- Volume 77, Issue 6, 1996

Avian sarcoma virus PR2257 contains 952 bp transduced from the left part of the 3′UTR of the chicken c-src oncogene. Deletion mutants were constructed to determine the effect of the 3′UTR on tumorigenicity in vivo and in vitro. In the presence of the 3′UTR, tumours were 3.4 times larger in vivo, and tumorigenicity was increased 2.5-fold in vitro. Several regulatory submotifs were also found within the 3′UTR. Parts of the 3′UTR were cloned into the LTR CAT plasmid and analysed for CAT expression. A 170 bp element was found to be responsible for the enhanced expression of the CAT gene. These results demonstrate the effect of the transduced 3′UTR sequence during long-term interaction between PR2257 virus and the chicken genome, and suggest a novel regulatory mechanism of the src oncogene.

The complete nucleotide sequence of a human T-cell lymphotropic virus type II isolate (HTLV-II-Gu) from an Italian injecting drug user was obtained, representing the first entire sequence of a European HTLV-II isolate. The HTLV-II-Gu genome was more similar to the HTLV-IIb-NRA isolate (98.4%) and HTLV-IIb-G12 (98.2%) than to HTLV-IIa-Mo (95.2%). The classification of HTLV-II-Gu as subtype IIb was confirmed by restriction analysis. Just as for HTLV-IIa strain Mo, HTLV-IIb-Gu cultured lymphocytes produce two additional mRNAs generated through alternative splicing in the pX region. A phylogenetic analysis was performed by using the methods of neighbour-joining and parsimony with bootstrapping, and maximum likelihood. The different gene regions were analysed separately, comparing Gu with all other HTLV-II strains presently available. In the LTR, as well as in other genome regions, a clear separation between IIa and IIb was evident, and within the IIb subtype three clusters were present of which two were well supported; one contained exclusively Amerindian strains and the other included all Italian and Spanish strains together with two strains obtained from New York drug users. All data clearly showed that HTLV-IIa and IIb subtypes are closely related and are equidistant from HTLV-I, suggesting that both groups evolved simultaneously. The results suggest that HTLV-II-Gu and other IIb South European isolates were probably derived from North American IIb isolates. The data also indicate that sequence analysis is necessary to further classify IIa and IIb subtypes.

It has been suggested that avian reovirus σ3 protein is analogous to σ1 trimer, the mammalian reovirus attachment protein. We have investigated the multimeric nature and cell binding ability of σ3 protein. The data presented here demonstrate that σ3 protein is a multimer in its undisrupted form as determined by SDS-PAGE in non-dissociating conditions. However, virion-associated σ3 protein and COS-7 cell-expressed protein behaved differently in SDS-PAGE, suggesting a need for virus-associated factor(s) to control the multimerization of the protein. The data also show that Escherichia coli expressed σ3 fusion protein (σ3-MBP) in its multimeric form is capable of attaching to Vero cells. The binding was found to be specific and receptor mediated by the fact that it was inhibited by a monoclonal antibody specific for σ3 protein and by competition with avian reovirus particles. As determined by a reverse experiment, σ3-MBP was also able to reduce the virus p.f.u. in monolayer cell cultures, indicating the important role of σ3 protein in the initiation of virus infection.

African horsesickness virus serotype 4 (AHSV-4) outer capsid protein VP2, or VP2 and VP5 plus inner capsid protein VP7, derived from single or dual recombinant baculovirus expression vectors were used in different combinations to immunize horses. When the proteins were purified by affinity chromatography, the combination of all three proteins induced low levels of neutralizing antibodies and conferred protection against virulent virus challenge. However, purified VP2 or VP2 and VP5 in the absence of VP7 failed to induce neutralizing antibodies and protection. Immunization with non-purified proteins enhanced the titres of neutralizing antibodies. Again, the combination of the three proteins was able to confer total protection to immunized horses, which showed absence of viraemia. The antigenicity of recombinant VP2 was analysed with a collection of 30 MAbs. Both purified and unpurified recombinant VP2 proteins showed different antigenic patterns in comparison to that of VP2 on virions. An immunization experiment with four more horses confirmed these results. The vaccine described here would not only prevent the disease, but would drastically reduce the propagation of the virus by vectors.

