Identification of the sodium-calcium exchanger as the major ricin-binding glycoprotein of bovine rod outer segments and its localization to the plasma membrane

After neuraminidase treatment the Na+/Ca2+ exchanger of bovine rod outer segments was found to specifically bind Ricinus communis agglutinin. SDS gel electrophoresis and Western blotting of ricin-binding proteins purified from rod outer segment membranes by lectin affinity chromatography revealed the existence of two major polypeptides of Mr 215K and 103K, the former of which was found to specifically react with PMe 1B3, a monoclonal antibody specific for the 230-kDa non-neuraminidase-treated Na+/Ca2+ exchanger. Reconstitution of the ricin affinity-purified exchanger into calcium-containing liposomes revealed that neuraminidase treatment had no significant effect on the kinetics of Na+/Ca2+ exchange activation by sodium. We further investigated the density of the Na+/Ca2+ exchanger in disk and plasma membrane preparations using Western blotting, radioimmunoassays, immunoelectron microscopy, and reconstitution procedures. The results indicate that the Na+/Ca2+ exchanger is localized in the rod photoreceptor plasma membrane and is absent or present in extremely low concentrations in disk membranes, as we have previously shown to be the case for the cGMP-gated cation channel. Previous reports describing the existence of Na+/Ca2+ exchange activity in rod outer segment disk membrane preparations may be due to the fusion of plasma membrane components and/or the presence of contaminating plasma membrane vesicles.     

The cGMP-gated channel and related glutamic acid-rich proteins interact with peripherin-2 at the rim region of rod photoreceptor disc membranes

The rod cGMP-gated channel is localized in the plasma membrane of rod photoreceptor outer segments, where it plays a central role in phototransduction. It consists of alpha- and beta-subunits that assemble into a heterotetrameric protein. Each subunit contains structural features characteristic of nucleotide-gated channels, including a cGMP-binding domain, multiple membrane-spanning segments, and a pore region. In addition, the beta-subunit has a large glutamic acid- and proline-rich region called GARP that is also expressed as two soluble protein variants. Using monoclonal antibodies in conjunction with immunoprecipitation, cross-linking, and electrophoretic techniques, we show that the cGMP-gated channel associates with the Na/Ca-K exchanger in the rod outer segment plasma membrane. This complex and soluble GARP proteins also interact with peripherin-2 oligomers in the rim region of outer segment disc membranes. These results suggest that channel/peripherin protein interactions mediated by the GARP part of the channel beta-subunit play a role in connecting the rim region of discs to the plasma membrane and in anchoring the channel.exchanger complex in the rod outer segment plasma membrane.     

The cGMP-gated channel of the rod photoreceptor cell characterization and orientation of the amino terminus.

The molecular properties and orientation of the cGMP-gated cation channel of bovine rod outer segment membranes were studied using biochemical and immunochemical methods. Western blots labeled with anti-channel monoclonal antibodies indicate that the channel has a subunit Mr of 63,000 in bovine rod outer segment membranes prepared in the presence and absence of protease inhibitors and in rod outer segments from other mammalian retinas. The channel has an apparent Mr of 78,000 in both COS-1 cells and Xenopus oocytes expressing the cloned cDNA. NH2-terminal sequence analysis indicates that the lower Mr of the channel in rod outer segments is caused by the absence of the first 92 amino acids predicted by cDNA sequence analysis. Immunofluorescent and immunogold labeling has confirmed that the 63,000 form of the channel is present in rod outer segments. These results indicate that photoreceptor cell-specific co-translational or post-translational cleavage of the NH2-terminal segment of the channel occurs prior or during the incorporation of the channel into the rod outer segment plasma membrane. Immunogold labeling studies using site-directed antibodies also indicate that the NH2-terminal segment of the rod outer segment channel is exposed on the cytoplasmic side of the plasma membrane.     

