Selective radiolabeling of cell surface proteins to a high specific activity

A procedure was developed for selective radiolabeling of membrane proteins on cells to higher specific activities than possible with available techniques. Cell surface amino groups were derivatized with 125I-(hydroxyphenyl)propionyl groups via 125I-sulfosuccinimidyl (hydroxyphenyl)propionate (125I-sulfo-SHPP). This reagent preferentially labeled membrane proteins exposed at the cell surface of erythrocytes as assessed by the degree of radiolabel incorporation into erythrocyte ghost proteins and hemoglobin. Comparison with the lactoperoxidase-[125I]iodide labeling technique revealed that 125I-sulfo-SHPP labeled cell surface proteins to a much higher specific activity and hemoglobin to a much lower specific activity. Additionally, this reagent was used for selective radiolabeling of membrane proteins on the cytoplasmic face of the plasma membrane by blocking exofacial amino groups with uniodinated sulfo-SHPP, lysing the cells, and then incubating them with 125I-sulfo-SHPP. Exclusive labeling of either side of the plasma membrane was demonstrated by the labeling of some marker proteins with well-defined spatial orientations on erythrocytes. Transmembrane proteins such as the epidermal growth factor receptor on cultured cells could also be labeled differentially from either side of the plasma membrane.     

Externally disposed plasma membrane proteins. I. Enzymatic iodination of mouse L cells

The enzymatic iodination technique has been utilized in a study of the externally disposed membrane proteins of the mouse L cell. Iodination of cells in suspension results in lactoperoxidase-specific iodide incorporation with no loss of cell viability under the conditions employed, less than 3% lipid labeling, and more than 90% of the labeled species identifiable as monoiodotyrosine. 90% of the incorporated label is localized to the cell surface by electron microscope autoradiography, with 5-10% in the centrosphere region and postulated to represent pinocytic vesicles. Sodium dodecylsulfate-polyacrylamide gels of solubilized L-cell proteins reveals five to six labeled peaks ranging from 50,000 to 200,000 daltons. Increased resolution by use of gradient slab gels reveals 15-20 radioactive bands. Over 60% of the label resides in approximately nine polypeptides of 80,000 to 150,000 daltons. Various controls indicate that the labeling pattern reflects endogenous membrane proteins, not serum components. The incorporated 125-I, cholesterol, and one plasma membrane enzyme marker, alkaline phosphodiesterase I, are purified in parallel when plasma membranes are isolated from intact, iodinated L cells. The labeled components present in a plasma membrane-rich fraction from iodinated cells are identical to those of the total cell, with a 10- to 20-fold enrichment in specific activity of each radioactive peak in the membrane.     

Organisation of the proteins of the chromaffin granule membrane

The organisation of the protein components of bovine chromaffin granules has been investigated by labelling or digesting intact granules or broken membranes with the following reagents: lactoperoxidase/Na¹²⁵I as a reagent for tyrosine residues, $\text{N-}\text{(iodoacetylaminoethyl)-5-naphthylamine-1-sulphonic}$ acid as a reagent for cysteine residues, pronase, and galactose oxidase/KB³H₄. Following treatment, membranes were purified and washed and proteins were examined by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate. Rather more than 60 bands were resolved, of which about 40 were relatively intense and reproducible. The bands were classified according to their molecular weights and sensitivity to reagents. Penetration of the membranes by the reagents was assessed by examination of intragranular proteins.The majority of chromaffin granule membrane polypeptides became labelled when intact granules were treated with impermeant reagents. Eleven were probably protected in the intact granules, reactive sites becoming exposed only on membrane lysis. By contrast, carbohydrate moieties of glycoproteins appear to be exposed only on the matrix side of the membrane. Two proteins were shown to span the membrane, although this is probably an underestimate.     

The interaction of lymphocyte membrane proteins with the lymphocyte cytoskeletal matrix

Lymphocyte membrane proteins are important in the transduction of signals across the plasma membrane. Visual and biophysical studies have shown that after ligand binding, membrane proteins may become immobile in the plane of the membrane and may cap. In intact cells, binding of cross-linking ligands to surface immunoglobulin converts it to a detergent-insoluble state (77% insoluble). This conversion is positively correlated with the transmission of a mitogenic signal. Class II histocompatibility proteins (Ia) and thy-1 remain predominantly detergent soluble (60 to 97% soluble). Insolubilized membrane proteins may be solubilized by incubating the detergent insoluble cytoskeletons with 0.34 M sucrose, 0.5 mM ATP, 0.5 mM dithiothreitol, 1 mM EDTA, or 3 X 10(-5) M DNAase I, 1 mM EDTA. To determine if the membrane-associated cytoskeleton contains the sufficient components for ligand-induced receptor insolubilization, experiments were done with a crude plasma membrane fraction. The results with whole cells or crude plasma membranes were comparable. These studies support the view that ligand-induced insolubilization of membrane proteins is due to their interaction with cytoskeletal structures.     

