The UDP-Glc:glycoprotein glucosyltransferase is a soluble protein of the endoplasmic reticulum.

N-linked oligosaccharides devoid of glucose residues are transiently glucosylated directly from UDP-Glc in the endoplasmic reticulum. The reaction products have been identified, depending on the organisms, as protein-linked Glc1Man5-9GlcNAc2. Incubation of right side-sealed vesicles from rat liver with UDP-[14C]Glc, Ca2+ ions and denatured thyroglobulin led to the glucosylation of the macromolecule only when the vesicles had been disrupted previously by sonication or by the addition of detergents to the glucosylation mixture. Similarly, maximal glucosylation of denatured thyroglobulin required disruption of microsomal vesicles isolated from the protozoan Crithidia fasciculata. Treatment of the rat liver vesicles with trypsin led to the inactivation of the UDP-Glc:glycoprotein glucosyltransferase only when proteolysis was performed in the presence of detergents. The glycoprotein glucosylating activity could be solubilized upon sonication of right side-sealed vesicles in an isotonic medium, upon passage of them through a French press or by suspending the vesicles in an hypotonic medium. Moreover, the enzyme appeared in the aqueous phase when the vesicles were submitted to a Triton X-114/water partition. Solubilization was not due to proteolysis of a membrane-bound enzyme. The enzyme could also be solubilized from C. fasciculata microsomal vesicles by procedures not involving membrane disassembly. About 30% of endogenous glycoproteins glucosylated upon incubation of intact rat liver microsomal vesicles with UDP-[14C]GLc could be solubilized by sonication or by suspending the vesicles in 0.1 M Na2CO3. These and previous results show that the UDP-Glc:glycoprotein glucosyltransferase is a soluble protein present in the lumen of the endoplasmic reticulum. In addition, both soluble and membrane-bound glycoproteins may be glucosylated by the glycoprotein glucosylating activity.     

ADP ribosylation factor and a 14-kD polypeptide are associated with heparan sulfate-carrying post-trans-Golgi network secretory vesicles in rat hepatocytes

Constitutive secretory vesicles carrying heparan sulfate proteoglycan (HSPG) were identified in isolated rat hepatocytes by pulse-chase experiments with [35S]sulfate and purified by velocity-controlled sucrose gradient centrifugation followed by equilibrium density centrifugation in Nycodenz. Using this procedure, the vesicles were separated from plasma membranes, Golgi, trans-Golgi network (TGN), ER, endosomes, lysosomes, transcytotic vesicles, and mitochondria. The diameter of these vesicles was approximately 100-200 nm as determined by electron microscopy. A typical coat structure as described for intra- Golgi transport vesicles or clathrin-coated vesicles could not be seen, and the vesicles were not associated with the coat protein beta-COP. Furthermore, the vesicles appear to represent a low density compartment (1.05-1.06 g/ml). Other constitutively secreted proteins (rat serum albumin, apolipoprotein E, and fibrinogen) could not be detected in purified HSPG-carrying vesicles, but banded in the denser fractions of the Nycodenz gradient. Moreover, during pulse-chase labeling with [35S]methionine, labeled albumin did not appear in the post-TGN vesicle fraction carrying HSPGs. These findings indicate sorting of HSPGs and albumin into different types of constitutive secretory vesicles in hepatocytes. Two proteins were found to be tightly associated with the membranes of the HSPG carrying vesicles: a member of the ADP ribosylation factor family of small guanine nucleotide-binding proteins and an unknown 14-kD peripheral membrane protein (VAPP14). Concerning the secretory pathway, we conclude from these results that ADP ribosylation factor proteins are not only involved in vesicular transport from the ER via the Golgi to the TGN, but also in vesicular transport from the TGN to the plasma membrane.     

Isolation and purification of extracellular matrix vesicles from blood vessels.

Extracellular membrane-bound vesicles (called matrix vesicles) which occur in abundance in atherosclerotic blood vessels are believed to be associated with lipid accumulation and calcification. A technique has been developed to isolate them from experimental aneurysms in sheep in which they are known to be plentiful. The matrix vesicles were isolated by differential centrifugation following extraction by hypotonic salt solution. Most of the vesicles were pelleted at 30,000g and fell within the size range of matrix vesicles in situ in the aneurysmal wall. Preliminary characterization of the enzymatic activities indicates that many of these vesicles are formed from cell membranes rather than being derived from lysosomes, mitochondria or endoplasmic reticulum. Morphologically they are similar to matrix vesicles of other mineralizing tissues.     

