Search

Here we report on the conformational changes that are responsible for the appearance of the Dicentrarchus labrax encephalitis virus (DlEV) coat protein as a doublet in SDS-PAGE. Wild-type and mutated forms of the coat protein cDNA were expressed in E. coli. The study of the resulting recombinant molecules excluded the possibility of the involvement of a precursor autocatalysis mechanism or a ribosomal frameshifting event in the doublet formation. The appearance of the coat protein doublet was found to be beta-mercaptoethanol sensitive. Based on this observation, we carried out substitution of all cysteine residues. The obtained results demonstrated the importance of intramolecular disulfide bonding between cysteines 187 and 201 on coat protein conformational changes.

Two pig endogenous retroviruses (PERV), PERV-A and -B, productively infect human cells and are therefore considered to constitute a potential risk in pig-to-human xenotransplantation. A PCR-based cloning technique to isolate infectious PERV proviruses was established. Overlapping 3' half and 5' halves of PERV proviral genomes were amplified using DNA extracted from human 293 cells infected with PERV-A or -B. These clones were fused at a unique restriction site in the overlapping region and tested for their infectivity. Representative constructs possessed the same infectious properties as their parent isolates. We also developed a polyclonal anti-PERV serum by using recombinant PERV capsid protein derived from one of the infectious constructs as immunogen and established an immunocytological method for detection and titration of PERV infection. This detection method proved to be more sensitive than the current method of choice (transfer of MLV-lacZ vectors) for infectivity assessment of PERV. These findings should be considered for future characterization of PERV isolates.

cDNA expression vectors of Potato leafroll virus (PLRV) were used to analyse specific mutations in the proteinase and replicase domains of the proteins encoded by ORF1 and ORF2. Agrobacterium-mediated DNA transfer was used to introduce a PLRV RNA expression unit, controlled by the 35S promoter of Cauliflower mosaic virus, into potato leaf cells. Expression of unmodified PLRV cDNA led to the replication of viral genomic and subgenomic RNAs and accumulation of the viral capsid protein, whereas alteration of amino acids GDD513-515 of the replicase to VHD abolished PLRV replication. Mutations in the presumed H-D-S catalytic triad of the viral proteinase abolished the formation of viral genomic and subgenomic RNAs as well as synthesis of the viral capsid protein. Co-agroinoculation of the GDD mutant along with any of the proteinase mutants restored virus replication in leaf discs, showing that these mutants are able to complement each other. Moreover, mutation of the postulated serine residue of the catalytic triad of the proteinase altered the pattern of proteins synthesized in vitro in comparison to wild-type, further supporting the relevance of the H-D-S motif.

Infectious pancreatic necrosis virus (IPNV), a member of the BIRNAVIRIDAE: with two double-stranded RNA genome segments, encodes five proteins designated VP1 to VP5. To study the function of the 17 kDa nonstructural protein VP5 during virus replication several mutated IPNV genome segments A were constructed and included in a reverse genetics system for IPNV to obtain recombinant virus. Mutations between nt 68 and 85 or nt 94 and 103 in the noncoding region failed to yield viable virus. Only mutations located between nt 86 and 92 and downstream of nt 104 were tolerated, and viable virus could be generated. All IPNV generated showed no difference in replication compared with the wild-type IPNV, indicating that the absence of expression of VP5 did not influence virus growth in vitro. Furthermore, the results presented here indicate that initiation of translation of VP5 occurs at position 113, the second in-frame start codon.

Human papillomavirus (HPV) infections are prevalent in human immunodeficiency virus (HIV)-positive individuals. To highlight the effect of HIV on HPV expression, HPV-18-positive HIV-permissive HeLa-T4 cells were either infected with HIV-1 or treated with Tat or with the cytokines IL-1alpha, IL-1beta, IL-6 and TNF-alpha. The presence of HPV-18 E1 (early) and L1 (late) transcripts was then determined by dot-blot or Northern blot hybridization with E1 and L1 or with genomic HPV-18 DNA probes, respectively. Protein extracts from parallel cultures were challenged by Western blotting with an antiserum raised against an L1-beta-galactosidase hybrid protein. Results indicated that HeLa-T4 cells constitutively express E1 and L1 transcripts. When cells were infected with HIV, the amounts of E1 and L1 RNAs increased with time, followed by the de novo appearance of L1 protein. E1 and L1 transcripts were also increased, in a dose-dependent manner, by treatment of uninfected cultures with Tat or with IL-6, but were not affected by IL-1alpha, IL-1beta and TNF- alpha. Neither Tat nor IL-6 could induce L1 translation. These findings raise the hypothesis that the increase of HPV shedding and of HPV-associated diseases in HIV-infected individuals could be due in part to a direct or cytokine-mediated action of HIV, in addition to the HIV-induced immunodeficiency.

In this study replication of A-type and B-type Epstein-Barr virus (EBV) strains has been assessed. A-type and B-type type lymphoblastoid cell lines (LCLs) were established by infecting B lymphocytes, isolated from five EBV-seropositive donors, with different A-type and B-type virus isolates. The presence of viral capsid antigens (VCA) in these LCLs was determined by immunofluoresence assay and by immunoblotting. All of the B-type EBV strains were capable of spontaneously generating virus regardless of the origin of the donor cells. In contrast the A-type strains, other than strain IARC-BL36, did not readily produce VCA in any of the different donor lymphocytes used. This study demonstrates another biological difference between the two virus types: their ability to spontaneously enter the lytic cycle.

A synthetic peptide (peptide 10) representing a surface-exposed, linear B cell epitope from outer-membrane (OM) protein F of Pseudomonas aeruginosa was shown previously to afford protection in mice from P. aeruginosa infection. This peptide was expressed in tandem with the protein F peptide 18 on each of the two coat proteins of cowpea mosaic virus (CPMV). The chimaeric virus particles (CVPs) expressing the peptides on the S (small) coat protein (CPMV-PAE4) and L (large) coat protein (CPMV-PAE5) were used to immunize mice. Following subcutaneous immunization in Freund's and QuilA adjuvants, CPMV-PAE4 induced antibodies predominantly against peptide 18, whereas CPMV-PAE5 produced antibodies exclusively against peptide 10, indicating that the site of peptide expression on CPMV influences its immune recognition. The anti-peptide antibodies elicited by CPMV-PAE5 were predominantly of the IgG2a isotype, indicating a highly polarized TH1-type response. The peptide-specific IgG2a strongly recognized the whole F protein, but more importantly, recognized protein F in all seven Fisher-Devlin immunotypes of P. aeruginosa. Furthermore, the peptide-specific IgG2a in CVP/QS-21 adjuvant-immunized mice was shown to bind complement and to augment phagocytosis of P. aeruginosa by human neutrophils in vitro. The ability of CPMV-PAE5 to induce P. aeruginosa-specific opsonic IgG2a gives it potential for further development as a protective vaccine against P. aeruginosa.

