Synthesis of glycoproteins in cells infected by the flavivirus Kunjin

The synthesis of glycoproteins in Vero cells infected by Kunjin virus was studied by labeling infected cells with radioactive mannose labeled glycoproteins by gel electrophoresis. In infected cells glycoproteins with approximate molecular weights of 53,000 (gp53), 59,000 (gp59), and 66,000 (gp66) were observed. Limited proteolytic digestion of gp53, gp59, and gp66 demonstrated that gp53 and gp59 contain similar amino acid sequences, and are probably related to gp66. Antiserum raised against the envelope protein V3 of purified virions selectively precipitated gp53, gp59, and the ³H-amino acid-labeled labeled protein of 51,000 daltons (p51), from extracts of infected cells; p51 is known to be almost identical with V3 by tryptic peptide mapping. Thus p51, gp53, gp59, and probably gp66 are intracellular forms of the envelope protein V3 of Kunjin virus. When infected cells were labeled with [³H]mannose for short time periods, gp59 was labeled within 5 min, gp66 within 10 min and gp53 within 20 min. After pulse labeling of infected cells with [³H]mannose followed by incubation in medium containing excess glucose, label was lost from gp66 and accumulated in gp53, demonstrating a possible precursor-product relationship between gp66 and gp53. The size of the glycopeptides derived from gp53, gp59, and gp66 by ehaustive digestion with Pronase were identical.     

Glycosylation sites of influenza viral glycoproteins Tryptic glycopeptides from the A/WSN (H0N1) hemagglutinin glycoprotein

The glycosylation sites of the hemagglutinin glycoprotein of influenza A/WSN virus were investigated by analysis of tryptic glycopeptides separated by ion exchange chromatography and gel filtration. Five major classes of glycosylated tryptic peptides were obtained from the HA₁ subunit, and a single major glycosylated peptide was obtained from the HA₂ subunit. Amino acid sequence analyses indicated that three of the tryptic glycopeptides were each obtained from a distinct glycosylation site on the HA glycoprotein molecule. The oligosaccharides linked to each glycopeptide class were characterized by radiolabeling with various sugar precursors, gel filtration analysis of Pronase digests, and determination of their susceptibility to cleavage by endoglycosidase H. The results indicate that complex oligosaccharides are present in three of the tryptic glycopeptides obtained from HA₁ as well as the glycopeptide obtained from HA₂. One of the tryptic glycopeptides of HA₁ contains a typical high-mannose oligosaccharide, whereas another HA, glycopeptide contains oligosaccharides that have characteristics intermediate between those of typical complex and high-mannose chains. The various tryptic glycopeptides also exhibit differences in extent of sulfation, and nonsulfated and nonsulfated forms of the same glycopeptides appear to be separable by ion exchange chromatography. Affinity chromatography on immobilized plant lectins indicated that each tryptic glycopeptide contained oligosaccharides that were heterogeneous in their exact structures.     

Immunoprecipitable polypeptides specified by varicella-zoster virus

Polypeptides encoded by varicella-zoster virus (VZV) in infected cell cultures have been identified by radioimmune precipitation techniques. Detergent-solubilized extracts of VZV-infected cells were reacted with highly specific VZV antisera raised in strain-2 guinea pigs immunized with sonicates of syngeneic virus-infected cells. Fractionation of the immunoprecipitates in acrylamide slab gels demonstrated an average of 16 polypeptides, which ranged in molecular weight from 32,000 to å200,000. These included the three major immunogenic glycoproteins (gp 62, gp 98, and gp 118) and a prominent higher molecular weight nonglycosylated polypeptide at 155,000. One of the [³⁵S]methionine-labeled polypeptides comigrated with purified actin. Not all polypeptides were visible in any one particular fluorogram, but comparative analysis of polypeptide profiles derived from electrophoreses performed with different gel concentrations and different crosslinkers (methylene-bisacrylamide and N,N-diallyltartardiamide) clearly established a consistent and reproducible pattern of radioactive bands. A low background of radio-activity was nonspecifically precipitated by the antigen-antibody-protein A complexes; however, with the exception of a common band comigrating with actin, the electrophoretic profiles representing virus-specific and nonspecific immunoprecipitates were easily distinguished.     

