Design and evaluation of hepatitis B virus inhibitors.

Antiviral therapy of chronic hepatitis B remains a major clinical problem worldwide. The design of new nucleoside analogs that inhibit hepatitis B virus (HBV) replication allowed their evaluation in in vitro and in vivo experimental models of HBV infection. This research has led to the discovery of the anti-HBV activity of lamivudine and its approval for the therapy of chronic hepatitis B. However, due to the development of viral resistance, strategies based on the combination of new inhibitors of HBV replication with immune modulatory approaches are urgently required.

Main properties of duck hepatitis B virus DNA polymerase: comparison with the human and woodchuck hepatitis B virus DNA polymerases.

The main properties of the duck hepatitis B virus (DHBV) DNA polymerase have been studied and compared with those of the human hepatitis B virus (HBV) and of the woodchuck hepatitis virus (WHV) DNA polymerases. All 3 enzymes are active under high salt conditions in the presence of high magnesium concentration. DHBV DNA polymerase was found less sensitive to ethanol and to operate at higher optimal pH than the HBV and WHV DNA polymerases. Like the other two viral endogenous DNA polymerases, the DHBV enzyme was strongly inhibited by phosphonoformic acid but not by aphidicolin, sulfhydryl group blockers or phosphonoacetic acid. Inhibition of DHBV DNA polymerase by the triphosphate derivatives of several nucleoside analogs appeared similar to that reported for HBV or WHV endogenous polymerase. FIACTP was the most, and ACVTP the least effective inhibitor; BVdUTP was of intermediary potency; araCTP and araTTP had a greater inhibitory effect on DHBV DNA polymerase than HBV or WHV DNA polymerase. The similarities in the properties of DHBV and HBV DNA polymerase justify the use of the duck hepatitis B polymerase model for screening and evaluation of potentially active drugs against HBV infection.

Inhibition of human and woodchuck hepatitis virus DNA polymerase by the triphosphates of acyclovir, 1-(2'-deoxy-2'-fluoro-beta-D-arabinofuranosyl)-5-iodocytosine and E-5-(2-bromovinyl)-2'-deoxyuridine.

The triphosphates of acyclovir (ACV), 1-(2'-deoxy-2'-fluoro-beta-D-arabinofuranosyl)-5-iodocytosine (FIAC) and E-5-(2-bromovinyl)-2'-deoxyuridine (BVdU) have been examined for their inhibitory effects on the endogenous DNA polymerase reactions of human hepatitis B virus (HBV) and woodchuck hepatitis virus (WHV). All three triphosphates (ACVTP, FIACTP and BVdUTP) inhibited the HBV and WHV DNA polymerases by competing with the corresponding natural substrates. FIACTP was the most potent inhibitor of HBV and WHV DNA polymerase while ACVTP was the least effective inhibitor. The inhibitory properties of these compounds were compared with those of the 5'-triphosphates of 1-beta-arabinofuranosyl-cytosine (ara-CTP) and 1-beta-arabinofuranosylthymine (ara-TTP). The 50% inhibitory doses for HBV and WHV DNA polymerases were in the following order: FIACTP less than BVdUTP less than ara-TTP less than ACVTP less than ara-CTP. BVdUTP appeared to be an efficient alternate substrate to dTTP for HBV DNA polymerase while FIACTP was much less efficient when substituted for dCTP. ACVTP did not act as an alternate substrate to dGTP and appeared to prevent DNA chain elongation.

Synthesis and antiviral activity of new carbonylphosphonate 2',3'-dideoxy-3'-thiacytidine conjugates.

The synthesis of new potential PFA-BCH-189 conjugate analogues is described and their molecular structure clearly identified through NMR and mass spectra techniques. The anti-HIV-1 activity was determined according to the inhibition of syncytium formation in MT-4 cells, while the anti-HBV activity was determined in infected duck hepatocytes. Both antiviral activities of the PFA-BCH-189 conjugates were much lower than those of the parent BCH-189 (2',3'-dideoxy-3'-thiacytidine) (1). Whereas a prodrug effect, following cleavage and release of the free BCH-189 and PFA, cannot be ruled out, poor cellular permeation of the drug seems to be the most likely reason for the reduced activities against HIV and DHBV. The presence of the PFA moiety appears to be detrimental for both the anti-HIV and anti-DHBV activity of PFA-BCH-189 cases.