Glyceraldehyde-3-phosphate dehydrogenase is a major protein associated with the plasma membrane of retinal photoreceptor outer segments

A major 38-kDa protein associated with bovine rod outer segment plasma membranes, but not disk membranes, has been identified as glyceraldehyde-3-phosphate dehydrogenase on the basis of its N-terminal sequence and specific enzyme activity. This enzyme was extracted from lysed rod outer segments or isolated rod outer segment plasma membrane with 0.15 M NaCl and purified to homogeneity by affinity chromatography on a NAD(+)-agarose column. A specific activity of 90-100 units/mg of protein is within the range of activity obtained for glyceraldehyde-3-phosphate dehydrogenase isolated from other mammalian cells. Enzyme activity measurements indicate that this enzyme makes up approximately 2% of the total rod outer segment protein and over 11% of the plasma membrane protein. Protease digestion and binding studies on purified rod outer segment plasma and disk membranes suggest that glyceraldehyde-3-phosphate dehydrogenase reversibly interacts with a protease-sensitive plasma membrane-specific protein of rod outer segments. The finding that glyceraldehyde-3-phosphate dehydrogenase is present in large quantities in rod outer segments suggests that at least some of the energy required for the synthesis of ATP and GTP for phototransduction and other processes of the outer segment is derived from glycolysis which takes place within this organelle.     

Glycolytic enzymes and a GLUT-1 glucose transporter in the outer segments of rod and cone photoreceptor cells

The presence of glycolytic enzymes and a GLUT-1-type glucose transporter in rod and cone outer segments was determined by enzyme activity assays, glucose uptake measurements, Western blotting, and immunofluorescence microscopy. Enzyme activities of six glycolytic enzymes including hexokinase, phosphofructokinase, aldolase, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, pyruvate kinase, and lactate dehydrogenase, were found to be present in purified rod outer segment (ROS) preparations. Immunofluorescence microscopy of bovine and chicken retina sections labeled with monoclonal antibodies against glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, and lactate dehydrogenase have confirmed that these enzymes are present in rod and cone outer segments and not simply contaminants from the inner segments or other cells. Rod outer segments were also found to contain glucose transport activity as detected by 3-O-[14C]methylglucose uptake and exchange. The glucose transporter had a Km of 6.3 mM and a Vmax of 0.15 nmol of 3-O-methylglucose/s/mg of ROS membrane protein for net uptake and a Km of 29 mM and a Vmax of 1.06 nmol of 3-O-methylglucose/s/mg of ROS membrane protein for equilibrium exchange. These Km values for net uptake and equilibrium exchange are similar to values obtained for human red blood cells and are characteristic of GLUT-1-type glucose transporter. The transport was inhibited by both cytochalasin B and phloretin. Western blot analysis and immunofluorescence microscopy using type-specific glucose transporter antibodies indicated that both rod and cone outer segment plasma membranes have a GLUT-1 glucose transporter of Mr 45K as found in red blood cells and brain microsomal membranes. Solid-phase radioimmune competitive inhibition studies indicated that rod outer segment plasma membranes contained 15% the number of glucose transporters found in human red blood cell membranes and had an estimated density of 400 glucose transporter per micron2 of plasma membrane. These studies support the view that outer segments can generate energy in the form of ATP and GTP by anaerobic glycolysis to supply at least some of the energy requirements for phototransduction and other metabolic processes.     

Purification of retinol dehydrogenase from bovine retinal rod outer segments

We purified retinol dehydrogenase from bovine rod outer segments using polyethylene glycol precipitation and hydroxylapatite, concanavalin A-Sepharose CL-4B, and Sepharose CL-6B column chromatography in the presence of NADP. We obtained 13-fold purification of retinol dehydrogenase with specific activity of 61.8 nmol/min/mg and 3.8% recovery. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that retinol dehydrogenase had a molecular mass of 37,000 daltons. The Km values of purified retinol dehydrogenase for all-trans retinol and all-trans retinal were 6.6 mM and 0.085 mM, respectively. The purified enzyme reacted with the all-trans retinal but not with 13-, 11-, and 9-cis compounds. In addition, we prepared antibody to retinol dehydrogenase using rat. The anti-retinol dehydrogenase antibody precipitated retinol dehydrogenase activity and was confirmed to bind to 37-kDa protein by Western blotting. We also found that anti-retinol dehydrogenase antibody bound to bovine rod outer segments specifically by immunohistochemical technique. The molar ratio of retinol dehydrogenase to opsin in rod outer segments estimated by enzyme-linked immunosorbent assay was 1:140.     

Glycoproteins specific for the retinal rod outer segment plasma membrane

Two ricin-specific glycoproteins have been identified on neuraminidase-treated rod outer segment plasma membranes of bovine retinal photoreceptor cells. Ricin-gold-dextran particles were observed by electron microscopy to densely label the surface of neuraminidase-treated rod outer segments. Western blotting of proteins separated by SDS-gel electrophoresis indicated that two ricin-binding glycoproteins of Mr 230,000 and 110,000 are specific for the plasma membrane and are not found in disk membranes. These glycoproteins can serve as specific probes for the purification of the rod outer segment plasma membrane.     

