Glycoprotein synthesis in the adult rat pancreas: II. Characterization of Golgi-rich fractions

Golgi-rich fractions were prepared from homogenates of adult rat pancreas by discontinuous gradient centrifugation. These fractions were characterized by stacks of cisternae associated with large, irregular vesicles and were relatively free of rough microsomes, mitochondria, and zymogen granules. The Golgi-rich fractions contained 50% of the UDP-galactose: glycoprotein galactosyltransferase activity; the specific activity was 12-fold greater than the homogenate. Such fractions represented < 19% of thiamine pyrophosphatase, uridine diphosphatase, adenosine diphosphatase, and Mg²⁺-adenosine triphosphatase. Zymogen granules and the Golgi-rich fractions were extracted with 0.2 m NaHCO3, pH 8.2, and the membranes were isolated by centrifugation. The glycoprotein galactosyltransferase could not be detected in granule membranes, while the specific activity in Golgi membranes was 25-fold greater than the homogenate.At least 35 polypeptide species were detected in Golgi membranes by polyacrylamide gel electrophoresis in 1% sodium dodecylsulfate. These ranged in molecular weight from 12,000 to <160,000. There were only minor differences between Golgi membranes and smooth microsomal membrane. In contrast, zymogen granule membranes contained fewer polypeptides. A major polypeptide, which represented 30–40% of the granule membrane profile, accounted for less than 3% of the polypeptides of Golgi membranes or smooth microsomal membranes.     

Distribution of protein-bound sugar residues in microsomal subfractions and Golgi membranes

Liver microsomal subfractions and Golgi membranes free from adsorbed and secretory proteins have a characteristic sugar composition. The ratio of mannose to galactose is largest in rough microsomes, smaller in smooth I microsomes, still smaller in smooth II microsomes, and smallest in Golgi membranes. There is about twice as much glucosamine in Golgi membranes and 3 times as much in smooth II microsomes as in the other microsomal subfractions. Golgi membranes are rich in sialic acid in comparison to rough microsomes and it is present at even higher levels in the two smooth microsomal subfractions. Increasing concentrations of deoxycholate preferentially remove protein-bound mannose and glucosamine, while releasing significantly less galactose. About half of the microsomal mannose and galactose can be liberated from the surface of intact microsomal vesicles by treatment with trypsin. When trypsin is added to permeable vesicles where the inside surface can be also attacked, an additional 20% of the total mannose but no additional galactose is liberated.     

Distribution of protein-bound sugar residues in microsomal subfractions and Golgi membranes.

Liver microsomal subfractions and Golgi membranes free from adsorbed and secretory proteins have a characteristic sugar composition. The ratio of mannose to galactose is largest in rough microsomes, smaller in smooth I microsomes, still smaller in smooth II microsomes, and smallest in Golgi membranes. There is about twice as much glucosamine in Golgi membranes and 3 times as much in smooth II microsomes as in the other microsomal subfractions. Golgi membranes are rich in sialic acid in comparison to rough microsomes and it is present at even higher levels in the two smooth microsomal subfractions. Increasing concentrations of deoxycholate preferentially remove protein-bound mannose and glucosamine, while releasing significantly less galactose. About half of the microsomal mannose and galactose can be liberated from the surface of intact microsomal vesicles by treatment with trypsin. When trypsin is added to permeable vesicles where the inside surface can be also attacked, an additional 20% of the total mannose but no additional galactose is liberated.     

Activity of acyl-CoA: cholesterol acyltransferase and 3-hydroxy-3-methylglutaryl-CoA reductase in subfractions of hepatic microsomes enriched with cholesterol

The influence of membrane cholesterol on the activities of acyl-CoA: cholesterol acyltransferase and 3-hydroxy-3-methylglutaryl-CoA reductase was examined in three microsomal subfractions (RNA-rich, RNA-poor, and smooth) that had been enriched with cholesterol by incubation with mixed lipoproteins from hypercholesterolemic rabbit serum. Acyl-CoA: cholesterol acyltransferase activity was significantly stimulated in the three subfractions, particularly in the RNA-rich microsomal component. 3-Hydroxy-3-methylglutaryl-CoA reductase, on the other hand, was suppressed (30%) in only one (RNA-poor) of the three microsomal subfractions, despite a 1.4-fold increase in the concentration of membrane cholesterol. An attempt was made to distinguish between an effect based exclusively on an increase in available cholesterol substrate and an activation of acyl-CoA: cholesterol acyltransferase in RNA-rich microsomes enriched with cholesterol. An experimental design was devised so that substrate cholesterol was provided in the form of heated smooth microsomes and acyl-CoA: cholesterol acyltransferase was provided as a separate preparation in the form of RNA-rich microsomes. Appropriate controls were carried out to test for transfer of cholesteryl ester between the two sets of particles. The results suggested that cholesterol enhanced acyl-CoA: cholesterol acyltransferase activity by serving both as a substrate and as a non-substrate modulator.     