HeLa cell plasma membranes : IV. Iodination of membrane proteins in synchronized cells

Intact HeLa cells and isolated HeLa cell plasma membranes were subjected to lactoperoxidase-catalysed iodination. The ¹²⁵I-labelled proteins were separated by SDS-polyacrylamide gel electrophoresis. Six protein species with apparent molecular weights from 32 000 to 200 000 were accessible to labelling from the outer cell surface, while most of the proteins present in the plasma membrane were labelled when isolated plasma membranes were iodinated. Iodination of synchronized intact cells revealed that the labelling obtained was cell cycle dependent with maximal labelling at mitosis. No changes in the distribution of radioactivity among the labelled proteins were observed when cells from different phases were iodinated.     

Iodination of plasma membrane proteins of BHK cells in different growth states

The surfaces of BHK cells in confluent monolayers, immediately after mechanical dispersal and in logarithmically growing suspension cultures have been iodinated with ¹²⁵I using the lactoperoxidase technique. Electrophoretic resolution of the labeled proteins revealed that the representation of plasma membrane proteins varies with the growth state. Trypsinization of the cells produced a drastic revision of the surfaces leaving behind root fragments of membrane components and exposing additional proteins for iodination. The rapid turnover of membrane proteins in growing BHK cells restored the plasma membrane to a state characteristic of the replicating cell within 10 h.     

Turnover of the plasma membrane proteins of hepatoma tissue culture cells.

The turnover of the plasma membrane proteins of hepatoma tissue culture cells was examined by three different methods--loss of polypeptides labeled in situ by lactoperoxidase-catalyzed iodination, loss of membrane polypeptides labeled with amino acid precursors, and loss from the membrane of fucose-labeled polypeptides. In both logarithmically growing and density-inhibited cells the proteins of the membrane are degraded with a half-life of about 100 hours. This is longer than the half-life of total cell protein, 50 to 60 hours, and longer than the doubling time of the cells, about 30 hours. Similar values for the rate of degradation of the membrane proteins were obtained by each of the three techniques. The same fucose-labeled polypeptides are present in the microsomal and the plasma membrane fractions of hepatoma tissue culture cells as analyzed by electrophoresis in dodecyl sulfate-acrylamide gels. But the fucose-labeled polypeptides were lost from the microsomal fraction at a faster rate than from the plasma membrane. Autoradiographic and double labeling techniques using 125I and 131I, or [3H]leucine and [14C]leucine were used to measure the relative rates of degradation of the proteins in the plasma membrane. All of the leucine-labeled polypeptides and the iodinated polypeptides had similar rates of degradation. These results support a model for the biogenesis of the plasma membrane in which the proteins are incorporated and removed in large structural units.     

Liposome-mediated transfer of integral membrane glycoproteins into the plasma membrane of cultured cells

Cultured mouse 3T3 cells treated with phosphatidylserine or phosphatidylserine/phosphatidylcholine (3: 7 mole ratio) liposomes containing ortho- and paramyxovirus envelope glycoproteins become susceptible to killing by virus-specific cytotoxic T lymphocytes indicating that the liposome-derived glycoproteins have been inserted into the cellular plasma membrane. Cells incubated with liposomes of similar lipid composition containing viral antigens plus a dinitrophenylated lipid hapten were killed by both virus- and hapten-specific T lymphocytes indicating that both protein and lipid components are inserted into the plasma membrane. We consider that assimilation of liposome-derived antigens into the plasma membrane results from fusion of liposomes with the plasma membrane. Cells incubated with phosphatidylcholine liposomes containing lipid haptens and viral glycoproteins were not killed by cytotoxic lymphocytes indicating that liposomes of this composition do not fuse with the plasma membrane. Liposome-derived paramyxovirus glycoproteins inserted into the plasma membrane retain their functional activity as shown by their ability to induce cell fusion. These experiments demonstrate the feasibility of using liposomes as carriers for introducing integral membrane (glyco)proteins into the plasma membrane of cultured cells and establish a new approach for studying the role of individual (glyco)proteins in the expression of specific cell surface properties.     

