- Volume 86, Issue 12, 2005

The pathogenesis of coxsackie B virus (CVB) infections is generally studied in mice by intraperitoneal (i.p.) injection, whereas the gastrointestinal tract is the natural porte d'entrée in humans. The present study was undertaken to compare systematically the influence of infection route on morbidity and pathology. Swiss Albino mice were infected with CVB3 (Nancy) at different doses (5×103, 5×105, 5×107, 5×109 TCID50), given either i.p. or orally. Virus could be isolated from several organs (heart, spleen and pancreas), indicating systemic infection, irrespective of the infection route. Virus titres were 1–2 logs higher after i.p. infection, but kinetics were largely independent of infection route. Organs became negative for virus isolation after 21 days, with the exception of spleen tissue, which remained positive for up to 49 days. Thereafter, virus was detected only by immunohistochemistry and PCR up to 98 days post-infection (oral route). Histopathology showed mild inflammation and necrosis in heart tissue of all mice during the acute phase, with repair at later stages. Strikingly, pancreatic lesions were confined to the exocrine pancreas and observed only after i.p. infection. Under all experimental conditions, the pancreatic islets were spared. In contrast, immunohistochemistry showed the presence of viral VP1, protein 3A and alpha interferon (IFN-α) in exocrine as well as endocrine pancreas of all mice, irrespective of route and dose of infection. It is concluded that infection via the oral route protects the pancreas from damage, but not from infection, a process in which IFN-α is not the only factor involved.

The complete nucleotide sequences of eight Human enterovirus B (HEV-B) strains were determined, representing five serotypes, E6, E7, E11, CVB3 and CVB5, which were isolated in the former Soviet Union between 1998 and 2002. All strains were mosaic recombinants and only the VP2–VP3–VP1 genome region was similar to that of the corresponding prototype HEV-B strains. In seven of the eight strains studied, the 2C–3D genome region was most similar to the prototype E30, EV74 and EV75 strains, whilst the remaining strain was most similar to the prototype E1 and E9 strains in the non-structural protein genome region. Most viruses also bore marks of additional recombination events in this part of the genome. In the 5′ non-translated region, all strains were more similar to the prototype E9 than to other enteroviruses. In most cases, recombination mapped to the VP4 and 2ABC genome regions. This, together with the star-like topology of the phylogenetic trees for these genome regions, identified these genome parts as recombination hot spots. These findings further support the concept of independent evolution of enterovirus genome fragments and indicate a requirement for more advanced typing approaches. A range of available phylogenetic methods was also compared for efficient detection of recombination in enteroviruses.

Hepatitis C virus (HCV) infection is associated with inflammation of liver endothelium, which contributes to the pathogenesis of chronic hepatitis. The mechanism of this endothelitis is not understood, since the virus does not appear to infect endothelial cells productively. Here, an ‘innocent bystander’ mechanism related to HCV proteins was hypothesized and it was investigated whether the binding of HCV particles to human endothelium induced functional changes in the cells. Exposure of human umbilical vein endothelial cells (HUVECs) to HCV-like particles (HCV-LPs) resulted in increased interleukin 8 (IL8) production and induction of apoptosis. The IL8 supernatants collected after stimulation of HUVECs with HCV-LPs, BV-GUS (control baculovirus containing β-glucuronidase) and appropriate controls were used to assay the transendothelial migration of neutrophils. This assay confirmed that HCV-LP-induced IL8 was functionally active. Using specific NF-κB inhibitors, it was also shown that HCV-LP-induced NF-κB activity mediated IL8 production in HUVECs. Apoptosis appeared to be mediated by the Fas/Fas-L pathway, as neutralizing antibodies for Fas and Fas-L significantly protected HUVECs against HCV-LP-induced apoptosis. Treatment of HUVECs with HCV-LPs also enhanced cellular Fas-L expression and augmented caspase-3 activation. This was confirmed by using a specific caspase-3 inhibitor, Z-Asp-Glu-Val-Asp-fluoromethyl ketone. As shown by blocking of specific chemokine receptors for IL8 on HUVECs, the induction of IL8 did not appear to contribute to HCV-LP-induced apoptosis. These results suggest that HCV proteins can trigger the release of inflammatory chemokines such as IL8 and cause endothelial apoptosis, thereby facilitating endothelitis.