Immunological studies of the functions of paramyxovirus glycoproteins

The HN and F glycoproteins of the paramyxovirus SV5 were purified and monospecific antibodies to each prepared. The effects of bivalent (IgG) and monovalent (Fab) forms of these antibodies on the biological activities of the glycoproteins were determined. Anti-HN antibodies inhibited adsorption of virus to erythrocytes, hemagglutination, and hemolysis. Inhibition of hemolysis was shown to be secondary to the inhibition of adsorption; anti-HN antibodies added after virus adsorption did not affect hemolysis. Anti-HN antibodies inhibited neuraminidase activity, and this inhibition was independent of substrate size in experiments in which virus and antibodies were allowed to react and substrates of different size added, i.e., neuraminlactose and fetuin. This result is in contrast to previous findings with influenza virus in which antibodies inhibited the action of the viral enzyme on the macromolecular substrate only. Thus with SV5, the antibodies appear to inhibit the hydrolytic process directly, rather than sterically hindering the access of the enzyme to large substrates, as in the case in influenza virus. Anti-F antibodies had no effect on virus adsorption or neuraminidase activity, but inhibited hemolysis when added either before or after absorption, confirming the direct involvement of the F protein in the hemolytic process. The inhibition of virus adsorption, neuraminidase, and hemolytic activities by the respective antibodies was accomplished by Fab fragments as well as IgG, indicating that in each case the inhibitory activity was a consequence of a direct effect on individual glycoprotein molecules rather than crosslinking of glycoproteins or aggregation of virions. Both anti-HN and anti-F antibodies neutralized virus infectivity in MDBK and CV-1 cells. Anti-F IgG and Fab, and anti-HN IgG caused essentially complete neutralization in both cell types; however, anti-HN Fab caused only partial neutralization in CV-1 cells.     

Viral membrane glycoproteins: comparison of the amino terminal amino acid sequences of the precursor and mature glycoproteins of three serotypes of vesicular stomatitis virus

The NH2-terminal amino acid sequences of the envelope glycoproteins and the in vitro synthesized, nonglycosylated precursors of the glycoproteins of three serotypes, namely Indiana (Toronto), Cocal, and New Jersey (Concan) of vesicular stomatitis virus were determined. A comparison of the sequences showed little homology in the signal peptides present in the nonglycosylated precursors except for their high hydrophobic amino acid content. In contrast, the NH2-terminal amino acid sequences of the mature envelope glycoproteins revealed extensive homology suggesting that this region is conserved and may be involved in essential biological function(s) of the rhabdovirus.     

Tryptic and chymotryptic methionine peptide analysis of the in Vitro translation products specified by the transforming region of adenovirus type 2

Early region 1 (El) cytoplasmic RNAs of adenovirus type 2 were translated in vitro. Structural relationships between these proteins were then established by tryptic and chymotryptic digestion and two-dimensional peptide mapping. This analysis also allowed a direct comparison of these proteins with polypeptides isolated from early infected cell extracts and previously assigned to E1 by indirect means (M., Green, W. S. M. Wold, K. H. Brackmann, and M. A. Cartas, 1979, Virology 97, 275–286). E1a (0–4.5%) RNA encodes five proteins in vitro: 35K, 41K, 47K and 53K at early times, and a 28K protein at later times in infection. These five proteins are all highly related by peptide mapping analysis, and are also related to a group of four proteins isolated from cell extracts prepared early in infection. E1b (4.5–11.0%) early RNA encodes 52K and one or more 15K proteins in vitro. The 52K protein shares tryptic peptides with a 53K immunoprecipitated T antigen. A 15K protein shares some peptides with the 52K protein and with several proteins isolated from infected cells. At intermediate to late times, E1b RNA encodes a 12K protein that corresponds to the virion structural polypeptide IX. These data also permit the correlation of E1 proteins synthesized in vitro with the polypeptides predicted by E1 sequence analysis.     