Antibodies against synthetic peptides used to determine the topology and site of glycosylation of the cGMP-gated channel from bovine rod photoreceptors.

Peptides corresponding to amino acids 321-339 (peptide GS21) and 416-431 (peptide GS31) of the cGMP-gated channel from bovine rod photoreceptors were synthesized and used as antigens for the preparation of polyclonal antibodies. After affinity purification, both antipeptide antibodies were found to bind specifically to the channel protein after Western blotting, but only the antibody against GS21 gave satisfactory results on enzyme-linked immunosorbent assay and electron microscopy. Using immunocytochemistry, we were able to localize amino acids 321-339 to the extracellular side of the rod photoreceptor plasma membrane. By synthesizing heptapeptides corresponding to amino acids 324-330 (peptide GS2s) and 420-426 (peptide GS3s), we were able to affinity purify antibodies specific for two N-glycosylation consensus sites in the channel protein. As assessed by Western blotting, antibodies against GS3s were found to bind to both the glycosylated and deglycosylated channel proteins, whereas antibodies against GS2s only bound to the channel protein after enzymatic deglycosylation. Together, these results allow the refinement of folding models for the cGMP-gated channel and implicate Asn-327 as being the sole site of N-glycosylation.     

Identification of frog photoreceptor plasma and disk membrane proteins by radioiodination

Several functions have been identified for the plasma membrane of the rod outer segment, including control of light-dependent changes in sodium conductance and a sodium-calcium exchange mechanism. However, little is known about its constituent proteins. Intact rod outer segments substantially free of contaminants were prepared in the dark and purified on a density gradient of Percoll. Surface proteins were then labeled by lactoperoxidase-catalyzed radioiodination, and intact rod outer segments were reisolated. Membrane proteins were identified by polyacrylamide gel electrophoresis and autoradiography. The surface proteins labeled included rhodopsin, the major membrane protein, and 12 other proteins. Several control experiments indicated that the labeled proteins are integral membrane proteins and that label is limited to the plasma membrane. To compare the protein composition of plasma membrane with that of the internal disk membrane, purified rod outer segments were lysed by hypotonic disruption or freeze-thawing, and plasma plus disk membranes were radioiodinated. In these membrane preparations, rhodopsin was the major iodinated constituent, with 12 other proteins also labeled. Autoradiographic evidence indicated some differences in protein composition between disk and plasma membranes. A quantitative comparison of the two samples showed that labeling of two proteins, 24 kilodaltons (kDa) and 13 kDa, was enriched in the plasma membrane, while labeling of a 220-kDa protein was enriched in the disk membrane. These plasma membrane proteins may be associated with important functions such as the light-sensitive conductance and the sodium-calcium exchanger.     

Solubilization, purification, and reconstitution of the sodium-calcium exchanger from bovine retinal rod outer segments

We have previously described a method for the solubilization and reconstitution of the cGMP-gated cation channel from the membranes of bovine rod outer segments (Cook, N. J., Zeilinger, C., Koch, K.-W., and Kaupp, U. B. (1986) J. Biol. Chem. 261, 17033-17039). Here we report that not only cGMP but also sodium is capable of releasing entrapped calcium from liposomes reconstituted with total rod outer segment membrane proteins. Other alkali cations tested were unable to induce calcium efflux; therefore, we concluded that the sodium-induced calcium efflux was due to the sodium-calcium exchanger. Sodium was found to activate calcium efflux from these liposomes with an EC50 of approximately equal to 35 mM, comparable to values reported for the sodium-calcium exchanger in native rod outer segment membranes. We found that reconstitution of the sodium-calcium exchanger is quantitative and used this method to assay the exchange protein during purification using conventional protein chromatographic techniques. In this way, we were able to purify and identify as the rod outer segment sodium-calcium exchanger a glycoprotein of apparent Mr = 220,000 to greater than 90% homogeneity. The specific activity of the purified protein at room temperature was 8.2 mumol of Ca2+ exchanged min-1 mg-1 of protein at 50 mM Na+, corresponding to a turnover number of approximately equal to 30 Ca2+ (or 90 Na+) s-1 exchanger-1. The Mr = 220,000 protein reported here appears to be distinct from another protein ("rim protein") with an identical Mr known to exist in these membranes.     