Two rotavirus strains, E210 and E212, implicated in epidemics of gastroenteritis in children in central and northern Australia during 1993–1994, exhibited the unusual combination of a ‘short’ RNA electrophoretic pattern and subgroup II specificity. The outer capsid protein VP7 was found by PCR typing and sequence analysis to be related to that of serotype G2 viruses. Both strains displayed a novel pattern of reactivity to G2-specific monoclonal antibodies that correlated with sequence variation in the antigenic regions of VP7. The VP4 serotype of E210 and E212 was determined as P1B in an enzyme immunoassay, consistent with other G2 viruses. Analysis of the VP6 gene indicated significant identity (98–99%) with other human subgroup II viruses. Northern hybridization analysis of E210 RNA using total genome probes derived from the prototype strains RV4 and RV5 indicated that E210 was derived from multiple gene reassortment between rotaviruses belonging to different genetic types.

Pathogenic serotype 1 strains of infectious bursal disease virus (IBDV) replicate efficiently in lymphoid cells of the bursa of Fabricius of chicken. Lymphoid cells in other organs are not susceptible. Apathogenic serotype 2 strains do not replicate in lymphoid bursa cells or in other lymphoid cells. Chicken embryo fibroblasts (CEF), however, efficiently replicate strains of either serotype. Binding studies showed that strains of both IBDV serotypes bind to lymphoid cells isolated from the bursa, thymus or spleen, indicating that restriction of IBDV replication to lymphoid B cells is not determined by the presence of specific receptor sites. The specificity of binding was demonstrated by saturation and competition experiments. These revealed the presence of different receptors: CEF had receptors common to both serotypes and specific ones for each serotype. Receptor sites common to both serotypes were also present on lymphoid cells; however, additional serotype-specific sites were only demonstrated for the apathogenic serotype 2 strain. Strains of both serotypes specificially bound to proteins with molecular masses of 40 kDa and 46 kDa, exposed on the surface of CEF and lymphoid cells. Competition experiments indicated that these proteins might represent the common receptor sites of IBDV.

Vaccinia virus (vv) recombinants expressing either wild-type (VA-F) or mutant forms (VA-FT, VA-FR47, VA-FS1 to VA-FS6) of the fusion (F) protein of respiratory syncytial (RS) virus were examined for their ability to elicit antibody, cytotoxic T lymphocytes (CTL) and protection against RS virus infection in BALB/c mice. Cells infected with the VA-F and VA-FT recombinants expressed the F protein on their surface and mice vaccinated with these recombinants developed RS virus neutralizing antibodies. The VA-FR47 recombinant expressed a mutant form of the F protein (with six amino acid changes from the wild-type) in which both proteolytic processing of the F0 precursor and its transport to the cell surface were inhibited. These mutants induced transient protection against RS virus infection although they did not induce RS virus neutralizing antibodies, or antibodies detectable by ELISA. All the vv recombinants were able to induce an RS virus-specific, MHC class I restricted CTL response. Vaccination of mice with a second set of vv recombinants expressing mutant forms of the F protein showed that the replacement Phe to Ser at amino acid 237 either alone or in combination with others abolished the neutralizing antibody response but did not affect priming of CTLs. These results demonstrate that long-term protection against RS virus infection in mice vaccinated with recombinant vv expressing the F protein is more dependent upon the induction of an antibody rather than a CTL response.