Transport of phosphate in membrane vesicles from mouse fibroblasts transformed by simian virus 40.

Membrane vesicles were prepared from mouse fibroblasts transformed by SV40 virus (SV3T3). Following disruption of the cells by nitrogen cavitation, the membrane vesicles were obtained by differential centrifugation. As measured by enzyme markers, they consist mainly of membrane from the plasma membrane and smooth and rough endoplasmic reticulum. The vesicles transport Pi by two separate, mediated systems: one is independent of Na+, and the other is secondary active transport driven by a Na+ gradient. Electrical and chemical energy can be provided by a Na+ gradient to drive the concentrative uptake of Pi by the vesicles, one or both forces being used to energize transport. Evidence is provided that both the electrical and chemical potentials produced by the asymmetric distribution of Na+ across the membrane of SV3T3 membrane vesicles are utilized to concentrate phosphate in the vesicles. Phosphate transport by the vesicles cannot be accounted for by a small contamination of this fraction with mitochondria (1 to 4%). The Pi transport properties of the membrane vesicles differ from those of the fraction enriched in mitochondria in the following respects: their kinetic properties, and their responses to a Na+ gradient, N-ethylmaleimide, mersalyl, and succinate/acetate. However, the membrane vesicles share some properties of Pi transport with mitochondria. Cyanide, azide, oligomycin, 2,4-dinitrophenol, and carbonyl cyanide m-cholophenylhydrazone, inhibitors of Pi transport by mitochondria, also inhibit membrane vesicle, Pi transport. The vesicles retain all the features of Pi transport by SV3T3 cells that have been examined. They provide a simplified system for a determination of the details of the mechanism of Pi transport under conditions where transport is dissociated from intracellular reactions and in the presence of a defined electrochemical driving force.     

Cholesterol-phospholipid vesicles in human bile: an ultrastructural study

Phospholipid vesicles, a newly described (bile salt independent) mode of cholesterol transport in human bile, were previously characterized by quasi-elastic light scattering and gel filtration. In the present study the ultrastructure of these vesicles was investigated by electron microscopy using freeze-fracture and negative-staining techniques. Vesicles of varying size were found in all 14 hepatic and 3 gallbladder biles examined. The diameter of the vesicles ranged from 25 to 75 nm by electron microscopy after freeze fracture and from 54 to 94 nm by quasi-elastic light scattering. They had a spherical shape and appeared to be unilamellar. The appearance of the vesicles in fresh hepatic and gallbladder biles as well as in chromatographic fractions was similar. Vesicles were dissolved by the addition of exogenous bile salts. Cholesterol is transported in human bile by both vesicles and micelles. The role of the vesicles may be particularly important in preventing cholesterol precipitation in dilute and supersaturated biles.     

Purification of a novel class of coated vesicles mediating biosynthetic protein transport through the Golgi stack

We describe a scheme for the purification of the nonclathrin-coated vesicles that mediate transport of proteins between Golgi cisternae and probably from ER to Golgi. These "Golgi-derived coated vesicles" accumulate when Golgi membranes are incubated with ATP and cytosol in the presence of GTP gamma S, a compound that blocks vesicle fusion. The coated vesicles dissociate from the Golgi cisternae in high salt and can then be purified by employing differential and density gradient centrifugation. Golgi-derived coated vesicles have a putative polypeptide composition that is distinct from both cytosol and Golgi membranes, as well as from that of clathrin-coated vesicles.     

The absorption of phospholipid vesicles by perfused rat liver depends on vesicle surface charge

Two types of sonicated vesicle have been prepared from dipalmitoylphosphatidylcholine (DPPC) by incorporation of phosphatidylinositol (PI) to give negatively charged vesicles and stearylamine to give positively charged vesicles. The absorption of the vesicles by rat liver has been investigated by perfusion techniques. A steady state of vesicle absorption is rapidly established in approx. 2 min and the initial rates of absorption decrease with PI content of the vesicles and increase with stearylamine content. In the steady state, the uptake of vesicles by the liver is similarly dependent on vesicle charge, being inhibited by PI and enhanced by incorporation of stearylamine in the vesicles. Fractionation of the liver into subcellular fractions following perfusion showed that most of the vesicular lipid could be found associated with a nuclear (plus plasma membrane) fraction. The suppression of vesicle absorption by PI may be of value as a means of bypassing the liver in relation to the use of vesicles as a delivery system.     

