Carbohydrate Structure of Vesicular Stomatitis Virus Glycoprotein* · 2002-12-11 · Carbohydrate Structure of Vesicular Stomatitis Virus Glycoprotein* (Received for publication,

THE JOURNAL OF BIOLM;ICAL CHEMIVTRY Vol. 253, No. 16. lseue of August 25, pp. 56W-5612. 1978 Printed m L’.S.A.

Carbohydrate Structure of Vesicular Stomatitis Virus Glycoprotein*

(Received for publication, February 27, 1978)

Christopher L. Reading,+ Edward E. Penhoet, and Clinton E. Ballou From the Department of Biochemistry, University of California, Berkeley, California 94720

The carbohydrate structure of the membrane glycoprotein of vesicular stomatitis virus (New Jersey serotype) has been determined on material purified from virions grown in spinner cultures of BHKzl cells in the presence of [3H]glucosamine. The intact 3H-glycoprotein contains 6 residues of N-acetyl-n-neuraminic acid, 6 of D-galactose, 6 of D-mannose, 2 of L-fucose, and 10 of N-acetyl-n-glucosamine. Neither D-ghCOSe nor N- acetyl-D-galactosamine was detected. The carbohydrate was not susceptible to fl elimination, and glycopeptides were obtained that yielded 1 mol of aspartic acid/m01 of glycopeptide after acid hydrolysis, both of which indicate that the sugar is linked to the amide nitrogen of asparagine in the glycoprotein. Hydrazin- olysis of the glycoprotein yielded a single major oligosaccharide that, after re-N-acetylation, contained 3 residues of N-acetylneuraminic acid, 3 of galactose, 3 of maxmose, 1 of fucose, and 6 of N-acetylglucosamine, suggesting that the glycoprotein contains two identical oligosaccharide chains. The linkages in the hydrazinolysis oligosaccharide were established by methylation analysis. The oligosaccharide was fragmented by partial acid hydrolysis and was specifically depolymerized at de-N-acetylated glucosamine residues by nitrous acid dean&nation, the fragments obtained at each step were purified, and their carbohydrate compositions were determined. A glycopeptide fraction, obtained by digestion of the glycoprotein with trypsin and pronase, was separated on the basis of amino acid differences into two components with identical sugar compositions that are assumed to represent units from the two different sites of glycosylation in the protein. The glycopeptides were subjected to sequential exoglycosidase digestion with a-D-neuraminidase, P-D-galactosidase, F-D-hexosaminidase, a-D-mannosidase, fi-n-mannosidase, and a-Mucosidase. After removal of the N-acetylneuraminic acid, the galactose, and part of the N- acetylglucosamine, the glycopeptides were cleaved with an endo-N-acetyl-/3-D-glucosaminidase. All fragments were reisolated after each digestion step and analyzed for carbohydrate composition or for linkages by methylation analysis. The glycoprotein, glycopeptides, and hydrazinolysis oligosaccharide were oxidized with periodate and subjected to Smith degradation, and the compositions and linkages of the products were determined. The results establish that the glycoprotein contains two identical carbohydrate chains that are composed of three outer chain units, a mannose-rich

* This work was supported by National Science Foundation Grant BMS74-18893 and United States Public Health Service Grant AI- 12522 (to C. E. B.) and by National Science Foundation Grant GB- 38658 and a grant from the Tuberculosis and Respiratory Disease Association of California (to E. E. P.). The costs of nublication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accord- ance with 18 U.S.C. Section 1734 solely to indicate this fact.

.j Present address, Department of Developmental and Cellular Biology, University of California, Irvine, Calif. 92717.

core, and a peptide linkage unit, with the structures

[crNeuNAc+3gGal+46G1cNAc+]3

aMardt?Matw'+~GlcNAc+ t3

aMan

crFuc

l?cNAc+Asn

and that these are interlinked to form an oligosaccharide with the following structure.

aNeuNAc+36Gal+4gG1cNAc aFuc +4 +6 I

aNeuNAc~3gGal+4gG1cNAc+2aMa~66Ma~4BG1cNAc~4G1cNAc-tAsn t3

aNeuNAc+3gGa1+46G1cNAc+2aMan I

Vesicular stomatitis virus is the prototype of the rhabdovi- ruses (1). The purified virions contain RNA, protein, carbohydrate, and lipid. The RNA is a single molecule with a molecular weight of 3.6 X lo6 (2). VSV’ contains five proteins, labeled L, G, N, NS, and M, with molecular weights of 150,000, 70,000, 50,000, 49,000, and 30,000, respectively (3). The L, N, and NS proteins are components of the ribonucleoprotein, whereas the G and M proteins are associated with the viral coat. The G protein is the only glycoprotein in the virus. The lipid and glycolipid compositions of the envelope generally reflect those of the host cell membrane (4,5). The glycoprotein is reported to contain N-SCetyl-D-glUCOSSmiUe, D-mannose, D-galactose, L-fucose, N-acetyl-D-neuraminic acid, as well as D-&COSe and N-acetyl-D-galactosamine (6-8).

The carbohydrate components of viral glycoproteins are thought to be involved in the hemagglutination of erythrocytes and the absorption of virus to host cells during infection (9-l 1). Removal of the G protein from VSV by digestion with proteases (12) renders the virions noninfectious, and infectivity can be restored by reconstitution with purified glycoprotein (13). The importance of terminal sialic acid residues on the G protein for hemagglutination and absorption is suggested by the fact that neuraminidase digestion drastically lowers VSV infectivity, agglutination of goose erythrocytes, and viral attachment to mouse L-cells, and these properties are restored by resialylation of the virus with a purified sialyltransferase (14-16).

A complete structure has not been reported for the carbohydrate component of the G protein. The evidence indicates that there are two nitrogen-linked oligosaccharide chains, with

’ The abbreviations used are: VSV, vesicular stomatitis virus; BHK, baby hamster kidney; DME medium, Dulbecco’s modified Eagle’s medium; Hepes, 4-(2lhydroxyethyl)-I-piperazine-ethanesulfonic &id; JME medium. Joklik’s modified Eagle’s medium; PBS, phosohate- buffered saline; Tricine, N-[tris(hydr&ymethyl)methyl]giycine; dansyl, 5-dimethylaminonaphthalene-1-sulfonyl.

5600

by guest on June 29, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Vesicular Stomatitis Virus Glycoprotein Structure

molecular weight about 3000, that are similar in structure to the chains of serum glycoproteins (17-19). In the present study, we show that VSV glycoprotein has two identical carbohydrate chains and we confii that they are linked to asparagine residues in the protein. Moreover, we have determined all of the sequences, linkages, and anomeric configu- rations, thus allowing the assignment of a unique structure.

EXPERIMENTAL PROCEDURES

Cell Culture-Baby hamster kidney cells, clone 21, subculture 13 (BHK cells) were obtained from Dr. Tad Wiktor, Wistar Institute of Anatomy and Biology, Philadelphia, Pa., and were maintained as a monolayer culture in glass prescription bottles using Dulbecco’s modified Eagle’s medium (Gibco H21) supplemented with 10% fetal bovine serum (Irvine Scientific Co.) and buffered with 15 rnM Hepes (Sigma).

A suspension culture of BHK cells (BHK spinner cells) was derived from the monolayer culture as follows. Ten 32-ounce bottles of confluent BHK monolayer cells (1.2 x 10’ cells) were incubated for 5 min at 23°C with 0.5% trypsin in phosphate-buffered saline (20) lacking Ca’+ and Mg” and containing 5 mu EDTA. The cells were pipetted up and down several times, to obtain a suspension of single cells, and resuspended in 100 ml of Joklik’s modified Eagle’s medium (Flow Laboratories) supplemented with 15 mM Tricine (Sigma), pH 7.4, and 20% fetal bovine serum. The cells were maintained in suspension in a 500~ml serum bottle with a 1.5-inch Teflon-coated seamless magnetic stirring bar rotated at 100 rpm. Faster rates killed the cells and slower rates allowed clumping. The cells were subcultured every 24 h by diluting the suspension into an equal volume of fresh medium. For large scale cultures (up to 40 liters), the cells were grown in 4- liter Erlenmeyer flasks, and the volumes were doubled daily with prewarmed medium.

Virus Production and Purification-The New Jersey serotype of vesicular stomatitis virus was obtained from Dr. John Holland, De- partment of Biology, University of California, San Diego, and selected free of defective virus by picking large plaques on three successive high dilution passages. Stock cultures were grown in glass prescription bottles by infecting PBS-washed, subconfluent, monolayer BHK cultures with virus at a multiplicity of infection of 0.01 (0.01 plaque- forming units/cell) in 2 ml of 2% calf serum in Hepes-buffered DME medium. The bottles were rocked periodically to bathe the cells with the virus suspension. After 30 min, the virus suspension was aspirated and the monolayer was washed with 5 ml of medium. After the wash liauid was asairated. 10 ml of 2% calf serum in Henes-buffered DME medium was-added and the cells were incubated at 37°C. After 24 h, the virus-containing medium was collected and centrifuged at 10,090 x g for 20 min at 4’C. The supernatant fluid was divided into portions and stored at -70°C.

Plaque Assays-Virus solutions were titered immediately after production on subconfluent monolayers of BHK cells (3.5 x lo” cells/well) in Linbro FB-16-240TC multidish trays (Flow Laborato- ries) as follows. The medium was aspirated from the wells, and 0.2 ml of the virus diluted in 2% calf serum in Hepes-buffered DME medium was added to three wells for each dilution. The plates were rocked periodically to bathe the cells with the virus suspension. After 30 min, 1 ml of 0.35% agarose in Hepes-buffered DME medium, supplemented with 2% calf serum, was added at 43°C. The agarose gelled at 23°C and the plates were placed in a 37°C incubator. After 16 h, the agarose was removed with a metal spatula, care being taken not to disturb the cell layer, and a few drops of 0.1% crystal violet in 50% ethanol were added to each well. After 2 to 5 min, the plates were rinsed gentlv with running tap water. Plaques appeared as l- to 2-mm clearings in the viol&-stained monolayers.

