- Volume 64, Issue 12, 1983

Human teratocarcinoma cells cultured in vitro can be induced to produce retroviruslike particles. The induction procedures are the same as those previously shown to induce the synthesis of animal retroviruses. Electron microscopical evidence is presented that the human teratocarcinoma-derived (HTD) particles are most closely related to the type C retrovirus strains. HTD particles can be banded at 1.16 g/ml in linear sucrose gradients, the characteristic density for retroviruses, and subsequently be used for negative staining and fine structure analysis.

It is well established that vaccinia virus infection induces the synthesis of virus-specific DNA in cytoplasmic ‘factories’, which are sites of virus-specific transcription. The present study demonstrates that vaccinia virus-specific DNA is synthesized also in the nuclei of infected cells with a similar time course. Direct observation and radiolabelling confirm the integrity of isolated nuclei. Reconstitution experiments and inhibitor studies demonstrate that virus-induced DNA is synthesized de novo within nuclei and does not result from cytoplasmic contamination. Cell-specific DNA synthesis is inhibited completely after infection and nuclei of infected cells then synthesize DNA which co-sediments with virus genomic DNA in denaturing gradients. Restriction endonuclease cleavage and hybridization with a virus-specific probe indicate that this is full-length, virus genomic DNA. The biological implications of this are discussed.

The spike protein (S; surface projection) of avian infectious bronchitis virus (IBV) strain M41 comprises two glycopolypeptides, S1 (mol. wt. 90 × 103) and S2 (mol. wt. 84 × 103), in equimolar proportions. The apparent mol. wt. of S was calculated as 354 (±17) × 103 following co-sedimentation with catalase in sucrose gradients. Incubation of radiolabelled IBV with urea resulted in the removal of most S1, but none of S2, from the virus particle. A similar result was obtained using low concentrations of SDS, although some nucleocapsid, but not matrix, protein was also released. 2% SDS alone was as effective as 2% SDS plus 2% 2-mercaptoethanol for the separation of S1 and S2 prior to SDS–polyacrylamide gel electrophoresis. Dithiothreitol did not remove S from virions but did decrease the buoyant density of the virus from 1.18 g/ml to 1.16 g/ml, and changed the configuration of S. It is concluded that IBV S protein is an oligomer comprising two copies of each of S1 and S2, although the possibility that there are three copies of each glycopolypeptide cannot be discounted. S is attached to the membrane by S2, while S1 has little or no contact with the membrane and may form the major part of the bulbous end of S. Interpeptide disulphide bonds do not occur in S, and the association of S1 and S2 is weak.

The genes of a field isolate of human rotavirus were cloned into pBR322. The strategy used involved polyadenylation of denatured double-stranded (ds) genomic RNA, followed by cDNA synthesis using reverse transcriptase. Oligo(dC) tails were added to the 3′ end of the single-stranded cDNA and then separated by alkaline agarose electrophoresis. Sized cDNA of both polarities were allowed to hybridize and inserted into the PstI site of pBR322. Transformations done with sized cDNA always resulted in the selection of hybrid plasmids carrying inserts with a size representative of the original cDNA. Four individual clones were selected for preliminary characterization. One clone has an insert of 1360 base pairs (bp) and corresponds to gene 6. The second clone has an insert of 1140 bp and corresponds to one of the genes in the triplet 7-8-9. The other two genes, with inserts of 780 and 660 bp, were identified as corresponding to dsRNA segments 10 and 11.

