- Volume 88, Issue 7, 2007

Although the WHO recommends a comprehensive genetic characterization, little is known about circulating strains and genotypes of rubella virus (RUBV) for many European countries. Studies investigating the genetic diversity of a sizeable number of strains from a certain location are rare. This study presents the first molecular characterization of isolates from Belarus. Throat swab and urine samples were collected between 2004 and 2005 from patients presenting in two infectious disease hospitals and three outpatient clinics in and around Minsk. In total, 14 isolates were obtained from this clinical material. Phylogenetic analysis of the E1 gene sequence of these isolates showed that three distinct groups of RUBV strains co-circulated. One group of isolates was assigned to genotype 1E, whereas the other two did not group with any of the recognized genotypes but grouped with a strain belonging to the provisional genotype 1g. Detailed analysis showed that the group comprising 1g strains also contained sequences formerly attributed to genotype 1B and could be divided into four subgroups, one of which might represent a putative novel provisional genotype of clade 1. These findings show that three distinct strains with limited variability are present in Belarus, suggesting independent introductory events. As there currently seem to be misattributions of strains to genotypes and unclear phylogenetic relationships, criteria for genotyping of RUBV should be clarified further.

Re-emergence of Chikungunya (CHIK), caused by CHIK virus, was recorded in India during 2005–2006 after a gap of 32 years, causing 1.3 million cases in 13 states. Several islands of the Indian Ocean reported similar outbreaks in the same period. These outbreaks were attributed to the African genotype of CHIK virus. To examine relatedness of the Indian isolates (IND-06) with Reunion Island isolates (RU), full-genome sequences of five CHIK virus isolates representative of different Indian states were determined. In addition, an isolate obtained from mosquitoes in the year 2000 (Yawat-2000), identified as being of the African genotype, and two older strains isolated in 1963 and 1973 (of the Asian genotype), were sequenced. The IND-06 isolates shared 99.9 % nucleotide identity with RU isolates, confirming involvement of the same strain in these outbreaks. The IND-06 isolates shared 98.2 % identity with the Yawat-2000 isolate. Of two crucial substitutions reported for RU isolates in the E1 region, M269V was noted in the Yawat-2000 and IND-06 isolates, whereas D284E was seen only in the IND-06 isolates. The A226V shift observed with the progression of the epidemic in Reunion Island, probably associated with adaptation to the mosquito vector, was absent in all of the Indian isolates. Three unique substitutions were noted in the IND-06 isolates: two (T128K and T376M) in the Nsp1 region and one (P23S) in the capsid protein. The two Asian strains showed 99.4 % nucleotide identity to each other, indicating relative stability of the virus. No evidence of recombination of the Asian and African genotypes, or of positive selection was observed. The results may help in understanding the association, if any, of the unique mutations with the explosive nature of the CHIK outbreak.

The replicase of Semliki Forest virus (SFV) consists of four non-structural proteins, designated nsP1–4, and is bound to cellular membranes via an amphipathic peptide and palmitoylated cysteine residues of nsP1. It was found that mutations preventing nsP1 palmitoylation also attenuated virus replication. The replacement of these cysteines by alanines, or their deletion, abolished virus viability, possibly due to disruption of interactions between nsP1 and nsP4, which is the catalytic subunit of the replicase. However, during a single infection cycle, the ability of the virus to replicate was restored due to accumulation of second-site mutations in nsP1. These mutations led to the restoration of nsP1–nsP4 interaction, but did not restore the palmitoylation of nsP1. The proteins with palmitoylation-site mutations, as well as those harbouring compensatory mutations in addition to palmitoylation-site mutations, were enzymically active and localized, at least in part, on the plasma membrane of transfected cells. Interestingly, deletion of 7 aa including the palmitoylation site of nsP1 had a relatively mild effect on virus viability and no significant impact on nsP1–nsP4 interaction. Similarly, the change of cysteine to alanine at the palmitoylation site of nsP1 of Sindbis virus had only a mild effect on virus replication. Taken together, these findings indicate that nsP1 palmitoylation as such is not the factor determining the ability to bind to cellular membranes and form a functional replicase complex. Instead, these abilities may be linked to the three-dimensional structure of nsP1 and the capability of nsP1 to interact with other components of the viral replicase complex.