Labeling of Virus-BHK spinner cells were infected with a multiplicity of infection of 1.0 and, 4 h later, 100 ml of the cell suspension w-as centrifuged at 250 x g in a clinical centrifuge to collect the cells. The medium was aspirated, and the cells were resuspended in 90 ml of a medium containing the normal components of JME medium with 15 mu Tricine, except that the “C-amino-acids were replaced with 0.25 mCi of a mixture of uniformly labeled “C-amino-acids (NEC-445, 10 Ci/mol, New England Nuclear) and fetal bovine serum that had been extensively dialyzed against PBS was used. The cells were stirred in a 100~ml serum bottle for 2 h, and then 10 ml of complete medium was added. The labeled virus was harvested at 24 h and virus from a parallel unlabeled culture was added as carrier before purification.

To label the cells with radioactive glucosamine, 100 ml of cell suspension was harvested and resuspended 12 h before infection in a

glucose-free medium containing 4 times the normal amount of amino acids and dialyzed fetal bovine serum. Four h after infection, 100 &i of n-[B-‘H]glucosamine HCl (NET-190, 10 Ci/mmol, New England Nuclear) was added in 1 ml of medium, and 2 h later 10 ml of complete medium was added. The virus was harvested 24 h after infection.

Purified virions were labeled with carrier-free Na? by the procedure of Cuatrecasas (21). The labeled virions were separated from the Na’*“I by gel filtration on a Sephadex G-25 column. Electron microscopy was performed on a Siemens Elmiskop 1A electron microscope by A. Taylor of this department.

General Methods-Protein was determined according to Lowry et al. (22), in the presence of 0.1% sodium dodecyl sulfate to avoid interference from Triton X-100 (23). Amino groups were estimated by the ninhydrin method (24). Galactose was determined with a Galactostat kit (Worthington), N-acetylglucosamine according to Ghosh et al. (25) without acetylation, fucose by the Dische-Shettles cysteine-sulfuric acid reaction (26), total hexose by the phenolsulfuric acid method (27), and reducing sugar by the Park-Johnson ferricya- nide method (28).

“H and 14C radioactivities were determined in a Beckman LS-3150 liquid scintillation counter with a dioxane-based counting solution (29). ““I radioactivity was measured in a Searle model 1185 Auto- Gamma system.

Column Chromatography and Electrophoresis-Bio-Gel P-2 (-400 mesh), P-6 (200 to 400 mesh), P-10 (200 to 400 mesh), P-100 (100 to 200 mesh), and AG 5OW-X2 (200 to 400 mesh) were from Bio- Rad. Gel filtration of the glycoprotein was done on a Bio-Gel P-100 column (2 x 20 cm) in 0.1% sodium dodecyl sulfate and oligosaccharides and their degradation products were separated on a Bio-Gel P- 6 column (1.6 x 120 cm) or a Bio-Gel P-2 column (1.6 x 120 cm) by elution with water, whereas the glycopeptides were fractionated on columns (1.6 x 120 cm) of Bio-Gel P-10, P-6, or P-2 by elution with 0.1 M ammonium acetate, pH 7.0. AU elutants contained 0.1% toluene to suppress bacterial growth.

Polyacrylamide gel electrophoresis was done on slab gels with a discontinuous buffer system (30). Protein-containing bands were vis- ualized with Coomassie brilliant blue R-250 stain (Bio-Rad), and carbohydrate-containing bands were detected with the periodate- Schiff reagent (31). Radioactivity was determined in a stained gel by drying it on Whatman 3MM paper (32) and cutting the individual lanes into l-mm horizontal strips. The strips were added to a vial along with 0.05 ml of water, followed by 10 ml of a fluor composed of 15.2 g of Omnifluor, 143 ml of NCS solubilizer, and 3 kg of toluene. The vials were capped and incubated at 50°C for 24 h, cooled at 4”C, and counted.

Gas-Liquid Chromatography and Mass Spectrometry- Monosaccharides were analyzed as volatile derivatives in a Varian 1400 gas-liquid chromatograph equipped with a flame ionization detector and a linear temperature programmer. A splitter valve was arranged to introduce part of the sample through a metal jet separator into a DuPont 21-491 mass spectrometer.

The 0-trimethylsilyl ethers of methyl glycosides were analyzed on a glass column (4 feet x l/8 inch) packed with 3% (w/w) SE-30 (Wilkins Instruments and Research, Inc.) on 100 to 110 mesh ABS (Analabs). The temperature was programmed from 120 to 18O’C at 0.5”C/min. Alditol acetates were separated on a glass column (4 feet X l/8 inch) of 3% (w/w) OV225 (Varian Aerograph) on 100 to 120 mesh Gas-Chrom Q (Applied Science Laboratories). The column temperature was maintained at 190°C for 32 min and then shifted to 210°C. Partially methylated alditol acetates were separated isother- mally between 150 and 200°C on a glass column (4 feet X l/8 inch) of Supelcoport coated with 3% (w/w) OV210 (Varian Aerograph). The trimethylsilyl derivatives of sialic acids were separated on the same column at 190°C.

Methanolysis and formation of trimethylsilyl ethers were done according to Clamp et al. (33), as modified by Etchison and Holland (7). Reactions were done in glass tubes with Teflon-lined screw caps, which were flushed with Nz, capped, and sealed with Teflon tape. Peak areas were measured relative to an internal mannitol standard and were corrected according to Clamp et al. (33).

To prepare alditol acetates, the oligosaccharide, containing 50 to 100 pg of carbohydrate, was dissolved in 2 ml of 1 N trifluoroacetic acid, the tube was flushed with Nz, capped, and sealed with Teflon tape. The sample was hydrolyzed at 12O’C for 90 min. and the solution was cooled, frozen, and lyophilized. The samples were reduced and acetylated according to Stellner et al. (34). Carbohydrate- containing compounds were permethylated by the procedure of Hak- omori (35), with the modifications for amino sugars described by


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

5602 Vesicular Stomatitis Virus Glycoprotein Structure

Stellner et al. (34). Sources of Glycopeptides, Oligosaccharides, and Sugar Deriva-

tiues-The disaccharide, ManGlcNAc, and the pentasaccharide, Man,GlcNAc, were isolated as described by Nakajima and Ballou (36) from yeast mannan FII core fragment. Immunoglobulin M complex glycopeptide (IgM.) was a gift from R. E. Cohen of this department. The glycopeptide, from the mixed glycosidase digestion of the IgM. glycopeptide with neuraminidase, P-n-galactosidase, and exo- P-n-hexosaminidase, was digested with endo-N-acetyl-P-u-glucosaminidase. A tetrasaccharide, MamGlcNAc, was obtained by gel filtration of the products. The trisaccharide, ManzGlcNAc, was supplied by A. Lundblad (University Hospital, Lund, Sweden) who isolated it from the urine of a patient with mannosidosis (37).

p-Nitrophenyl WL- and /.I-n-fucopyranosides, b-D-ghCOpyKUIO- sides, a- and P-n-gaIactopyranosides, a- and /3-N-acetyl-2-acetamido- 2-deoxy-n-glucopyranosides, a- and /I-N-acetyl-2-acetamido-2-deoxy- n-galactopyranosides, and cu-n-mannopyranoside were obtained from Sigma. p-Nitrophenyl P-n-mannopyranoside was prepared according to Rosenfeld and Lee (38). N-[‘H]Dansyl ovalbumin glycopeptides were prepared by the method of Tarentino and MaIey (39).

Enzyme Sources-Trypsin (type XII) was from Sigma and pronase (B grade) from Calbiochem. a-Neuraminidase (40) from Clostridium perfringens strain 10543, /I-Neuraminidase (41) and a-n-mannosidase (42) from jack bean meal, and endo-N-acetyl-P-D-glucosaminidase (43) from Diplococcus pneumoniae attenuated strain 645 were purified according to the published procedures.

/?-D-Galactosidase from Aspergillus niger (44) was purified from Calbiochem P-D-galactosidase (B grade), and a+fucosidase (45) was isolated from bovine kidney (46). The enzyme was further purified by affinity chromatography (47) on l -aminocaproylfucopyranosylamine (48) coupled to Sepharose 4B activated with CNBr (49).

P-n-Mannosidase was obtained from Glusulase (Endo Laborato- ries) that was free of o-D-mannosidase activity but contained about 30 units/ml of /.I-D-mannosidase activity (1 unit of enzyme will hydro- lyze 1 pmol ofp-nitrophenyl P-D-mannopyranoside/min at 37°C in 50 mu sodium citrate buffer, pH 3.5). The enzyme was eluted from a Sephadex G-75 column just after the first protein peak, which contained most of the other glycosidase activities present. It was only slightly retarded on a Sephadex A-50 column when loaded in 50 mM sodium phosphate buffer, pH 7.0, but it was eluted ahead of the majority of the protein, which was more strongly absorbed. The eluted peak of activity was concentrated by dialysis against polyethylene glycol 6000, dialyzed against 10 mM sodium citrate buffer, pH 6.0, containing 0.1 M NaCl and 0.1% toluene, and stored at 4°C. The purified enzyme had a specific activity of 60 units/mg of protein.

RESULTS

Studies on Intact Glycoprotein Production and Purification of Vesicular Stomatitis Vi-

rus-For the production of large amounts of VSV, 3-liter spinner cultures of BHK cells were infected at a density of lo6 cells/ml with a multiplicity of infection of 0.1, and the cultures were diluted to 4 liters with prewarmed medium. After 24 h, the cells were pelleted at 10,000 x g for 20 min at 4°C. The supematant liquid containing the virus (up to 10” plaque forming units/ml) was decanted and the virus was precipitated in 7.5% polyethylene glycol 6000 (PEG 6000) and 0.5 M NaCl according to McSharry and Benzinger (50). The precipitate was collected by centrifugation at 10,000 x g for 20 min at 4”C, and the viral pellet was resuspended by repeated pipetting in 10 mu Tris-Cl, pH 7.4, containing 0.1 M NaCl and 1 mu EDTA. The virus suspension was sonicated (five l-s bursts in a Branson W185 cell disrupter), then centrifuged at 2,060 x g for 5 min to remove any insoluble material, and the milky white viral suspension was centrifuged at 28,000 rpm in a Spinco 30 rotor for 1 h at 4°C. The viral pellet was resuspended by pipetting in the Tris buffer. Seven milliliters of this suspension was layered onto a preformed 25-ml NaK tartrate linear gradient (15 to 33% w/w) and centrifuged to isopycnic equilibrium (at least 6 h) in a SW 25.1 rotor at 22,500 rpm. The dense white viral band was collected by piercing the tube with an 18-guage needle attached to a lo-ml plastic syringe

and withdrawing the virus in a volume of 5 to 8 ml. The virus suspension from the tartrate gradient was diluted with 3 volumes of the Tris buffer and pelleted by centrifugation at 28,000 rpm in a Spinco 30 rotor for 1 h at 4°C. The overall yield of infectivity was about 60%.

