- Volume 84, Issue 8, 2003

The genomes of umbraviruses differ from those of most other viruses in that they do not encode a coat protein, and thus no virus particles are formed in infected plants. Protection of umbraviral RNA outside the host plant, during vector transmission, utilizes the coat protein of an assistor luteovirus, but this review focuses on the mechanisms that compensate for the lack of a coat protein in processes within the host plant. As well as an RNA-dependent RNA polymerase, umbravirus genomes encode two other proteins from almost completely overlapping open reading frames. One of these is a cell-to-cell movement protein that can mediate the transport of homologous and heterologous viral RNAs through plasmodesmata without the participation of a coat protein. The other, the ORF3 protein, binds to viral RNA to form filamentous ribonucleoprotein particles that have elements of helical structure. It serves to stabilize the RNA and facilitates its transport through the vascular system of the plant. It may also be involved in protection of the viral RNA from the plant's defensive RNA-silencing response, although it is not a suppressor of silencing. The ORF3 protein also enters the cell nucleus, specifically targeting the nucleolus. Although the function of this localization is unknown, the ORF3 protein may provide a valuable tool for investigating plant nucleolar function.

After passive transfer of measles virus (MV)-specific antibodies, vaccine-induced seroconversion and subsequent protection is inhibited in cotton rats (Sigmodon hispidus). In this system, an attenuated, recombinant vesicular stomatitis virus expressing the MV haemagglutinin (VSV-H) was found previously to induce neutralizing antibodies and protection against MV challenge after intranasal (i.n.) immunization. Here it is demonstrated that, after i.n. immunization, VSV-H is found in both lung and brain tissue in the absence of clinical signs. Intratracheal inoculation, which does not lead to infection of the brain, proved that immunization via the lung mucosa is sufficient to protect. To reduce or eliminate infection of the brain after i.n. inoculation, stepwise-attenuated VSV-H mutants with truncated cytoplasmic tails of the G protein were tested in cotton rats. A mutant with 9 aa in the G cytoplasmic tail was found at much lower levels in the brain and was protective in the absence or presence of MV-specific antibodies. A more attenuated mutant containing only 1 aa in its tail was not found in brain tissue after inoculation, but it still induced protective antibody to measles in the absence of MV-specific antibody. However, its ability to induce MV-neutralizing antibodies in the presence of passively transferred MV-specific antibodies and its protective capacity was abolished unless higher-dose immunizations were used. This study demonstrates that a lower degree of attenuation is required to be able to immunize in the presence of MV-specific antibodies.

Parainfluenza virus type 3 (PIV3) and respiratory syncytial virus (RSV) are the main causes of ubiquitous acute respiratory diseases of infancy and early childhood, causing 20–25 % of pneumonia and 45–50 % of bronchiolitis in hospitalized children. The primary goal of this study was to create an effective and safe RSV vaccine based on utilizing attenuated bovine PIV3 (bPIV3) as a virus vector backbone. bPIV3 had been evaluated in human clinical trials and was shown to be attenuated and immunogenic in children as young as 2 months of age. The ability of bPIV3 to function as a virus vaccine vector was explored further by introducing the RSV attachment (G) and fusion (F) genes into the bPIV3 RNA genome. The resulting virus, bPIV3/RSV(I), contained an insert of 2900 nt, comprising two translationally competent transcription units. Despite this increase in genetic material, the virus replicated to high titres in Vero cells. This recombinant virus expressed the RSV G and F proteins sufficiently to evoke a protective immune response in hamsters upon challenge with RSV or human PIV3 and to elicit RSV neutralizing and PIV3 haemagglutinin inhibition serum antibodies. In effect, a bivalent vaccine was produced that could protect vaccinees from RSV as well as PIV3. Such a vaccine would vastly reduce the respiratory disease burden, the associated hospitalization costs and, most importantly, decrease morbidity and mortality of infants, immunocompromised individuals and the elderly.

The nucleocapsid protein (NP) of Newcastle disease virus expressed in E. coli assembled as ring- and herringbone-like particles. In order to identify the contiguous NP sequence essential for assembly of these particles, 11 N- or C-terminally deleted NP mutants were constructed and their ability to self-assemble was tested. The results indicate that a large part of the NP N-terminal end, encompassing amino acids 1 to 375, is required for proper folding to form a herringbone-like structure. In contrast, the C-terminal end covering amino acids 376 to 489 was dispensable for the formation of herringbone-like particles. A region located between amino acids 375 to 439 may play a role in regulating the length of the herringbone-like particles. Mutants with amino acid deletions further from the C-terminal end (84, 98, 109 and 114 amino acids) tended to form longer particles compared to mutants with shorter deletions (25 and 49 amino acids).

