Carbohydrate Composition of Bovine Rhodopsin”

The carbohydrate content of bovine rhodopsin was investigated and found to be different from previously reported values. Rod outer segments were isolated from dark-adapted bovine retinas by sucrose flotation and purified by sucrose density centrifugation. Rhodopsin was extracted with detergents and purified by chromatographic procedures involving calcium phosphate/celite chromatography followed by affinity chromatography on concanavalin A-Sepharose (or in some cases, gel filtration on agarose). Purified preparations of rhodopsin had A218/A,98 ratios of 1.6 to 2.0. After treatment of the rhodopsin with chloroform/methanol (2/l) to remove lipids and detergents, the carbohydrate content was measured by gas-liquid chromatography, calorimetric and enzymatic analyses, paper chromatography, and electrophoresis. Rhodopsin was found to have about 9 mol of manriose and 5 mol of glucosamine per mol of visual pigment. A molar ratio of mannose/glucosamine of about 2 was also found in samples of rhodopsin obtained from two other laboratories. The amino acid analysis was similar to previously published values. visual pigment for the

It has been estimated that visual pigment accounts for the bulk of the protein of the discs in the rod outer segments of the mammalian retina (1). Heller (2) and Heller and Lawrence (3) have reported that the bovine visual pigment, rhodopsin, is a glycoprotein containing mannose and glucosamine. Their analyses showed the presence of 3 mol of each of these two sugars per mol of rhodopsin. Apart from those reports, little information has been published concerning the carbohydrates of this molecule. Although the detailed structure of the carbohydrate groups in rhodopsin is not known, other glycoproteins that contain a single carbohydrate chain composed of this amount of mannose and glucosamine have not been described as yet. The present report is a reinvestigation of the carbohydrate content of bovine rhodopsin. While our amino acid analysis was similar to that reported previously (2, 4-6).
the quantitative carbohydrate composition of the constituent sugars, mannose and glucosamine, was different, and is more in accord with that described for other glycoproteins containing only these two sugars.
* This work was supported by Public Health Service Research Grant EY 00393 from the National Eye Institute, United States Public Health Service, and the Ohio Lions Eye Research Foundation. A preliminary report of these findings was presented at the Spring, 1975  when compared to the crude preparation. These values are minimal since they refer only to detergent-extractable protein. When Con A-Sepharose is used, there appears to be little advantage in the use of the calcium phosphate/celite step. In addition to removing extraneous protein, the affinity column also removed non-rhodopsin carbohydrate components ( Table  I) that were not separated from rhodopsin by calcium phosphate/celite.* Absorption Spectra, SDS Polyacrylamide Gel Electrophoresis, Purity-Bovine rhodopsin, purified by these procedures, had the typical absorption spectrum (5) in the native state and after bleaching (Fig. 2). These same spectral properties were observed in all of the fractions from the purified rod outer segments through the remaining steps in the fractionation procedure outlined in Table I The spectra of this preparation in 0.05 M Tris-HCI, pH 7.0, containing 0.3% Ammonyx LO was obtained before (-) and after (---) photobleaching.
phoresis of rhodopsin purified by these procedures showed patterns similar to that described previously for purified rhodopsin (7). Under the conditions used3 a molecular weight of about 35,000 can be calculated for rhodopsin. The criteria for the purity of the rhodopsin used in these studies (obtained by well established procedures of rod purification, extraction of rhodopsin, and chromatographic techniques) were: the low A278/A,08 ratios, the spectral properties before and after bleaching, and the results of SDS-polyacrylamide gel electrophoresis. The procedures used here produced a product whose purity was as high as or higher than that published klsewhere for bovine rhodopsin.

