Orientation of glycoprotein galactosyltransferase and sialyltransferase enzymes in vesicles derived from rat liver Golgi apparatus

UDP-galactose: N-acetylglucosamine galactosyltransferase (GT) and CMP- sialic:desialylated transferrin sialyltransferse (ST) activities of rat liver Golgi apparatus are membrane-bound enzymes that can be released by treatment with Triton X-100. When protein substrates are used to assay these enzymes in freshly prepared Golgi vesicles, both activities are enhanced about eightfold by the addition of Triton X-100. When small molecular weight substrates are used, however, both activities are only enhanced about twofold by the addition of detergent. The enzymes remain inaccessible to large protein substrates even after freezing and storage of the Golgi preparation for 2 mo in liquid nitrogen. Accessibility to small molecular and weight substrates increases significantly after such storage. GT and ST activities in Golgi vesicles are not destroyed by treatment with trypsin, but are destroyed by this treatment if the vesicles are first disrupted with Triton X-100. Treatment of Golgi vesicles with low levels of filipin, a polyene antibiotic known to complex with cholesterol in biological membranes, also results in enhanced trypsin susceptibility of both glycosyltransferases. Maximum destruction of the glycosyltransferase activities by trypsin is obtained at filipin to total cholesterol weight ratios of approximately 1.6 or molar ratios of approximately 1. This level of filipin does not solubilize the enzymes but causes both puckering of Golgi membranes visible by electron microscopy and disruption of the Golgi vesicles as measured by release of serum albumin. When isolated Golgi apparatus is fixed with glutaraldehyde to maintain the three-dimensional orientation of cisternae and secretory vesicles, and then treated with filipin, cisternal membranes on both cis and trans faces of the apparatus as well as secretory granule membranes appear to be affected about equally. These results indicate that liver Golgi vesicles as isolated are largely oriented with GT and ST on the luminal side of the membranes, which corresponds to the cisternal compartment of the Golgi apparatus in the hepatocyte. Cholesterol is an integral part of the membrane of the Golgi apparatus and its distribution throughout the apparatus is similar to that of both transferases.     

Ganglioside biosynthesis. Characterization of CMP-N-acetylneuraminic acid : lactosylceramide sialyltransferase in Golgi apparatus from rat liver.

An enzyme that transfers sialic acid from GMP-sialic acid to lactosylceramide was concentrated 40-50 times in Golgi apparatus from rat liver relative to total homogenates. This enzyme required detergents as dispersing agents. Of the numerous detergents tested, the combination Tween 80-Triton CF-54 (1 : 2, w/w) was the most effective in stimulating the reaction. Two apparent pH optima, at 6.35 and 5.5, were observed. The enzyme showed no requirement for a divalent cation. The Km values calculated for CMP-N-acetylneuraminic acid and lactosylceramide were 2.7 - 10(-3) and 1.3 - 10(-4) M, respectively. The enzyme could not be dissociated from Golgi apparatus fractions by treatment with ultrasound, indicating that it is tightly associated with the membrane. The newly synthesized GM3, the product of the reaction, was incorporated into or became tightly associated with the membranes of the Golgi apparatus.     

The 17-residue transmembrane domain of beta-galactoside alpha 2,6- sialyltransferase is sufficient for Golgi retention

beta-Galactoside alpha 2,6-sialyltransferase (ST) is a type II integral membrane protein of the Golgi apparatus involved in the sialylation of N-linked glycans. A series of experiments has shown that the 17-residue transmembrane domain of ST is sufficient to confer localization to the Golgi apparatus when transferred to the corresponding region of a cell surface type II integral membrane protein. Lectin affinity chromatography of chimeric proteins bearing this 17-residue sequence suggests that these chimeric proteins are localized in the trans-Golgi cisternae and/or trans-Golgi network. Further experiments suggest that this 17-residue sequence functions as a retention signal for the Golgi apparatus.     

