glucose transporter in human red blood cells with Kmranging from 8.1 to 38 mM (Miller, 1968; Eilam and Stein, 1972; Edwards, 1974;Eilam, 1975; Brahm, 1983) and Vmax from 130 to 217 nmol/s/mg ghost protein(Miller, 1968; Eilam and Stein, 1972; Wheeler and Hinkle, 1981). ROS glucosetransport differs from the GLUT-2 type glucose tranport in liver which has a higherKm of 18.1 mM for net uptake and a Km of 20.2 mM for equilibrium exchange(Craik and Elliot, 1979).Lopez-Escalera et al. (1991) have also detected glucose transport activity inisolated ROS. Under the net uptake condition, ROS 3-O-methylglucose transportwas found to have a Km of 10 mM and a Vmax of 0.038 mol/s/mol rhodopsin (or 0.90nmol/s/mg ROS membrane protein assuming that rhodopsin constitutes 90 % ofthe total ROS membrane protein and has a molecular weight of 38,000 gjmol).Under the equilibrium exchange condition, 3-0-methylglucose transport alsoshowed an increased Km of 38 mM and Vmax of 0.13 mol/s/mol rhodopsin (3.08nmol/s/mg ROS membrane protein). Different values of 3-O-methylglucosetransport Km and Vmax found by Lopez-Escalera et al. (1991) may reflect differencesin conditions of transport assay. Nevertheless, both sets of data suggest that theROS glucose transporter exhibits asymmetric transport kinetics (Km and Vmax ofthe transporter is dependent on the transport condition) similar to that observed inerythrocytes.ROS plasma membrane contains a GLUT-1 type glucose transporter:Antibodies against several different types of glucose transporters were used withWestern blotting to identify and localize the glucose transporter of ROS. As shownin Figure 24 (lanes a-c), anti-GLUT-1 antibodies labeled a diffuse protein band of8 2Fig. 24. Western blots of red blood cell membranes, brain microsomes and ROSmembranes labeled for the GLUT-1 type glucose transporter.Lanes a, human red blood cell ghosts; lanes b, rat brain microsomes; lanes c,bovine ROS; lanes d, purified bovine ROS plasma membrane; and lanes e, purifiedbovine ROS disk membranes were subjected to SDS gel electrophoresis and eitherstained with Coomassie Blue (CB) or transferred to mtrocellulose and labeled withpolyclonal anti-GLUT-1 antiserum and 1 I-labeled goat anti-rabbit Ig forautoradiography. Coomassie Blue-stained gels contained 15-20 ug of membraneproteins; Western blots contained 3 ug of human red blood cell membrane proteinand 30 ug of brain or ROS membrane protein.8 345-50 kDa in human red blood cells, bovine brain microsomes, and isolated ROSmembranes. Western blot analysis of isolated plasma and disk membranes furtherindicated that the GLUT-1 transporter is present in the ROS plasma membrane,and not in disk membranes (Figure 24, lanes d and e). An anti-GLUT-2 antiserumand anti-GLUT-4 antiserum which labeled glucose transporter in rat livermicrosomes and adipocyte membranes respectively, did not label any protein in ratROS membranes as analyzed by Western blot analyses (data not shown).Quantitative comparison of the GLUT-1 type glucose transporter in ROSand red blood cell plasma membranes: The rate of 3-0-methylglucose uptake intoROS was compared with that into human and bovine red blood cells. At an external3-0-methylglucose concentration of 45 uM, the rate of uptake was 738 pmol 3-0-methylglucose/s/mg membrane protein for human red blood cells; 12 pmol 3-0-methylglucose/s/mg membrane protein for bovine red blood cells; and 46 pmol 3-0-methylglucose/s/mg plasma membrane protein for ROS. The low glucosetransport activity for bovine red blood cells compared to human red blood cells hasbeen previously reported (Hoos et al., 1972).The quantity of GLUT-1 transporter in ROS and human red blood cellghosts was also compared by solid-phase radioimmune competition assays. Asshown in Figure 25, Triton X-100 solubilized red blood cell membranes and ROSmembranes inhibited GLUT-1 antibody binding to immobilized glucose transporter.Half-maximum inhibition was obtained at a red blood cell membrane proteinconcentration of 1.2 ug/ml and at a ROS membrane protein concentration of 160ug/ml. Since the ROS plasma membrane constitutes about 5% of the total ROSmembrane, half maximum inhibition would be expected to be obtained at a 20 foldlower plasma membrane protein concentration or about 8 ug/ml. On this basis theROS plasma membrane would contain about 15% the number of glucosetransporters found in the human red blood cell plasma membrane. In agreement-0C0Ea.10Eco1 0080604020010 1^102^103^104^105Protein Concentration (ng/mI)84Figure 25. Solid-phase radioimmune competitive inhibition assay of the GLUT-1glucose transporter in ROS and in red blood cell ghosts.Anti-GLUT-1 glucose transporter antiserum was pre-incubated with serially-diluted Triton X-100 solubilized bovine ROS (A), human red blood cell ghosts (e)and bovine red blood cell ghosts (• ) for 2 h. The unbound antibody was thenquantified by solid phase radioimmune assay using human red blood cell ghosts asimmobilized antigen and 'I-labeled goat anti-rabbit Ig as secondary antibody. Theassay was carried out in triplicate.8 5with the glucose uptake studies, bovine red blood cells contain much smalleramounts of transporter than either human red blood cells or bovine ROS.Localization of the GLUT-1 type glucose transporter in bovine photoreceptorcells by immunofluorescence microscopy: Immunofluorescence microscopy of thephotoreceptor layer of bovine retinal sections labeled with anti-GLUT-1 glucosetransporter antiserum is shown in Figure 26. The ROS, as well as the rod innersegment and cell body, are labeled. Labeling pattern of the less numerous conecells can not be readily seen against the fluorescently labeled rod cells. In controlstudies, adsorption of the anti-GLUT-1 antiserum with human red blood cell ghostswhich contain large quantities of the GLUT-1 transporter abolished labeling ofphotoreceptor cells. Adsorption of the anti-GLUT-1 antiserum with bovine redblood cell membranes which contain very low amounts of transporter did notdecrease the labeling of retina sections.33.3. CONE OUTER SEGMENTS ALSO CONTAIN GLYCOLYTIC ENZYMESAND A GLUT-1 TYPE GLUCOSE TRANSPORTERThe presence of glycolytic enzymes and a GLUT-1 glucose transporter incone cells was studied in chicken retinal sections which contain predominantly conecells (Szel et al., 1986). As shown in fluorescent micrographs in figure 27, the coneouter and inner segment of the photoreceptor layer were intensely labeled with thegpd 3E11 monoclonal antibody against G3PD, pgk 1C9 monoclonal antibody againstPGK, MAb65 monoclonal antibody against LDH and anti-GLUT-1 antiserumagainst the GLUT-1 glucose transporter. Labeling was essentially abolished incontrols in which antibodies in the hybridoma cell culture fluid and the antiserumwere removed by immunoprecipitation with goat anti-mouse Ig-Sepharose andhuman red blood cell ghosts respectively (data not shown). These results indicate8 6Figure 26.^Localization of the GLUT-1 glucose transporter in bovinephotoreceptors by immunofluorescence microscopy.Cryosections of bovine retina were labeled with the anti-GLUT-1 typeglucose transporter antiserum pre-adsorbed with (a) bovine red blood cell ghostswhich contain very low levels of glucose transporter and (b) with human red bloodcell ghosts which contain high levels of glucose transporter to serve as a control.Panel (a) shows the labeling of photoreceptor outer segments (os), inner segments(is) and cell bodies (cb) by the anti-GLUT-1 type glucose transporter antiserum.87Figure 27. Localization of glycolytic enzymes and a GLUT-1 glucose transporter inthe photoreceptor cell layer of chicken retina by immunofluorescence microscopy.Cryosections of chicken retina were sequentially labeled with the indicatedmonoclonal antibody or antiserum and FITC-conjugated goat anti-mouse or goatanti-rabbit Ig. Fluorescent micrographs of photoreceptor layer labeled with (a)anti-G3PD monoclonal antibody gpd 3E11; (b) anti-PGK monoclonal antibody pgk1C9; (c) anti-LDH monoclonal antibody MAb65; and (d) anti-GLUT 1 glucosetransporter antiserum. Phase contrast micrograph of photoreceptor layer showinthe outer segment (as), the inner segment (is) and the cell body (cb) is shown in (f)and a fluorescent micrograph of outer segment labeled with anti-rhodopsin antibodyrho 3D6 which cross-reacts with cone opsin is shown in (e).88that cone outer segments, as well as rod outer segments, contain a glucosetransporter and glycolytic enzymes for anaerobic glycolysis.3.4. DISCUSSIONGlycolytic enzymes in photoreceptor outer segments: Enzyme activitymeasurements and glucose uptake studies indicate that isolated bovine ROS containglycolytic enzymes and a glucose transporter required for anaerobic glycolysis. Thepossibility that these activities result from contaminants of the rod inner segment orfrom other retinal cells is excluded on the basis of immunofluorescence labelingstudies. Monoclonal antibodies against G3PD, PGK and LDH and site-directedpolyclonal antiserum against the GLUT-1 glucose transporter clearly label the outersegment layer, as well as the inner segment and cell body layer of bovine retinalsections. Immunofluorescent labeling of cone dominant chicken retina sectionsindicate that cone outer segments also contain glycolytic enzymes and the GLUT-1glucose transporter. Although glycolytic enzyme activities in photoreceptor outersegments are low compared to other retinal layers (Lowry et al., 1956 and 1961),they are comparable to that found in human red blood cells.Glucose transport in photoreceptor outer segments: Biochemical andimmunochemical studies carried out here indicate that rod and cone outer segmentscontain a GLUT-1 type glucose transporter (review, Gould and Bell, 1990) as foundin human red blood cell membranes (Kasahara and Hinkle, 1977; Mueckler et al.,1985; Allard and Lienhard, 1985) and in brain (Birnbaum et al., 1986; Gerhart et al.,1989). The transporter in ROS has a Km which is similar to that for GLUT-1 typetransporters in other cells and is inhibited by both cytochalasin B and phloretin.8 9More importantly, a GLUT-1 specific antibody labels the 45 kDa protein in ROSwhich closely corresponds to the 45 kDa glucose transporter in human red bloodcells and brain microsomal membranes (Figure 24). Small differences in thebanding pattern of the glucose transporter from the different cell types is likely dueto differences in glycosylation of this transporter (Gorga et al., 1979; Wang, 1987).Western blotting of ROS plasma membrane