Evidence for an Association of the 15-kDa Proteolipid of Mediatophore with a 14-kDa Polypeptide

The present report shows that mediatophore, a nerve terminal membrane protein that translocates acetylcholine on calcium action, forms a complex with a 14-kDa polypeptide. The complex was identified based on the following results. (a) A polyclonal antimediatophore antiserum that immunoprecipitates activity precipitates both the 15- and 14-kDa polypeptides. (b) After HPLC purification of mediatophore, both antigens were found in the same peak. (c) After 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate solubilization of presynaptic membranes or of the purified mediatophore, an immunoaffinity column made with the anti-14-kDa antigen monoclonal antibody retained both the 14-kDa and the 15-kDa polypeptide. Similarly, immunoprecipitation experiments using protein A-coated beads sedimented an immunocomplex in which both antigens were found. (d) The 14-kDa antigen could be localized in the synaptosomal membrane where mediatophore and its 15-kDa component are found.     

Proteomics of photoreceptor outer segments identifies a subset of SNARE and Rab proteins implicated in membrane vesicle trafficking and fusion

The outer segment is a specialized compartment of vertebrate rod and cone photoreceptor cells where phototransduction takes place. In rod cells it consists of an organized stack of disks enclosed by a separate plasma membrane. Although most proteins involved in phototransduction have been identified and characterized, little is known about the proteins that are responsible for outer segment structure and renewal. In this study we used a tandem mass spectrometry-based proteomics approach to identify proteins in rod outer segment preparations as an initial step in defining their roles in photoreceptor structure, function, renewal, and degeneration. Five hundred and sixteen proteins were identified including 41 proteins that function in rod and cone phototransduction and the visual cycle and most proteins previously shown to be involved in outer segment structure and metabolic pathways. In addition, numerous proteins were detected that have not been previously reported to be present in outer segments including a subset of Rab and SNARE proteins implicated in vesicle trafficking and membrane fusion. Western blotting and immunofluorescence microscopy confirmed the presence of Rab 11b, Rab 18, Rab 1b, and Rab GDP dissociation inhibitor in outer segments. The SNARE proteins, VAMP2/3, syntaxin 3, N-ethylmaleimide-sensitive factor, and Munc 18 detected in outer segment preparations by mass spectrometry and Western blotting were also observed in outer segments by immunofluorescence microscopy. Syntaxin 3 and N-ethylmaleimide- sensitive factor had a restricted localization at the base of the outer segments, whereas VAMP2/3 and Munc 18 were distributed throughout the outer segments. These results suggest that Rab and SNARE proteins play a role in vesicle trafficking and membrane fusion as part of the outer segment renewal process. The data set generated in this study is a valuable resource for further analysis of photoreceptor outer segment structure and function.     

Binding of the cGMP-gated channel to the Na/Ca-K exchanger in rod photoreceptors

The intracellular Ca(2+) concentration in rod outer segments of vertebrate photoreceptors is controlled by Ca(2+) influx through cGMP-gated channels and by Ca(2+) efflux driven by Na/Ca-K exchangers. Previously, we suggested that channel and exchanger are associated (Bauer, P. J., and Drechsler, M. (1992) J. Physiol. (Lond. ) 451, 109-131). This suggestion has been thoroughly examined using a variety of biochemical approaches. First, we took advantage of the fact that cGMP-gated channels bind calmodulin (CaM). Using CaM affinity chromatographic purification of the channel in 10 mm CHAPS, a significant fraction of exchanger was co-eluted with the channel indicating a binding affinity between channel and exchanger. Binding of channel and exchanger was examined more directly by cross-linking of proteins in the rod outer segment membranes. Activation of the channel with cyclic 8-bromo-GMP lead to exposure of a cysteine, which allowed cross-linking of the channel to the exchanger with the thiol-specific reagent dl-1,4-bismaleimido-2,3-butanediol. Cleavage of the cross-links and electrophoretic analysis indicated that a cross-link between the alpha-subunit of the channel and the exchanger formed. Furthermore, a cross-link between two adjacent alpha-subunits of the channel was found, suggesting that the alpha-subunits of the native channel are dimerized. Further support for an interaction between alpha-subunit and exchanger was obtained by in vitro experiments. Specific binding of the exchanger to the alpha-subunit but not to the beta-subunit of the channel was observed in Western blots of purified channel incubated with purified exchanger. This study suggests that two exchanger molecules bind to one cGMP-gated channel and, more specifically, that binding of exchanger molecules occurs at the alpha-subunits, which in the native channel are dimerized. The implications of these findings regarding the possibility of local Ca(2+) signaling in vertebrate photoreceptors will be discussed.     