The membrane-binding domain of a 23-kDa G-protein is carboxyl methylated

We have purified to homogeneity a 23-kDa protein from bovine brain membranes using [35S]guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S) binding as an assay. GTP gamma S binding to the purified protein is inhibited by GDP, GTP, and GTP analogs but not by cGMP, GMP, or adenine nucleotides, consistent with the nucleotide-binding behavior of members of the family of GTP-binding regulatory proteins. On addition of the methyl donor S-adenosyl-L-methionine and a methyltransferase present in bovine brain membranes, the purified 23-kDa G-protein is carboxyl methylated. When subjected to limited tryptic proteolysis, the 23-kDa protein is converted to a 22-kDa major fragment with concomitant release of a carboxyl methylated protein fragment of 1 kDa. Furthermore, when the cleaved protein is reconstituted with stripped bovine brain membranes, the small carboxyl-methylated fragment but not the 22-kDa major fragment is found to reassociate with the membranes. These results indicate that the site of carboxyl methylation and the region responsible for membrane anchoring, most likely, are localized to a small region at the carboxyl terminus. It is attractive to speculate that carboxyl methylation and membrane anchoring are interrelated processes and play key roles in the function of this small G-protein.     

The streptozotocin-prostaglandin interaction in Golgi apparatus of rat liver

16,16'-Dimethylprostaglandin E2 was administered to rats in three doses: 30 min prior and 24 h and 48 h after a single intraperitoneal injection of streptozotocin. The Golgi membrane fraction was analyzed 6 days after streptozotocin injection. It has been found that prostaglandin restores the Golgi membrane fraction and the activity of UDP-Gal----GlcNAc transferase, both significantly decreased upon treatment with streptozotocin alone. Morphology of the liver Golgi apparatus studied by the electron microscopy was similar to that of control from untreated rats although streptozotocin alone significantly decreased the size of this organelle.     

Substrate specificities of a bacterial sialidase and rat liver ganglioside GM3 sialyltransferase

Sialidases cleave off sialic acid residues from the oligosaccharide chain of gangliosides in their catabolic pathway while sialyltransferases transfer sialic acid to the growing oligosaccharide moiety in ganglioside biosynthesis. Ganglioside G_M3 is a common substrate for both types of enzymes, for sialidase acting on ganglioside G_M3 as well as for ganglioside G_D3 synthase. Therefore, it is possible that both enzymes recognize similar structural features of the sialic acid moiety of their common substrate, ganglioside G_M3. Based on this idea we used a variety of G_M3 derivatives as glycolipid substrates for a bacterial sialidase (Clostridium perfringens) and for G_D3 synthase (of rat liver Golgi vesicles). This study revealed that those G_M3 derivatives that were poorly degraded by sialidase also were hardly recognized by sialyltransferase (G_D3 synthase). This may indicate similarities in the substrate binding sites of these enzymes.     

Glycolipid membrane-binding domain of human erythrocyte acetylcholinesterase

The membrane-binding domain of human erythrocyte acetylcholinesterase is a small hydrophobic structure at the COOH-terminus of the enzyme subunits. Papain digestion cleaves a COOH-terminal dipeptide linked to the hydrophobic structure with the sequence His-Gly-ethanolamine-Z, where the ethanolamine is in amide linkage to the glycine and Z is a partially characterized glycolipid. This glycolipid includes a second residue of ethanolamine and a residue of glucosamine, both of which have free primary amino groups accessible to radiomethylation. The glycolipid also contains a carbohydrate residue or residues that bind to concanavalin A and nearly stoichiometric amounts of both palmitate and C22 unsaturated fatty acids. Similarities in this membrane-binding structure to those reported for trypanosome variant surface glycoproteins and Thy-1 glycoprotein suggest an important new category of posttranslational modifications involving the attachment of COOH-terminal glycolipid.     

Lactosylceramide molecular species specificity of rat liver CMP-N-acetylneuraminate:lactosylceramide sialyltransferase

Six naturally occurring and three synthetic molecular species of lactosylceramide (LacCer) were used to examine the molecular species specificity of CMP-N-acetylneuraminate:lactosylceramide alpha 2,3-sialyltransferase in a Golgi-rich fraction of rat liver. The enzyme molecular species specificity was determined either in the presence of nonspecific lipid transfer protein or in the presence of detergents. Assays performed in the presence of transfer protein showed that for those lactosylceramide molecular species with either d18:1 or d18:0 long chain base the enzyme activity decreased linearly as the effective carbon number of the fatty acid increased. An increase in the carbon number of the long chain base decreased the activity of the enzyme twice as much as a corresponding increase in the carbon number of the fatty acid. On the other hand, when the enzyme activity was assayed in the presence of detergents, there was no significant difference in activity among the various molecular species of lactosylceramide based upon the carbon number of the fatty acid or on the presence of a double bond in the long chain base. However, the decrease in enzyme activity with an increase in the carbon number of the long chain base persisted. These results demonstrate that sialyltransferase has binding specificity with respect to the long chain base, but not the fatty acid. The apparent molecular species towards the fatty acid is related to the aqueous solubility of the various LacCer molecular species.     

