- Volume 70, Issue 2, 1989

The evolutionary relationships of epidemic influenza A (H1N1) viruses isolated between 1982 and 1986 have been examined by oligonucleotide mapping and partial DNA sequencing. The T1 mapping studies confirmed our previous report that the evolution of the influenza virus genome generally results in an average of four to six oligonucleotide changes per year. Between 1982 and 1986, however, two apparent exceptions to this finding occurred. H1N1 antigenic variants (including the A/Chile/83 and A/Victoria/83 reference strains) that caused influenza outbreaks and epidemics from 1983 to 1984 differed by 20 to 30 oligonucleotides from viruses isolated during the previous influenza season. T1 mapping of individual RNA segments and sequencing revealed that all six internal genes of a representative 1983 A/Chile-like virus were more closely related to genes of non-reassortant H1N1 viruses that circulated from 1977 to 1982 than to genes of H3N2 viruses. Therefore, the 1983 variant viruses were not H1N1-H3N2 reassortants. The A/Taiwan/86-like H1N1 antigenic variants that emerged in south-east Asia in the spring of 1986 and caused epidemic activity the following winter also exhibited changes of 20 to 30 oligonucleotides from the A/Chile/83-like or A/Victoria/83-like H1N1 viruses that circulated during the previous influenza season. Fewer oligonucleotide changes were observed between the 1986 A/Taiwan/86-like and H1N1 viruses isolated before 1983, however, suggesting that the former evolved from viruses that circulated before the 1983 antigenic variants became the predominant H1N1 epidemic virus strains. This was confirmed by sequencing the HA1 domain of the haemagglutinin genes of three A/Taiwan/86-like viruses. These studies provide evidence that other genes of influenza A viruses, in addition to the haemagglutinin gene, may evolve concurrently along two or more separate pathways.

Mice were immunized with eight different serotypes of avian paramyxovirus (PMV). The number of splenocytes that formed antibody to PMV-infected Madin–Darby bovine kidney cells was determined by immunoperoxidase staining. Microscopic examination revealed brown plaques of cell surface or inclusion body-like antigen comprising the glycoprotein or nucleoprotein-polymerase complex, respectively. The IgG response to virus glycoprotein was priming-dependent and 99% serotype-specific, a value which exceeds that of PMV typing using avian serology. The response to inclusion body-like antigens was more variable and less priming-dependent and suggested cross-reactivities between PMV-1 and PMV-3 or -9. The primary IgM response also included a 1 to 10% non-specific antibody response to host antigens revealed by neuraminidase treatment.

Groups of mice were immunized intraperitoneally with HEp-2 lysate or respiratory syncytial virus (RSV)-infected HEp-2 cells with or without adjuvant(s). The animals were subsequently challenged intranasally (i.n.) with RSV and/or HEp-2 lysate and studied for RSV shedding in the lung, antigen-specific antibody response and pulmonary histopathology. A significant decrease in virus shedding was detected in each of four groups of animals immunized with a virus preparation and in two of four groups that received a HEp-2 lysate. Mice immunized with adjuvant(s) developed higher antibody-specific responses. All vaccinated animals developed pulmonary histopathology only on subsequent i.n. inoculation with RSV and/or HEp-2 lysate. Denatured extracts of purified RSV, HEp-2, BALB/c lung, cotton rat lung, Buffalo green monkey kidney and human buccal epithelium were tested for reactivity against the sera of immunized mice by an immunoblot method. Sera from all groups of immunized mice reacted with the extracts tested. The data suggest a possible role of reactivity to viral as well as non-viral components in the pathogenesis of RSV vaccine-induced pulmonary inflammation in the mouse model system.

The replication of flaviviruses results in the secretion of four virus-coded proteins into the extracellular environment. Three of these proteins, E, C and M (or pre-M), are found in purified virions. A fourth virus-specified extracellular protein which was not present in either the slowly sedimenting haemagglutinin particles or in virions is described. The relationship of this protein to the intracellular NS1 polypeptide was investigated along with its similarity to the soluble complement-fixing antigen (SCF) reported for mosquito-borne flaviviruses. The difference in the M r of NS1 and SCF is the result of additional glycosylation of SCF, mostly by the addition of fucose molecules. The synthesis of E and NS1 is sequential but their secretion is simultaneous, suggesting a role for NS1 in virion protein transport or virion release.