Hybridomas were formed between mouse myeloma cells and spleen cells from mice immunized with Marek’s disease virus (MDV) or with herpesvirus of turkeys (HVT). Three monoclonal antibodies were obtained, two (M26 and M34) from MDV clones and one (H9) from an HVT clone, all of which were specific for cross-reactive membrane antigen (MA) expressed on the surface of cells infected with MDV or HVT. All three antibodies also reacted with MDV- and HVT-specific glycoproteins in the molecular weight (mol. wt.) ranges 54K to 70K (MDV-gp54/70) and 50K to 64K (HVT-gp50/64), respectively. These glycoproteins constitute the putative ‘A’ antigens which are found in the medium of cultures infected with MDV or HVT. These results suggest that the cross-reactive MA may correspond to ‘A’ antigen. Pulse-chase experiments using monoclonal antibodies revealed the presence in virus-infected cells of precursor and processed forms of MDV-gp54/70 and HVT-gp50/64 which differ in size. Moreover, by two-dimensional gel electrophoresis we found that MDV and HVT glycoproteins were separated to heterogeneous spots by electric charge as well as mol. wt. The several spots with higher mol. wt. and with more acidic isoelectric points among them were lost by treatment with neuraminidase, suggesting that the processing was, at least in part, due to the addition of sialic acid to the precursor forms. Tunicamycin blocked the surface expression of cross-reactive HVT-MA on HVT-infected cells. Phosphonoacetic acid inhibited both the appearance of HVT-MA on the cell surface and synthesis of HVT-gp50/64, indicating that the MA and secreted glycoprotein were late gene products of the HVT genome.

We performed antigenic analysis of the haemagglutinin and neuraminidase subunits of a recombinant virus (A/swine/Kanagawa/2/78) isolated from a pig in Japan in 1978, using a series of monoclonal antibodies to H1 (Hsw1) haemagglutinin and N2 neuraminidases of H2N2 and H3N2 viruses. Results obtained in haemagglutination inhibition tests with five monoclonal antibodies to the haemagglutinin of A/NJ/8/76 (H1N1) revealed that the haemagglutinin of three H1N1 and the recombinant viruses were indistinguishable from that of A/NJ/8/76. The neuraminidase of A/swine/Kanagawa/2/78 was found to be antigenically similar to A/Kumamoto/22/76 (H3N2, A/Victoria/3/75-like strain). The oligonucleotide maps of the entire RNAs of H1N1, H1N2 and H3N2 viruses showed that A/swine/Kanagawa/2/78 (H1N2) virus was more similar to swine (H1N1) virus than to A/Kumamoto/22/76 (H3N2) virus. Radioactive cDNA was prepared by reverse transcription of the recombinant virus RNA using a dodecadeoxyribonucleotide primer and used in DNA–RNA hybridization experiments. The results obtained in molecular hybridization based on blotting procedures showed that all cDNA segments except gene 6 hybridized efficiently with RNAs of swine (H1N1) influenza virus. The sixth cDNA segment was homologous to the corresponding RNA segment of H3N2 virus. The genetic relatedness of A/swine/Kanagawa/2/78 (H1N2) with either A/swine/Kanagawa/4/78 (H1N1) or A/Kumamoto/22/76 (H3N2) was clearly established by hybridization between the cDNA segment probes and viral RNA. It was concluded that the neuraminidase gene of A/swine/Kanagawa/2/78 (H1N2) was derived from a human H3N2 virus, while the seven other genes were from a swine H1N1 virus.

Temperature-sensitive (ts) mutants obtained by undiluted passages of fowl plague virus at 33 °C have their defects located mainly in RNA segments 3, 4 and 8 as determined by rescue to wild-type with standard ts mutants. This result is different from that obtained after treatment of virus with mutagens, where the frequency of mutations follows roughly the target size of the RNA segments. Many isolates generated after undiluted passages at 33 °C, which seem to have mutations in RNA segments 3 and 4, can be rescued to wild-type. This occurs, however, with certain defined standard ts mutants having a defect in RNA segment 4, but not by other segment 4 mutants. One such mutant, ts 1/93 (ts defect in segment 3), interferes with the multiplication of ts 227 (ts defect in segment 4) at the permissive temperature, presumably at the level of vRNA synthesis, preventing reassortment to wild-type. Similarly, ts 263 (ts defect in segment 3) interferes with the multiplication of ts 1/1 (ts defect in segment 4). For other such interfering mutants, the mechanism preventing reassortment to wild-type is different from that of ts 1/93 or ts 1/1, but is not yet understood. Thus, the number of mutations as determined by rescue with standard ts mutants in isolates obtained by undiluted passages is overestimated due to intrinsic interference.

