- Volume 86, Issue 2, 2005

With the human immunodeficiency virus type 1 (HIV-1) epidemic expanding at increasing speed, development of a safe and effective vaccine remains a high priority. One of the most central vaccine platforms considered is plasmid DNA. However, high doses of DNA and several immunizations are typically needed to achieve detectable T-cell responses. In this study, a Semliki Forest virus replicon DNA vaccine designed for human clinical trials, DREP.HIVA, encoding an antigen that is currently being used in human trials in the context of a conventional DNA plasmid, pTHr.HIVA, was generated. It was shown that a single immunization of DREP.HIVA stimulated HIV-1-specific T-cell responses in mice, suggesting that the poor immunogenicity of conventional DNA vaccines may be enhanced by using viral replicon-based plasmid systems. The results presented here support the evaluation of Semliki Forest virus replicon DNA vaccines in non-human primates and in clinical studies.

Within human immunodeficiency virus type 1 (HIV-1)-infected patients, there are those who have been infected for more than 10 years with a CD4+ cell count of >500 cells μl−1 and who remain asymptomatic without antiretroviral therapy; these patients are designated long-term non-progressors (LTNPs). In a set of 16 LTNPs, viral dating, DNA viral load, quasispecies heterogeneity and antibody (Ab) titres against gp160 and β 2 microglobulin (β 2m) were determined. Plasma viral RNA and CD4+ and CD8+ T-cell numbers were estimated in more than three samples per patient. Host genetic characteristics, such as Δ32-CCR5 genotype and human leukocyte antigen (HLA) genotype and supertypes, and clinical–epidemiological factors were evaluated. Dating of global populations and of DNA and RNA viral quasispecies identified two subsets of patients: one displaying only ancestral sequences and the other displaying predominantly modern sequences. The ancestral patients displayed a significant reduction in RNA and DNA viral loads, quasispecies heterogeneity, CD8+ cell number, anti-gp160 Ab titres and β 2m level, and they were also associated with better use of safe-sex practices and higher presence of the HLA sB58 supertype than the modern subset. Viral dating has therefore permitted the segregation of LTNPs into two subsets that show very different virological, immunological, host and clinical–epidemiological characteristics. Moreover, whereas the modern subset displayed low levels of virus replication, the ancestral group displayed not only a very limited virus replication, often to undetectable levels, but also very slow or arrested viral evolution, maintaining the close relationship of the viral population to the transmitted virus.

Measles virus (MV) infection and vaccination induce long-lasting immunity and neutralizing-antibody responses that are directed against the MV haemagglutinin (H) and the fusion (F) protein. A new MV genotype, D7, emerged recently in western Germany and rapidly replaced the long-term endemically circulating genotypes C2 and D6. Analysis of the H gene of C2, D6, D7 and vaccine viruses revealed uniform sequences for each genotype. Interestingly, a consistent exchange of seven distinct amino acids in the D7 H was observed when compared with residues shared between C2, D6 and vaccine viruses, and one exchange (D416→N) in the D7 H was associated with an additional N-linked glycosylation. In contrast, the F gene is highly conserved between MVs of these genotypes. To test whether the D7 H protein escapes from antibody responses that were raised against earlier circulating or vaccine viruses, the neutralizing capacity of mAbs recognizing seven distinct domains on the H of an Edmonston-related MV was compared. The mAbs revealed a selective and complete loss of two neutralizing epitopes on the D7 H when compared with C2, D6 and vaccine viruses. To assess whether these alterations of the D7 H affect the neutralizing capacity of polyclonal B-cell responses, genotype-specific antisera were produced in cotton rats. However, no significant genotype-dependent difference was found. Likewise, human sera obtained from vaccinees (n=7) and convalescents (n=6) did not distinguish between the MV genotypes. Although the hypothesis of selection of D7 viruses by pre-existing neutralizing antibodies is compatible with the differing pattern of neutralizing epitopes on the H protein, it was not confirmed by the results of MV neutralization with polyclonal sera.