The genes encoding the putative small hydrophobic (SH), attachment (G) and polymerase (L) proteins of the Colorado isolate of subgroup C avian pneumovirus (APV) were entirely or partially sequenced. They all included metapneumovirus (MPV)-like gene start and gene end sequences. The deduced Colorado SH protein shared 26·9 and 21·7 % aa identity with its counterpart in human MPV (hMPV) and APV subgroup A, respectively, but its only significant aa similarities were to hMPV. Conserved features included a common hydrophobicity profile with an unique transmembrane domain and the conservation of most extracellular cysteine residues. The Colorado putative G gene encoded several ORFs, the longer of which encoded a 252 aa long type II glycoprotein with aa similarities to hMPV G only (20·6 % overall aa identity with seven conserved N-terminal residues). The putative Colorado G protein shared, at best, 21·0 % aa identity with its counterparts in the other APV subgroups and did not contain the extracellular cysteine residues and short aa stretch highly conserved in other APVs. The N-terminal end of the Colorado L protein exhibited 73·6 and 54·9 % aa identity with hMPV and APV subgroup A, respectively, with four aa blocks highly conserved among Pneumovirinae. Phylogenetic analysis performed on the nt sequences confirmed that the L sequences from MPVs were genetically related, whereas analysis of the G sequences revealed that among MPVs, only APV subgroups A, B and D clustered together, independently of both the Colorado isolate and hMPV, which shared weak genetic relatedness at the G gene level.

A formerly undescribed virus has been isolated from the gills of farmed Atlantic salmon post-smolts in Norway suffering from gill disease. Cytopathic effects appeared in RTgill-W1 cells 9 weeks post-inoculation with gill tissue material. Virus production continued for an extended period thereafter. Light and electron microscopic examination revealed inclusions and replication in the cytoplasm. The viral nucleocapsid consisted of approximately 17 nm thick filaments in a herringbone pattern. Certain areas of the plasma membrane were thickened by the alignment of nucleocapsids on the internal surface and projections of 10 nm long viral glycoprotein spikes on the external surface. Virus assembly and release was achieved by budding through the modified plasma membrane. Negatively stained virions were spherical and partly pleomorphic with a diameter of 150–300 nm as seen by electron microscopy. The virus was sensitive to chloroform, heat and low and high pH, and replication was not inhibited by Br-dU or IdU indicating an RNA genome. Both haemagglutination and receptor-destroying enzyme activity were associated with the virions and the formation of syncytia in infected cultures indicated fusion activity. The receptor-destroying enzyme was identified as neuraminidase. The virus contained five major structural polypeptides with estimated molecular masses of 70, 62, 60, 48 and 37 kDa. Its buoyant density was 1·18–1·19 g ml−1 in CsCl gradients. From the observed properties we conclude that this new virus belongs to the Paramyxoviridae and suggest the name Atlantic salmon paramyxovirus (ASPV). Furthermore, replication occurred at 6–21 °C, suggesting a host range confined to cold-blooded animals.

Apoptosis was detected in Vero cell cultures expressing transfected dengue virus type 2 (DENV-2) genes. Approximately 17·5 and 51·5 % of cells expressing NS3 serine protease and NS2B-NS3185 serine protease precursor protein [NS2B-NS3185(pro)] genes, respectively, were apoptotic. The percentage of apoptotic cells was significantly higher in cell cultures expressing NS2B-NS3185(pro). NS2B-NS3185(pro) was detected as NS2B-NS3185(pro)-EGFP fusion protein in cytoplasmic vesicular structures in the apoptotic cells. Site-directed mutagenesis which replaced His51 with Ala within the protease catalytic triad significantly reduced the ability of the expressed NS3 and NS2B-NS3185(pro) to induce apoptosis. Results from the present study showed that DENV-2-encoded NS3 serine protease induces apoptosis, which is enhanced in cells expressing its precursor, NS2B-NS3185(pro). These findings suggest the importance of NS2B as a cofactor to NS3 protease-induced apoptosis.

