Search

We have previously reported a functional role associated with hepatitis C virus (HCV) E1 glycoprotein using vesicular stomatitis virus (VSV)/HCV pseudotype. In this study, we have investigated the role of glycosylation upon intracellular transport of chimeric E1-G, and in infectivity of the pseudotyped virus. Interestingly, surface expressed E1-G exhibited sensitivity to Endoglycosidase H (Endo H) treatment, which was similar to full-length E1, suggesting that additional complex oligosaccharides were not added while E1-G was in transit from the endoplasmic reticulum (ER) to the mammalian cell surface. As a next step, each of the four potential N-linked glycosylation sites located at amino acid position 196, 209, 234, or 305 of the E1 ectodomain were mutated separately (asparagine --> glutamine), or in some combination. FACS analysis suggested that mutation(s) of the glycosylation sites affect the translocation of E1-G to the cell surface to different extents, with no single site being particularly essential. VSV pseudotype virus generated from glycosylation mutants exhibited a decrease in titer with an increasing number of mutations at the glycosylation sites on chimeric E1-G. In a separate experiment, N-glycosidase F treatment of pseudotype generated from the already synthesized E1-G or its mutants decreased virus titer by approximately 35%, and the neutralization activity of patient sera was not significantly altered with N-glycosidase F-treated pseudotype virus. Taken together, our results suggested that E1-G does not add complex sugar moieties during transport to the cell surface and retain the glycosylation profile of its parental E1 sequence. Additionally, the removal of glycans from the E1-G reduced, but does not completely impair, virus infectivity.

Cyanovirin-N (CV-N) is under development as a topical (vaginal or rectal) microbicide to prevent sexual transmission of human immunodeficiency virus (HIV), and an economically feasible means for very large-scale production of the protein is an urgent priority. We observed that N-glycosylation of CV-N in yeast eliminated the anti-HIV activity, and that dimeric forms and aggregates of CV-N occurred under certain conditions, potentially complicating the efficient, large-scale manufacture of pure monomeric CV-N. We therefore expressed and tested CV-N homologs in which the glycosylation-susceptible Asn residue at position 30 was replaced with Ala, Gln, or Val, and/or the Pro at position 51 was replaced by Gly to eliminate potential conformational heterogeneity. All homologs exhibited anti-HIV activity comparable to wild-type CV-N, and the Pro51Gly homologs were significantly more stable proteins. These glycosylation-resistant, functional cyanovirins should be amenable to large-scale production either in bacteria or in eukaryotic hosts.

PNGase F is a widely used deglycosidase, secreted in small amounts by the gram-negative bacterium Flavobacterium meningosepticum. We have designed a T7 promoter-based Escherichia coli expression system to provide a high-yield source of recombinant enzyme. When expressed intracellularly, the enzyme was produced in a largely insoluble state. However, when expressed as a fusion with the leader sequence from the ompA gene, hexahistidine-tagged PNGase F was efficiently processed and exported to the E. coli periplasm. Single-step purification using immobilized metal affinity chromatography yielded 8 mg of pure enzyme per liter of culture, which is fully active on a range of protein and peptide substrates., (Copyright 2002 Elsevier Science (USA).)

The cell--cell adhesion molecule 1 (C-CAM1) plays an important role as a tumor suppressor for prostate cancer. Decreased expression of C-CAM1 was detected in prostate, breast, and colon carcinoma. Reexpression of C-CAM1 in prostate and breast cancer cell lines was able to suppress tumorigenicity in vivo. These observations suggest that C-CAM1 may be used as a marker for cancer detection or diagnosis. To generate monoclonal antibodies specific to C-CAM1, we have overexpressed full-length human C-CAM1 in Sf9 cells using a baculovirus expression system. The protein was purified 104-fold using nickel affinity chromatography. About 0.4 mg purified C-CAM1 was obtained from 200 mg of infected cells. When the purified protein was digested with peptidyl-N-glycosidase, the apparent mobility of the protein on SDS--PAGE changed from 90 to 58 kDa, which is close to the molecular weight predicted from the cloned cDNA sequence. This observation suggests that C-CAM1 was glycosylated on asparagine residues when expressed in Sf9 cells. Western blotting and internal protein sequencing analysis confirmed that the purified protein is human C-CAM1. Biochemical and functional assays indicate that this protein expressed in Sf9 cells displays characteristics similar to those of native protein, including adhesion function and glycosylation modification. Using this protocol, sufficient quantity of this protein can be produced with purity suitable for monoclonal antibody generation and biochemical study., (Copyright 2001 Academic Press.)