Purity of the virus preparation was assessed in several ways. Transmission electron microscopy of samples, negatively stained with 2% phosphotungstic acid without prior fixation, revealed only virions. Only the five proteins of VSV were observed when sodium dodecyl sulfate gel electropherograms were stained for protein (Fig. 1). If the virus was grown in the presence of [“Hlglucosamine, or the purified virions were radioiodinated using conditions under which only the exposed membrane proteins are iodinated (51), a single radioactive peak was observed on electropherograms, at a position identical to that of the glycoprotein. A single band in the same position was observed if the gel was stained for carbohydrate (data not shown). These tests indicate that the virus preparation was not contaminated by a significant amount of cellular protein.

Isolation of Glycoprotein-The viral pellet was resuspended in 50 ml of 10 mu T&-Cl, pH 8.0, containing 10 pM

phenylmethylsulfonyl fluoride. Five milliliters of 20% Triton X-100 was added, and the suspension was stirred at 23’C for 1 h to release the glycoprotein from the virions (52). The viral cores were removed by centrifugation at 48,000 rpm in a Spinco 50 rotor for 1 h at 4”C, and the supematant solution was lyophilixed.

The lyophilized Triton extract was resuspended in 1 ml of water and the glycoprotein was precipitated with 10 volumes of acetone. The precipitate was collected by centrifugation (1,000 x g, 5 min) and washed in hot absolute ethanol. The precipitate was suspended in water:I-butanol (1:l) and sonicated, and the two phases were separated by centrifugation. The precipitated glycoprotein, which collected at the interface, was isolated and this extraction with the butanol/water mixture was repeated twice. The glycoprotein was further extracted in the two-phase system of chloroformmeth-

L*

DYE+ FIG. 1. Polyacrylamide gel electrophoresis in sodium dodecyl sul-

fate of the viral components before and after treatment with Triton X-100. Purified virus was treated with Triton, the solution was ultra- centrifuged, and samples of the resuspended pellet and the supernatant fraction were analyzed. 1, purified virus prior to the addition of Triton; 2, resuspended viral cores after centrifugation of the Triton- extracted virus; 3, supernatant fraction after centrifugation of the T&on-extracted virus. L, G, N, NS, and M, VSV proteins; DYE, tracking dye.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Vesicular Stomatitis Virus Glycoprotein Structure 5603

anokwater (3:2:1), and the material at the interface was collected and sonicated in chloroform:methanol:water (3:3:1). The precipitate was collected by centrifugation, dried in a stream of Ne, and stored at -20°C.

The glycoprotein preparation was examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. After Tri- ton X- 100 treatment and ultracentrifugation, the supernatant extract contained only the glycoprotein, whereas the other virion proteins were in the pelleted fraction (Fig. 1). Glyco- protein, isolated from VSV produced in the presence of 14C- ammo-acids, was the only radioactive protein observed in the Triton supematant fraction.

Carbohydrate Composition-Carbohydrate was determined by gas-liquid chromatography of the trimethylsilyl ethers of the methyl glycosides obtained by methanolysis. Fig. 2 shows a representative chromatogram, and the averaged values for seven determinations on the intact glycoprotein are listed in Table I. The values for N-acetylglucosamine were close to 10 mol/mol, based on the dry weight of the sample and assuming M, = 69,000 for the glycoprotein. The values are normalized to 10.0 residues of N-acetylglucosamine/ glycoprotein molecule on the assumption of two sites of glycosylation, with 5 hexosamine residues/site. There were also approximately 2 mol of fucose, 6 of mannose, 6 of galactose,

17

19 n II 16

and 6 of sialic acid/IO mol of N-acetylglucosamine. Glucose and N-acetylgalactosamine were absent.

Periodate Oxidation-The glycoprotein was oxidized with 4 mu periodic acid in 1 ml of 2% Triton X-100 in the dark at 23°C for 24 h, and the reaction was terminated by a drop of glycerol. The oxidized glycoprotein was precipitated with 10 ml of acetone, washed in 2 ml of 1 M acetic acid, collected by centrifugation, and washed in hot ethanol. Perlodate treatment destroyed 83% of the fucose, 91% of the sialic acid, and 33% of the mannose (Table I). Galactose and N-acetylglucosamine were unaffected.

Alkali Stability of Glycosidic Linkage to Protein-The purified [3H]glucosamine-labeled glycoprotein in 1% sodium dodecyl sulfate was treated with 0.1 M NaBHd in 0.1 M NaOH at 23°C for 24 h, conditions that cause /3 elimination of sugars linked to serine and threonine. The solution was acidified and fractionated in 0.1% sodium dodecyl sulfate on a Bio-Gel P- 100 column (2 x 20 cm). The radioactivity was eluted in a single peak near the void volume of the column, indicating that no sugar was released that would have derived 3H from [3H]glucosamine (N-acetylglucosamine, N-acetylgalactosamine, or sialic acid). The recovered alkali-treated glycoprotein had a slightly decreased sialic acid content (Table I), but this sugar is known to be destroyed in alkaline solution (53). A

RETENTION TIME (MIN) FIG. 2. Gas-liquid chromatogram tracing of the trimethylsilyl and N-acetylneuraminic acid, 19. The bottom panel shows the com-

methyl glycosides and polyols. The top panel illustrates a typical ponent sugars of a sample of VW glycoprotein. Note the absence of separation of an equimolar mixture of standards. Peak identifications Peaks 10 and 11 (glucose) and 16 (N-acetylgalactosamine) in this are: arabinitol, 1 and 5, fucose, 2,3, and 4, mannose, 6 and 7; galactose, tracing. The multiple peaks of the methyl glycosides represent differ- 7, 8, 9, and I@ glucose, 9, 10, and II; mannitol, 12 and 13; N- ent anomers and ring forms, whereas the double peaks for the polyols acetylgalactosamine, 14 and 1s; N-acetylglucosamine, 15, 17, and 18; probably reflect incomplete substitution.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


lower value of fucose was obtained after /3 elimination, but this can be explained by loss of the volatile derivative during analysis (see “Discussion”). All of the mannose, galactose, and N-acetylglucosamine survived the treatment and, therefore, represent carbohydrate attached to protein by N-glycosidic linkages.

Studies on Isolated Oligosaccharides and Glycopeptides Hydrazinolysis and Nitrous Acid Deamination-The

[3H]gluco samine-labeled glycoprotein (9.0 X 10’ cpm, 450 nmol) was subjected to hydrazinolysis according to Yosizawa et al. (54) to break the N-glucosyl asparagine amide linkages and release the intact oligosaccharides, during which the acetamido linkages of the amino sugars were also cleaved (see Fig. 3). After the hydrazine was sublimed according to Schroe- der (55), the reaction product was desalted on a Bio-Gel P-6 column (1.6 X 120 cm) by elution with water, and 860 nmol of oligosaccharide was recovered. A portion of the recovered oligosaccharide (450 nmol) was re-N-acetylated in 1 ml of saturated NaHC03 with 0.05 ml of acetic anhydride for 1 h at 23”C, and subjected to gel filtration on a Bio-Gel P-6 column (1.6 X 20 cm) by elution with water. The carbohydrate composition of the re-N-acetylated oligosaccharide, normalized to 5.0 residues of N-acetylglucosamine (Table II), was almost identical to that of the glycoprotein.

A portion (410 nmol) of the hydrazinolysis oligosaccharide, which had not been re-N-acetylated, was deaminated with nitrous acid as described by Horton and Phillips (56) to depolymerize the molecule at the positions occupied by glu-

TABLE I Carbohydrate comnosition of VW &coprotein

Residuesreenaining

Monosaccharide 572%: n Range /3 Elimi- Perio- nation re-

action d;;&oi

Fucose Mannose &lactose N-Acetylglucosa-

mine N-Acetylneura-

minic acid

1.7 f 0.3 1.5-2.1 1.2 0.3 6.0 + 0.4 5.6-6.3 6.3 4.0 6.2 f 0.2 5.9-6.4 6.0 6.3

10.0” 10.oh lO.Ob

5.8 f 0.3 5.6-6.1 2.6 0.5

” Mel of monosacchtide/mol of glycoprotein, assuming M, = 69,000 for the glycoprotein.