Isolation of plasma membranes and Golgi apparatus from a single chicken liver homogenate

Chicken liver plasma membranes, minimally contaminated with Golgi apparatus-derived vesicles, were prepared from a low-speed (400 g) pellet by means of flotation in isotonic Percoll solution, followed by a hypotonic wash and flotation in a discontinuous sucrose gradient. Based on the analysis of suitable marker enzymes, alkaline phosphatase and alkaline phosphodiesterase, two plasma membrane fractions were isolated with enrichments, depending on the equilibrium density and marker of 28-97 and with a total yield of 4-5%. Golgi apparatus fractions were prepared by flotation of microsomes, obtained from the same homogenate as the low-speed pellet, in a discontinuous sucrose gradient. The trans-Golgi marker galactosyltransferase was 27-fold enriched in a fraction of intermediate density (d=1.077-1.116 g/ml). Approximately 12% of galactosyltransferase was recovered in the membranes equilibrating d=1.031-1.148 g/ml. Contamination with plasma membrane fragments was low in the light (d=1.031-1.077 g/ml) and intermediate density Golgi vesicles. The isolation of purified plasma membranes and Golgi vesicles from one liver homogenate will enable future studies on receptor cycling between these cell organelles.     

CHARACTERIZATION AND DEVELOPMENTAL CHANGES OF UDP-GALACTOSE-CERAMIDE GALACTOSYL TRANSFERASE IN A RAT CNS AXOLEMMA-ENRICHED FRACTION. DIFFERENCES AND SIMILARITIES OF THE ENZYME ASSOCIATED WITH THE MICROSOMAL AND MYELIN FRACTIONS

Abstract— The properties of rat CNS UDP-galactose-ceramidc galactosyltransferase in an axolemma-enriched fraction (AXL), microsomes, and myelin simultaneously isolated with the AXL was characterized using a newly developed assay system. The microsomal enzyme utilized either magnesium or manganese equally well as the divalent cation at 3.3 m m , while both the myelin and AXL enzyme preferred manganese over magnesium at this concentration. The microsomal enzyme was more stable to heat inactivation than the myelin or AXL enzyme. The AXL galactosyltransferase had the highest specific activity at 15 days (8-fold higher than that of the microsomes) and dramatically decreased in specific activity with development. The developmental profile of the myelin enzyme paralleled that of the AXL although the absolute specific activity was lower than that of AXL. In contrast, the specific activity of microsomal enzyme was quite low at the earliest age then sharply increased to 25 days and gradually decreased with further development. The specific activity of the enzyme in AXL isolated from Quaking mouse was dramatically decreased (about 5% of control levels) whereas both whole homogenate and microsomal specific activity were decreased to 35% of control levels. These data indicate that AXL and myelin contain a galactosyltransferase with properties which are unique relative to those of the microsomal fraction. The possible functional significance of these findings with respect to myelination is discussed.     

Distribution of the integral membrane protein NADH-cytochrome b5 reductase in rat liver cells, studied with a quantitative radioimmunoblotting assay.

The intracellular localization of the post-translationally inserted integral membrane protein, NADH-cytochrome b5 reductase, was investigated, using a quantitative radioimmunoblotting method to determine its concentration in rat liver subcellular fractions. Subcellular fractions enriched in rough or smooth microsomes, Golgi, lysosomes, plasma membrane and mitochondrial inner or outer membranes were characterized by marker enzyme analysis and electron microscopy. Reductase levels were determined both with the NADH-cytochrome c reductase activity assay, and by radioimmunoblotting, and the results of the two methods were compared. When measured as antigen, the reductase was relatively less concentrated in microsomal subfractions, and more concentrated in fractions containing outer mitochondrial membranes, lysosomes and plasma membrane than when measured as enzyme activity. Rough and smooth microsomes had 4-5-fold lower concentrations, on a phospholipid basis than did mitochondrial outer membranes. Fractions containing Golgi, lysosomes and plasma membrane had approximately 14-, approximately 16, and approximately 9-fold lower concentrations of antigen than did mitochondrial outer membranes, respectively, and much of the antigen in these fractions could be accounted for by cross-contamination. No enzyme activity or antigen was detected in mitochondrial inner membranes. Our results indicate that the enzyme activity data do not precisely reflect the true enzyme localization, and show an extremely uneven distribution of reductase among different cellular membranes.     