SPC1 (furin/PACE), an enzyme belonging to the S8 group of serine endoproteases, is a type I integral membrane protein that catalyzes the processing of a multitude of precursor proteins. We report here the use of transfected Drosophila melanogaster Schneider 2 cells to produce milligram amounts of two forms of recombinant human SPC1. In order to investigate the role of the cysteine-rich region (CRR) of SPC1, we compared the biochemical and enzymatic properties of hSPC1/714 that has the C-terminal tail and transmembrane region of the native enzyme removed with that of hSPC1/585 which had, in addition, the CRR deleted. Two stable cell lines were established. The S2-hSPC1/714 line secreted a major form of apparent molecular weight of 83 kDa and a minor form of 80 kDa whereas the S2-hSPC1/585 line secreted a single 59-kDa protein. PNGase F treatment of the different forms demonstrated that the enzymes were glycosylated. Automated NH(2)-terminal sequencing revealed that all purified forms resulted from processing at the expected zymogen activation site. Removal of the CRR resulted in a broadening of the enzyme's pH range, a shift of K(0.5) for Ca(2+), and a shorter enzymatic half-life when compared to the longer form, which suggest that the CRR of hSPC1 may help in stabilizing the enzyme's proteolytic activity. The use of this high-level expression system will meet the demand for material necessary to perform biochemical and structural studies that are needed to further our understanding of this and other SPCs at the molecular level., (Copyright 2000 Academic Press.)

Complete carbohydrate composition analysis of glycoprotein hormones, their subunits, and oligosaccharides isolated from individual glycosylation sites can be accomplished using high-pH anion-exchange chromatography combined with pulsed amperometric detection. Neutral and amino sugars are analyzed from the same hydrolyzate by isocratic chromatography on a Dionex CarboPAC PA1 column in 16 mM NaOH. Sialic acid is quantified following mild hydrolysis conditions on the same column in 150 mM sodium acetate in 150 mM NaOH. Ion chromatography on a Dionex AS4A column in 1.8 mM Na(2)CO(3)/1.7 mM NaHCO(3); postcolumn, in-line anion micromembrane suppression; and conductivity detection can be used to quantify sulfate, a common component of pituitary glycoprotein hormone oligosaccharides. Mass spectrometric analysis before and after elimination of oligosaccharides from a single glycosylation site can provide an estimate of the average oligosaccharide mass, which facilitates interpretation of oligosaccharide composition data. Following release by peptide N-glycanase (PNGase) digestion and purification by ultrafiltration, oligosaccharides can be characterized by a high-resolution oligosaccharide mapping technique using the same equipment employed for composition analysis. Oligosaccharide mapping can be applied to the entire hormone, individual subunits, or individual glycosylation sites by varying PNGase digestion conditions or substrates. Oligosaccharide release by PNGase is readily monitored by SDS-PAGE. Site-specific deglycosylation can be confirmed by amino acid sequence analysis. For routine isolation of oligosaccharides, addition of 2-aminobenzamide at the reducing terminus facilitates detection; however, the oligosaccharide retention times are altered. Composition analysis is also affected as the 2-aminobenzamide-modified GlcNAc peak overlaps the fucose peak., (Copyright 2000 Academic Press.)

Human herpesvirus 8 (HHV-8) or Kaposi's sarcoma-associated herpesvirus (KSHV) is the only known human member of the Rhadinovirus genus of the gammaherpesvirus subfamily. Antibodies against peptides representing portions of the amino- and carboxyl-termini of HHV-8 gB were produced and used to detect gB expression in Vero cells transfected with the gB gene, in the HHV-8-harboring cell line, BCBL-1, and in purified virions. Expression of gB was detected in approximately 3% of uninduced BCBL-1 cells, while up to 30% of the cells expressed gB after 12-O-tetradecanoylphorbol-13-acetate (TPA) induction of virus replication. Indirect immunofluorescence assays and confocal microscopy showed that gB was distributed throughout the cytoplasm of BCBL-1 cells and transfected Vero cells. Immunoblot analyses of virion preparations revealed the presence of full-length as well as two smaller than full-length gB-derived species corresponding to the amino- and carboxy-terminal portions of gB, respectively. Biochemical analysis of the gB carbohydrate moieties using glycosylation inhibitors revealed that gB contained N-linked oligosaccharides of the high-mannose type, characteristic of precursor carbohydrate chains added in the endoplasmic reticulum., (Copyright 2000 Academic Press.)