Stimulation of precipitation of calcium phosphate by matrix vesicles.

The ability of matrix vesicles isolated from the epiphysial growth plate of 6-week-old chicks to facilitate the precipitation of calcium phosphate was studied in vitro. The vesicles lowered the minimum concentration product [ca2+]X[p1] needed to induce crystal formation, thereby showing the vesicles are nucleators of crystallization. After freezing and thawing the vesicles at pH6.0, part but not all of this ability to nucleate disappeared. Freezing and thawing markedly decreased the Ca and Pi content of the vesicles, suggesting that part of the nucleating activity may have been due to mineral already present. After removal of the mineral the residual nucleating activity could be destroyed by extracting the vesicles with lipid solvents or by treatment with enzymes such as phosphoilipase C, neuraminidase or proteinase. Matrix vesicles obtained from chicks treated with 1-hydroxyethane-1, 1-diphosphonate, a compound that inhibits calcification in vivo, showed impaired nucleating activity, both before and after treatment at pH6.0. The vesicle preparation bound some diphosphonate in vitro, probably to the mineral present in the preparation, since no binding could be detected in vesicles preincubated at pH6.0. No difference was found in the nucleating activity of vesicles isolated from rachitic chicks which had or had not received cholacalciferol 48 h before death. These results suggest that matrix vesicles possess intrinsic nucleating activity that may be important in biological calcification.     

Formation of multilamellar vesicles by addition of tannic acid to phosphatidylcholine-containing small unilamellar vesicles

Tannic acid induces aggregation and formation of multilamellar vesicles when added to preparations of small unilamellar vesicles, specifically those containing phosphatidylcholine. Aggregation and clustering of vesicles was demonstrated by cryo-electron microscopy of thin films and by freeze-fracture technique. Turbidity measurements revealed an approximately one-to-one molar ratio between tannic acid and phosphatidylcholine necessary for a fast and massive aggregation of the small unilamellar vesicles. When tannic acid-induced aggregates were dehydrated and embedded for conventional thin-section electron microscopy, multilamellar vesicles were retrieved in thin sections. It is concluded from morphological studies, as well as previous tracer studies, that tannic acid, at least to a great extent, prevents the extraction of phosphatidylcholine. Multilamellar vesicles were also observed in tannic acid-treated vesicles prepared from total lipid extracts from either rabbit or rat hearts. Substantially more multilamellar vesicles were retrieved in the rabbit vesicle preparation. This difference can probably be explained by the difference in the proportion of the plasmalogen phosphatidylcholine, and possibly the content of sphingomyelin, in lipid extracts of rabbit and rat hearts. It is concluded that the dual effect (reduced extraction and aggregation) of tannic acid on phosphatidylcholines should be taken into consideration when tannic acid is used in tissue preparation.     

Characterization of plasma membrane domains enriched in lipid metabolites.

A subpopulation of plasma membrane vesicles enriched in membrane lipid metabolites has been isolated from petals of carnation flowers and leaves of canola seedlings. This was achieved by immunopurification from a microsomal membrane preparation using region-specific antibodies raised against a recombinant polypeptide of the plasma membrane H(+)-ATPase. The properties of this subpopulation of vesicles were compared with those of purified plasma membrane isolated by partitioning in an aqueous dextran-polyethylene glycol two-phase system. The lipid composition of the immunopurified vesicles proved to be clearly distinguishable from that of phase-purified plasma membrane, indicating that they represent a unique subpopulation of plasma membrane vesicles. Specifically, the immunopurified vesicles are highly enriched in lipid metabolites, including free fatty acids, diacylglycerol, triacylglycerol and steryl and wax esters, by comparison with the phase-purified plasma membrane. These findings can be interpreted as indicating that lipid metabolites generated within the plasma membrane effectively phase-separate by moving laterally through the plane of the membrane to form discrete domains within the bilayer. It is also apparent that these domains, once formed, are released as vesicles into the cytosol, presumably by microvesiculation from the surface of the plasmalemma. Such removal may be part of normal membrane turnover.     

