- Volume 85, Issue 3, 2004

Japanese encephalitis (JE) is the most common mosquito-borne encephalitis in the Asia–Pacific region. Patients with JE usually present neuronal involvement, but other organ involvement is relatively rare. Employing human neuroblast-derived (NB) cell lines and different blood cells (erythrocytes, lymphocytes, granulocytes and monocytes), the neurotropism and persistency of Japanese encephalitis virus (JEV) in human cells was investigated. It was found that JEV could not replicate in erythrocytes, granulocytes or lymphocytes. Monocytes and NB cell lines could support replication of JEV as demonstrated by expression of viral NS3 antigen and virus plaque-forming units (p.f.u.). JEV could replicate more efficiently in neuroblastoma (HTB-11) cells than in monocytes after infection for 48 h (2·1±1·2×107 vs 2·8±0·7×102 p.f.u. ml−1). Two different strains of JEV revealed a similar infectivity to different leukocytes and four NB cell lines. In a kinetic study, it was found that JEV-infected monocytes possessed a high viability (90 %) after infection for 5 days, while JEV-infected neuroblastoma cells suffered cell apoptosis in 2 days and decreased viability to less than 1 % in 5 days. Further studies showed that monocytes could take up JEV rapidly, displaying a log scale increase of intracellular JEV titres in 9 h after infection. Significantly, extracellular production of JEV by monocytes started in 12 h, peaked in 3 days and persisted for more than 3 weeks. These results suggest that JEV-infected monocytes may play an important role in harbouring JEV for eventual transmission to NB cells and that modulation of JEV-induced NB cell apoptosis may be useful in treating patients with JE.

White tail disease (WTD) causes a high mortality rate in the freshwater prawn Macrobrachium rosenbergii. The pathogenic agent is a small virus, 25 nm in diameter, Macrobrachium rosenbergii nodavirus (MrNV), associated with extra small virus (XSV), a virus-like particle,15 nm in diameter. Sequencing of the XSV genome showed that it consists of a linear single-stranded RNA of 796 nucleotides, encoding a single structural protein, the capsid CP-17. The genome is in sense orientation, ended by a short poly(A) tail at the 3′-end. Sequence comparison did not allow XSV to be affiliated to known virus families. The hypothesis that XSV is a satellite virus, such as those described in the plant kingdom, is put forward based on its characteristics. It would constitute, therefore, the first satellite virus associated with a nodavirus.

So-called fowl glioma is a retroviral infectious disease caused by avian leukosis virus subgroup A (ALV-A). We determined the complete nucleotide sequence of the virus genome. The full-length sequence was consistent with a genetic organization typical of a replication-competent type C retrovirus lacking viral oncogenes. The coding sequences were well conserved with those of replication-competent viruses, but the 3′ noncoding regions including LTR were most related to those of replication-defective sarcoma viruses. The U3 region of the LTR had a few deletions and several point mutations compared to that of other ALVs. The promoter activities of the LTRs of glioma-inducing ALV and ALV-A standard strain, RAV-1, were equivalent in chick embryo fibroblasts (CEF), while that of glioma-inducing ALV was significantly lower than that of RAV-1 in human astrocytic cells. These subtle differences of the promoter activity of the LTR may be related to the induction of glial neoplasm.

Heparan sulphate (HS) has been found to serve as receptor for initial cell binding of numerous viruses. Different glycosaminoglycans (GAGs), including heparin and HS, were analysed for their ability to bind swine vesicular disease virus (SVDV), a picornavirus with close homology to human coxsackie B5 virus. Binding of SVDV was established by heparin-affinity chromatography. In addition, infection of IB-RS-2 epithelial porcine cells was inhibited by treating the virus with soluble HS, heparin, and chondroitin sulphate B (CS-B), as well as by enzymic digestion of cell surface GAGs. Analysis of the infection course showed that SVDV uses cellular HS for its binding to the cell surface and that this interaction occurs during attachment of the virus, prior to its internalization into the cell. Sequence analysis of SVDV variants selected for their lack of sensitivity to heparin inhibition in vitro led to the identification of two residues (A2135V and I1266K) potentially involved in heparin/HS interaction. The location of these residues in a three-dimensional model shows that they are clustered in a well-exposed region of the capsid, providing a physical mechanism that could account for the heparin-binding phenotype.