Biotinidase deficiency is an autosomal recessive disorder of biotin recycling. Biotinidase cleaves the biotin from biocytin or short biotinyl-peptides to replenish the free biotin pool, or it can transfer the vitamin to specific proteins. The cDNA for human serum biotinidase has two in-frame start codons, potentially allowing for the synthesis of an enzyme with a signal peptide (SP) consisting of either 21 or 41 amino acids. In order to examine the requirements of the signal peptide region for the production and secretion of biotinidase, three different forms of the normal human serum biotinidase gene were constructed that encode either the 21-amino-acid SP (SP21-NL) or the 41-amino-acid SP (SP41-NL) or without a SP (NoSP-NL). These constructs were expressed in insect cells via a baculovirus expression system. Biotinidase from cells with SP41-NL and SP21-NL had immunoreactive and biotinyl-hydrolase-active enzyme in lysates and expression media. Cells with NoSP-NL had about 3% of the immunoreactive material and no enzyme activity in lysates and no immunoreactive protein or enzymatic activity in the expression medium. Lack of biotinidase from cells with NoSP-NL may be due to translation inefficiency or increased susceptibility of this species to protease degradation than the secreted forms. We have demonstrated that the 21-amino-acid signal peptide is sufficient to result in glycosylated, secreted biotinidase, but we cannot determine if the glycosylated biotinidase in the lysates or secreted in the medium of cells with SP41-NL use the first, second, or both ATGs in the SP region. Because this particular expression system has no mechanism for timing the movement of newly translated biotinidase protein, we cannot draw conclusions about the relative efficiency of SP41-NL versus SP21-NL, but it is possible that either is used in vivo depending on particular cellular conditions., (Copyright 2000 Academic Press.)

A prominent secretory glycoprotein was detected in the culture medium of porcine thyrocytes which was identified as clusterin by microsequencing. Treatment of thyrocytes with thyroid stimulating hormone revealed a tight regulation of both synthesis and secretion of clusterin, with a distinct fraction of clusterin being always associated with the cells. At least three N-bound glycans were found on each subunit of clusterin, receiving most of the incorporated [(32)P]phosphate-label. Binding of clusterin to the immobilized cation-independent mannose 6-phosphate (M6P) receptor indicated that part of the phosphate label was contained in M6P moieties. Immunolabeling of cultured thyrocytes and of thyrocytes in situ showed clusterin on the apical cell surfaces where it colocalized with gp330/megalin, which is known to serve as a binding site for clusterin. The association with the apical plasma membrane, which, in thyrocytes, carries the iodinating system, was confirmed by biosynthetic iodination, an as yet unknown posttranslational modification of clusterin. On the basolateral plasma membranes clusterin was found within distinct, bipartite patches, suggesting that it is a constituent of cell-adhesion complexes and that it participates in cell-cell and cell-matrix interactions., (Copyright 1999 Academic Press.)

Peptide-N4-(N-acetyl-beta-d-glucosaminyl asparagine amidase) from Aspergillus tubigensis (PNGase At) was expressed in baculovirus-infected insect cells. The recombinant PNGase At was secreted and purified to homogeneity with a yield of 9.5 mg per liter of infected cell medium. Recombinant PNGase At migrated upon SDS-PAGE as a single-chain protein with a molecular mass of 78 kDa. This contrasts with the native Aspergillus enzyme which is "nicked" and migrates as two subunits each with a molecular weight about 43 kDa. Quantitation of total sugar by phenol-sulfuric acid suggests that the enzyme expressed in baculovirus-infected insect cells was substituted with 8-10 chains of carbohydrate of which 75% was released by Endoglycosidase F1. ESI-MS analysis of the oligosaccharides released from the recombinant PNGase At revealed similarity in the number of glycosylated residues but a significant difference in their composition, when compared to the carbohydrates of the native PNGase At. Despite differences in the primary structure and in the composition of glycan residues, the recombinant enzyme had the same specific activity as the native enzyme., (Copyright 1998 Academic Press.)

Keratinocyte growth factor (KGF) is a member of the fibroblast growth factor family that acts specifically on epithelial cells in a paracrine mode. We employed a mammalian expression system to synthesize recombinant human KGF and isolated two preparations, KGF-a and KGF-b, from medium conditioned by Chinese hamster ovary cells. On an SDS-PAGE gel, KGF-a migrates as two bands near 25-29 kDa and contains both N- and O-linked sugar moieties attached near the N-terminus. Detailed structural characterization confirms that KGF-a contains a single amino acid sequence predicted from cDNA sequence and the molecule has two intramolecular disulfide bridges, Cys1-Cys15 and Cys102-Cys106. An additional Cys at position 40 is free and resides in a solvent-inaccessible environment. Mass spectrometric analyses of KGF-a peptides verify the occurrence of several post-translational modifications in the molecule, including partial oxidation at Met28, partial sulfation at Tyr27, and glycosylation at Asn14 and Thr22. The Asn-linked carbohydrate structures are heterogeneous, which include biantennary, triantennary, and tetraantennary structures with none or up to four sialic acids attached to various structures, while the Thr-linked carbohydrates contain typical mucin-type structures. KGF-b is an N-terminally truncated form of KGF-a posttranslationally processed at Arg23 and is not glycosylated. Both KGF-a and KGF-b forms are capable of stimulating DNA synthesis in quiescent Balb/MK mouse epidermal keratinocytes.