Hantaviruses are maintained in nature in persistently infected rodents and can also persistently infect cultured mammalian cells, causing little or no cytopathology. An unexpected outcome of this study was the observation of cytopathic effects (CPE) in the hantavirus-infected human embryonic kidney cell line HEK293. It was confirmed that hantaviruses induce apoptosis in HEK293 cells, although apoptosis appeared mostly in uninfected, bystander cells and rarely in infected HEK293 cells. Although studies by others suggest that the nucleocapsid protein of Puumala virus interacts with the Fas-mediated apoptosis enhancer Daxx at the gene expression level, it was determined that members of the TNF receptor superfamily did not contribute to the apoptosis observed in infected HEK293 cells. The observation of CPE in HEK293 cells might lead to a better understanding of the mechanisms of persistence and pathogenesis in hantavirus infections.

RNA genomes of enteroviruses and rhinoviruses contain a 5′-terminal structure, the cloverleaf (CL), which serves as signal in RNA synthesis. Substitution of the poliovirus [PV1(M)] CL with that of human rhinovirus type 2 (HRV2) was shown previously to produce a viable chimeric PV, whereas substitution with the HRV14 CL produced a null phenotype. Fittingly, the HRV14 CL failed to form a complex with PV-specific proteins 3CDpro–3AB or 3CDpro–PCBP2, considered essential for RNA synthesis. It was reported previously ( Rohll et al., J Virol 68, 4384–4391, 1994 ) that the major determinant for the null phenotype of a PV/HRV14 chimera resides in subdomain Id of the HRV14 CL. Using a chimeric PV/HRV14 CL in the context of the PV genome, stem–loop Id of HRV14 CL was genetically dissected. It contains the sequence C57 UAU 60-G, the underlined nucleotides forming the loop that is shorter by 1 nt when compared to the corresponding PV structure (UUGC 60 GG). Insertion of a G nucleotide to form a tetra loop (C57 UAU 60 GG61) did not rescue replication of the chimera. However, an additional mutation at position 60 (C57 UAC 60 GG61) yielded a replicating genome. Only the mutant PV/HRV14 CL with the UAC 60 G tetra loop formed ternary complexes efficiently with either PV proteins 3CDpro–3AB or 3CDpro–PCBP2. Thus, in the context of PV RNA synthesis, the presence of a tetra loop in subdomain D of the CL per se is not sufficient for function. The sequence and, consequently, the structure of the tetra loop plays an essential role. Biochemical assays demonstrated that the function of the CL element and the function of the cis-acting replication element in the 3Dpol–3CDpro-dependent uridylylation of VPg are not linked.

Despite numerous studies on human immunodeficiency virus type 1 (HIV-1) fitness, many key conceptual and technical questions are still unsolved. For example, the proper system to determine virus fitness of HIV-1 is still unknown. In this study, an assay was developed to estimate HIV-1 fitness based on growth competition experiments and TaqMan real-time PCR. This novel technique was compared with several methods (i.e. virus growth kinetics, growth competition/heteroduplex-tracking analysis and single-cycle replication capacity assay) in order to analyse the impact of various genomic regions and overall genetic background on virus fitness. HIV-1 primary isolates and three different sets of recombinant viruses [i.e. recombinant clones carrying protease (PR), reverse transcriptase (RT) or the 3′ end of Gag, PR and RT (3′Gag/PR/RT), sequences amplified by PCR from the same primary isolates)] were evaluated. Here, it is demonstrated that, in spite of intrinsic differences, both growth competition/TaqMan and single-cycle replication assays detected a significant reduction in HIV-1 fitness as a consequence of drug-resistant mutations in pol. However, this new assay, based on HIV-1 isolates, may be useful to quantify replicative fitness in viruses from patients treated simultaneously with antiretroviral drugs targeting different genomic regions of HIV-1 (e.g. pol and env).