To achieve specific gene transfer into human CD4+ cells, murine leukaemia virus (MLV)-based pseudotype vector particles were generated employing Env variants derived from human or simian immunodeficiency virus (HIV-1 or SIVagm). Here, we describe the generation of full-length onco/lentivirus hybrid genomes comprising components of MLV and HIV-1 or SIVagm, respectively, to assess the possibility of replication-competent hybrid virus formation. The env reading frame of an infectious molecular clone of MLV was replaced with the analogous coding regions of HIV-1 or SIVagm encompassing the env gene and accessory genes. Resulting MLV/HIV-1 or MLV/SIVagm hybrid genomes were transfected into 293T cells. Expression of viral proteins and budding of retroviral particles was shown by specific immunostaining and electron microscopy. The viral particles mediated CD4- and co-receptor-specific infection of human cells as demonstrated by PCR and immunostaining in the respective target cells. However, no productive infection resulting in the generation of infectious virus was detected in these cells. Thus, these onco/lentivirus hybrids, although able to initiate single-round infection, were not replication competent. Thus, MLV-based pseudotype vectors carrying Env variants of HIV-1 or SIVagm are not prone to form replication-competent retroviruses, suggesting a favourable safety profile for MLV-based CD4-specific pseudotype vectors.

The putative G glycoprotein genes of 25 human metapneumovirus (hMPV) field isolates obtained during five consecutive epidemic seasons (1997 to 2002) were sequenced. Sequence alignments identified two major genetic groups, designated groups 1 and 2, and two minor genetic clusters within each major group, designated subgroups A and B. Extensive nucleotide and deduced amino acid sequence variability was observed, consisting of high rates of nucleotide substitutions, use of alternative transcription-termination codons and insertions that retained the reading frame. Deduced amino acid sequences showed the greatest variability, with most differences located in the extracellular domain of the protein: nucleotide and amino acid sequence identities for the entire open reading frame ranged from 52 to 58 % and 31 to 35 %, respectively, between the two major groups. Like the closely related avian pneumovirus and human and bovine respiratory syncytial viruses, the predicted G protein of hMPV shared the basic features of a type II mucin-like glycosylated protein. However, differences from these related viruses were also observed, e.g. lack of conserved cysteine clusters as seen in human respiratory syncytial virus and avian pneumovirus. The displacement of genetic groups of hMPV observed during the study period suggests that potential antigenic differences in the G glycoprotein, which have evolved in response to immune-mediated pressure, may influence the circulation patterns of hMPV strains.

Phosphoprotein P of rinderpest virus (RPV), when expressed in E. coli, is present in the unphosphorylated form. Bacterially expressed P protein was phosphorylated by a eukaryotic cellular extract, and casein kinase II (CK II) was identified as the cellular kinase involved in phosphorylation. In vitro phosphorylation of P-deletion mutants identified the N terminus as a phosphorylation domain. In vivo phosphorylation of single or multiple serine mutants of P protein identified serine residues at 49, 88 and 151 as phospho-acceptor residues. The role of P protein phosphorylation in virus replication/transcription was evaluated using the RPV minigenome system and replication/transcription of a reporter gene in vivo. P protein phosphorylation was shown to be essential for in vivo replication/transcription since phosphorylation-null mutants do not support expression of a reporter gene. Transfection of increased amounts of phosphorylation-null mutant did not support minigenome replication/transcription in vivo.

Avian hepatitis E virus (HEV), a novel virus identified from chickens with hepatitis–splenomegaly (HS) syndrome, is genetically and antigenically related to human HEV. Recently, it was found that avian HEV antibody is also prevalent in healthy chickens. A prospective study was done on a known seropositive but healthy chicken farm to identify avian HEV isolates from healthy chickens. Fourteen chickens were randomly selected, tagged and monitored under natural conditions for 19 weeks. All 14 chickens were seronegative at the beginning of the study at 12 weeks of age. By 21 weeks of age, all 14 chickens had seroconverted to avian HEV antibody. None of the chickens had any sign of HS syndrome. Partial helicase gene and capsid gene sequences of avian HEV isolates recovered from a healthy chicken were determined and found to share 75–97 % nucleotide sequence identity with the corresponding regions of avian HEV isolates from chickens with HS syndrome. Thus far, only one strain of avian HEV from a chicken with HS syndrome has been genetically characterized for its capsid gene, therefore the capsid gene region of an additional 14 isolates from chickens with HS syndrome were also characterized. The capsid genes of avian HEV isolates from chickens with HS syndrome were found to be heterogeneic, sharing 76–100 % nucleotide sequence identity with each other. This study indicates that avian HEV is enzootic in chicken flocks and spreads subclinically among chickens in the United States and that the virus is heterogeneic.