A procedure for purifying the Torpedo californica nicotinic acetylcholine receptor subunits is proposed which involves preparative SDS-PAGE followed by reverse-phase HPLC on a C4 column in an acetonitrile-isopropanol system. By this method, the alpha-subunit can be completely separated from the 43-kDa protein which migrates very close to it on SDS-PAGE, and the delta-subunit can be isolated free from the beta-subunit of Na+, K(+)-ATPase comigrating with it on SDS-PAGE. The purity of all acetylcholine receptor subunits thus obtained was verified by Edman degradation and MALDI mass-spectrometric analysis which could be performed quite easily on the HPLC-purified samples. In general, we observed a good correlation between the experimentally determined molecular masses and those calculated from the amino acid sequences and when known, posttranslational modifications (glycosylation and phosphorylation) of individual receptor subunits. Transfer of the isolated receptor subunits into 1% octyl-beta-D-glucopyranoside generates samples suitable for functional studies and enzymatic proteolysis or deglycosylation.

The human monocyte chemoattractant protein-1 (MCP-1) was expressed at high levels in Pichia pastoris with the alcohol oxidase promoter. It was secreted from the yeast when either its natural signal sequence or the Saccharomyces cerevisiae alpha-factor signal peptide was used. SDS-PAGE and Western blot revealed two immunoreactive MCP-1 species at 15 and 8.5 kDa designated MCP-1H and MCP-1L, respectively; both were purified by cation-exchange chromatography. MCP-1H could be converted to MCP-1L by treatment with peptide N-glycosidase F, showing that the former is an N-glycosylated form of the latter. Laser desorption mass spectrometry showed that MCP-1L actually consisted of a mixture of three polypeptides of 8449, 8614, and 8780 Da and MCP-1H showed a broad peak at 11,134 Da. N-terminal peptide sequencing indicated that nearly half of MCP-1L lacked the two N-terminal amino acids found in the native protein. Both MCP-1H and MCP-1L could induce monocyte migration and calcium influx in THP-1 monocytic leukemia cells, although these activities were about 10- to 100-fold lower than those of MCP-1 produced in insect cells.

A Pichia pastoris expression clone has been developed to produce the human cytokine interleukin-17 (hIL-17). Characterization of purified recombinant hIL-17 made with this clone demonstrated that it shared many characteristics with hIL-17 produced in mammalian cells. The hIL-17 produced in Pichia had the correct N-terminus of natural mature hIL-17 and a glycosylation pattern similar to hIL-17 produced in mammalian cells; both Pichia and human cells add approximately 5 kDa of sugars via N-linked glycosylation and both express a mixture of the glycosylated and nonglycosylated forms. Gel filtration provides evidence that the Pichia produced hIL-17 exists as a dimer in solution. A combination of cation-exchange and gel-filtration chromatography yielded 3.5 mg of highly purified and biologically active hIL-17 from a 10-liter fermentation. These results show that P. pastoris is a useful system to produce recombinant hIL-17 in structure/function studies of this molecule.

A protective surface antigen (200 kDa) on C. salmositica was detected using a monoclonal antibody (mAb-001). Enzymatic studies on the epitope indicated that it was sensitive to nonspecific protease K and to site-specific trypsin and protease V8 but not to alpha-chymotrypsin. The reactivity of the epitope with mAb-001 was not affected when the antigen was denatured with 8 M urea; however, reduction of the antigen with dithiothreitol destroyed the epitope. The epitope was susceptible to sodium m-periodate oxidation and N-glycosidase F, but not to O-glycosidase or neuraminidase. It was also sensitive to mild potassium hydrochloride hydrolysis and to phospholipase C, which is specific for phosphatidylinositol. These results suggest that the epitope consists of a polypeptide, a carbohydrate, and probably a phospholipid. The asparagine-bound N-glycosidically linked hybrid-type carbohydrate chain has the minimum length of a chitobiose core unit. There is probably a phosphatidylinositol residue which anchors the polypeptide to the surface membrane. The antigen is extensively posttranslationally modified.

