- Volume 72, Issue 5, 1991

Different passages of the vaccinia virus strain Ankara (CVA wild-type) during attenuation to MVA (modified vaccinia virus Ankara) have been analysed to detect alterations in the genome. Physical maps for the restriction enzymes HindIII and XhoI have been established. Six major deletions relative to the wild-type strain CVA could be localized. They reduce the size of the entire genome from 208 kb (CVA wild-type) to 177 kb for the MVA strain. Four deletions occurred during the first 382 passages and the resulting variant (CVA 382) displays an attenuated phenotype similar to that of the MVA strain. The deletions are located in both terminal fragments, affect two-thirds of the host range gene K1L and eliminate 3.5 kb of a highly conserved region in the HindIII A fragment. During the next 190 passages leading to MVA two additional deletions appeared. Again, one is located in the left terminal fragment, and the other includes the A-type inclusion body gene. Neither of the deletions appear to participate in further attenuation of the virus. Rescue of the partially deleted host range region with the corresponding wild-type DNA restored the ability of the attenuated strains MVA and CVA 382 to grow in some non-permissive tissue cultures. Nevertheless, the complete host range of the wild-type strain was not recovered. Also, plaque-forming behaviour and reduced virulence were not influenced. From the data presented it may be concluded that the partially deleted host range gene is not solely responsible for attenuation.

A simple method has been developed for detecting a broad range of genital human papillomavirus (HPV) types using the polymerase chain reaction (PCR). We utilized two consensus sequence primer pairs within the E6 and E7 open reading frames to amplify HPV DNA; malignant HPV DNA (from HPV-16, -18, -31, -33, -52b and -58) was amplified using the pU-1M/pU-2R primer pair whereas benign HPV DNA (from HPV-6 and -11) was amplified using the pU-31B/pU-2R primer pair. Identification of the amplification product was confirmed by restriction enzyme digestion. In this study, a pU-1M/pU-2R-mediated PCR was successfully applied to 39 cervical carcinoma specimens; HPV-16 was detected in 19 cases, HPV-18 in five cases, HPV-31 in two cases, HPV-33 in two cases, HPV-52b in one case, HPV-58 in three cases, and an unknown type(s) was detected in four cases. Overall, the prevalence of HPV was 84.6%. The results indicate that this detection system is useful for the detection of HPVs not only of known types but also of new types.

The complete nucleotide sequence of the gene encoding the matrix protein (M) of the porcine paramyxovirus LPMV has been determined. The gene is 1376 nucleotides long including 5′ and 3′ non-coding sequences with a protein-coding sequence of 1107 nucleotides. The deduced protein, containing 369 amino acids with a calculated M r of 41657, is hydrophobic overall with a net positive charge of +17.5. Comparative sequence analysis revealed high amino acid homology to other paramyxovirus M proteins, with the highest degree of identity (46%) with the human mumps virus. This is strong evidence that the porcine paramyxovirus LPMV is a genuine member of the paramyxovirus genus.

Chemical and enzymic cleavages of the F1 subunit of the fusion (F) protein of respiratory syncytial (RS) virus showed that the sequence 184-Gly to 314-Trp reacted with neutralizing monoclonal antibodies (MAbs). Twelve synthetic peptides covering a part of this sequence were analysed for their immunoreactivity with neutralizing MAbs and anti-RS virus rabbit serum. Two sequential antigenic domains corresponding to amino acids 200 to 225 and 255 to 278 were defined with anti-RS virus rabbit serum. The peptides 205–225 and 259–278, belonging to these antigenic domains, inhibited binding to the F protein and the neutralizing activity of the anti-RS virus rabbit serum. One MAb (RS-348) reacted with peptides containing amino acids 200 to 225. Moreover, the peptide 205–225 induced an anti-peptide rabbit serum neutralizing RS virus in vitro. These results indicate that the sequence from residues 200 to 225 was present in one of the immunodominant sites of the F protein.

A group A rotavirus designated L338 was isolated from the faeces of a diarrhoeic foal and was compared to 11 standard G serotype strains of group A rotaviruses by cross-neutralization. It was clearly distinct from serotypes G1 to G11 and thus representative of a novel rotavirus G serotype tentatively designated G13. The nucleic acid sequence of the virion protein 7 (VP7) coding region was determined and the deduced amino acid sequence compared to published sequences. Within VP7 regions A and B, L338 was clearly distinct from serotypes G1 to G12 (excluding G7 which has not been sequenced), but region C was very similar to those of G3 and G8. This further questions the significance of region C in determining serotype specificity of at least three distinct rotavirus G serotypes.