The phosphorylation status of the small hydrophobic (SH) protein of respiratory syncytial virus (RSV) was examined in virus-infected Vero cells. The SH protein was isolated from [35S]methionine- and [33P]orthophosphate-labelled RSV-infected cells and analysed by SDS-PAGE. In each case, a protein product of the expected size for the SH protein was observed. Phosphoamino acid analysis and reactivity with the phosphotyrosine specific antibody PY20 showed that the SH protein was modified by tyrosine phosphorylation. The role of tyrosine kinase activity in SH protein phosphorylation was confirmed by the use of genistein, a broad-spectrum tyrosine kinase inhibitor, to inhibit SH protein phosphorylation. Further analysis showed that the different glycosylated forms of the SH protein were phosphorylated, as was the oligomeric form of the protein. Phosphorylation of the SH protein was specifically inhibited by the mitogen-activated protein kinase (MAPK) p38 inhibitor SB203580, suggesting that SH protein phosphorylation occurs via a MAPK p38-dependent pathway. Analysis of virus-infected cells using fluorescence microscopy showed that, although the SH protein was distributed throughout the cytoplasm, it appeared to accumulate, at low levels, in the endoplasmic reticulum/Golgi complex, confirming recent observations. However, in the presence of SB203580, an increased accumulation of the SH protein in the Golgi complex was observed, although other virus structures, such as virus filaments and inclusion bodies, remained largely unaffected. These results showed that during RSV infection, the SH protein is modified by an MAPK p38-dependent tyrosine kinase activity and that this modification influences its cellular distribution.

The aim of this study was to gain more detailed insights into the genetic evolution and variability of Borna disease virus (BDV). Phylogenetic analyses were performed on field viruses originating from naturally infected animals, the BDV vaccine strain ‘Dessau’, four widely used laboratory strains and the novel BDV subtype No/98. Four regions of the BDV genome were analysed: the complete p40, p10 and p24 genes and the 5′-untranslated region of the X/P transcript. BDV isolates from the same geographical area exhibited a clearly higher degree of identity to each other than to BDV isolates from other regions, independent of host species and year of isolation. Five different clusters could be established within endemic areas, corresponding to the geographical regions from which the viruses originated: (i) a Swiss, Austrian and Liechtenstein Rhine valley group, related closely to the geographically bordering Baden-Wurttemberg and Bavaria II group (ii) in the western part of Germany; (iii) a third group, called Bavaria I group, limited in occurrence to Bavaria; (iv) a southern Saxony-Anhalt and bordering northern Saxony group, bound to the territories of these federal states in the eastern part of Germany; and (v) a mixed group, consisting of samples from different areas of Germany; however, these were mainly from the federal states of Thuringia and Lower Saxony. The laboratory strains and the vaccine strain clustered within these groups according to their geographical origins. All field and laboratory strains, as well as the vaccine strain, clearly segregated from the recently described and highly divergent BDV strain No/98, which originated from an area in Austria where Borna disease is not endemic.

Borna disease virus (BDV) can persistently infect the central nervous system and induce CD8+ T-cell-mediated neurological disease in MRL mice. To determine whether specific immune priming would prevent disease, a prime–boost immunization protocol was established in which intramuscular injection of a recombinant parapoxvirus expressing BDV nucleoprotein (BDV-N) was followed by intraperitoneal infection with vaccinia virus expressing BDV-N. Immunized wild-type and perforin-deficient mice remained healthy after intracerebral infection with BDV and contained almost no virus in the brain at 5 weeks post-challenge. Immunization failed to induce resistance against BDV in mice lacking mature CD8+ T cells. Immunization of perforin-deficient mice with a poxvirus vector expressing mutant BDV-N lacking the known CD8+ T-cell epitope did not efficiently block multiplication of BDV in the brain and did not prevent neurological disease, indicating that vaccine-induced immunity to BDV in wild-type and perforin-deficient mice resulted from the action of CD8+ T cells.