A human alpha-1,3-fucosyltransferase (Fuc-TVII) was expressed by recombinant baculovirus-infected insect Sf9 cells as a secretory fusion protein. The fusion protein consisted of the human granulocyte colony-stimulating factor signal peptide followed by an IgG-binding domain of protein A, a Fuc-TVI-derived peptide, and the putative catalytic domain of Fuc-TVII. The signal peptide was correctly cleaved and the recombinant Fuc-TVII was secreted into the culture medium at a concentration of 10 micrograms/ml. The recombinant Fuc-TVII could be highly purified in a single-step purification procedure, i.e., IgG-Sepharose column chromatography. The enzymatic properties of the Sf9-produced Fuc-TVII were compared with the properties of that expressed by a human B-cell line, Namalwa KJM-1, transfected with an episomal plasmid carrying the fusion Fuc-TVII cDNA. Both recombinant proteins showed alpha-1,3-fucosyltransferase activity toward a type II oligosaccharide with a terminal alpha-2,3-linked sialic acid among various acceptors. The apparent Km values of Sf9-produced Fuc-TVII for GDP-fucose and its acceptor substrate were slightly lower than those of the Fuc-TVII produced by Namalwa KJM-1 cells. Sf9-produced Fuc-TVII has N-linked carbohydrate chains whose molecular weights are lower than those linked to Namalwa KJM-1-produced Fuc-TVII. This difference in carbohydrate structure hardly affects the thermal stability of Fuc-TVII. The baculovirus expression system is available for high-level expression of stable and enzymatically active secretory Fuc-TVII.

Previously we have shown that Lelystad virus (LV) contains five structural proteins, a nucleocapsid protein N, an integral membrane protein M, and three glycoproteins GP3, GP4, and GP5. In this study we identified a sixth structural protein of Lelystad virus. The protein has an apparent molecular weight of 29 to 30 kDa, was recognized by an ORF2-specific antipeptide serum in Western immunoblotting using sucrose gradient purified LV virions, and was shown to be N-glycosylated. It was therefore designated GP2. The GP2 protein was also immunoprecipitated by ORF2-specific antipeptide serum from lysates and extracellular virus of CL2621 cells infected with LV. A fraction of the GP2 protein present in these lysates contained an intrachain disulfide bond. Endoglycosidase H treatment of immunoprecipitates indicated that the endoglycosidase H-sensitive N-glycans of the GP2 protein become endoglycosidase H-resistant during passage through the Golgi compartment in cells infected with LV. In contrast, the N-glycans of the GP2 protein expressed individually in a recombinant Semliki Forest virus remained endoglycosidase H-sensitive, indicating that the GP2 protein was retained in the endoplasmic reticulum. LV is the first arterivirus which has been demonstrated to contain six virion-associated proteins, one nucleocapsid protein, one integral membrane protein, and four glycoproteins.

Human cytomegalovirus (HCMV), strain AD169, contains four genes (US27, US28, UL33, and UL78) that encode putative homologues of cellular G protein-coupled receptors (GCRs). GCRs transduce extracellular signals to alter intracellular processes, and there is evidence that HCMV may elicit such changes at early times following infection. The US27, US28, and UL33 genes are transcribed during infection, and the US28 gene product has been found to be a functional receptor for the beta-chemokine class of immune modulators. The US27, UL33, and UL78 gene products have not been described and we have concentrated on identifying the UL33 protein because it is the most highly conserved of the GCR homologues among the human beta and gamma herpesviruses. We report here cloning UL33 into a recombinant baculovirus (rBV) and expressing it in insect cells; constructing a mutant HCMV with a disrupted UL33 gene; and identifying the UL33 protein in HCMV-infected cells and virus particles. Our results demonstrate that the UL33 protein (i) is expressed as a approximately 36-kDa, heat-aggregatable protein in rBV-infected cells, (ii) is modified heterogeneously by asparagine-linked glycosylation and expressed as a > or = 58-kDa glycoprotein that is present in the region of the cytoplasmic inclusions in HCMV-infected fibroblasts, (iii) is present in virions and two other enveloped virus particles, and (iv) is not essential for growth of HCMV in human foreskin fibroblast cultures.

Affinity chromatography on immobilized ribonuclease (RNase) inhibitor was developed for purification of mammalian RNase. Human placental RNase inhibitor was conjugated to CNBr-activated Sepharose in the presence of dithiothreitol. About 80% of the immobilized RNase inhibitor was capable of binding bovine pancreatic RNase A. The bound RNase A was eluted with 3 M NaCl at pH 5.0. Two 25-kDa and 18-kDa RNases, which were obtained from human liver using a cellulose phosphate column, were bound to the immobilized RNase inhibitor and recovered in a pure and active form after treatment of the resin with p-hydroxymercuribenzoate. These enzymes were considered to be nonsecretory-type RNases with different sugar contents.