To study the replication of Nipah virus (NiV), a minigenome replication assay that does not require the use of infectious virus was developed. The minigenome was constructed to encode a NiV vRNA analogue containing the gene for chloramphenicol acetyltransferase (CAT) under the control of putative NiV transcription motifs and flanked by the NiV genomic termini. CAT protein was detected only when plasmids encoding the NiV minigenome, nucleocapsid protein (N), phosphoprotein (P) and polymerase protein (L) were transfected into CV1 cells. To determine whether NiV conforms to the rule of six, a series of plasmids encoding minigenomes that differed in length by a single nucleotide was tested in the replication assay. CAT production was detected only with the minigenome whose length was an even multiple of six. The replication assay was also used to show that the N, P and L proteins of NiV recognize cis-acting sequences in the genomic termini of Hendra virus (HeV) but not measles virus. While these results suggest that NiV uses a replication strategy that is similar to those of other paramyxoviruses, they also support the inclusion of NiV and HeV in a separate genus within the subfamily Paramyxovirinae.

The human respiratory syncytial virus (HRSV) matrix (M) protein is a structural internal membrane protein. Here we have shown that, like its orthomyxovirus and rhabdovirus counterparts, it has RNA-binding capacity, as determined by retardation of 32P-labelled riboprobes in gel electrophoresis, cross-linking with UV light and Northern-Western assays. Its binding to RNA was neither sequence-specific nor showed a length requirement, although it had cooperative kinetics with a K d of 25 nM and probably two different types of RNA-binding sites. After preparative cross-linking of 32P-labelled riboprobes with purified, renatured HRSV Long strain M protein (256 residues), the residues in contact with RNA were located between amino acids 120 and 170, in the central part of the molecule. Lysine (positions 121, 130, 156 and 157) and arginine (position 170) residues located within this region and conserved among pneumovirus M proteins of different origins were found to be essential for RNA contact. M protein expression did not affect the replication and transcription of HRSV RNA analogues in vivo (except when expressed in large amounts), in contrast to the in vitro transcription inhibition described previously. In addition, M protein was found to aggregate into high-molecular-mass oligomers, both in the presence and absence of its RNA-binding activity. The formation of these structures has been related in other viruses to either viral or host-cell RNA metabolism.

The hepatitis C virus (HCV) non-structural NS5A protein has been shown to associate with a variety of cellular signalling proteins. Of particular interest is the observation that a highly conserved C-terminal polyproline motif in NS5A was able to interact with the Src-homology 3 (SH3) domains of the adaptor protein Grb2. As it has previously been shown that specific polyproline motifs can interact with a range of SH3 domains, we investigated whether NS5A was capable of interacting with other SH3 domain-containing proteins. We show here that NS5A interacts with the SH3 domains of members of the Src family of tyrosine kinases: a combination of in vitro binding assays and co-immunoprecipitation experiments revealed an interaction between NS5A and Hck, Lck, Lyn and Fyn, but interestingly not Src itself. Mutational analysis confirmed that the polyproline motif responsible for binding to Grb2 also bound to the SH3 domains of Hck, Lck, Lyn and Fyn. Furthermore, a previously unidentified polyproline motif, adjacent to the first motif, was also able to mediate binding to the SH3 domain of Lyn. Using transient transfections and Huh-7 cells harbouring a persistently replicating subgenomic HCV replicon we demonstrate that NS5A bound to native Src-family kinases in vivo and differentially modulated kinase activity, inhibiting Hck, Lck and Lyn but activating Fyn. Lastly, we show that signalling pathways controlled by Src-family kinases are modulated in replicon cells. We conclude that the interactions between NS5A and Src-family kinases are physiologically relevant and may play a role in either virus replication or pathogenesis.

