- Volume 79, Issue 6, 1998

Two mutant virus strains in which the novel P225H mutation appeared in a V106A reverse transcriptase (RT)-mutated genetic background upon treatment of human immunodeficiency virus type 1 (HIV-1) with quinoxaline S-2720 were isolated. Surprisingly, the addition of the P225H mutation to the V106A RT mutant genetic background resensitized the V106A RT mutant virus to the nonnucleoside RT inhibitor (NNRTI) BHAP U-90152, but not to other NNRTIs. Construction of both recombinant viruses and recombinant RTs containing the V106A, P225H and V106A P225H mutations revealed that P225H was indeed responsible for the marked potentiation of the antiviral activity of BHAP against the P225H single-mutant virus and the V106A P225H double-mutant virus when compared to wild-type and V106A single-mutant viruses, respectively. An explanation for the markedly increased sensitivity of the P225H mutant HIV-1 RT to BHAP and not to the other NNRTIs was provided by the unique features of the X-ray structure of the RT-BHAP complex.

Tuberculosis has emerged as an epidemic, extended by the large number of individuals infected with human immunodeficiency virus type 1 (HIV-1). The major goal of this study was to determine whether the mycobacterial cell wall component mannose-capped lipoarabinomannan (ManLAM) of Mycobacterium tuberculosis (M. tuberculosis) could activate transcription of HIV-1 in T cells with the use of an in vitro cell culture system. These experiments are of prime importance considering that CD4-expressing T lymphocytes represent the major virus reservoir in the peripheral blood of infected individuals. Using the 1G5 cell line harbouring the luciferase reporter gene under the control of the HIV-1 LTR, it was first found that culture protein filtrates (CFP) from M. tuberculosis or purified ManLAM could activate HIV-1 LTR-dependent gene expression unlike similarly prepared CFP extracts devoid of ManLAM. The implication of protein tyrosine kinase(s), protein kinase A and/or protein kinase C was highlighted by the abrogation of the ManLAM-mediated activation of HIV-1 LTR-driven gene expression using herbimycin A and H7. It was also determined, using electrophoresis mobility shift assays, that M. tuberculosis ManLAM led to the nuclear translocation of the transcription factor NF-κB. M. tuberculosis ManLAM resulted in clear induction of the luciferase gene placed under the control of the wild-type, but not the κB-mutated, HIV-1 LTR region. Finally, the ManLAM-mediated activation of HIV-1 LTR transcription was found to be independent of the autocrine or paracrine action of endogenous TNF-α. The results suggest that M. tuberculosis can upregulate HIV-1 expression in T cells and could thus have the potential to influence the pathogenesis of HIV-1 infection.

Human immunodeficiency virus type 1 (HIV-1) establishes latent infection of a certain population of CD4 host cells which could be long-term reservoirs for HIV-1. The expression of viral genes in such long-term infected cells is strongly regulated by cellular status, such as the phase of the cell cycle or stage of cell differentiation. Here, viral gene expression in synchronized U1 cells, a monocytic cell clone latently infected with HIV-1, was characterized. The expression of HIV-1 antigens was detected exclusively at G2/M phase in U1 cells, irrespective of phorbol myristate acetate (PMA) treatment. The induction of HIV-1 gene expression in PMA-treated cells was due to the recruitment of NF-κB with DNA-binding activity at G2/M phase. Activated NF-κB was induced only by PMAtreatment during the late G1 to S, but not after entering G2 phase, indicating that the transcriptional factor(s) involved in viral gene expression is also largely regulated by the host cell cycle. In contrast, the enhancement of antigen expression by treatment with tumour necrosis factor-alpha (TNF-α) was cell cycle-independent. In fact, NF-κB was activated 2 h after TNF-α treatment at all stages of the cell cycle. Thus, the mechanisms of HIV-1 activation from latency in U1 cells by PMA and TNF-α treatment are different. The model system using U1 cells shown here may provide insight into the mechanisms responsible for HIV-1 gene expression from latency.

Japanese studies have defined the discrete 22092248 amino acid region of the non-structural 5A protein (NS5A2209–2248) of hepatitis C virus genotype 1b (HCV 1b) isolates as the interferon sensitivity determining region (ISDR). European studies did not confirm these results since most of the ISDR sequences harboured an intermediate profile. Recently, a direct interaction between the NS5A protein, involving the ISDR, and the interferon-induced protein kinase (PKR) has been reported and presented as a possible explanation of HCV interferon resistance. In the present study, the entire NS5A amino acid sequence from 11 resistant and eight sensitive strains from European HCV 1b isolates was inferred from direct sequencing. The previously described important amino acid stretches and positions in NS5A were compared between the resistant and sensitive groups. Although some variations were observed, no clear differences could be directly correlated with the interferon sensitivity. However, sensitive strains were different, owing to more amino acid changes when compared to a consensus sequence from all strains. The carboxy- terminal region and especially the previously reported NS5A/V3 region showed most of the variations. Moreover, the conformational analysis of NS5A by secondary structure prediction allowed the differentiation of most sensitive strains from resistant ones. It was concluded that other regions different from ISDR were involved in resistance to interferon maybe via the interaction between NS5A and PKR.