The haemagglutinin molecule on the bovine enteric coronavirus has been identified as a glycoprotein of 140K composed of disulphide-linked subunits of 65K. In this study, we have shown the subunits to be identical by demonstrating an unambiguous amino-terminal amino acid sequence. The unglycosylated subunit was found to have an M r of 42·5K and to undergo rapid disulphide linkage and glycosylation. Glycosylation was found to be of the asparagine-linked type and some of the oligosaccharides underwent processing to complex forms. Studies with inhibitors of glycosylation suggested that a processing of the haemagglutinin oligosaccharide takes place on the virion whilst it is in the Golgi apparatus. Each haemagglutinin subunit on the mature virion was estimated to possess six or seven carbohydrate chains of either the high-mannose or hybrid type, and three or four chains of the complex type.

Using a focus assay we have shown that the entire human papillomavirus type 16 (HPV-16) genome is capable of cooperating with an activated ras oncogene to transform secondary rat embryo fibroblast (REF) cells as indicated by focus formation of unselected cells. However, this assay failed to detect any similar activity with either the whole HPV-6 genome or with subgenomic fragments. In contrast, transformed colonies appeared when G418 (geneticin)-resistant colonies were selected after cotransfection with activated ras DNA and either the entire HPV-6 genome or subgenomic fragments containing the E6/E7 open reading frames (ORFs) of HPV-6 or HPV-16. The transformation assessment was based on the development of a ras-transformed appearance in G418-resistant colonies. The appearance of this morphology did not imply the ability of transformed cells to produce colonies in semi-solid agarose (anchorage-independent growth), and extended culture for about 10 to 20 population doublings was necessary before transfected cells exhibited anchorage-independent growth. Transformation of REF cells was not observed with the E5 ORF of HPV-16 under the control of an exogenous promoter (the long terminal repeat of Rous sarcoma virus) in cooperation with activated ras DNA. No transformation was observed using an activated myc oncogene with either HPV-6 or HPV-16 DNA.

Retinoic acid (RA)-induced differentiation of human teratocarcinoma (T2) cells results in a change from a normally non-permissive phenotype for human cytomegalovirus (HCMV) infection to cells which are fully permissive. We have used this system to analyse factors associated with differentiation which may regulate HCMV gene expression. Differentiation of T2 cells results in an increase of c-ras expression. Consequently, we have introduced ras expression vectors into T2 cells. We find that, as with RA induction, transfection of T2 cells with oncogenic human Ha-ras results in cells which are permissive for HCMV infection and gene expression. However, unlike RA which induces a cessation of cell proliferation and terminal differentiation, ras transfection only appears to result in changes associated with early events in RA-induced differentiation of T2 cells.

A series of herpes simplex virus isolates were recovered from a bone marrow transplant patient who received prolonged acyclovir therapy for indolent herpes simplex mouth and throat ulceration. Of 14 isolates received 10 were resistant to acyclovir and partially resistant to phosphonoacetic acid. Biochemical characterization revealed that resistance was due to an alteration in the virus DNA polymerase. DNA sequence analysis of the polymerase gene of a plaque-purified resistant virus isolate revealed a single nucleotide change when compared with the sequence of the gene of a plaque-purified sensitive isolate. This single base change resulted in a predicted amino acid substitution of Gly to Ser at residue number 841, a putative functional region of the polymerase.

A gene in equine herpesvirus 1 (EHV-1 ; equine abortion virus) equivalent to the gB glycoprotein gene of herpes simplex virus (HSV) has been identified by DNA hybridization and nucleotide sequencing. A 4·3 kbp EHV-1 PstI-ClaI sequence (0·40 to 0·43 map units) contained an open reading frame flanked by appropriate control elements and was capable of encoding a polypeptide of 980 amino acids. This had 50 to 60 % identity over a 617 amino acid conserved region with the gB gene products of HSV and three other alphaherpesviruses, and 20 to 30% identity with those of human cytomegalovirus and Epstein–Barr virus. Analysis of the amino acid sequence predicts a long signal peptide, hydrophobic and hydrophilic domains and N-glycosylation sites, and has identified a probable internal proteolytic cleavage site. The EHV-1 gB open reading frame appears to be overlapped at its 5′ end by 135 nucleotides of the 3′ end of an upstream open reading frame the potential translation product of which has approximately 50% identity with HSV gene ICP 18·5 and VZV gene 30 products.