Structure and glycosylation of tacaribe viral glycoproteins

The virion glycoprotein (G) of Tacaribe virus was separated into multiple components by two-dimensional electrophoresis. The most acidic glycoprotein species, which migrated most rapidly toward the anode, were found to differ from less acidic species primarily in their sialic acid content. A previously undetected basic component yielded a peptide map dissimilar to the map of the acidic component; however, after neuraminidase digestion, peptide maps of the basic and acidic component were very similar. We have detected at least two glycopeptide size classes by analysis of Pronase digests of Tacaribe G protein as well as a precursor designated gp70. The larger size class (～ 3300 daltons) was labeled with glucosamine, galactose, or mannose and appeared to contain neuraminic acid as indicated by sensitivity to neuraminidase. The smaller size class (1600–1900 daltons) contained a high amount of mannose in addition to glucosamine and was sensitive to endo-β-N-acetylglucosaminidase H. The G glycoprotein was composed mainly of large glycopeptides, whereas the gp70 glycoprotein contained mostly the smaller mannose-rich glycopeptides.     

Processing of virus-specific glycoproteins of varicella zoster virus

Monoclonal antibodies to varicella zoster virus (VZV) glycoproteins were used to study the processing of three glycoproteins with molecular weights of 83K-94K (gp 2), 64K (gp 3), and 55K (gp 5). Immunoprecipitation experiments performed with VZV-infected cells, pulse labeled with [3H]glucosamine in the presence of tunicamycin, suggest that O-linked oligosaccharide is present on the glycoprotein of gp 2. Use of the enzyme endo-beta-N-acetylglucosaminidase H revealed that the fully processed form of gp 3 had high-mannose type and that of gp 5 had only complex type of N-linked oligosaccharides. Experiments with monensin suggest that the precursor form (116K) of gp 3 is cleaved during the processing from Golgi apparatus to cell surface membrane. The extension of O-linked oligosaccharide chain and the complex type of N-linked oligosaccharide chains also occurs during this processing.     

Purification and characterization of a virus from the aphid Rhopalosiphum padi

A 27-nm icosahedral virus was purified from the oat bird cherry aphid, Rhopalosiphum padi (L.). The virus had an s(20,w) of 162 +/- 2 S, and bouyant densities of 1.37 in CsCl and 1.35 in Cs2SO4. It contained one ssRNA of 31 +/- 2 S and three major proteins. The relationship of the R. padi virus to other small RNA invertebrate viruses is unclear.     

Protein structure of lymphocytic choriomeningitis virus: evidence for a cell-associated precursor of the virion glycopeptides.

Immunoprecipitates of cells infected with lymphocytic choriomeningitis virus (LCMV) contain a 74,000- to 75,000-dalton glycopeptide (GP-C) which is not found in the virion. The relationship between GP-C and the virus structural glycopeptides GP-1 and GP-2 was studied by two-dimensional peptide mapping and pulse-chase labeling. GP-C contained the peptides of both GP-1 and GP-2 indicating that it is a proteolytic cleavage precursor of the structural glycopeptides. Radiolabeled glucosamine or methionine incorporated into GP-C during a 1-hr pulse period was demonstrated in GP-1 and GP-2 after a chase interval of 6 hr. GP-1 and GP-2 are apparently the only major products of the cleavage of GP-C, as mixtures of the digests of GP-1 and GP-2 essentially reproduce the map of GP-C. Carbohydrate label incorporation into the viral glycopeptides was studied, and GP-C was found to be relatively rich in mannose and glucosamine but poor in fucose and galactose. In constrast, GP-1 contained glucosamine, fucose, and galactose, while GP-2 contained glucosamine and fucose. By surface iodination and immunoprecipitation, only GP-1 was found on the surfaces of infected cells.     