Culture supernatant harvested from Daudi cells, a lymphoplastoid cell line, after 58 days of infection with the H77 strain of hepatitis C virus (HCV), was inoculated into a chimpanzee. HCV RNA, as detected by RT-PCR, first appeared in the serum and liver 5 and 6 weeks, respectively, after inoculation. Peripheral blood mononuclear cells (PBMC) collected on week 7 were also positive for HCV RNA. The major sequences of hypervariable region 1 (HVR1) of the viral genome recovered from the inoculated chimpanzee were the ones which were the majority in the original H77 inoculum and not those which were in the majority in the culture supernatant. Only the sequence recovered from PBMC was the same as the major one found in the cell culture.

Feline aminopeptidase N (fAPN) is a major cell surface receptor for feline infectious peritonitis virus (FIPV), transmissible gastroenteritis virus (TGEV), human coronavirus 229E (HCV 229E) and canine coronavirus (CCV). By using chimeric molecules assembled from porcine, human and feline APN we have analysed the determinants involved in the coronavirus receptor function of fAPN. Our results show that amino acids 670–840 of fAPN are critically involved in its FIPV and TGEV receptor function whereas amino acids 135–297 are essential for the HCV 229E receptor function. We also demonstrate that a chimeric molecule assembled from human and porcine APN is able to act as a receptor for FIPV. This is surprising as neither human nor porcine APN by themselves mediate FIPV infection. These results suggest that different determinants in the APN protein are involved in mediating the coronavirus receptor function.

The fusion glycoprotein (FB) gene of five strains of the B subtype of avian pneumovirus (APV; turkey rhinotracheitis virus) has been sequenced. The length of the FB protein was 538 amino acids, identical to that of the F protein of subtype A virus, with which it had 74% and 83% overall nucleotide and deduced amino acid identities, respectively. The FB and FA ectodomains had 90% amino acid identity, very similar to the 91% identity between the ectodomains of the F proteins of subtype A and B human respiratory syncytial virus (HRSV). As with HRSV, the F2 polypeptide was less conserved (83% identity) than F1 (94%). In contrast to the ectodo- main, the transmembrane and cytoplasmic domains of the two APV subtypes were much less conserved (30% and 48% identity, respectively) than those of HRSV (92% and 87%, respectively). Comparisons within all the genera of the Paramyxoviridae (Pneumovirus, Morbillivirus, Paramyxovirus and Rubulla- virus) show that low amino acid identity between F protein transmembrane domains is a feature of different species of virus rather than of strain differences. This may indicate that the two subtypes of APV have evolved in different geographical regions and/or different avian species. This is the first report of an F gene sequence from a subtype B APV.

The interaction of bovine respiratory syncytial virus (BRSV) nucleocapsid protein (N) with itself and phosphoprotein (P) was investigated using the yeast two-hybrid system. N-P interaction was abolished by any of a series of internal deletions or deletions at the C terminus. In contrast, removal of up to 32 amino acids from the N terminus had little effect. Interestingly, while removal of the C-terminal 32 amino acids ablated interaction, it was largely restored by a second deletion removing up to 32 amino acids from the N terminus. Many of these interactions of the BRSV N protein demonstrated a pattern that was similar to those occurring in the N protein of related viruses. N-N interaction was abolished by any of the internal deletions; however, removal of up to 32 amino acids from the N terminus or C terminus was tolerated and increased the strength of the interaction between the two N proteins.

During the 1996 influenza epidemic in Vienna we obtained influenza A virus specimens (Vienna/ 47/96, Vienna/81/96) which grow efficiently in African green monkey kidney (Vero) cells but not in embryonated chicken eggs. Amplification of the specimens in Vero cells resulted in progeny that agglutinated human but not chicken erythrocytes. Reassortment analysis suggested that the haem- agglutinin (HA) might be responsible for the host restriction. Vero cells were infected with the Vienna/47/96 virus and then transfected with reconstituted ribonucleoprotein complexes containing HA genes from egg-adapted strains. Subsequent selective passages in embryonated chicken eggs resulted in selection of transfectant viruses, growing in eggs and containing the transfected HAs. The results demonstrate that host restriction of the Vero-adapted Vienna/47/96 virus is due to its HA. Moreover, the experiments showed that the Vienna/47/96 strain can be used as helper virus for reverse genetics experiments.