To gain more insight into the structure of the African horsesickness virus (AHSV) core particle, we have cloned, partially characterized and expressed the two major core proteins, VP3 and VP7, of AHSV-9. VP7 was found to be highly conserved amongst different serotypes. The VP3 and VP7 genes were subsequently expressed in insect cells by means of recombinant baculoviruses. VP7 was synthesized to very high levels and aggregated into distinctive crystals. Co-expression of VP3 and VP7 resulted in the intracellular formation of core-like particles which structurally resembled empty AHSV cores.

Experiments were done to test cell lines for their capacity to express human immunodeficiency virus type 1 (HIV-1) proteins in a stable manner. Marked differences were seen in the ability to stably express and export viral Gag and Pol proteins. Two cell lines, one suspension (MDS) and one monolayer (SW480), were established which exported these proteins at high level. Two other cell lines, HeLa and THP-1, showed poorer expression and very limited particle release. Single cell cloning was used to select the optimal producing clones from the lines. These produced large quantities of viral core particles pelletable from the supernatants. Cell lines were constructed from these clones which stably expressed in addition either the HIV-1 Envelope or a packageable HIV-based vector. The vector was shown to be packaged within the viral core particles. Transient transfection of envelope expressing constructs into a gag-pol plus vector cell line, or the vector into a gag-pol plus envelope expressing cell line resulted in gene transfer to CD4+ target cells. These cell lines provide useful tools with which to study the assembly and export of viral proteins and RNA, for assay of alternative envelope proteins to pseudotype HIV cores, for assessment of antiviral drugs and as a source of correctly processed proteins for immunological studies.

The role of the 3' noncoding (NC) region of Physalis mottle tymovirus genomic RNA in the multiplication of the virus was examined using an in vivo protoplast assay system. Coat protein (CP) synthesis was specifically inhibited by sense 3' NC region transcripts. To establish the role of the pseudoknot structure present in the NC region in virus multiplication, four site-specific mutants, two of which disrupted the pseudoknot structure while the other two restored the structure, were constructed. Interestingly, none of the four sense mutant transcripts inhibited CP synthesis, suggesting that the specific sequence representing the 3' terminal pseudoknot structure may play an important role in virus multiplication. However, the wild-type antisense 3' NC transcript as well as the four antisense mutant transcripts inhibited CP synthesis, suggesting that the inhibitions by antisense transcripts could be due to the formation of RNA-RNA hybrids at the 3' end of the genomic RNA.

Transgenic potato plants were generated which express a dicistronic mRNA corresponding to the 8 kDa protein and coat protein (CP) genes of potato virus X (PVX). Northern blot analysis of RNA isolated from these plants indicated that both the 8 kDa protein and the CP are translated from this dicistronic mRNA. Expression of both of these proteins in transgenic plants was demonstrated by Western blot analysis, and suggests that translation of CP takes place either by initiation by internal entry of ribosomes or by a termination/reinitiation mechanism. CP was detected in protoplasts electroporated with RNA transcripts of the dicistronic construct in the presence or absence of a stable hairpin structure at the 5' terminus, suggesting that the former model is more likely. Similar results were obtained when a DNA fragment containing the PVX CP leader sequence was placed between two reporter genes in the transient assay system. These results suggest that potexvirus CP expression may take place from the genomic RNA as well as from the subgenomic RNA.

The 2B proteins of coxsackievirus and poliovirus (PV) share significant structural similarity and exhibit similar biochemical activities, namely inhibition of protein secretion and modification of membrane permeability. Both proteins contain two hydrophobic domains in the carboxy-terminal two-thirds of their sequence, of which one has the potential to form a cationic amphipathic alpha-helix. To gain more insight into the structural requirements of enterovirus protein 2B for its functioning in viral RNA replication, a chimeric cDNA approach was used. Chimeric coxsackie B3 virus (CBV3) genomes were constructed that expressed either the entire PV 2B protein or hybrid proteins in which specific segments of CBV3 2B were substituted by their corresponding PV counterparts. In vitro synthesis and processing of the chimeric polyproteins showed no abnormalities. CBV3 genomes carrying the entire PV 2B gene failed to replicate. A chimeric genome that expressed a hybrid 2B protein consisting of the amino-terminal one-third of PV and the remainder of CBV3 yielded viable viruses. In contrast, a 2B protein consisting of the amino-terminal one-third of CBV3 and the remainder of PV failed to drive replication. These data imply that a sequence-specific interaction with another viral protein is required to drive RNA replication and suggest that the proposed sites of contact reside in the carboxy-terminal two-thirds of 2B. Hybrid genomes in which either the amphipathic alpha-helix or the other hydrophobic domain was replaced failed to replicate. The potential contribution of these domains to the structure and functioning of protein 2B are discussed.

Polioviral genes coding for P1, the precursor for the structural proteins, and 3CD, the viral protease, were cloned in a Saccharomyces cerevisiae inducible expression system. N-antigenic empty capsids could be isolated from the yeast cell extract provided that pirodavir, a capsid-binding compound and capsid stabilizer, was added during the induction period and during purification. Purification was by immunoaffinity chromatography. The purified empty capsids had the same immunogenicity as poliovirus virions. The techniques described might be useful for the production of new non-infectious vaccines.

VP6, the major capsid protein of rotavirus, oligomerizes into trimers that constitute the intermediate shell of the virions. In order to map functional domains in this protein, we introduced seven internal in-frame deletions within the coding region of gene 6 of human rotavirus strain Wa. Regions of homology among the VP6 proteins of group A and group C rotaviruses were targeted for deletion mutagenesis. The mutant VP6 proteins were expressed in mammalian cells using the recombinant vaccinia virus system and were examined for their ability to oligomerize into trimers as well as to assemble into double-layered virus-like particles upon coexpression with the rotavirus core protein VP2. Deletions that abolished trimerization defined a domain (residues 246 to 314) that maps within a larger region previously found to be critical for oligomerization (amino acids 105 to 328). When the capacity of each mutant to assemble into double-layered virus-like particles was analysed, three different assembly phenotypes were observed. Phenotype I was represented by two deletion mutants lacking residues 246 to 250 and 308 to 314 that produced particles with efficiencies similar to that of wild-type VP6. Phenotype II, characterized by a moderate decrease in the efficiency of particle assembly with respect to that of wild-type VP6, included two mutants with deletions at the C terminus of the protein. Phenotype III was exhibited by three mutants whose abilities to assemble into double-layered virus-like particles were drastically impaired. Two of these mutants define a previously unidentified assembly domain (amino acids 122 to 147) at the N terminus of rotavirus VP6.