‘Normalized to a value of 10.0 for N-acetylglucosamine, as described in the text.

cosamine (see Fig. 3) and produce fragments terminated at the reducing end by 2,5-anhydro-n-mannose (57). The products were reduced with borohydride and separated in water on a Bio-Gel P-2 column (1.6 x 120 cm) (Fig. 4). About 85% of the initial radioactivity was recovered in four fractions with the compositions in Table II. Fraction A contained a compound with a retention time by gas chromatography near that of sialic acid; it probably represents a deaminated form of this sugar. Due to its negative charge, it was eluted near the void volume of the Bio-Gel P-2 column. Fraction B was eluted in the tetrasaccharide region and contained approximately 3 mannoses/2,5-anhydromannitol, as well as small amounts of galactose and N-acetylglucosamine. This product represents the mannoses attached to N-acetylglucosamine in the core. The 2,5-anhydromannitol arises from the reduction of the 2,5- anhydromannose produced by the deamination of the glucosamine. Fraction C contained approximately 4 galactose residues and 1 fucose residue/4 residues of 2,5-anhydromannitol. Since this peak was eluted in the disaccharide region of the Bio-Gel P-2 column, it was probably a mixture of galactosyl + 2,5-anhydromannitol and fucosyl + 2,5-anhydromannitol, with the fit being 3 times as prevalent as the second. Fraction D contained only a trace of 2,5-anhydromannitol. The results indicate that the oligosaccharide contained both galactosyl-

aNeuNAc+36Gal+46GlcNAc aFuc

(A) aNe~NAc~16GaIt*6GIcNAc~z~~~~~6~~~-46G~cNAc+'t(cNAc-~~n

aNeuNAc+'6Gal+"6GlcNAc+%Man

1 a

aNeuNHZ+36Gal+46G1cNH2 aFuc

(B) oNeuNH2-)6Ga,+L6GlcN~*~z~~~~66~~~~6G,cN"~~4~~cN"*

aNeuNHp36Gal+46G1cNHpz~Man

J,

b aFuc

Peak C 6Gal+2,5-Anhydrcmannitol + 2,5-~~hydrmannltol

Peak B aMa~66:~IF+2.5-Anhydromannitol

oMan

Peak A Degraded slalic acid fragment

FIG. 3. Nitrous acid deamination of the hydrazinolysis oligosaccharide. The amide bonds in the glycoprotein (A) were degraded by hydrazinolysis (a), and the product (B) was degraded with nitrous acid ( b) to yield derivatives of 2,5-anhydromannose that were reduced with borohydride to give 2,5-anhydromannitol derivatives. The reaction mixture was separated into four peaks (as shown in Fig. 4) with the indicated compositions.

TABLE II Carbohydrate compositions of hydrazinolysis oligosaccharide and degradation products

Substance N-Acetyl- neuraminic acid Galactose MaIIIIW2 N-Acetyl-

glucosamine Fucose P&Anhydro- mannitol

Intact glycoprotein” 2.9 3.1 3.0 5.0 0.9 Hydrazinolysis oligosaccharide” 3.0 3.2 2.5 5.0 0.7 Periodate-oxidized hydrazinolysis oligosac- 0.5 2.7 1.9 5.0 0

charide” Nitrous acid deamination fragments*

Fraction A -( Fraction B I, d - - 3.1 1.0 Fraction C 1.1 0.3 1.0 Fraction D -p

Partial acid hydrolysis fragments’ Fraction A 3.0 5.0 Fraction B 3.0 3.0 Fraction C -p

n Molar ratios normalized to 5.0 mol of N-acetylglucosamine. p Contained only 2,5-anhydromannitol. ’ Molar ratios normalized to 2,5-anhydromannitol. Fractions are ‘Molar ratios normalized to 3 or 5 residues of N-acetylglucosamine

from Fig. 4. to agree with the size of the fragments assessed from gel filtration in ’ Contained only a degradation product of sialic acid. Fig. 5. ’ Present in a trace amount. p Contained only N-acetylglucosamine.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE III Methylation analysis of hydrazinolysis oligosaccharide

Peak on Area chromat- Methyl&d sugar under Linkages in 01

nmam” peak h gosaccharide i-

1 2,3,4-Trimethylfucose 0.3 Fuc -+ 2 2,3,4,6-Tetramethylga- 0.3 Gal-+

lactose 3 2,4,6-Trimethylgalactose 2.5 4 “Gal+ 4 3,4$Trimethylmannose 1.3 Man --)

T’ 5 3,6-Dimethylmannose 0.6 -+ 4Man +

t’ 6 2,4-Dimethylmannose 1.0 --t “Man 4

T” 7 3,6-Dimethyl-N-acetyl- 1.8 + 4GlcNAc -+

glucosamine 8 3-Methyl-N-acetylglu- JR

cosamine 0.2 + 4GlcNAc -+

n Separated as partially methylated alditol acetates and character- ized by retention times and fragment patterns in the mass spectrometer.

b These values have not been corrected for differences in detector response nor for differential destruction or loss during processing. Thus, their quantitation is not meant to reflect the composition of the original oligosaccharide.

,A. B

15 1

1 14c

Fraction

FIG. 4. Gel filtration of the products of nitrous acid deamination of the hydrazinolysis [“Hloligosaccharide. After reduction, the reaction mixture was fractionated on a column of Bio-Gel P-2 (-400 mesh, 1.6 x 120 cm) with water, and l&ml fractions were collected with V” at Fraction 40 and V, at Fraction 110. Samples of 0.1 ml were assayed for radioactivity in the sugars derived from virus labeled with [6-“HI- glucosamine. Analyses of the fractions are given in Table II. Fraction A was a degradation product of sialic acid, Fraction B was a tetrasaccharide (Man< -+ anhydromannitol), Fraction C was a mixture of disaccharides (Gal + anhydromannitol) and Fuc 4 anhydromannitol, and Fraction D was anhydromannitol.

and fucosyl-N-acetylglucosamine linkages, as well as an N- acetylglucosamine unit with 3 attached mannose units.

Partial Acid Hydrolysis-The re-N-acetylated hydrazinolysis 3H-oligosaccharide (350 nmol) was subjected to partial acid hydrolysis (50 mu HCI at 100°C for 4 h), and the products were separated on a Bio-Gel P-6 column (Fig. 5). A dark brown color which developed during the hydrolysis stayed at the top of the column. Three fractions were collected and each was purified on a Bio-Gel P-2 column, and their compositions are given in Table II. Fraction A contained mannose and N-acetylglucosamine in the molar ratio of 35; Fraction B contained mannose and N-acetylglucosamine in the molar

6 F

L

5605

Fraction FIG. 5. Gel filtration of the partially acid-hydrolyzed re-N-acety-

lated hydrazinolysis “H-oligosaccharide. The products were chromatographed in water on a Bio-Gel P-6 column, and analyses of the peak fractions are given in Table II. Fraction A is composed of 3 mannose and 5 N-acetylglucosamine residues, Fraction B is 3 mannose and 3 N-acetylglucosamine residues, and Fraction C is N-acetylglucosamine.

mine. A labeled compound containing fucose was not observed, and the radioactive sialic acid was not eluted from the column. Since sialic acid is destroyed during acid hydrolysis (53), the brown material that remained at the top of the column may have contained the degradation products. The recovery of radioactivity in glucosamine was 98%. The fact that sialic acid, galactose, and fucose were rapidly released indicates that they occur near the nonreducing termini of the oligosaccharide.

Periodate Oxidation-A sample of the re-N-acetylated hydrazinolysis oligosaccharide (100 nmol) was oxidized with 4 mM periodic acid in 1 ml of 0.1 M sodium acetate, pH 4.5, at 23°C in the dark. After 24 h, a drop of glycerol was added and the product was chromatographed on a Bio-Gel P-6 column (1.6 x 120 cm). Two radioactive peaks were observed. The first peak, which was eluted near the void volume of the column, contained 98 nmol of the oxidized oligosaccharide, and it was treated with NaBEL to reduce the aldehydes generated by the oxidation. The second peak eluted in the salt volume of the column and lost all radioactivity when it was lyophilized. Presumably it contained [“HIformaldehyde that came from carbon 9 of [9-3H]sialic acid, derived from [6- 3H]glucosamine. The composition of the periodate-oxidized oligosaccharide (Table II) indicates that 1 residue of mannose, all of the fucose, and almost all of the sialic acid were destroyed by the periodate. The galactose and N-acetylglucosamine remained intact. This agrees with the results of periodate treatment of the intact glycoprotein.

Methylation Analysis of Re-N-acetylated Hydrazinolysis OZigosaccharide-The results, summarized in Table III, show that almost all of the galactose was substituted in position 3, whereas alI of the fucose was unsubstituted. Mannose appeared as the 2-substituted and the 3,6- and 2,4-disubstituted derivatives. N-Acetylglucosamine was substituted in position

ratio of 3:3; and Fraction C contained only N-acetylglucosa- 4 except for a small portion that was 4,6-disubstituted. Sialic


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


acid does not survive the procedures involved in methylation analysis, and no peak was observed for it.

Glycopeptide Obtained by Trypsin Digestion-Glyco- protein (1 pmol), labeled with [6-3H]glucosamine (2.0 x 10” cpm), was digested with 5 mg of trypsin in 2 ml of 0.1 M Tris- Cl, pH 7.8, for 24 h at 37°C under toluene. The digest was centrifuged and the supernatant fraction was removed and frozen. The pellet, which contained undigested glycoprotein, was redigested as above 5 times until at least 95% of the radioactivity had been solubilized. The reaction mixture was fractionated on a Bio-Gel P-10 column (1.6 x 120 cm) in 0.1 M ammonium acetate, pH 7, containing 0.1% toluene. The single radioactive peak obtained contained 1.9 pmol of glycopeptide and had a sugar composition similar to that of the glycoprotein (Table IV).

Pronase Digestion-Tryptic glycopeptide (1.6 pmol) was digested with 5 mg of pronase in 1 ml of 0.1 M NH4HC03, pH 7.8, at 37°C for 24 h under toluene, and the products were separated on a Bio-Gel P-6 column (1.6 X 120 cm) in 0.1 M

ammonium acetate, pH 7 (Fig. 6, open circles). The compositions of the recovered peaks (1.5 pmol) (Table IV) were similar to those of the tryptic glycopeptide. The major component,

TABLE IV Carbohydrate compositions of glycopeptides and oligosaccharides

formed during sequential enzymic digestions Refer to Fig. 7 for structures of substances identified by capital

italic letters. Molar ratios were normalized to 1, 2, 4, or 5 residues of N-acetvlelucosamine. as indicated.