UDP-galactose-ceramide galactosyltransferase in rat brain myelin subfractions during development.

The localization and activity of the enzyme UDP-galactose-hydroxy fatty acid-containing ceramide galactosyltransferase is described in rat brain myelin subfractions during development. Other lipid-synthesizing enzymes, such as cerebroside sulphotransferase, UDP-glucose-ceramide glucosyltransferase and CDP-choline-1,2-diacylglycerol cholinephosphotransferase, were also studied for comparison in myelin subfractions and microsomal membranes. The purified myelin was subfractionated by isopycnic sucrose-density-gradient centrifugation. Four myelin subfractions, three floating respectively on 0.55 M- (light-myelin fraction), 0.75 M- (heavy-myelin fraction) and 0.85 M-sucrose (membrane fraction), and a pellet, were isolated and purified. At all ages, 70--75% of the total myelin proteins was found in the heavy-myelin fraction, whereas 2--5% of the protein was recovered in the light-myelin fraction, and about 7--12% in the membrane fraction. Most of the galactosyltransferase was associated with the heavy-myelin and membrane fractions. Other lipid-synthesizing enzymes studied appeared not to associate with purified myelin or myelin subfractions, but were enriched in the microsomal-membrane fraction. During development, the specific activity of the microsomal galactosyltransferase reached a maximum when the animals were about 20 days old and then declined. By contrast the specific activity of the galactosyltransferase in the heavy-myelin and membrane fractions was 3--4 times higher than that of the microsomal membranes in 16-day-old animals. The specific activity of the enzyme in the heavy-myelin fraction sharply declined with age. Chemical and enzymic analyses of the heavy-myelin and membrane myelin subfractions at various ages showed that the membrane fraction contained more proteins in relation to lipids than the heavy-myelin fraction. The membrane fraction was also enriched in phospholipids compared with cholesterol and contrined equivalent amounts of 2':3'-cyclic nucleotide 3'-phosphohydrolase compared with heavy- and light-myelin fractions. The membrane fraction was deficient in myelin basic protein and proteolipid protein and enriched in high-molecular-weight proteins. The specific localization of galactosyltransferase in heavy-myelin and membrane fractions at an early age when myelination is just beginning suggests that it may have some role in the myelination process.     

Biochemical, kinetic and cytochemical approaches resolve Golgi subcompartments of IgM-secreting rat myeloma cells

The rat myeloma cells chosen for study (IR202) are highly specialized toward the synthesis and secretion of immunoglobulin M (IgM). In [35S]methionine pulse-chase protocols the half-time for secretion of newly synthesized [35S]Ig at 37 degrees C is approximately 2 1/2 h. No degradation of [35S]Ig was detected in such experiments. Pulse-chase experiments with [3H]galactose show that addition of this terminal sugar occurs only approximately 2 min before discharge. The intracellular pool of Ig bearing mature oligosaccharides is therefore very small. Incubation at 20 degrees C stops secretion of the [35S]- and [3H]Ig. We describe a subcellular fractionation protocol for these cells which results in the recovery of a total microsomal fraction by gel filtration. This fraction includes approximately 1/4 of the galactosyltransferase and uridine diphosphatase (UDPase) of the homogenate. By employing two cytological Golgi markers (an "overosmicatable material" and UDPase), galactosyltransferase activity and [35S]methionine and [3H]galactose pulse-chase protocols with the chase at 15 degrees C we document the partial resolution of Golgi subcompartments in isopycnic sucrose gradients used to subfractionate the total microsomal fraction. Electron microscopic and enzymologic examination of the fractions resolved by these gradients confirm that rough microsomes are well separated from Golgi membranes and that the fractions most highly enriched in galactosyltransferase activity have a protein-based specific activity approximately 10 times that of the total microsomal fraction. These studies, therefore, form the basis for an analysis of the composition of the membranes of the Golgi Complex and document the location of proximal Golgi elements, as defined by cytological criteria, in isopycnic gradients.     