A cellular receptor for the haemagglutinating enteroviruses (HEV), and the protein that mediates haemagglutination, is the membrane complement regulatory protein decay accelerating factor (DAF; CD55). Although primate DAF is highly conserved, significant differences exist to enable cell lines derived from primates to be utilized for the characterization of the DAF binding phenotype of human enteroviruses. Thus, several distinct DAF-binding phenotypes of a selection of HEVs (viz. coxsackievirus A21 and echoviruses 6, 7, 11–13, 29) were identified from binding and infection assays using a panel of primate cells derived from human, orang-utan, African Green monkey and baboon tissues. These studies complement our recent determination of the crystal structure of SCR34 of human DAF [ Williams, P., Chaudhry, Y., Goodfellow, I. G., Billington, J., Powell, R., Spiller, O. B., Evans, D. J., Lea, S. (2003). J Biol Chem 278, 10691–10696 ] and have enabled us to better map the regions of DAF with which enteroviruses interact and, in certain cases, predict specific virus–receptor contacts.

Parvoviruses were isolated from Python regius and Boa constrictor snakes and propagated in viper heart (VH-2) and iguana heart (IgH-2) cells. The full-length genome of a snake parvovirus was cloned and both strands were sequenced. The organization of the 4432-nt-long genome was found to be typical of parvoviruses. This genome was flanked by inverted terminal repeats (ITRs) of 154 nt, containing 122 nt terminal hairpins and contained two large open reading frames, encoding the non-structural and structural proteins. Genes of this new parvovirus were most similar to those from waterfowl parvoviruses and from adeno-associated viruses (AAVs), albeit to a relatively low degree and with some organizational differences. The structure of its ITRs also closely resembled those of AAVs. Based on these data, we propose to classify this virus, the first serpentine parvovirus to be identified, as serpentine adeno-associated virus (SAAV) in the genus Dependovirus.

We have previously developed an antigen-delivery system based on hybrid recombinant porcine parvovirus-like particles (PPV-VLPs) formed by the self-assembly of the VP2 protein of PPV carrying a foreign epitope at its N terminus. In this study, different constructs were made containing a CD8+ T-cell epitope of chicken ovalbumin (OVA) to analyse the influence of the sequence inserted into VP2 on the correct processing of VLPs by antigen-presenting cells. We analysed the presentation of the OVA epitope inserted without flanking sequences or with either different natural flanking sequences or with the natural flanking sequences of a CD8+ T-cell epitope from the lymphocytic choriomeningitis virus nucleoprotein, and as a dimer with or without linker sequences. All constructs were studied in terms of level of expression, assembly of VLPs and ability to deliver the inserted epitope into the MHC I pathway. The presentation of the OVA epitope was considerably improved by insertion of short natural flanking sequences, which indicated the relevance of the flanking sequences on the processing of PPV-VLPs. Only PPV-VLPs carrying two copies of the OVA epitope linked by two glycines were able to be properly processed, suggesting that the introduction of flexible residues between the two consecutive OVA epitopes may be necessary for the correct presentation of these dimers by PPV-VLPs. These results provide information to improve the insertion of epitopes into PPV-VLPs to facilitate their processing and presentation by MHC class I molecules.

The acidic activation domains (AADs) of the baculovirus transactivators IE1 and IE0 are essential for transcriptional transactivation. To compare the relative transcriptional activation potentials of IE1 and IE0 AADs of Autographa californica multiple nucleopolyhedrovirus (AcMNPV) and Orgyia pseudotsugata MNPV (OpMNPV), we constructed two ecdysone receptor (EcR)-based inducible expression systems to analyse six baculovirus AADs in two insect cell lines (Ld652Y and Sf9) and two mammalian cell lines (NIH-3T3 and CHO). For insect cell expression, the AADs were fused to the C, D, E and F domains of the spruce budworm Choristoneura fumiferana EcR. For mammalian cell expression the AADs were fused to the E and F domains of mammalian Mus musculus retinoid X receptor. In Ld652Y and Sf9 cells, chimeric proteins containing the AcMNPV AADs activated gene expression to higher levels than those containing the OpMNPV AADs. In NIH-3T3 cells, chimeras containing AcMNPV IE1 and IE0 AADs consistently activated gene expression to higher levels than the archetypal mammalian herpesvirus VP16 AAD. In contrast, OpMNPV AADs only activated expression by 5–15 % relative to the VP16 AAD. In CHO cells, both AcMNPV and OpMNPV AADs exhibited intermediate transactivation levels relative to VP16 AAD. These results show that the baculovirus AADs are functional for transcriptional activation in mammalian cells and that AcMNPV AADs generally appear to be more potent than OpMNPV AADs in both insect and mammalian cells.