and disk membrane fractions alsoindicate that the GLUT-1 transporter is only present in the plasma membrane ofROS. Thus, the glucose transporter, like the cGMP-gated channel (Cook et al.1987) and the Na+/Ca 2 +-K+ exchanger (Reid et al. 1990), must be selectively sortedto the plasma membrane and not to disk membranes during ROS morphogenesis.Two other types of glucose transporter, GLUT-2 found in liver (Fukumoto et al.,1988; Thorens et al., 1988) and GLUT-4 found in adipose cells (James et al., 1988),are not present in outer segments.An estimate of the amount of glucose transporter in ROS plasma membranecan be made by comparison with the amount reported in human red blood cellghosts. On the basis of cytochalasin B binding studies, it has been estimated that thehuman red blood cell contains approximately 250,000-500,000 glucose transporters(Lin and Snyder, 1977; Zoccoli and Lienhard, 1977; Allard and Lienhard, 1985);using a value of 350,000 per red blood cell, this translates to a density of about 2600transporters/um2 for a red blood cell surface area of 135 um 2 (Evans and Fung,1972). Solid phase competition assays indicate that ROS membranes contain about0.75% of the transporter found in red blood cell ghost. If the plasma membraneconstitutes 5% of the ROS membrane, then the amount of glucose transporter inthe ROS plasma membrane is about 15% that of red blood cell ghosts. On this basisthe density of the transporter in ROS plasma membrane is estimated to be on theorder of 400/um2 . Glucose uptake studies indicate that ROS plasma membranescontains about 6% of transport activity of human red blood cell membranes. The9 0lower value relative to that observed by the radioimmune competition assay can beattributed to the presence of unsealed ROS in the preparations used for glucoseuptake measurements.The finding that glycolytic enzymes and a glucose transporter are present inouter segments further supports the view that outer segments have the capability ofgenerating energy in the form of ATP by anaerobic glycolysis. Energy for outersegment function is also probably provided by a phosphocreatine shuttle betweenthe inner and outer segments since it has been shown that outer segments containsignificant amounts of creatine kinase (Walliman et al. 1986) and that rod outersegment preparations can efficiently transfer a phosphate from creatine phosphateto ADP and AMP (Dontsov et al., 1978; Schnetkamp and Daemen, 1981). Therelative contributions of glycolysis and the phosphocreatine shuttle system to thegeneration of ATP and GTP for outer segment function will be discussed in Chapter4.91CHAPTER 4GLUCOSE METABOLISM IN THE OUTER SEGMENTS OF RETINALPHOTORECEPTOR CELLS: ITS INVOLVEMENT IN THE MAINTENANCE OFTHE PHOTOTRANSDUCTION PROCESS4.1. MATERIALS4.1.1. Animal tissues: Fresh bovine eyes were obtained from J & L Meats (Surrey,British Columbia).4.1.2. Chemicals: D-[ 14C(U)]-glucose (11.5 mCi/mmol), Hyamine solution andAquasol-2 scintillation cocktail were purchased from New England Nuclear(Boston, MA). Hydrazine was from Eastman Organic Chemicals (Rochester, NY).Bicinchoninic acid (BCA) protein assay reagents were purchased from Pierce(Rockford, IL). All other chemicals were from Sigma (St. Louis, MO) or BritishDrug House Chemical Co. (Vancouver, British Columbia).4.2. METHODS42.1. Preparation of bovine ROSROS were prepared under dim red light from freshly dissected bovine retinasby sucrose gradient centrifugation according to a modified method of Molday et. al.,1987. Briefly, one hundred retinas were placed in 40 ml of homogenizing solutionconsisting of 20 % (w/v) sucrose, 0.25 mM MgC1 2, 10 mM taurine, and 20 mMKH2PO4, pH 7.0. After gently shaking the retinas for 1 min, the tissue suspensionwas filtered through a Teflon filter. The filtrate was layered onto six 22 ml 27 - 45% (w/v) continuous sucrose gradients containing 10 mM taurine, 0.25 mM MgC12and 20 mM KH 2PO4, pH 7.0. After centrifugation in a SW-28 rotor (BeckmanInstruments, Inc., Palo Alto, CA) at 25,000 rpm for 1 h at 4 °C, the pink ROS band9 2was collected and diluted to a final volume of 150 ml with a resuspension bufferconsisting of 20 % (w/v) sucrose and 10 mM KH 2PO4, pH 7.0. Followingcentrifugation in a Sorvall SS-34 rotor (E.I. DuPont de Nemours & Co., Inc., SorvallInstruments Div., Newtown, CT), the ROS pellets were resupended at a proteinconcentration of 14 - 16 mg/ml in a lysis buffer containing 10 mM KH 2PO4, pH 7.0.The isolated ROS membranes were stored in the dark at 4 °C and used forexperimentation within 8 h. Protein concentration of the ROS preparation wasdetermined using the Bicinchoninic acid protein assay (Pierce).4.2.2. Quantitation of glycolysis, hexose monophosphate pathway and retinalreduction in bovine ROS preparationUnbleached ROS resuspended in 10 mM KH 2PO4, pH 7.0 at a proteinconcentration of 8 mg/ml were diluted with an equal volume of assay bufferconsisting of 10 mM MgSO 4, 2 mM NaC1, 5 mM ATP, 2 mM ADP, 2 mM NAD, and90 mM KH2PO4, pH 7.0. The mixture was incubated with shaking in a 37 °C waterbath for 10 min at a distance of 1.8 m from two 34 watt fluorescent lights (lightcondition) or at at a distance of 1 m from a dim red light (dark condition) prior toassay initiation. All assays were carried out using three different ROS preparations.Glycolysis: The rate of glycolytic flux was measured by monitoring theproduction of lactate using a modified method of Clegg and Jackson (1988). Theassay was carried out at 37 °C. One ul of 0.5 M glucose in 10 mM KH2PO4, pH 7.0was placed at the bottom of a 15 x 85 mm glass test tube shaking in a 37 °C waterbath at 60 rev/min and the assay was initiated by the addition of 100 ul of ROS tothe glucose solution. At the desired time, the assay was terminated by the additionof 100 ul of 12 % of perchloric acid. Zero-time controls were carried out by addingthe perchloric acid to the ROS suspension before initiating the assay. Theprecipitated ROS suspension was transferred to an Eppendorf tube and pelleted by9 3centrifugation in an Eppendorf Centrifuge 5415C (Canlab; Mississauga, Ontario) at14,000 rpm for 3 min. One hundred-and-fifty ul of supernatant was neutralized with100 ul of 1.6 M NaOH and added to 250 ul of lactate detection reagent. Each ml oflactate detection reagent contained 25 units of lactate dehydrogenase, 4 mg of NAD,37.6 mg glycine and 17.6 ul of 95 % hydrazine, pH 9.5. The mixture was incubated at37 °C for 90 min. Absorbance at 340 nm due to the production of NADH from theoxidation of lactate to pyruvate was measured at the end of incubation. Lithiumlactate was used as standard for lactate calibration. Only background absorbance at340 nm similar to that observed for zero-time controls was detected if lactatedehydrogenase was omitted from the lactate detection reagent. This indicates thatthe observed absorbance at 340 nm was due to the presence of lactate, not NADH,produced by glycolysis during the assay period.Hexose monophosphate pathway: The rate of glucose oxidation by thehexose monophosphate pathway was determined by monitoring the production of14c-2u from D-[ 14C(U)]glucose using the method of Thomson and Richards (1977).D-[14C(U)]glucose in 90 % methanol was dried down under vacuum andresuspended with unlabelled glucose in 10 mM KH 2PO4, pH 7.0 to a final specificactivity of 4.7 x 105 dpm per umol glucose. One ul of resuspended glucose wasplaced at the bottom of a 15 x 85 mm test tube and the assay was initiated asdescribed above. Two-hundred ul of Hyamine was added to the center well(Kontes; Owens-Kimble Unit 15, Scarborough, Ontario) 20 sec before the assay wasterminated by the addition of 100 ul of 12 % perchloric acid to the ROS mixture.After 4 h incubation at 25 °C with constant shaking to allow the trapping of 14CO2 byHyamine, the center well was cut directly into 10 ml of Aquasol-2 scintillationcocktail and counted in a scintillation counter. 14c,2ki production was not detectedif ROS were omitted from the assay mixture indicating that no spontaneous9 4breakdown of 14C[U]-glucose to 14CO2 occurred during the assay period. Zero-timecontrols were carried out as described for the glycolytic flux measurements.Retinal reduction: The rate of retinal reduction was determined bymonitoring the disappearance of retinal in a ROS suspension using a modifiedmethod of Futterman and Saslaw (1961). The assay was initiated as described forglycolytic flux measurements and was terminated by the addition of 0.4 ml ofethanol. Procedures below were carried out under dim light. Retinal in the ROSmixture was assayed by incubation with 0.14 ml of thiobarbiturate and 0.14 ml ofthiourea reagents in an Eppendorf tube for 30 min at room temperature. At the endof incubation, the mixture was centrifuged at 14,000 rpm for 5 min in an EppendorfCentrifuge 5415C and the supernatant was collected for absorbance measurement at530 nm Purified all-trans-retinal was used as standard for retinal calibration.4.2.3. Glycolytic enzyme assaysThe efficiency of pig ROS hexokinase, phosphofructokinase,phosphoglycerate kinase and pyruvate kinase to use guanine and adeninenucleotides as their substrates were assessed by comparing the K., and Vmax of theseglycolytic enzymes in the presence of corresponding nucleotide substrates. Pig ROSinstead of bovine ROS were used for this study because pig ROS contain muchhigher glycolytic enzyme activities than bovine ROS (maximally 5 fold for pyruvatekinase, unpublished data). Enzyme activities were monitored at 25 °C by followingthe consumption or production of NADH at 340 nm through coupled enzymereactions as described previously (Hsu and Molday, 1991). The hexokinase assaywas carried out according to the method of Bucher et al. (1953); thephosphofructokinase assay according to the method of Harris et al. (1982); thephosphoglycerate kinase assay according to the method of Kulbe and Bojanovski(1982) except 100 mM Tris instead of 100 mM triethanolamine was used as the9 5buffer; and the pyruvate kinase assay according to the method of Kahn and Marie(1982). Changes in absorbance at 340 nm were monitored using a SLM Amincospectrophotometer. One unit of enzyme activity is defined as the production of 1umol of NADH at 25 °C/min/mg of ROS protein using 6.22 x 10 3 N4-1 cm' as theextinction coefficient for NADH.4.2.4. Assessment of ROS purityThe extent of bleaching and purity of ROS membrane preparations wereassessed by measuring the ROS A 400/A500 and A280/A500 ratios respectively.Hypotonically