Kinetics of intracellular transport and sorting of lysosomal membrane and plasma membrane proteins

We have used monospecific antisera to two lysosomal membrane glycoproteins, lgp120 and a similar protein, lgp110, to compare the biosynthesis and intracellular transport of lysosomal membrane components, plasma membrane proteins, and lysosomal enzymes. In J774 cells and NRK cells, newly synthesized lysosomal membrane and plasma membrane proteins (the IgG1/IgG2b Fc receptor or influenza virus hemagglutinin) were transported through the Golgi apparatus (defined by acquisition of resistance to endo-beta-N-acetylglucosaminidase H) with the same kinetics (t1/2 = 11-14 min). In addition, immunoelectron microscopy of normal rat kidney cells showed that lgp120 and vesicular stomatitis virus G-protein were present in the same Golgi cisternae demonstrating that lysosomal and plasma membrane proteins were not sorted either before or during transport through the Golgi apparatus. To define the site at which sorting occurred, we compared the kinetics of transport of lysosomal and plasma membrane proteins and a lysosomal enzyme to their respective destinations. Newly synthesized proteins were detected in dense lysosomes (lgp's and beta-glucuronidase) or on the cell surface (Fc receptor or hemagglutinin) after the same lag period (20-25 min), and accumulated at their final destinations with similar kinetics (t1/2 = 30-45 min), suggesting that these two lgp's are not transported to the plasma membrane before reaching lysosomes. This was further supported by measurements of the transport of membrane-bound endocytic markers from the cell surface to lysosomes, which exhibited additional lag periods of 5-15 min and half-times of 1.5-2 h. The time required for transport of newly synthesized plasma membrane proteins to the cell surface, and for the transport of plasma membrane markers from the cell surface to lysosomes would appear too long to account for the rapid transport of lgp's from the Golgi apparatus to lysosomes. Thus, the observed kinetics suggest that lysosomal membrane proteins are sorted from plasma membrane proteins at a post-Golgi intracellular site, possibly the trans Golgi network, before their delivery to lysosomes.     

Immunoglobulin of T lymphoma cells. Biosynthesis, surface representation, and partial characterization.

Four cloned continuously cultured mouse T lymphoma cell lines, WEHI-22.1, WEHI-7.1, S49.1, and EL-4.1, were examined for immunoglobulin biosynthesis and the presence of immunoglobulin on the cell surface. Incorporation of [-3H]leucine into cellular proteins followed by serological analysis showed that immunoglobulin constituted between 0.1 and 1.1 percent of protein synthesized by the different cell lines during a 6-hr period. Under the same conditions cultured cells of nonlymphoid origin, the mastocytoma P-815 X-2.1, did not synthesize any detectable immunoglobulin. Lactoperoxidase-catalyzed radioiodination was used to label proteins on the surface of viable lymphoma and mastocytoma cells. Although the lymphoma lines lacked immunoglobulin as assessed by fluorescent antibody staining, immunoglobulin was detected in surface proteins of all four lymphoma lines. Estimates of the number of immunoglobulin molecules on the cell surface were 1.1 times 10-4/cell for S49.1 and EL-4.1, 1.7 times 10-4 for WEHI-7.1, and 4.3 times 10-4 for WEHI-22.1. Electrophoretic mobilities in sodium dodecyl sulfate polyacrylamide gel indicated that intact cell surface immunoglobulin was slightly larger than IgG, and on disulfide bond reduction to dissociate into two components, one with the mobility of serum immunoglobulin light chain, the other with a mobility similar to that of mu heavy chain. The heavy chain from the T lymphoma cells possessed an apparent molecular weight of about 65,000 compared with 70,000 for mu chain, although both chains shared antigenic determinants characteristic of mu chains. These findings are interpreted as support for other reports that T lymphocytes carry immunoglobulin on their surface and as direct evidence that thymus-derived lymphoid cells synthesize an immunoglobulin resembling the 7S subunit of IgM.     