Orungo and Lebombo orbivirus isolates were examined for their intra- and intergroup genetic relatedness by blot hybridization and gene reassortment; blot hybridization was also used to examine the relatedness of selected Orungo and Lebombo isolates to known orbiviruses. Among the Orungo isolates, >74% sequence similarity was shown in the majority of their genes. Gene 2 was the most divergent gene, with four unique types identified, and genes 5, 6 and 10 were variant among the isolates. Plaque reduction neutralization tests revealed at least four serotypes, a result which correlated with the hybridization data. Gene reassortment was shown between two representative Orungo isolates. Among the Lebombo isolates, two hybridization types were identified between which gene reassortment was demonstrated. Unique genes were not shown, whereas genes 2, 5 and 10 exhibited minor sequence variability. Geographic distribution correlated with relatedness among the Lebombo isolates, which was not the case among the Orungo isolates. Orungo and Lebombo viruses did not cross-hybridize or reassort their genes in vitro, in intergroup studies. In blot hybridization tests of Orungo and Lebombo isolates with known orbivirus serogroups and ungrouped orbiviruses, no strong cross-hybridization was seen. These results demonstrate that Orungo and Lebombo are distinct from each other and from other orbiviruses, and should therefore be recognized as two new Orbivirus serogroups.

The segment 10 (S10) genes of African horsesickness virus (AHSV), Palyam virus and epizootic haemorrhagic disease virus were translated in vitro in a rabbit reticulocyte lysate system. Each of the S10 genes encoded two proteins, NS3 and NS3A, which were shown to be related by peptide mapping. Cloned copies of the S10 genes of two AHSV serotypes (AHSV-3 and AHSV-9) and Palyam virus were sequenced and compared to each other and to the nucleotide sequence of bluetongue virus (BTV) gene S10. Two in-phase ATG translation initiation codons reported for the S10 genes of BTV-10 and BTV-1 were conserved in the S10 genes of AHSV-3, AHSV-9 and Palyam virus, and would be able to initiate synthesis of NS3 and NS3A respectively. Comparison of the amino acid sequences of NS3 of AHSV-3 and AHSV-9 identified two areas of approximately 45 amino acids which displayed high (98%) similarity. One of these areas corresponded to the only region which displayed more than 50% amino acid similarity between NS3 of BTV, AHSV and Palyam virus. This region could represent an important structural or catalytic site of the protein. The overall amino acid similarity outside this conserved region was between 13% and 29%.

A subgenomic cDNA clone from hepatitis A virus strain HM175, composed of the last eight nucleotides of the 5′ non-translated region and the first 2248 nucleotides of the coding sequence (P1 region), was inserted into a vector under the control of the T7 promoter. Restriction enzyme digestion at sites within the structural region and subsequent transcription in vitro yielded RNA products which were translated efficiently in rabbit reticulocyte lysates to produce proteins of the predicted sizes. The translation products were specifically precipitated with antipeptide antisera; these reactions were not affected by denaturation of the antigens by boiling in 1% SDS. The translated proteins were also precipitated by antivirion antisera, but recognition was totally abolished following denaturation. Thus antivirion antisera recognized conformation-dependent epitopes expressed on the translated products exclusively.

A peptide from the carboxyl-terminal region of the Mengo virus capsid protein VP1, representing residues 259 to 277, can induce serum neutralizing (SN) antibodies in both the mouse and guinea-pig. This peptide, termed F164, also induces high levels of protective neutralizing antibodies in mice subsequent to immunization; 87 to 100% of mice are refractory to the effects of an intraperitoneal challenge of 100 LD50 of Mengo virus. The mouse model discussed herein will prove useful for studying the immune response to Mengo virus and evaluating the immunogenicity of individual viral components.