The uptake of the GTP analogue guanylyl(β,γ-methylene)diphosphonate (GppCH2p) is the same in Semliki Forest virus (SFV)-infected BHK cells as in mock-infected cells, in spite of the fact that protein synthesis is inhibited by GppCH2p more markedly in SFV-infected cells than in control cells. A possible explanation for this difference is that infected cells have a lower concentration of GTP and a lower ratio of GTP:GDP than uninfected cells, and the analogue may thus be a more effective competitive inhibitor of translation. We conclude that in this system, the difference between infected and uninfected cells lies not at the plasma membrane but within the cytoplasm.

Treatment of Daudi cells with successively increasing concentrations of interferon-α resulted in the selection of a cell population which multiplied in the continued presence of 104 units/ml of interferon-α. A number of clones of interferon-resistant Daudi cells were isolated from this population. Two clones, DIF2 and DIF3, were found to exhibit moderate and pronounced resistance, respectively, to both the antiviral and antiproliferative actions of human interferons-α and -β. These clones were also less responsive to the enhancement by interferon of Epstein—Barr virus early antigen expression. Both the surface antigens and karyotype of the interferon-resistant clones were similar to those of parental Daudi cells. After prolonged cultivation in the absence of interferon, DIF3 cells were found to ‘revert’ to an intermediate interferon sensitivity. The interferon sensitivity of clone DIF2 remained unchanged even after more than 1 year in culture.

The mechanism of interferon resistance was studied in two clones of Daudi cells, DIF2 and DIF3, which exhibit respectively moderate and pronounced resistance to both the antiviral and antiproliferative actions of human interferons-α and -β. Clones DIF2 and DIF3 were found to possess specific high affinity interferon receptors similar to those of parental Daudi cells. However, DIF2 cells, which have a tetraploid karyotype, had approximately twice as many interferon-binding sites as either DIF3 or parental Daudi cells. One of the first detectable changes in Daudi cells following interferon treatment is a rapid increase in the intracellular concentration of cyclic GMP. No increase in cyclic GMP was observed in DIF2 or DIF3 cells treated with interferon-α. However, neither DIF2 nor DIF3 cells respond to sodium azide, a non-physiological inducer of cyclic GMP. Interferon treatment was found to induce the production of 2′-5′-oligo-isoadenylate synthetase in DIF2 and DIF3 cells in a manner similar to parental Daudi cells, indicating that these cells possess functional interferon receptors. The levels of 2′-5′-oligo-isoadenylate synthetase and 2′-5′ A phosphodiesterase activity were similar in all three cell lines, suggesting that the interferon resistance of clones DIF2 and DIF3 was not due to a deficiency of pp(A2′ p)nA.

We describe the characteristics of a general assay for eukaryote transcription-control sequences using the herpes simplex virus (HSV) thymidine kinase (tk) gene. After transfection of cultured cells with tk-containing recombinant plasmids, two assays were used to measure gene expression: short term or transient levels of tk mRNA and TK enzyme activity, and the rate of biochemical transformation from a TK− to a TK+ phenotype in selective growth medium (HAT). Deletion of the endogenous tk promoter results in 500-fold inactivation of gene expression. Replacement with exogenous transcription-control sequences from the human epsilon globin, mouse β major globin, simian virus 40 and Moloney murine sarcoma virus (MoMuSV) genomes results in reactivation of gene expression. The presence of enhancers or activators of gene expression can also be conveniently measured. The transient expression assay ranged over two orders of magnitude while the transformation assay was almost two orders of magnitude more sensitive using the same recombinants. Analysis of the transcriptioncontrol domains in the MoMuSV LTR sequences shows the presence of both an enhancer and a promoter whose activity equalled that of the tk endogenous promoter. Insertion of the LTR promoter between the LTR enhancer and the tk promoter had little effect on modulating gene expression, suggesting no absolute preference for proximal promoters by this element. The different levels of gene expression obtained appears to be mediated by transcriptional control of full-length tk mRNA. There was an apparent correlation between the results obtained with the transient expression and transformation assays. However, cultured transformed cells all contained roughly the same levels of tk DNA, tk mRNA and tk enzyme activity. We propose that initial expression levels have a major effect in determining the transformation efficiency but that additional genetic controls are superimposed in cells grown in selective HAT medium.