Sequence variations are seen in the JC virus promoter/enhancer in virus taken from progressive multifocal leukoencephalopathy (PML) brains and it has been hypothesized that the variations arise in the host at some point in the development of PML. These rearrangements may be adaptations for enhanced growth in glial cells; if so, transcription or replication levels should differ between archetypal and rearranged PML-type promoters. The archetype and four PML-type promoters were analysed in human glial cells for early and late transcriptional activity in the absence or presence of virus T antigen, and for DNA replication. CAT reporter expression differed within a fivefold range and the archetype was intermediate in strength to the PML-type regulatory regions. The archetype differed from rearranged promoters in that the late promoter was less responsive to T antigen and the shift from early to late activity with T antigen was less pronounced. All five regulatory regions demonstrated similar levels of DNA replicating activity. Rearrangement of the archetype was not required for activity in glial cells, but the potential for differences in the regulation of the late capsid genes was found.

Potato leafroll luteovirus (PLRV) movement through phloem of PLRV-resistant potato clones was examined in experiments in which stem pieces were grafted either between infected rootstocks and virus-free susceptible scions or between infected scions and virus-free susceptible rootstocks. These test plants permitted either upwards or downwards virus movement into the susceptible tissue. Resistant potato clones had either host gene-mediated resistance (H-MR) or transgene-mediated resistance (T-NR, conferred by transformation with the PLRV coat protein gene) to PLRV accumulation. The rate of PLRV movement was similar whether stem tissue was taken from H-MR, T-MR or susceptible potato clones. Virus movement through two graft unions began around 7 days after grafting and was generally complete by about 14 to 16 days. Virus movement occurred soon after acquiring functional phloem continuity across grafts as demonstrated by tracing with 6(5)-carboxyfluorescein, a phloem-mobile dye. Most of the delay in virus detection after grafting probably resulted from the time necessary to develop new phloem strands across graft unions; subsequent movement of PLRV was rapid suggesting a passive process. PLRV infection was largely excluded from external phloem bundles in stem tissue of clones with either H-MR or T-MR. This trait was less pronounced as tissue aged. The mechanism limiting PLRV invasion of external phloem bundles of the T-MR clones appears to be similar to that operating in the H-MR clones. Results are discussed in the context of a proposed model of PLRV movement.

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.

The role of T cells in immunity to foot-and-mouth disease virus is still poorly defined compared to that of the humoral response. In this paper we describe a systematic, longitudinal study on the cellular recognition of FMDV and its subunit protein VP1 by bovine peripheral blood T lymphocytes. Multiple vaccination with a single virus serotype induced a serotype cross-reactive proliferative T cell repertoire that varied in magnitude between individual animals and with the serotype of the vaccine used. Primary proliferative T cell responses of vaccinated and acutely infected cattle were weak relative to the magnitude of responses determined for the same animals after boosting. In contrast, the level of circulating antibody produced after both primary and secondary exposure to virus was good. Phenotypic analysis of lymphocytes from vaccinated or infected cattle showed a small increase in CD8+ T cells after infection compared to vaccination. However, in general the profiles of circulating lymphocytes elicited were similar. Thus, we were not able to use proliferative or phenotypic analyses to distinguish between vaccinated and convalescent cattle. T cell recognition of VP1 by multiply-vaccinated cattle was serotype-specific implying that the cross-reactive responses observed with whole virus may be attributed to proteins other than VP1. In contrast to other studies, immunization with recombinant VP1 induced only low levels of neutralizing antibody and failed to elicit profound proliferative responses or protection ever after two immunizations.

cDNA cassettes encoding the foot-and-mouth disease virus (FMDV) structural protein precursor (P1-2A) together with the 3C protease, which cleave this molecule to 1AB, 1C and 1D, were constructed. These cassettes were introduced into vaccinia virus (VV) transfer vectors. Attempts to isolate recombinant VVs constitutively expressing these cassettes were unsuccessful. However, when the P1-2A-3C cassette was placed under the control of the bacteriophage T7 promoter, stable VV/FMDV recombinants were isolated. Co-infection with recombinant VV vTF7-3 (which expresses T7 RNA polymerase) led to the production of correctly processed FMDV capsid proteins. Analysis by sucrose gradient centrifugation showed that material which co-sedimented with natural empty capsid particles (70S) was formed. Electron microscopy revealed empty capsid-like particles with diameters of about 30 nm. Studies using monoclonal antibodies specific for conformational epitopes indicated that the antigenicity of the synthetic particles was similar to whole virions and natural empty capsid particles. Surprisingly, merely the modification of a single amino acid residue within the myristoylation consensus sequence at the N terminus of P1-2A allowed the isolation of a recombinant VV which constitutively expressed the correctly processed proteins. However, the capsid proteins expressed from this mutant cassette failed to assemble into 70S empty particles.

A full-length copy of a single genomic component of the whitefly-transmitted geminivirus ageratum yellow vein virus (AYVV) has been cloned from an extract of infected Ageratum conyzoides originating from Singapore. Sequence analysis shows that the genomic component encodes two virion-sense (V1 and V2) and four complementary-sense open reading frames (C1-C4), typical of DNA A of whitefly-transmitted geminiviruses from the Eastern hemisphere. A genomic component equivalent to DNA B was not detected in extracts of infected A. conyzoides. The cloned genomic component produced a systemic infection in Nicotiana benthamiana, Phaseolus vulgaris and Lycopersicon esculentum when introduced into plants by agroinoculation, and symptoms were identical to those produced by wild-type virus introduced into these hosts using viruliferous whiteflies. However, attempts to re-establish a systemic infection in A. conyzoides either by agroinoculation or by whitefly transmission of the cloned progeny were unsuccessful, suggesting that additional factors are required for infection of the natural host. The significance of A. conyzoides as a reservoir host for the economically important geminivirus diseases is discussed.

The ability of rinderpest virus (RPV) to replicate in vitro in adherent peripheral blood monocytes and monocyte-derived macrophages under non-stimulation conditions was investigated. When flow cytometry analysis on bovine peripheral blood mononuclear cells (PBMC) was performed, monocytic cells were seen to be targets for infection by the cell culture-attenuated RBOK vaccine strain of RPV. Viral glycoprotein (H) and nucleoprotein (N) expression in adherent blood monocytes and monocyte-derived macrophages was compared with the infection in Vero cells, in which a productive infection typical of morbilliviruses is obtained. In both cell types, the infection was m.o.i.-dependent, but the rate of viral protein accumulation was slower in monocytes/macrophages. Double-labelling experiments with monoclonal antibodies against RPV and the myeloid marker CD14 confirmed that the infected blood adherent cells were monocytes and macrophages. Productive infection of monocytes was confirmed by progeny virus titration. Permissiveness to infection was not dependent on macrophage differentiation: in vitro maturation of monocytes to macrophages before infection, did not increase the susceptibility of these cells to RPV infection. With the virulent Saudi RPV isolate, similar results were obtained, although the Saudi virus apparently had a higher rate of replication compared to the attenuated virus. These observations demonstrate clearly that bovine blood monocytes and monocyte-derived macrophages serve as hosts for a relatively slow but productive infection by rinderpest virus.