lysed unbleached ROS were solubilized in 1.1 % cetyltrimethylammonium bromide (CTAB) at a protein concentration of 0.5 -1.5 mg/ml in thedark. Absorbances were measured by an Ultrospec II spectrophotometer (LKB,Milwaukee, WI) in the dark. The rhodopsin concentration of prepared ROS wasdetermined by the method of Wald and Brown (1953) with few exceptions. ROSsolubilized in CTAB at a protein concentration of 0.5-1.5 mg/ml were incubatedwith a final concentration of 20 mM hydroxylamine for 2 min. Absorbance ofunbleached and bleached hydroxylamine-treated ROS solution at 500 nm wasmeasured. The rhodopsin content in ROS was calculated from the difference inA500 values of unbleached and bleached ROS solutions using a molar extinctioncoefficient of 40,600 for rhodopsin. ROS membrane preparations used in this studyhad A400/A500 ratios of 0.20-0.22, A280/A500 ratios of 2.1-2.3 and retinaldehydecontent of 16-20 nmol/mg total ROS protein.4.3. RESULTS4.3.1. Glycolytic flux in ROSAssay condition: Because isolated ROS are not as well sealed as intact cells,metabolic cofactors such as adenine and guanine nucleotides are slowly lost from9 6ROS during the isolation procedure (Robinson and Hagins, 1979). In order toinitiate glycolysis in ROS preparations, ROS have to be lysed in a sucrose-freebuffer supplemented with cofactors such as NAD, ATP and ADP required tosupport glycolysis. Glycolytic flux was measured in a potassium-rich, phosphate-buffered assay medium similar in ionic composition to the intracellular fluid ofmuscle cells (Gregersen, 1961) in an effort to mimic the intracellular environmentof ROS. The concentrations of supplied cofactors such as ATP, ADP and NAD inthe assay buffer were similar to their corresponding intracellular concentrations inROS as previously reported by Matchinsky (1968) and Robinson and Hagins (1979).Glycolytic rate: The rate of glycolysis was quantitated by measuring the rateof lactate production. Fig. 28 shows a typical time course of lactate production byisolated bovine ROS. After a four minute lag period, lactate accumulationproceeded at a steady rate of 44.0 ± 6.4 nmol/min/mg ROS protein for at leasteight minutes (three ROS preparations). This rate is higher than that of 27 nmollactate/hr/mg ROS protein (0.45 nmol lactate/min/mg ROS protein) observed byMcConnell et al. (1969) or 1.4 umo1/10 mg dry wt./hr (4.7 nmol lactate/min/mgROS protein) observed by Futterman et al. (1970), assuming that proteins constitute50 % of the total ROS dry weight (Nielsen et al., 1970). A higher lactate productionrate observed here may be due to the omission of sucrose and Triton X-100 in theassay buffer and the measurement of the initial, rather than the overall, lactateproduction rate. Triton X-100 and high concentrations of sucrose inhibit glycolyticenzyme activities and decrease the overall glycolytic rate (unpublished observation).The overall rate of lactate production may be lower than the initial lactateproduction rate due to inactivation or end product inhibition of enzymes during aone-hour assay period.The rate of lactate production was decreased by 20 % to 35.0 ± 4.9nmol/min/mg ROS protein when 0.2 mM NADP was included in the assay buffer to9750013 4000eclaE• 300.,.ZEc•—• 2000a4-,U_1a 10000^5^10^15Time (min)Fig. 28. Glycolytic flux in isolated bovine ROS.A typical time course of glycolytic lactate production by isolated bovine ROSin the presence 0 and the absence (v) of 5 mM glucose.9 8stimulate the hexose monophosphate pathway (three ROS preparations). Thisdecrease in glycolytic rate is most likely due to channeling of glucose-6-phosphatefrom the glycolytic pathway to the hexose monophosphate pathway. The rate ofglycolysis in ROS was not significantly affected by light. The lactate production ratewas only maximally 10 % higher in light than in dark. This small increase in theglycolytic rate under light condition was probably due to the stimulation of glycolysisby ATP consumption during visual excitation. Lactate production was abolishedwhen glucose was omitted from the assay buffer (Fig. 28). The presence ofiodoacetate, an inhibitor of glycolytic enzymes triose phosphate isomerase andglyceraldehyde-3-phosphate dehydrogenase, in the assay buffer also inhibitedglycolytic flux (data not shown).GTP production by glycolysis: Because guanosine phosphates are present insimilar concentrations as adenosine phosphates in ROS (Robinson and Hagins,1979), the possibility of a direct GTP production by glycolysis to regenerate cGMPhydrolyzed by the cGMP-phosphodiesterase was investigated (Table X). Glycolysishas two energy-requiring steps catalyzed by hexokinase (HK) andphosphofructokinase (PFK) and two energy-generating steps carried out byphosphoglycerate kinase (PGK) and pyruvate kinase (PK, Fig. 9). GTP was used byphosphofructokinase but not by hexokinase for glucose phosphorylation. Figure 29illustrates the dependence of phosphofructokinase activity on GTP and ATPconcentrations. Both phosphoglycerate kinase and pyruvate kinase, however, werecapable of using GDP for the production of GTP. All three glycolytic enzymesexhibited greater preference for ADP(ATP) over GDP(GTP) as their substrates.Since guanine and adenine nucleotides are present in similar concentrations inROS, it may be estimated that GTP could be produced from 36 % and 29 % of totalphosphorylation reactions catalyzed by phosphoglycerate kinase and pyruvate kinaserespectively. The specificity of phosphoglycerate kinase and pyruvate kinase for99TABLE XSpecificity of Glycolytic Enzymes for Guanine and Adenine Nucleotides'Enzyme NucleotidesubstrateKm(uM)Vmax(U/mg ROS)Enzyme Specificitykcat(")/KnP.mbkcat(Am/Km(ATp)HK ATP 150 0.023 0GTP Oe OcPFK ATP 35 0.088 0.26GTP 55 0.056PGKd ATP 110 0.524 0.57GTP 155 0.596PK ADP 350 3.76 0.41GDP 530 3.53a Average values from two pig ROS preparations.This indicates the ratio ofspecificity of enzyme using GDP or GTP as the substratespecificity of enzyme using ADP or ATP as the substrate.Hexokinase activity was not detectable.d PGK activity was assayed in a backward direction using ATP and GTP as itssubstrates.••0 6ccrnEc40E(4020 0^0.1^0.2^0.3^0.4S (mut)00.0^0.1 0 . 2S (mM)0.3 0.4100Fig. 29. Specificity of phosphofructokinase for ATP and GTP.The activity of phosphofructokinase in isolated pig ROS was determined bymeasuring the initial rate of NADH production in coupled enzyme reactions (Harriset al., 1982) initiated by the addition of of ATP or GTP. Michaelis-Menten graphshows the phosphofructokinase activity (V) as a function of ATP (•) or GTP (n.)concentrations (S). Insert shows Hane's plot analysis of the Michaelis-Mentengraph.101guanine over adenine nucleotides, as determined by the kcat/Kin ratios shown inTable X, were 0.57 and 0.41, respectively. Thus approximately 11 % and 89 % ofnucleoside triphosphates produced by glycolysis (specifically by reactions catalyzedby phosphoglycerate and pyruvate kinases) are GTP and ATP respectively. Thiswould translate to a GTP production rate of 4.8 nmol/min/mg ROS protein byglycolysis since 22 nmol glucose/min/mg ROS protein was consumed by glycolysis.4.3.2. Hexose monophosphate pathwayThe rate of the hexose monophosphate pathway (HMP) in isolated bovineROS was determined from CO 2 production by ROS. In the absence of addedNADP, HMP proceeded at a rate of 2.9 ± 0.2 nmol CO 2/min/mg ROS protein or 6nmol NADPH/min/mg ROS protein (three ROS preparations; Fig. 30). Thedetected CO 2 production was likely due to the presence of endogenous NADP inROS (Schnetkamp et al., 1979) rather than the presence of contaminating pyruvatedehydrogenase and tricarboxylic acid cycle enzymes isocitrate and a-ketoglutaratedehydrogenases from rod inner segments because the addition of unlabelledpyruvate, a substrate for the citric acid cycle, did not inhibit 14CO2 production (seeFig. 9 for tricarboxylic acid cycle). In the presence of 2 mM iodoacetate whichinhibited over 90 % of glycolysis, inclusion of 5 mM unlabelled pyruvate in the assaybuffer actually increased the 14CO2 production by 14 % (data not shown). Thisincrease in CO2 production was probably due to the consumption of NADPHformed by the HMP by lactate dehydrogenase in the presence of excess pyruvate(Cohen and Noell, 1960). Futterman (1963) had also detected CO 2 production byisolated bovine ROS in the absence of added NADP. A lower CO 2 production rateof 26 nmol/5mg dry wt./30min or 0.35 nmol/mg ROS protein/min observed byFutterman may be in part due to his measurement of an overall rather than theinitial rate of CO2 production for reasons discussed above in ••7•ZZz 0 0 og 2000cc0 vEOE100..._.,.400.7oo001 I^ I4^8Time (min)12Fig. 30. Hexose monophosphate pathway in isolated bovine ROS.The rate of glucose oxidation to CO2 via the hexose monophosphate pathwayin ROS was quantitated by measuring the rate of 14CO2 production by isolatedbovine ROS supplied with tracer amount of D-[ 14C(U)J-glucose in the presence (0)and the absence (^) of 0.2 mM NADP.103The addition of 0.2 mM NADP in the assay buffer resulted in a 6-7 foldincrease in CO2 production by HMP to 19.8 ± 1.1 nmol CO 2 produced/min/mgROS protein or about 40 nmol NADPH produced/min/mg ROS protein (threeROS preparations; Fig. 30). This increase in the HMP rate could be due to eitheran increase in glucose consumption by ROS or a rapid channeling of glucoseintermediates through both the HMP and glycolysis so that both NADPH andlactate can be produced from each molecule of glucose. However, since lactateproduction using an HMP intermediate ribose-5-phosphate at 5 mM concentrationwas only 10 % of that using 5 mM glucose as the substrate, it is likely that theobserved increase in the HMP rate was due to an increase in both the glucoseconsumption by ROS and the proportion of glucose-6-phosphate entering the HMP(Table XI). In the presence of a low concentration of NADP, approximately 24.9nmol glucose/min/mg ROS protein was phosphorylated by hexokinase, and 12 % ofphosphorylated glucose (2.9 nmol/min/mg ROS protein) was channeled to theHMP. In the presence of excess NADP, the rate of glucose phosphorylation wasincreased by 50 % to 37.5 nmol/min/mg ROS protein and 54 % of phosphorylatedglucose (20 nmol/min/mg ROS protein) was used by the HMP. The extent ofbleaching of ROS did not have a detectable effect on the rate of CO2 production bythe HMP.4.3.3. RETINAL REDUCTIONROS retinal content: To assess the potential of HMP in ROS to supportretinal reduction, the rate of reduction of endogenous retinal