A variant of arrestin-1 binds rod outer segment membranes in a light-independent manner

A 50-kDa-polypeptide band peripherally bound to retinal rod outer segment (ROS) membranes was purified by anion-exchange chromatography. When the 50-kDa protein was compared with purified arrestin-1, it was observed that: (1) both proteins comigrated on sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and were recognized by either anti-50-kDa protein polyclonal antibodies or anti-arrestin-1 monoclonal antibodies; (2) protein fragments and peptide fingerprint maps obtained following limited and complete proteolysis with specific proteases were very similar for both molecules; and (3) several chromatographically-purified tryptic peptides from the 50-kDa protein possessed the same amino acid composition as tryptic peptides deduced from the reported arrestin-1 primary structure. Consequently, arrestin-1 and the purified 50-kDa protein must correspond to variants of the same molecule. However, in contrast to arrestin-1 that associated to the ROS membranes only in the presence of light and ATP, the 50-kDa protein interacted with the ROS membranes in a light-independent manner, either in the presence or absence of ATP. These results clearly established that phosphorylated and illuminated rhodopsin is not the membrane anchor for this variant of arrestin-1.     

Preparation and characterization of monoclonal antibodies to several frog rod outer segment proteins

Monoclonal antibodies to proteins important in phototransduction in the frog rod outer segment have been obtained. These include 6 different antibodies to rhodopsin, 50 to a guanine nucleotide binding protein (G-protein; 40,000 daltons), and 2 to cytoplasmic proteins. The antigens used were Percoll-purified rod outer segments, a rod outer segment soluble protein fraction, or a soluble plus peripheral membrane protein fraction. Antibodies were assayed by solid phase assay using a fluorogenic detection system. Proteins to which antibodies bound were assayed on Western blots, and the sensitivities of three different detection systems were compared. Most antibodies bound to only one rod outer segment protein band on Western blots. Immunofluorescence microscopy demonstrated binding of both anti-rhodopsin and anti-G-protein to isolated frog rod outer segments. Antibodies were purified from either culture supernatants or ascites fluid on protein A affinity columns. Two purified anti-G-protein antibodies have binding affinities to 125I-labeled G-protein of less than 10(-6) M-1. Of 11 antibodies to frog or bovine G-protein tested in solid phase and Western blot assays, all bind to the alpha rather than the beta or gamma subunits. Procedures developed here are being used in preparing other antibodies that affect reactions in the phototransduction pathway.     

Organization of cGMP sensing structures on the rod photoreceptor outer segment plasma membrane

A diffusion barrier segregates the plasma membrane of the rod photoreceptor outer segment into 2 domains; one which is optimized for the conductance of ions in the phototransduction cascade and another for disk membrane synthesis. We propose the former to be named “phototransductive plasma membrane domain," and the latter to be named “disk morphogenic plasma membrane domain." Within the phototransductive plasma membrane, cGMP-gated channels are concentrated in striated membrane features, which are proximally located to the sites of active cGMP production within the disk membranes. For proper localization of cGMP-gated channel to the phototransductive plasma membrane, the glutamic acid-rich protein domain encoded in the β subunit plays a critical role. Quantitative study suggests that the disk morphogenic domain likely plays an important role in enriching rhodopsin prior to its sequestration into closed disk membranes. Thus, this and our previous studies provide new insight into the mechanism that spatially organizes the vertebrate phototransduction cascade.     