The U3 region of the human endogenous retrovirus W long terminal repeat (HERV-W LTR) contains several putative regulatory sequences that might not only regulate transcription of viral genes but also influence the expression of neighbouring cellular genes. In this study, we analysed the U3 region in detail in order to understand the transcriptional regulatory mechanism of HERV-W. Two transcription factor (TF) binding sites for Oct-1 and C/EBP were important as a silencer and an enhancer, respectively, for transcriptional regulation. Furthermore, it was possible to divide the HERV-W LTR isolates into two groups depending on their promoter strength, which might be determined by the integrity of the two TF binding sites. However, neither the Oct-1 binding site nor the CAAT-box was required for the cell type-specific activity of the HERV-W LTR. Instead, the 3′ terminus of U3 from 191 to 260, which includes a TATA box, was sufficient for specificity, suggesting that the efficiency of assembly of basic transcription machinery at the TATA box of HERV-W LTR might determine the cell type specificity.

A DNA vaccination regime was investigated previously in rhesus macaques using a full-genome human immunodeficiency virus type 1 (HIV-1) plasmid, which, due to mutations in the nucleocapsid (NC) proteins, produced only non-infectious HIV-1 particles ( Akahata et al., Virology 275, 116–124, 2000 ). In that study, four monkeys were injected intramuscularly 14 times with the plasmid. All of them showed immunological responses against HIV-1 and partial protection from challenge with a simian immunodeficiency virus/HIV (SHIV) chimeric virus. To improve this DNA vaccination regime, the plasmid used for vaccination was changed. In the present study, four macaques were injected intramuscularly eight times with a full-genome SHIV plasmid that produces non-infectious SHIV particles. CTL activities were higher than those observed in monkeys vaccinated previously with the HIV-1 plasmid. In all macaques vaccinated, peak plasma virus loads after homologous challenge with SHIV were two to three orders of magnitude lower than those of the naive controls, and virus loads fell below the level of detection at 6 weeks post-challenge. This suggested that the vaccination regime in this study was partially effective and better than the previous regime.

The complete genome sequence of a new isolate of enzootic nasal tumour virus (ENTV-2), associated with enzootic nasal adenocarcinoma (ENA) of goats, was determined. The genome exhibits a genetic organization characteristic of β-retroviruses. ENTV-2 is closely related to the retrovirus (ENTV-1) associated with enzootic adenocarcinoma of sheep, and to jaagsiekte retrovirus. The main sequence differences between these viruses reside in orfX, the U3 LTR, two small regions in gag and the transmembrane (TM) region of env. Sequence analysis of the TM region of env from several sheep and goats naturally affected by ENA suggested that ENTV-1 and ENTV-2 are distinct viruses rather than geographical variants. Although both viruses transform secretory epithelial cells of the ethmoid turbinate, the study of their tissue distribution using specific PCRs showed that ENTV-2 establishes a disseminated lymphoid infection whereas ENTV-1 is mainly confined to the tumour.

Moloney murine leukaemia virus (MoMLV) enters host cells by membrane fusion between the viral envelope and the host cell membrane. The cytoplasmic tail (R peptide) of the MoMLV envelope protein (Env) is cleaved by the viral protease during virion maturation. R peptide-truncated Env induces syncytia in susceptible cells but R peptide-containing Env does not, indicating that the R peptide inhibits membrane fusion. To examine the function of amino acid residues at the R peptide cleavage site in virus entry, mutant Env expression plasmids containing amino acid substitutions at these cleavage site residues were constructed. Some of these mutants induced syncytia in NIH 3T3 cells, even though they expressed the R peptide, indicating the importance of these residues for membrane fusion inhibition by the R peptide. Some mutants in which R peptide cleavage was detected had comparable transduction efficiency to wild-type Env, but mutants in which R peptide cleavage was not detected had lower transduction efficiency than wild-type Env. This result strongly supports that R peptide cleavage is required for virus entry.

Development of a protective host response against intracellular pathogens requires innate and cell-mediated immune responses, with cytokines playing an important role in host defences. Different studies in mice have shown that IL-12 can promote protective immunity to a variety of viruses but, during virus infection, little is known about the in vivo function of IL-18 alone or in combination with IL-12. Using recombinant vaccinia viruses (rVVs) expressing IL-12 and IL-18, the antiviral role of both cytokines in mice has been analysed. The specific anti-VV immune response elicited and the persistence of the virus in target tissues were compared in BALB/c mice inoculated with rVVs expressing IL-12 and IL-18 either singly or in combination. Delivery of IL-12 and IL-18 by rVVs in mice induced a significant enhancement in virus clearance from ovaries and spleen, greater than that expected from the sum of action of both cytokines. Virus clearance involved NK and T cells, as demonstrated in mice depleted of NK cells and in immunodeficient SCID animals. Th1 parameters (CD8+ T cell response and IgG2a : IgG1 ratios) were increased in mice inoculated with rVVs expressing both IL-12 and IL-18 as compared to those animals receiving a single cytokine. These findings indicate that when IL-12 and IL-18 are delivered by rVVs, different mechanisms involving both the innate and specific arms of the immune system act as mediators in the synergistic action of IL-12 and IL-18, leading to VV clearance. These results are of interest for the design of prophylactic as well as therapeutic VV-based strategies.

