- Volume 84, Issue 11, 2003

Group B coxsackieviruses (CVBs) have a well-established association with type 1 diabetes but the mechanism of depletion of beta-cell mass following infection has not yet been defined. In this report we show that the major difference in pathogenesis between the E2 diabetogenic strain of CVB4 and the prototypic JVB strain in SJL mice is not in tropism for islet cells but in the degree of damage inflicted on the exocrine pancreas and the resulting capacity for regeneration of both acinar and islet tissue by the host. Both strains replicated to a high titre in acinar tissue up to day 3 post-infection (p.i.), while the islets of Langerhans were largely spared. However, the pancreas in the JVB-infected animals then regenerated and many small islets were seen throughout the tissue by day 10 p.i. In contrast, the acinar tissue in E2-infected mice became increasingly necrotic until all that remained by day 21 p.i. were large islets containing varying numbers of dead cells, caught up in strands of connective tissue. Surviving beta cells were found to synthesize little insulin, although islet amyloid polypeptide was detected and glucagon synthesis in alpha cells appeared normal or enhanced. Our results suggest that the key to CVB-E2-induced damage lies in the exocrine tissue and prevention of islet neogenesis rather than from direct effects on existing islets.

A cytopathic agent was isolated using Vero cells from the culture medium of HeLa cells that had been used for more than 30 years in our laboratory. This agent, termed U-1 strain, was serially passed in Vero cells with distinct CPE. Particles of U-1 strain negatively stained with phosphotungstic acid exhibited a distinct surface that resembled Aichi virus. The RNA genome of U-1 strain comprises 8374 nt, with a genome organization analogous to that of picornaviruses. Possible cleavage sites of the large ORF, which encoded a leader protein prior to the capsid protein region, were assigned following amino acid alignment with Aichi virus. The virus sequence had 33 and 75 % amino acid identity with the Aichi virus VP1 and 3D regions, respectively, but no more than 23 and 36 % with those of the prototype strains of other Picornaviridae. The dendrogram based on the P1, P2 and P3 proteins indicated that U-1 strain is genetically included in the genus Kobuvirus but is distinct from Aichi virus. Of 72 cattle sera, 43 (59·7 %) were positive for neutralizing antibody against U-1 strain at a titre of 1 : 8 or more. However, sera from 190 humans, 242 monkeys, 139 pigs, 5 horses, 22 dogs and 9 cats did not neutralize U-1 strain at a 1 : 4 dilution. RNA corresponding to U-1 strain was detected in 12 (16·7 %) of 72 faecal samples from cattle by RT-PCR. These results indicated that U-1 strain, suspected to be a contaminant from calf sera, is a new species of the genus Kobuvirus, now termed bovine kobuvirus.

Because Rabbit haemorrhagic disease virus (RHDV) is highly pathogenic for rabbits, farmers illegally introduced it as a bio-control agent onto New Zealand farms in 1997. The virus was dispersed rapidly, initially causing high fatality rates in rabbits. Nevertheless, many survived and these surviving rabbits have been investigated for evidence of infection by RHDV. Livers from healthy rabbits contained RHDV-specific RNA, as shown by nested RT-PCR sequencing. The sequences of the viral capsids were related closely to the released Czech strain of RHDV, although the sequence from one rabbit was related most closely to a Spanish strain of RHDV. Phylogenetic analysis of the capsid sequences of 38 samples implied that there have been at least two introductions of the Czech virus into New Zealand, probably corresponding firstly to the original illegal introduction by farmers and secondly to the introduction of the same virus under governmental control. Genomic length sequence of two samples was obtained, suggesting that they may have retained the potential to be infectious, although this has not yet been demonstrated. The detection of genomic-length RNA in the liver of healthy rabbits suggests that even though a highly virulent virus was introduced into New Zealand, it rapidly established persistent or latent infections in a proportion of rabbits. This might account for their ability to survive in the face of virulent released virus. Moreover, the co-circulation of other strains of RHDV in the same rabbit population, such as the Spanish strain, might also impact on their susceptibility to the bio-control agent.