CD8+ T-cell responses are central for the resolution of hepatitis C virus (HCV) infection, and viral escape from these CD8+ T-cell responses has been suggested to play a major role in HCV persistence. However, the factors determining the emergence of CD8 escape mutations are not well understood. Here, the first identification of four HLA-A26-restricted CD8+ T-cell epitopes is reported. Of note, two of these four epitopes are located in the NS3/4A and NS5A/5B cleavage sites. The latter epitope is targeted in all (three of three) patients with acute, resolving HCV infection and in a relatively high proportion (four of 14) of patients with chronic HCV infection. Importantly, the epitope corresponding to the NS5A/5B cleavage site is characterized by the complete absence of sequence variations, despite the presence of functional virus-specific CD8+ T cells in our cohort. These results support previous findings that showed defined functional constraints within this region. They also suggest that the absence of viral escape may be determined by viral fitness cost and highlight an attractive target for immunotherapies.

Hepatitis C virus (HCV) is an important pathogen causing both acute and chronic infections in humans. The HCV polyprotein is synthesized by cap-independent translation initiation after ribosome binding to the highly structured internal ribosome entry site (IRES). The HCV IRES has been shown to have a low requirement for translation initiation factors and the ability to bind directly to the 40S ribosomal subunit. A novel yeast bicistronic reporter system, suitable for sensitive and accurate analysis of IRES activity, has been developed. It employs signal amplification based on the Gal4p transcription factor-mediated activation of a variety of secondary reporter genes. The system has a broad dynamic range and, depending on the nature of the particular secondary reporter, can be used both for precise measurements of IRES activity and for selection and screening for novel IRES variants and IRES trans-acting factors. By using this novel bicistronic system, it was shown that the HCV IRES is functional in yeast cells. Mutational analysis of the IRES loop IV and the adjacent region revealed that, in yeast, as in mammalian cells, translation initiates preferentially at the authentic 342AUG codon and that disruption of the HCV IRES loop IV abrogates its function, whilst minor positional changes or substitutions of the initiation codon within loop IV are largely tolerated. These findings bring more general insights to translation initiation, but also open the door for utilization of yeast and its sophisticated genetics for searching for new antiviral drugs and HCV IRES trans-acting proteins.

The promising potential of RNA interference-based antiviral therapies has been well established. However, the antiviral efficacy is largely limited by genomic diversity and genetic instability of various viruses, including human enterovirus B (HEB). In this work, the first evidence supporting the anti-HEB activity of the small interfering RNA (siRNA) targeting the highly conserved cis-acting replication element (CRE) within virus coding region 2C is presented. HeLa cells pre-treated with siRNA complementary to the conserved sequence of the loop region of CRE(2C) were effectively rescued from the cytopathic effects of HEBs. Downregulation of virus replication and attenuation of cytotoxicity were consistently observed in various reference strains and clinical isolates. Cells treated with this siRNA were resistant to the emergence of viable escape mutants and showed sustained antiviral ability. Collectively, the data suggest that the siRNA based on the disordered structure within the highly conserved cis-acting coding region has potential as a universal, persistent anti-HEB agent. The same strategy can be successfully applied to the development of siRNA with consistent antiviral effects in other virus groups possessing similar RNA elements.