Isolation and structural analysis of influenza C virion glycoproteins

Influenza C virions possess a single glycoprotein that is cleaved into two disulfide-linked subunits, gp65 and gp30. When analyzed under nonreducing conditions, the uncleaved (gpI) and cleaved (gpI) glycoproteins differ significantly in apparent molecular weight; however, we observed no difference in their tryptic peptide patterns. We have isolated the glycoproteins by selective solubilization with Triton X-100 or octylglucoside; only preparations obtained with the latter detergent showed hemagglutinating activity. In purified glycoprotein samples, gp65 was routinely observed as a doublet on SDS-polyacrylamide gels. Analysis of tryptic peptides by ion-exchange chromatography demonstrated that the two gp65 bands have indistinguishable polypeptide backbones; they appear to differ, however, in carbohydrate content. The uncleaved glycoprotein as well as gp65 were resistant to Edman degradation indicating the presence of blocked amino termini, whereas gp30 was observed to have the N-terminal tripeptide sequence NH₂-Ile-Phe-Gly. This sequence is homologous to a sequence at the N termini of influenza A and B HA₂ glycoproteins, except for the presence of an additional terminal glycine residue in these viruses. The influenza C glycoproteins form a regular hexagonal lattice on the viral envelope. This arrangement is sometimes maintained in disrupted virus preparations and in glycoprotein subunits released from the envelope by limited proteolysis, indicating that direct interactions between the glycoprotein molecules are responsible, at least in part, for the observed arrangement. Observations of clustered surface projections on plasma membranes of infected cells, and of released virus particles apparently devoid of internal nucleoproteins, are consistent with the suggestion that lateral interactions between the influenza C glycoproteins may be important in virus assembly.     

Assembly of membrane glycoproteins studied by phenotypic mixing between mutants of vesicular stomatitis virus and retroviruses

Temperature sensitive (ts) mutations of vesicular stomatitis virus (VSV), Indiana serotype, which belong to complementation group V (tsV) have been shown to affect the viral envelope glycoprotein, or G protein. When ts V mutants are grown in cells producing avian leukosis viruses, the titers of infectious VSV obtained at the nonpermissive temperature are 10⁴-fold higher than in control cells. Cells releasing murine leukemia viruses or avian reticuloendotheliosis virus rescue VSV ts V mutants much less efficiently. The rescued virions have the properties of envelope pseudotypes in that their host range is restricted to that of the helper retrovirus, they are neutralized by anti-retrovirus antibodies but not anti-VSV antibodies, and they are not thermolabile. Sensitive serological techniques, including the use of complement-mediated virolysis, immunoprecipitation, and monoclonal antibody reacting with G protein, show that VSV pseudotypes produced at the nonpermissive temperature have no detectable G protein, whereas VSV particles released from retrovirus infected cells at the permissive temperature have mosaic envelopes bearing both VSV G protein and retrovirus glycoprotein. In mixed infections of Rous sarcoma virus (RSV) and VSV ts V mutants, pseudotype particles with RSV genomes and VSV envelope antigens are produced only at the permissive temperature. In contrast, substantial yields of RSV(VSV) pseudotypes but no VSV(RSV) pesudotypes are obtained at the nonpermissive temperature with VSV carrying mutations in complementation group III, which affect M protein. Thermolabile VSV tsV mutants form RSV(VSV) pseudotypes which also are thermolabile. The kinetics of heat inactivation of G protein function in tsV mutants is the same in VSV particles with unmixed envelopes and with mosaic envelopes. From these studies of phenotypic mixing we draw the following conclusions: (i) The synthesis of functional M protein but not G protein is essential for the maturation of VSV virions. (ii) VSV M protein is not required for the assembly of G protein into retrovirus virions. (iii) The thermolabile nature of tsV VSV mutants is an intrinsic property of the G protein, independent of the type of virion into which it is incorporated and of other viral glycoproteins which may be assembled into the envelope of the same virion.     