Substance

N-Acetyl- W?“ra- G&C-

N-Acetyl- Man- gluco- FU- minic tose n*se samine CO.%! acid

Glycoprotein Tryptic glycopep-

tide Pronase-treated

tryptic glycopeptide (A)

Periodate-oxidized

Smith-degraded (0

c+Neuraminidase- digested pro-

glyco- LTide (B)

Periodate-oxidized

Smith-degraded (J)

P-o-Galactosid- ase-digested (0

8-nHexosamini- dase-digested (D)

Endo-oligosaccharide (E)

a-D-hlannosi- dase-digested (F)

P-o-Mannosi- dase-digested

Endo-glycopeptide-1 (G)

a-L-Fucosidase- digested

Endo-glycopeptide-11 (G)

a-L-Fucosidase- digested

2.9 3.1 3.0 2.7 2.9 3.0

2.5 3.3 3.4

0.2 3.0 2.1

2.1 1.9

3.0 3.3 3.2 3.0

0.1 2.0

0.2 2.0

3.2

2.9

2.9

0.9

a Analyzed as alditol acetates.

5.0 5.0

5.0

5.0

4.0

5.0 5.0

5.0

4.0

5.0

2.0

1.0

1.0

1.0

1.0

1.0

1.0

1.0

0.9 0.6

0.7

0.8 0.7”

0.5

0.5

0.2

0.8

0.3

0.2 %

0.1 9 8

0

Fraction

FIG. 6. Gel filtration on a Bio-Gel P-6 column of the products formed during sequential enzymic digestion of the tryptic “H-glycopeptide as outlined in Fig. 7. Peaks A and B (0- - -0) show the pattern given by the pronase-digested tryptic “H-glycopeptide. Fol- lowing neurarninidase digestion, a single ‘H-glycopeptide (Peak C) is observed (A-A), along with free [3H]sialic acid (Peak E). Peak C is converted to 3H-glycopeptide Peak D by the action of /3-o-galactosidase (0~~ .. .Cl), with the release of galactose (Peak F, [1--o), which was determined by the change in A426 using the Galactostat assay.

Fraction A (tubes 41 to 56), contained 3 residues of sialic acid/glycopeptide, and Fraction B (tubes 57 to 68) contained only 2 residues of sialic acid/glycopeptide, indicating a slight heterogeneity in the sugar compositions of the glycopeptide. These were recombined to yield 1.45 pmol for the following studies in which the glycopeptide was degraded sequentially according to the scheme in Fig. 7.

Smith Degradation of Pronase-digested Glycopeptide (Fig. 7h)-A portion (200 nmol) of the recombined pronase-digested glycopeptide fraction was oxidized with 0.1 M periodic acid in 0.5 M sodium acetate buffer, pH 4.5, for 24 h at 23’C in the dark. The sample was treated with 1 ml of 1 M NaBH4 for 5 h at 23’C and then acidified with 0.2 ml of glacial acetic acid. The periodate-treated glycopeptide was then chromatographed on a Bio-Gel P-6 column (1.6 X 120 cm). Two radioactive peaks were observed. The fist contained the glycopeptide fraction (184 nmol), and its composition (Table IV) shows that the fucose, most of the sialic acid, and 1 residue of mannose were destroyed by periodate. The second peak appeared in the included volume of the column, and the radioactivity w’as lost by lyophilization. This radioactivity probably was in methanol formed by reduction of the [3H]formaldehyde produced by periodate oxidation of the [9-3H]sialic acid.

The oxidized and reduced glycopeptide fraction (101 nmol) was subjected to mild acid hydrolysis (1 M HCl at 23°C for 24 h) (Smith degradation), and the products were fractionated on a Bio-Gel P-6 column (1.6 X 120 cm). A glycopeptide fraction (100 nmol) was recovered that contained no fucose or sialic acid, 2 residues of mannose, about 2 residues of galactose, and 4 residues of N-acetylglucosamine (Table IV). The methylation analysis in Table V demonstrates that the 2-substituted mannose was destroyed and that the 3,6-disubstituted mannose was no longer present, but that a 6-substituted


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


aNeuNA&8Gal-c48G1cNAc aFuc

aNeuNAc+38Gal-LBGlcNAc+2~~n~s~~~~48G,c~Ac~4~~cNAc~sn h

aNeuNAc+38Ga1+48G1cNAc+2aMan \ 6GaL48G1cNAc

1

a (I) 8Ga1+4f3G1cNAc+2atk68Man+48GlcNAc+4GlcNAc+Asn

gGal+48GlcNAc aFuc +

(B) 8Gal+48G1cNAc+20ksgt4;n-48G1cNAc+4~~cNAc+Asn 8GaL4GlcNAc

i

8Ga1+48GlcNAc+2aMan

i

\ 8GlcNAc

b (J) ~GlcNAc+2tkn+68Marw48GlcNAc+4G1cNAc+Asn

8GlcNAc aFuc +

J4 c (C) 8GlcNAc+2aMan+68Man+4~G1cNAc+4G!cNAc+Asn

GlcNAc

t3 BGlcNAc+2aMan

1

C aFuc

(D) ~Man+6f$h4gG1cNAc+4~~cNAc+Asn

aMan

1 d

aFuc t6

(E) aMan+6BMatw4G1cNAc + GlcNAc-dsn t3

(G)

aMan e 9

5607

(F) 8Marh4G1cNAc GlcNAc+Asn (H)

GlcNAc

FIG. 7. Sequential degradation of VSV glycopeptide. Structures are identified with capital letters, whereas reaction steps are identified with lower case letters. The intact glycopeptide (A) was treated sequentially, with reisolation of the product at each step, with a-D- neuraminidaae (a), /3-n-galactosidase ( b), exo-N-acetyl-p-n-hexosa-

mannose had replaced it, a consequence of the removal of the mannose unit by periodate oxidation. A second radioactive peak was eluted in the disaccharide region of the column and contained equal proportions (100 nmol) of galactose and N- acetylglucosamine, presumably the disaccharide /?Gal + 4GlcNAc.

Neuraminidase Digestion of Pronase-treated Glycopep- tide-The pronase-treated glycopeptide (1.25 pmol) was digested with 1 unit of ru-D-neuraminidase in 1 ml of 0.1 M sodium acetate, pH 4.5, at 37°C for 24 h under toluene (Fig. 7a), and the products were separated on a Bio-Gel P-6 column (1.6 X 120 cm) (Fig. 6, A). Peak C (1.2 pmol) had a composition similar to that of the starting glycopeptide minus sialic acid (Table IV), whereas Peak E was sialic acid (2.9 pmol). The siahc acid released by the neuraminidase was converted to the per-0-trimethylsilyl derivative (58) and analyzed by gas-liquid chromatography and mass spectrometry. It had the retention time of authentic N-acetylneuraminic acid, but it was eluted before a sample of N-glycolylneuraminic acid under conditions that separate N-acetyl-, N-glycolyl-, N-glycolyl-0-acyl-, and N-acetyl-0-acylneuraminic acid. The mass spectrum was similar to that of the authentic N-acetylneuraminic acid, but it differed from that of the N-glycolylneuraminic acid. The most prominent features of the spectrum were large peaks at m/e = 726 (M-15), and m/e = 622 (M-15-104), which arise from the loss of CH2 and trimethylsilyl-0-CHa.

minidase (c), endo-PI-acetyl#-D-glucosaminidase (d), a-n-mannosidase (e), P-D-mannosidase (f), and a-L-fucosidase (g). Glycopeptides A and B were subjeced to Smith degradation (sequential periodate oxidation, reduction and partial acid hydrolysis, steps h and i).

Smith Degradation of Neuraminidase-digested Pronase- treated Glycopeptide (Fig. 7i)-The neuraminidase-digested glycopeptide (200 nmol) was oxidized with periodate and reduced with NaBI&, and the products were separated on a Bio-Gel P-6 column as described for the pronase-digested glycopeptide. The oxidized and reduced glycopeptide fraction was eluted in a single radioactive peak and contained only a trace of galactose, 2 residues of mannose, and 5 residues of N- acetylglucosamine (Table IV). Mild acid hydrolysis performed on 120 nmol, as described for the pronase-digested glycopeptide, yielded two radioactive peaks that were separated by gel filtration on a Bio-Gel P-2 column (1.6 X 120 cm). The first contained the Smith-degraded glycopeptide fraction (110 nmol), and had a composition of 2 residues of mannose and 4 of N-acetylglucosamine, The methylation analysis is given in Table V. The second peak contained only N-acetylglucosamine (105 nmol).

Digestion with P-Galactosidase-The neuraminidase-digested glycopeptide (800 nmol) was incubated with 14 units of ,8-D-galactosidase in 1 ml of 50 mM sodium citrate, pH 3.5, for 24 h at 37°C under toluene (Fig. 7b), and the products were separated on a Bio-Gel P-6 column (Fig. 6, Cl). The composition of the glycopeptide fraction (750 nmol) (Table IV) is that expected for selective removal of all galactose, confirming that all of this hexose was in a terminal location after removal of the sialic acid.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE V Linkages from methylation analysis of glycopeptides and

oligosaccharides Oligo-

Smith- saccha- de- ride

Hydra- Smith- Neura- graded pi-o-

Sugar and link +- de- mini- ~~“~El- duced

ages lysls oli- graded dase-di- pronase gested mini- by en-

gosac- doglu- charide glyco- glyco- dase-di-

gested COSB- peptide peptide glyco- mini-

peptide dase diges-

Fuc + +

Marl-, + t’

+‘MLul+ + t’

+ ‘Man 4 + t”

+ ‘GlcNAc -+ +

lb + ‘GlcNAc + +

+

+ +

+

+

+

+

+

+ +

+ +

+ + + +

+

Digestion with P-Hexosaminidase-The glycopeptide recovered from the fi-galactosidase digestion (720 nmol) was treated with 3 units of j?-n-hexosaminidase in 1 ml of 50 mM sodium citrate, pH 5.0, at 37°C for 24 h under toluene (Fig. 7c), and the products were separated on a Bio-Gel P-2 column (Fig. 8, 0). The first peak (A) contained 700 nmol of a glycopeptide containing 3 residues of mannose, 2 of N-acetylglucosamine, and 0.5 residue of fucose (Table IV). The second peak (D) contained only N-acetylglucosamine (2.01 pmol), and it possessed 58% of the “H present in the starting glycopeptide, which shows that 3 of the 5 N-acetylglucosamine residues were accessible to attack by the enzyme and must be in terminal positions (Fig. 7C).