The envelope glycoprotein located at the outermost surface of the flavivirus particle mediates entry of virus into host cells. In this study, the involvement of domain III of West Nile virus (WNV-DIII) envelope protein in binding to host cell surface was investigated. WNV-DIII was first expressed as a recombinant protein and purified after a solubilization and refolding procedure. The refolded WNV-DIII protein displays a content of β-sheets consistent with known homologous structures of other flavivirus envelope DIII, shown by using circular dichroism analysis. Purified recombinant WNV-DIII protein was able to inhibit WNV entry into Vero cells and C6/36 mosquito cells. Recombinant WNV-DIII only partially blocked the entry of dengue-2 (Den 2) virus into Vero cells. However, entry of Den 2 virus into C6/36 was blocked effectively by recombinant WNV-DIII. Murine polyclonal serum produced against recombinant WNV-DIII protein inhibited infection with WNV and to a much lesser extent with Den 2 virus, as demonstrated by plaque neutralization assays. Together these results provided strong evidence that immunoglobulin-like DIII of WNV envelope protein is responsible for binding to receptor on the surface of host cells. The data also suggest that similar attachment molecule(s) or receptor(s) were used by WNV and Den 2 virus for entry into C6/36 mosquito cells.

Persistent infection of mouse neuroblastoma NB41A3 cells with yellow fever 17D virus generates viral variants which exhibit defective cell penetration, poor cell-to-cell spread, small plaque size and reduced growth efficiency, caused by substitution of glycine for aspartic acid or glutamic acid at positions 360 and 362 in the envelope protein. These positions occur within a charge cluster, Asp360-Asp361-Glu362, located in domain III, near its interface with domain I. To characterize further the molecular basis for the variant phenotype, a series of mutant viruses containing substitutions at position 360, 361 and 362, were studied for effects on the cell culture properties typical of the neuroblastoma-adapted variant. Most substitutions at position 360 gave rise to viruses that were very defective in cell penetration, growth efficiency and cell-to-cell spread, whereas substitution with glutamic acid yielded a virus indistinguishable from parental yellow fever 17D. Substitution with lysine was not tolerated and substitution with asparagine resulted in frequent wild-type revertants. A glycine residue was not tolerated at position 361, but substitution at 362 yielded a small plaque virus, similar to the effect of substitution at position 360. These data indicate that the yellow fever virus E protein contains a locus within domain III where a negative-charge cluster is important for optimal function of this domain in virus-cell interactions beyond the stage of virus attachment. Modelling predictions suggest that the mutations alter the local properties of the loop within domain III, and may compromise interactions of this domain with an adjacent region of domain I during conformational changes that occur in the E protein in association with virus entry.

The complete genomes of three human H5N1 influenza isolates were characterized, together with the haemagglutinin (HA) and neuraminidase (NA) genes from two additional human isolates and one chicken isolate. These six influenza isolates were obtained from four different provinces of Thailand during the avian influenza outbreak in Asia from late 2003 to May 2004. All six Thailand isolates contained multiple basic amino acids at the cleavage site in the HA gene. Amino acid residues at the receptor-binding site of the five human viruses were similar to those of the chicken virus and other H5N1 viruses from Hong Kong. The presence of amantadine resistance in the Thailand viruses isolated during this outbreak was suggested by a fixed mutation in M2 and confirmed by a phenotypic assay. All genomic segments of the Thailand viruses clustered with the recently described genotype Z. The Thailand viruses contained more avian-specific residues than the 1997 Hong Kong H5N1 viruses, suggesting that the virus may have adapted to allow a more efficient spread in avian species.