Protein translation in noroviruses requires translational processing of a polyprotein precursor by the viral protease. So far, the molecular mechanisms of catalytic cleavage by the viral protease are poorly understood. In this study, the catalytic activities and substrate specificities of the viral protease were examined in vitro by using synthetic peptides (11–15 residues) corresponding to the cleavage sites of the norovirus polyprotein. Both predicted forms of the viral protease, the 3C-like protease (3Cpro) and the 3CD-like protease polymerase protein (3CDpropol), displayed a specific trans cleavage activity of peptides bearing Gln–Gly at the scissile bond. In contrast, peptides bearing Glu–Gly at the scissile bond (p20/VPg and 3Cpro/3Dpol junctions) were resistant to trans-cleavage by 3Cpro and 3CDpropol. Interestingly, the VPg/3Cpro scissile bond (Glu–Ala) was cleaved only by 3CDpropol, and examination of relative cleavage efficiencies revealed significant differences in processing of peptides, indicating differential cleavage patterns for 3Cpro and 3CDpropol.

The nucleocapsid (N) protein of all morbilliviruses has a highly conserved central region that is thought to interact with and encapsidate the viral RNA. The C-terminal third of the N protein is highly variable among morbilliviruses and is thought to be located on the outer surface and to be available to interact with other viral proteins such as the phosphoprotein, the polymerase protein and the matrix protein. Using reverse genetics, a chimeric rinderpest virus (RPV)/peste-des-petits-ruminants virus (PPRV) was rescued in which the RPV N gene open reading frame had been replaced with that of PPRV (RPV–PPRN). The chimeric virus maintained efficient replication in cell culture. Cattle vaccinated with this chimeric vaccine showed no adverse reaction and were protected from subsequent challenge with wild-type RPV, indicating it to be a safe and efficacious vaccine. The carboxyl-terminal variable region of the rinderpest N protein was cloned and expressed in Escherichia coli. The expressed protein was used to develop an indirect ELISA that could clearly differentiate between RPV- and PPRV-infected animals. The possibility of using this virus as a marker vaccine in association with a new diagnostic ELISA in the rinderpest eradication programme is discussed.

Both rhesus and cynomolgus macaques have been used as animal models for measles vaccination and immunopathogenesis studies. A number of studies have suggested that experimental measles virus (MV) infection induces more-characteristic clinical features in rhesus than in cynomolgus monkeys. In the present study, both macaque species were infected with two different wild-type MV strains and clinical, virological and immunological parameters were compared. The viruses used were a genotype C2 virus isolated in The Netherlands in 1991 (MV-Bil) and a genotype B3 virus isolated from a severe measles case in Sudan in 1997 (MV-Sudan). Following infection, all rhesus monkeys developed a skin rash and conjunctivitis, which were less obvious in cynomolgus monkeys. Fever was either mild or absent in both species. Virus reisolation profiles from peripheral blood mononuclear cells and broncho-alveolar lavage cells and the kinetics of MV-specific IgM and IgG responses were largely identical in the two animal species. However, in animals infected with MV-Sudan, viraemia appeared earlier and lasted longer than in animals infected with MV-Bil. This was also reflected by the earlier appearance of MV-specific serum IgM antibodies after infection with MV-Sudan. Collectively, these data show that cynomolgus and rhesus macaques are equally susceptible to wild-type MV infection, although infection in the skin seems to follow a different course in rhesus macaques. MV-Sudan proved more pathogenic for non-human primates than MV-Bil, which may render it more suitable for use in future pathogenesis studies.

As pigs are susceptible to infection with both avian and human influenza A viruses, they have been proposed to be an intermediate host for the generation of pandemic virus through reassortment. Antigenic and genetic characterization was performed for five swine H9N2 influenza viruses isolated from diseased pigs from different farms. The haemagglutinin (HA) antigenicity of swine H9N2 viruses was different from that of chicken H9N2 viruses prevalent in northern China. Genetic analysis revealed that all five isolates had an RLSR motif at the cleavage site of HA, which was different from those of A/duck/Hong Kong/Y280/97 (Dk/HK/Y280/97)-like viruses established in chickens in China. Phylogenetic analyses indicated that the five swine H9N2 viruses formed novel HA and neuraminidase sublineages that were related closely to those of earlier chicken H9 viruses and were also consistent with the extent of the observed antigenic variation. The six internal genes of the isolates possessed H5N1-like sequences, indicating that they were reassortants of H9 and H5 viruses. The present results indicate that avian to porcine interspecies transmission of H9N2 viruses might have resulted in the generation of viruses with novel antigenic and genetic characteristics; therefore, surveillance of swine influenza should be given a high priority.