The coronavirus nucleocapsid (N) protein is a viral RNA-binding protein with multiple functions in terms of virus replication and modulating cell signalling pathways. N protein is composed of three distinct regions containing RNA-binding motif(s), and appropriate signals for modulating cell signalling. The subcellular localization of severe acute respiratory syndrome coronavirus (SARS-CoV) N protein was studied. In infected cells, SARS-CoV N protein localized exclusively to the cytoplasm. In contrast to the avian coronavirus N protein, overexpressed SARS-CoV N protein remained principally localized to the cytoplasm, with very few cells exhibiting nucleolar localization. Bioinformatic analysis and deletion mutagenesis coupled to confocal microscopy and live-cell imaging, revealed that SARS-CoV N protein regions I and III contained nuclear localization signals and region II contained a nucleolar retention signal. However, cytoplasmic localization was directed by region III and was the dominant localization signal in the protein.

Recently, class II fusion proteins have been identified on the surface of alpha- and flaviviruses. These proteins have two functions besides membrane fusion: they generate an isometric lattice on the viral surface and they form ion-permeable pores at low pH. An attempt was made to identify inhibitors for the ion pores generated by the fusion proteins of the alphaviruses Semliki Forest virus and Sindbis virus. These pores can be detected and analysed in three situations: (i) in the target membrane during virus entry, by performing patch-clamp measurements of membrane currents; (ii) in the virus particle, by studying the entry of propidium iodide; and (iii) in the plasma membrane of infected cells, by Fura-2 fluorescence imaging of Ca2+ entry into infected cells. It is shown here that, at a concentration of 0·1 mM, rare earth ions block the ion permeability of alphavirus ion pores in all three situations. Even at a concentration of 0·5 mM, these ions do not block formation of the viral fusion pore, as they do not inhibit entry or multiplication of alphaviruses. The data indicate that ions flow through the ion pores into the virus particle in the endosome and from the endosome into the cytoplasm after fusion of the viral envelope with the endosomal membrane. These ion flows, however, are not necessary for productive infection. The possibility that the ability of class II fusion proteins to form ion-permeable pores reflects their origin from protein toxins that form ion-permeable pores, and that entry via class II fusion proteins may resemble the entry of non-enveloped viruses, is discussed.

Two (2·3 %) of 87 wild-caught boars in Japan had detectable hepatitis E virus (HEV) RNA. The two boar HEV isolates (wbJTS1 and wbJYG1) obtained in the present study and a previously reported isolate (wbJSG1) whose partial sequence had been determined were sequenced over the entire genome. The wbJSG1, wbJTS1 and wbJYG1 isolates comprised 7225 or 7226 nt, excluding the poly(A) tail, and segregated into genotype 3. They differed by 8·5–11·2 % from each other and by 8·6–18·4 % from 17 reported genotype 3 HEV isolates, including one boar isolate, in the full-length sequence. When compared with 191 reported genotype 3 HEV isolates whose partial sequences were known, these three boar isolates were closer to Japanese isolates than to isolates of non-Japanese origin (89·2±2·6 vs 85·9±2·2 %; P<0·0001). A proportion of wild boars in Japan are infected with markedly heterogeneous HEV strains that are indigenous to Japan and may serve as reservoirs of HEV.

Crimean-Congo hemorrhagic fever virus (CCHFV), a member of the genus Nairovirus of the family Bunyaviridae, causes severe disease in humans with high rates of mortality. The virus has a tripartite genome composed of a small (S), a medium (M) and a large (L) RNA segment; the M segment encodes the two viral glycoproteins, GN and GC. Whilst relatively few full-length M segment sequences are available, it is apparent that both GN and GC may exhibit significant sequence diversity. It is unknown whether considerable antigenic differences exist between divergent CCHFV strains, or whether there are conserved neutralizing epitopes. The M segments derived from viral isolates of a human case of CCHF in South Africa (SPU 41/84), an infected tick (Hyalomma marginatum) in South Africa (SPU 128/81), a human case in Congo (UG 3010), an infected individual in Uzbekistan (U2-2-002) and an infected tick (Hyalomma asiaticum) in China (Hy13) were sequenced fully, and the glycoproteins were expressed. These novel sequences showed high variability in the N-terminal region of GN and more modest differences in the remainder of GN and in GC. Phylogenetic analyses placed these newly identified strains in three of the four previously described M segment groups. Studies with a panel of mAbs specific to GN and GC indicated that there were significant antigenic differences between the M segment groups, although several neutralizing epitopes in both GN and GC were conserved among all strains examined. Thus, the genetic diversity exhibited by CCHFV strains results in significant antigenic differences that will need to be taken into consideration for vaccine development.