Herpesvirus saimiri-specific proteins from the nuclear fractions of productively infected owl monkey kidney cells were dissociated from virus and host DNA by treatment with 2 m-NaCl or separation on Urografin density gradients. Empty virus capsids remained intact and could be separated from major non-structural proteins (110K, 51K and 48K) and from a subset of structural proteins (130K, 29K and 12K), either by Urografin gradient sedimentation or differential centrifugation. The DNA in such soluble extracts of nuclear proteins was efficiently removed by spermine precipitation, together with the host cell histones and large fractions of the 130K and 12K structural proteins. Proteins in the spermine-soluble fraction were analysed by affinity chromatography on columns of single-stranded calf thymus DNA coupled to cellulose. Two major structural proteins (130K and 12K), whose synthesis was sensitive to phosphonoacetic acid (PAA), and one minor PAA-resistant structural protein (29K) bound to DNA-cellulose. The major PAA-resistant 110K non-structural protein and the PAA-resistant non-structural 51K and 48K phosphoproteins were efficiently released into the spermine-soluble fraction and also bound to DNA-cellulose as did the 76K protein and minor species of 42K, 39K, 34K, 25K and 21K. Virus-specific proteins were eluted from such columns by buffers containing 0.4 m-NaCl or by heparin in lowsalt buffers. Polypeptides from virus particles, infected cell extracts, or samples of eluates from DNA-cellulose chromatography, were separated by SDS-polyacrylamide gel electrophoresis, transferred onto nitrocellulose filters and probed for their ability to bind labelled polynucleotides. The non-structural 51K phosphoprotein, the 12K and 29K structural proteins and a 100K virion polypeptide all bound labelled DNA. However, the binding activities of the 130K protein from virions or purified by affinity chromatography and of the 110K polypeptide could not be demonstrated reproducibly after transfer from SDS gels to nitrocellulose. Comparisons of the present results on the properties of the herpesvirus saimiri-specified DNA-binding proteins with published accounts of the DNA-binding proteins of other herpesviruses, suggest some striking similarities with the DNA-binding proteins of the Epstein—Barr virus.

The distribution of restriction endonuclease (RE) sites was compared for 84 herpes simplex virus type 1 (HSV-1) isolates obtained from the ganglia, facial lesions, genital lesions and from brain tissue from herpes encephalitis cases. The isolates came from Canada, the U.K., the U.S.A. and Japan. Out of a total of 224 sites identified, 87 were variable. Three of the 30 most variable sites were at significantly (P < 0.05) different frequencies in groups of isolates from distinct anatomical sites of isolation; one of these, and a further two sites, were at significantly different frequencies in groups from distinct geographical origins. There are at least two inter-related linkage groups. However, most of the site combinations appear to be random. The variability of RE sites in contiguous genome segments, which include both non-coding and coding sequences, show a marked heterogeneity, indicating that some viral gene sequences are more variable than others. The three RE sites at different frequencies in viral groups from distinct anatomical sites of isolation are in two genome segments: map units 27 to 35 and 50 to 57. We infer from the observed associations with anatomical site of viral isolation that part of at least one of these segments may modulate viral virulence in man following infection.

In contrast to other viral glycoproteins, the herpes simplex virus (HSV) glycoprotein C (gC) binds to the N-acetylgalactosamine-specific Helix pomatia lectin (HPA). In the present paper gC was purified by affinity chromatography with monospecific antibodies and the purified glycoprotein was subjected to protease digestion. HPA-binding protease-resistant glycopeptides were isolated by lectin affinity chromatography. The isolated structures did not bind to concanavalin A and seemed to lack charged groups as determined by ion-exchange chromatography. In gel filtration, the glycopeptides appeared in two peaks with molecular weights higher than 4000. The HPA-binding structures of gC were synthesized in the presence of tunicamycin, indicating that they belong to the O-glycosyl class of oligosaccharides. In addition to HPA-binding oligosaccharides, synthesis of tunicamycin-resistant wheat germ lectinbinding gC oligosaccharides was demonstrable. These were sensitive to sialidase and apparently unrelated to the HPA-binding oligosaccharides.