Fusion between cell membrane and liposomes containing the glycoproteins of influenza virus

Glycoproteins of influenza virus strains were incorporated into liposomes by a dialysis procedure, using octylglucoside as detergent. Liposomes containing either the cleaved or uncleaved hemagglutinin of the virus were tested for fusion activity with cellular membranes. Electron microscopic examination as well as microinjection studies revealed that liposomes containing the cleaved hemagglutinin could fuse with cell membranes. In contrast, liposomes containing the uncleaved hemagglutinin were merely adsorbed to the cell surface and fusion occurred only after treatment with trypsin. Native virus particles with the cleaved hemagglutinin could be shown to fuse with liposomes containing cellular receptors of influenza virus. From these results and the known correlation existing between cleavage of hemagglutinin and infectivity of influenza virus, it is suggested that fusion may be an important step in penetration of the nucleocapsid of influenza virus into host cells.     

Isolation and characterization of low-molecular-weight DNA-binding proteins from retroviruses

The DNA-binding proteins from Rauscher leukemia virus, baboon leukemia virus, endogenous cat virus, woolly monkey virus, bovine leukemia virus, equine infectious anemia virus, Rous sarcoma virus, Prague strain, mouse mammary tumor virus, and Mason-Pfizer monkey virus were purified to homogeneity by molecular sieving chromatography and DNA affinity chromatography. The amino acid composition shows each to be a basic protein.     

Biogenesis of poxviruses: Preliminary characterization of conditional lethal mutants of vaccinia virus defective in DNA synthesis

Characterization of six is mutants of the DNA^? phenotype of vaccinia 1HD-W is described. Complementation analysis revealed that four of these map into different complementation groups while one was able to complement, albeit inefficiently, with only one among the other ts mutants tested. Judging by the rate of viral DNA synthesis it became evident that all DNA^? mutants except ts 6389 became more “leaky” for DNA synthesis at the restrictive temperature in single-cycle infection if adsorption was carried out at 4° when compared with values obtained after initiation of infection at 39.5°. This suggested that these mutants are defective in some function required very early in the virus life cycle. This notion was further supported by the observation that multiple rounds of viral DNA synthesis could occur at 39.5° provided that the temperature was raised after initiating the infection at a lower temperature. Employing density-shift experiments coupled with analysis of the size of replicative intermediates by means of alkaline sucrose gradients, it was found that progeny DNA synthesized by all but ts 6389 was indistinguishable from wild-type DNA. By contrast, DNA synthesis in ts 6389-infected cytoplasm was is throughout the period of viral DNA replication, regardless of the time when temperature was shifted up, relegating this mutant to the “fast-stop” category. This idea is supported by evidence that all of the DNA^? isolates except ts 6389 are deficient at 39.5° in some of the virus-specified DNA-binding proteins. Therefore, ts 6389 is the only isolate at hand with the characteristics expected of a mutant with a defective protein related to the growing fork of the vaccinia DNA replication apparatus.     

Carbohydrate-induced conformational changes of Semliki forest virus glycoproteins determine antigenicity

Cells infected with Semliki forest virus contain the glycoproteins E₁ and p62. Both are glycosylated and undergo a maturation process in which the antigenicity is changed from that of the nonglycosylated counterpart to that of E₁ and E₂ occurring in the virion. Antigenic conversion of E₁ and p62 in infected cells includes modification of the carbohydrate chains. The conversion is interpreted as a change in conformation of the glycoproteins causing exposure of different antigenic determinants, which are not identical with the sugar moiety. The results favor the idea that carbohydrate chains are responsible for establishment and maintenance of specific conformations, thus determining indirectly the antigenicity of E₁ and p62.     