Development in mammalian cells of a recombinant expression system that mimics the rotavirus capsid assembly process would be advantageous for studying the structural requirements for particle formation. To this end, we investigated the ability of a recombinant vaccinia virus system to produce double-layered virus-like particles. The genes coding for VP2 and VP6 proteins of the human rotavirus strain Wa were cloned and used to generate recombinant vaccinia viruses. Metabolic labelling of CV-1 cells infected with these recombinant viruses followed by immunoprecipitation with a polyclonal antiserum directed to Wa virus showed that VP2 and VP6 were efficiently expressed. The recombinant proteins were similar in size and immunoreactivity to authentic rotavirus proteins. Biochemical and electron microscopy analyses demonstrated that simultaneous expression of VP2 and VP6 in mammalian cells resulted in the formation of intracellular spherical particles resembling double-layered rotavirus particles.

Human adenovirus (Ad) vectors are being used increasingly for a variety of applications in vaccination and gene therapy. The ability of vectors to enter cells and the efficiency of promoters expressing the therapeutic gene or vaccine antigen are critical to the outcome of such experiments. To identify promoters which might be suitable for use under a variety of conditions we have investigated the expression of a rotavirus antigen, VP7sc, employing several commonly used promoters carried in E1-substituted Ad vectors both in cell types which support virus replication and in cells which do not. Although not all gene constructions were identical, wide variations in promoter function were evident even in human 293 cells which support virus replication. The simian virus type 40 (SV40) early and beta-actin promoters expressed poorly; the SV40 late promoter was somewhat better. The human IE94 cytomegalovirus (CMV) promoter and a modified Ad major late promoter were best, functioning equally well but with different kinetics. In other human cell lines the CMV promoter was more versatile, generally providing sustained expression at a significant level, in one case for at least 6 days. In addition, as mouse, rabbit and pig models of rotavirus infection are under investigation and VP7sc is a vaccine antigen, we also investigated the ability of the recombinant adenovirus to infect cells from these and other sources. VP7sc expression was detected in several heterologous cell types, illustrating the ubiquity of the human Ad receptor and the versatility of human Ad as vectors when suitable promoters are used.

Antisera raised against oriented peptide conjugates were used to identify and partially characterize a 24 kDa protein product expressed by chicken anaemia virus (CAV). The peptides derived from the N and C termini of the protein were shown to react against the native protein, expressed within virus-infected cells, by immunofluorescence, immunoperoxidase and immunogold thin section electron microscopy techniques. The protein product was located by immunogold single labelling in intranuclear inclusions similar to those described previously for the 13 kDa CAV protein, which causes apoptosis. The 24 kDa protein was co-localized to the nuclear inclusions with the CAV 13 kDa protein by simultaneous dual labelling immunogold electron microscopy. Following isolation of the CAV proteins by nuclei isolation and SDS-PAGE, the antisera were used to probe for the protein by immunoblotting. The antisera recognized an expressed protein product of apparent molecular mass 30 kDa. An immunofluorescence time course study of CAV protein expression was carried out and the peptide antisera reacted against the protein at 12 h post-infection. Antisera against the 13 kDa protein reacted at similar times post-infection. This was in contrast to antisera raised against the 52 kDa capsid protein which is detectable by immunofluorescence only after 24 h. The 13 kDa and 24 kDa proteins thus appear to be early antigens produced by CAV during infection.

We have identified viruses in several isolates of the filamentous ascomycete Atkinsonella hypoxylon. The virus from one isolate of the fungus, 2H, was selected for genomic characterization. Purified virus particles contained three dsRNAs with sizes estimated by gel electrophoresis to be 2.2, 2.1 and 1.8 kb. A library of cDNA clones representing the three dsRNA segments of isolate 2H was synthesized, mapped and sequenced. The three segments had no significant similarity to each other, as determined by Northern blot analysis, and had sizes of 2180, 2135 and 1790 nt as determined by nucleotide sequence analysis. Long open reading frames were deduced from the sequences of dsRNAs 1 (molecular mass 78 kDa) and 2 (74 kDa), but not from dsRNA 3. Both terminal regions of dsRNA 1 and dsRNA had similar nucleotide sequences, as determined from 5' RACE clones. Comparisons of the amino acid sequence deduced from dsRNA 1 revealed similarities with viral RNA-dependent RNA polymerases. Translation in vitro of full-length cDNA clones representing dsRNAs 1 and 2 each yielded single major products of > 70 kDa by analysis on polyacrylamide gels. Based on properties of its dsRNA segments, the virus of A. hypoxylon strain 2H fits into the Partitiviridae family, and represents the first member of this family from a fungal host completely characterized at the level of primary nucleotide sequence.

Serial passage of nuclear polyhedrosis viruses (NPVs) through cultured cell lines results in the appearance of mutants with a complex phenotype referred to as the 'few polyhedra' (FP) phenotype. The altered plaque morphology and reduced occlusion production associated with the FP phenotype have been observed in Autographa californica multinucleocapsid nuclear polyhedrosis virus (AcMNPV) bearing mutations in the gene encoding the 25 kDa protein (25K gene). In this study, we sequenced the 25K genes of four spontaneously occurring AcMNPV FP mutants. These mutants, together with an artificially generated FP mutant (AcFP beta gal, in which the gene for beta-galactosidase is fused in frame with the 25K ORF), were examined at the ultrastructural level to see if they exhibited the reduced virion occlusion and intranuclear envelopment which is associated with the FP phenotype. Observations on Spodoptera frugiperda Sf9 cells infected with the FP mutants revealed that all five mutants were impaired in virion occlusion and intranuclear nucleocapsid envelopment. The 25K mutants were also found to release two- to fivefold more infectious virus (p.f.u.) into the media of infected Sf9 cells. Marker rescue of AcFP beta gal restored wild-type virion occlusion, intranuclear envelopment and levels of infectious virus production.

