- Volume 72, Issue 8, 1991

Complementary DNA representing the genomic M RNA segment of the Prospect Hill (PH) Hantavirus was cloned and its nucleotide sequence determined. The PH virus M RNA segment is 3707 nucleotides in length and has a long open reading frame in the viral complementary-sense RNA with a coding capacity of 1142 amino acids. The predicted gene product of the PH virus M segment was compared with the corresponding gene products of Hantaan virus strain 76–118 (Hantaan), Sapporo rat virus strain SR-11 (SR) and Puumala virus strain Hällnäs B1 (Hällnäs). The amino acid sequence identities between the G1 and G2 proteins of PH virus and Hällnäs virus are respectively 74% and 79%. In contrast, the amino acid sequence similarities between the G1 proteins of PH virus and SR virus or Hantaan virus are only 50%. However the G2 proteins of SR and Hantaan viruses were more closely related to the G2 protein of PH virus with amino acid sequence similarities of approximately 62%. The G1 proteins of all four viruses had three potential asparagine-linked glycosylation sites conserved and there was one conserved site in the G2 proteins. Hydrophilicity plots of the four virus glycoproteins were very similar. The region of greatest hydrophilicity was conserved in the Hällnäs, SR and Hantaan viruses, and was located near the C terminus of the G1 protein, whereas the region of greatest hydrophilicity in the PH virus glycoprotein precursor is located closer to the N terminus of the G1 protein. Our data demonstrate that despite differences in the serotypic profiles and virulence of PH and Hällnäs viruses, their G1 and G2 proteins are closely related. We conclude that PH and Hällnäs viruses may have evolved along a separate evolutionary pathway in the Hantavirus genus from that of SR and Hantaan viruses.

Three anti-VP7 monoclonal antibodies (MAbs) which neutralized only two strains (K8 and S12) of five serotype 1 human rotaviruses (HRVs) were obtained, and neutralization epitopes recognized by these ‘monotype-specific’ MAbs were analysed by epitope mapping and sequencing of the VP7 genes. Neutralization-resistant mutants of K8 and S12 were selected by the monotype-specific MAbs and serotype 1-specific MAbs prepared previously. Cross-neutralization tests between MAbs and neutralization-resistant mutants of K8 and S12 indicated that epitopes of monotype-specific MAbs operationally overlap with those of serotype 1-specific and cross-reactive MAbs recognizing the S1 region. Sequence analyses of the VP7 genes indicated that VP7s of strains K8 and S12, which belong to a monotype of serotype 1 viruses, possessed amino acids at positions 42 and 87 different from other serotype 1 HRVs. Furthermore, amino acid substitution sites of representative mutants of K8, selected by the monotype-specific MAbs, were identified at positions 96, 97 and 100. These results imply that amino acids in variable region B (amino acids 87 to 101) are involved in the monotype-specific neutralization epitope as well as serotype-specific neutralization epitopes.

A total of 24 monoclonal antibodies (MAbs) against the haemagglutinin (H) and the fusion protein (F) of rinderpest virus (RPV) were used to characterize their antigenic structure and biological properties, and to analyse natural variation in the envelope proteins of morbilliviruses. The anti-H and anti-F MAbs defined seven and three distinct antigenic sites, respectively. The MAbs to six sites on H were able to neutralize the infectivity of RPV. The addition of guinea-pig complement or anti-mouse immunoglobulin increased the virus-neutralizing antibody titre of most of the anti-H MAbs, including those lacking neutralizing activity. One of the antigenic sites on H was conserved among morbilliviruses and the MAbs to this site had haemagglutination inhibition activity against measles virus (MV). The remaining sites were specific for RPV and varied antigenically between strains of RPV. The anti-F MAbs lacked neutralizing activity, but two of the five MAbs did show activity in the presence of complement or anti-mouse immunoglobulin. On the whole, the antigenic sites on F were conserved in some strains of MV, but not in canine distemper virus. All of the sites on the surface proteins were sensitive to SDS and, although those on F were not affected by 2-mercaptoethanol, five of the seven sites on H were destroyed by it. These results suggest that the epitopes on the envelope proteins are conformation-dependent.