B2 is a mAb that recognizes a conformational determinant on the active form of native tryptase, but does not recognize native tryptase that spontaneously loses activity in physiologic buffer. Precursor forms of recombinant human (rh) alpha- and rh beta-tryptase have been expressed in a baculovirus system. In each case, multiple electrophoretic forms were detected in both culture media and cell lysates of infected insect cells by Western blots developed with the G3 mAb made against native human tryptase. Although only 4 of 30 amino acids in the leader sequences of alpha- and beta-tryptase differ, rh alpha-tryptase appeared predominantly in the cell lysates, rh beta-tryptase predominantly in the culture media. B2 recognized rh alpha-tryptase and rh beta-tryptase found in the culture media of infected Sf-9 cells, but not that in cell lysates. Secreted forms of tryptase were purified to homogeneity by B2-immunoaffinity chromatography. From 1 liter of culture fluid 1.5 to 3 mg of rh-tryptase could be purified. Each rh-tryptase precursor was enzymatically inactive with synthetic substrates. Analysis of the N-terminal amino acid sequences of purified rh alpha- and rh beta-tryptase precursors (APAPVQA and APAPGQA, respectively) indicated that the initial 18 amino acids of the 30-amino-acid leader sequence had been removed. Differential N-glycosylation was found in both rh alpha-tryptase (one or two carbohydrate groups per molecule) and rh beta-tryptase (zero or one carbohydrate group per molecule). Thus, the baculovirus expression system is a useful tool for generation of rh alpha- and rh beta-tryptase precursors that exhibit a conformational epitope also present on natural tryptase and that are preferentially secreted into the culture media of infected cells.

Glycoproteins are microheterogeneous with respect to their attached oligosaccharides. When these oligosaccharides contain sialic acid, the oligosaccharide microheterogeneity will impart charge heterogeneity to the glycoprotein. We found that two commercial preparations of human serum transferrin (HST), a sialylated glycoprotein, have very different chromatographic profiles when the samples are separated by pellicular anion-exchange chromatography. Each anion-exchange profile contained multiple peaks, which suggested that both glycoproteins have charge heterogeneity. High-pH anion-exchange chromatography with pulsed amperometric detection (HPAEC/PAD) analysis of sialic acids and oligosaccharides released from the two preparations of HST revealed that the two preparations differed in sialic acid and sialylated oligosaccharide content. When oligosaccharides were released from the anion-exchange fractions of the two HST preparations, the HPAEC/PAD oligosaccharide profiles showed that protein retention was directly related to sialylated oligosaccharide content (i.e., the longer a fraction was retained, the greater its sialylated oligosaccharide content). Therefore, the anion-exchange profiles of the two HST preparations are related to their sialylated oligosaccharide content. We believe that pellicular anion-exchange chromatography can be used to quickly monitor gross changes in the sialylation of sialylated glycoproteins due to physiological state, or in the case of recombinant glycoproteins, culture conditions and/or purification.

Subgroup B adenoviruses (Ad3, -7, -11, -35) contain two open reading frames (ORFs) in the early E3 transcription unit that are not present in subgroup C adenoviruses (Ad2, Ad5). The product of one of these ORFs, a 20,500-kDa (20.5K) protein, was shown previously to be expressed as two diffuse 22K and 36K bands on SDS-PAGE; the 22K appeared to be the precursor to the 36K species. As judged by its predicted sequence, 20.5K is a type I membrane glycoprotein with two potential sites for N-glycosylation and a transmembrane domain near its COOH-terminus. Here we show that when Ad3- or Ad7-infected cells were radiolabeled in the presence of tunicamycin, which prevents the addition of N-linked oligosaccharides, both the 22K and the 36K forms of 20.5K showed increased mobility in SDS-PAGE, indicating that both forms contain N-linked sugars. Both the 22K and the 36K forms were sensitive to digestion by endoglycosidase F and N-glycanase, again indicating that they both contain N-linked sugars. Only the 22K species was sensitive to endoglycosidase H, indicating that it contains high-mannose-type oligosaccharides and that the 36K species contains complex-type carbohydrates. The 36K form was sensitive to neuraminidase, indicating that its sugars contain terminal sialic acid. When digested with N-glycanase and neuraminidase, the 36K form was sensitive to O-glycanase, indicating that the 36K form has O-linked oligosaccharides. The 22K form was labeled with [3H]mannose and the 36K form was labeled with [3H]glucosamine and to a much lesser extent by [3H]mannose. Altogether these results indicate that the 20.5K protein is cotranslationally modified with N-linked high-mannose oligosaccharides, then the protein moves into the Golgi and trans-Golgi network where it acquires O-linked and complex N-linked oligosaccharides.