Plants infected by citrus tristeza virus (CTV) contain, in addition to the 2000 nm full-length thread-like virions, a heterogeneous population of smaller particles. The CTV particles and RNA extracts from purified CTV preparations were fractionated by sucrose gradient centrifugation and the RNA molecules from different fractions translated in a reticulocyte translation system. Fractions containing predominantly an RNA band of approximately 3.2 kb directed the synthesis of CTV coat protein (CP), which in SDS-PAGE had an estimated molecular mass of 28 kDa. Three additional polypeptides, with estimated sizes of 21 kDa, 23 kDa and 27 kDa, were translated from a range of RNA molecules smaller than 3.2 kb. Hybridization with cDNA to the CP gene (CTV-CPG) and with a 350 base clone complementary to the 3' and 5' termini of the genes for CTV p20 and p23.5, respectively, indicated that preparations of CTV particles contain, in addition to the genomic (20 kb) RNA, two sub-genomic RNA molecules of 3.2 kb and 2.4 kb and probably also two smaller molecules of 1.6 kb and 0.9 kb. Only the 3.2 kb RNA, its corresponding dsRNA molecule and populations of larger RNAs, including the 20 kb genomic RNA, hybridized with a CTV-CPG probe, thus conflicting with our previous assignment of the CTV-CPG to the 0.8 kbp dsRNA. Based on these results we propose that distinct populations of CTV particles encapsidate smaller RNAs which were formed as a nested set of subgenomic RNAs. Sequence analysis of 2540 nucleotides downstream to CTV-CPG of strain VT revealed four open reading frames (ORFs) potentially encoding in the 5' to 3' direction, 18 kDa (p18), 13 kDa (p13), 20 kDa (p20) and 23.5 kDa (p23.5) proteins. The CTV-VT ORFs showed variable but usually close levels of homology with the corresponding ORFs of CTV-T36 from Florida.

Using site-directed mutagenesis, we tested whether a potential integrin-binding site, (composed of the amino acids RGD) which is predicted in the adeno-associated virus 2 (AAV-2) capsid open reading frame (ORF), plays a role in the infectivity of AAV-2. Nucleotide sequencing of wild-type and mutant capsid protein-coding sequences, however, revealed discrepancies with the published sequence data at several positions, including a frameshift in the carboxy terminus which cancels the RGD motif and extends the capsid ORF by 27 amino acids. This sequence was confirmed by protein sequencing of proteolytic fragments of VP3. Thus, the virus mutant (pTAV-p), in which the intention was to exchange D of the putative RGD motif for E, resulted in replacing I480 by S in the newly established ORF. A second virus mutant (pTAV-d), in which the intention was to delete the RGD peptide, in fact gave a shift into the ORF of the originally published sequence. The pTAV-p mutant showed a strongly reduced infectivity compared to wild-type AAV-2, whereas pTAV-d was not infectious at all. Neither mutant accumulated viral ssDNA as detected by Hirt extraction. Analysis of virus particle formation and subcellular localization of the capsid proteins revealed a defect of the mutant capsid proteins in capsid assembly. This shows that the newly established C-terminal sequence of the AAV capsid proteins plays an important role in viral assembly.

To identify Mengo virus-specific T cell epitopes in mice (the natural host for the virus), lymph node cells were obtained from BALB/c (H-2d) mice, previously immunized with u.v.-inactivated virus, and stimulated in vitro with each of 116 overlapping peptides (10 to 18 residues long) covering the entire capsid coding region (834 amino acids). T cell epitopes were defined on the basis of specific peptide-induced lymphocyte proliferation. Where proliferation occurred, immunological characterization showed that it was the CD4+ T helper (Th) cell subpopulation that was responsible for the Mengo virus-specific response. Surprisingly, no Mengo virus Th cell epitopes were found in capsid protein VP1 or VP4. Six peptides in VP2 (residues 1 to 15, 99 to 108, 118 to 132, 133 to 147, 227 to 236 and 247 to 256) identified the positions of separate Th cell epitopes, and two overlapping peptides (residues 173 to 182 and 178 to 192) defined an additional Th cell immunogenic sequence. Three individual peptides in VP3 (residues 46 to 58, 136 to 150 and 198 to 212) and two overlapping peptides (residues 1 to 15 and 11 to 20) also represent Th cell epitopes. Similar assays with C57BL/6 (H-2b) and SJL/J (H-2s) mice showed that the pattern of recognition of these peptides was H-2 restricted. Each of the previously identified sites of B cell antigenicity in VP2 and VP3 are associated with one Th epitope. Comparison of the experimentally determined Th epitopes with potential T cell epitopes identified by several predictive strategies revealed only a low correlation between authentic and predicted epitopes.

Measles virus infection of microvascular endothelium in vivo and ensuing endothelial cell activation may be important in the pathogenesis of subsequent inflammation in target organs. This study investigated the capacity of measles virus to induce procoagulant activity, in vitro, in endothelial cells isolated from human umbilical cord veins. Endothelial cells were infected with a clinical isolate of measles virus propagated in Vero cells. Cells were also incubated with bacterial lipopolysaccharide (10 micrograms/ml), herpes simplex virus type 1, cytomegalovirus or culture medium alone as positive and negative controls, respectively. Endothelial cell procoagulant activity was measured in a one-stage clotting assay. Measles virus stimulated both a time and dose-dependent endothelial cell procoagulant response by the induction of tissue factor synthesis, confirmed by both immunocytochemistry and its dependence on factor VII for activity. This activity was reduced by u.v.-irradiation of the virus. Infected cells were analysed by double immunofluorescent staining for both tissue factor and measles virus N-protein, and examined using confocal scanning laser microscopy. Cells expressing tissue factor were also positive for the measles virus N-protein. Low levels of interleukin-1 were detected in some viral inocula derived from measles virus-infected Vero cells, however neutralising antibody to interleukin-1 failed to inhibit the endothelial cell procoagulant response to measles virus, whereas it significantly reduced procoagulant activity induced in endothelial cells by recombinant interleukin-1. The capacity of measles virus to induce endothelial tissue factor in vitro, may be relevant to the thrombotic vasculopathy associated with measles virus infection in vivo.