Analysis of chromatin preparations from [3H]glucosamine-labelled human foreskin fibroblasts revealed that a chromatin-associated glycopolypeptide with the approximate mol. wt. 130000 (130K) is induced in response to either infection with human cytomegalovirus (HCMV) or serum treatment. Comparative limited proteolysis suggested that the [3H]glucosamine-labelled 130K polypeptides induced by these different stimuli were not identical. This observation was in contrast to results obtained by immunoprecipitation with antisera raised against the 130K glycopolypeptide from serum-induced cells which favoured a relatedness to the 130K polypeptide from virus infected cultures. Two-dimensional separation by isoelectric focusing and SDS–polyacrylamide gel electrophoresis subsequently showed that the 130K glycopolypeptide from serum-induced cells consists of two components, one of which is identical to the single component observed in samples from HCMV-infected cultures. Experiments on the effect of glycosylation inhibitors on DNA replication in HCMV-infected as well as in serum-induced cells support the view that the host-specific chromatin-associated glycopolypeptide may be involved in DNA replication in infected cells.

The nerves supplying the pinna of the ear of mice latently infected in the 2nd and 3rd cervical ganglia were sectioned. Immediately after neurotomy, or some days later, the denervated ears were stripped with cellophane tape to induce recurrent disease. The cervical ganglia and skin of the ears were tested for the presence of infectious virus at different times after neurotomy. Nerve section induced a low incidence of reactivation of virus in ganglia. After neurotomy, infectious virus was isolated from the skin very rarely and recurrent disease was not seen.

Inhibition constants (K is) were used as an estimate of the ability of various nucleoside analogues to be recognized as substrates by the deoxythymidine kinases (dTKs) of a 5-methoxymethyldeoxyuridine-resistant (MMdUr) mutant of herpes simplex virus type 1 (HSV-1) and its parent wild-type (wt). It was found that the K is for the 5-position analogues MMdU, [E]-5-(2-bromovinyl)deoxyuridine, bromodeoxyuridine and iododeoxyuridine were increased approximately three-to fivefold, suggesting that they were poorer substrates for the MMdUr dTK than for the wt dTK. With the 2′ analogues arabinosylthymine and 2′ fluoro 5-methylarabinosyluracil, however, the K is were increased to a much greater extent, 80- and 240-fold, respectively. These findings suggest that the resistance of the mutant MMdUr to these analogues may be due to a mutation(s) in the viral dTK that directly affects binding at the 2′ recognition site and indirectly at the 5, while still allowing substantial activity with the natural substrate deoxythymidine.

The 7.2 kb genomic DNA of bovine papillomavirus type 3 (BPV-3) was molecularly cloned using its unique EcoRI site, and the 7.3 kb genome of BPV-4 was cloned using its single BamHI site. The viral genomes were compared by liquid hybridization, Southern blot hybridization and heteroduplex mapping. Low stringency hybridization conditions revealed that the genomes are colinear but the sequences are extensively mismatched. The relative alignment of the restriction endonuclease maps of the two viral genomes has been determined. It was found that the genomes as linearized for cloning are out of phase by 1.7 kb, so that the single EcoRI site of BPV-3 appears to coincide with the BPV-4 EcoRI site at 0.22 map units. It is concluded that the genomes of BPV-3 and BPV-4, both of which cause true epithelial warts, share the same physical organization but exhibit sequence divergence.

The interferon receptors of C3H/10T mouse cells respond differently to α and β interferons under certain conditions. If C3H/10T cells which have been maintained in logarithmic growth phase are exposed to trypsin or Pronase immediately before they are treated with mouse interferons, they evince an antiviral response to α interferon but not to β interferon. In contrast, contact-inhibited C3H/10T cells, L-929 mouse cells or human HEL cells lost the ability to respond to both α and β interferons after treatment with trypsin or Pronase. When L-929 cells are incubated at 37 °C following exposure to these proteolytic enzymes, they completely regain their ability to respond to mouse β interferon within 2 h. These observations suggest that the receptors for α and β interferons are different in their topographical distribution in C3H/10T cells.