Digestion with Endo-N-acetyl$- D-glucosaminidase-The glycopeptide recovered from the ,&hexosaminidase digest (300 nmol) was incubated with 1 unit of endo-N-acetyl-b-n-glucosaminidase in 0.5 ml of 100 IIIM sodium citrate, pH 6.0, for 24 h at 37°C under toluene, which cleaves the di-N-acetylchitobiose linkage (Fig. 7d), and the products were desalted on a Bio-Gel P-2 column. The neutral oligosaccharide released was removed from the newly formed glycopeptide fraction on a Dowex 50 column. This oligosaccharide (300 nmol) gave a single radioactive peak when chromatographed on a Bio-Gel P-2 column (1.6 X 120 cm) in buffer (Fig. 8, A) and it contained 3 mannose residues to 1 of N-acetylglucosamine (Table IV).

Authentic reference samples of N-acetylglucosamine (GlcNAc), a disaccharide ManGlcNAc, a trisaccharide Man?GlcNAc, a tetrasaccharide MamGlcNAc, and a pentasaccharide MamGlcNAc were chromatographed separately on the Bio-Gel P-2 column used for chromatography of the neutral oligosaccharide peak in Fig. 8. The structures of the oligosaccharides and their elution coefficients (&) are presented in Table VI. A plot of the elution coefficient vs the log10 of the molecular weight for these samples gave a linear

curve. The oligosaccharide (Fig. 7E) from the VSV glycopeptide had an elution coefficient identical with that of MansGlcNAc (Table VI).

The glycopeptide fraction was eluted from the Dowex 50 colunn with cold 2 N HCl, lyophilized, and chromatographed on a Bio-Gel P-2 column (1.6 x 120 cm) in buffer. Two glycopeptide peaks, GP-I (102 nmol) and GP-II (117 nmol), were obtained that had similar carbohydrate compositions (Table IV). Each contained a fraction of a fucose residue and I of N-acetylglucosamine.

Fraction FIG. 8. Gel filtration on a Bio-Gel P-2 column of the ‘H-glycopep-

tide products formed during sequential glycosidase digestion of the pronase-, neuraminidase-, and galactosidase-digested tryptic ‘H-glycopeptide (Peak D from Fig. 6). Exo-N-acetyl-P-n-hexosaminidase yields ‘H-glycopeptide (Peak A, 0. ‘0) with the release of 60% of the label as N-[3H]acetylglucosamine (Peak D). Glycopeptide A is cleaved by endo-N-acetyl-P-n-glucosaminidase to release the neutral 3H-oligosaccharide (Peak B, A-A) containing half the label of Glycopeptide A, that is degraded by a-n-mannosidase to disaccharide (Peak C, M), which in turn is converted to free N- [“H]acetylglucosamine (Peak E, O--D) by the action of p-n-mannosidase.

TABLE VI Determination of core oligosaccharide size by gel filtration

Filtration was performed on a Bio-Gel P-2 column as described under “Experimental Procedures.”

Ratio of mannose to N-ace-

tylglucosa- Kd’ mineh

Compound”

GlcNAc Man + 4GlcNAc Man + 4GlcNAc T” Man Man + ’ Man + “GlcNAc

t” Man

Man 4 ’ Man + 4GlcNAc t” Man t2 Man

Core oligosaccharide (E in Fig. 7)

o-Mannosidaskdigested E (F in Fig. 7)

L3-Mannosidsse-digested F

0 0.73 1.07 0.62 1.83 0.52

2.79

3.97 0.42

2.94 0.48

0.90 0.62

0 0.75

0.48

(L Sources are given under “Experimental Procedures.” b Determined as the trimethylsilyl ethers. ’ Kd = V, - VO/V,, where V, is the elution volume, V, is the column

void volume, and V, is the column internal volume. A plot of log K,, against molecular weights gave a straight line.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Vesicular Stomatitis Virus Glycoprotein Structure 5609

Digestion with a- o-Mannosidase-The neutral oligosaccharide MamGlcNAc from the endoglucosaminidase digestion (90 nmol) was digested with 6 units of a-n-mannosidase in 0.5 ml of 0.1 M sodium citrate, pH 4.5, for 24 h at 37°C under toluene (Fig. 7e), and the products were separated on a Bio- Gel P-2 column (Fig. 8, 0). A single radioactive peak was obtained that contained 1 residue of mannose/N-acetylglucosamine (87 nmol) (Table IV) and had a Kd identical to that of ManGlcNAc (Table VI).

Digestion with p- D-Mannosidase-The disaccharide, recovered from the a-mannosidase digestion (68 nmol), was digested with 5.3 units of /?-n-mannosidase in 0.5 ml of 50 XDM

sodium citrate, pH 3.5 (Fig. 7f), and the products were separated by chromatography on a Bio-Gel P-2 column (Fig. 8, Cl). The radioactive peak contained only N-acetylglucosamine (66 nmol) (Table IV).

Digestion with /?- L-Fucosidase-The glycopeptides, GP-I and GP-II, were digested separately with one unit of bovine kidney a-L-fucosidase in 0.5 ml of 50 mM sodium citrate, pH 5.0, at 5O’C for 24 h under toluene (Fig. 7g). N-Acetylglucos- amine was the only sugar observed in acid hydrolysates of the recovered glycopeptide fractions. The amino acid compositions of GP-I and GP-II (Table VII) show that each contained 1 mol of aspartic acid/m01 of glucosamine and confii the earlier conclusion that amino acid differences account for their separation from each other.

CONCLUSIONS

From the studies presented under “Results,” a unique structure can be derived for the oligosaccharide chains of the glycoprotein of vesicular stomatitis virus. The two chains are identical and the common structure is presented in Fig. 7A.

Structure of the Outer Chain Trisaccharide Units-Taken together, the results indicate that the oligosaccharide contains three outer chains with the common structure aNeuNAc + ‘fiGal+ 4PGlcNAc +. The sialic acid residues are displayed as terminal N-acetyl-o-n-neuraminic acid because this sugar was observed when the monosaccharide released by digestion with purified a-n-neuraminidase was analyzed. The quanti- tative release of sialic acid by neuraminidase suggests that all of this sugar occupies a terminal position in the oligosaccharide. Periodate oxidation of the 3H-oligosaccharide and 3H- glycopeptide, obtained from glycoprotein labeled with [6- 3H]glucosamine, released all of the radioactivity in the sialic acid as a volatile derivative, probably formaldehyde. Since the label would be in position 9 of the sialic acid, this indicates that none of the sialic acid was substituted at position 8.

According to the methylation data, 85% of the galactose

TABLE VII

Amino acid compositions of limit glycopeptides from VSV glycoprotein

Residues per residue of glucosamine. Glycopeptides were hyclro- lyzed at 110°C for 12 h in 2 N HCl under N1. Analyses were done by P. Hombeck, of this department, on a Beckman model 120C Amino acid analyzer.

Component GP-I GP-II

Aspartic acid 1.03 1.21 Threonine 0.52 0.51 Serine 0.70 0.79 Glutamic acid 0.40 0.60 Glycine 0.36 0.51 Alanine 0.70 Isoleucine 0.18 Methionine 0.52 Histidine 0.65 Arginine 0.20 Glucosamine 1.0 1.0

was substituted in position 3 and the remainder was terminal. After neuraminidase digestion, all of the galactose became terminal as evidenced by the methylation data and by the fact that all of the galactose was removed by a purified B-D- galactosidase. Periodate oxidation of the hydrazinolysis oligosaccharide and Smith degradation of the pronase-treated glycopeptides confiied these conclusions. Galactose residues that are substituted in positions 3 or 4 are resistant to periodate, but terminal or 6-substituted galactose residues are oxidized. Since galactose survived periodate treatment origi- nally, but was destroyed after removal of sialic acid, the terminal linkage must be a-2 + 3, rather than a-2 + 6.

When the hydrazinolysis oligosaccharide was subjected to nitrous acid deamination and reduction, a disaccharide was obtained that yielded mostly galactose and 2,5-anhydromannitol on hydrolysis. Since nitrous acid produces fragments with 2,5-anhydromannose at the reducing end from the positions occupied by glucosamine residues in the oligosaccharide, we conclude that galactose was bound to N-acetylglucosamine in the glycoprotein. This is confiied by the sequential exoglycosidase digestions and methylation analyses. Before the removal of galactose, most of the N-acetylglucosamine was substituted at position 4. Afterwards, three-fifths of the N- acetylglucosamine became terminal since it could be removed by digestion with the purified exo-/3-n-hexosaminidase. In agreement, the methylation data show that most of the N- acetylglucosamine was substituted at position 4, and the resistance of this sugar to periodate oxidation is consistent with this structure.

Structure of Core Fragment-The terminal trisaccharide units containing N-acetyl-n-neuraminic acid, D-gdaCtOSe, and N-acetyl-n-glucosamine are linked to a core that contains D-

mannose, N-acetyl-n-glucosamine, and L-fucose and has the structure:

The partial structure (Fig. 7E) was deduced for the neutral oligosaccharide obtained by digesting the core glycopeptide with the endo-N-acetyl-P-n-glucosaminidase. From the known enzyme specificity, it is apparent that this product resulted from cleavage of the glycosidic bond in a di-N-acetylchitobiose unit that was substituted in position 4 with a P-linked mannose unit that, in turn, was substituted at position 3 with an unsubstituted mannose residue (Fig. 70). In agreement, methylation analysis of this fragment indicated that the core contained 2 terminal mannose residues, a 3,6-disubstituted mannose unit, and a 4-substituted N-acetylglucosamine.

It was established by several criteria that the core contains 3 mannose residues. These include the composition of the tryptic glycopeptide and the hydrazinolysis oligosaccharide, the size determined by gel filtration, and the composition of the core fragment obtained by nitrous acid deamination of the hydrazinolysis product. Moreover, when the oligosaccharide, released by the action of the endoglucosaminidase, was digested with purified a-D-mannosidase, 2 residues of mannose were released leaving a disaccharide of mannose and N-acetylglucosamine. This disaccharide was hydrolyzed by a purified fl-n-mannosidase.