Osteoblast-specific factor 2 (OSF-2) is a approximately 90-kDa protein selectively expressed in bone. OSF-2 cDNA was recently isolated from mouse and human cDNA libraries and shows limited sequence homology with fasciclin I, a cell adhesion protein expressed in insect nerve cells. Here we describe the expression of recombinant murine OSF-2 (rmOSF-2) in a baculovirus/insect cell system. Western blotting analysis employing polyclonal antiserum raised against a C-terminal synthetic OSF-2 peptide detected a protein of approximately 90-kDa as early as 2 days after infection of Sf9 cells with the recombinant virus. Tunicamycin treatment of infected cells resulted in a mobility shift of OSF-2 (approximately 90-kDa band) on Western blots. N-Glycanase digestion resulted in the same mobility shift of OSF-2, indicating that rmOSF-2 expressed in insect cells is N-glycosylated. However, OSF-2 was insensitive to endoglycosidase H digestion while a major fraction of this protein had affinity for concanavalin A. Finally, it was demonstrated that rmOSF-2 was able to bind to heparin. This finding suggests that OSF-2 might be associated with the bone extracellular matrix after secretion by osteoblasts and participate in cell adhesion and/or cell communication. The establishment of the baculovirus expression system with a high productivity of recombinant OSF-2 (around 40 micrograms/ml at maximum) and its heparin binding properties should allow us to obtain large amounts of rmOSF-2.

Previous studies have found Kunjin (KUN) virus isolates from within Australia to be genetically homogenous and that the envelope protein of the type strain (MRM61C) was unglycosylated and lacked a potential glycosylation site. We investigated the extent of antigenic variation between KUN virus isolates from Australia and Sarawak using an immunoperoxidase assay and a panel of six monoclonal antibodies. The glycosylation status of the E protein of each virus was also determined by N glycosidase F (PNGase F) digestion and limited sequence analysis. The results showed that KUN viruses isolated within Australia oscillated between three antigenic types defined by two epitopes whose expression was influenced by passage history and host cell type. In contrast an isolate from Sarawak formed a stable antigenic type that was not influenced by passage history and was distinct from all Australian isolates. PNGase F digestions of KUN isolates indicated that 19 of the 33 viruses possessed a glycosylated E protein. Nucleotide sequence of the 5' third of the E gene of selected KUN isolates revealed that a single base change in PNGase F sensitive strains changed the tripeptide N-Y-F (amino acids 154-156 of the published sequence) to the potential glycosylation site N-Y-S. Further analysis revealed that passage history also had a significant influence on glycosylation.

The predicted amino acid sequence of herpes simplex virus type 1 glycoprotein C (HSV-1 gC) shows that it has the features of a typical type 1 integral membrane protein: a cleavable N-terminal signal sequence, a glycosylated ectodomain, a single transmembrane domain, and a small, charged cytoplasmic domain. In an earlier investigation of the function of the gC cytoplasmic domain, it was shown that the gC synthesized by a gC mutant, dl2, which lacked the last three residues of the transmembrane domain and the entire cytoplasmic domain, was initially synthesized as a membrane bound glycoprotein, but was not stably anchored in the plasma membrane. In this study, we generated a panel of four HSV-1 gC mutants with novel cytoplasmic domains in order to further delineate the role of this domain in stable anchoring and to investigate the role of charged residues in this process. The cytoplasmic domain mutants produced significant quantities of a novel precursor (pgC-86K), approximately 6K smaller than the wild-type gC precursor. The quantity of pgC-86K correlated with the number of positive charges in the cytoplasmic domain. Although the nature of the novel form of the precursor is unclear, its correlation with cytoplasmic domain charge suggests that an important function of this domain is to influence gC processing. Significantly, restoration of the carboxy-terminal 3 amino acids of the gC transmembrane domain restored the wild-type anchoring phenotype to gC, indicating that the cytoplasmic domain is not required for membrane anchoring. Further characterization of dl2 gC confirmed that this glycoprotein is not released from the membrane by proteolysis. We suggest that the addition of three additional hydrophobic residues to the dl2 transmembrane domain increases its hydrophobicity enough to stabilize membrane anchoring of the glycoprotein.