Pre-weaning diarrhoea is a well-known problem in mink farming in Europe, causing morbidity that varies between farms, regions and season. Different causalities for the disease have been proposed, but only most recently has a novel astrovirus been identified as an important risk factor. In this report, the molecular characterization, origin and evolution of this novel astrovirus of mink are discussed. The polyadenylated, positive-stranded RNA genome was sequenced and found to contain 6610 nt, organized into three ORFs and two short UTRs. A ribosomal frameshift sequence links the 5′ two ORFs, containing sequence motifs for a serine protease (ORF1a) and an RNA-dependent RNA polymerase (ORF1b). The structural proteins are encoded by ORF2 and, presumably, are expressed as a polyprotein precursor to be cleaved into the mature capsid proteins. These results indicate that mink astrovirus (MiAstV) has all of the features typical of members of the Astroviridae. Phylogenetic analyses revealed that MiAstV is distantly related to established astroviruses, showing less than 67 % similarity at the nucleotide level with its closest relative, ovine astrovirus, and even lower identities at the predicted amino acid level. Nevertheless, sequence analysis of MiAstV isolates from geographically distinct Swedish and Danish farms showed much less diversity. This suggests either the spread in the mink population of a virus that has evolved a long time ago or the recent introduction of an ancient virus into a new host species.

The endothelial cell line ECV304, derived from human umbilical cord and identified to be susceptible to dengue virus type 2 (DEN-2) infection, was used to study the molecular mechanism of DEN-2 binding to endothelial cells. DEN-2 was found by virus overlay protein-binding assays (VOPBAs) to bind to three ECV304 cell membrane proteins with molecular masses of 29, 34 and 43 kDa. Only a single protein of 29 kDa was observed when VOPBAs were carried out using preparations of trypsin-treated ECV304 cells. Pre-incubation of live ECV304 cells in culture or cell membrane proteins in modified VOPBAs with the recombinant DEN-2 envelope glycoprotein (rEgp) inhibited DEN-2 infection and blocked virus binding to the three proteins identified. These results indicate that DEN-2 rEgp could bind to three proteins on the surface of ECV304 cells. This virus–cell interaction may be associated with the receptor complex specific for DEN-2 infection of endothelial cells.

Borna disease virus (BDV) is an enveloped virus. Its non-segmented, negative-stranded RNA genome has the coding capability for six main polypeptides and has an organization characteristic of members of the order Mononegavirales. However, based on its unique genetics and biological features, BDV is considered to be the prototypic member of a new virus family, Bornaviridae. Here, the establishment of a reverse genetics system for BDV is described. Intracellular synthesis of a BDV RNA analogue or minigenome (MG) from a plasmid was driven by RNA polymerase I. Co-transfection with plasmids expressing the BDV polymerase (L), nucleoprotein (N) and phosphoprotein (P) under the control of RNA polymerase II allowed for BDV MG replication and expression. This process depended on a delicate N : P ratio, whereas the L : P ratio was less critical. Two isoforms of N, Np40 and Np38, are present in BDV-infected cells but only Np40 was strictly required for virus polymerase activity. BDV p10 polypeptide encoded by the P gene exhibited a strong inhibitory effect on BDV MG expression.

The influenza virus matrix protein (M1) possesses a cysteine and histidine (CCHH) motif in the helix 9 (H9) and adjacent region (148 CATCEQIADSQHRSH 162). The CCHH motif has been proposed as a putative zinc finger motif and zinc-binding activity has been implicated in virus uncoating as well as transcription inhibition and mRNA regulation. The function of the CCHH motif in the influenza virus life cycle was investigated by site-directed mutagenesis (alanine replacement) and by rescuing mutant viruses by reverse genetics. Mutant viruses containing an alanine replacement of the cysteine and histidine residues, either individually or in combination, were seen to exhibit wt phenotype in multiple virus growth cycles and plaque morphology. In addition, synthetic peptides containing the putative zinc finger motif did not inhibit virus replication in MDCK cells. However, mutation of Ala155 in H9 was lethal for rescuing infectious virus. These data show that the CCHH motif does not provide a critical function in the influenza virus life cycle in cell culture and that the zinc-binding function may not be involved in virus biology. However, the lethal phenotype of the Ala155 mutation shows that the H9 region of M1 provides some other critical function(s) in virus replication.