Due to their extremely high genetic diversity, which is a direct consequence of high mutation rates, RNA viruses are often described as molecular quasispecies. According to this theory, RNA virus populations cannot be understood in terms of individual viral clones, as they are clouds of interconnected mutants, but this prediction has not yet been demonstrated experimentally. The goal of this study was to determine the fitness of individual clones sampled from a given RNA virus population, a necessary previous step to test the above prediction. To do so, limiting dilutions of a vesicular stomatitis virus population were employed to isolate single viral clones and their initial growth dynamics were followed, corresponding to the release of the first few hundred viral particles. This technique is useful for estimating basic fitness parameters, such as intracellular growth rate, viral yield per cell, rate at which cells are infected and time spent in cell-to-cell transmission. A combination of these parameters allows estimation of the fitness of individual clones, which seems to be determined mainly by their ability to complete infection cycles more quickly. Interestingly, fitness was systematically higher for initial clones than for their derived populations. In addition to environmental changes, such as cellular defence mechanisms, these differences are attributable to high RNA virus mutation rates.

Molecular methods have enabled the rapid identification of new enterovirus (EV) serotypes that would have been untypable using existing neutralizing antisera. Nineteen strains of four new EV types termed EV76 (11 isolates), EV89 (two isolates), EV90 (four isolates) and EV91 (two isolates), isolated from clinical specimens from patients in France (one isolate) and Bangladesh (18 isolates), are described. Nucleotide sequences encoding the VP1 capsid protein (882–888 nt) are less than 65 % identical to the homologous sequences of the recognized human EV serotypes, but within each group the sequences are more than 78 % identical. The deduced amino acid sequences of the complete capsid (P1) region are more than 94 % identical within type but less than 76 % identical to those of the recognized serotypes. For both VP1 and P1, the 19 isolates are monophyletic by type with respect to all other EV serotypes. Using the proposed molecular typing scheme, these data support their identification as four new types within the species Human enterovirus A (HEV-A). In almost all cases, the VP1 sequences were more similar to those of some simian EVs than to the human EVs. Partial 3D sequences of all 19 isolates also clustered within HEV-A; they were monophyletic as a group, but not by type, suggesting that recombination has occurred among viruses of these four types. Partial 3D sequences were more closely related to those of simian EVs than to human viruses in HEV-A. These results suggest that the four new types may represent a new subgroup within HEV-A, in addition to the existing human and simian subgroups.

Bluetongue viruses (BTVs) are economically important arboviruses that affect sheep and cattle. The overwintering mechanism of BTVs in temperate climates has eluded researchers for many years. Many arboviruses overwinter in their invertebrate vectors. To test the hypothesis that BTVs overwinter in their vertically infected insect vectors, Culicoides sonorensis larvae were collected from long-term study sites in northern Colorado, USA, and assayed for the presence of BTV RNA by nested RT-PCR. Sequences from BTV RNA segment 7 were detected in 30 % (17/56) of pools composed of larvae and pupae collected in 1998 and in 10 % (31/319) of pools composed of adults reared from larvae collected in 1996. BTV was not isolated from the insects. Additionally, Culicoides cell-culture lines derived from material collected at one of the sites, or derived from insect samples collected during a BTV outbreak, contained BTV RNA segment 7. In contrast, segment 2 RNA was detected at half the rate of segment 7 RNA in the field-collected larvae and was only detected in the Culicoides cell lines with one of two primer sets. These data suggest that BTVs could overwinter in the insect vector and that there is reduced expression of the outer capsid genes during persistent infection.

Glycoprotein G-1 (gG-1) of herpes simplex virus type 1 (HSV-1) and gG-2 of HSV-2 are the only known HSV proteins that induce type-specific human antibody responses. Recently, it was shown that purified human anti-gG-1 and anti-gG-2 antibodies presented a type-specific reactivity to immunogenic stretches with high similarity between gG-1 and gG-2. In this study, the molecular basis for this type-specific recognition was investigated employing synthetic peptides covering the indicated regions, including substitutions of the type-specific residues. The results revealed that single or dual type-specific residues localized within regions of high similarity could induce significant structural differences, explaining the type-specific recognition of the human antibody response to the gG proteins.