Tubulin anchoring to glycolipid-enriched, detergent-resistant domains of the neuronal plasma membrane

After incubation of intact living cultured rat cerebellar granule cells at 37 degrees C with a new GM1 ganglioside analog, carrying a diazirine group and labeled with (125)I in the ceramide moiety, followed by photoactivation, a relatively small number of radiolabeled proteins were detected in a membrane-enriched fraction. A protein of about 55 kDa with a pI of about 5 carried a large portion of the radioactivity even if incubation and cross-linking were performed at 4 degrees C and in the presence of inhibitors of endocytosis, suggesting that it is cross-linked at the plasma membrane. Immunoprecipitation and Western blotting experiments showed the positivity of this protein for tubulin. Trypsin treatment of intact cells ruled out the involvement of a plasma membrane surface tubulin. Release of radioactivity from cross-linked tubulin after KOH treatment (but not hydroxylamine treatment) suggested that the photoactivated ganglioside reacts with an ester-linked fatty acid anchor of tubulin. Low buoyancy, detergent-resistant membrane fractions, isolated from cells after incubation with the GM1 analogue and photoactivation, proved their enrichment in endogenous and radioactive GM1 ganglioside, sphingomyelin, cholesterol, signal transduction proteins, and tubulin. It is noteworthy that radioactive tubulin was also detected in this fraction, indicating the presence of tubulin molecules carrying a fatty acid anchor in detergent-resistant, ganglioside-enriched domains of the plasma membrane. Parallel experiments carried out with a phosphatidylcholine analogue, also carrying a diazirine group and labeled with (125)I in the fatty acid moiety, showed the specificity of tubulin interaction with GM1. Taken together, these results indicate that some tubulin molecules are associated with a lipid anchor to detergent-resistant glycolipid-enriched domains of the plasma membrane. This novel feature of membrane domains can provide a key for a better understanding of their biological role.     

Sperm surface proteins persist after fertilization

Certain sperm components labeled with fluorescein isothiocyanate or its radioactive derivative, 125I-diiodofluorescein isothiocyanate (125IFC), are transferred at fertilization to the egg, where they persist throughout early cleavage stages at a localized site in the embryo cytoplasm (Gabel, C. A., E. M. Eddy, and B. M. Shapiro, 1979, Cell, 18:207-215; Gundersen, G. G., C. A. Gabel, and B. M. Shapiro, 1982, Dev. Biol., 93:59-72). By using image intensification we have extended these observations in the sea urchin to the pluteus larval stage, in which greater than 60% of the embryos have localized fluorescent sperm components. Because of the unusual persistence of the sperm components in the embryo, a characterization of the nature of the labeled species in sea urchin sperm was undertaken. Approximately 10% of the 125IFC was in sperm polypeptides of Mr greater than 15,000. These proteins were on the sperm surface as shown by their sensitivity to externally added proteases. The remainder of the 125IFC in sperm was in several low- molecular-weight species, none of which was 125IFC-derivatized phospholipid. To determine if any labeled sperm polypeptides remained intact in the embryo after fertilization, 125IFC-labeled sperm proteins were recovered from one-cell and late gastrula stage embryos by using an anti-IFC immunoadsorbent. Most of the labeled sperm proteins were degraded shortly after fertilization; however, distinct sets of labeled polypeptides were recovered from both one-cell and gastrula stage embryos. Six of the labeled polypeptides recovered from both embryonic stages had identical SDS gel mobilities as labeled sperm polypeptides. Other polypeptides in the embryos appeared to arise from limited proteolysis of sperm proteins. Thus, in this physiological cell fusion system, individual sperm proteins are transferred to the egg at fertilization, and some persist intact or after specific, limited degradation long after gamete fusion, until at least the late gastrula stage.     

Fate of surface proteins of rabbit polymorphonuclear leukocytes during phagocytosis. I. Identification of surface proteins

To study the fate of external membrane proteins during phagocytosis, rabbit peritoneal neutrophils were labeled by enzymatic iodination. Iodine was incorporated into at least 13 proteins ranging in size from approximately 250,000 to 18,000 daltons as judged from autoradiography of gels after SDS-polyacrylamide gel electrophoresis of labeled cells. The major contractile proteins of neutrophils, actin and myosin, were not labeled when intact cells were iodinated but were labeled when homogenates of these cells were iodinated. Nine of the iodinated proteins were released by mild protease treatment of intact cells. A plasma membrane-rich fraction was isolated by density centrifugation. This fraction was enriched at least 10-fold for lactoperoxidase-labeled acid-insoluble proteins. It was enriched to the same extent for the presence of iodinated wheat germ agglutinin that had been bound to intact cells at 4 degrees C before homogenization. Analysis of SDS-polyacrylamide gel electrophoresis revealed that the proteins of this fraction were predominantly of high molecular weight. However, only 8 of the 13 proteins iodinated on intact cells were found in this fraction. The remaining five were enriched in a dense fraction containing nuclei, intact cells, and membranous vesicles, and may represent a specialized segment of the neutrophil cell surface.     