The entire nucleotide sequence of the G gene of three human respiratory syncytial virus (HRSV) isolates (antigenic group B) has been determined. These three viruses (named BA viruses) were isolated in Buenos Aires in 1999 from specimens collected in different hospitals and at different dates. BA viruses have an exact duplication of 60 nucleotides in the G gene, starting after residue 791. This duplication is flanked by a repeat of four nucleotides (GUGU) and can fold into a relatively stable secondary structure. These features suggest a possible mechanism for the generation of a duplicated G segment. The predicted polypeptide is lengthened by 20 amino acids (residues 260–279) and this is reflected in the slower electrophoretic mobility of the G protein precursor of BA viruses compared with related viruses. The changes reported here expand the examples of drastic genetic alterations that can be introduced into the G protein sequence of HRSV while it replicates in its natural host.

Newcastle disease virus (NDV) possesses two envelope spike glycoproteins: the haemagglutinin–neuraminidase (HN) protein and the fusion (F) protein. The HN protein, which is responsible for virus attachment to sialic acid-containing receptors, varies in length due to differences in the sizes of the ORFs. An HN protein precursor of 616 aa has been found in avirulent but not in virulent NDV strains, whereas an HN protein of 571 aa can be detected in highly virulent strains only. An HN protein of 577 aa is present in virulent and avirulent strains. The F protein, which mediates virus–cell fusion, requires proteolytic activation at an internal cleavage site, whose amino acid composition determines cleavability by various proteases. Here, the functional significance of the length of the HN protein in combination with F protein cleavage sites typical for virulent (velogenic and mesogenic) or avirulent (lentogenic) strains was investigated. To this end, site-directed mutagenesis was used to construct recombinant NDV on the basis of an infectious clone of the lentogenic vaccine virus Clone-30. Only recombinant NDV expressing an F protein with a multibasic cleavage site typical of virulent strains was able to spread efficiently in cell culture, irrespective of the size of the HN protein. Moreover, as determined by the intracerebral pathogenicity index (ICPI) in 1-day-old, specific-pathogen-free chickens, pathogenicity was influenced by the cleavability of the F protein and not by the length of the HN protein. The maximum ICPI value obtained for these recombinants was 1·3, as compared to a possible maximum of 2. This demonstrates that the modifications introduced did not result in the conversion of the lentogenic Clone-30 to a velogenic strain with an ICPI value of >1·5 and suggests the involvement of additional virulence determinants that contribute to the pathogenicity of NDV.

In this study, the cell cycle modulation of retrovirus vector production and transduction was analysed. Retrovirus vector expression was found to be similar in all phases of the cell cycle and, in contrast to some other virus promoters shown previously to be upregulated by G2/M arrest, Moloney murine leukaemia virus LTR-driven expression was upregulated neither by G2/M growth arrest nor by G1/S growth arrest. In contrast, cultures enriched for S phase cells produced more infectious virions, apparently by modulation of stages consequent to provirus expression. In terms of retrovirus transduction, limitations appear to be slow progression through the cell cycle and short half-life of the virus. Synchronization of cells prior to mitosis can increase transduction efficiency. Cell cycle modulation can be used to modify retrovirus vector production and transduction and can allow short transduction periods.

Human papillomavirus type 16 (HPV-16) is the prototype strain among the malignant types of HPV in the western world. The main promoter, P97, located in front of the E6 ORF, has been shown to control expression of the oncogenes E6 and E7. These oncogenes are expressed continuously in HPV-16-transformed cells. In contrast to malignant HPV types, non-malignant HPV types have separate promoters driving the expression of E6 and E7. Experiments have shown that the translation of E7 is more efficient from monocistronic than bicistronic transcripts encoding both E6 and E7. Here, identification of a cluster of transcription start sites located in the E6 ORF of HPV-16 is presented. Transcripts from this region contain the E7 ORF as the first reading frame. The cluster consists of multiple transcription start sites located around nt 441. Additional transcription start sites were identified in a cluster around nt 480. A transcription start site has been identified previously at nt 480 but has never been characterized further. The region responsible for transcription activity was mapped to nt 272–448. Mutational analysis showed that initiation of transcription is independent of a TATA-box element, which is consistent with the finding of multiple transcription start sites. Furthermore, it is shown that proteins from HeLa and SiHa nuclear cell extracts bind to the two regions at nt 291–314 and 388–411, and that these two regions influence transcription activity in a cell type-dependent manner.