Vesicular stomatitis (VS) outbreaks of unknown origin occur at 8–10-year intervals in the south-western USA with the most recent outbreak beginning in 2004. A previous study has suggested that strains causing US outbreaks are closely related to strains causing outbreaks in Mexico [Rodriguez (2002) Virus Res 85, 211–219]. This study determined the phylogenetic relationships among 116 vesicular stomatitis New Jersey virus (VSNJV) strains obtained from the 2004 outbreak and from endemic areas in Mexico. All 69 US viruses showed little sequence divergence (≤1.3 %), regardless of their location or time of collection, and clustered with 11 Mexican viruses into a genetic lineage not previously present in the USA. Furthermore, viruses with identical phosphoprotein hypervariable region sequences to those causing the US outbreaks in 1995–1997 and 2004–2005 were found circulating in Mexico between 2002 and 2004. Molecular adaptation analysis provided evidence for positive selection in the phosphoprotein and glycoprotein genes during a south-to-north migration among 69 US viruses collected between the spring and autumn of 2004 and 2005. Phylogenetic data, temporal–spatial distribution and the finding of viral strains identical to those causing major outbreaks in the USA circulating in Mexico demonstrated that VS outbreaks in the south-western USA are the result of the introduction of viral strains from endemic areas in Mexico.

Bovine leukemia virus (BLV) is a valuable model system for understanding human T-lymphotropic virus 1 (HTLV-1); the availability of an infectious BLV clone, together with animal-model systems, will help to explore anti-HTLV-1 strategies. Nevertheless, the specificity and inhibitor sensitivity of the BLV protease (PR) have not been characterized in detail. To facilitate such studies, a molecular model for the enzyme was built. The specificity of the BLV PR was studied with a set of oligopeptides representing naturally occurring cleavage sites in various retroviruses. Unlike HTLV-1 PR, but similar to the human immunodeficiency virus 1 (HIV-1) enzyme, BLV PR was able to hydrolyse the majority of the peptides, mostly at the same position as did their respective host PRs, indicating a broad specificity. When amino acid residues of the BLV PR substrate-binding sites were replaced by equivalent ones of the HIV-1 PR, many substitutions resulted in inactive protein, indicating a great sensitivity to mutations, as observed previously for the HTLV-1 PR. The specificity of the enzyme was studied further by using a series of peptides containing amino acid substitutions in a sequence representing a naturally occurring HTLV-1 PR cleavage site. Also, inhibitors of HIV-1 PR, HTLV-1 PR and other retroviral proteases were tested on the BLV PR. Interestingly, the BLV PR was more susceptible than the HTLV-1 PR to the inhibitors tested. Therefore, despite the specificity differences, in terms of mutation intolerance and inhibitor susceptibility of the PR, BLV and the corresponding animal-model systems may provide good models for testing of PR inhibitors that target HTLV-1.

Rotavirus non-structural protein (NSP) 4 can induce aqueous secretion in the gastrointestinal tract of neonatal mice through activation of an age- and Ca2+-dependent plasma membrane anion permeability. Accumulating evidence suggests that nitric oxide (NO) plays a role in the modulation of aqueous secretion and the barrier function of intestinal cells. This study investigated transcriptional changes in inducible NO synthase (iNOS), an enzyme responsible for NO production, after rotavirus infection in mice and after treatment of intestinal cells with NSP4. Diarrhoea was observed in 5-day-old CD-1 mice from days 1 to 3 after inoculation with 107 focus-forming units of different rotavirus strains. Ileal iNOS mRNA expression was induced as early as 6 h post-inoculation, before the onset of clinical diarrhoea in infected mice, and was upregulated during the course of rotavirus-induced diarrhoea. Ex vivo treatment of ilea excised from CD-1 suckling mice with NSP4 resulted in upregulation of ileal iNOS mRNA expression within 4 h. Furthermore, NSP4 was able to induce iNOS expression and NO production in murine peritoneal macrophages and RAW264.7 cells. The specificity of NSP4 inducibility was confirmed by the inhibitory effect of anti-NSP4 serum. Using a series of truncated NSP4s, the domain responsible for iNOS induction in macrophages was mapped to the reported enterotoxin domain, aa 109–135. Thus, rotavirus infection induces ileal iNOS expression in vivo and rotavirus NSP4 also induces iNOS expression in the ileum and macrophages. Together, these findings suggest that NO plays a role in rotavirus-induced diarrhoea.