Human placental trophoblast cultures produce a mixture of interferons (IFNs) when challenged with Sendai virus. High-performance dye-ligand and immunoaffinity chromatography of a trophoblast IFN (tro-IFN) preparation enabled the isolation of three antigenically distinct IFNs, α I, α II1 and β, with M rs of 16K, 22K and 24K respectively, by reducing and non-reducing SDS-PAGE. The major IFN, responsible for 75% of the total antiviral activity, was tro-IFN-β, with the remaining activity being due to tro-IFN-α I and tro-IFN-α II1, as determined by an antiviral neutralization test using specific anti-human IFN antibodies. The antiviral activities of the tro-IFNs were stable at pH 2.0 for 24 h and tro-IFN-α II1 and -β were shown to be glycoproteins. The three tro-IFNs showed different antiviral activities when assayed on human and bovine cell species; tro-IFN-α I and -α II1 protected both human and bovine (MDBK) cells from virus infection, whereas tro-IFN-β showed a high degree of species specificity, protecting only the human cell types tested.

The antiviral action of cyclopentenone prostaglandins (PGs) is generally associated with alterations in the synthesis and/or maturation of specific virus proteins. In particular, inhibition of Sendai virus (SV) replication in African green monkey kidney cells by PGA1 has been shown to be a cell-mediated event, due to alterations in SV protein glycosylation and accompanied by the induction of a cellular polypeptide of M r 74K. In this report we identify this protein as a heat shock protein (HSP) related to the major 70K HSP group (HSP70). Induction of HSP70 synthesis by PGA1 was found to be dose-dependent, and an accumulation of HSP70 comparable to that occurring after heat shock could be obtained at concentrations of PGA1 that did not inhibit macromolecular synthesis in uninfected cells, but caused a dramatic block of virus replication in SV-infected cells. Induction of HSP70 by PGA1 occurred at the transcriptional level and was not affected by SV infection. HSP70 synthesis was evident between 2 and 3 h after PGA1 treatment, maximal at 12 h and went back to control levels by 26 h after the addition of PGA1, thus preceding virus protein synthesis. Finally, of several PGs tested, only those which possess antiviral activity induced the synthesis of HSP70. These results, together with the observation that suppression of HSP70 synthesis by actinomycin D also abolishes the PGA1-induced alteration of SV glycoproteins, suggest that HSP70 could play a role in the block of virus replication by cyclopentenone PGs.

The effects of vaccination of sheep with a recombinant vaccinia virus (rVV) expressing the bovine leukaemia virus (BLV) envelope glycoprotein (gp60) were studied by determining BLV titres in peripheral blood leukocytes after vaccination and challenge. The proliferation of BLV was suppressed markedly, not only when rVV was inoculated prior to challenge with BLV, but also when it was inoculated after challenge. These results indicate that vaccination with rVV induces protective immunity that can suppress the growth of BLV in carrier animals. Since rVV induced a strong anti-BLV delayed-type hypersensitivity response without producing detectable levels of binding or neutralizing antibodies, and there was no apparent correlation between the humoral immune response and BLV proliferation, a cell-mediated immune response was assumed to play a major role in protective immunity.

We have isolated a maedi-visna-like virus from the peripheral blood mononuclear cells of a British sheep displaying symptoms of arthritis and pneumonia. After brief passage in fibroblasts this virus (designated EV1) was used to infect choroid plexus cells. cDNA clones of the virus were prepared from these cells and sequenced. Gaps between non-overlapping clones were filled using gene amplification by the polymerase chain reaction. The genome structure is similar to that described for visna virus strain 1514, and differs from that described for visna virus strain SA-OMVV in not having a W reading frame. Overall the genome differs by about 20% between each of these strains, but there is fivefold variation in the amount of divergence of derived amino acid sequences of different open reading frames. Two sequenced EV1 clones each contain only one copy of the 43 bp repeat, with paired AP-1 sites, which is a feature of other ruminant lentiviral long terminal repeats (LTRs). However, analysis of viral DNA in infected cells by gene amplification shows that LTRs with two repeats do occur, albeit at a relatively low frequency.