to retinol in ROS wasmonitored using a thiobarbiturate assay previously described by Futterman andSaslaw, 1961 (Fig. 31). According to this assay, bovine ROS preparations used inthis study contained approximately 18-19 nmol retinal/mg total ROS protein (threeROS preparations), in agreement with that of 16-20 nmol rhodopsin/mg total ROS104TABLE XIGlucose Utilization by Bovine ROSCondition^Glycolysis^HMPa^ROS glucose consumption(nmol/min/mg ROS protein)-NADP 22.0 2.9 24.9+NADP (0.2mM) 17.5 20.0 37.5a Hexose monophosphate pathway.10520.9,cn0w16Q,E0Ec%-, 121:3ccotr80^2^4^6^8^10Time (min)Fig. 31. Retinal reduction in isolated bovine ROS.The rate retinal reduction to retinol in ROS was quantitated by measuringthe disappearance of retinal in isolated bovine ROS in the presence (0) and absence(A) of 5 mM glucose.106protein quantitated using the spectrophotometric method described above in 4.2.4assuming that each rhodopsin molecule has one molecule of bound retinal.Assuming that rhodopsin has a molecular weight of 38,000 g/mol and comprises 70% of the total ROS protein, isolated bovine ROS should contain about 20 nmolrhodopsin or retinal/mg total ROS protein. Thus isolated bovine ROS used in thisstudy contained approximately 85-100 % unbleached rhodopsin.ROS retinal reduction: Figure 31 shows a typical time course of retinalreduction in bovine ROS. Very little retinal (0.09 ± 0.02 nmol retinal/min/mgROS protein) was reduced when glucose was omitted from the assay buffer or whenROS were unbleached. In dark, it is likely that 11-cis-retinal bound to rhodopsin isunavailable for reduction. In addition, retinol dehydrogenase has been shown toreduce free all-trans-retinal 10 times more rapidly than it does free 11-cis-retinal(ref. Lion et al., 1975). Under the light condition, an initial retinal reduction rate of1.2 ± 0.2 nmol retinal/min/mg ROS protein was observed when bleached ROSwere supplemented with 5 mM glucose and 0.2 mM NADP (three ROSpreparations). Nicotra and Livrea (1982) have reported a retinal reduction rate of5.8 nmol retinal reduced/min/mg ROS protein when retinal reduction was carriedout in the presence of 0.5 mM NADPH. This higher rate is likely an overestimate ofretinal reduction rate in ROS because the intracellular ratio of NAPDH to NADPin monkey and rabbit ROS is approximately 1:4 (Matschinsky, 1968) and NADP is acompetitive inhibitor of retinal reduction process (Nicotra and Livrea, 1982).Substitution of glucose by fructose-1,6-bisphosphate did not support retinalreduction, suggesting that very little or no gluconeogenesis takes place in ROS.4.4. DISCUSSIONThe existence of glycolytic and hexose monophosphate pathways in isolatedbovine ROS preparation suggests that glycolysis in ROS supplies both NADPH and107energy in the form of ATP and GTP for the maintenance of the phototransductionprocess. Energy-requiring processes in phototransduction include phosphorylationof photoactivated rhodopsin, replenishment of GTP hydrolyzed by transducin andregeneration of cGMP hydrolyzed by cGMP-phosphodiesterase (described in moredetail in Chapter 1, 1.4). The rate of cGMP regeneration, one of the most energy-consuming processes in phototransduction, has been determined in intact rabbitretina by Ames et al. (1986). In dark, cGMP hydrolysis occurs at a basalintracellular rate of 1.7 mM cGMP/min. This translates to an ATP consumption of3.4 mM/min since 2 ATP molecules are required to regenerate 1 molecule cGMPfrom 5'GMP in reactions catalyzed by guanosine monophosphate kinase, nucleosidediphosphate kinase and guanylate cyclase (Chapter 1, Fig. 5). Uponphotoexcitation, the rate of cGMP hydrolysis increases by a maximal 4.5 fold to anATP consumption of 15.3 mM/min Glycolysis in ROS produces about 44 nmollactate/min/mg ROS protein or 5.7 mM ATP/min, assuming that rhodopsinconstitutes 70 % of the total ROS protein, has a molecular weight of 38,000 g/mol,and is present at an intracellular concentration of 2.4 mM in ROS (Ames et al.,1986). This value is slightly higher compared to the human erythrocyte intracellularglycolytic flux rate of 6.76 umo1/10 min/100 mg hemoglobin (Harrison et al., 1991)or 1.7 mM ATP/min assuming that each red blood cell has a volume of 116 um 3(Evans and Fung, 1972), that there are 15 g hemoglobin/d1 blood and that there are5.1 x 106 red blood cells/ul blood (Berne and Levy, 1988). Thus, glycolysis in ROShas the capacity to maintain cGMP regeneration in dark but not in light. A directGTP production by glycolysis at a rate of 4.8 nmol/min/mg ROS protein or 0.6mM/min alone is not sufficient to support cGMP regeneration in dark. It is likelythat most cGMP is synthesized from GTP derived from ATP produced by glycolysis.In this regard, enzyme activities which rapidly transfer high-energy phosphate108groups between adenine and guanine nucleotides have been detected in ROS(Dontsov et al., 1978).NADPH produced by the HMP is required to support retinal reduction andthe glutathione redox cycle in ROS. In the absence of exogenous NADP, about 12% of the total glucose utilized by ROS is oxidized by the HMP. This ratio is thesame as that calculated from Futterman's data for bovine ROS (1963, 1970) and issimilar to the ratio of 10-11 % observed in human erythrocytes (Brin andYonemoto, 1958; Murphy, 1960). This basal level of NADPH production by theHMP (5 8 nmol NADPH/min/mg ROS protein) is more than sufficient to supportretinal reduction (1 2 nmol retinal/min/mg ROS protein) in ROS. The necessity ofROS glucose metabolism to support retinal reduction is reflected in the observationthat the substitution of glucose by pyruvate, a substrate that supports the aerobicKrebs cycle in rod inner segment but not anaerobic glycolysis in ROS, cannotmaintain the photoreceptor electrical activity at the optimal level (Winkler, 1981a).Upon stimulation by excess exogenous NADP, as much as 54 % of glucoseconsumed by ROS can be oxidized by the HMP. This large excess of NADPHproduced by the HMP may be used to protect ROS against oxidative damage via theglutathione redox cycle (Winkler et al., 1986; and Winkler, 1987).Thus, glucose metabolism in both inner and outer segments of photoreceptorcells are required for the maintenance of the phototransduction process. Glucosemetabolism in the outer segment has the capacity to maintain the basal rate ofphototransduction in the dark, to buffer against sudden changes in ROS ATPconcentration upon visual excitation and to support rhodopsin regeneration duringvisual recovery. Additional energy demand beyond the capacity of ROS glycolysisunder light condition is satisfied by the phosphocreatine shuttle channeling energyfrom inner segment mitochondria to the outer segment.109SUMMARYPhotoreceptor cells are highly specialized neurons involved in mediatingvision. Each photoreceptor cell is elongated in shape and is divided into fourcompartments: the outer segment, the inner segment, the cell body and the synapticterminus. The process of phototransduction takes place in the outer segment whilemost anaerobic and all aerobic glucose metabolism in the photoreceptor cell occurin the inner segment. This compartmentalization of the photoreceptor cell has ledto a general view that all the energy and nucleotides required by thephototransduction process are supplied by the inner segment. This belief issupported by three findings: Firstly, activities of enzymes involved in glycolysis andthe hexose monophosphate pathway in the outer segment are less than 15 % ofthose in the inner segment in photoreceptor cells of both mammals and amphibians(Lowry et al., 1956, 1961; Lolley and Hess, 1969). Secondly, mitochondrion- andbrain-type creatine kinase isozymes have been detected in both cone and rodphotoreceptor cells. The mitochondrion-type isozyme is found only in the innersegment where all the mitochondria of the photoreceptor cell are located. Thebrain-type isozyme, on the other hand, is distributed in both inner and outersegments. This differential localization of creatine kinase isozymes suggests thatthere is a phosphocreatine shuttle analogous to that found in muscle, spermatozoaand electric organ of Torpedo marmorata (Wallimann et al., 1986) channeling energyderived from mitochondrial respiration to sites of energy-requiring processes in bothinner and outer segments. In sea urchin spermatozoa, the phosphocreatine shuttlehas been shown to provide all the energy required for sperm tail motility (Tombesand Shapiro, 1985). Thirdly, inhibition of aerobic glucose metabolism in intactretina by KCN resulted in a 60 % decrease in the photoreceptor current (Winkler,1981) in the presence of physiological concentrations of glucose. This observation110suggests that anaerobic glucose metabolism in the photoreceptor outer segmentalone is not sufficient to maintain phototransduction.The results in this thesis suggest that there is glucose metabolism takingplace in the outer segments of photoreceptor cells and that some energy andnucleotides required by the phototransduction process can be supplied by glucosemetabolism in the outer segment. Activities of glycolytic enzymes, including the firstand last enzymes in the glycolytic pathway have been detected in isolatedphotoreceptor rod outer segments (ROS). Although the specific activities of theseglycolytic enzymes in ROS are much lower than that found in the photoreceptorinner segments, they are higher than that found in red blood cells. In fact, oneglycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase (G3PD), was foundto be a major protein associated with the ROS plasma membrane constituting up to11-17 % of the total ROS plasma membrane protein. The binding site for thisenzyme on the ROS plasma membrane is not known although it has been shown tobe a trypsin- and chymotrypsin-sensitive protein that associates with G3PD in anelectrostatic manner similar to that observed between G3PD and an anion channelBand 3 in human erythrocytes (Yu and Steck, 1975). Binding of G3PD to Band 3has been shown to inhibit G3PD activity (Tsai et al., 1982) and dissociation of G3PDdue to tyrosine phosphorylation of Band 3 was found to be accompanied by aconcomitant elevation in the erythrocyte glycolytic rate (Harrison et al., 1991). It ispossible that such enzyme release/activation mechanism could also play a role inregulating the glycolytic rate in the photoreceptor outer segment.Immunohistochemical localization of glycolytic enzymes and detection of aGLUT-1 type glucose transporter in