The genus Nairovirus of the family Bunyaviridae includes the Crimean-Congo haemorrhagic fever (CCHF) species group. The species is predominated by the hazard-group 4 pathogens, from which the name and majority of strain entries are derived. Additionally, the species embraces hazard-group 2 viruses that are classified as members by antigenic cross-reactivity. CCHF viruses have a tripartite RNA genome consisting of large (L), medium (M) and small (S) segments. Here, the sequence characterization of previously undescribed L and S segments from novel strains originating in the Middle East and Africa is reported. Further scrutiny of this data with phylogenetic tools, in the context of other publicly available sequence information, reveals analogous grouping patterns between the L and S segments. These groups correlate with the geographical distribution of strain isolation and indicate that the L and S segments of CCHF viruses have evolved together.

Seven of the eight genes in the avian pneumovirus (APV) genome contain a conserved 9 nt transcriptional start sequence with the virus large (L) polymerase gene differing from the consensus at three positions. The sequence requirements of the APV transcriptional gene start sequence were investigated by generating a series of mutations in which each of the nine conserved bases was mutated to each of the other three possible nucleotides in a minigenome containing two reporter genes. The effect of each mutation was assessed by measuring the relative levels of expression from the altered and unaltered gene start sequences. Mutations at positions 2, 7 and 9 significantly reduced transcription levels while alterations to position 5 had little effect. The L gene start sequence directed transcription at levels approximately 50 % below that of the consensus gene start sequence. These data suggest that there are common features in pneumovirus transcriptional control sequences.

Rinderpest, or cattle plague, is caused by Rinderpest virus (RPV), which is related most closely to human Measles virus (MV), both being members of the genus Morbillivirus, a group of viruses known to have strong immunosuppressive effects in vitro and in vivo. Here, it was shown that peripheral blood mononuclear cells (PBMCs) isolated from cattle experimentally infected with either wild-type or vaccine strains of RPV impaired the proliferation of PBMCs derived from uninfected animals; however, in contrast to either mild or virulent strains of wild-type virus, the inhibition induced by the vaccine was both weak and transient. Flow-cytometric analysis of PBMCs obtained from cattle infected with different strains of RPV showed that the proportion of infected cells was virus dose-dependent and correlated with lymphoproliferative suppression.

The matrix (M) protein of vesicular stomatitis virus plays a key role in both assembly and budding of progeny virions. In vitro experiments have shown a strong propensity of M protein to bind to vesicles containing negatively charged phospholipids. In vivo, it has also been demonstrated that recruitment of some cellular proteins by M protein is required for efficient virus budding and release of newly synthesized virions in the extracellular medium. The ability of M protein to deform target membranes in vitro was investigated in this study. It was shown that incubation of purified M protein with giant unilamellar vesicles results in the formation of patches of M protein at their surface, followed by deformations of the membrane toward the inside of the vesicle, which could be observed in phase-contrast microscopy. This provides the first evidence that M protein alone is able to impose the correct budding curvature on the membrane. Using confocal microscopy, patches of M protein that colocalized with negatively charged lipid domains a few minutes after vesicle injection were observed. After a longer incubation period, membrane deformations appeared in these domains. At this time, a strict colocalization of M protein, negatively charged lipids and membrane deformation was observed. The influence on this process of the basic N-terminal part of the protein and of the previously identified hydrophobic loop has also been investigated. Interestingly, the final fission event has never been observed in our experimental system, indicating that other partners are required for this step.

Most retroviruses induce severe immunosuppression during acute infection. We have used the Friend retrovirus mouse model to demonstrate that immunostimulatory B-type CpG oligodeoxynucleotides (ODN) have a protective effect against retrovirus-induced suppression of antibody responses to potent B-cell antigens. CD8+ T cells were critical for effective treatment with CpG-ODN, since in vivo depletion of these cells from treated mice impaired protection from retrovirus-induced immunosuppression. Protection also required IFN-γ, as neutralization of this cytokine abolished the therapeutic effect of CpG-ODN. These findings may have implications for the treatment of immunosuppressive virus infections.