One mechanism by which viral tropism may be controlled is by the expression of a specific virus receptor on the cell surface. This paper reports the identification of a putative cellular receptor for visna virus, the prototype virus of the family Lentiviridae. Using a virus overlay protein blot assay we identified a group of polypeptides of apparent M r 30K to 33K which interacts with visna virus and is present on permissive but not non-permissive cells. A rat polyclonal anti-ovine major histocompatibility complex (MHC) class II antigen (Ag) serum raised to immunopurified MHC class II Ag, but not preimmune serum, blocked the interaction of visna virus with these polypeptides. In an ELISA, immunopurified MHC class II Ag bound to visna virus but not to bovine parainfluenza 3 virus. Preincubation of visna virus with immunopurified soluble MHC class II Ag resulted in a marked decrease in virus-induced syncytium formation, i.e. preincubation with class II Ag inhibited infection with visna virus, but we have been unable to inhibit infection using class II Ag-specific antisera. These results suggest that ovine MHC class II Ag acts as a component of a cellular receptor for visna virus. This is of particular interest owing to the close similarities between visna virus and human immunodeficiency virus (HIV), and the relationship between MHC class II and CD4, the cellular receptor for HIV. It is also of relevance to recent reports that a growing number of viruses utilize polypeptides of the Ig supergene family as receptors.

Eighteen monoclonal antibodies (MAbs) directed against the purified human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) protein were produced. The antibodies were characterized by competitive ELISAs and Western blot experiments, and with nested, nine amino acid long peptides representing the whole 560 amino acid RT protein. By ELISA, the MAbs react with a minimum of seven epitopes of the protein. Four of the epitopes were located on the N-terminal 51K subunit and the remaining three epitopes were located at the C-terminal end of the protein. Using synthetic peptides, two epitopes at the N-terminal part were located at amino acids 294 to 302 and 350 to 354, respectively, from the N-terminal start of the protein. One epitope was located at amino acids 442 to 450, just after the cleavage site between the N-terminal and C-terminal subunit at position 440. Antibodies located at amino acids 294 to 302 could inhibit the RT enzymic activity of the protein. Two other MAbs, directed at the N-terminal and C-terminal parts of the protein, could also inhibit RT activity.

Apparently conflicting results have been reported regarding the role of env glycoprotein glycans in human immunodeficiency virus type 1 (HIV-1) infectivity and cytopathogenicity. Whereas we have shown that enzymic removal of carbohydrates from mature envelope glycoproteins has only limited effect on the ability of HIV-1 to bind to CD4 and to infect target cells, sugar analogues that interfere with the glycosylation process of the nascent molecule markedly reduce virus infectivity. Here we have investigated the effect of a glucosidase inhibitor, 1-deoxynojirimycin (dNM), on the bioactivity and immunoreactivity of precursor gp160 produced by recombinant vaccinia virus-infected BHK-21 cells (rgp160). dNM (4 mm) did not affect the amount of rgp160 recovered nor its secretion from the cells. As described by other authors the effect of dNM was incomplete, resulting in the production of rgp160, the glycosylation of which was heterogeneous with respect to apparent M r distribution and to sensitivity to endoglycosidase H and endoglycosidase F, all the species being susceptible to N-glycanase. A major reduction of the binding to CD4+ cells was noted with rgp160 produced by dNM-treated cells using a quantitative indirect immunofluorescence assay and labelling with polyclonal human anti-HIV IgG. Similarly, dNM treatment altered the accessibility to murine monoclonal antibody 110-4 of the exposed V3 loop of HIV-1 gp120 by at least 10-fold, as determined by either ELISA capture assay or immunoaffinity purification. Such bioactivity and conformation modifications, which result from the abnormal folding of the nascent glycoprotein due to aberrant glycosylation, may account for the impaired HIV-1 infectivity elicited by dNM.