Samples of neoplastic and normal urothelium were obtained from cows originating from areas of southern Italy, a region in which chronic enzootic haematuria is endemic and bracken fern infestation is widespread. Specimens were analysed for bovine papillomavirus type 2 (BPV-2) DNA, BPV-2 E5 expression and telomerase activity. A total of 46 of 60 tumours and 17 of 34 normal bladder mucosa samples harboured BPV-2 DNA. Analysis of a subset of samples showed E5 protein expression and telomerase activity in tumour tissue only. No normal samples positive for BPV DNA showed E5 protein expression or telomerase activity, suggesting the presence of DNA in a latent state. Taken together, these data on naturally occurring bovine bladder tumours corroborate the hypothesis of their virus origin.

Murine polyomavirus (MPyV) infection occurs through recognition of sialic acid (SA) residues present on the host cell membrane, but the nature of the molecules involved and the exact role of this interaction in virus cell entry still need to be clarified. In this work, mutations at residues R77 or H298 of the MPyV VP1 protein were shown to lead to a complete loss of virus infectivity, which, however, could be restored by lipofection of virus particles into the cytoplasm of the host cells. Using virus-like particles (VLPs), it was demonstrated that the non-infectivity of these mutants was due to impaired cell entry caused by total abrogation of SA-dependent cell binding. This indicates that SA residues are essential primary cell receptors for MPyV. As the α4β1 integrin has been identified recently as a cell receptor for MPyV, the relationship, if any, was investigated between SA-containing and α4β1 integrin receptors. The ability of mutants R77Q and H298Q and wt VLPs to bind to cells overexpressing the α4β1 integrin was studied in SA-positive (BALB/c 3T3 cells and Pro-5 cells) and SA-deficient (Pro5-derived Lec-2 cells) backgrounds. Overexpression of α4β1 integrin did not restore binding of mutant VLPs in any of these cell lines or, indeed, that of wt VLPs in a SA-deficient background. Moreover, evidence is provided that overexpression of the sialylated α4β1 integrin enhances wt VLP cell binding, suggesting that, in addition to its function at a post-attachment level, α4β1 integrin acts also as one of the SA-containing receptors for initial cell binding.

Porcine circovirus type 2 (PCV-2) has been identified as the causal agent of postweaning multisystemic wasting syndrome and has been associated with several other disease syndromes in pigs. To date, however, little is known regarding the mechanism(s) underlying the pathogenesis of PCV-2-induced diseases and the interaction of the virus with the host immune system. In the present study, oligodeoxynucleotides (ODNs), with central CpG motifs selected from the genome of PCV-2, were demonstrated to modulate the immune response of porcine PBMCs. Four of the five ODNs tested were demonstrated to act in a stimulatory manner via induction of IFN-α production, whereas only one of the five ODNs showed inhibitory activity. Also, this inhibitory ODN was demonstrated to completely inhibit IFN-α production induced by the other stimulatory ODNs and showed a variable degree of inhibitory action on other known inducers of IFN-α. Although no single common characteristic among resistant or susceptible inducers could be identified, the presence of immune modulatory sequences in the genome of PCV-2 may represent an underlying mechanism of the pathogenesis of PCV-2-associated diseases.