Open reading frame UL12 of herpes simplex virus type 1 (HSV-1) encodes an alkaline nuclease that has previously been implicated in processing the complex, branched, viral DNA replication intermediates and allowing egress of DNA-containing capsids from the nucleus. This report describes experiments using the HSV-1 UL12 null mutant ambUL12, which aim to explain the approximately 200- to 1000-fold decrease in the yield of infectious virus, compared with wild-type (wt) HSV-1, from non-complementing cells. A detailed examination revealed that both DNA replication and encapsidation were affected in ambUL12-infected cells, resulting in an approximately 15- to 20-fold reduction in the amount of packaged DNA. In contrast to previous reports, the absence of UL12 function did not greatly impair capsid release into the cytoplasm, and virus particles were readily detected in the supernatant medium from ambUL12-infected cells. The released virus, however, exhibited much higher particle/p.f.u. ratios than wt HSV-1, and this made a further important contribution to the overall reduction in yield. Gel analyses of packaged ambUL12 and wt DNAs revealed the presence of structural abnormalities. The DNA obtained from extracellular ambUL12 virions was non-infectious in transfection assays, and both ambUL12 DNA and virus particles exerted a dominant inhibitory effect on the growth of wt virus. These results suggest that ambUL12 virions produced in non-complementing cells have a greatly reduced ability to initiate new cycles of infection, and that this defect results from the encapsidation of abnormal genomes.

The prototype strain of bovine adenovirus (BAdV) type 10 and four additional isolates that were indistinguishable in serum-neutralization tests have been shown to have remarkable variation in their genome size and restriction maps. In the present study, more than 40 % of the DNA sequence of the BAdV-10 isolate with the longest genome was determined. A biased base composition resulting in low (<41 %) GC content was noticed. Analysis of the genes of the DNA-binding protein, 100K, 33K, pVIII and fibre proteins, as well as early regions E3 and E4, which are encoded by the genome fragment examined, confirmed that BAdV-10 is different from the other known BAdV types regarding its phylogenetic distance and the organization of its exceptionally short E3 region, apparently containing only two genes. A comparative analysis of the E3 and E4 regions of BAdV-10 with various animal adenoviruses revealed interesting features accounting for the very short genome of BAdV-10. In the examined BAdV-10 isolate, duplicated sequences were localized in and around the fibre gene. Since BAdV-10 appears to be pathogenic to cattle and is genetically distant from the other BAdVs, we suggest that BAdV-10 is not a genuine bovine virus, but has recently switched host and is now undergoing an adaptation process in its new host. In accordance with this hypothesis, the remarkable predominance of AT-rich codons along with the variable fibre gene might be signs of adaptation.

SR protein ASF/SF2 is a general pre-mRNA splicing factor as well as a regulator of alternative splicing. Data presented here show that ASF/SF2 is efficiently recruited to sites in the nucleus where adenovirus genes are transcribed and the resulting pre-mRNAs are processed. At the intermediate stages of a productive infection, ASF/SF2 colocalizes with small nuclear ribonucleoprotein particles (snRNPs), splicing factors in ring-like structures surrounding viral replication centres and, at late stages of the infection, in enlarged speckles. Results presented here demonstrate that ASF/SF2 requires only one of the two RNA-recognition motifs (RRMs) present in the protein for its efficient recruitment to the ring-like structures, where viral pre-mRNAs are transcribed and processed, and that the arginine/serine-rich (RS) domain in ASF/SF2 is both redundant and insufficient for the translocation of the protein to active viral RNA polymerase II genes in adenovirus-infected cells.

A murine gammaherpesvirus (MHV-68) containing a deletion of the putative plasmid maintenance protein ORF73 exhibits a severe latency deficit. In this study the ability of an ORF73 deletion mutant (Δ73) to confer in vivo protection against subsequent challenge with wild-type virus has been examined. Vaccination studies have shown that Δ73 vaccination reduced latent infection of wild-type challenge virus to a level below the limit of detection. These results indicate that a live-attenuated gammaherpesvirus that cannot persist is an effective vaccine.