Coxsackievirus B3 (CVB3)-induced myocarditis was studied in euthymic (nu/+) and athymic (nu/nu) C3H/HeN (H-2k) mice. Mice were inoculated intraperitoneally with 10(6) p.f.u. of CVB3 (Nancy strain) and sacrificed at intervals up to 92 days post-inoculation (p.i.). Viraemia peaked at day 2 to 3 p.i. and ceased at day 5 to 7 p.i. in a synchronized manner in both sets of mice. Very few infectious particles were detected in the blood of nu/nu mice after day 14 p.i. In nu/nu mice, CVB3 persisted in myocardial tissue with constant titres between 2.7 +/- 1.9 x 10(4) and 7.6 +/- 5.2 x 10(4) p.f.u./mg from day 3 to 92 p.i., which were comparable to those of nu/+ mice in the acute phase. In nu/+ mice, the virus was recovered from all animals examined by day 11 p.i. and from three out of 13 mice between days 14 and 21 p.i., yet no virus was recovered from nu/+ mice at day 42 p.i. In nu/nu mice, sense and antisense RNA for CVB3 was detected in the myocardial tissue up to day 42 p.i. by in situ hybridization and up to day 92 p.i. by reverse transcriptase-PCR. Neither sense nor antisense RNA was detected after day 21 p.i. in nu/+ mice with the same techniques. Myocardial tissue damage was analysed morphologically. At day 92 p.i., the area of myocardial injury peaked at 23% of the section in nu/nu mice. In contrast, less than 0.6% of tissue sections contained lesions in nu/+ mice. A neutralizing antibody response to CVB3 was observed in both nu/nu and nu/+ mice. The mean titre of neutralizing antibody was significantly higher at day 21 p.i. in nu/+ mice, but similar at day 42 p.i. with nu/nu and nu/+ mice. Perforin-producing natural killer-like cells, which are considered to play an important role in causing acute myocarditic lesions in immunocompetent mice, were found in the lesions of nu/nu mice persistently infected with CVB3. Prolonged tumour necrosis factor-alpha mRNA synthesis detected in nu/nu mice appears to reflect the continuous activation of macrophages, which extend phagocytic reactions to virus-infected myocytes. These immunological results suggested that the host immune response devoid of antigen-specific T cell function is not sufficient to terminate CVB3 infection in nu/nu mice. Also, it appears that competent cellular immunity, on the whole, plays a role in curing rather than in aggravating myocarditis in nu+mice.(ABSTRACT TRUNCATED AT 400 WORDS)

The complete nucleotide sequence of genome segment 8 (S8) of Nilaparvata lugens reovirus (NLRV) was determined. It consisted of 1802 nucleotides containing a long open reading frame (562 amino acids), which was expressed in Escherichia coli as a fusion protein. The expressed S8 product, a 62K protein, was detected by Western blotting using IgG directed against intact NLRV particles. This result indicates that S8 encodes the major outer capsid protein of NLRV. The protein exhibited 18.6% amino acid sequence identity with the predicted translation product of S10 of rice black-streaked dwarf virus.

Diagnosis of Crimean-Congo haemorrhagic fever (CCHF) virus infections is hampered by the problems of handling this human pathogen, which requires the highest levels of biological containment. Recombinant antigens were examined for their potential as non-hazardous diagnostic reagents. The nucleocapsid (N) gene of the Greek AP92 isolate of CCHF virus was sequenced from cloned PCR products and the open reading frame was identified by homology to the N protein of a Chinese isolate of CCHF virus. The N protein was expressed to high levels in a baculovirus expression system. Three N protein-derived peptides were expressed in Escherichia coli as fusions with glutathione S-transferase and the antigenicities of these proteins and the baculovirus-expressed protein were tested by ELISA. When tested with laboratory animal sera representing all seven serogroups of nairoviruses, the only reactive sera were those raised to CCHF virus (Greek, Nigerian and Chinese isolates) and, more weakly, Hazara virus. When tested with a panel of known positive and negative human sera, the baculovirus-expressed N protein, and the peptide derived from the central region of the N protein, proved to be the best for identifying CCHF virus-specific IgG.

Baculovirus-expressed human papillomavirus type 11 (HPV-11) major capsid protein (L1) virus-like particles (VLPs) were produced in insect cells and purified on CsCl density gradients. The VLPs retained conformational neutralizing epitopes that were detected by a series of HPV-11-neutralizing monoclonal antibodies. Electron microscopy determined that the HPV-11 L1 VLPs were variable in size with a surface topography similar to that of infectious HPV-11. The VLPs were very antigenic, and induced high titres of neutralizing antibodies in rabbits and mice when used as an immunogen without commercial preparations of adjuvant. These VLP reagents may be effective vaccines for protection against HPV infections.

The nucleotide sequence of the RNA 2 of a strawberry isolate (H) of strawberry latent ringspot virus (SLRSV) comprised 3824 nucleotides and contained one long open reading frame with a theoretical coding capacity of 890 amino acids equivalent to a protein of 98.8K. The N-terminal amino acid sequences of virion-derived proteins were determined by Edman degradation allowing the capsid coding regions to be located and serine/glycine cleavage sites to be identified within the polyprotein. The amino acid sequence in the capsid coding region of an isolate of SLRSV from flowering cherry in New Zealand was 97% identical to that of SLRSV-H. Except in the 3' and 5' terminal non-coding sequences, computer-based alignment and comparison algorithms did not reveal any substantial homologies between RNA 2 of SLRSV-H and the equivalent genomic segments in the nepoviruses arabis mosaic, cherry leaf roll, grapevine fanleaf, raspberry ringspot, grapevine hungarian chrome mosaic, tomato blackring, tomato ringspot, tobacco ringspot, or in the comoviruses cowpea mosaic and red clover mottle. Despite the similarities in overall genome organization, data from RNA 2 remain insufficient for unambiguous positioning of SLRSV in relation to species/genera in the Comoviridae.

We describe the cloning, nucleotide sequencing and expression in Escherichia coli of the major capsid component (VP60) from the Spanish field isolate AST/89 of rabbit haemorrhagic disease virus (RHDV). The sequence of the 3'-terminal 2483 nucleotides of the genome was found to be 95.4% identical to the German RHDV strain, showing ten changes in the deduced VP60 amino acid sequence. The gene coding for this structural polypeptide has been expressed in bacteria as a beta-galactosidase fusion protein or using a T7 RNA polymerase-based system. The VP60 fusion protein showed only partial antigenic similarity with native VP60 and did not confer protective immunity. The recombinant VP60 produced in the T7 RNA polymerase-based system was antigenically similar to the viral polypeptide as determined using polyclonal and monoclonal antibodies. When used to immunize rabbits the recombinant VP60 was able to protect the animals against a lethal challenge using purified RHDV.

Purified virions of lettuce infectious yellows virus (LIYV), a tentative member of the closterovirus group, contained two RNAs of approximately 8500 and 7300 nucleotides (RNAs 1 and 2 respectively) and a single coat protein species with M(r) of approximately 28,000. LIYV-infected plants contained multiple dsRNAs. The two largest were the correct size for the replicative forms of LIYV virion RNAs 1 and 2. To assess the relationships between LIYV RNAs 1 and 2, cDNAs corresponding to the virion RNAs were cloned. Northern blot hybridization analysis showed no detectable sequence homology between these RNAs. A partial amino acid sequence obtained from purified LIYV coat protein was found to align in the most upstream of four complete open reading frames (ORFs) identified in a LIYV RNA 2 cDNA clone. The identity of this ORF was confirmed as the LIYV coat protein gene by immunological analysis of the gene product expressed in vitro and in Escherichia coli. Computer analysis of the LIYV coat protein amino acid sequence indicated that it belongs to a large family of proteins forming filamentous capsids of RNA plant viruses. The LIYV coat protein appears to be most closely related to the coat proteins of two closteroviruses, beet yellows virus and citrus tristeza virus.