In most stages of human immunodeficiency virus (HIV) infection, cell-free viral particles can be detected in germinal centres (GCs) that are principally retained, in the form of immune complexes, on the surface of follicular dendritic cells (FDCs). The source of this virus remains unknown, although it is agreed that the FDCs themselves are not infected productively. By sequencing HIV viral DNA, genomic RNA and spliced mRNA isolated from individual splenic white pulps, it was shown here that the majority of HIV-1 viral particles are produced locally within the supporting lymphoid structure and do not result from trapping of circulating viruses or immune complexes. These findings underline the exquisite spatial organization of HIV-1 replication in vivo, suggesting a local origin for viruses trapped in splenic GCs.

In vivo depletion of CD8+ T cells results in an increase in viral load in macaques chronically infected with simian immunodeficiency virus (SIVmac239Δnef). Here, the cellular and humoral immune responses associated with this transient period of enhanced viraemia in macaques infected with SIVmac239Δnef were characterized. Fourteen days after in vivo CD8+ T-cell depletion, two of six macaques experienced a 1–2 log10 increase in anti-gp130 and p27 antibody titres and a three- to fivefold increase in gamma interferon-secreting SIV-specific CD8+ T cells. Three other macaques had modest or no increase in anti-gp130 antibodies and significantly lower titres of anti-p27 antibodies, with minimal induction of functional CD8+ T cells. Four of the five CD8-depleted macaques experienced an increase in neutralizing antibody titres to SIVmac239. Induction of SIV-specific immune responses was associated with increases in CD8+ T-cell proliferation and fluctuations in the levels of signal-joint T-cell receptor excision circles in peripheral blood cells. Five months after CD8+ T-cell depletion, only the two high-responding macaques were protected from intravenous challenge with pathogenic SIV, whilst the remaining animals were unable to control replication of the challenge virus. Together, these findings suggest that a transient period of enhanced antigenaemia during chronic SIV infection may serve to augment virus-specific immunity in some, but not all, macaques. These findings have relevance for induction of human immunodeficiency virus (HIV)-specific immune responses during prophylactic and therapeutic vaccination and for immunological evaluation of structured treatment interruptions in patients chronically infected with HIV-1.

To investigate why human immunodeficiency virus type 2 (HIV-2) is less virulent than HIV-1, the evolution of coreceptor usage, autologous neutralization, envelope sequence and glycosylation was studied in sequentially obtained virus isolates and sera from four HIV-2-infected individuals. Neutralization of primary HIV-2 isolates was tested by a cell line-based assay and IgG purified from patients' sera. Significant autologous neutralization was observed for the majority (39 of 54) of the HIV-2 serum–virus combinations tested, indicating that neutralization escape is rare in HIV-2 infection. Furthermore, sera from 18 HIV-2 patients displayed extensive heterologous cross-neutralization when tested against a panel of six primary HIV-2 isolates. This indicates that HIV-2 is intrinsically more sensitive to antibody neutralization than HIV-1. In line with earlier reports, HIV-2 isolates could use several alternative receptors in addition to the major coreceptors CCR5 and CXCR4. Intrapatient evolution from CCR5 use to CXCR4 use was documented for the first time. Furthermore, CXCR4 use was linked to the immunological status of the patients. Thus, all CXCR4-using isolates, except one, were obtained from patients with CD4 counts below 200 cells μl−1. Sequence analysis revealed an association between coreceptor usage and charge of the V3 loop of the HIV-2 envelope, as well as an association between the rate of disease progression and the glycosylation pattern of the envelope protein. Furthermore, HIV-2 isolates had fewer glycosylation sites in the V3 domain than HIV-1 (two to three versus four to five). It is proposed here that HIV-2 has a more open and accessible V3 domain than HIV-1, due to differences in glycan packing, and that this may explain its broader coreceptor usage and greater sensitivity to neutralizing antibodies.

Group A rotaviruses are major intestinal pathogens that express potential α4β1 and α4β7 integrin ligand sequences Leu–Asp–Val and Leu–Asp–Ile in their outer capsid protein VP7, and Ile–Asp–Ala in their spike protein VP4. Monkey rotavirus SA11 can use recombinant α4β1 as a cellular receptor. In this study a new potential α4β1, α4β7 and α9β1 integrin ligand sequence, Tyr–Gly–Leu, was identified in VP4. It was shown that several human and monkey rotaviruses bound α4β1 and α4β7, but not α9β1. Binding to α4β1 mediated the infectivity and growth of monkey rotaviruses, and binding to α4β7 mediated their infectivity. A porcine rotavirus interacted with α4 integrins at a post-binding stage to facilitate infection. Activation of α4β1 increased rotavirus infectivity. Cellular treatment with peptides containing the α4 integrin ligand sequences Tyr–Gly–Leu and Ile–Asp–Ala eliminated virus binding to α4 integrins and infectivity. In contrast, rotavirus recognition of α4 integrins was unaffected by a peptide containing the sequence Leu–Asp–Val or by a mutation in the VP7 Leu–Asp–Val sequence. VP4 involvement in rotavirus recognition of α4β1 was demonstrated with rotavirus reassortants. Swapping and point mutagenesis of α4 surface loops showed that rotaviruses required the same α4 residues and domains for binding as the natural α4 integrin ligands: mucosal addressin cell adhesion molecule-1, fibronectin and vascular cell adhesion molecule-1. Several rotaviruses are able to use α4β7 and α4β1 for cell binding or entry, through the recognition of the same α4-subunit domains as natural α4 ligands.