Incorporation of galactose from UDP-galactose into microsomal and golgi membranes of rat liver

Rough and smooth microsomes and Golgi membranes were incubated with UDP[¹⁴C]galactose and the incorporation of radioactivity into the lipid extract and into endogenous protein acceptors were measured. Antagonistic pyrophosphatases were inhibited with ATP and interference from β-galactosidase activity was greatly decreased by carrying out the incubation at pH 7.8. After incubation the particles were centrifuged to remove free oligosaccharide residues. Radioactivity was found in the lipid extract from Golgi membranes but not from rough and smooth microsomes. This radioactivity, however, was not associated with dolichol or retinyl phosphates. The incorporation of radioactivity into proteins of the Golgi fraction was more than double than that of the microsomal fractions. In addition, the transferases in these two types of particles exhibited different properties. Trypsin treatment of intact rough microsomal vesicles, smooth vesicles and Golgi membranes removed about 5, 15 and 50%, respectively, of newly incorporated protein-bound galactose, indicating that the proportion of the newly galactosylated proteins, which are localized at the cytoplasmic surface of the membrane, is lowest in rough microsomes, intermediate in smooth, and highest in Golgi membranes.     

Activation of rat liver microsomal glutathione S-transferase by hydrogen peroxide: role for protein-dimer formation.

The mechanism of oxygen radical-dependent activation of hepatic microsomal glutathione S-transferase by hydrogen peroxide was studied. Glutathione S-transferase activity in liver microsomes was increased 1.5-fold by incubation with 0.75 mM hydrogen peroxide at 37 degrees C for 10 min, and the increase in activity was reversed by incubation with dithiothreitol. Purified glutathione S-transferase was also activated by hydrogen peroxide after incubation at room temperature, and the increase in the activity was also reversed by dithiothreitol. Immunoblotting with anti-microsomal glutathione S-transferase antibodies after sodium dodecyl sulfate-polyacrylamide gel electrophoresis of hydrogen peroxide-treated microsomes or purified glutathione S-transferase revealed the presence of a glutathione S-transferase dimer. These results indicate that the hydrogen peroxide-dependent activation of the microsomal glutathione S-transferase is associated with the formation of a protein dimer.     

Smooth microsomes. a trap for cholesteryl ester formed in hepatic microsomes

Acyl-CoA:cholesterol acyltransferase was found predominantly (85%) in RNA-rich microsomes, the rest being in RNA-poor and smooth microsomes. However, the esterified cholesterol concentration of smooth microsomes was 2-fold greater than that of RNA-rich microsomes, suggesting the possibility of an interaction between RNA-rich and smooth microsomes. The distribution of cholesteryl ester between microsome subfractions was examined after incubation of a mixture of RNA-rich and smooth microsomes with [1-14C]palmitoyl-CoA. Based upon specific acyl-CoA:cholesterol acyltransferase activities of the individual fractions, only 31 +/- 3% of the total cholesteryl ester radioactivity should have been found in the smooth component. However, the smooth microsomes contained 54 +/- 3% (p < 0.01) of the radioactive cholesteryl esters. The entrapment of radioactive cholesteryl ester in the smooth microsomes could not be accounted for by passive transfer of cholesteryl ester from RNA-rich microsomes to smooth microsomes. It is proposed that cholesterol in the smooth microsomal membranes may have been esterified by acyl-CoA:cholesterol acyltrasferase located on the surface of RNA-rich microsomes with the resulting cholesteryl ester retained in the smooth microsomes. This hypothesis was strengthened by the observation that acyl-CoA:cholesterol acyl-transferase was located on the cytoplasmic surface of the RNA-rich microsomal vesicle.     

Hydroxymethylglutaryl CoA reductase and the modulation of microsomal cholesterol content by the nonspecific lipid transfer protein

The influence of membrane cholesterol content on 3-hydroxy-3-methylglutaryl CoA reductase (HMG-CoA reductase, EC 1.1.1.34) in rat liver microsomes was investigated. Microsomes were enriched in cholesterol by incubation with egg phosphatidylcholine-cholesterol vesicles and the nonspecific lipid transfer protein from rat liver. By this method, the microsomal cholesterol content was 2.5-fold enhanced up to final concentrations of 140 nmol cholesterol per mg microsomal protein. In another experiment, microsomes isolated from rats fed a cholesterol-rich diet were depleted of cholesterol by incubation with egg phosphatidylcholine vesicles and the transfer protein. Both cholesterol enrichment and depletion had virtually no effect on the microsomal HMG-CoA reductase activity. In another set of experiments, normal rat liver microsomes were incubated with human serum, resulting in a rise of microsomal cholesterol content. This was reflected in an increase of acyl-CoA:cholesterol acyltransferase activity but failed to have an effect on HMG-CoA reductase.     