photoreceptor outer segments suggest that themeasured glycolytic enzyme activities were not due to contaminating enzymes from111inner segments or other retinal cells. Three glycolytic enzymes, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase and lactate dehydrogenase havebeen localized in the outer segment layer of frozen retina sections byimmunofluorescence microscopy. The presence of a glucose transporter on theplasma membrane of photoreceptor outer segments was investigated by the glucosetransport assay, Western blot analyses and immunofluorescence microscopy. Bothtransport kinetics and Western blot analyses indicated that the outer segmentglucose transporter is of the GLUT-1 type similar to that found in erythrocytes andbrain cells. No GLUT-2 (liver-type) or GLUT-4 (adipocyte-type) glucosetransporters were detected in photoreceptor outer segments by Western blottinganalyses. Concurrently, Lopez-Escalera et al. (1991) have also detected glucosetransport activity in rod outer segments exhibiting similar transport kinetics. It ispossible that the photoreceptor outer segment also contains a GLUT-3 or a noveltype glucose transporter in addition to the GLUT-1 isoform. Nevertheless, thepresence of at least a GLUT-1 type glucose transporter in the photoreceptor outersegment suggests that this organelle can potentially obtain glucose from the bloodsupply to maintain its glucose metabolism.The existence of a GLUT-1 type glucose transporter as well as enzymesinvolved in glucose metabolism in the photoreceptor outer segment raises questionsof whether glucose metabolic pathways are active in the outer segment and how theymay contribute to the maintenance of the phototransduction process. Threehypotheses have been put forward to define the function of glucose metabolism inROS. First, ROS glucose metabolism may be required to produce NADPH for apartial maintenance of the retinal reduction process during the visual recoveryphase of phototransduction (Futterman et al., 1970). Second, glycolysis in ROS mayspecialize in phosphosphorylating GDP to form GTP via pyruvate kinase to112regenerate cGMP hydrolyzed by PDE (Lopez-Escalera et al., 1991). Third, glucosemetabolism in ROS may be involved in supplying energy to maintain housekeepingbut not a phototransduction function in ROS (McConnell et al., 1969). Presentstudies showed that glucose metabolism in ROS produces both the ATP andNADPH required by phototransduction. Glycolysis in ROS has the capacity tosupport cGMP regeneration, one of the most energy-consuming processes inphototransduction, under dark but not light conditions. Although ROS glycolysiscan produce both ATP and GTP, only ATP is produced in sufficient quantity tomaintain cGMP regeneration in the dark. GTP required for cGMP synthesis cantherefore be produced from glycolysis-derived ATP since enzyme activities whichrapidly transfer high-energy phosphate groups between adenine and guaninenucleotides have been detected in ROS (Dontsov et al., 1978). The hexosemonophosphate pathway in ROS produces sufficient NADPH to support retinalreduction during visual recovery. Excess NADPH produced by the hexosemonophosphate pathway may be used to maintain the glutathione redox cycle toprotect ROS from oxidative damage.Thus glucose metabolism in both inner and outer segments of photoreceptorcells participate in the maintenance of phototransduction (Fig. 32). Glucosemetabolism in the outer segment supplies NADPH and some energy in the form ofATP required by the phototransduction process. The hexose monophosphatepathway in the outer segment provides all the NADPH required for retinalreduction during visual recovery. Glycolysis, on the other hand, supplies energy inthe form of ATP to maintain the basal level of cGMP regeneration in the dark andto buffer against sudden changes in ATP concentration upon illumination. Duringvisual excitation, additional energy required for processes such as rhodopsinphosphorylation, transducin activation and a higher rate of cGMP regeneration can113be channeled from the inner segment to the outer segment by the phosphocreatineshuttle.ADPPCr ISPHOTO-TRANSDUCTION OS114Fig. 32. Maintenance of phototransduction by glucose metabolism in photoreceptorinner and outer segments.In the photoreceptor outer segment (OS), glucose metabolism is supportedby glucose obtained from the blood supply through a GLUT-1 type glucosetransporter (GT) located on the outer segment plasma membrane. The hexosemonophosphate pathway (..) in this organelle provides NADPH required for retinalreduction to retinol (for rhodopsin regeneration) during visual recovery. Theglycolytic pathway in this organelle, on the other hand, supplies energy in theform of ATP to maintain phototransduction in the dark and to buffer against suddenchanges in ATP concentration upon light illumination. One glycolytic enzyme,glyceraldehyde-3-phosphate dehydrogenase (G3PD) associates reversibly with anouter segment plasma membrane specific protein (BP). The function of thisassociation is not known.Additional energy required by the phototransduction process during visualexcitation is provided by aerobic glucose metabolism in the photoreceptor innersegment (IS) via a phosphocreatine shuttle. ATP produced by aerobic glucosemetabolism in the mitochondrion (M) is transphosphorylated to creatine (Cr) by amitochondrion-type creatine kinase (CKm) to produce phosphocreatine (PCr).Phosphocreatine diffuses to the outer segment where it is used to phosphorylateADP by a brain-type creatine kinase (CKb) to produce ATP required to supportphototransduction during visual excitation. Creatine diffuses back to the innersegment and the shuttle repeats again.115REFERENCESAllard, W.J., and Lienhard, G.E. (1985) Monoclonal Antibodies to the GlucoseTransporter from Human Erythrocytes. J. Biol. Chem. 260, 8668-8675.Ames, A. III, Walseth, T.F., Heyman, R.A., Barad, M., Graeff, R.M. and Goldberg,N.D. (1986) Light-induced Increases in cGMP Metabolic Flux Correspond withElectrical Responses of Photoreceptors. J. Biol. Chem. 261, 13034-13042.Arcari, P., Martinelli, R., and Salvatore, F. (1984) The Complete Sequence of a FullLength cDNA for Human Liver Glyceraldehyde-3-phosphate Dehydrogenase:Evidence for Multiple mRNA Species. Nucleic Acids Res. 12, 9179-9189.Arshaysky, V.Y. and Bownds, M.D. (1992) Regulation of Deactivation ofPhotoreceptor G Protein by Its Target Enzyme and cGMP. Nature 357, 416-417.Baldwin, S.A. and Lienhard, G.E. (1981) Glucose Transport Across PlasmaMembranes: Facilitated Diffusion Systems. TIBS 6, 208-211.Ballas, S.K., Kliman, H.J., and Smith, E.D. (1985) Glyceraldehyde-3-phosphateDehydrogenase of Rat Erythrocytes Has No Membrane Component. Biochim.Biophys. Acta 831, 142-149.Bascom, R.A., Manara, S., Collins, L., Molday, R.S., Kalnins, V.I., and McInnes,R.R. (1992) Cloning of the cDNA for a Novel Photoreceptor Membrane Proteinfrom-1) Identifies a Disk Rim Protein Family Implicated in Human Retinopathies.Neuron 8, 1171-1184Baylor, D.A., Lamb, T.D. and Yau, K.-W. (1979a) The Membrane Current of SingleRod Outer Segments. J. Physiol. 288, 589-611.Baylor, D.A., Lamb, T.D. and Yau, K.-W. (1979b) Responses of Retinal Rods toSingle Photons. J. Physiol. 288, 613-634.Bell, G.I., Kayano, T., Buse, J.B., Burant, C.F., Takeda, J., Lin, D., Fukumoto, H.and Seino, S. (1990) Molecular Biology of Glucose Transporters. Diabetes Care 13,198-208.Bennett, N., Michel-Villaz, M., Kuhn, H. (1982) Light-induced Interaction BetweenRhodopsin and the GTP-Binding Protein. Eur. J. Biochem. 127, 97-103.Berger, S.J., DeVries, G.W., Carter, J.G., Schulz, D.W., Passonneau, P.N., Lowry,0.H., and Ferrendelli, J.A. (1980) The Distribution of the Components of the CyclicGMP Cycle in Retina. J. Biol. Chem. 255, 3128-3133.Berne, R.M., and Levy, M.N. (1988) Physiology, 2nd ed., p. 361, The C.R. MosbyCompany, St. Louis.Bessman, S.P. and Geiger, P.J. (1981) Transport of Energy in Muscle: ThePhosphorylcreatine Shuttle. Science 211, 448-452.116Birnbaum, M.J., Haspel, H.C., and Rosen, O.M. (1986) Cloning andCharacterization of a cDNA Encoding the Rat Brain Glucose-transporter Protein.Proc. NatL Acad. Sci. U.S.A. 83, 5784-5788.Bloch, R. (1973) Inhibition of Glucose Transport in the Human Erythrocyte byCytochalasin B. Biochemistry 12, 4799-4801.Bok, D. (1985) Retinal Photoreceptor-Pigment Epithelium Interactions. Invest.OphthalmoL Vis. Res. 26, 1659-1694.Bok, D. and Heller, J. (1976) Transport of Retinol from the Blood to the Retina:Autoradiographic Study of the Pigment Epithelial Cell Surface Receptor for PlasmaRetinol-binding Protein. Exp. Eye Res. 22, 395-402.Bownds, D., Dawes, J., Miller, J., and Stahlman, M. (1972) Phosphorylation of FrogPhotoreceptor Membranes Induced by Light. Nature New BioL 237, 125-127.Brahm, J. (1983) Kinetics of Glucose Transport in Human Erythrocytes. J. Physiol.339, 339-354.Brin, M., and Yonemoto, R.H. (1958) Stimulation of the Glucose Oxidative Pathwayin Human Erythrocyte by Methylene Blue. J. BioL Chem. 230, 307-317.Brown, K.T. (1968) The Electroretinogram: Its Components and Their Origins.Vision 8, 633-677.Biicher et aL (1953) Z. Natwforsch from protocol supplied by ICN BiomedicalsCanada Ltd., St. Laurent, Quebec.Caswell, A.H., and Corbett, A.M. (1985) Interaction of Glyceraldehyde-3-phosphateDehydrogenase with Isolated Microsomal Subfractions of Skeletal Muscle. J. BioLChem. 260, 6892-6898.Charbonneau, H., Prusti, R.K., LeTrong, H., Sonnenburg, W.K., Mullaney, P.J.,Walsh, K.A., and Beavo, J.A. (1990) Identification of a Noncatalytic cGMP-bindingDomain Conserved in Both the cGMP-stimulated and Photoreceptor CyclicNucleotide Phosphodiesterases. Proc. Nati Acad. Sci. U.S.A. 87, 288-292.Cervetto, L., Lagnado, L., Perry, R.J., Robinson, D.W. and McNaughton, P.A.(1989) Extrusion of Calcium from Rod Outer Segments Is Driven by Both Sodiumand Potassium Gradients. Nature (London) 337, 740-743.Clegg, J.S. and Jackson, S.A. (1988) Glycolysis in Permeabilized Cells. Biochem. J.255, 335-344.Cohen, L.H. and Noell, W.K. (1960) Glucose Catabolism of Rabbit Retina Beforeand After Development of Visual Function. J. Neurochem. 