Monosaccharide Transport in Protein-depleted Vesicles from Erythrocyte Membranes*

Treatment of human erythrocyte membranes with dilute alkali (pH 11.5) generates sealed, protein-depleted vesicles that can be isolated by density gradient centrifugation. The vesicles are 0.5 to 2.0 micrometers in diameter, and their membranes are predominantly oriented inside-out. The vesicles lack protein bands 1, 2, 5, and 6 (nomenclature of Steck, T.L. (1974) J. Cell Biol. 62, 1-19) of the erythrocyte membrane. L-Sorbose, a substrate of the monosaccharide transport system in erythrocytes, is transported by the vesicles. Based on comparisons between erythrocytes and vesicles with regard to specificity, temparture dependence, and effects of inhibitors, we conclude that sorbose uptake into the vesicles occurs by way of the monosaccharide transport system. The specific activity of the transport system in vesicles, as determined by initial rate measurements of sorbose uptake, averaged 58% of that in erythrocytes. This finding indicates that the major polypeptides of Bands 1, 2, 5, and 6 do not play an obligatory role in monosaccharide transport.

we conclude that sorbose uptake into the vesicles occurs by way of the monosaccharide transport system. The specific activity of the transport system in vesicles, as determined by initial rate measurements of sorbose uptake, averaged 58% of that in erythrocytes.
This finding indicates that the major polypeptides of Bands 1, 2, 5, and 6 do not play an obligatory role in monosaccharide transport.
The monosaccharide transport system of the human erythrocyte catalyzes the equilibration of n-glucose and other sugars across the cellular membrane. A great deal is known about the kinetics, specificity, and inhibition of this system of facilitated diffusion (l-3). However, neither the identity of the erythrocyte membrane protein that is the major component of this system nor the possible role(s) of other erythrocyte membrane proteins in modulating the activity of the system has been, as yet, unequivocally determined (2,3). As an approach to this problem, we have made use of the finding of Steck and Yu that treatment of erythrocyte membranes (ghosts) with dilute alkali selectively elutes four of the major proteins ( Bands 1,2,5,and 61 (4). We have isolated sealed vesicles from the membrane pellet after alkali treatment and compared the activity of the monosaccharide transport system in these vesicles with that in intact erythrocytes. were added with stirring to 280 ml of water, and the pH raised to 11.5-11.6 by the addition of about 0.4 ml of 1 N NaOH. This operation was performed with a pH meter that had been carefully standardized at pH values of 9.39 and 12.00 through the use of 10 rnM K,B,O, .5H,O and 25 rnM sodium phosphate buffers, respectively, at 5" (6). Immediately after adjustment of the pH, the suspension was centrifuged at 40,000 x g for 20 min. The supernatant was discarded and the pellet was resuspended in 40 ml of 2.6 rnM sodium phosphate, pH 8.0. The pH was readjusted to 8.0 with 2.6 rnM disodium phosphate and the membrane fragments were separated again by centrifugation.
Vesicles were isolated from the pellet by density gradient centrifugation after the method of Steck for the isolation of sealed erythrocyte membrane vesicles (7). The pellet was resuspended to give 9 ml in 2.6 rnrvr sodium phosphate, pH 8, and 1.5-ml aliquots were layered on top of 11.3 ml of a 1.00 to 1.06 g/cm3 linear gradient of dextran T-70 in 0.5 rnM sodium phosphate, pH 8.0. Centrifugation was carried out for 90 min at 37,000 rpm in a Beckman SW41 Ti rotor. The protein-depleted vesicles were collected from the gradient and washed two times by suspension in 40 ml of 2.6 rnM sodium phosphate, pH 8, followed by centrifugation at 48,000 x g for 15 min. All the above operations were carried out at O-3". The vesicles were finally adjusted to a protein concentration of 1.4 mglml in 2.6 mM sodium phosphate, pH 8, and stored at 3". SDS'-polyacrylamide gel electrophoresis of the vesicles, after storage for various periods at 12.5", revealed the disappearance of less than 10% of Band 3 after 24 h, of about 20% after 53 h, and of 60% after 78 h, probably as the result of the action of a protease in the preparation (8 (c) Vesicles that had equilibrated with L-114C1sorbose were isolated by centrifugation as well as by Millipore filtration. The two methods gave the same value for the amount of sorbose within the vesicles, after correction for the extravesicular space in the pellet through the use of the slowly permeant compound myo-PHlinositol as a marker.
(d) Filters from assays performed at 10% and 100% equilibration of the sorbose were shaken with 0.2 ml of 3 mM Lsorbose/l% Triton X-100. The extracts were subjected to thin layer chromatography on silica gel plates with ethanol/water (4/l by volume) as the solvent (9). Ninety per cent or more of the radioactivity was recovered at the position of sorbose. Generally this procedure was used to determine the initial rates of uptake by sampling at live time intervals over the period required for about 10% equilibration of the sorbose (see Fig. 3). In order to determine the amount of sorbose taken up at equilibrium, the assay mixtures were subsequently left at 25" for 14 to 18 h and sampled again. The percentage of the total radioactivity that was bound to the filter rose from about 0.2% at zero time to 0.4 to 0.8% during the measurement of the initial rates to 4 to 6% at equilibrium. The initial rate of sorbose uptake (micromoles/min) decreased by a factor of 2 when the concentration of proteins was halved and increased by a factor of 2 when the sorbose concentration was doubled. Since the monosaccharide transport system of erythrocytes is inhibited by thiol reagents (3) and so could be inactivated by disulfide formation, one preparation of vesicles was made and assayed in the presence of 1 mM dithiothreitol.
Its initial rate of sorbose uptake did not differ significantly from that for vesicles prepared from the same batch of ghosts and assayed in the absence of dithiothreitol.
In some experiments the uptake of myo-inositol as well as Lsorbose was followed by including 0.5 mM myo-["Hlinositol (10 FCi) in the assay mixture and measuring both the 3H and 14C bound to the filters. Here the quench solution contained 0.50 rnM myo-inositol.