Lectin binding and enzymatic deglycosylation of the cGMP-gated channel from bovine rod photoreceptors

In order to investigate the lectin-binding properties of the photoreceptor cGMP-gated channel, solubilized and purified channel protein was incubated with immobilized lectins followed by reconstitution of unbound proteins. Of the lectins tested, only concanavalin A (ConA) was able to specifically sediment channel activity. A 240-kDa protein, which copurifies with the 63-kDa channel protein but does not bind ConA, was also found to be sedimented by the ConA-affinity matrix, thereby implicating that it is associated with the channel complex. Treatment of the purified channel protein with the enzyme glycopeptidase F in the presence of the denaturing detergent sodium dodecyl sulfate resulted in a rapid reduction of the apparent molecular mass by 1.90 kDa, and the abolition of ConA-binding. No intermediate molecular weight species were observed, suggesting that the channel protein is N-glycosylated at one site only. Under nondenaturing conditions, the kinetics of deglycosylation were distinctly two-phased: 50-60% deglycosylation was achieved rapidly; however, prolonged incubation was required to arrive at complete deglycosylation. Reconstitution experiments showed that deglycosylation had no significant effect on the kinetics of channel protein activation by cGMP.     

The cGMP-gated channel of bovine rod photoreceptors is localized exclusively in the plasma membrane

Although there have been several reports pertaining to the existence of the cGMP-gated channel in the disk membrane of rod photoreceptors, its density there relative to that of the photoreceptor plasma membrane is unknown. Using immunoblotting, immunohistochemical, and reconstitution techniques on purified disk and plasma membrane preparations, we found that the density of channels in the plasma membrane was at least 50-fold higher than that of the disk membrane. Purification of membrane fractions without prior digestion of cytoskeletal components by mild trypsinization was found to increase the amount of channel protein present in disk membrane preparations. We propose that the presence of the channel protein in rod disk membrane preparations is an artifact arising from fusion of plasma membrane components during permeabilization of the photoreceptor cell.     

A 26 kd calcium binding protein from bovine rod outer segments as modulator of photoreceptor guanylate cyclase.

The resynthesis of cGMP in vertebrate photoreceptors by guanylate cyclase is one of the key events leading to the reopening of cGMP-gated channels after photoexcitation. Guanylate cyclase activity in vertebrate rod outer segments is dependent on the free calcium concentration. The basal activity of the enzyme observed at high concentrations of free calcium (greater than 0.5 microM) increases when the free calcium concentration is lowered into the nanomolar range (less than 0.1 microM). This effect of calcium is known to be mediated by a soluble calcium-sensitive protein in a highly cooperative way. We here show that this soluble protein, i.e. the modulator of photoreceptor guanylate cyclase, is a 26 kd protein. Reconstitution of the purified 26 kd protein with washed rod outer segment membranes containing guanylate cyclase revealed a 3- to 4-fold increase of cyclase activity when the free calcium concentration was lowered in the physiological range from 0.5 microM to 4 nM. Guanylate cyclase in whole rod outer segments was stimulated 10-fold in the same calcium range. The activation process in the reconstituted system was similar to the one in the native rod outer segment preparation, it showed a high cooperativity with a Hill coefficient n between 1.4 and 3.5. The half-maximal activation occurred between 110 and 220 nM free calcium. The molar ratio of the modulator to rhodopsin is 1:76 +/- 32. The protein is a calcium binding protein as detected with 45Ca autoradiography. Partial amino acid sequence analysis revealed a 60% homology to visinin from chicken cones.     

Ultrastructural localization of rhodopsin in the vertebrate retina

Early work by Dewey and collaborators has shown the distribution of rhodopsin in the frog retina. We have repeated these experiments on cow and mouse eyes using antibodies specific to rhodopsin alone. Bovine rhodopsin in emulphogene was purified on an hydroxyapatite column. The purity of this reagent was established by spectrophotometric criteria, by sodium dodecyl sulfate (SDS) gel electrophoresis, and by isoelectric focusing. This rhodopsin was used as an immunoadsorbent to isolate specific antibodies from the antisera of rabbits immunized with bovine rod outer segments solubilized in 2% digitonin. The antibody so prepared was shown by immunoelectrophoresis to be in the IgG class and did not cross-react with lipid extracts of bovine rod outer segments. Papain-digested univalent antibodies (Fab) coupled with peroxidase were used to label rhodopsin in formaldehyde-fixed bovine and murine retinas. In addition to the disk membranes, the plasma membrane of the outer segment, the connecting cilium, and part of the rod inner segment membrane were labeled. We observed staining on both sides of the rod outer segment plasma membrane and the disk membrane. Discrepancies were observed between results of immunolabeling experiments and observations of membrane particles seen in freeze-cleaved specimens. Our experiments indicate that the distribution of membrane particles in freeze cleaving experiments reflects the distribution of membrane proteins. Immunolabeling, on the other hand, can introduce several different types of artifact, unless controlled with extreme care.