Using an adenovirus-hepatitis B virus (HBV) recombinant, expression of the HBV surface antigen (HBsAg) genes was examined in various cell lines using S1 nuclease mapping and radioimmunoassay. The steady-state level of the 2.4 kb RNA encoding the large HBsAg was much greater than, or the same as, that of the 2.0 kb RNA, encoding the middle and major HBsAgs, in primate cells, but was negligible in non-primate cells, as is the case in most expression systems. According to the amount of 2.4 kb RNA expressed, cells were classified into three groups: those in which (1) the amount of 2.4 kb RNA was much greater than that of 2.0 kb RNA (HepG2 and JHH-4), (2) the amount of 2.4 kb RNA was the same as that of 2.0 kb (Hul-1, HeLa and other non-hepatic primate cells), and (3) the amount of 2.4 kb RNA was less than one-tenth of that of 2.0 kb RNA (rodent cells). Radioimmunoassay revealed that most HBsAg is located intracellularly in primate cells, but is secreted into the culture medium of rodent cells. The expression of 2.4 kb RNA was unaffected by an inhibitor of DNA synthesis in HepG2 cells, which are of human liver origin, whereas it was strongly inhibited in human non-hepatic HeLa cells.

The cis-acting elements for the early and late promoters, as well as the enhancer in the prototype strains of human polyomavirus BK (BKV) are located within a 500 bp intergenic region. We previously studied the specificity of protein binding in this region in vitro and showed that the interaction of proteins of the nuclear factor-1 (NF-1) family is crucial for early promoter activity. We have now extended our study to the BKV late promoter. We show that the late promoter activity in HeLa cell extracts is poor compared to the activity of the early promoter. Using a high template to protein ratio, multiple start sites were detected by primer extension analysis. DNase I protection experiments revealed the presence of three NF-1 binding sites in the late side, in addition to those identified previously in the 68 bp repeats and C element. Competition transcription assays using binding sites for NF-1, AP-1, Sp-1 and a complete 68 bp repeat indicated that only the 68 bp repeat and the NF-1 binding site competed significantly with the late promoter activity. A point mutation in the NF-1 binding site, which destroys the ability of the oligonucleotide to bind NF-1, also impaired its capacity to compete with the late promoter. The ability of NF-1 to activate both the early and late promoters suggests that the proteins of this family act as a bidirectional transcriptional activator in this virus.

Serial undiluted passage of polyoma virus derived from transfection of mouse fibroblasts with well defined wild-type genomes results in the appearance of a very heterogeneous population of defective virus particles. Many of these variants show duplications which not only contain the cis-acting region minimally required for efficient replication, but also other regions. A detailed analysis of the duplication patterns appearing in high multiplicity infections is presented. We performed heteroduplex analyses of duplicated fragments using mung bean nuclease and demonstrated that the pattern of duplication junctions is conserved qualitatively and quantitatively. However, the distribution of fragment sizes varied in a number of independently derived virus stocks. Amplification of viral nucleotide sequences is an early event in virus replication, occurring at least as early as 3 days post-transfection. The pattern of duplication did not change significantly in successive early passages at high multiplicity. Although duplication was accompanied by deletion of various parts of the viral genome, a sequence bordering the duplications at the late side of the origin of replication was retained as a single copy in all defective viruses. The relevance of these findings to the mechanism that creates the duplications and the biological activity of defective virus is discussed.

Herpesvirus (h.) saimiri, an infectious agent of squirrel monkeys, is capable of persisting in Tlymphocytes of various primate species. It has been used as a vector for the functional analysis of regulatory genes in primary human T lymphocytes. As it is not yet known whether other cell types are capable of supporting viral persistence, various human cell lines were investigated using selectable h. saimiri recombinants. The lines chosen represent cells from the epithelium and connective tissue as well as from all haematopoietic lineages, i.e. cells of B and T lymphoid origin as well as myeloid-, fibroblast- and carcinoma-derived cultures converted to Geneticin or hygromycin B resistance, and harbouring episomal DNA of the selectable recombinants. The Burkitt's lymphoma-derived cell line Raji also contained simultaneously persisting episomes of the Epstein—Barr virus. Most of the cell cultures except a pancreatic carcinoma line and foreskin fibroblasts did not produce infectious virus. These observations show that a herpesvirus genome can persist episomally in a broad range of cultured cell types. The variety of infectable cell types and species suggests the presence of a widely distributed and well conserved virus receptor for h. saimiri. Thus the h. saimiri genome could be applied more generally as a vector.