Purification by affinity chromatography and some properties of microsomal galactosyltransferase from pig thyroid.

Membrane-bound 4-beta-galactosyltransferase (lactose synthase; UDP galactose: D-glucose 4-beta-galactosyltransferase, EC 2.4.1.22) was purified 1500-fold to near homogeneity from pig thyroid microsomes with about 30% yield. The purified enzyme behaved as a lipophilic protein, rapidly losing activity and aggregating if not supplemented with either Triton X-100 or serum albumin (both of these were equally effective for long-term stabilization). The enzyme preparation showed an absolute requirement for Mn2+, which could not be replaced by other cations. Catalytic properties were very similar to those reported for soluble forms of the enzyme in biological fluids. The purified galactosyltransferase showed a major protein band of approx. 74,000 daltons on sodium dodecyl sulfate gel electrophoresis. On gel filtration, enzyme activity was eluted at approx. 70,000 daltons. It is concluded that the membrane-bound thyroid galactosyltransferase is a monomeric protein significantly larger than the soluble forms of this enzyme described earlier; but it resembles recently reported galactosyltransferases from sheep mammary Golgi membranes and liver microsomes.     

Galactosyl- and fucosyltransferases in etiolated pea epicotyls: product identification and sub-cellular localisation

Particulate membrane preparations from etiolated pea epicotyls were found to contain fucosyltransferases, which transferred fucose from GDP-fucose onto xyloglucan and N-linked glycoprotein, and galactosyltransferases, which transferred galactose from UDP-galactose onto galactan, xyloglucan, and N-linked glycoprotein. The products were characterised by specific enzyme degradation and by acid and alkaline hydrolysis. All the enzymes were found to be concentrated in the Golgi apparatus. The Golgi apparatus was further fractionated into membranes of low, medium and high-density. The glycoprotein fucosyltransferase activity was present in highest amounts in the medium-density Golgi membranes, while the majority of the xyloglucan fucosyltransferase was present in the low-density Golgi membranes. The majority of the galactan galactosyltransferase (galactan synthase) was found in the low-density membranes, while the glycoprotein galactosyltransferase was equally distributed in all three subfractions.     

Phosphatidylcholine synthesis for incorporation into membranes or for secretion as plasma lipoproteins by Golgi membranes of rat liver

Phosphatidylcholine, the major phospholipid of very low density lipoproteins, is packaged with triglyceride in the Golgi cisternae. CTP-phosphocholine cytidyltransferase and CDP-choline phosphotransferase activities of Golgi subfractions were higher than those of rough or smooth microsomes measured under the same conditions, indicating that phosphatidylcholine synthesis can occur in Golgi membranes. Consistent with this, the specific activity of phosphatidylcholine of Golgi membranes rose more rapidly than that of rough and smooth microsomes after injection of [14C]choline in vivo. The specific activity of the Golgi content phosphatidylcholine (non-membrane fraction) remained low. The S-adenosylmethionine phosphatidylethanolamine methyltransferase activity of Golgi subfractions was also higher than that of rough or smooth microsomes. After injection of [3H]methyl-labeled methionine in vivo, the specific activity of phosphatidylcholine of the Golgi membranes rose in parallel with that of the rough and smooth microsomes. The specific activity of the Golgi content phosphatidylcholine rose above that of the Golgi membranes and exhibited a different pattern, suggesting that this pathway may selectively label phosphatidylcholine which is secreted as lipoproteins. These observations indicate that the Golgi membranes have the enzymes necessary for synthesis of phosphatidylcholine, and incorporation of lipid precursors indicates that synthesis of phosphatidylcholine by Golgi membranes occurs in vivo.     