The membrane-binding domain of the Rous sarcoma virus Gag protein.

The Gag protein of Rous sarcoma virus (RSV) can direct particle assembly and budding at the plasma membrane independently of the other virus-encoded products. A previous deletion analysis has suggested that the first 86 amino acids of RSV Gag constitute a large membrane-binding domain that is absolutely required for these processes. To test this hypothesis, we inserted these residues in place of the N-terminal membrane-binding domain of the pp60v-src, a transforming protein whose biological activity requires plasma membrane localization. The ability of the Src chimera to induce cellular transformation suggests that the RSV sequence indeed contains an independent, functional domain.     

The lipids of the Golgi apparatus subfractions from rat liver.

Golgi apparatus were isolated from untreated rat liver and separated into three fractions. One consisted mainly of vesicles, a second of tubular particles (dictyosomes) and the third was a mixed fraction. Large differences between these fractions could be seen in the electron microscope and by enzyme analysis. The total lipid content of the vesicles was 3.5-times greater than that of the dictyosomes and the neutral lipid value was 7-times greater. The ratio of phospholipids to protein was approximately the same in the three fractions. However, the phospholipid patterns differed between the vesicle and dictyosome fractions.     

The modulating effects of lipids on purified rat liver Golgi galactosyltransferase

Galactosyltransferase was purified from rat liver Golgi membranes. The Triton X-100, used to solubilize the enzyme was removed immediately prior to the lipid interaction studies. In lipid vesicles, prepared from a variety of phosphatidylcholines (PCs), including egg PC, DOPC, DMPC, DPPC and DSPC, the ability of the lipids to stimulate the enzyme decreased in the order egg PC greater than DOPC greater than DMPC greater than DPPC greater than DSPC, i.e. the lower the transition temperature (Tc) the greater the stimulation of the enzyme. A second, neutral lipid, phosphatidylethanolamine was used to permit a comparison of the effect of a different head group of the same net charge at neutral pH. The PEs included, egg PE, soy PE, Pl-PE, PE(PC) and DPPE in order of increasing Tc. The effect of the PEs was opposite to that of the PCs, i.e. the higher the Tc, the greater the stimulation of the enzyme. In fact egg PE and soy PE which have the lowest Tc values were inhibitory. Thus the modulation of the Golgi membrane galactosyltransferase by these lipids was different from that reported earlier for the bovine milk galactosyltransferase. The effects of two acidic lipids, egg phosphatidic acid (PA) and egg phosphatidylglycerol (PG) were studied also. Both totally inhibited the enzyme even at low concentrations of lipid, however, the PA was more effective than PG. In mixtures of neutral lipid (PC) and acidic lipid (PA or PG), the effect of the acidic lipid dominated. Even in the presence of excess PC, total inhibition of the enzyme was observed. It was concluded that the enzyme bound the acidic lipid preferentially to itself. The choice of the lipids allowed us to make several direct comparisons concerning the effect of the nature of the lipid head group on the activity of the enzyme. For example PE(PC), egg PA and egg PG would have fatty acid chains identical to egg PC since these three lipids are all prepared by modification of egg PC. As well, DPPE differs from DPPC only by nature of the head group. These comparisons indicated that not only the net charge but also chemical nature of the head group were important in the lipid modulation of Golgi galactosyltransferase.     

Targeting of active sialyltransferase to the plant Golgi apparatus.

Glycosyltransferases in the Golgi apparatus synthesize cell wall polysaccharides and elaborate the complex glycans of glycoproteins. To investigate the targeting of this type of enzyme to plant Golgi compartments, we generated transgenic Arabidopsis plants expressing alpha-2,6-sialyltransferase, a glycosyltransferase of the mammalian trans-Golgi cisternae and the trans-Golgi network. Biochemical analysis as well as immunolight and immunoelectron microscopy of these plants indicate that the protein is targeted specifically to the Golgi apparatus. Moreover, the protein is predominantly localized to the cisternae and membranes of the trans side of the organelle. When supplied with the appropriate substrates, the enzyme has significant alpha-2,6-sialyltransferase activity. These results indicate a conservation of glycosyltransferase targeting mechanisms between plant and mammalian cells and also demonstrate that glycosyltransferases can be subcompartmentalized to specific cisternae of the plant Golgi apparatus.     