Cytoplasmic extracts prepared from avian reovirus (ARV) strain S1133-infected chicken embryo fibroblasts were examined for the presence of RNA- binding proteins in order to identify and characterize ARV RNA-binding proteins. Analysis of binding activity to poly(A)-Sepharose indicated that infected cells contained significant amounts of a protein that co-migrated with ARV protein σNS present in total virus-infected cell extracts. Determination of the N-terminal amino acid sequence of several peptide fragments generated by V8 protease digestion of the poly(A)-Sepharose- purified protein confirmed that this viral protein was σNS. Competition assays showed that singlestranded RNA from the unrelated avian pathogen infectious bursal disease virus was able to compete for binding of σNS to poly(A)-Sepharose. These data suggest that ARV σNS binds to single-stranded RNA in a nucleotide sequence non-specific manner and is functionally similar to its counterpart specified by mammalian reovirus.

During vaccinia virus (VV) morphogenesis intracellular mature virus (IMV) is wrapped by two additional membranes to form intracellular enveloped virus (IEV). IEV particles can nucleate the formation of actin tails which aid movement of IEVs to the cell surface where the outer IEV membrane fuses with the plasma membrane forming cell- associated enveloped virus (CEV) which remains attached to the cell, or extracellular enveloped virus (EEV) which is shed from the cell. In this report, we have used a collection of VV mutants lacking individual EEV-specific proteins to compare the roles of these proteins in the formation of IEV and IEV-associated actin tails and fusion of infected cells after a low pH shock. Data presented here show that p45–50 (A36R) is not required for IEV formation or for acid-induced cell-cell fusion, but is required for formation of IEV-associated actin tails. In contrast, gp86 (Δ 56R), the virus haemagglutinin, is not required for formation of either IEV or IEV- associated actin tails. Data presented also confirm that p37 (gene F13L), gp42 (B5R) and gp22–24 (A34R) are needed for formation of IEV-associated actin tails and for cell-cell fusion after low pH shock. The phenotypes of these mutants were not affected by the host cell type as similar results were obtained in a range of different cells. Lastly, comparisons of the phenotypes of VV strains Western Reserve, ΔA34R and ΔA36R demonstrate that actin tails are not required for low pH-induced cell-cell fusion.

The cell tropism, organ distribution and resultant pathology of African swine fever were compared in domestic pigs infected with lethal (Malawi) and sublethal (Malta) isolates of African swine fever virus (ASFV). After infections with both isolates, ASFV was predominantly localized in cells of the mononuclear phagocytic system and was not observed in endothelial cells in lymphoid tissue. More severe tissue destruction and cell depletion, associated with high levels of infected macrophages, were seen in lymphoid tissues from domestic pigs infected with the virulent Malawi isolate compared to the less virulent Malta isolate of ASFV. The abundant lymphocyte death was caused by apoptosis and not necrosis. In the spleen, as early as 3 days post-infection (p.i.), many lymphocytes in the B and T cell areas of the white and red pulp were apoptotic. Apoptosis in the T cells of the periarteriolar lymphoid sheaths in the spleen, however, occurred later, at 5–7 days p.i. In lymph nodes apoptosis was observed in T lymphocytes as early as 4 days p.i. and extended to B lymphocytes in the follicles later in infection. In pigs recovered from infection with the sublethal Malta isolate, virus was found to persist in lymph nodes and tonsils for up to 48 days p.i. and was located in cells, surrounded by apoptotic lymphocytes, in the paracortex of lymph nodes up to 32 days p.i. Taken together, these observations suggest that apoptosis of uninfected lymphocytes was induced by cytokines or apoptotic mediators released from ASFV infected macrophages.

Bushpigs and warthogs are natural reservoir hosts of African swine fever virus (ASFV) in the wild, showing no clinical signs of disease when infected with the same highly virulent isolates of ASFV that induce rapid, haemorrhagic death in domestic pigs. In contrast to domestic pigs, infection of bushpigs with Malawi isolate results in low levels of virus replication and lymphocyte apoptosis within the spleen, and a relatively low spread of virus to other lymphoid tissues. However, at 10 days post-infection, a high degree of apoptosis was seen in B lymphocytes of the B cell follicles in bushpig lymph nodes. Virus infected cells were present amongst the apoptotic B cells of these follicles, suggesting that indirect factors released from ASFV infected macrophages signal surrounding lymphocytes to enter apoptosis. The susceptibility/resistance of domestic pigs/bushpigs to ASFV may serve as a unique veterinary model for the recently emerging haemorrhagic disease of man.