Glycoprotein B (gB) is the most conserved glycoprotein among herpesviruses and it plays important roles in virus infectivity. In most herpesviruses, including pseudorabies virus (PRV), gB is cleaved by a cellular protease into two disulfide-linked subunits. In the present study, I found that the PRV gB generated in human colon carcinoma LoVo cells, which lack the ubiquitous protease furin, remained in the uncleaved form and the virus replicated in these cells without cell fusion. The uncleaved gB was converted into its subunits after furin digestion. The virus also replicated in Madin–Darby bovine kidney cells without cell fusion in the presence of a furin inhibitor, whereas distinct syncytia were formed in the absence of the inhibitor. LoVo cells constitutively expressing furin showed cell fusion when they were infected with the virus. Penetration kinetics assays revealed that the virus carrying uncleaved gB penetrated the cells at the same rate as the virus carrying cleaved gB. These results indicate that PRV gB is cleaved by furin and that the cleavage is dispensable for virus replication in vitro. Furthermore, gB cleavage is involved in syncytium formation but not in penetration kinetics, suggesting that different mechanisms operate between cell–cell fusion and virus–cell fusion by PRV.

Apoptosis has recently been associated with herpes simplex virus 1 (HSV-1) latency and disease severity. There is an intricate balance between pro- and anti-apoptotic processes during HSV-1 infection. When anti-apoptotic pathways are suppressed, this balance is upset and the cells die by apoptosis, referred to here as HSV-1-dependent apoptosis (HDAP). It has been observed previously that HeLa cancer cells exhibit an enhanced sensitivity to HDAP. Here, a series of specific patient-derived cancer cells was utilized to investigate the cell-type specificity of HDAP. The results showed that a human mammary tumour cell line was sensitive to HDAP, whilst syngeneic normal cells were resistant. Furthermore, low-passage-number primary human mammary epithelial cells were resistant to HDAP. When the susceptibility of human colon, brain, breast and cervical cancer cells was assessed, the only cells insensitive to HDAP were those resistant to all environmental stimuli tested. This implies that the HDAP resistance was probably due to mutations in the cellular apoptotic machinery. Thus, the susceptibility of cancer cells to HDAP requires that they possess a functional ability to undergo programmed cell death.

Since ‘constitutive activation’ of STAT1 was first described in Epstein–Barr virus (EBV)-immortalized lymphoblastoid cell lines (LCLs), there has been controversy regarding the molecular identity of the STAT1 DNA-binding complex found in these cells. The post-translational modifications of STAT1 in LCLs have been analysed and an LMP1-induced STAT1 DNA-binding complex, different from that generated by alpha interferon (IFN) stimulation and not involving tyrosine phosphorylation, is demonstrated. STAT1 is serine-phosphorylated downstream of PI3K and MEK in LCLs and this modification restricts IFN-stimulated STAT1–DNA binding. These data suggest that EBV induces a distinct form of DNA-bound STAT1 in virus-infected cells.