Developmental control of N-CAM sialylation state by Golgi sialyltransferase isoforms

A rat brain Golgi sialyltransferase activity capable of the differentiation-dependent control of N-CAM sialylation state is described. The specific activity of Golgi sialyltransferase was found to be developmentally regulated with respect to both endogenous and exogenous protein acceptors, with a particular elevation on postnatal days 10-12 when the heavily sialylated or 'embryonic' form of N-CAM is re-expressed. The subsequent developmental decrease in activity was associated with a significant decrease in apparent Km for the CMP-NeuNAc substrate, but not for the asialofetuin exogenous acceptor, which could not be attributed to the temporal expression of an endogenous competitive inhibitor. The apparent Vmax remained constant for CMP-NeuNAc but was significantly reduced for asialofetuin. Sialyltransferase activity, which was optimal at pH 7.0-7.5, was also modulated by various cations. Zinc abolished enzyme function, in contrast to ferric ions which stimulated activity fourfold-sevenfold. The marked activation of the adult form of the enzyme by potassium and magnesium ions, together with the alterations in kinetic constants, suggested this activity to be distinct from that derived from postnatal day-12 tissue. The kinetics of [14C]sialic acid incorporation into immuno-precipitated N-CAM demonstrated the individual polypeptides to be sialylated, possibly by addition of polysialosyl units, in a developmental sequence. The presence of four distinct sialyltransferase activities was demonstrated by non-denaturing gel electrophoresis followed by solid-phase enzyme assay. These isoforms were temporally expressed during development, two being correlated with the postnatal reexpression of the 'embryonic' form of N-CAM.     

Purification of detergent-solubilized form and membrane-binding domain of rat gamma-glutamyltransferase by immuno-affinity and hydrophobic chromatography

A new method to purify papain- or detergent-solubilized form (papain or detergent form) of gamma-glutamyltransferase from rat hepatomas as well as from rat kidney by immuno-affinity column chromatography is presented. The antibody-column was prepared by coupling the anti-kidney papain form antibody, which had been purified by using a kidney papain form-Sepharose column, to CNBr-activated Sepharose 4B. The enzyme bound to the antibody-column was eluted with 0.04 M NH4OH. By this method, detergent forms were purified 300 and 1600-fold in approx. 50% yields from rat kidney and rat ascites hepatoma AH 13, respectively, and the papain form was also purified 16 000-fold in a similar yield from primary hepatoma which has a very low activity of this enzyme. Preparations thus obtained apparently did not contain any peptide other than heavy and light subunit peptides of this enzyme on SDS-polyacrylamide gel electrophoresis. The detergent form of kidney enzyme was preferentially absorbed to a hydrophobic column of aminooctyl-Sepharose, while the papain form was not, suggesting that the detergent form might be adsorbed to the column through hydrophobic interaction of the membrane-binding domain. The domain peptide was also purified by the hydrophobic column after release from the detergent form by papain treatment. The molecular weight of the peptide was estimated to be about 16 000 on SDS-polyacrylamide gel electrophoresis. On double immunodiffusion, the domain peptide reacted with anti-detergent form antibody but not with anti-papain form antibody. The domain-specific antibody was also purified from the anti-detergent form antibody.     

The effect of cycloheximide on galactosyltransferase of rat liver Golgi membranes.

An inhibitory effect of cycloheximide on the initial rate of galactosyltransferase of rat liver Golgi membranes has been demonstrated. Cycloheximide was effective in inhibiting the activity of the enzyme when added directly to the assay medium or after pre-incubation of the membranes with the drug. The inhibition observed with different concentrations of nucleotide sugar was shown to be competitive at higher concentrations of the nucleotide sugar (0.10-1.0 mumol). The inhibition observed with different concentrations of acceptor, N-acetylglucosamine was complex and could not be analysed further with the present data. Washing the Golgi membranes previously incubated with cycloheximide with water failed to reverse the inhibition. Washing with UDPgalactose partially reversed the inhibition only. These results, together with the observation that serum galactosyltransferase was not inhibited by cycloheximide supported the view that the cycloheximide effect may be primarily on the membrane system.     