The amino acid sequence for the envelope protein(s) predicted from the nucleotide sequence of the E and E2/NS1 regions of the hepatitis C virus (HCV) genome is enriched with an N-linked glycosylation site motif, Asn-X-Thr/Ser, suggesting oligosaccharide moieties are present on the virion surface. We attempted to characterize the sugar moiety on the surface of HCV virions recovered from sera of infected humans to assess the natural properties of the virus. Six kinds of lectins were used to bind HCV virions in affinity column chromatographies: RCAI, WGA, Con A, AAL, LCA, and PNA. Lectin-bound virions were identified by detecting HCV RNA in eluted chromatography fractions with a polymerase chain reaction (PCR) method. Our results showed that HCV was similar to hepatitis B virus (HBV) in characteristics of binding to lectins: HCV showed a strong binding to RCAI and WGA, weak binding to Con A, and no detectable binding to AAL, LCA, or PNA. Treatment of the HCV virion preparation with an enzyme, glycopeptidase A, or a detergent, NP-40, resulted in a significant decrease in the ability to bind these lectins. Our results suggest that asparagine-linked sugar chains are present on the surface of native virions of HCV, very similar to those for HBV.

The influence of the carbohydrate groups of rhodopsin on its ability to regenerate upon incubation with 11-cis retinaldehyde after photobleaching was examined. Rhodopsin was deglycosylated enzymatically with peptide-N-glycosidase F (PNGase F). Verification of deglycosylation was established by: (a) SDS-PAGE; (b) carbohydrate compositional analysis using high performance anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD); (c) isolation and carbohydrate analysis by HPAEC-PAD and fast atom bombardment-mass spectrometry of the oligosaccharides liberated from rhodopsin; and (d) absence of reactivity with lectins. Deglycosylated rhodopsin, when present either in rod outer segments or after purification, exhibited the same absorption spectrum as the native molecule. After photobleaching, deglycosylated rhodopsin reacted with 11-cis retinaldehyde in a manner similar to the native material, restoring the spectral properties lost after light-exposure. The carbohydrate portion, therefore, was not required for expressing the spectral properties of rhodopsin nor for regeneration of the photobleached visual pigment.

Monoclonal antibodies were used to characterize vaccinia virus glycoproteins known to be incorporated into the envelope of extracellular enveloped vaccinia (EEV) virus. The 89K hemagglutinin, 42K, and 23-28K glycoproteins were predominantly expressed as late vaccinia proteins. The 89K glycoprotein was sulfated and phosphorylated but not acylated. 89K precursor proteins of 32K, 41.5K, and 52K were detected. The former had a molecular weight expected from the deduced amino acid sequence of the hemagglutinin gene. A 76K glycoprotein that did not contain methionine and was not sulfated or phosphorylated was precipitated late in infection by the anti-hemagglutinin monoclonal antibody. The appearance of this protein was inhibited by rifampicin and it may thus result from 89K cleavage. A 220K complex contained some or all of the hemagglutinin gene products linked by disulfide bonds. The 42K glycoprotein was not sulfated or phosphorylated but was acylated. This glycoprotein was disulfide bonded with the EEV 37K nonglycosylated envelope protein. The 23-28K glycoprotein was not sulfated but was both phosphorylated and acylated. The 23-28K glycoprotein group of five proteins had a common protein backbone that was differentially glycosylated. Pulse-chase, glycosylation inhibition with tunicamycin, and glycosidase experiments established that the precursor to the 23-28K glycoproteins was a 21K protein. Members of this protein family formed dimers of approximately 55K through disulfide bonds.

We have investigated the action of three endo N-acetylglucosaminidases on rhodopsin. The oligosaccharide chains of native and denatured opsin and rhodopsin, both solubilized and membrane-bound, were shown to be cleaved by endohexosaminidase H, endohexosaminidase F, and peptide-N-glycosidase F (PNGase F) as revealed by SDS-PAGE. These enzymes were shown to be free of protease activity. Under correct conditions, the endoglycosidases could release one or both carbohydrate chains. Rhodopsin and opsin at concentrations between 2 and 65 nmol ml-1 were cleaved, with more complete deglycosylation occurring at the higher concentrations.

Asparagine-linked oligosaccharides were released quantitatively by N-oligosaccharide glycopeptidase (almond) digestion from human thyroglobulins prepared from thyroid glands of normal subjects and patients with several pathological conditions. The pyridylamino derivatives of the oligosaccharides were prepared and analyzed by high-performance liquid chromatography. The content of high-mannose-type oligosaccharides was comparable to that of the complex type in normal thyroglobulins. Man9GlcNAc2 was the predominant component in the high-mannose-type region, while biantennary oligosaccharides with fucose were the major components in the complex-type region. High-mannose-type oligosaccharides were markedly decreased in thyroglobulins prepared from patients with various disorders, such as Basedow's disease, papillary carcinoma, and adenomatous goiter, whereas they were appreciably increased in thyroglobulin from diffuse goiter.