We have described two types of oligosaccharide modification influencing the antigenicity of the herpes simplex virus type 1 (HSV-1)-specified glycoprotein C (gC-1). First, the expression of several epitopes belonging to antigenic site II of gC-1 is dependent on the peripheral galactose of N-linked oligosaccharides. We have also shown that treatment of HSV-1-infected cells with 5-n-propyl-2′-deoxyuridine (PdU) under certain circumstances results in other modifications of peripheral carbohydrate determinants, which are associated with increased antigenic activity of gC-1. In the present study we have mapped and characterized the epitopes susceptible to PdU induction by analysing the reactivity to a number of monoclonal antibodies defining several epitopes of antigenic sites I and II. The results indicate that the strict galactose dependence of epitopes and the PdU-induced increase of antigenic activity are independent and unrelated phenomena. Thus, we identified galactose-dependent epitopes that were not PdU-inducible and vice versa, and some epitopes were both galactose-dependent and PdU-inducible. The results support a model where PdU treatment blocks synthesis of an antigen-masking carbohydrate determinant. In addition, PdU treatment of HSV-1-infected cells seemed to increase the antigenic activity of other HSV-1 glycoproteins.

The gene regulatory immediate early protein, pp89, of murine cytomegalovirus interacts with both DNA-associated and isolated histones in vitro. We characterized the histone-binding region of pp89 and its cellular localization during cell division to examine the possible interaction between pp89 and chromatin. pp89 expressed constitutively in cell line BALB/c 3T3 IE1 does not interact with condensed chromatin. As observed in infected cells, pp89 is localized within the nucleus of cells during interphase but spreads throughout the cell plasma following degradation of the nuclear membrane during early mitosis. In late telophase, pp89 is reorganized within the nucleus. Analysis of pp89 deletion mutants and of fragments generated by cleavage at pH 2.5 revealed that the regions responsible for association with histone are located between amino acids 71 and 415, and are not identical with the domain that shows homology to histone H2B or the highly acidic carboxy-terminal region. A potential gene-activating role of the high affinity of pp89 for isolated histones and the low affinity for DNA-associated histones is discussed.

The envelope of human cytomegalovirus contains a family of disulphide-linked glycoprotein complexes designated gC-I which contain two glycoproteins of 52000 M r (gp52) and 93000 to 130000 M r (gp93-130). Epitopes recognized by several of our gC-I gp52-specific monoclonal antibodies (MAbs) were previously assigned to three domains based on reactivity with gC-I in a competitive binding assay. In this report, we have used additional gC-I MAbs to characterize three distinct discontinuous epitopes in the gC-I complexes. Two of these epitopes were in Domain I and one in Domain III. These epitopes were resistant to proteolysis, heat denaturation and SDS treatment. However, the discontinuous epitopes were lost after reduction of disulphide bonds. After digestion of gC-I complexes with chymotrypsin,: two fragments of 43000 (43Κ) M r and 34000 (34Κ) M r were obtained which contained all discontinuous and continuous epitopes recognized by our gp52 MAbs. The M r of these fragments could not be reduced further by longer digestion or by use of other proteases such as trypsin or pronase. The 43Κ fragment contained N-linked oligosaccharides not detected in the 34Κ fragment. These oligosaccharides may have prevented a complete proteolytic digestion so that the 34Κ fragment was not always obtained. It was established that 80 to 90% of the mass of these fragments was contributed by gp52. Thus the discontinuous epitopes were composed primarily of gp52 and not gp93-130.

The heterogeneity of low passage human cytomegalovirus (HCMV) strains was determined by HindIII typing of 28 clinical isolates from transplant patients. These data have shown that, in general, each patient's strain has a unique restriction profile, usually comprising combinations of HindIII sites present in one or more of the tissue culture-adapted strains AD169, Towne and Davis. To map sequence changes in a more refined manner we performed detailed analyses of 33 low passage clinical isolates, including those aforementioned, analysing a sequence within glycoprotein B containing a major neutralizing epitope. A 149 bp sequence containing the epitope (amino acids 608 to 625) was amplified using the polymerase chain reaction, the products were cloned and their DNA sequence was determined. Comparison of the DNA and deduced amino acid sequences with those of HCMV strain AD169 revealed that there was a high degree of conservation of the epitope between the 33 clinical isolates. However 10 of the isolates possessed silent mutations and three isolates contained mutations producing amino acid changes within the neutralizing epitope. The possible functional significance of these changes is discussed.