Topology of mannosidase II in rat liver Golgi membranes and release of the catalytic domain by selective proteolysis

The orientation of mannosidase II, an integral Golgi membrane protein involved in asparagine-linked oligosaccharide processing, has been examined in rat liver Golgi membranes. Previous studies on mannosidase II purified from Golgi membranes revealed an intact subunit of 124,000 daltons, as well as a catalytically active 110,000-dalton degradation product generated during purification (Moremen, K. W., and Touster, O. (1985) J. Biol. Chem. 260, 6654-6662). In Triton X-100 extracts of Golgi membranes, the intact enzyme was cleaved by a variety of proteases to generate degradation products similar to those observed previously. At appropriate concentrations, chymotrypsin, pronase, and proteinase K generated 110,000-dalton species, while trypsin and Staphylococcus aureus V8 protease generated 115,000-dalton forms. Cleavage by chymotrypsin under mild conditions (10 micrograms/ml, 10 min, 20 degrees C) resulted in a complete conversion to a catalytically active 110,000-dalton form of the enzyme which was extremely resistant to further degradation. Attempts to demonstrate these protease digestions in nonpermeabilized Golgi membranes were unsuccessful, a result suggesting that the protease-sensitive regions are not accessible on the external surface of the membrane. In Golgi membranes permeabilized by treatment with 0.5% saponin, mannosidase II could readily be cleaved to the 110,000-dalton form by digestion with chymotrypsin under conditions similar to those which result in a proteolytic inactivation of galactosyltransferase, a lumenal Golgi membrane marker. Although mannosidase II catalytic activity was not diminished by this chymotrypsin digestion, as much as 90% of the enzyme activity was converted to a nonsedimentable form. To examine the effect of the proteolytic cleavage on the partition behavior of the enzyme, control and chymotrypsin-treated Triton X-114 extracts of Golgi membranes were examined by phase separation at 35 degrees C. The undigested enzyme partitioned into the detergent phase consistent with its location as an integral Golgi membrane protein, while the 110,000-dalton chymotrypsin-digested enzyme partitioned almost exclusively into the aqueous phase in a manner characteristic of a soluble protein. These results suggest that mannosidase II catalytic activity resides in a proteolytically resistant, hydrophilic 110,000-dalton domain. Attachment of this catalytic domain to the lumenal face of Golgi membranes is achieved by a proteolytically sensitive linkage to a 14,000-dalton hydrophobic membrane anchoring domain.     

The purification and characterization of mannosidase IA from rat liver Golgi membranes

Rat liver Golgi membranes contain two alpha 1,2-specific mannosidases (IA and IB) (Tulsiani, D. R. P., Hubbard, S. C., Robbins, P. W., and Touster, O. (1982) J. Biol. Chem. 257, 3660-3668). Mannosidase IA has now been purified to apparent homogeneity by detergent extraction and (NH4)2SO4 precipitation, followed by Sephacryl S-300, ion-exchange, and hydroxylapatite chromatography. The enzyme was homogeneous by nondenaturing polyacrylamide gel electrophoresis with different gel concentrations, and Ferguson plot analysis indicated an Mr of 230,000 for the native enzyme. Although electrophoresis under denaturing conditions generally gave a subunit Mr of 57,000, electrophoresis of less than 1 microgram of protein yielded a faint doublet of Mr 57,000 and 58,000. Thus, the enzyme appears to be a tetramer with four very similar subunits. The enzyme bound to concanavalin A-Sepharose 4B only when it was kept in contact with the lectin for 16 h. Endoglycosidase H treatment resulted in loss of its binding to the lectin, without leading to a detectable change in the size of the enzyme subunit. On electrophoretic gels, the enzyme gave a faint positive stain with periodic acid-Schiff's base. The enzyme contained about 0.9% hexose by direct analysis. It did not bind to affinity resins specific for neuraminic acid, galactose, or N-acetylglucosamine. All these studies suggest that the enzyme is a glycoprotein containing only one or two clusters of high mannose oligosaccharide. Mannosidase IA is active toward oligosaccharides containing alpha 1,2-linked mannosyl residues. [3H]Man9GlcNAc, [3H] Man8GlcNAc, [3H]Man7GlcNAc, and [3H]Man6GlcNAc are good substrates. Man9GlcNAc, the best substrate, yields Man8, Man7, and Man6 derivatives with structures suggesting that the sequence of release of mannose residues is rather specific. Immunoprecipitation studies using polyclonal antibody (IgG) prepared against homogeneous mannosidase IA cross-reacted with mannosidase IB, a result suggesting that these two enzymes share antigenic determinants. However, no cross-reactivity was observed with rat liver cytosolic and lysosomal alpha-D-mannosidases or with Golgi mannosidase II.