Immunization of BALB/c mice with a single dose of the Sabin type 1, type 2 or type 3 poliovirus vaccine strains stimulated cross-reactive T helper cell responses detected by both in vitro proliferation and interleukin (IL)-2/IL-4 production. Although the polyclonal T cell responses were cross-reactive, the results also suggest that a proportion of the T cells were directed against serotype-specific determinants. In contrast, neutralizing antibodies, assayed in the serum from the same animals, were predominantly serotype-specific and only reached significant titres after secondary immunization. A comparison of the immunogenicity of poliovirus administered subcutaneously in Freund’s complete adjuvant or intraperitoneally as an alum precipitate or without adjuvant, showed that optimum responses were obtained by immunization with virus in the presence of alum. An examination of the effect of heterotypic priming showed that immunization with type 2 virus primed for a secondary antibody response to each of the three serotypes, whereas priming with type 1 or type 3 viruses could only generate a secondary antibody response to the homologous virus or to type 2 virus.

A recombinant baculovirus expressing the A antigen (A Ag) of Marek’s disease virus (MDV) was constructed. In Spodoptera fruiperda (Sf) cells infected with the recombinant virus, A Ag expression was localized to the cell surface. Only a small amount of recombinant A Ag was detected in the culture supernatant of infected Sf cells, but authentic A Ag is mainly secreted into the culture supernatant of MDV-infected chicken embryo fibroblasts. Cell surface-associated recombinant A Ag seemed to be slightly larger than authentic A Ag, whereas the secreted recombinant A Ag seemed to be smaller. The recombinant A Ag was shown to be reactive with the sera of MDV-infected chickens by immunodiffusion studies and ELISA. Sera of chickens immunized with recombinant A Ag formed a precipitin line with the culture supernatant of MDV-infected chicken embryo fibroblasts in an immunodiffusion test. These results indicate that the recombinant A Ag expressed by the recombinant baculovirus retains the antigenic and immunogenic properties of the authentic A Ag.

An oligonucleotide complementary to the splice donor sequence of the 1.8 kb gene family produced from the BamHI-H region of Marek's disease virus (MDV) DNA inhibited the proliferation of the MDV-derived lymphoblastoid cell line, MDCC-MSB1 (MSB-1), but not that of the avian lymphoid leukosis-derived lymphoblastoid cell line, LSCC-BK3. Colony formation in soft agar was also inhibited by treatment of MSB-1 cells with the antisense oligonucleotide. It is hypothesized that expression of the 1.8 kb gene family produced from the BamHI-H region is directly associated with the maintenance of the tumorigenic state of transformed Marek's disease-derived lymphoblastoid cells.

The DNA sequence of the HindIII T fragment of human cytomegalovirus strain AD169 has been determined. This 6225 bp sequence has been analysed for transcription signals and probable open reading frames. Similarities with herpes simplex virus, varicella-zoster virus and Epstein-Barr virus genes were observed for three of the predicted open reading frames; a virion protein and a unique DNA helicase are believed to be the functional products of two of these open reading frames. Two other open reading frames are novel in that no homologues could be found, either in the known herpesvirus sequences or in the Protein Identification Resource database. Both of these open reading frames also lie in the genomic coding region of a 5.0 kb RNA which is transcribed throughout the infectious cycle.

Infection with human herpesvirus 6 (HHV-6) was found to up-regulate expression of human immuno-deficiency virus and human T cell leukaemia virus type I (HTLV-I) long terminal repeat sequence (LTR), and herpes simplex virus type 1 (HSV-1) gD chloramphenicol acetyltransferase (CAT) constructs transfected into the T cell line, J. Jhan. Activation by HHV-6 was due to one or more viral proteins produced early in infection and, in the case of the HTLV-I LTR, was synergistic to induction mediated by the HTLV-I tax gene product. Neither the HTLV-I enhancer nor basal promoter elements of the HSV-1 gD gene were essential for activation and no increase in accumulated HTLV-I mRNA was observed due to HHV-6 infection. Induction by HHV-6 was found to be dependent on the reporter construct used, because the CAT gene and, to a lesser extent, the HSV-1 thymidine kinase gene were responsive to HHV-6 infection although no significant activation of growth hormone constructs was observed. Our results bear a strong resemblance to those obtained for the Epstein-Barr virus BMLF1 gene, indicating that the major HHV-6 trans-activator may be a homologue of this gene.