5, 253-276.Cohen, L.H. and Noe11, W.K. (1965) Relationships between Visual Function andMetabolism. In Biochemistry of the Retina. C. Graymore (ed.) Academic Press,London, 36-50.117Connell, G.J. and Molday, R.S. (1990) Molecular Cloning, Primary Structure, andOrientation of the Vertebrate Photoreceptor Cell Protein Peripherin in the RodOuter Segment Disk Membrane. Biochemistry 29, 4691-4698.Cook, NJ., Hanke, W. and Kaupp, U.B. (1987) Identification, Purification, andFunctional Reconstitution of the Cyclic GMP-dependent Channel from RodPhotoreceptors. Proc. Natl. Acad. Sci. U.S.A. 84, 584-589.Cook, N.J. and Kaupp, U.B. (1988) Solubilization, Purification, and Reconstitutionof the Sodium-Calcium Exchanger from Bovine Retinal Rod Outer Segments. J.Biol Chem. 263, 11382-11388.Cook, NJ., Molday, L.L., Reid, D., Kaupp., U.B. and Molday, R.S. (1989) ThecGMP-gated Channel of Bovine Rod Photoreceptors Is Localized Exclusively in thePlasma Membrane. J. Biol. Chem. 264, 6996-6999.Craik, J.D. and Elliott, K.R.F. (1979) Kinetics of 3-O-Methyl-D-Glucose Transportin Isolated Rat Hepatocytes. Biochem. J. 182, 503-508.Crane, K. and Ball, E.G. (1951) Relationship of 14CO2 Fixation to Carbohydratemetabolism in Retina. J. Biol. Chem. 189, 269.Czok, R., and Biicher, Th. (1960) Crystallized Enzymes from the Myogen of RabbitSkeletal Muscle. Adv. Protein Chem. 15, 315-415.Dagher, S.M., and Deal., W.C., Jr. (1977) Pig Liver Glyceraldehyde-3-PDehydrogenase: Purification, Crystallization, and Characterization. Arch. Biochem.Biophys. 179, 643-653.Deigner, P.S., Law, W.C., Canada, F.J. and Randon, R.R. (1989) Membranes as theEnergy Source in the endergonic Transformation of Vitamin A to 11-cis-retinol.Science 244, 968-971.Dizhoor, A.M., Ray, S., Kumar, S., Niemi, G., Spencer, M., Brolley, D., Walsh, K.A.,Philipov, P.P., Hurley, J.B., and Stryer, L. (1991) Recoverin: A Calcium SensitiveActivator of Retinal Rod Guanylate Cyclase. Science 251, 915-918.Dontsov, A.E., Zak, P.P. and Ostrovskii, M.A. (1978) Regeneration of ATP in OuterSegments of Frog Photoreceptors. Biochemistry U.S.S.R 43, 471-474.Easterby, J.S. and Qadri, S.S. (1982) Hexokinase Type II from Rat Skeletal Muscle.Methods EnzymoL 90, 11-15.Edwards, P.A.W. (1974) A Test for Non-specific Diffusion Steps in TransportAcross Cell Membranes, and Its Application to Red Cell Glucose Transport.Biochim. Biophys. Acta 345, 373-386.Eilam, Y. (1975) Two-carrier Models for Mediated Transport. II. Glucose andGalactose Equilibrium Exchange Experiments in Human Erythrocytes as a Test forSeveral Two-Carrier Models. Biochim. Biophys. Acta 401, 364-369.Eilam, Y. and Stein, W.D. (1972) A simple Resolution of the Kinetic Anomaly inthe Exchange of Different Sugars Across the Membrane of the Human Red BloodCell. Biochim. Biophys. Acta 266, 161-173.118Emeis, D., Kuhn, H., Reichert, J., Hofmann, K.P. (1982) Complex FormationBetween Metarhodopsin II and GTP-binding Protein in Bovine PhotoreceptorMembranes Leads to a Shift of the Photoproduct Equilibrium. FEBS Lett. 143, 29-34.Erickson, M.A., Robinson, P., and Lisman, J. (1992) Deactivation of visualTransduction Without Guanosine Triphosphate Hydrolysis by G Protein. Science257, 1255-1258.Evans, E. and Fung, Y.-C. (1972) Improved Measurements of the ErythrocyteGeometry. Microvascular Res. 4, 335-347.Fairbanks, G., Steck, T.L., and Wallach, D.F.H. (1971) Electrophoretic Analysis ofthe Major Polypeptides of the Human Erythrocyte Membranes. Biochemistry 10,2606-2617.Fesenko, E.E., Kolesnikov, S.S. and Lyubarsky, A.L. (1985) Induction by CyclicGMP of Cationic Conductance in Plasma Membrane of Retinal Rod OuterSegment. Nature 313, 310-313.Fort, P., Marty, L., Piechaczyk, M., El Sabrouty, S., Dani, C., Jeanteur, P., andBlanchard, J.M. (1985) Various Rat Adult Tissues Express Only One Major mRNASpecies from the Glyceraldehyde-3-phosphate Dehydrogenase Multigenic Family.Nucleic Acids Res. 13, 1431-1442.Fowles, C., Akhtar, M. and Cohen, P. (1989) Interplay of Phosphorylation andDephosphorylation in Vision: Protein Phosphatases of Bovine Rod Outer Segments.Biochemistry 28, 9385-9391.Fukumoto, H., Seino, S., Imura, H., Seino, Y., Eddy, R.L., Fukushima, Y., Byers,M.G., Shows, T.B., and Bell, G.I. (1988) Sequence, Tissue Distribution, andChromosomal Localization of mRNA Encoding a Human Glucose Transporter-likeProtein. Proc. Natl. Acad. Sci. U.S.A. 85, 5434-5438.Fung, B.K.-K. (1983) Characterization of Transducin from Bovine Retinal RodOuter Segments. Separation and Reconstitution of the Subunits. J. Biol. Chem. 258,10495-10502.Fung, B.K. and Griswold-Prenner, I. (1989) G Protein-Effector Coupling: Binding ofRod Phosphodiesterase Inhibitory Subunit to Transducin. Biochemistry 28, 3133-3137.Fung, B.K.-K., Hurley, J.B., Stryer, L. (1981) Flow of Information in the Light-triggered Cyclic Nucleotide Cascade of Vision. Proc. Natl. Acad. Sci. USA 78, 152-156.Fung, B.K.-K. and Stryer, L. (1980) Photolyzed Rhodopsin Catalyzes the Exchangeof GTP for GDP in Retinal Rod outer Segment Membranes. Proc. Natl. Acad. Sci.U.S.A. 77, 2500-2504.Futterman, S. (1963) Metabolism of the Retina: III. The Role of ReducedTriphosphopyridine Nucleotide in the Visual Cycle. J. Biol. Chem. 238, 1145-1150.119Futterman, S. and Kinoshita, J. (1959) Metabolism of the Retina. I. Respiration ofCattle Retina. J. Biol. Chem. 234, 723-726.Futterman, S., Hendrickson, A., Bishop, P.E., Rollins, M.H., and Vacano, E. (1970)Metabolism of Glucose and Reduction of Retinaldehyde in Retinal Photoreceptors.J. Neurochem. 17, 149-156.Futterman, S., and Saslaw, L.D. (1961) The Estimation of Vitamin A Aldehyde withThiobarbituric Acid. J. Biol. Chem. 236, 1652-1657.Gerhart, D.Z., LeVasseur, R.J., Broderius, M.A., and Drewes, L.R. (1989) GlucoseTransporter Localization in Brain Using Light and Electron Immunocytochemistry.J. Neuroscience Res. 22, 464-472.Gorga, F.R., Baldwin, S.A., and Lienhard, G.E. (1979) The MonosaccharideTransporter from Human Erythrocytes is Heterogeneously Glycosylated. Biochem.Biophys. Res. Commun. 91, 955-961.Gould, G.W. and Bell, G.I. (1990) Facilitative Glucose Transporters: an ExpandingFamily TIBS 15, 18-23.Graymore, C. (1959) Metabolism of the Developing Retina. I. Aerobic andAnaerobic Glycolysis in the Developing Rat Retina. Br. J. Opthalmol. 43, 34-39.Graymore, C. (1970) Biochemistry of the Retina in Biochemistry of the Eye. C.Graymore (ed.) Academic Press, London, 645-735.Graymore, C.N. and Towlson, M.J. (1965) The Metabolism of the Retina of theNormal and Alloxan Diabetic Rat. Vis. Res. 5, 379-389.Gregersen, M.I. (1961) Total Body Water and Fluid Compartment in MedicalPhysiology (Philip Bard, ed), 11th ed., p. 307, The C.V. Mosby Co., St. Louis.Hagins, W.A., Pen, R.D. and Yoshikami, S. (1970) Dark Current and Photocurrentin Retinal Rods. Biophys. J. 10, 380-412.Harrington, W.F., and Karr, G.M. (1965) Subunit Structure of Glyceraldehyde-3-phosphate dehydrogenase. J. MoL BioL 13, 885-893.Harris, J.I., and Perham, R.N. (1965) Glyceraldehyde-3-Phosphate Dehydrogenases:I. The Protein Chains in Glyceraldehyde-3-phosphate Dehydrogenase from PigMuscle. J. Mol. Biol. 13, 876-884.Harris, B.G., Starling, J.A., and Hofer, H.W. (1982) Phosphofructokinase fromAscaris suum Muscle. Methods Enzymol. 90, 45-49.Harrison, M.L., Rathnavelu, P., Arese, P., Geahlen, R.L. and Low, P.S. (1991) Roleof Band 3 Tyrosine Phosphorylation in the Regulation of Erythrocyte Glycolysis. J.BioL Chem. 266, 4106-4111.HediAer, M.A., Coady, M.J., Ideda, T.S. and Wright, E.M. (1987) ExpressionCloning and cDNA Sequencing of the Na + -glucose Co-transporter. Nature 330, 379-381.120Heinz, F., and Kulbe, K.D. (1970) Glyceraldehydephosphate Dehydrogenase fromLiver, I. Isolation and Characterization of the Bovine Liver Enzyme. Hoppe-Seyler'sZ. Physiol. Chem. 351, 249-262.Hicks, D. and Molday, R.S. (1986) Differential Immunogold-Dextran Labeling ofBovine and Frog Rod and Cone Cells Using Monoclonal Antibodies Against BovineRhodopsin. Exp. Eye Res. 42, 55-71.Hodges, R.S., Heaton, R.J., Parker, J.M.R., Molday, L., and Molday, R.S. (1988) J.Biol. Chem. 263, 11768-11775.Hodgkin, A.L., McNaughton, P.A. and Nunn, B.J. (1985) The Ionic Selectivity andCalcium Dependence of the Light-Sensitive Pathway in Toad Rods. J. Physiol.(London) 358, 447-468.Holmgren, F. (1865-6) Method Att Objectivera Effecten av Ljusintryck pa Retina.Upsala Lakaref 1, 177-191.Hoos, R.T., Tarpley, H.L., and Regen, D.M. (1972) Sugar Transport in BeefErythrocytes. Biochim. Biophys. Acta 266, 174-181.Hsu, S.-C. and Molday, R.S. (1990) Glyceraldehyde-3-phosphate Dehydrogenase Isa Major Portein Associated with the Plasma Membrane of Retinal PhotoreceptorOuter Segments. J. Biol. Chem. 265, 13308-13313.Hsu, S.-C. and Molday, R.S. (1991) Glycolytic Enzymes and a GLUT-1 GlucoseTransporter in the Outer Segments of Rod and Cone Photoreceptor Cells. J. Biol.Chem. 266, 21745-21752.Hsu, Y.-T. and Molday, R.S. (1993) Modulation of the CGMP-gated Channel ofRod Photoreceptor Cells by Calmodulin Nature 361, 76-79.Huitorel, P., and Pantaloni, P. (1985) Bundling of Microtubules by Glyceraldehyde-3-phosphate Dehydrogenase and Its Modulation by ATP. Eur. J. Biochem. 150, 265-269.Ishiguro, 5.-I., Suzuki, Y., Tamai, M. and Mizuno, K. (1991) Purification of RetinolDehydrogenase from Bovine Retinal Rod Outer Segments. J. Biol. Chem. 266,15520-15524.James, D.E., Brown, R., Navarro, J., Pilch, P.F. (1988) Insulin-regulatable TissuesExpress a Unique Insulin-Sensitive Glucose Transport Protein. Nature 333, 183-185.Johnson, L.V. and Blanks, J.C. (1984) Application of Acrylamide as an EmbeddingMedium in Studies of Lectin and Antibody Binding in the Vertebrate Retina.Current Eye Res. 3, 969-973.Kamps, K.M.P., DeGrip, W.J., and Daemen, F.J.M. (1982) Use of a DensityModification Technique for Isolation of the Plasma Membrane of Rod OuterSegments. Biochim. Biophys. Acta 687, 296-302.Kahn, A., and Marie, J. (1982) Pyruvate Kinases from Human Erythrocytes andLiver. Methods Enzymol. 90, 131-140.121Kant, J.A., and Steck, T.L. (1973) Specificity in the Association of Glyceraldehyde-3-phosphate Dehydrogenase with Isolated Human Erythrocyte Membranes. J. Biol.Chem. 248, 8457-8464.Kasahara, M. and Hinkle, P.C. (1977) Reconstitution and Purification of the D-Glucose Transporter from Human Erythrocytes. J. Biol. Chem. 252, 7384-7390.Kasanicki, M.A. and Pilch, P.F. (1990) Regulation of Glucose-Transporter Function.Diabetes Care 13, 219-227.Kaupp, U.B. and Koch, K.-W. (1986) Mechanism of Photoreception in VertebrateVision. TIBS 11, 43-47.Kawamoto, R.M., and Caswell, A.H. (1986) Autophosphorylation ofGlyceraldehydephosphate Dehydrogenase and Phosphorylation fo Protein fromSkeletal Muscle Microsomes. Biochemistry 25, 656-661.Keokitichai, S., and Wrigglesworth, J.M. (1980) Association of Glyceraldehyde-3-phosphate Dehydrogenase with the Membrane of the Intact Human Erythrocyte.Biochem. J. 187, 837-841.Kliman, H.J. and Steck, T.L. (1980) Association of Glyceraldehyde-3-phosphateDehydrogenase with the Human Red Cell Membrane. J. Biol. Chem. 255, 6314-6321.Koch, K.