The glycopeptides, separated after the endoglucosaminidase digestion, differed in amino acid content, but they had the general structure indicated in Fig. 7G. The fucose was released by digestion with a purified a-L-fucopyranosidase, which establishes the ring form and linkage of this sugar. The


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


hydrazinolysis oligosaccharide yielded a disaccharide of fucose and 2,5-anhydromannitol on deamination, indicating that it contained a fucosyl + N-acetylglucosamine unit in which the fucose was either terminal or substituted with another N- acetylgluco samine residue. The methylation data establish that the fucose was unsubstituted and that the N-acetylglucosamine residue to which it was attached was substituted in positions 4 and 6. From the known specificity of the endoglucosaminidase, the substitutent at position 4 must have been N-acetylglucosamine, so the L-fucose must be located at position 6.

Attachment of Outer Chain Trisaccharide Units to Core-We have concluded that the terminal trisaccharide units are linked to the core in the following manner:

in which R = aNeuNAc -+ ‘pGal+ 4flGlcNAc. Partial acid hydrolysis of the re-N-acetylated hydrazinolysis oligosaccharide released a fragment that contained 3 mannose residues and 5 N-acetylglucosamine residues. Nitrous acid deamination of the hydrazinolysis oligosaccharide produced a tetrasaccharide with 3 mannose units and a single 2,5-anhydromannose unit (Fig. 3). Two-thirds of the mannose residues in Fragment C of Fig. 7 became susceptible to hydrolysis by (Y- o-mannosidase after exo-P-n-hexosaminidase digestion, indicating that they became exposed in a terminal position by this treatment. This was verified by methylation analysis. The tryptic glycopeptide and the hydrazinolysis oligosaccharide contained 2-substituted, 2,4-disubstituted, and 3,6-disubstituted mannose residues in approximately equal proportions. After exo-P-n-hexosaminidase digestion, only terminal and 3,6-disubstituted mannose units were found, indicating that the N-acetylglucosamine residues were linked to position 2 of one mannose and to positions 2 and 4 of another.

Two arrangements are possible for the attachment of the outer chains to the core depending on which mannose is disubstituted. To distinguish between them, Smith degrada- tions were performed on both the intact and the neuraminidase-digested glycopeptides. Since the mannose residue that is substituted only at position 2 can be oxidized by periodate and the attached fragment can then be released by mild acid hydrolysis after reduction, it is possible to determine which of the mannose residues was substituted only in position 2. With the intact glycopeptide (Fig. 7A), the released fragment was a disaccharide that contained galactose and N-acetylglucosamine in equal proportions. Methylation analysis of the glycopeptide remaining after Smith degradation (Fig. 7n revealed the presence of a 2,4-disubstituted and a 6substituted mannose residue. When the neuraminidase-digested glycopeptide was Smith-degraded, the only small fragment released contained N-acetylglucosamine, whereas a large fragment with the structure shown in Fig. 7 J was produced. Therefore, the mannose that is linked to position 6 of the 3,6-disubstituted mannose residue is substituted at both positions 2 and 4 by N-acetylgluco samine, and the mannose linked to position 3 of the 3,gdisubstituted mannose has a single N-acetylglucosamine attached at position 2 (Fig. 7B).

DISCUSSION

The spinner culture of BHKPl cells we used allowed isolation of IOO-mg quantities of virus from which IO-mg amounts of glycoprotein were obtained. Following radioiodination of the intact purified virus, under conditions such that only the G

protein of the virion is accessible to the reagent (51), gel electrophoresis of the total protein extract revealed only labeled glycoprotein. This demonstrates that the sample was free of cellular membranes. To avoid contamination of the glycoprotein or other fractions by extraneous sugars, procedures were avoided that employed sucrose density gradient separations or carbohydrate-containing gels, and the isolated glycoprotein was extensively extracted to remove viral glyco- lipids. The glycoprotein was free of contaminating proteins that could be detected by protein stains of gels or by analyzing the radioactivity of glycoprotein gels from virions labeled with “C-amino-acids.

The only sugars detected in an acid hydrolysate of the purified glycoprotein were fucose, mannose, galactose, N-acetylglucosamine, and sialic acid. In contrast, McSharry and Wagner (6) reported over 20% glucose in the glycoprotein of VSVindiana grown in L-cells, whereas Etchison and Holland (7) grew VSVindisna in BHKPl cells and found 3 to 5 residues of glucose/glycoprotein molecule. Because these authors employed sucrose density gradients, Sephadex gel chromatography, and cellulose dialysis tubing in their work, contamination from these sugar-containing materials is a possibility. We have avoided these sources of contamination and have employed samples of such a size that minor contaminants did not interfere. We also failed to detect the presence of N-acetylgalactosamine, in agreement with the finding of Burge and Huang (59) and McSharry and Wagner (6), although Etchison and Holland (8) reported 1 to 2 residues of this amino sugar per glycoprotein molecule. We have no explanation for this discrepancy, although cellular membrane sphingolipids are an obvious source of this amino sugar.

A third difference between the published compositions and our results concerns the fucose content. McSharry and Wag- ner (6) reported that fucose represented less than 0.1% of the carbohydrate in the purified glycoprotein from virus grown in L cells. Our finding of 1.7 residues of fucose/glycoprotein molecule parallels that of Etchison and Holland (7), who reported 1.2 residues. These differences appear to depend on the cell line from which the virus is isolated since Etchison and Holland (8) reported 1.83 residues of fucose/glycoprotein molecule for virus grown in MDBK cells, 1.21 residues for virus from BHKPl cells, 0.8 residues for virus from HeLa cells, and only 0.12 residue for virus from L29-cells. The fractional value we found for fucose suggests that there is heterogeneity in the substitution by this sugar, but an alternative explanation exists; because the fucosyl derivatives used in quantitation are the most volatile of those in the mixture, this sugar may be lost preferentially, thus giving low values.

Discrepancies in some other reported sugar ratios are more difficult to interpret. McSharry and Wagner (6) found almost 6 times as much glucosamine and neuraminic acid as mannose and galactose. The composition we found, however, is in good agreement with that reported by Etchison and Holland (7) for galactose and N-acetylglucosamine, and in fair agreement for mannose and sialic acid. The slightly lower sialic acid content reported by Etchison and Holland may reflect a difference between the BHK monolayer cell line and the spinner line in the degree of sialylation of glycoproteins.

Etchison and Holland (7) reported about 8 residues of mannose/molecule of glycoprotein, which differs significantly from the value of 6 reported here. This could result from analytical discrepancies, from differences in method of virus production, or it could reflect differences in virus serotype. The evidence for precisely 3 mannose residues/oligosaccharide, or G/glycoprotein molecule, found in this work will be discussed later, but recent studies indicate that there are 3 mannose units in the core of most “complex” nitrogen-linked


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


oligosaccharides in serum-type glycoproteins (60). The compositions of the hydrazinolysis oligosaccharide and

the tryptic glycopeptide from VSV glycoprotein are consistent only with there being two oligosaccharide chains/glycoprotein molecule, and this is in agreement with previous authors (7, 17, 18). The data suggest that that the two oligosaccharide chains have identical sugar compositions, and the same conclusion has been reached by Etchison et al. (19).

The oligosaccharide from VSVn,, ~~~~~~ grown in the BHK spinner cell line was surprisingly homogeneous, showing only a small fraction of the chains with incomplete sialylation. In contrast, other authors have obtained glycopeptides missing 1, 2, or sometimes all 3 sialic acid residues (17, 19, 61). The glycopeptide appeared nearly homogeneous at each step in the sequential removal of the sialic acid, galactose, and N- acetylglucosamine. However, after digestion with the endo-N- acetyl-P-n-glucosaminidase, some heterogeneity was observed, probably due to amino acid differences because the two glycopeptides showed different properties even after removal of the fucose residue.

cates that the terminal sialic acid of VSV glycoprotein is required for binding of the virus to cells and erythrocytes, the binding site must involve more than just the sialic acid residues because binding is not inhibited solely by sialic acid or sialyllactose. In experiments not described here, we have found that the re-N-acetylated hydrazinolysis oligosaccharide from fetuin, at 5 pg/ml, inhibits plaque formation by VSV on monolayers of BHKzl cells. Purified complex sialoglycopep- tide isolated from IgM does not inhibit, even at 50 pg/ml. Since fetuin and VSV appear to have the same terminal trisaccharide structure, but both differ from that of the immunoglobulins, the trisaccharide unit aNeuNAc + “PGal + 4/3GlcNAc may play a role in the binding. Alternatively, the 3 sialic acid residues in close proximity found in fetuin and the VSV glycoprotein, but not in IgM, may be important.

REFERENCES 1.

The oligosaccharide released by endo-N-acetyl-fl-u-glucosaminidase digestion from the neuraminidase-, galactosidase- and exohexosaminidase-digested tryptic glycopeptide had the properties of MannGlcNAc. Digestion of the MamGlcNAc with a-mannosidase yielded a single disaccharide, PMan + 4GlcNAc. In contrast, Moyer and Summers (62) and Moyer et al. (18) reported heterogeneity in the core region of the VSV glycopeptide based on an apparent resistance to digestion with endo-N-acetylglucosaminidase. This may have been due to incomplete removal of the terminal sugars which would inhibit the digestion.

2. 3.

4. 5. 6. 7. a.

9.

10.

All of our studies indicate the presence of 3 mannose residues/oligosaccharide chain, rather than 4. Thus, the core oligosaccharide migrated on chromatography with authentic Man,GlcNAc, and it was clearly separated from MamGlcNAc and MamGlcNAc. In addition, its composition was similar to authentic MamGlcNAc, but not to the ManzGlcNAc or the MamGlcNAc references. Finally, the methylation data were similar to those of the MamGlcNAc standard, which shows the presence of 2 terminal mannose residues and one 3,6- disubstituted mannose attached to position 4 of an N-acetylglucosamine residue. Our results demonstrate that the VSV glycopeptide core is identical to that of glycopeptides from fetuin, human al-acid glycoprotein, human lactotransferrin and human serotransferrin (60). This same structure has also been reported for IgE, IgM, IgG, and some GM,-gangliosidosis oligosaccharides.

