Genetic analysis of the compatibility between polymerase proteins from human and avian strains of influenza A viruses.

In order to determine how efficiently the polymerase proteins derived from human and avian influenza A viruses can interact with each other in the context of a mammalian cell, a genetic system that allows the in vivo reconstitution of active ribonucleoproteins was used. The ability to achieve replication of a viral-like reporter RNA in COS-1 cells was examined with heterospecific mixtures of the core proteins (PB1, PB2, PA and NP) from two strains of human viruses (A/Puerto Rico/8/34 and A/Victoria/3/75), two strains of avian viruses (A/Mallard/NY/6750/78 and A/FPV/-Rostock/34), and a strain of avian origin (A/Hong Kong/156/97) that was isolated from the first human case of H5N1 influenza in Hong Kong in 1997. In accordance with published observations on reassortant viruses, PB2 amino acid 627 was identified as a major determinant of the replication efficiency of heterospecific complexes in COS-1 cells. Moreover, the results showed that replication of the viral-like reporter RNA was more efficient when PB2 and NP were both derived from the same avian or human virus or when PB1 was derived from an avian virus, whatever the origin of the other proteins. Furthermore, the PB1 and PB2 proteins from the A/Hong- Kong/156/97 virus exhibited intermediate properties with respect to the corresponding proteins from avian or human influenza viruses, suggesting that some molecular characteristics of PB1 and PB2 proteins might at least partially account for the ability of the A/Hong Kong/156/97 virus to replicate in humans.

Comparative analysis of the ability of the polymerase complexes of influenza viruses type A, B and C to assemble into functional RNPs that allow expression and replication of heterotypic model RNA templates in vivo.

Influenza viruses type A, B, and C are human pathogens that share common structural and functional features, yet they do not form natural reassortants. To determine to what extent type-specific interactions of the polymerase complex with template RNA contribute to this lack of genotypic mixing, we investigated whether homotypic or heterotypic polymerase complexes support the expression and replication of model type A, B, or C RNA templates in vivo. A plasmid-based expression system, as initially described by Pleschka et al. [(1996) J. Virol. 70, 4188-4192] for influenza A virus, was developed for influenza viruses B/Harbin/7/94 and C/Johannesburg/1/66. The type A core proteins expressed heterotypic model RNAs with similar efficiencies as the homotypic RNA. The influenza B virus model RNA was efficiently expressed by all three types of polymerase complexes. Although no functional polymerase complex could be reconstituted with heterotypic P protein subunits, when the influenza A virus P proteins were expressed together with heterotypic nucleoproteins, significant, albeit limited, expression of RNA templates of all influenza virus types was detected. Taken together, our results suggest that less strict type-specific interactions are involved for the polymerase complex of influenza A compared with influenza B or C viruses.

New quantitative assay of hepatitis B and C viruses by competitive PCR using alternative internal sequences.

A competitive PCR was developed for quantitation of hepatitis B virus (HBV) DNA and hepatitis C virus (HCV) RNA, alternatively, using only two constructions containing both priming sites. DNAs corresponding to the HBV-S gene and the HCV-5' non-coding region were introduced into distinct plasmids. HBV plasmid was used as a standard for HBV-DNA quantitation, in competition with the HCV plasmid as internal control. HBV and HCV plasmids also served as template for transcription of HBV-RNA, and HCV-RNA, which was used as internal control and standard, respectively, in competition for HCV-RNA quantitation. The analyzed samples for HBV and HCV quantitation were processed in the same way in competition with the internal controls and to the respective calibration curves obtained by serial dilutions of the mimic standard. This method showed very good specificity and sensitivity, allowing absolute quantitation in a large linear range from 5 viral genomic copies per assay up to 10(6) copies, in sera of chronically HBV and HCV infected patients, as well as in supernatants of cell cultures inoculated with these viruses.

Interplay between a new HNF3 and the HNF1 transcriptional factors in the duck hepatitis B virus enhancer.