Evidence for the presence of repeating antigenic determinants in the major polypeptides derived from hepatitis B surface antigen

The immunogenicity of the four major and seven minor hepatitis B surface antigen (HBsAg)-derived polypeptides was evaluated in guinea pigs and compared to that of 22-nm HBsAg particles. Both humoral and cell-mediated immune responses were determined. All antisera reacted with homologous HBsAg/ayw and P22/ayw. However, the glycosylated polypeptides (P25, P27, P31, and P52) generally elicited the most intense immune responses. At least nine polypeptides induced antibodies which also reacted with heterologous HBsAg/adw particles and P22/adw. Whole HBsAg/ayw particles induced transformation in lymphocytes obtained from guinea pigs immunized with the major or minor polypeptides. These data tend to demonstrate that all HBsAg-derived polypeptides contain in their structure the group-specific and at least one of the subtype-specific antigenic determinants, and these determinants are also exposed on the surface of the HBsAg particle. It seems possible that a limited nucleotide sequence confers the same virus specificities to most if not all of the constitutive proteins of HBsAg.     

Association of the envelope glycoproteins of influenza virus with liposomes--a model study on viral envelope assembly.

Envelope glycoproteins of virus N (an avian influenza virus) were artificially associated with liposomes, using octylglucoside as detergent which could be readily removed by dialysis. The prepared liposomes, about 0.1-1 μm in diameter, contained specific hemagglutinin and neuraminidase activities. The total amounts of proteins incorporated into liposomes were about the same irrespective of the lipid composition. Similarly, the neuraminidase activity of liposomes remained unchanged by altering lipids. However, the hemagglutinating and cell-adsorbing properties of liposomes varied greatly depending on the composition of lipids. As both lipid and protein components can be altered, the present system may be elaborated for the study of protein-lipid interactions and functions of viral membranes.     

Cell lines derived from tumors induced in syngeneic rats by FR 3T3 SV40 transformants no longer synthesize the early viral proteins

Several SV40-transformed FR 3T3 rat cell lines formed tumors upon inoculation to syngeneic immunocompetent Fisher rats. These tumors, which appeared only after a long latency period, showed a fast rate of growth. Tumor-derived (TD) cell lines were established in culture from several tumors induced by independent transformants, and their properties were analyzed. Though TD cells were highly tumorigenic, their level of transformation in culture was similar to that of the original transformants. They did not synthesize detectable amounts of the two early viral gene products, the large-T and small-T polypeptides. However, the transformation-associated cellular p53 protein was detected in all of them by [35S]methionine labeling and immune precipitation with monoclonal antibodies directed against the mouse p53. Growth in the animal apparently counterselected the cells expressing the early viral proteins, and hence, possibly, the tumor-specific transplantation antigen. This selection was mediated at least in part by the T-cell immune response, as the tumors induced by the same transformants in nu/nu mice still expressed the nuclear T-antigen. Absence of expression of the early viral region was frequently correlated with the loss of the integrated SV40 DNA. Some tumors, however, still contained early viral DNA sequences, which were, in even fewer cases, transcribed into RNA. These results altogether suggest that tumor formation by the FR 3T3-SV40 transformed cells in immunocompetent rats requires two events, the selection for the acquisition of a high tumorigenic potential, and against the expression of the early viral genes. Only the first of these two events was observed upon tumor formation in nude mice.     

Isolation and characterization of a papovavirus from cynomolgus macaque kidney cells

A papovavirus was isolated from uninoculated kidney cell cultures of the Cynomolgus Macaque (Macaca fascicularis), in which it produced extensive cytopathologic changes after prolonged periods of culturing. On basis of virion morphology, genome size, and restriction endonuclease data, this virus was identified as a new isolate of the Stump-Tailed Macaque Virus (STMV) and subsequently named CK-strain (Cynomolgus kidney) of STMV. Without inducing c.p.e. the virus replicated also in BSC-1, VERO, human embryonic, and calf kidney cells. The relationship between STMV and SV40 was investigated by cross-blot hybridization between DNA fragments of both viruses. The region containing the SV40 regulating sequences showed strong homology with STMV DNA. No antigenic relationship between STMV and SV40 could be demonstrated. Antibodies to STMV were not found in sera from 20 Cynomolgus Macaques, but were detected in 25 out of 57 cattle sera and 6 out of 26 bovine colostrum samples. The serologic findings indicate that STMV is not a latent simian virus, but a virus of bovine origin.