On the preparation and some properties of closed membrane vesicles from hog duodenal and jejunal brush border

Closed and nearly spherical vesicles were obtained from both hog duodenum and jejunum after mucosa homogenization in the absence of EDTA and a series of fractional centrifugations. The vesicles were found to contain large amounts of two of the characteristic enzyme markers of the brush border membrane (aminopeptidase and alkaline phosphatase). They were seen by electron microscopy on thin sections or after negative staining to be composed of an apparently intact, 90–100 Å-thick membrane overlaid by the fuzzy coat and to be partly filled by a fibrous material tentatively identified with the cross-filaments of the microvilli. This filling was not removed by 5 mM EDTA or/and 1 M Tris unless the structure of the vesicles was largely destroyed. Very few empty vesicles were obtained at the end of these treatments.The vesicles from hog duodenum and jejunum were observed to contain nearly 2 molecules of cholesterol for 1 molecule of phospolipids. Specific differences were noted between both types of vesicles at the level of their sugar composition and associated enzyme activities. For instance, the jejunal vesicles contained no sialic acid and no enterokinase. They contain, respectively, 2 and 4 times as much alkaline phosphatase and aminopeptidase as duodenal vesicles.     

Isolation and properties of two classes of low-density vesicles from Saccharomyces cerevisiae.

A mixture of small (0.43-mum diameter) and large (0.62-mum diameter) low-density vesicles from spheroplasts of Saccharomyces cerevisiae was fractionated by rate centrifugation in a gradient of 0 to 8% (wt/vol) Ficoll to yield fractions rich (90 to 95%) in small or large vesicles. The large, but not small, vesicles swelled when diluted into mannitol solutions containing less than 0.4 M mannitol. The pH-electrophoretic mobility curve of the large vesicles showed that they are probably enclosed in a phospholipid-protein membrane. The dyes neutral red and toluidine blue, accumulated into large vesicles by intact cells and spheroplasts, were largely lost from large vesicles when these were separated from stained spheroplasts. Sudan black III stained small and large vesicles, both classes of vesicle retaining the stain on separation. Fractions rich in large vesicles contained proportionately more phospholipid and less free sterols, diacylglycerols, and free fatty acids compared with those enriched in small vesicles. The two classes of vesicles contained about the same proportions of esterified sterols and triacylglycerols. The free fatty acids in both small and large vesicles were free from unsaturated fatty-acyl residues; diacylglycerols and triacylglycerols contained appreciable proportions of unsaturated fatty-acyl residues. Small vesicles were richer in lipase activity, whereas the larger vesicles contained greater beta-glucanase and alpha-mannosidase activities. Phospholipase activity could not be detected in any of the fractions.     

Vesicles of variable sizes produced by a rapid extrusion procedure

Previous studies from this laboratory have shown that large unilamellar vesicles can be efficiently produced by extrusion of multilamellar vesicles through polycarbonate filters with a pore size of 100 nm (Hope, M.J., Bally, M.B., Webb, G. and Cullis, P.R. (1985) Biochim. Biophys. Acta 812, 55-65). In this work it is shown that similar procedures can be employed for the production of homogeneously sized unilamellar or plurilamellar vesicles by utilizing filters with pore sizes ranging from 30 to 400 nm. The unilamellarity and trapping efficiencies of these vesicles can be significantly enhanced by freezing and thawing the multilamellar vesicles prior to extrusion. This procedure is particularly applicable when very high lipid concentrations (400 mg/ml) are used, where extrusion of the frozen and thawed multilamellar vesicles through 100 and 400 nm filters results in trapping efficiencies of 56 and 80%, respectively. Freeze-fracture electron microscopy revealed that vesicles produced at these lipid concentrations exhibit size distributions and extent of multilamellar character comparable to systems produced at lower lipid levels. These results indicate that the freeze-thaw and extrusion process is the technique of choice for the production of vesicles of variable sizes and high trapping efficiency.     