Herpes simplex virus (HSV) capsids assemble, mature and package their viral genome in the nucleoplasm. They then exit the nucleus into the cytoplasm, where they acquire their final tegument and envelope. The molecular mechanism of cytoplasmic envelopment is unclear, but evidence suggests that the viral glycoprotein tails play an important role in the recruitment of tegument and capsids at the final envelopment site. However, due to redundancy in protein–protein interactions among the viral glycoproteins, genetic analysis of the role of individual glycoproteins in assembly has been difficult. To overcome this problem, a glutathione S-transferase fusion protein-binding assay was used in this study to test the interaction between the cytoplasmic tail of one specific viral glycoprotein, gD, and tegument proteins. The study demonstrated that the 38 kDa tegument protein VP22 bound specifically to the gD tail. This association was dependent on arginine and lysine residues at positions 5 and 6 in the gD tail. In addition, HSV-1 capsids bound the gD tail and exhibited a similar sequence dependence. It is concluded that VP22 may serve as a linker protein, mediating the interaction of the HSV capsid with gD.

Sequence data for eight genes, together with time-course Northern blotting and 3′- and 5′-RACE (rapid amplification of cDNA ends) analysis for some mRNAs from a 12 kb region upstream from the major immediate-early (MIE) genes of the English isolate of rat cytomegalovirus (RCMV), are presented. The results identified important differences compared to both murine cytomegalovirus (MCMV) and the Maastricht isolate of RCMV. A striking finding is the presence of a highly conserved, rightwards-oriented homologue of the rat cellular CD200 (OX2) gene immediately to the right of the MIE region, which replaces either the leftwards-oriented AAV REP gene of RCMV (Maastricht) or the upstream spliced portions of the immediate-early 2 gene (ie2) in MCMV. From the presence of other homologues of MCMV- and RCMV-specific genes, such as the β-chemokine MCK-2, SGG1 and an Fcγ receptor gene, as reported here, the basic architecture of the MIE region (reported previously) and the level of IE2 and DNA polymerase (POL) protein conservation in phylogenetic analyses, it is clear that the English strain of RCMV is also a member of the genus Muromegalovirus, but is a β-herpesvirus species that is very distinct from both MCMV and RCMV (Maastricht). Both the lack of a CD200 homologue in the other two rodent viruses and the depth of sequence divergence of the rodent CMV IE2 and POL proteins suggest that these three viruses have evolved as separate species in the genus Muromegalovirus since very early in the host rodent lineage.

Human cytomegalovirus (HCMV) genetic determinants of endothelial-cell tropism and virus transfer to leukocytes (both polymorphonuclear and monocyte) have been recently identified in the UL131–128 genes. Here it is documented that the same genetic determinants of HCMV are responsible for monocyte-derived dendritic-cell (DC) tropism, i.e. all endotheliotropic and leukotropic strains of HCMV are also DC-tropic (or dendrotropic). In fact, all recent clinical HCMV isolates and deletion mutants sparing the UL131–128 locus as well as the endotheliotropic revertants AD169 and Towne were able to productively infect DC following co-culture with infected endothelial cells. On the contrary, the same clinical isolates extensively propagated in human fibroblasts, the UL131–128 deletion mutants and the reference laboratory strains were not. Peak extracellular virus titres in DC were reached 4–7 days post-infection (p.i.). Viral proteins pp65 and p72 were detected 1–3 h p.i., involving the great majority of DC 24 h p.i., while gB was abundantly detected 96 h p.i., when a cytopathic effect first appeared. Infection of DC with cell-free virus released into the medium could only be achieved with HCMV strains extensively adapted to growth in endothelial cells, reaching the peak titres 10 days p.i. DC infected for 24 h with cell-free virus and incubated for 16 h with autologous peripheral blood mononuclear cells were found to act as a potent stimulator of both HCMV-specific CD4+- and CD8+-mediated immune responses, as determined by cytokine flow cytometry. DC incubated with inactivated crude whole viral antigen preparations were only capable of eliciting a significant CD4+-mediated immune response.