Depolymerization of microtubules alters membrane potential and affects the motional freedom of membrane proteins

Two independent lines of evidence were obtained indicating that microtubule depolymerization affects the functions and the physical state of membranes in intact Chinese hamster ovary cells. The first type of evidence was obtained by using the dye dihexyloxacarbocyanine iodide to measure membrane potential before and after treatment with several microtubule active agents. Microtubule depolymerization resulted in a decrease in cell fluorescence, whereas stabilization of microtubules with taxol resulted in an increase in cell fluorescence. These effects of the drugs were due to their interactions with microtubules and not to direct effects of the drugs on the plasma membranes for the following reasons: effects were time dependent and required entry into the cells as indicated by the lack of fluorescence change in a multi-drug-resistant mutant that does not accumulate antimicrotubule drugs and a colcemid-resistant tubulin mutant did not show these effects on cell fluorescence. Evidence for altered motional freedom of membrane proteins in the plasma membrane was obtained by using electron spin resonance analysis of maleimide spin probe labeled cells. This study showed that depolymerization of microtubules results in increased motional freedom of maleimide-labeled sulfhydryl group containing proteins. Taken together, these data argue that microtubules function in mammalian cells to regulate the physical state of membranes and modulate membrane potential generated across cell membranes.     

Antibodies against the plasma membrane 3,3',5-triiodo-L-thyronine binding protein of rat pituitary GH3 cells: partial characterization and cross-species immunoreactivity

To develop antibodies against the plasma membrane 3,3',5-triiodo-L-thyronine (T3) binding protein (M.W. 55,000), rabbits were immunized with formalin-fixed GH3 cells or highly purified plasma membranes from these cells. Antibodies were screened by immunoprecipitation using detergent solubilized N-bromoacetyl-[125I]T3-labeled 55K protein. Among the nine detergents tested, 0.18% CHAPS was found to be the best in its solubilization efficiency and its ability to maintain the integrity of the antigenicity of the 55K protein. The N-bromoacetyl-[125I]T3-labeled 55K protein was also immunoprecipitated by anti-T3 antibodies. The anti-55K protein antibodies cross-reacted with plasma membrane T3 binding proteins from cultured cells and tissues of human and rodent origin. These results indicate that structural similarities exist in human and rodent plasma membrane T3 binding proteins. These antibodies should provide a powerful tool in the characterization and in probing the function(s) of the plasma membrane T3 binding protein in cells.     

Lymphoma-vesicle interactions: vesicle adsorption, membrane fragmentation, and intermembrane protein transfer

Sonicated dimyristoylphosphatidylcholine vesicles interact with cultured murine lymphoma (BL/VL3) to generate complexes of vesicle and cell membrane components. Cell-free supernatants harvested after cell-vesicle incubations contain three distinct lipid species that can be separated by density gradient centrifugation. Analysis of protein and lipid composition and assays for cell and vesicle lumen contents reveal that the densest of the three lipid species comprises sealed plasma membrane fragments complexed with vesicles, while the least dense species is indistinguishable from pure phospholipid vesicles. The third, intermediate density species consists of topologically intact vesicles with associated plasma membrane proteins but without detectable cell lipids or cytoplasmic components. The membrane fragmentation and cell-to-vesicle protein transfer observed during lymphoma-vesicle incubations are examined as functions of cell and vesicle concentrations and incubation time.     