Membrane vesicles shed by murine melanoma cells selectively inhibit the expression of Ia antigen by macrophages

We previously demonstrated that membrane vesicles shed by the F10 variant of the murine B16 melanoma cell line inhibited the induction by interferon-gamma (IFN) of murine macrophage immune response region-associated (Ia) antigen expression. In this paper we present evidence that the inhibition of macrophage Ia antigen expression is a selective effect of vesicles and characterize its temporal requirements. Membrane vesicles shed from F10 cells did not affect the expression of macrophage H-2K or H-2D antigens under conditions shown to profoundly inhibit Ia antigen expression. Similarly, the induction of plasminogen activator and interleukin 1 from macrophages was not inhibited by the vesicles. The vesicles did not measurably decrease total cellular RNA or protein synthesis. Macrophages were sensitive to the inhibitory effects of the vesicles during the induction and maintenance phases of Ia expression. Pretreatment of macrophages with vesicles before culture with IFN did not reduce the induction of Ia. The rate of decline of Ia expression after removal of IFN was unaffected by the presence of vesicles. Removal of vesicles from cultures of IFN-treated macrophages resulted in only a partial recovery of Ia expression, suggesting that the inhibition of Ia expression may be a slowly reversible process. The selective and partially reversible inhibition of Ia expression by vesicles shed from the plasma membrane of tumor cells is a possible mechanism whereby tumor-bearing hosts may become immunocompromised.     

Processing and secretion in the neurohypophysis. Stability of isolated secretory vesicles and role of internal pH.

A possible role of low pH in secretory vesicles for processing and secretion in the neurohypophysis was investigated. Subcellular fractionation of guinea-pig neural lobes revealed that a proton present in the membranes from this tissue could not be ascribed to secretory vesicles. However, a proton pump was found in coated microvesicles. Secretory vesicles isolated from rats and guinea pigs were stable under conditions known to lyse secretory vesicles from the adrenal medulla owing to the generation of a proton gradient. These results suggest that the internal pH of secretory vesicles from the neurohypophysis is closer to neutral than is the pH in chromaffin secretory vesicles. Processing of a neurophysin-glycopeptide intermediate from the biosynthesis of vasopressin in intact secretory vesicles incubated in vitro was activated by the addition of NH4Cl, known to increase the intravesicular pH. This activation of neurohormone processing was also apparent in isolated nerve endings incubated in the presence of NH4Cl, suggesting that NH4Cl can also be used to increase the intravesicular pH in intact nerve endings. However, NH4Cl did not affect the secretion of neurohormones, indicating that a low intravesicular pH is not important for exocytosis in the neurohypophysis. Our results indicate that a low pH generated during processing by mechanisms other than ATP-dependent proton transport may inhibit the processing enzymes, thereby preventing extensive breakdown of neurohormone precursors.     

Transport-Specific isolation of large channels reconstituted into lipid vesicles

Summary To develop a technique for purifying and identifying pore-forming membrane proteins, we used a transport-specific increase in buoyant density to select for lipid vesicles containing voltage-dependent anion channels (VDAC). Monodisperse, single-walled vesicles were formed by gel filtration from a detergent-solubilized mixture of lipid and protein in a urea buffer. The vesicles were layered on a linear iso-osmolar density gradient formed of urea and sucrose buffers. Since VDAC is open at zerotrans-membrane voltage and is permeable to urea and sucrose, vesicles containing functional VDAC should become more dense as sucrose enters and urea leaves, while those lacking open channels should maintain their original density. Vesicles formed in the absence of VDAC migrated to a characteristic density, while vesicles formed in the presence of VDAC fractionated into two populations in the gradients, one migrating to the same density as the vesicles formed without VDAC, and one at a significantly greater density. In contrast to the lower density vesicles, the higher density vesicles showed a high permeability to calcein, and contained functional VDAC channels (shown by electrophysiological recordings following fusion with a planar bilayer). Thus, vesicles containing open channels were separable from those that did not by a transport-specific shift in density. This technique may be useful for the enrichment of channels of known permeability properties from impure, material.     

Transition vesicle-enriched fractions isolated from rat liver exhibit a unique peptide pattern

Summary Isolated fractions enriched in transition vesicles isolated by preparative free-flow electrophoresis from transition elements of endoplasmic reticulum incubated with ATP and a cytosol fraction in a cell free system, exhibited a polypeptide composition distinct from that of the original transitional endoplasmic reticulum fractions. The transition vesicles were deficient in peptides of apparent molecular mass between 200 and 270 kD, in the vicinity of 65 kD and less than 40 kD. Immunoblot analysis suggested a 140 kD protein to be concentrated in transition vesicles and low or absent from the endoplasmic reticulum fractions from which the transition vesicles were derived.     