The activation of NF-κB has long been considered a positive factor for human cytomegalovirus (HCMV) replication. The HCMV immediate-early promoter, the initial transcriptional element in the HCMV replication cycle, is activated by the transcription factor NF-κB, and several HCMV gene products have been demonstrated to activate this transcription factor. However, the role of NF-κB in the full replication cycle of the virus has not been carefully examined. A series of experiments that demonstrate an important inhibitory role of NF-κB for HCMV replication in fibroblasts is presented here. Using both genetic and pharmaceutical methods, it was shown that blocking NF-κB activation in cell culture does not inhibit HCMV replication, but rather leads to a modest increase in replication. Two cytokines inhibitory for HCMV, tumour necrosis factor-α and interferon-γ, no longer inhibit HCMV when NF-κB activation is blocked. Furthermore, forced expression of the NF-κB activating IκB kinase β (IKKβ), but not a kinase inactive mutant, also inhibits HCMV replication. In addition, it was shown that NF-κB signalling is essential for the production of an anti-viral factor in the supernatant of HCMV-infected fibroblasts, and identified interferon-β as this factor. Thus, the role of NF-κB in fibroblasts is to activate a host defence against HCMV.

The human cytomegalovirus (HCMV) UL78 ORF is considered to encode a seven-transmembrane receptor. However, neither the gene nor the UL78 protein has been characterized so far. The objective of this study was to investigate the UL78 gene and to clarify whether it is essential for replication. UL78 transcription was activated early after infection, was inhibited by cycloheximide but not by phosphonoacetic acid, and resulted in a 1·7 kb mRNA. Later in the replication cycle, a second mRNA of 4 kb evolved, comprising the UL77 and UL78 ORFs. The 5′ end of the UL78 mRNA initiated 48 bp upstream of the translation start and the polyadenylated tail started 268 bp downstream of the UL78 translation stop codon within the UL79 ORF. By using bacterial artificial chromosome technology, a recombinant HCMV lacking most of the UL78 coding region was constructed. Successful reconstitution of the UL78-deficient virus proved that the gene was not essential for virus replication in fibroblasts. The deletion also did not reduce virus replication in ex vivo-cultured sections of human renal arteries. Analysis of viral proteins at different stages of the replication cycle confirmed these results. Among clinical HCMV isolates, the predicted UL78 protein was highly conserved. However, an accumulation of different single mutations could be found in the N-terminal region and at the very end of the C terminus. Due to the absence of an in vivo HCMV model, the role of UL78 in the pathogenesis of HCMV infection in humans remains unclear.

Phylogenetic relationships within the subfamily Gammaherpesvirinae of the family Herpesviridae were investigated for three species in the genus Lymphocryptovirus (or γ1 group) and nine in the genus Rhadinovirus (or γ2 group). Alignments of amino acid sequences from up to 28 genes were used to derive trees by maximum-likelihood and Bayesian Monte Carlo Markov chain methods. Two problem areas were identified involving an unresolvable multifurcation for a clade within the γ2 group, and a high divergence for Murid herpesvirus 4 (MHV4). A robust final tree was obtained, which was valid for genes from across the virus genomes and was rooted by reference to previous analyses of the whole family Herpesviridae. This tree comprised four major lineages: the γ1 group of primate viruses; a clade of artiodactyl γ2 viruses; a clade of perissodactyl γ2 viruses; and a clade of γ2 viruses with a multifurcation at its base and containing Old World and New World primate viruses, Bovine herpesvirus 4 and MHV4. Developing previous work it was proposed, on the basis of similarities between the gammaherpesvirus tree and the tree of corresponding mammalian hosts, that the first three of these major viral lineages arose in a coevolutionary manner with host lineages, while the fourth had its origin in an ancient interspecies transfer. Transfer of dates from mammalian palaeontology then allowed estimation of dates for nodes in the gammaherpesvirus tree.