The early inflammatory response to a virus may be critical in restricting infection and in shaping the subsequent adaptive immune response. In this study we have examined the early inflammatory response of mice following infection with vaccinia virus (VV) strain Western Reserve (WR). Respiratory challenge of BALB/c mice with VV led to early virus replication in the lung and upper respiratory tract followed by dissemination of virus to other visceral organs and to the brain. The number of inflammatory cells, largely macrophages and T lymphocytes, recovered from bronchoalveolar lavage (BAL) fluid increased markedly during infection and coincided with the expression of CC chemokine ligands (CCL) 3, 2 and 11 and CXC chemokine ligands (CXCL) 1 and 2/3 in BAL. The peak of the inflammatory response occurred around day 10 and declined thereafter. The antiviral cytokines IFN-γ and TNF-α, and the reactive nitrogen intermediate nitric oxide (NO), were also detected in BAL from VV-infected mice. A markedly different inflammatory response was observed after intradermal inoculation of WR into the ear pinnae of mice. Intradermal challenge was followed by highly localized virus replication and by a cellular influx, consisting largely of neutrophils and T lymphocytes, into the dermal compartment of the infected ear. Together these findings highlight differences in the pathogenesis and in the cellular inflammatory response to WR following intranasal and intradermal inoculation of mice.

Lumpy skin disease virus (LSDV), a capripoxvirus with a host range limited to ruminants, was evaluated as a replication-deficient vaccine vector for use in non-ruminant hosts. By using the rabies virus glycoprotein (RG) as a model antigen, it was demonstrated that recombinant LSDV encoding the rabies glycoprotein (rLSDV-RG) was able to express RG in both permissive (ruminant) and non-permissive (non-ruminant) cells. The recombinant LSDV, however, replicated to maturity only in permissive but not in non-permissive cells. Recombinant LSDV-RG was assessed for its ability to generate immunity against RG in non-ruminant hosts (rabbits and mice). Rabbits inoculated with rLSDV-RG produced rabies virus (RV) neutralizing antibodies at levels twofold higher than those reported by the WHO to be protective. BALB/c mice immunized with rLSDV-RG elicited levels of RV-specific cellular immunity (T-cell proliferation) comparable with those of mice immunized with a commercial inactivated rabies vaccine (Verorab; Pasteur Merieux). Most importantly, mice immunized with rLSDV-RG were protected from an aggressive intracranial rabies virus challenge.

Epstein–Barr virus (EBV) is a human herpesvirus associated with a number of malignancies. EBV establishes a latent infection in human B cells in vitro, and infected lymphoblastoid cells proliferate indefinitely as a result of virus activation of cellular signalling pathways. Latently infected cells express a viral oncoprotein called the latent membrane protein-1 (LMP-1). LMP-1 signals both proliferative and survival signals to the infected B cell. The switch from latency to lytic replication is associated with upregulation of an N-terminally truncated LMP-1, called lytic LMP-1 (lyLMP-1). To understand better the relationship between LMP-1 protein function and the virus life cycle, LMP-1 and lyLMP-1 were precisely localized in infected B cells. Immunoelectron microscopy of latently infected cells revealed LMP-1 localized in discrete patches in the plasma membrane. Unexpectedly, immunogold-labelled LMP-1 was found in vesicles budding from the plasma membrane into the extracellular space and in small membrane vesicles accumulating in conditioned medium from infected cells. LyLMP-1 immunolabelling was observed only in B95-8 cells harbouring detectable intracellular virus particles and was abundant in the nuclear membrane early, and in the plasma membrane late, following lytic cycle induction. LyLMP-1 immunoreactivity was also observed at sites of virus budding and associated with intracellular virions, suggesting that lyLMP-1 might be incorporated into cytoplasmic virions when budding through the nuclear membrane.