A possible role of Golgi membrane-associated galactosyltransferase in the formation of zymogen granule glycoproteins

A galactosyltransferase activity in smooth microsomes and Golgi membrane-rich fractions from rat pancreas glycosylated endogenous acceptors during incubation with UDP-[¹⁴C]galactose in the absence of exogenous glycoproteins. To evaluate the role of this activity in secretion, the endogenous products were partially characterized. Galactose-labeled fractions were sequentially extracted in 0.2 m NaHCO₃ and 0.25 m NaBr to prepare membranes and soluble acceptors. Bound radioactivity was equally distributed between these two fractions. Analysis by polyacrylamide gel electrophoresis in sodium dodecyl sulfate indicated that the particulate galactose-labeled polypeptides were distinct from the soluble galactose acceptors. Rabbit antisera against highly purified zymogen granule membranes precipitated approximately 40% of the radioactivity of the particulate fraction when solubilized in nonionic detergents. In polyacrylamide gels, the galactose-labeled species of the immunoprecipitate migrated with zymogen granule membrane glycoproteins. Rabbit antisera against secretory proteins cross-reacted with less than 5% of the galactose-labeled soluble acceptors. Mature zymogen granule membranes neither contained detectable galactosyltransferase activity nor served as galactosyltransferase acceptors. These results suggest that galactosyltransferase activity associated with membranes derived from the Golgi complex glycosylated zymogen granule membrane precursors. Analysis of [¹⁴C]galactolipids did not implicate lipid intermediates in this process.     

Isolation of Golgi fractions from colchicine-treated rat liver. II. Electrophoretic characterization.

1. Intact Golgi fractions, three from colchicine- or ethanol-treated rat livers and two from a control, were analyzed by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis. All the fractions showed very similar electrophoretic profiles with 33 protein bands, some of which, especially albumin, had rather higher density in the secretory vesicle fraction than those in the cisternal fraction. 2. Using albumin as the content marker, the Golgi fractions were subfractionated into membranes and contents by freezing-thawing and sonication followed by centrifugation. Distribution of galactosyltransferase among these membrane preparations showed that this enzyme was more enriched in the Golgi cisternal membranes than in the secretory vesicle membranes. 3. All the membrane preparations from the Golgi complex showed very similar patterns on electrophoresis, which were distinctly different from those of microsomal membranes and of plasma membrane. Furthermore, all the Golgi content subfractions had similar protein components, most of which were also found in serum. The microsomal contents, however, showed a considerably different pattern from those of the Golgi contents. 4. From these results it could be concluded that the secretory vesicles are indeed a member of the Golgi complex despite their different appearance and morphology.     

Characterization of microsomal subfractions from porcine endometrium cells

Crude microsomes from porcine endometrium and three subfractions obtained by a modification of Rothschild's technique were characterized by RNA/protein ratio, marker enzyme activities and morphological appearance. The microsomes were devoid of glucose-6-phosphatase activity. They contained approximately 10% of arylesterase-, approximately 30% of both NADPH-cytochrome reductase- and UDPgalactose-N-acetyl-glucosamine beta-D-galactosyltransferase- and approximately 60% of 5'-nucleotidase activities present in the homogenates. Subfraction I (smooth membranes) had twice the galactosyltransferase activity of Subfraction II (smooth and rough membranes + free ribosomes); both subfractions were rich in 5'-nucleotidase and cytochrome reductase activities. Subfraction III (rough membranes) had very low marker activities but exhibited the highest RNA/protein ratio, which was lowest in I.     

Incorporation of N-acetylglucosamine from UDP-N-acetylglucosamine into proteins and lipid intermediates in microsomal and golgi membranes from rat liver

Rough and smooth microsomes and Golgi membranes incorporate $\text{N-}\text{acetylglucosamine}$ from $\text{UDP}\text{-N-}\text{acetylglucosamine}$ into endogenous protein acceptors. A lipid intermediate of the dolichol phosphate type participates in this transfer reaction in the case of both microsomal subfractions, but the nature of lipid glycosylation is different in these two fractions. Glucosamine transfer in Golgi membranes does not appear to involve a lipid intermediate. In contrast to the results obtained under in vivo conditions, no glucosamine label is recovered in nascent ribosomal proteins or on luminal secretory proteins after incubation in vitro. Proteolysis of intact vesicles of the subfractions removes glycosylated dolichol phosphate and protein acceptors to various extents and interferes with transferase activities. This finding suggests the possibility that glycosylation at the cytoplasmic side of the membrane of the endoplasmic reticulum may involve a system separate from that acting at the luminal side of the same membrane.