Protein G of respiratory syncytial virus (RSV) is an envelope glycoprotein that is structurally very different from its counterparts (haemagglutinin—neuraminidase and haemagglutinin) in other paramyxoviruses. In this study, we put forward a model for this unique viral envelope protein. We propose that protein G of RSV contains several independently folding regions, with the ectodomain consisting of a conserved central hydrophobic region located between two polymeric mucin-like regions. The central conserved region is probably the only relatively fixed and folded part of the ectodomain of RSV-G. This central conserved region contains four conserved cysteine residues which can form two disulphide bridges. Analysis of the proteolytic digestion products of a peptide corresponding to the central conserved region showed that one of the three theoretically possible combinations of disulphide connections could be eliminated. The final disulphide bridge assignment was established by affinity measurements with peptide variants in which different disulphide connections were formed. Additionally, peptide binding studies were used to map the binding site, at the amino acid level, of a monoclonal antibody directed against the central conserved region. These studies indicated the level of surface exposure of the amino acid side-chains. The surface exposure agreed with the structural model. The proteolytic digestion, the peptide binding studies and the affinity measurements with structural peptide variants support a structural model with disulphide connections that correspond to a structural motif called a cystine noose. This model provides a structural explanation for the location and molecular details of important antigenic sites.

Sequence analysis of a 795 nucleotide region of the fish rhabdovirus viral haemorrhagic septicaemia virus (VHSV) genome revealed one complete and one partial ORF of 369 and 153 nucleotides, respectively. The latter ORF probably encodes the amino-terminal part of the L (polymerase) protein. The former ORF potentially encodes a 122 amino acid protein. The location of this ORF as well as the size and deduced structure of the translation product indicate that it represents a homologue of the non-virion (NV) protein of the related infectious haematopoietic necrosis virus (IHNV). Antisera raised against prokaryotically expressed NV protein of VHSV and IHNV were used to detect NV expression in VHSV- and IHNV-infected cells by Western Blot and immunofluorescence analyses. We present here the sequence of the VHSV NV gene and demonstrate the presence of IHNV and VHSV NV proteins in virus-infected cells.

The subgenomic mRNAs (sg mRNA) of porcine reproductive and respiratory syndrome virus (PRRSV) were characterized. The number of sg mRNAs, which form a 3′-coterminal nested set in PRRSV-infected cells, varied from six to seven among PRRSV isolates with differing virulence. The additional species of sg mRNA in some isolates of PRRSV was designated as sg mRNA 3-1. The leader-mRNA junction sequences of sg mRNAs 3, 3-1 and 4 were found to contain a similar six nucleotide sequence motif, U(G)UA(G/C)ACC. By comparing the 5′-terminal sequence of sg mRNA 3-1 with the genomic sequence of two isolates, ISU79 and ISU1894, it was found that a point mutation, from U in isolate ISU1894 to C in isolate ISU79, led to the acquisition of a new leader-mRNA junction sequence (UUGACC) in isolate ISU79, and therefore an additional species of sg mRNA 3-1. A small ORF (3-1) was identified at the 5′ end of sg mRNA 3-1.

The nucleotide sequence of a 3266 bp region encompassing open reading frames (ORFs) 2 through 7 of the porcine reproductive and respiratory syndrome virus (PRRSV) was determined for 10 isolates recovered from the midwestern United States. Pairwise comparisons showed that genetic distances between isolates ranged from 2.5% to 7.9% (mean 5.8% ± 0.2%) whereas the Lelystad strain from Europe was, on average, 34.8% divergent from US clones. Thus, US and European PRRSV isolates represent genetically distinct clusters of the same virus. ORF 5, which encodes the envelope glycoprotein, was the most polymorphic [total nucleotide diversity (π) = 0.097 ± 0.007] and ORF 6, encoding the viral M protein, was the most conserved (π = 0.038 ± 0.003). The substantial differences in nucleotide diversity among ORFs suggests that the virus is evolving by processes other than simple accumulation of random neutral mutations. In support of this hypothesis, statistical analyses of the nucleotide sequence provided strong evidence for intragenic recombination or gene conversion in ORFs 2, 3, 4, 5 and 7, but not in ORF 6. An excess of synonymous (silent) substitutions was observed in all six ORFs, indicating an evolutionary pressure to conserve amino acid sequences. Taken together, the data indicate that despite intragenic recombination among extant PRRSV isolates, purifying selection has acted to maintain the primary structure of individual ORFs.