A body cavity lymphoma-derived cell line (BC1), known to carry both Epstein-Barr virus (EBV) and human herpes virus type 8 (HHV-8; or Kaposi’s sarcoma-associated herpesvirus, KSHV), was analysed for the expression of EBV-encoded, growth transformation-associated antigens and cellular phenotype by immunofluorescence staining, Western blotting, RT-PCR and flow cytometry. A similar phenotypic analysis was also performed on another body cavity lymphoma line, BCBL1, that is singly infected with HHV-8. Phenotypically, the two lines were closely similar. Although both lines are known to carry rearranged immunoglobulin genes, they were mostly negative for B-cell surface markers. Both expressed the HHV-8-encoded nuclear antigen (LNA1). Similarly to Epstein-Barr nuclear antigen type 1 (EBNA1), LNA1 was associated with the chromatin in interphase nuclei and the mitotic chromosomes in metaphase. It accumulated in a few well-circumscribed nuclear bodies that did not colocalize with EBNA1. BC1 cells expressed EBNA1, LMP2A and EBV-encoded small RNAs but not EBNA2-6,LMP1 and LMP2B. They were thus similar to type I Burkitt ’s lymphoma cells and latently infected peripheral B-cells. Analysis of the splicing pattern of the EBNA1-encoding message by RT-PCR showed that BC1 cells used theQUK but not the YUK splice, indicating that the mRNA was initiated from Qp and not from Cp or Wp.

The putative hexon gene of bovine adenovirus type 4 (BAV-4), encoding 910 amino acid residues, has been identified and sequenced. A characteristic codon usage biased towards the use of AT-rich triplets was observed. Comparative analysis with other hexon sequences detected a high level of amino acid identity in the regions corresponding to the pedestals of the hexon. Substitutions, insertions and deletions were identified mainly in the variable regions forming the loops which are exposed on the outer surface of the virion. In these variable regions, BAV-4 shared similarity only with egg drop syndrome (EDS) virus and ovine adenovirus isolate 287 (OAV287). The close relationship of these viruses was also demonstrated by phylogenetic analysis of the hexon gene. In addition to the two groups of the Mastadenovirus and Aviadenovirus genera, a third cluster appeared comprising BAV-4, OAV287 and EDS virus.

Adenovirus-based gene therapy vectors now in use cannot be targeted to specific cell types in vivo and are immunogenic, properties which limit their clinical utility. Improved vectors lacking the genes for viral structural proteins may overcome these limitations. We have developed cell lines which stably express the adenovirus type 5 (Ad5) fibre protein in its native trimeric form. These cells can complement an Ad5 mutant with a defect in the fibre gene, and are capable of incorporating the Ad5 fibre into particles of a different Ad serotype. As the fibre protein is responsible for the initial binding of virus to cells, packaging cell lines expressing different or modified fibre proteins will be useful in studying the mechanism by which adenovirus infects different cell types.

Simian virus 40 (SV40) binds to MHC class I molecules anywhere on the cell surface and then enters through caveolae. The fate of class I molecules after SV40 binding is not known. Sensitivity of 125I-surface-labelled class I molecules to papain cleavage was used to distinguish internalized class I molecules from class I molecules remaining at the cell surface. Whereas the caveolae-enriched membrane microdomain was found to also be enriched for class I molecules, no internalized papain-resistant 125I-surface-labelled class I molecules could be detected at any time in either control cells or in cells preadsorbed with saturating amounts of SV40. Instead, 125I-surface-labelled class I molecules, as well as preadsorbed 125I-labelled anti-class I antibodies, accumulated in the medium, coincident with the turnover of class I molecules at the cell surface. The class I heavy chains that accumulated in the medium were truncated and their release was specifically prevented by the metalloprotease inhibitor 1,10-phenanthroline. Thus, whereas class I molecules mediate SV40 binding, they do not appear to mediate SV40 entry.

Sequence determination of full-length cDNA clones of genomic segments 7–10 (S7-S10) of oat sterile dwarf fijivirus (OSDV) revealed that the 5′ and 3′ ends of the plus strands of these segments had the same conserved terminal sequences, 5′ AACGAAAAA and UUUUUUUAGUC 3′. These sequences are similar, but not identical, to the conserved terminal nucleotide sequences of the genomic segments of rice black streaked dwarf fijivirus (RBSDV) and maize rough dwarf fijivirus (MRDV). The coding strands of S7 and S10 each contained two large nonoverlapping open reading frames (ORFs), as do RBSDV S7 and S9, MRDV S6 and S8 and Nilaparvata lugens reovirus (NLRV; a putative member of Fijivirus) S9. These results strongly suggest that the dicistronic nature of certain genomic segments is characteristic of fijiviruses. Computer analyses revealed sequence homology between RBSDV S7 ORF2, MRDV S6 ORF2 and OSDV S7 ORF2, suggesting that this protein is conserved among plant fijiviruses. No counterparts were found in the genome of NLRV, which is a nonphytopathogenic insect reovirus. Furthermore, phylogenetic trees derived from multiple sequence alignments of each of the homologous proteins from OSDV, RBSDV, MRDV and NLRV suggest that NLRV did not evolve from either Fijivirus group 2 (RBSDVand MRDV) or group 3 (OSDV).