-W. and Kaupp, U.B. (1985) Cyclic GMP Directly Regulates a CationConductance in Membranes of Bovine rods by a Cooperative Mechanism. J. Biol.Chem. 260, 6788-6800.Koch, K.-W. and Stryer, L. (1988) Highly Cooperative Feedback Control of RetinalRod Guanylate Cyclase by Calcium Ions. Nature 334, 64-66.Kokame, K., Fukada, Y., Yoshizawa, T., Takao, T., and Shimonishi, Y. (1992) LipidModification at the N Terminus of Photoreceptor G-protein a-Subunit. Nature 359,749-752.Korenbrot, J.I. and Miller , D.L. (1989) Cytoplasmic Free Calcium Concentration inDark-adapted Reinal Rod Outer Segments. Vis. Res. 29, 939-948.Krebs, E.G., Rafter, G.W., and Junge, J.M. (1953) Yeast Glyceraldehyde-3-phosphate Dehydrogenase: Yeast Protein II. J. Biol. Chem. 200, 479-492.Kuhn, H. (1980) Light- and GTP-regulated Interaction of GTPase and OtherProteins With Bovine Photoreceptor Membranes. Nature 283, 587-589.Kuhn, H., Bennett, N., Michel-Villaz, M., Chabre, M. (1981) Interactions BetweenPhotoexcited Rhodopsin and GTP-binding Protein: Kinetics and StoichiometricAnalyses from Light-Scattering Changes. Proc. Natl. Acad. Sci. USA 78, 6873-6877.Kuhn, H. and Dreyer, W.J. (1972) Light Dependent Phosphorylation of Rhodopsinby ATP. FEBS Lett. 20, 1-6.122Kiihn, H., Hall, S.W. and Wilden, U. (1984) Light-induced Binding of 48-kDaProtein to Photoreceptor Membranes Is Highly Enhanced by Phosphorylation ofRhodopsin. FEBS Lett. 176, 473-478.Kulbe, K.D., and Bojanovski, M. (1982) 3-Phosphoglycerate Kinase from BovineLiver and Yeast. Methods Enzymol. 90, 115-120.Kulbe, K.D., Foellmer, H., and Fuchs, J. (1982) Simultaneous Purification ofGlyceraldehyde-3-phosphate Dehydrogenase, 3-Phosphoglycerate Kinase, andPhosphoglycerate Mutase from Pig Liver and Muscle. Methods Enzymol. 90, 498-511.Kulbe, K.D., Jackson, K.W., and Tang, J. (1975) Structural Evidence for a Liver-specific Glyceraldehyde-3-phosphate Dehydrogenase. Biochim. Biophys. Res.Commun. 67, 35-42.Kuwabara, T. and Cogan, D.G. (1961) Retinal Glycogen. Am. J. Ophthalmol. 66,680-688.Kuwata, 0 , Imamoto, Y., Okano, T., Kokame, K. Kojima, D., Matsumoto, H.,Morodome, A., Fukada, Y., Shichida, Y., Yasuda, K., Shimura, Y., and Yoshizawa,T. (1990) The Primary Structure of Iodopsin, a Chicken Red-sensitive ConePigment. FEBS Lett. 272, 128-132.Laemmli, U.K. (1970) Cleavage of Structural Proteins During the Assembly of theHead of Bacteriophage T4. Nature 227, 680-685.Lamb, T.D., McNaughton, P.A. and Yau, K.-W. (1981) Spatial Spread of Activationand Background Desensitization in Toad Rod Outer Segments. J. Physiol. 319, 463-496.Lambrecht, H.-G. and Koch, K.-W. (1991) A 26 kd Calcium Binding Protein fromBovine Rod Outer Segments as Modulator of Photoreceptor Guanylate Cyclase.EMBO 10, 793-798.Lee, C.-Y., Yuan, J.H., and Goldberg, E.58.Lane, R.D., Crissman, R.S. and Ginn, S.for Producing Monoclonal AntibodiesEnzymol. 121, 183-192.from Mouse. Methods Enzymol. 89, 351-3(1986) High Efficiency Fusion ProcedureAgainst Weak Immunogens. Methods(1982) Lactate Dehydrogenase IsozymesLeFevre, P.G. and Marshall, J.K. (1959) The attachment of Phloretin andAnalogues to Human Erythrocytes in Connection with Inhibition of SugarTransport. J. Biol. Chem. 234, 3022-3026.Liebman, P.A. and Pugh, E.N.Jr. (1980) Phosphodiesterase Activation in VisualReceptor Membranes. Nature 287, 734-736.Lin, S. and Snyder, C.E.Jr. (1977) High Affinity Cytochalasin B Binding to Red CellMembrane Proteins Which Are Unrelated to Sugar Transport. J. Biol. Chem. 252,5464-5471.123Lion, F., Rotmans, J.P., Daemen, F.J.M. and Bonting, S.L. (1975) BiochemicalAspects of the Visual Process: XXVII. Stereospecificity of Ocular RetinolDehydrogenases and the Visual Cycle. Biochim. Biophys. Acta 384, 283-292.Liou, G.I., Bridges, C.C.B., Fong, S.-L., Alvarez, R.A. and Gonzalez-Fernandez, F.(1982) Vitamin A Transport between Retina and Pigment Epithelium - AnInterstitial Protein Carrying Endogenous Retinol (Interstitial Retinol-bindingProtein). Vision Res. 22, 1457-1467.Lolley, R.N. and Hess, H.H. (1969) The Retinal Rod Outer Segment of the Frog:Detachment, Isolation, Phosphorus Fraction and Enzyme Activity. J. CellularPhysiol. 73, 9-24.Lolley, R.N. and Lee, R.H. (1990) Cyclic GMP and Photoreceptor Function.FASEB J. 4, 3001-3008.Lopez-Escalera, R., Li, X.-B., Szerencsei, R.T., and Schnetkamp, P.P.M. (1991)Glycolysis and Glucose Uptake in Intact Outer Segments Isolated from BovineRetinal Rods. Biochemistry 30, 8970-8976.Lowe, A.G. and Walmsley, A.R. (1985) A Quenched-flow Technique for theMeasurement of Glucose Influx into Human Red Blood Cells. Anal. Biochem. 144,385-389.Lowry, O.H., Roberts, N.R., and Lewis, C. (1956) The Quantitative Histochemistryof the Retina. J. Biol. Chem. 220, 879-892.Lowry, O.H., Roberts, N.R., Schulz, D.W., Clow, J.E., and Clark, J.R. (1961)Quantitative Histochemistry of Retina: II. Enzymes of Glucose Metabolism. J. Biol.Chem. 236, 2813-2820.MacKenzie, D., and Molday, R.S. (1982) Organization of Rhodopsin and a HighMolecular Weight Glycoprotein in Rod Photoreceptor Disc Membranes UsingMonoclonal Antibodies. J. Biol. Chem. 257, 7100-7105.March, S.C., Parikh, I., and Cuatrecasas, P. (1974) A simplified Method forCyanogen Bromide Activation of Agarose for Affinity Chromatography. Anal.Biochem. 60, 149-152,Maretzki, D., Reimann, B., and Rapoport, S.M. (1989) A Reappraisal of the Bindingof Cytosolic Enzymes to Erythrocyte Membranes. TIBS 14, 93-96.Matchinsky, F.M. (1968) Quantitative Histochemistry of Nicotinamide AdenineNucleotides in Retina of Monkey and Rabbit. J. Neurochem. 15, 663-657.Matesic, D., and Liebman, P.A. (1987) CGMP-Dependent Cation Channel ofRetinal Rod Outer Segments. Nature 326, 600-603.McConnell, D.G. (1965) The Isolation of Retinal Outer Segment Fragments. J. CellBiol. 27, 459.McConnell, D.G., Ozga, G.W. adn Solze, D.A. (1969) Evidence for Glycolysis inBovine Retinal Microsomes and Photoreceptor Outer Segments. Biochim. Biophys.Acta 184, 11-28.124McNaughton, P.A., Cervetto, L. and Nunn, B.J. (1986) Measurement of theIntracellular Free Calcium Concentration in Salamander Rods. Nature 322, 261-263.Miller, D.M. (1968) The Kinetics of Selective Biological Transport. Biophys J. 8,1329-1338.Miller, J.L., Fox, D.A. and Litman, B.J. (1986) Amplification of PhosphodiesteraseActivation is Greatly Reduced by Rhodopsin Phosphorylation. Biochemistry 25,4983-4988.Miller, R.F. and Dowling I.E. (1970) Intracellular Responses of the Muller (Glial)Cells of Mudpuppy Retina. Their Relation to b -Wave of the Electroretinogram. J.Neurophysiol. 33, 323-341.Mizuno, K. and Sato, K. (1975) Reassessment of Histochemistry of RetinalGlycogen. Exp. Eye Res. 21, 489-497.Molday, L.L., Cook, N.J., Kaupp, U.B. and Molday, R.S. (1990) The cGMP-gatedCation Channel of Bovine Rod Photoreceptor Cells Is Associated with a 240-kDaProtein Exhibiting Immunochemical Cross-reactivity with Sepctrin. J. Biol. Chem.265, 18690-18695.Molday, R.S., Hicks, D., and Molday, L. (1987) Peripherin, A Rim-specificMembrane Protein of Rod Outer Segment Discs. Invest. Ophthalmol. Vis. Sci. 28,50-61.Molday, R.S. and MacKenzie, D. (1983) Monoclonal Antibodies to Rhodopsin:Characterization Cross-reactivity, and Application as Structural Probes.Biochemistry 22, 653-660.Molday, R.S. and Molday, L.L. (1987) Differences in the Protein Composition ofBovine Retinal Rod Outer Segment Disk and Plasma membranes Isolated by aRicin-Gold-Dextran Density Perturbation Method. J. Cell Biol. 105, 2589-2601.Morjaria, B. and Voaden, M.J. (1979) The Formation of Glutamate, Aspartate andGABA in the Rat Retina; Glucose and Glutamine as Precursors. Exp. Eye. Res. 21,489-497.Mueckler, M. (1990) Family of Glucose-Transporter Genes: Implications forGlucose Homeostasis and Diabetes. Diabetes 39, 6-11.Mueckler, M., Caruso, C., Baldwin, S.A., Panico, M., Blench, I., Morris, H.R.,Allard, W.J., Lienhard, G.E., Lodish, H.F. (1985) Sequence of Structure of a HumanGlucose Transporter. Science 229, 941-945.Murphy, J.R. (1960) Erythrocyte Metabolism: II Glucose Metabolism and Pathways.J. Lab. Clin. Med. 55, 286-302.Murthy, S.N.P., Liu, T., Kaul, R.K., Kohler, H., and Steck, T.L. (1981) The Aldolase-binding Site of the Human Erythrocyte Membrane Is at the NH 2 Terminus of Band3. J. Biol. Chem. 256, 11203-11208.125Nataki, K. and Yau, K.-W. (1988a) Calcium and Magnesium Fluxes Across thePlasma Membrane of the Toad Rod Outer Segment. J. Physiol. 395, 695-792.Nataki, K. and Yau, K.-W. (1988b) Guanosine 3'-5'-Cyclic Monophosphate-activated Conductance Studied in a Truncated Rod Outer Segment of the Toad. J.Physiol. (London) 395, 731-753.Neubert, T.A., Johnson, R.S., Hurley, J.B. and Walsh, K. (1992) J. Biol. Chem. 267,18274-18277.Nicotra, C., and Livrea, M.A. (1982) Retinol Dehydrogenase from Bovine RetinalRod Outer Segments. J. Biol. Chem. 257, 11836-11841.Nielsen, N.C., Fleischer, S., and McConnell, D.G. (1970) Lipid Composition ofBovine Retinal Outer Segment Fragments. Biochim. Biophys. Acta 211, 10-19.Nowak, K., Wolny, M., and Banas, T. (1981) The Complete Amino Acid Sequenceof Human Muscle Glyceraldehyde-3-phosphate Dehydrogenase. FEBS Lett. 134,143-146.Ogden, T.E. (1973) The Oscillatory Waves of the Primate Electroretinogram.Vision Res. 13, 1059-1074.Palczewski, K., Hargrave, P.A., McDowell, J.H. and Ingebritsen, T.S. (1989) TheCatalytic Subunit of Phosphatase 2A Dephosphorylates Phosphoopsin. Biochemistry28, 415-419.Pan, L., Beverley, P.C.L., Bobrow, L.G., Swallow, D.M., and Isaacson, P.G. (1989)Production of Monoclonal Antibodies to Lactate Dehydrogenase (LDH)Isoenzymes for Immunohistochemical Study on Fixed Tissue Section. Histochem. J.21, 638-644.Papermaster, D.S. and Dreyer, W.J. (1974) Rhodopsin Content in the OuterSegment Membranes of Bovine and Frog Retinal Rods. Biochemistry 13, 2438-2444.Penn, R.D. and Hagins, W.A. (1972) Kinetics of the Photocurrent of Retinal Rods.Biophys. J. 12, 1073-1094.Pfeffer, B., Wiggert, B., Lee, L., Zonnenberg, B., Newsome, D. and Chader, G.(1983) The Presence of a Soluble Interphotoreceptor Retinol-binding Protein(IRBP) in the Retinal Interphotoreceptor Space. J. Cell Physiol. 117, 333-341Pugh, E.N.Jr. and Lamb, T.D. (1990) Cyclic GMP and Calcium: The InternalMessengers of Excitation and Adaptation in Vertebrate Photoreceptors. Vision Res.30, 1923-1948.Rahman, M.A. and Kerly, M. (1961) pathways of Glucose Metabolism of Ox RetinaIn Vitro. Biochem. J. 78, 536-540.Ratto, G.M., Payne, R., Owen, W.G., and Tsien, R.Y. (1988) The Concentration ofCytosolic Free Calcium in Vertebrate Rod Outer Segments Measured with Fura-2.J. Neurosci. 