Characterization of calpain I-binding proteins in human erythrocyte plasma membrane

The calpain-binding components on the plasma membrane were characterized using calpain I. 125I-labeled calpain was bound to inside-out membrane vesicles from human erythrocyte in a Ca2(+)-dependent manner, but not to right-side-out membrane vesicles. The maximum binding was observed at more than 5 microM Ca2+. The binding amount of calpain to the inside-out membrane vesicles was decreased when the vesicles were pretreated with 100 micrograms/ml of trypsin, chymotrypsin, elastase, or pronase P for 30 min at 37 degrees C, although the binding to the vesicles pretreated with 200 micrograms/ml of phospholipase A2 or C was not affected. Calpain-binding proteins in the membrane were analyzed by using a modified immunoblotting for calpain. Immunostained bands of 240, 220, 89, 72, 52, and 36 kDa were detected as the calpain-binding proteins in the native membrane. All of these bands had disappeared in trypsin-treated membrane. The disappearance of bands was dose-dependent with respect to trypsin and paralleled the reduction of binding amount of calpain to the trypsinized membrane. In calpain-treated membrane, the 240 and 36 kDa bands were retained in the blotting, though the other bands disappeared dose-dependently with respect to calpain. These results suggested that the significant component in the inner surface of plasma membrane for binding of calpain was proteinaceous and the calpain-binding proteins could be classified into two species, i.e. substrates of calpain (220, 89, 72, and 52 kDa protein) and non-substrates (240 and 36 kDa protein).     

Characterization of the mycoplasma membrane proteins. V. Release and localization of membrane-bound enzymes in Acholeplasma laidlawii.

The peripheral membrane protein fraction released by washing Acholeplasma laidlawii membranes with low-ionic strength buffers contained about 50% of the total membrane-bound ribonuclease and deoxyribonuclease activities. The ATPase, NADH oxidase and p-nitrophenylphosphatase activities remained bound to the membrane even when EDTA was added to the wash fluids, and thus appear to belong to the integral membrane protein group. Serving as a marker for peripheral membrane proteins, the membrane-bound ribonuclease activity was solubilized by bile salts much more effectively than the integral membrane-bound enzymes. On the other hand, the solubilized ribonuclease showed a much lower capacity to reaggregate with other solubilized membrane components to membranous structures. Yet, most of the ribonuclease molecules which were bound to the reaggregated membranes could not be released by low-ionic strength buffer. The reaggregated membranes differed from the native membranes in the absence of particles on their fracture faces obtained by freeze cleaving, and by their much higher labeling by the [125-I]lactoperoxidase iodination system. These results suggest that most of the proteins are exposed on the reaggregated membrane surfaces, with very little, if any, protein embedded in its lipid bilayer core. Enzyme disposition in the A. laidlawii membrane was studied by comparing the activity of isolated membranes with that of membranes of intact cells after treatment with pronase or with an antiserum to membranes. The data indicate the asymmetrical disposition of these activities, the ATPase and NADH oxidase being localized on the inner membrane surface, while the nucleases are exposed on the external membrane surface.     

Lactoperoxidase-catalyzed iodination of surface membrane lipids

Lactoperoxidase-catalyzed iodination of intact cells, is known to label predominantly, if not exclusively, the exposed tyrosine residues of cell surface proteins. The present study demonstrates that during this iodination process surface membrane lipids are also iodinated through an enzyme-dependent step. Phosphatidylcholine, cholesterol-phosphatidylcholine liposomes and confluent secondary cultures of chick embryo cells were iodinated by the lactoperoxidase-glucose oxidase-glucose [¹²⁵I] procedure. Liposomes were efficiently labeled. In the cells, 20–30% of the radioactivity was found in proteins and 20–30% in the lipids. Both neutral and polar lipids were found to bind [¹²⁵I] covalently. Controls in which lactoperoxidase was omitted showed < 6% of the radioactivity found in liposomes or cells labeled with the enzyme.     

Transbilayer mapping of membrane proteins using membranes isolated on polylysine-coated polyacrylamide beads

Erythrocyte and HeLa cell plasma membranes were isolated on polylysinecoated polyacrylamide beads and the transbilayer disposition of their proteins was investigated.When membranes of intact erythrocytes were isolated on beads and then labelled by lactoperoxidase-catalysed iodination, their labelling pattern was similar to that of inside-out vesicles in solution.When the membranes of intact HeLa cells were isolated on beads and then labelled by galactose oxidase-[³H]borohydride treatment, no glycoprotein or glycolipid sugars were accessible. On the other hand, when the HeLa cell membranes were isolated on beads and then labelled by the lactoperoxidase-catalysed iodination, all of the major membrane proteins were iodinated. These experiments confirmed for HeLa cell membranes what had previously been shown for erythrocyte membranes: when the membranes of intact cells are isolated on beads, the accessibility of their surfaces to enzymatic probes is the same as would be expected of inside-out vesicles in suspension. Double-label experiments, in which the HeLa cell membranes were labelled first on the intact HeLa cells and again after isolation on beads, identified several