Serial passage of yellow fever virus (YF17D) in mouse brain enhances neurovirulence, causing a reduction in survival time after intracerebral inoculation of adult mice. To study the biological and genetic basis for this phenomenon, we compared neurovirulence properties of the neuroadapted Porterfield strain (PYF) to a YF17D strain generated from a full-length YF cDNA template (YF5.2iv). Adult mice were infected by olfactory bulb inoculation, which results in widespread distribution of virus throughout the central nervous system. Although PYF and YF5.2iv spread rapidly throughout the neuraxis, maximal titres of PYF in the brain and spinal cord were 1000- to 10000-fold higher than those of YF5.2iv. Paralysis and death occurred earlier with the PYF strain. Several cDNA clones of the E/NS1 region of the PYF strain were sequenced. Three predicted amino acid changes were consistently observed in the envelope protein of the PYF strain compared to YF5.2iv. Common substitutions were also identified in NS1 and NS2A. The potential contribution of these genetic differences to neurovirulence was evaluated by generating recombinant, intertypic PYF/YF5.2iv viruses. Physical signs of disease and mean spinal cord titres after inoculation of one recombinant were not different from the YF5.2iv parent. Our data indicate that PYF and YF5.2iv differ significantly in their virulence properties, however, common amino acid substitutions in the E/NS1 region of the PYF strain do not determine its enhanced neurovirulence. Other regions of the viral genome may contribute dominant effects on the virulence properties of the PYF strain.

The lack of an effective animal model has been a major obstacle in attempts to define the role of humoral and cellular immune responses in protection against flavivirus infection. We have used F1 hybrid mice (BALB/c × C3H/RV) that are heterozygous for the flavivirus resistance allele Flvr and show reduced virus replication in the brain after intracerebral inoculation. F1 hybrid mice challenged by intracerebral inoculation with Murray Valley encephalitis (MVE) virus developed encephalitis 2–3 days later than a genetically susceptible strain (BALB/c) but showed a similar mortality rate. This delay in the onset of disease provided more opportunity for virus clearance by primed immune responses. Using F1 hybrid mice we were able to demonstrate protective immunity induced by structural and non-structural proteins of MVE virus by immunization with pure NS1 protein or recombinant vaccinia viruses that expressed various regions of the MVE genome. These constructs included VV-STR (C-prM-E-NS1-NS2A), VV-ΔC (prM-E-NS1-NS2A) and VV-NS1 (NS1-NS2A). VV-ΔC vaccinated mice were completely protected (100% survival) from challenge with 1000 infectious units of MVE virus, while mice inoculated with VV-STR, VV-NS1 or pure NS1 were partially protected (40%, 47% and 85% respectively). Analysis of prechallenge sera and in vivo depletion studies revealed that the solid protection induced by VV-ΔC was mediated by neutralizing antibody to the E protein and did not require a CD8+ T cell response. The partial protection provided by VV-STR, VV-NS1 and pure NS1 occurred after induction of antibody to NS1. However, depletion of CD8+ cells prior to virus challenge ablated the protection provided by VV-NS1 indicating some requirement for class I restricted cytotoxic T cells.

Pestiviruses initiate infection of susceptible cells by receptor-mediated endocytosis. Cellular plasma membrane or endosomal molecules involved in translocation of these viruses into the cytosol have not been unequivocally identified. We reported previously that a mutant cell line derived from Madin-Darby bovine kidney (MDBK) cells, termed CRIB-1, was resistant to infection with bovine viral diarrhoea virus. CRIB-1 cells were also resistant to infection with classical swine fever virus and border disease virus of sheep, suggesting that entry of these three different pestiviruses into bovine cells requires a common cell membrane function. The resistance is pestivirus-specific: CRIB-1 cells were as susceptible as the parental MDBK cells to 14 other viruses of cattle and swine belonging to unrelated families. The resistance of CRIB-1 cells to pestivirus infection involves a block in virus entry since transfection of virus RNA or virus inoculation in the presence of PEG resulted in productive infection. Furthermore, quantitative analyses of the outcome of PEG-mediated infection of CRIB-1 cells indicated that the intracellular milieu was fully permissive for pestivirus replication. Binding studies revealed that virus attachment to CRIB-1 cells was not completely abrogated. These results indicate that entry of pestiviruses into MDBK cells depends on a common plasma membrane or endosomal function, which is lacking in CRIB-1 cells.