8, 3240-3246.Rawn, J.D. (1989) in Biochemistry. Neil Patterson Publishers, North Carolina.126Reid, D.M., Friedel, U., Molday, R.S., and Cook, N.J. (1990) Identification of theSodium-calcium Exchanger as the Major Ricin-binding Glycoprotein of Bovine RodOuter Segments and Its Localization to the Plasma Membrane. Biochemistry 29,1601-1607.Rich, G.T., Dawson, A.P., and Pryor, J.S. (1984) Lack of Binding of Glyceraldehyde-3-phosphate Dehydrogenase to Erythrocyte Membranes Under In Vivo Conditions.Biochem. J. 221, 197-202.Rich, G.T., Pryor, J.S., and Dawson, A.P. (1985) Glyceraldehyde-3-phosphateDehydrogenase Release from Erythrocytes During Haemolysis. Biochim. Biophys.Acta 817, 61-66.Riley, M.V. and Voaden, M.J. (1970) The Metabolism of the Isolated Retina.Ophthalmic Res. 1, 58-64.Robinson, W.E., and Hagins, W.A. (1979) GTP Hydrolysis in Intact Rod OuterSegments and the Transmitter Cycle in Visual Excitation. Nature 280, 398-400.Rogalski, A.A., Steck, T.L., and Waseem, A. (1989) Association of Glyceraldehyde-3-phosphate Dehydrogenase with the Plasma Membrane of the Intact Human RedBlood Cell. J. Biol. Chem. 264, 6438-6446.Saari, J.C. and Bredberg, L. (1988) CoA- and Non-CoA-dependent RetinolEsterification in Retinal Pigment Epithelium. J. Biol. Chem. 263, 8084-8090.Saari, J.C. and Bredberg, L. (1989) Lecithin:Retinol Acyltransferase in RetinalPigment Epthelial Micromomes. J. Biol. Chem. 264, 8636-8640.Schnetkamp, P.P.M. (1986) Sodium-calcium Exchange in the Outer Segments ofBovine Rod Photoreceptors. J. Physiol. (London) 373, 25-45.Schnetkamp, P.P.M., and Daemen, F.J.M. (1981) Transfer of High energyPhosphate in Bovine Rod Outer Segments: A Nucleotide Buffer System. Biochim.Biophys. Acta 672, 307-312.Schnetkamp, P.P.M., Klompmakers, A.A., and Daemen, F.J.M. (1979) The Isolationof Stable Cattle Rod Outer Segments With An Intact Plasma Membrane. Biochim.Biophys. Acta 552, 379-389.Schroder, W.H. and Fain, G.L. (1984) Light-dependent Calcium Release fromPhotoreceptors Measured by Laser Micro-mass Analysis. Nature 309, 268-270.Scopes, R.K. and Stoter, A. (1982) Purification of All Glycolytic Enzymes from OneMuscle Extract. Methods Enzymol. 90, 479-490.Shichi, H. (1983) in Biochemistry of Vision. Academic Press, New York.Shimke, R.T. (1959) Effects of Prolonged Light Deprivation on the Development ofRetinal Enzymes in the Rabbit. J. Biol. Chem. 234, 700-703.127Shin, B.C., and Carraway, K.L. (1973) Association of Glyceraldehyde-3-phosphateDehydrogenase with the Human Erythrocyte Membrane. J. Biol. Chem. 248, 1436-1444.Sillman, A.J., Ito, H. and Tomita, T. (1969) Studies on the Mass Receptor Potentialof the Isolated Frog Retina: On the Basis of the Ionic Mechanism. Vision Res. 9,1443-1451.Silverman, M. (1991) Structure and Function of Hexose Transporters. Annu. Rev.Biochem. 60, 757-794.Sitaramayya, A. and Liebman, P.A. (1983) Mechanism of ATP Quench ofPhosphodiesterase Activation in Rod Disk Membranes. J. Biol. Chem. 258, 1205-1209.Solti, M., Bartha, F., Halasz, N., Toth, G., Sirokman, F., and Friedrich, P. (1981)Localization of Glyceraldehyde-3-phosphate Dehydrogenase in Intact HumanErythrocytes. J. Biol. Chem. 256, 9260-9265.Steck, T.L. (1974) Preparation of Impermeable Inside-out and Right-side-outVesicles from Erythrocyte Membranes. Methods Membr. Bio. 2, 245-281.Stein, W.D. (1986) in Facilitated Diffusion Across Cell Membranes, p. 231-306,Academic Press, New York.Steinberg, R.H., Fisher, S.K. and Anderson, D.H.(1980) Disc Morphogenesis inVertebrate Photoreceptors. J. Comp. Neurol. 190, 501-518.Stirling, C.E. and Lee, A. (1980) [ 3H]Oubain Autoradiography of Frog Retina. J.Cell Biol. 85, 313-324.Strapazon, E., and Steck, T.L. (1976) Binding of Rabbit Muscle Aldolase to Band 3,the Predominant Polypeptide of the Human Erythrocyte Membrane. Biochemistry15, 1421-1424.Strapazon, E., and Steck, T.L. (1977) Interaction of Aldolase and the Membrane ofHuman Erythrocytes. Biochemistry 16, 2966-2971.Stryer, L. (1986) Cyclic GMP Cascade of Vision. Annu. Rev. Neurosci. 9, 87-119.Stryer, L. (1988) in Biochemistry. W.H. Freeman and Co., New York.Stryer, L. (1991) Visual Excitation and Recovery. J. Biol. Chem. 266, 10711-10714.Szel, A., Takacs, L., Monostori, E., Diamantstein, T., Vigh-Teichmann, I , andReshlich, P. (1986) Monoclonal Antibody-Recognizing cone Visual Pigment. Exp.Eye Res. 43, 871-883.Thomson, M.J., and Richards, J.F. (1977) Ornithine Decarboxylase and ThymidineKinase Activity in Tissues of Prolactin-treated RAts: Effect of Hypophysectomy.Life Sciences 22, 337-344.Thorens, B., Charron, M.J. and Lodish, H.F. (1990) Molecular Physiology ofGlucose Transporters. Diabetes Care 13, 209-218.128Thorens, B., Sarkar, H.K., Kaback, H.R., and Lodish, H.F. (1988) Cloning andFunctional Expression in Bacteria of a Novel Glucose Transporter Present in Liver,Intestine, Kidney, and B-Pancreatic Islet Cells. Cell 55, 281-290.Tombes, R.M. and Shapiro, B.M. (1985) Metabolite Channeling: APhosphorylcreatine Shuttle to Mediate High Energy Phosphate Transport BetweenSperm Mitochondrion and Tail.Toyoda, N., Robinson, F.W., Smith, M.M., Flanagan, J.E., and Kono, T. (1986)Apparent Translocation of Glucose Transport Activity in Rat EpididymalAdipocytes by Insulin-like Effects of High pH or Hyperosmolarity. J. Biol. Chem.261, 2117-2122.Vidaver, G.A. (1966) Inhibition of Parallel Flux and Augmentation of Counter FluxShown by Transport Models Not Involving a Mobile Carrier. J. Theoret. Biol. 10,301-306.Wald, G. (1968) Molecular Basis of Visual Excitation. Science 162, 230-239.Wald, G. and Brown, P.K. (1953) The Molecular Extinction of Rhodopsin. J. Gen.Physiol. 37, 189-200.Wallimann, T., Wegmann, G., Moser, H., Huber, R., and Eppenberger, H.M. (1986)High Content of Creatine Kinase in Chicken Retina: CompartmentalizedLocalization of Creatine Kinase Isoenzymes in Photoreceptor Cells. Proc. Natl.Acad. Sci. U.S.A. 83, 3816-3819.Walmsley, A.R. (1988) The Dynamics of the Glucose Transporter. TIBS 13, 226-231.Wang, C. (1987) The D-Glucose Transporter is Tissue-specific. J. Biol. Chem. 262,15689-15695.Wang, C.-S., and Alaupovic, P. (1980) Glyceraldehyde-3-phosphate Dehydrogenasefrom Human Erythrocyte Membranes. Kinetic Mechanism and CompetitiveSubstrate Inhibition by Glyceraldehyde-3-phosphate. Arch. Biochem. Biophys. 205,136-145.Wensel, T.G. and Stryer, L. (1990) Activation Mechanism of Retinal Rod CyclicGMP Phosphodiesterase Probed by Fluorescein-labeled Inhibitory Subunit.Biochemistry 29, 2155-2161.Wheeler, G.L. and Bitensky, M.W. (1977) A Light-activated Cyclic GMPPhosphodiesterase. Proc. Natl. Acad. Sci. U.S.A. 74, 4238-4242.Wheeler, T.J. and Hinkle, P.C. (1981) Kinetic Properties of the ReconstitutedGlucose Transporter from Human Erythrocytes. J. Biol. Chem. 256, 8907-8914.Whitesell, R.R. and Gliemann, J. (1979) Kinetic Parameters of Transport of 3-0-Methylglucose and Glucose in Adipocytes. J. Biol. Chem. 254, 5276-5283.129Wilden, U. and Kuhn, H. (1982) Phosphodiesterase Activation by PhotoexcitedRhodopsin Is Quenched When Rhodopsin Is Phosphorylated and Binds the Intrinsic48-kDa Protein of Rod Outer Segments. Biochemistry 21, 3014-3022.Winkler, B.S. (1972) The Electroretinogram of the Isolated Rat Retina. Vision Res.12, 1183-1198.Winkler, B.S. (1981a) Glycolytic and Oxidative Metabolism in Relation to RetinalFunction. J. Physiol. 77, 667-692.Winkler, B.S. (1981b) The Intermediary Metabolism of the Retina: Biochemical andFunctional Aspects. In Biochemistry and Physiology of the Eye. R.E. Anderson (ed.),227-242, American Academy of Ophthalmology, San Francisco.Winkler, B.S. (1987) In Vitro Oxidation of Ascorbic Acid and Its Prevention byGSH. Biochim. Biophys. Acta 925, 258-264.Winkler, B.S., DeSantis, N., and Solomon, F. (1986) Multiple NADPH-ProducingPathways Control Glutathione (GSH) Content in Retina. Exp. Eye Res. 43, 829-847.Wisloki, G.B. and Sidman, R.L. (1954) The Chemical Morphology of the Retina. J.Comp. Neurol. 101, 53-91.Wong, S. and Molday, R.S. (1986) A Spectrin-like Protein in Retinal Rod OuterSegments. Biochemistry 25, 6294-6300.Woodruff, M.L., Bownds, M.D. (1979) Amplitude, Kinetics, and Reversibility of aLight-induced Decrease in Guanosine 3',5'-Cyclic Monophosphate in FrogPhotoreceptor Membranes. J. Gen. Physiol. 73, 629-653.Yamazaki, A., Sen, I., Bitensky, M., Casnellie, J.E., and Grrengard, P. (1980) CyclicGMP-specific High-affinity, Noncatalytic Binding Sites on Light-activatedPhosphodiesterase. I Biol. Chem. 255, 11619-11624.Yamazaki, A., Hayashi, F., Tatsumi, M., Bitensky, M.W. and George, J.S. (1990)Interactions between the Subunits of cyclic GMP Phosphodiesterase in RanaCatesbiana Rod Photoreceptors. J. Biol. Chem. 265, 11539-11548.Yau, K.-W. and Baylor, D.A. (1989) Cyclic GMP-activated Conductance of RetinalPhotoreceptor Cells. Annu. Rev. Neurosci. 12, 289-327.Yau, K. -W., Haynes, L.W. and Nakatani, K. (1988) Study of the PhototransductionMechanism in Rods and Cones in Proceedings of the Retina Rsearch FoundationSymposium. D.M.-K. Lam (ed.), 41-58, Portfolio Publishing Company, Inc. Texas.Yau, K. -W. and Nakatani, K. (1984) Electrogenic Na-Ca Exchange in Retinal RodOuter Segment. Nature 311, 661-663.Yau, K. -W. and Nakatani, K. (1985) Light-induced Reduction of Cytoplasmic FreeCalcium in Retinal Rod Outer Segment. Nature 313, 252-255.Yee, R. and Liebman, P.A. (1978) Light-activated Phosphodiesterase of the RodOuter Segment. Kinetics and Parameters of Activation and Deactivation. J. Biol.Chem. 253, 8902-8909.130Yeltman, D.R. and Harris, B.G. (1982) Fructose-bisphosphate Aldolase fromHuman Erythrocytes. Methods Enzymol. 90, 251-254.Yokishima, S. and Hagins, W.A. (1970) Ionic Basis of Dark Current andPhotocurrent of Retinal Rods. Biophys. J. 10, 60a.Yu, J., and Steck, T.L. (1975) Associations of Band 3, the Predominant Polypeptideof the Human Erythrocyte Membranes. J. Biol. Chem. 250, 9176-9184.Zoccoli, M.A. and Lienhard, G.E. (1977) Monosaccharide Transport in Protein-depleted Vesicles from Erythrocyte Membranes. J. Biol. Chem. 252, 3131-3135.Zuckerman, R., Buzdygon, B., Philip, N., Liebman, P., Sitaramayya, A. (1985)Arrestin: An ATP/ADP Exchange Protein that Regulates cGMP PhosphodiesteraseActivity in Retinal Rod Disk Membranes (RDM). Biophys. J. 47, 37a.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items