Nude and normal BALB/c mice were primed by intravenous inoculation of purified, infectious foot-and-mouth disease virus (FMDV) type A24, strain Cruzeiro. Frequency estimation of antigen-specific antibody-secreting cells (ASC) and Thy 1+ T cells in the spleens of immunized mice identified that the IgM response was similar for both nude and normal mice, whereas substantial numbers of both IgG ASC and Thy 1+ cells were present in normal mice only. In contrast, nude and normal mouse sera both contained IgG although the nude mouse serum was deficient in IgG1. Antigen-specific antibody could not be induced in spleen cell cultures from C57BL/6 mice after depletion of T cells with monoclonal antibody plus complement. However, the antibody response could be reconstituted if either a source of exogenous lymphokines or T cells from primed but not unprimed mice were added. Similarly, polyclonal stimulation or unprimed T cells could restore the in vitro response and thus complemented the finding of a low frequency of helper T cells in unprimed mice. Taken together, these data identify that the induction of IgG in FMDV-immunized mice is T cell-dependent and regulated by lymphokines. Furthermore, in nude mice a site other than the spleen must be responsible for the synthesis of the observed serum IgG.

Wild-type (wt) strains of vesicular stomatitis virus (VSV) strain Indiana are poor to non-inducers of interferon (IFN) which express IFN induction-suppressing activity. At non-permissive temperatures, temperature-sensitive (ts) mutants of this virus are either like their wt parents, or they are good to excellent inducers of IFN. IFN inducibility and IFN induction-suppressing activity are mutually exclusive phenotypes in VSV-Indiana. With one exception, all Orsay ts mutants derived by A. Flamand (CNRS, Gif-sur-Yvette, France), representing the five complementation groups, were poor to non-inducers of IFN and were also capable of suppressing IFN induction by other viruses. In contrast, all Glasgow ts mutants derived by C. R. Pringle (University of Warwick, Coventry, U.K.) were excellent inducers of IFN. We demonstrate that this difference in acquisition of IFN inducibility relates primarily to the origin of the mutations; spontaneous for Orsay, and mutagen-derived for Glasgow. Tests with newly generated spontaneous and mutagen-derived mutants, and temperature-stable revertants of IFN-inducing ts mutants indicate that IFN inducibility results from non-ts, multiple mutations rarely acquired spontaneously, but generated frequently upon mutagenesis with 5-fluorouracil. The capacity of VSV-Indiana to induce IFN is considered intrinsic to the virus, but is only manifested when the dominant IFN induction-suppressing phenotype is lost through mutagenesis. Thus, non-ts mutations appear to regulate the expression of the IFN induction-suppressing phenotype, and hence the IFN inducibility of VSV-Indiana.

We have constructed recombinant human adenovirus (Ad) vectors containing the glycoprotein gene of vesicular stomatitis virus (VSV). The structural gene of the VSV glycoprotein was modified by the addition of promoter and poly(A) addition sequences from the herpes simplex virus type 1 thymidine kinase (TK) gene and inserted, in either orientation, into early region 3 (E3) of human Ad type 5. The recombinant vectors were fully infectious and replicated in HeLa cells in culture. The TK promoter was functional in both insert orientations and responsive to trans-activation by herpesvirus infection; however production of VSV glycoprotein in readily detectable amounts was only obtained with the vector having an insert in the E3 parallel orientation (AdG 12), and depended principally on transcripts initiating within upstream Ad sequences. The onset of expression of the glycoprotein in AdG 12-infected cells was detectable at about the same time as the Ad 72K DNA-binding protein encoded by E2, and its synthesis was not prevented by blocking viral DNA synthesis. The VSV glycoprotein produced by AdG12 was fully processed and could function to direct low pH-induced fusion of infected cells. These Ad vectors have considerable potential utility for the expression of antigens in cell culture and for the immunization of animals in studies of immunity and protection.