It has been suggested that heparan sulphate has a receptor function in the initial phase of the attachment of herpes simplex virus (HSV) to cells. We have studied the influence of glycosaminoglycans on cell adsorption of, and plaque formation by, HSV-1 and HSV-2, with regard to the role of saccharide structure, chain length and charge density. Heparin and highly-sulphated heparan sulphate (1.5 sulphate groups/disaccharide unit), but neither chondroitin sulphate nor dermatan sulphate, were found to compete with the cellular receptor for attachment of HSV. Heparan sulphate preparations of low sulphate content (0.5 and 0.7 sulphate groups/disaccharide unit) failed to show any significant interaction with HSV. Oligosaccharides generated by partial deaminative cleavage of heparin were used to determine the minimum molecular size required for the binding of virus; the smallest oligosaccharide which reacted with HSV was composed of 10 monosaccharide units. The importance of charge density was demonstrated more directly by subfractionation of the heparin dodecasaccharide fraction by anion-exchange HPLC. The virus-binding capacities of the four resulting dodecasaccharide subfractions increased from the least sulphated to the most heavily sulphated fraction. The results reported are discussed in relation to virus-receptor interactions involved in the attachment of HSV, including the reported binding of HSV to the fibroblast growth factor receptor.

Computer-assisted comparison of the herpes simplex virus type 2 (HSV-2) ribonucleotide reductase large subunit (RR1) sequence with the known primary structures of other RR1 proteins revealed a motif consisting of five leucines occurring at every seventh residue between positions 409 to 437. This motif is specific to HSV RR1 proteins. A synthetic oligopeptide (LA-4) corresponding to 15 residues in the internal portion of the motif inhibited HSV-2 RR activity. In immunoprecipitation experiments, LA-4 disrupted a complex consisting of RR1, the small RR subunit and a previously uncharacterized 180K protein, apparently of cellular origin. We deduce that the LA-4 sequence represents a critical RR1 site involved in RR complex formation and enzymic activity.

Genomes of herpesvirus aotus type 2 (HVA-2) and bovine herpesvirus type 4 (BHV-4) have previously been shown to be closely similar. Moreover, preliminary serological data indicated that HVA-2 is antigenically related to BHV-4. To extend this study, structural components of four BHV-4 strains and HVA-2 were compared by SDS-PAGE, radioimmunoprecipitation and Western blotting. The overall pattern of structural proteins was the same for HVA-2 and BHV-4 but variations were observed in electrophoretic profiles of glycoproteins, mainly of the two major ones (gp6/gp10/gp17 and gp11/VP24). Variations between HVA-2 and BHV-4 glycoproteins were greater than those observed among BHV-4 strains.

Capsid protein VP4 of poliovirus is acylated with myristic acid via an amide linkage to its N-terminal glycine residue. Our previous studies suggested that myristic acid plays a role in poliovirus assembly and in the early events of infection. In order to understand better its role in the assembly process, we introduced a Gly1 to Ala amino acid substitution in the myristoylation signal sequence of VP4. This substitution prevented VP0 myristoylation in vivo and abolished the infectivity of genomic transcripts harbouring the mutation. These mutated RNAs were still able to replicate in the transfected cells but the assembly processes were inefficient and no mature virions could be detected.

In a study of the cleavage specificity of poliovirus proteinase 3Cpro, two mutant polioviruses were constructed to include putative 3Cpro cleavage sites in the BC loop of VP1. The BC loop of VP1 in the wild-type virus is the neutralization antigenic site IA, consisting of a continuous chain of nine amino acids (ASTTNKDKL). The first mutant, W1-1D-BC1, has four altered amino acids in the BC loop (ASTQGPGKL); the second mutant, W1-1D-BC2, has an insertion of nine amino acids in the BC loop (ASTGTAKVOGPGNKDKL). W1-1D-BC1 and W1-1D-BC2 were viable, grew to high titre and produced plaques of normal size. W1-1D-BC1 virions were resistant to proteolytic cleavage of the BC loop in vivo as well as upon incubation with a large excess of 3Cpro in vitro, although a synthetic decapeptide (PASTQGPGKL) containing the amino acids of the BC loop in W1-1D-BC1 was cleaved by 3Cpro. In contrast, W1-1D-BC2 yielded virus the VP1 of which was cleaved partially in vivo and completely when incubated with 3Cpro in vitro. Our results showed that an insertion of nine amino acids into the antigenic loop of poliovirus, representing a synthetic 3Cpro cleavage site, renders the loop susceptible to cleavage by proteinase 3Cpro, but that this cleavage is restricted if the loop is the length of that in the native virion. This result implies that, in this case, structural restrictions override sequence determinants for cleavage of the BC loop by 3Cpro.