The genome of pea enation mosaic virus (PEMV) is composed of two taxonomically unrelated RNAs, interacting to create what has traditionally been considered a bipartite virus. The cohesiveness of this interaction was assessed by examining the autonomy of each RNA in viral replication, coat protein expression and systemic invasion. Using a pea protoplast system, in vitro transcripts of RNA1 were found to be capable of initiating RNA2-independent replication, including the formation of the distinctive nuclear membrane-based replication complex associated with wild-type PEMV infection. Western blotting and electron microscopic analysis demonstrated that the synthesis of the RNA1-encoded coat protein, as well as virion assembly, was also independent of RNA2-directed functions. Mechanical inoculations with transcripts of RNA1 failed to establish a systemic RNA1 infection, whereas inoculations with RNA2 were able to establish a largely asymptomatic systemic infection. Combined inoculum containing RNA1 and RNA2 transcripts were able to recreate wild-type PEMV symptomatology, demonstrating the dependence of RNA1 on RNA2 for mechanical passage. With the notable exception of the adaptation of PEMV to establish a true systemic invasion, these data further strengthen the analogy between PEMV and the helper-dependent complexes associated with members of the luteovirus group.

Immature or B capsids of herpes simplex virus type 1 (HSV-1) are composed of seven proteins encoded by six viral genes. The proteins encoded by UL18 (VP23), UL19 (VP5), UL35 (VP26) and UL38 (VP19C) are components of the outer capsid shell whereas those specified by UL26 (VP21 and VP24) and UL26.5 (VP22a), are involved in scaffold formation. We have used a panel of recombinant baculoviruses, each expressing one of the capsid protein genes, to examine the requirements for capsid assembly. Coexpression of the six genes in insect cells resulted in the formation of capsids that were indistinguishable in appearance and protein composition from those made during HSV-1 infection of mammalian cells. This demonstrates that the proteins encoded by the known capsid genes contain all the structural information necessary for capsid assembly and that other virus-encoded proteins are not required for this process. Omission of single recombinant baculoviruses from this system allowed the role of individual HSV-1 proteins in capsid assembly to be determined. Thus, capsid assembly did not take place in the absence of VP23, VP5 or VP19C, whereas lack of VP26 had no discernible effect on capsid formation. Capsids assembled in the absence of the UL26 gene products had a large-cored phenotype resembling that previously described for the HSV-1 mutant ts1201 which has a lesion in this gene. Some apparently intact capsid shells were also made in the absence of the major scaffolding protein, VP22a, whereas the omission of both UL26 and UL26.5 resulted in the appearance of large numbers of partial and deformed capsid shells.

It has been shown previously that the viral glycoprotein VP7 is a major target of cytotoxic T lymphocytes (CTLs) induced by bovine rotavirus RF in C57BL/6 mice. Here we show that these RF-specific CTLs recognize target cells infected with a recombinant vaccinia virus (rVV) expressing the SA11 (simian rotavirus) VP6 but not those infected with rVVs that express RF VP1, VP2 or VP3 core proteins. After immunization of mice with insect cells infected with recombinant baculoviruses that express the corresponding RF proteins a strong specific CTL response was generated against target cells infected with VP6 rVV and VP3 rVV, a weak one against the VP2 rVV but none against the VP1 rVV. Kb-restricted CTL epitopes were identified on VP6 and VP3 using allele-specific motifs. When peptides corresponding to these epitopes were used to restimulate in vitro the spleen cells from RF-immunized mice, CTLs specific for each peptide and its corresponding rVV were induced.

The VP-2 gene of mink enteritis parvovirus (MEV) was amplified by the polymerase chain reaction using MEV DNA isolated from the faeces of a naturally infected mink. Subsequently the VP-2 gene was cloned into a baculovirus expression vector. Recombinant baculoviruses were isolated and the MEV VP-2 gene product was characterized after expression in Sf9 insect cells. The MEV VP-2 product had the same size as that reported for the wild-type MEV VP-2 protein and was recognized by convalescent sera from MEV-infected mink and a panel of monoclonal antibodies reactive to MEV. Furthermore, the VP-2 protein was able to form parvovirus-like particles, which had haemagglutinating properties comparable with the wild-type MEV. The cloned VP-2 gene was sequenced and only five nucleotide differences were found after alignment with the known sequences of the MEV type 1 and type 2 isolates. Surprisingly, the VP-2 gene encoded a valine and a tyrosine at amino acid positions 232 and 234, identical to the situation found in MEV type 1, but at position 300 there was a valine which is a determinant of MEV type 2. Immunization of mink with approximately 40,000 haemagglutinating units of recombinant MEV VP-2 induced a measurable antibody response as tested by haemagglutination inhibition. Furthermore, the immunized mink did not excrete virus and did not develop clinical disease upon challenge with a virulent isolate of MEV.

Transgenic Samsun NN tobacco plants expressing the coat protein of tobacco rattle virus were exposed to mechanical leaf inoculation with tobacco rattle virus and to viruliferous trichodorid vector nematodes. Whereas plants were resistant to mechanical inoculation the vector nematodes successfully transmitted tobacco rattle virus to the roots as well as to the leaves of these plants. It is suggested that transgenic resistance is overcome either because vector nematodes inject relatively large numbers of virus particles into a cell or because they inject destabilized particles. The results indicate that coat protein-mediated resistance is unlikely to be of value for controlling tobacco rattle virus in field crops.

Sunn-hemp mosaic tobamovirus (SHMV) facilitated the spread of the cowpea strain of southern bean sobemovirus (SBMV-C) only in inoculated leaves of common bean (Phaseolus vulgaris L. cv. Bountiful), a resistant host for SBMV-C. Tissue prints of bean primary leaves doubly inoculated with SHMV and SBMV-C, developed by Western blotting, showed the presence of the SBMV-C capsid antigen in the mesophyll and epidermis, but no antigen was detected in the conducting bundles. Typical SBMV-C virions were not seen in electron micrographs of immunogold-labelled mesophyll cells; instead, specifically labelled, amorphous protein clumps were found in the vacuole. Particles of smaller diameter than that of typical SBMV-C virions were specifically trapped by SBMV antibodies following immunosorbent electron microscopy of extracts from doubly infected leaves. SBMV-C coat protein from infected Vigna unguiculata L. (cowpea) and bean plants showed no difference in its mobility following electrophoresis in denaturing SDS-polyacrylamide gels. Lack of efficient assembly of SBMV-C virions does not impede cell-to-cell movement of the virus in doubly infected leaves of bean, yet it is probably an important factor in determining the inability of SBMV-C to move into and/or through the vascular system of this host.