An orbivirus designated Yunnan orbivirus (YUOV) was isolated from Culex tritaeniorhynchus mosquitoes collected in the Yunnan province of China. Electron microscopy showed particles with typical orbivirus morphology. The YUOV genome was sequenced completely and compared with previously characterized orbivirus genomes. Significant identity scores were detected between proteins encoded by the segments (Seg-1 to Seg-10) of YUOV and those encoded by their homologues in insect-borne and tick-borne orbiviruses. Analysis of VP1 (Pol) and VP2 (T2, which correlates with the virus serogroup) indicated that YUOV is a new species of the genus Orbivirus that is unrelated to the other insect-borne orbiviruses. The replication of YUOV in mosquito cell lines was restricted to Aedes albopictus cells and the virus failed to replicate in mammalian cell lines. However, intraperitoneal injection of virus into naïve mice resulted in productive, non-lethal virus replication and viraemia. Infected mice developed serum neutralizing antibodies and were protected against a new infection challenge. Sequence analysis of clones from the segments encoding outer coat proteins (Seg-3 and Seg-6) of YUOV recovered from mouse blood did not show significant changes in the sequences. The availability of the complete genome sequence will facilitate the development of sequence-specific PCR assays for the study of YUOV epidemiology in the field.

Deformed wing virus (DWV) is a honeybee viral pathogen either persisting as an inapparent infection or resulting in wing deformity. The occurrence of deformity is associated with the transmission of DWV through Varroa destructor during pupal stages. Such infections with DWV add to the pathology of V. destructor and play a major role in colony collapse in the course of varroosis. Using a recently developed RT-PCR protocol for the detection of DWV, individual bees and mites originating from hives differing in Varroa infestation levels and the occurrence of crippled bees were analysed. It was found that 100 % of both crippled and asymptomatic bees were positive for DWV. However, a significant difference in the spatial distribution of DWV between asymptomatic and crippled bees could be demonstrated: when analysing head, thorax and abdomen of crippled bees, all body parts were always strongly positive for viral sequences. In contrast, for asymptomatic bees viral sequences could be detected in RNA extracted from the thorax and/or abdomen but never in RNA extracted from the head. DWV replication was demonstrated in almost all DWV-positive body parts of infected bees. Analysing individual mites for the presence of DWV revealed that the percentage of DWV-positive mites differed between mite populations. In addition, it was demonstrated that DWV was able to replicate in some but not all mites. Interestingly, virus replication in mites was correlated with wing deformity. DWV was also detected in the larval food, implicating that in addition to transmission by V. destructor DWV is also transmitted by feeding.

Canine adenovirus type 2 (CAV2) has become an attractive vector for gene therapy because of its non-pathogenicity and the lack of pre-existing neutralizing antibodies against this virus in the human population. Additionally, this vector has been proposed as a conditionally replicative adenovirus agent under the control of an osteocalcin promoter for evaluation in a syngeneic, immunocompetent canine model with spontaneous osteosarcoma. In this study, a CAV2 vector labelled with the fluorescent capsid fusion protein IX–enhanced green fluorescent protein (pIX–EGFP) was developed. Expression of the fluorescent fusion-protein label in infected cells with proper nuclear localization, and incorporation into virions, could be detected. The labelled virions could be visualized by fluorescence microscopy; this was applicable to the tracking of CAV2 infection, as well as localizing the distribution of the vector in tissues. Expression of pIX–EGFP could be exploited to detect the replication and spread of CAV2. These results indicate that pIX can serve as a platform for incorporation of heterologous proteins in the context of a canine adenovirus xenotype. It is believed that capsid-labelled CAV2 has utility for vector-development studies and for monitoring CAV2-based oncolytic adenovirus replication.