11. 12.

Wagner, R. R. (1974) in Comprehensive Virology (Fraenkel-Con- rat, H., and Wagner, R. R., eds) Vol. 4, pp. I-80, Plenum Press, New York

Mudd, J. A., and Summers, D. F. (1970) Virology 42,958-968 Wagner, R. R., Kiley, M. P., Snyder, R. M., and Schnaitman, C.

A. (1972) J. Viral. 9, 672-683 McSharry, J. J., and Wagner, R. R. (1971) J. Viral. 7, 59-70 Klenk, H:-D., and Chop&, P. W. (1971) J. Virol. 7.416-417 McSharrv. J. J.. and Waener. R. R. (1971) J. Viral. 7. 412-415 Etchison:J., anh Hollani, J. j. (1974) Virology 60, 217-229 Etchison, J., and Holland, J. J. (1974) Proc. Natl. Acad. Sci. U.

S. A. 71,4011-4014 Gottschalk, A. (1966) in The Amino Sugars (Balazs, E. A., and

Jeanloz, R. W., eds) Vol. 2B, pp. 337-359, Academic Press, New York

Balazs, E. A., and Jacobson, B. (1966) in The Amino Sugars (Balazs, E. A., and Jeanloz, R. W., eds) Vol. 2B, pp. 361-395, Academic Press, New York

Howe, C., and Lee, L. T. (1972) Adv. Virus Res. 17, I-50 Cartwright, B., Smale, C. J., and Brown, F. (1969) J. Gen. Virol.

5, I-10 13.

14. 15. 16.

Bishop, D. H. L., Repix, P., Obijeski, J. F., Moore, N. F., and Wagner, R. R. (1975) J. Virol. 16, 75-64

Schloemer, R. H., and Wagner, R. R. (1974) J. Virol. 14.270-281 Schloemer, R. H., and Wagner, R. R. (1975) J. Viral. 15.882-893 Schloemer, R. H., and Wagner, R. R. (1975) J. Virol. 15,

1029-1032 17. Robertson, J. S., Etchison, J. R., and Summers, D. F. (1976) J.

Viral. 19,871-878 18. Moyer, S. A., Tsang, J. M., Atkinson, P. H., and Summers, D. F.

(1976) J. Virol. 18, 167-175 19. Etchison, J. R., Robertson, J. S., and Summers, D. F. (1977)

Virology 78,375-392

The outer chains of the VSV oligosaccharide are similar or identical to those in fetuin, transferrin and al-acid glycoprotein (66). They differ from those of several immunoglobulins (63) in that the sialic acid is linked 2 + 3 rather than 2 + 6 to galactose, and that the galactose is linked only to position 4 of N-acetylglucosamine. The fucose in the VSV glycoprotein core is attached in the same position as that reported for IgG (64), IgE (65) and IgA (66), IgM (67), mouse histocompatibil- ity-2 alloantigen (68), and glycopeptide Sl of glycoprotein E2 of Sindbis virus (69).

20. 21. 22.

23. 24. 25.

26. 27.

Dulbecco, R., and Vogt, M. (1954) J. Exp. Med. 99, 167-182 Cuatrecasas, P. (1973) Biochemistry 12, 1312-1323 Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193,265-275 Wang, C., and Smith, R. (1975) Anal. Biochem. 63,414-417 Spies, J. R. (1957) Methods Enzymol. 3,467-477 Ghosh, S., Blumenthal, H. J., Davidson, E., and Roseman, S.

(1960) J. Biol. Chem. 235, 1265-1273 Spiro, R. G. (1966) Methods Enzymol. 8, 11-13 Dubois, M., Gilles, K. A., Ham&on, J. K., Rebers, P. A., and

Smith. F. (1956) Anal. Chem. 28. 350-356

The oligosaccharide of VSV glycoprotein is similar to that of many known glycoproteins. A similar structure with two outer chains has been reported for a Sindbis virus glycopeptide (69). The same attachment of the outer two chains to position 2 of the a-linked mannose residues has been reported for IgG, IgM, IgE, and IgA (66) and for human transferrin glycopeptide (70). In serum glycoproteins with four outer chains, these are probably linked to positions 2 and 4 of both of the a-linked mannose residues (63).

28. Dische, Z. (1962) Methods Carbohidr. Chem. 1, 512-514 29. Bray, G. A. (1960) Anal. Biochem. 1, 279-285 30. Laemmli, U. K. (1970) Nature 227,680-685 31. Fairbanks, G., Steck, T. L., and Mallach, 0. F. L. (1971) Biochem-

istry 10,2606-2617 32. Maizel, J. V., Jr. (1971) Methods Virol. 5, 179-246 33. Clamp, J. R., Bhatti, T., and Chambers, R. E. (1971) Methods

Biochem. Anal. 19,229-344 34. Stellner, K., Saito, H., and Hakomori, S. (1973) Arch. Biochem.

Biophys. 155, 464-472 35. Hakomori, S. (1964) J. Biochem. (Tokyo) 55,205-208 36. Nakajima, T., and Ballou, C. E. (1974) J. Biol. Chem. 249,

7685-7694 Although the work of Schloemer and Wagner (14-16) indi- 37. Norden, N. E., Lundblad, A., Svensson, S., Ackerman, P.-A., and


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


Autio, S. (1973) J. Biol. Chem. 242,6210-6215 38. Rosenfeld, L., and Lee, Y. C. (1976) Curbohydr. Res. 46, 155-158 39. Tarentino, A. L., and Maley, F. (1974) J. Biol. Chem. 249,811-817 40. Cassidy, J. T., Jourdian, G. W., and Roseman, S. (1965) J. Biol.

Chem. 240,3501-3506 41. Li, S.-C., and Li, Y.-T, (1970) J. Biol. Chem. 245,5153-5160 42. Li, Y.-T. (1967) J. Biol. Chem. 242,5474-5480 43. Koide, N., and Muramatsu, T. (1974) J. Biol. Chem. 249,

4897-4904 4.4. Bahl, 0. P., and Agrawal, K. M. L. (1969) J. Biol. Chem. 244,

2970-2978

46. Wied&schai& G. Y., and Rosenfeld, E. L. (1971) Biochem. Bio- phys. Res. Commun. 44,1008-1014

47. Alhadeff. J. A., Miller, A. L., Wenaas, H., Vedvick, T., and

45. Levw. G. A.. and McAllan, A. (1961) Biochem. J 80.435439

54. Yosizawa, A., Sato, T., and Schmid, K. (1966) Biochim. Biophys. Acta 121, 157-162

55. Schroeder, W. (1972) Methods Enzymol. 26, 138-143 56. Horton, D., and Phillips, K. (1972) Carbohydr. Res. 21,417-419 57. Bayard, B., and Foumet, B. (1976) Curbohydr. Res. 46, 75-86 58. Buscher, H., Casals-Stenzel, J., and Schauer, R. (1974) Eur. J.

B&hem. 60.71-82

217-237 64. Komfeld, R., Keller, J., Baenziger, J., and Komfeld, S. (1971) J.

Biol. Chem. 246,3259-3268

59. Burge, B. W., and Huang, A. S. (1970) J. Viral. 6, 176-182 66. Bavard, B., and Roux, D. (1975) FEBS Lett. 55.206-207 61. Sefton,.B. (1976) J. Viral. 17,85-93 62. Moyer, S. A., and Summers, D. F. (1974) Cell 2,63-70 63. Kornfeld, R., and Komfeld, S. (1976) Annu. Rev. Biochem. 45.

C’S&, J. S.‘(1975) J. Biol. Chem. 250, 7106-7113 48. Bhnnbere. S.. Hildesheim. J.. Yariv. J., and Wilson. K. J. (1972)

B&h&. Biophys. Acta 264, 171-176 49. March, S. C., Parikh, I., and Cuatrecasas, P. (1974) Anal. Bio-

&em. 60.149-152 50. McSharry, J. J., and Benzinger, R. (1970) Virology 40, 745-749 51. Moore, N. F., Kelley, J. M., and Wagner, R. R. (1974) Virology

61.292-296 52. Kellev. J.. Emerson, S., and Wagner, R. R. (1972) J. Viral. 10.

123111235 53. Gottschalk, A (1966) in Glycoproteins: Their Composition, Struc-

ture and Function (Gottschalk, A., ed) pp. 96-111, Elsevier, New York

65. Baensiger, J., Komfeld, S., and Kochwa, S. (1974) J. Biol. Chem. 249, 1897-1903

66. Baenziger, J., and Komfeld, S. (1974) J. Biol. Chem. 249, 7270-7281

67. Hickman, S., Komfeld, R., Osterland, C. K., and Komfeld, S. (1972) J. Biol. Chem. 247,2156-2163

68. Nathenson, S. G., and Muramatau, T. (1971) in Glycoproteins of Blood Cells and Plasma (Jamieson, G. A., and Greenwalt, T. J., eds) pp. 245-262, J. B. Lippincott, Philadelphia

69. Burke, D. J. (1976) Doctoral dissertation, State University of New York, Stony Brook

70. Spik, G., Bayard, G., Fournet, B., Strecker, G., Bouquelet, S., and Montreuil, J. (1975) FEBS Lett. 50, 296-299


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

C L Reading, E E Penhoet and C E BallouCarbohydrate structure of vesicular stomatitis virus glycoprotein.

1978, 253:5600-5612.J. Biol. Chem.

http://www.jbc.org/content/253/16/5600.citation

Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/253/16/5600.citation.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/253/16/5600.citation

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;253/16/5600&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/253/16/5600.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=253/16/5600&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/253/16/5600.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/253/16/5600.citation.full.html#ref-list-1

http://www.jbc.org/

Documents

Carbohydrate Structure of Vesicular Stomatitis Virus Glycoprotein* · 2002-12-11 · Carbohydrate Structure of Vesicular Stomatitis Virus Glycoprotein* (Received for publication,