We identified a new hepatocyte nuclear factor 3 (HNF3) binding site in the DHBV enhancer. This site is close to the hepatocyte nuclear factor 1 (HNF1) binding site, responsible for most of the enhancing activity. No differences in the migrating properties were found between this new site and the two other HNF3 sites recently described in this enhancer. Factor HNF1 strongly inhibits binding of the HNF3 factor in this newly characterized site. The two factors were never detected simultaneously on the DNA fragment, even when their respective concentrations were modified. Competition persisted after enlarging by 5 and 10 nucleotides the space between the two sites. On the contrary, when the HNF3 binding site was changed into the perfect consensus site, binding of the HNF3 factor was not inhibited any longer by HNF1 and a supershift, corresponding to the binding of both factors, was observed. Thus a limited mismatching appears to modulate the interaction between transcriptional proteins and DNA and allows a second transcriptional protein to interplay with the former one.

Binding of nuclear factors to functional domains of the duck hepatitis B virus enhancer.

We have analyzed the structures, relative organization, and activities of binding sites for nuclear factors in the duck hepatitis B virus (duck HBV) enhancer. DNase I footprinting analysis and mobility shift assays demonstrate that this enhancer of 192 bp contains at least three binding sites for transcription factors: one for hepatocyte-adipocyte C/EBP, a second for the liver-specific transactivator hepatocyte nuclear factor 1 HNF-1, and a third for a factor, called F3, which binds to a DNA sequence bearing some resemblance to that for the ubiquitous factor EF-C. Analysis of transcriptional activity reveals that oligonucleotides corresponding to the individual binding sites, inserted upstream from a heterologous promoter, display very weak enhancer activity, whereas the enhancer encompassing these three sites displays very high activity. Analysis of duck HBV enhancer mutants indicates that the deletion of any of these sites leads to a modification of transcriptional enhancer activity. The hepatocyte nuclear factor 1 binding site is crucial, since an internal deletion of 14 bp abolishes the activity. The C/EBP site can act as repressor, and the F3 site is required for full activity. Comparative analysis reveals that the nuclear factors are similar to those bound to the human HBV enhancer but that the organization of their binding sites in the duck HBV enhancer is different.

Identification of a strong enhancer element upstream from the pregenomic RNA start site of the duck hepatitis B virus genome.

The genome of the duck hepatitis B virus (DHBV) contains an enhancer element. This sequence, of 192 bp, is located in the 3'-terminal coding region of the DNA polymerase gene (nucleotides 2159 to 2351), upstream from the pregenomic RNA start site. This enhancer potentiates a marked increased activity from the heterologous thymidine kinase promoter in an orientation-independent manner and at a proximal, as well as a distal, location. The DHBV enhancer activates transcription in a relatively cell-type-independent manner. Sequence homologies with the nuclear factor EF-C binding site are located in the DHBV enhancer. By using the HepG2 nuclear extracts and the DHBV enhancer as probes, a complex was observed in mobility shift assays.

Protein F. A novel F(ab)-binding factor, present in normal liver, and largely released in the digestive tract during hepatitis.

Significant percentages of patients suffering from non-A non-B hepatitis (43%) and B hepatitis (35%) were found to release an Ig-binding factor in their stools. This factor, which we called "protein F" was less frequently observed (20%) in patients suffering from other liver disorders, and was found in only 6.7% of healthy subjects (p less than 10(-7), less than 10(-4), and less than 0.03, respectively). The specificity of the detection test (a nonimmune ELISA-like assay) was confirmed by inhibition experiments. Binding was located on the F(ab) fragment of Ig, irrespectively of their isotype. Protein F was inactivated by pepsin, neuraminidase, and high concentrations of subtilisin, whereas it was resistant to trypsin and chymotrypsin. Molecular sieving by HPLC indicated an apparent molecular mass of 175 kDa. In preparative SDS-PAGE, the molecular mass was 85 kDa in favor of a dimer disrupted under dissociating conditions. Preparative IEF showed the isoelectric charge to lie between 3.9 and 4.1. Analysis of liver extracts from two patients suffering fron non-A non-B hepatitis, and from a transplant donor, revealed the presence of the factor in the three cases.