The Epstein–Barr virus (EBV)-encoded tumour-associated latent membrane protein 1 (LMP1) gene expression is transactivated by EBV nuclear antigen 2 (EBNA2) in human B cells. We have previously identified a cyclic AMP-responsive element (CRE) in the B95-8 LMP1 promoter that is essential for transcription activation. Sequencing of LMP1 promoter in the P3HR1-derived EREB2.5 cell line revealed 25 single base pair substitutions in comparison to the B95-8 virus, one of them localized in the CRE element. Sequence variations in this element have been identified in several EBV isolates of both African and Asian origins. The effect of the P3HR1 CRE site variation on binding of factors to the LMP1 promoter sequence (LRS) and promoter activation was investigated with electrophoreticmobility-shift assays and reporter gene transfection assays. ATF1 and CREB1 transcription factors bound with reduced efficiency to the P3HR1 variant and below the detection level to the other tested variants. Accordingly, reporter plasmids carrying the P3HR1 CRE sequence in a B95-8 LRS context displayed 50 % lower activity in all tested cell lines. The impaired ability to activate transcription caused by the C to A substitution in CRE was not apparent when the mutated site was placed in a P3HR1 LRS context and the reporter transfected into Jijoye cells, most likely as a consequence of the other base pair substitutions in P3HR1 LRS. Overall, our results suggest that the mutations in the LRS CRE site have been conserved to adjust LMP1 expression to levels that favour cell survival in certain cellular and environmental contexts.

Many viruses have evolved mechanisms to evade host immunity by subverting the function of dendritic cells (DCs). This study determined whether murine gammaherpesvirus-68 (γHV-68) could infect immature or mature bone-marrow-derived DCs and what effect infection had on DC maturation. It was found that γHV-68 productively infected immature DCs, as evidenced by increased viral titres over time. If DCs were induced to mature by exposure to LPS and then infected with γHV-68, only a small percentage of cells was productively infected. However, limiting-dilution assays to measure viral reactivation demonstrated that the mature DCs were latently infected with γHV-68. Electron microscopy revealed the presence of capsids in the nucleus of immature DCs but not in mature DCs. Interestingly, infection of immature DCs by γHV-68 did not result in upregulation of the co-stimulatory molecules CD80 and CD86 or MHC class I and II, or induce cell migration, suggesting that the virus infection did not induce DC maturation. Furthermore, γHV-68 infection of immature DCs did not result in elevated interleukin-12, an important cytokine in the induction of T-cell responses. Finally, lipopolysaccharide and poly(I : C) stimulation of γHV-68-infected immature DCs did not induce increases in the expression of co-stimulatory molecules and MHC class I or II compared with mock-treated cells, suggesting that γHV-68 infection blocked maturation. Taken together, these data demonstrate that γHV-68 infection of DCs differs depending on the maturation state of the DC. Moreover, the block in DC maturation suggests a possible immunoevasion strategy by γHV-68.

Since 1980 there has been global eradication of smallpox due to the success of the vaccination programme using vaccinia virus (VACV). During the eradication period, distinct VACV strains circulated, the Lister strain being the most commonly employed in Europe. Analysis of the safety of smallpox vaccines has suggested that they display significant heterogeneity. To gain a more detailed understanding of the diversity of VACV strains it is important to determine their genomic sequences. Although the sequences of three isolates of the Japanese Lister original strain (VACV-LO) are available, no analysis of the relationship of any Lister sequence compared to other VACV genomes has been reported. Here, we describe the sequence of a representative clonal isolate of the Lister vaccine (VACV-List) used to inoculate the French population. The coding capacity of VACV-List was compared to other VACV strains. The 201 open reading frames (ORFs) were annotated in the VACV-List genome based on protein size, genomic localization and prior characterization of many ORFs. Eleven ORFs were recognized as pseudogenes as they were truncated or fragmented counterparts of larger ORFs in other orthopoxviruses (OPVs). The VACV-List genome also contains several ORFs that have not been annotated in other VACVs but were found in other OPVs. VACV-List and VACV-LO displayed a high level of nucleotide sequence similarity. Compared to the Copenhagen strain of VACV, the VACV-List sequence diverged in three main regions, one of them corresponding to a substitution in VACV-List with coxpox virus GRI-90 strain ORFs, suggestive of prior genetic exchanges. These studies highlight the heterogeneity between VACV strains and provide a basis to better understand differences in safety and efficacy of smallpox vaccines.