Rates of Transport
into Whol e Cells-The uptake of sorbose by intact erythrocytes was measured by a modification of the procedure described by Levine et al. (10). Erythrocytes were washed at room temperature three times with 6 volumes of 150 mM NaC1/5 mM sodium phosphate, pH 8.0, by suspension and centrifugation. The uptake of sorbose was initiated by mixing an aliquot of L-l'4C1sorbose in 150 rnM NaCl, 5 rnM sodium phosphate, pH 8, with a suspension of cells that was equilibrated at the desired temperature in a water bath. The mixtures contained 5.9 mM L-['4Clsorbose in the medium initially and 25% cells (v/v). At various time intervals 1.67-ml aliquots were withdrawn and mixed with 9 ml of ice-cold 1% NaCl, 2 rnM HgCl?, 1.25 mM KI (stopping solution) in a 12-ml glass centrifuge tube. The cells were sedimented by centrifugation, the supernatant was carefully aspirated, and the centrifuge tube was washed through the addition and removal of 10 ml of stopping solution without disturbing the pellet. After the last traces of supernatant had been taken off with a filter paper wick, the pellet was mixed in a Vortex with 1 ml of 10% trichloroacetic acid. The precipitated protein was packed by centrifugation, and aliquots of the supernatant were counted.
The initial rate of sorbose uptake was determined at 12.5" by taking six aliquots at 2-min intervals after sorbose addition. The plots of sorbose in the cell pellet against time were linear and rose from about 4% to 12% of the value at equilibrium, which was obtained from aliquots of the assay mixture that were kept at 37" for 3 h before stopping.
Duplicate determinations of the rate agreed within ?5% of the average value. For these experiments, the number of cells per ml was carefully measured in order that the initial rates of uptake per mg of membrane protein could be calculated. For the determination of the activation energy for sorbose transport, it proved more convenient to follow the uptake of sorbose by assaying at six time intervals over the period required for 90% equilibration. This data gave linear first order plots from which the first order rate constants for equilibration were calculated (see Ref. 10).
Other Methods-The number of vesicles per mg of protein was determined by counting both the vesicles and beads in an admixture of vesicles with a known number of l.Ol-pm polystyrene beads (Dow Chemical Co.), through the use of a Zeiss Universal microscope (11).
The vesicular volume per mg of protein was calculated from the amount of sorbose within the vesicles at equilibrium. Erythrocytes and ghosts were counted with both a hemacytometer and a ZBI Coulter Counter; the two methods agreed within 5%. SDS-polyacrylamide gel electrophoresis was performed as described by Steck and Yu (4). Protein was routinely measured by the Lowry method with crystallized, lyophilized bovine serum albumin (Sigma) as the standard (12). The protein contents of ghosts and vesicles were also determined by amino acid analysis after hydrolysis for 24 and 48 h in 6 N HCl, 0.1% phenol at 105", with norleucine as an internal standard. These values were 76% and 69% of those found for ghosts and vesicles, respectively, in the Lowry assay. The values for protein reported herein are given on the basis of amino acid content. Acetylcholinesterase activity was measured as described by Steck and Kant (5). Phospholipids were extracted with chloroform/methanol (2:l) (13), and total phosphorus in the extract was determined by the method of Bartlett (14).  The vesicles were found to take up sorbose to an equilibrium level over a period of several hours (Fig. 2). Fig. 3 illustrates data for the initial rates of this uptake. The following findings provide evidence that entry into the vesicles occurs by the monosaccharide transport system.

Second, an Arrhenius
plot of the initial rates of sorbose uptake at the various temperatures (Fig. 3) gives an activation energy of 36 kcabmol. This value is similar to the value of 39 kcabmol that we have obtained from the temperature dependence of the first order rate constant for equilibration of sorbose with intact erythrocytes. Our values for this rate constant are 0.91,2.0,4.0, 6.9, and 12 x lo-" min' at 5.4, 8.3, 10.5, 12.5, and 15.5", respectively. First, sorbose enters the vesicles at least 10 times more rapidly than the cyclic hexahydroxy compound myo-inositol (Fig. 2). Erythrocytes exhibit the same specificity (16, 17).
Third, n-glucose decreases the initial rate of sorbose uptake (Table II). On the assumption that glucose and sorbose compete for the transport system, the expression for the dissocia-  Effects ofD-glucose, cytochalasin B, and phloretm on initial rates ofLsorbose uptake by protein-depleted vesicles, at 12.5" and pH 8.0 In each case the initial rates in the presence and absence of inhibitor were determined with the same preparation of vesicles. Each compound was added as a small aliquot from a concentrated solution in buffer, and was included at the same final concentration in the quench solution (see under "Experimental Procedures"). In the case of n-glucose, the vesicles were incubated with and without glucose for 6 h at 12.5" before the addition of sorbose. This period is sufficient for complete equilibration of the glucose (19). All initial rates were measured in duplicate, and the duplicates agreed within +lO% of the average.  Table  I and "Results"). Similar increases in activity, expressed per mg of the specific protein, are demanded for Bands 2, 5, and 6 (see Table I  Also, the vesicles may serve as a convenient starting material for purification of the transport system in membranous form. One possible approach is fusion of the vesicles with excess phospholipid followed by sonication to yield vesicles of 300 to 500-A diameter that have, on the average, only one protein per vesicle (32,33). The vesicles containing the transport system would then be separated from the others using some property of this system, such as its affinity for glucose derivatives.