Cytotoxic T lymphocytes (CTLs) may play a role in host defence against hepatitis C virus (HCV) infection, and HCV-specific CTL epitopes may be included in vaccines to induce protective CTLs. We identified three new epitopes within the HCV nucleoprotein recognized by CTLs. HCV nucleoprotein residues 28–37 are the minimal epitope recognized by CTLs in association with the class I human leukocyte antigen B60, and epitopes in HCV nucleoprotein residues 111–130 and 161–180 are both recognized by CTLs in association with the class II human leukocyte antigen DRB1*08032.

By analysing the nucleotide sequence data generated from both the E2 (gp55) and the NS5B genes of classical swine fever virus (CSFV), in addition to previously published data from the 5′NCR, we were able to divide 115 CSFV isolates into two major groups, five subgroups and two disparate isolates. Further discrimination was possible by analysis of sequence data from the E2 region. The three sequencing based methods were compared to monoclonal antibody (MAb) typing and to limited restriction enzyme (RE) mapping. Although both MAb and RE methods confirmed the previous classification the resolution was inferior. We estimated an approximate evolution rate for CSFV from an analysis of the virus variation observed in a single geographical area over a 6 year period. Applying this proposed rate to each of our deduced CSFV subgroups enabled us to calculate the approximate dates of divergence for each subgroup.

We have recently shown that the Semliki Forest virus (SFV) subgenome contains a translation enhancer region in the coding part of its RNA. The enhancer increases the translation of the viral structural genes approximately ten-fold and thereby plays a key role in virus assembly. In this study we conclusively show that this translation enhancer represents an adaptation to the

Expression from the human parvovirus B19p6 promoter fused to the firefly luciferase (‘Luc’) reporter gene was evaluated in a non-erythroid human nasopharyngeal carcinoma cell line, KB, and a human megakaryocytic leukaemia cell line, MB-02, known to become permissive for B19 replication following erythroid-differentiation. The B19p6-Luc construct was introduced into KB and MB-02 cells, both in undifferentiated and differentiated states, either via DNA-mediated transfection, or via infection with recombinant adeno-associated virus 2 (AAV), a non-pathogenic human parvovirus known to possess a broad host-range. Although Luc activity was readily detected in KB cells following transfection of the B19p6-Luc plasmid DNA, no expression from the B19p6 promoter was observed following infection with recombinant virus. In addition, transfection of the reporter plasmid resulted in high-level expression of Luc in differentiated but not in undifferentiated MB-02 cells. However, no Luc activity was detected, even in differentiated MB-02 cells, following infection with recombinant virus. Further studies with an additional recombinant as well as wild-type (wt) AAV revealed that MB-02 cells were non-permissive for AAV infection. A second human megakaryocytic leukaemia cell line, M07e, was likewise resistant to infection by recombinant as well as wt AAV. Taken together, these studies identify the first human cell type that cannot be infected by AAV. They indicate that expression from the B19p6 promoter, in the context of an AAV genome, is restricted to primary human haematopoietic cells, perhaps because parvoviral DNA replication and transcription are intrinsically coupled.

Hepatitis B virus (HBV) wild-type pre-C RNA directs the synthesis of the HBeAg precursor but does not serve as mRNA for translation of the adjacent downstream C gene which encodes the core protein. Using bicistronic mRNA constructs that mimick pre-C RNA, we have demonstrated that this RNA likewise does not serve as messenger for translation of the P gene, which is located downstream of the C gene. However, when the pre-C RNA contains a translational stop codon at position 2 or 28 of the pre-C sequence (as in certain HBV mutants), it no longer directs synthesis of the HBeAg precursor but instead translation is initiated at downstream C and P gene AUGs. We propose that this occurs by a translation reinitiation mechanism.