Comparative electrophysiological study of reconstituted giant vesicle preparations of the rabbit skeletal muscle sarcoplasmic reticulum K+ channel

Sarcoplasmic reticulum (SR) membranes isolated from rabbit skeletal muscle were reconstituted into two types of giant vesicles: (1) Giant proteoliposomes prepared by freeze-thawing of a mixture of SR vesicles and sonicated phospholipid vesicles without the use of detergent. (2) Giant SR vesicles prepared by fusion of SR vesicles using poly(ethylene glycol) (PEG) as a fusogen and without the addition of exogenous lipid. These giant vesicles were patch-clamped and properties of the single voltage-dependent potassium channel in the excised patch were studied. Single-channel conductance in a symmetrical solution of 0.1 M KCl and 1 mM CaCl2 was 140.0 +/- 10 pS (n = 5) for freeze-thawed vesicles and 136.4 +/- 15 pS (n = 7) for PEG vesicles. Both types of vesicles exhibited a sub-conductance state having 55% of the fully open state conductance. The voltage-dependence of open-channel probability could be expressed in terms of thermodynamic parameters of delta Gi = 0.95 kcal/mol and z = -0.77 for freeze-thawed vesicles and delta Gi = 0.92 kcal/mol and z = -0.87 for PEG vesicles. These values correlated well with previous data obtained by fusion of native SR vesicles with a planar lipid membrane. Channel orientation was found to be conserved in both types of vesicles used in the present study.     

A mechanism for the formation of inside-out membrane vesicles. Preparation of inside-out vesicles from membrane-paired randomized chloroplast lamellae

Inside-out thylakoid membrane vesicles can be isolated by aqueous polymer two-phase partition of Yeda press-fragmented spinach chloroplasts (Andersson, B. and Åkerlund, H.-E. (1978) Biochim. Biophys. Acta 503, 462–472). The mechanism for their formation has been investigated by studying the yield of inside-out vesicles after various treatments of the chloroplasts prior to fragmentation. No inside-out vesicles were isolated during phase partitioning if the chloroplasts had been destacked in a low-salt medium prior to the fragmentation. Only in those cases where the chloroplast lamellae had been stacked by cations or membrane-paired by acidic treatment did we get any yield of inside-out vesicles. Thus, the intrinsic properties of chloroplast thylakoids seem to be such that they seal into right-side out vesicles after disruption unless they are in an appressed state. This favours the following mechanism for the formation of inside-out thylakoids. After press treatment, a ruptured membrane still remains appressed with an adjacent membrane. Resealing of such an appressed membrane pair would result in an inside-out vesicle.If the compartmentation of chloroplast lamellae into appressed grana and unappressed stroma lamellae is preserved by cations before fragmentation, the inside-out vesicles are highly enriched in photosystem II. This indicates a granal origin which is consistent with the proposed model outlined. Inside-out vesicles possessing photosystem I and II properties in approximately equal proportions could be obtained by acid-induced membrane-pairing of chloroplasts which had been destacked and randomized prior to fragmentation. Since this new preparation of inside-out thylakoid vesicles also exposes components derived from the stroma lamellae it complements the previous preparation.It is suggested that fragmentation of paired membranes followed by phase partitioning should be a general method of obtaining inside-out vesicles from membranes of various biological sources.     

Translocation of UDP-N-acetylglucosamine into vesicles derived from rat liver rough endoplasmic reticulum and Golgi apparatus

A mixture of UDP-N-acetylglucosamine labeled with different radioisotopes in the uridine and glucosamine was used to show that the intact sugar nucleotide was translocated across the membrane of vesicles derived from rat liver rough endoplasmic reticulum (RER) and Golgi apparatus. Translocation was dependent on temperature, saturable at high concentrations of sugar nucleotide, and inhibited by treatment of vesicles with proteases, suggesting protein carrier mediated transport. Translocation of UDP-GlcNAc by RER-derived vesicles appeared to be specific since these vesicles were unable to translocate UDP-galactose, in contrast to those derived from the Golgi apparatus. Preliminary results suggest that the mechanism of UDP-GlcNAc translocation into RER-derived vesicles is via a coupled exchange with lumenal nucleoside monophosphate. This is similar to the recently postulated mechanism for translocation of sugar nucleotides into vesicles derived from the Golgi apparatus.