An infectious recombinant human adenovirus type 5 (Ad5) vector, AdG12, which carries the glycoprotein gene of vesicular stomatitis virus (VSV) and expresses that gene in cultured HeLa cells was used to examine the host range of insert expression by human Ad vectors. The VSV glycoprotein was expressed in bovine, canine and murine cells when infected with AdG 12 in culture. These cell lines are respectively permissive, non-permissive and semi-permissive for human Ad5 replication. Administration of the AdG 12 vector to calves, piglets or dogs by either the subcutaneous or oral route resulted in the production of high titres of neutralizing antibodies to VSV. Mice injected intraperitoneally with the vector produced neutralizing antibodies and were protected against subsequent intravenous challenge with normally lethal doses of VSV. This work demonstrates the utility of human adenoviral vectors for antigen expression in a number of non-human cell lines and for the induction of an immune response to the delivered antigen in a number of species.

Three distinct antigenic determinants on the nucleoprotein (NP) of measles virus were localized. These epitopes were defined by three monoclonal antibodies, one of which recognized all measles virus field strains examined, whereas the other two were variable. A measles virus NP cDNA subclone representing 502 of the 525 amino acids was cloned into a bacterial expression vector plasmid (pRIT) and expressed as a Protein A-NP fusion protein in Escherichia coli. The expressed protein reacted with all three monoclonal antibodies. A series of NP gene deletions was constructed in order to locate the antigenic sites. The antigenic site identified on all measles virus strains studied, which was designated site I, was located between amino acids 122 and 150. The two variable epitopes were located at the C terminus of the protein (site II at 457 to 476; site III at 519 to 525). The structural and biological implications of these observations are discussed.

We found a tandem repeat sequence in the region (designated BS7) in which restriction fragments BamHI D and SalI B overlap each other, near the centre of the unique long sequence (UL) of the herpes simplex virus type 1 (HSV-1) strain F genome. The SmaI physical map of BS7 was constructed, and the position of a heterogeneous SmaI subfragment from HSV-1 isolates and plaque-purified clones from a single strain was defined on the map. The maximum size difference in the SmaI subfragment was estimated to be 600 bp between these isolates and 100 bp between the clones. The 0·23 kb SmaI subfragment recloned from BS7 was sequenced, and was shown to contain a tandem repeat sequence consisting of 15 units of 12 bp, 5' TTGGGGCTGGGG 3'. These results suggest that the fragment length heterogeneity in the UL of HSV-1 isolates and clones is attributable to copy number variation of the tandem repeat sequence.

An insertion mutant of herpes simplex virus type 1 has been constructed which carries the lacZ gene from Escherichia coli within the coding sequence of gene UL2, which is in the long unique region of the genome. In a one-step growth curve experiment this recombinant (called in 1601) grew as well as the wild-type (wt) parent virus, indicating that the UL2 gene is dispensable for growth in tissue culture. Analysis of in 1601 DNA with restriction endonucleases showed no detectable changes from the wt apart from the insertion. Extracts of cells infected with in 1601 possessed levels of viral DNA polymerase and alkaline exonuclease activities similar to those infected with the wt, but unlike the wt had negligible uracil–DNA glycosylase activity, suggesting strongly that the product of the UL2 gene is the uracil–DNA glycosylase. The sequence of the uracil–DNA glycosylase gene of E. coli was recently published, and the encoded amino acid sequence of this shows clear similarity to that of UL2, confirming our results.

Seven monoclonal antibodies (MAbs) were produced which recognized equid herpesvirus type 1 (EHV-1). Three MAbs neutralized the subtype 1 virus (strain Army 183) in the presence of complement, but did not neutralize the subtype 2 virus (strain MD). All three MAbs immunoprecipitated an M r 83K glycoprotein from a detergent-solubilized virion envelope preparation of the subtype 1 virus. The target antigens of the four non-neutralizing MAbs (6F11, 2A4, 1F10 and 8D9) were identified by immunoblotting against purified EHV-1 virions and had respective apparent M r values of > 205K, > 205K, 97K and 13K.