Chloroform/Methanol
Treatment; Amino Acid Analysis-It was not possible to analyze the carbohydrate content of rhodopsin directly by GLC because of interference by the detergents present in the buffers and the lipids associated with this molecule. The interfering substances were effectively removed by treating rhodopsin preparations with 20 volumes of 'Samples containing 0.25 to 1.0 nmol of rhodopsin were preincubated with 2.5% SDS (7) and applied to gels containing 5.6% acrylamide.
by guest on March 24, 2020 http://www.jbc.org/ Downloaded from chloroform/methanol (2/l). Rhodopsin is insolubilized by this process. The distribution of protein was determined by a complete amino acid analysis of the starting material (purified rhodopsin, in the presence of detergent lipids, and buffer) and of the soluble and insoluble fractions after treatment with chloroform/methanol (2/l). Glucosamine was also measured in these same fractions by means of the amino acid analyser. As can-be seen in Table II, there was essentially complete recovery of protein and glucosamine in the precipitate that was formed.
The amino acid analysis of rhodopsin purified by these techniques is presented in Table III, and is similar to that reported by several other laboratories (2,(4)(5)(6). The data presented in Table III are average values from hydrolyses carried out for 24, 48, and 72 hours, not extrapolated to zero time since an increase of about 9% in the yield of total amino acids was observed over this period of time (data not shown). Presented also are calculations based on residues per 100 residues. There is good agreement between these values and the average value of several reports in the literature. Tryptophan was not determined but a value of 5 residues/m01 has been reported elsewhere (2).
Carbohydrate Analyses: GLC, Enzymatic, Colorimetric-After rhodopsin was treated with chloroform/methanol, mannose and glucosamine were the only sugars detected in the insoluble residue. Table I shows analyses of a representative study from various stages in the purification of rhodopsin. About 34% of the total mannose and glucosamine and 45% of the rhodopsin present in the purified rod outer segments were recovered in the final purified product. At this stage there were approximately 9 mol of mannose and 5 mol of glucosamine per mol of rhodopsin. A molar ratio of mannose to glucosamine of about 2 was observed also at each step after the isolation of the purified rod outer segments. Additional fractionation of rhodopsin on columns of agarose (see under "Materials and Methods") did not alter the carbohydrate analysis (data not shown).
A summary of carbohydrate analyses of rhodopsin purified by chromatography on columns of calcium phosphate/celite followed by Con A-Sepharose is presented in Table IVA. Given also in this table are analyses for mannose and glucosamine by means of different techniques of measurement. Relatively good agreement was obtained among the several diverse methods, indicating the general reliability of the measurements. Data from GLC (a procedure common for both of the sugars) indicated the presence of about 9 mol of mannose and 5 mol of glucosamine per mol of bovine rhodopsin. As a control, analysis of the carbohydrate content of ovalbumin by these techniques' showed the presence of mannose and glucosamine in amounts similar to those reported by other authors (23).
Heterogeneity of the carbohydrate components of glycoproteins is a widely observed phenomenon, and may explain some of the variation observed in these investigations. The data in Table IVA were obtained for rhodopsin purified by a rigorously controlled procedure, described under "Materials and Methods." In preliminary experiments, many variations were tried, such as the nature of the buffers, types and concentrations of detergents, different procedures for isolating the rod outer segments, and additions and deletions to the purification process. When the GLC data for all of these ' Ovalbumin (five times recrystallized, Nutritional Biochemicals, Cleveland, Ohio) was shown to contain (moles/mol of glycoprotein): mannose, 4.6 (GLC), 4.4 (enzymatic), 4.9 (anthrone); glucosamine, 3.2 (Elson-Morgan), 3.0 (GLC).  preparations (some of which had A~,~/A,~~ ratios greater than the preparations presented in Table IVA) were combined with those presented in Table IV the following values were calculated from 13 separate preparations of rhodopsin (moles/mol of rhodopsin * SE.): mannose, 7.7 + 0.59; glucosamine, 4.1 + 0.38. The molar ratio of mannose/glucosamine was 1.9 * 0.08.
It is perhaps not surprising that these procedural variations as well as the use of different batches of eyes over a period of over 2 years from cattle not reared under controlled laboratory conditions may have resulted in rhodopsin molecules showing somewhat different amounts of components present in relatively minor concentrations. However, in spite of these differences, samples of rhodopsin obtained by a variety of isolation and purification procedures contain amounts of mannose and glucosamine that differ substantially from the values reported previously (2, 3) with respect to both the absolute amounts of these two sugars and their molar ratio.
Purified rhodopsin was also obtained from two other labora-