The assembly of adenovirus particles is a multistep process, in which viral genomic DNA is selected and subsequently inserted into preformed empty capsids. The selective encapsidation of the adenovirus genome is directed by cis-acting packaging motifs, termed A repeats due to their AT-rich character in DNA sequence. A repeats are usually located at the left end of the viral genome. In this report, the construction and analysis of bovine adenovirus type 3 (BAdV-3) mutants containing deletion mutations introduced into the AT-rich regions are described. The main cis-acting packaging domains of BAdV-3 were localized between nt 224 and 540 relative to the left end of the viral genome. They displayed a functional redundancy and followed a hierarchy of importance. In addition, the results demonstrated that not all of the AT-rich units functioned as cis-acting packaging motifs.

The expression of the genes of varicella-zoster virus (VZV) is regulated by self-encoded viral as well as cellular transcription factors. A potential candidate with an ability to influence the transcription of VZV genes is USF (upstream stimulatory factor), which recognizes the consensus E-box motif. Quantitative RT-PCR and immunoblot assays indicate stable expression of both USF1 and USF2 throughout infection. It was also found that USF binds to a variety of E-boxes (consensus and closely related motifs) within the promoters of ORF 8/9 (two elements), ORF 22 and ORF 67. Co-immunoprecipitation experiments and His-tag protein affinity pull-down assays indicate that a direct physical interaction occurs between USF and the major virus transactivator IE62. To study the general effects of USF in the replication of VZV, a cell line expressing a dominant–negative form of USF (A-USF), which inhibits binding of USF to its recognition sites, was created. A significant decrease in virus replication was detected when this cell line was infected with cell-free virus, indicating that USF is an important cellular factor that regulates the transcription of VZV genes.

Blood monocytes infected with pseudorabies virus (PRV), a swine alphaherpesvirus, are not eliminated efficiently by antibody-dependent immunity and may occasionally transport PRV to the pregnant uterus of vaccinated animals. This study examines in vitro the long-term fate of PRV-infected monocytes cultivated in the presence of porcine PRV-specific antibodies. All monocytes were infected and expressed viral late proteins, and 30 % of PRV-infected monocytes cultivated with PRV-specific antibodies survived up to 194 h post-infection (p.i.), the end of the experiment (compared to 0 % for cells cultivated with PRV-negative antibodies). Of these surviving cells, ±75 % no longer expressed microscopically detectable viral late proteins from 144 h p.i. onwards. Remarkably, monocytes infected with a PRV gB-null virus did not survive in the presence of PRV-specific antibodies. These data suggest that PRV-specific antibodies suppress viral protein levels in infected monocytes, perhaps helping the virus to persist and reach internal organs in vaccinated animals.

During neuronal latency of herpes simplex virus (HSV)-1, the latency-associated transcript (LAT) is the only viral gene readily detectable. LAT is required for the high-level reactivation phenotype in animal models. LAT's anti-apoptotic activity was recently demonstrated by our group and it was proposed that LAT's anti-apoptotic function is involved in enhancing the reactivation phenotype. Recently, using chimeric virus CJLAT, it was shown that the reactivation phenotype of LAT− mutant dLAT2903 can be restored to wild-type levels by inserting the bovine herpes virus (BHV)-1 latency-related (LR) gene into the LAT locus of this HSV-1 LAT deletion mutant. Although transcription of the LR gene, like LAT, inhibits apoptosis, LR appears to be multifunctional. To investigate whether the LR gene's anti-apoptotic function was responsible for restoring the high-reactivation phenotype, a mutated BHV-1 LR gene was inserted into the LAT locus of HSV-1 generating the chimeric virus CJLATmut. This mutation consists of three stop codons inserted just after the ATG of the first LR open reading frame (ORF2). In plasmids and in a BHV-1 mutant, this mutation eliminated the LR gene's anti-apoptotic activity, strongly suggesting that ORF2 encodes a protein responsible for LR's anti-apoptotic activity. Reactivation of the CJLATmut virus, in both rabbits and mice, was significantly lower than in wild-type McKrae virus (P=0·0001 and P=0·0003, respectively) and CJLAT virus, containing wild-type LR in place of LAT (P<0·0001) and was similar to LAT− dLAT2903 (P=0·8 and P=0·7, respectively). Thus, disruption of BHV-1 LR ORF2 eliminated the high-reactivation phenotype.