During an infection with cowpea mosaic virus (CPMV) both virion assembly and formation of tubules associated with plasmodesmata are required for cell-to-cell movement. These functions are encoded by the M-RNA of CPMV. To study the mechanism of CPMV movement, mutant N123 was used in complementation studies with sunn-hemp mosaic virus (SHMV), a legume-infecting tobamovirus. Previous studies have shown that N123 fails to spread in cowpea plants because of mutation(s) in its M-RNA. However, the mutant was efficiently replicated in cowpea protoplasts, in which virions were formed and tubular transport structures were induced. After high-dose inoculation of cowpeas with N123, only a few infected protoplasts could be isolated, indicating that cell-to-cell transport of N123 was greatly impaired, if not completely abolished. Upon coinoculation with SHMV, mutant N123 infected cowpea plants systemically and accumulated to levels which were comparable to those of wild-type CPMV. In contrast, separate B-RNA of CPMV and a CPMV deletion mutant lacking the tubule-inducing function, were complemented by SHMV to only low levels. It is concluded that SHMV-facilitated spread of CPMV in the non-virion tobamovirus mode is inefficient and that spread of mutant N123 is probably in the CPMV mode, SHMV providing an as yet unidentified helper function.

The gene encoding the major nucleocapsid, N, protein of measles virus has been inserted into a baculovirus vector under the control of the polyhedrin promoter. Insect cells infected with this recombinant baculovirus synthesize high levels of measles N protein, up to 40% of total soluble cell protein. The recombinant protein is recognized by sera from convalescent patients, vaccinees and patients with subacute sclerosing panencephalitis and thus could form the basis of a simple diagnostic assay. Nucleocapsid-like structures, similar to those found in mammalian cells infected with measles virus, can be observed in both the nucleus and cytoplasm of the infected insect cells. These have many structural features in common with nucleocapsids found in measles virus-infected cells, but are longer (up to 2 microns) and have a lower buoyant density. Measles N protein thus appears to be capable of assembling into nucleocapsid-like structures in the absence of measles virion RNA or other viral proteins.

Multiple copies of the Johnsongrass mosaic virus coat protein synthesized in Escherichia coli can readily assemble to form potyvirus-like particles. This E. coli expression system has been used to identify some of the key amino acid residues, within the core region of the coat protein, required for assembly. The two charged residues R194 and D238 previously proposed theoretically to be involved as a pair in the construction of a salt bridge crucial for the assembly process were targeted for site-directed mutagenesis. The results from our experiments suggest that the two residues are required for the assembly process but are not necessarily involved as a pair in a common salt bridge.

Bluetongue virus (BTV) VP6 is often resolved into two closely migrating bands by SDS-PAGE (VP6 and VP6a). RNA segment 9 of BTV-serotype 1 South Africa (encoding VP6) has been cloned as cDNA, and the complete sequence has been determined. Expression of this clone both in vitro and in tissue culture produced the same polypeptide doublet as seen previously in extracts from BTV-infected cells. Modification of the cDNA, including the removal of the first initiation codon, demonstrated that the two forms of VP6 are derived from initiation of protein synthesis at two distinct sites and not by post-translational modification.

The ATCC66 strain of brome mosaic virus (BMV) (propagated at Kyoto University) contained two types of coat protein in its virion whereas the Russian strain of BMV has been known to contain a single coat protein; both strains have two initiation codons for coat protein in the same reading frame at the 5'-proximal end of the gene in RNA 4. Comparative studies on the nucleotide sequences of the ATCC66 and Russian strains of BMV demonstrated that in the ATCC66 strain, two adjacent adenine residues were absent from RNA 3 in the leader sequences of the coat protein gene just a few nucleotides 5' to the first initiation codon of the coat protein gene. Using biologically active cDNA clones of BMV RNA of the ATCC66 strain, we inserted two adjacent adenine residues into the cDNA of RNA 3 to obtain an RNA 3 transcript which has the same nucleotide sequence as the Russian strain in the non-coding leader sequence of the coat protein gene. Barley protoplasts inoculated with this RNA 3 transcript together with RNA 1 and 2 produced a single coat protein. To obtain further insight into the mechanism of translation of the BMV coat protein, we constructed several types of RNA 4 by changing the sequence surrounding the first AUG codon in the coat protein gene and analysed the in vitro translation products of the mutant RNA 4. The results confirmed that the absence of the two adjacent adenine residues was responsible for the production of two types of coat protein in the ATCC66 strain. The deletion of the two adjacent adenine residues in ATCC66 resulted in a base substitution of A with U three nucleotides 5' to the first AUG in the coat protein gene. The base substitution reduced translational activity from the first AUG codon and concomitantly increased translational activity from the second AUG codon from which a truncated coat protein was translated.

RNA purified from the feline calicivirus (FCV) F9 vaccine strain was used to prepare a cDNA library in the expression vector lambda gt11. The library was screened for expression of FCV antigen using a rabbit antiserum prepared against purified FCV. A 330 bp cDNA clone was identified and used as a probe to obtain a larger overlapping clone of 1369 bp. Comparative sequence analysis with the CFI and F4 strains showed that the clones were derived from the 3' open reading frame encoding the capsid protein. The region encoded by the 330 bp clone was shown to be variable in the three strains compared, and therefore the probable location of major antigenic variation. This clone was expressed in a bacterial system and antiserum to the recombinant protein was used in immunoblots to confirm that this clone was derived from the gene encoding the capsid protein. From these immunoblots, several other capsid-related polypeptides were identified. Comparison with immunoblots using post-vaccination cat sera showed the antibody response in the cat was directed mainly against the capsid protein. Antiserum to the recombinant protein was shown to be effective in neutralizing the infectivity of FCV, indicating that at least one major neutralizing epitope had been cloned.

Bovine viral diarrhoea virus (BVDV) belongs to the pestivirus group, a genus within the Flaviviridae family. It possesses a positive-sense ssRNA genome with a single large open reading frame (ORF) encoding about 4000 amino acids. Here we report the continuation of our studies of pestivirus protein biogenesis, involving expression from the viral non-structural protein-encoding region. The 3'-terminal 60% of the BVDV ORF was cloned into a plasmid transfer vector which was then used to construct a recombinant baculovirus. Infection of Spodoptera frugiperda Sf9 cells with this recombinant virus resulted in the production of the expected mature viral proteins. Polyprotein processing by the BVDV p80 proteinase appeared to be nearly identical to that observed in authentic BVDV-infected bovine cells, and as previously shown to occur when expression of the same region was studied in a mammalian cell transient expression system. However, one viral proteolytic cleavage did not occur in the baculovirus-infected insect cells; the viral p80 proteinase failed to act at its own N terminus. This recombinant baculovirus/insect cell expression system provides an abundant source of BVDV non-structural proteins. Therefore we explored the utility of the proteins produced in this system for the detection of anti-BVDV antibodies in bovine sera. In preliminary experiments using these antigens in an ELISA we found a positive correlation between the presence of ELISA-reactive antibody and virus-neutralizing activity.