The SATE pronucleotide approach applied to acyclovir: part II. Effects of bis(SATE)phosphotriester derivatives of acyclovir on duck hepatitis B virus replication in vitro and in vivo.

The in vitro and in vivo antiviral activities of two mononucleoside phosphotriester derivatives of acyclovir (ACV) incorporating S-acyl-2-thioethyl (SATE) groups are reported using the duck model of hepatitis B (DHBV). In primary duck hepatocyte cultures, the described phosphotriesters significantly inhibited the replication of DHBV at submicromolar concentrations. They were found to be more potent than the parent nucleoside. This result was in agreement with our data concerning the anti-HBV activity of these pronucleotides in HepG2.2.15 cells (previous paper). In vivo, the studied SATE pronucleotide was also found to be more efficient than ACV in infected ducklings upon short-term oral therapy, while intraperitoneal treatment showed high anti-DHBV activity with both ACV and its SATE pronucleotide in this animal model. These findings demonstrate the potential of SATE pronucleotides of ACV as anti-HBV agents.

Use of the cross-reactivity between hepatitis B and non-A, non-B viruses for the identification and detection of non-A, non-B 'e' antigen.

The morphological similarities between B and non-A, non-B hepatitis viruses prompted the search for common antigenic determinants between the two agents. Major antigenic cross-reactivity was demonstrated by immunodiffusion between hepatitis B e/3 and the most common circulating antigen previously reported in the serum of patients with non-A, non-B hepatitis. Since it is the equivalent of an e antigen, this antigen will be hereafter referred to as non-A, non-B/e antigen. Screening with hepatitis B e/3 antigen or anti-hepatitis B e/3 by immunodiffusion may be used as an easy and efficient way to test for non-A, non-B hepatitis in the absence of specific reagents since it detected up to 91% of non-A, non-B antigen and 86% of anti-non-A, non-B/e. Antigenic cross-reactivity may be used to look for additional viruses belonging to the hepatitis B group.

Use of the cross-reactivity with hepatitis B virus antigens and antibodies for the demonstration of a woodchuck hepatitis virus 'e' antigen-antibody system.

Woodchucks hepatitis virus (WHV)-associated antigens and antibodies were studied using current sensitive radio- or enzyme immunoassays (RIA, EIA). A significant cross-reactivity was observed between hepatitis B surface antigen (HBsAg) and woodchuck hepatitis surface antigen (WHsAg) using RIA or EIA (Abbott Laboratories, North Chicago, Ill., U.S.A.) although not with two other commercial EIA tested (Organon Technika, Oss, The Netherlands; Behringwerke AG, Marburg, F.R.G.). A weak but significant reactivity was also found when woodchuck sera positive for WHsAg or anti-WHs by immunodiffusion were tested for HBeAg and anti-HBe by RIA, suggesting the existence of a WHeAg-anti-WHe system in infected woodchucks. The specificity of this e-anti-e reactivity in the woodchuck was further confirmed by successful absorption experiments. WHsAg and WHeAg could be distinguished serologically by immunodiffusion and separated from each other by ultracentrifugation and ammonium sulphate precipitation. A WHeAg preparation was used to boost the presumed natural antibody activity of an immune woodchuck. The specific anti-HBe response detected by RIA during the immunization experiments demonstrated the existence of a soluble WHeAg cross-reacting with the human HBe-anti-HBe system. This was confirmed in immunodiffusion by a partial identity between the precipitin lines formed by the WHeAg-anti-Whe and HBeAg-anti-HBe reaction. Whether the WHe-Ag-anti-WHe system wil mimick HBeAg and anti-HBe in all their clinico-pathological correlations, deserves further study.

A novel vector for the study of hepatitis delta virus replication.

In order to study HDV replication without the difficulties caused by the use of a multimeric construction and to obtain a selectable expression vector, a minimal amount of antigenomic HDV cDNA, sufficient to initiate RNA dependent replication was cloned into the plasmid pUTSV1. The first plasmid, pUTdelta1.7, contained 1.7 genomes of HDV cDNA. After transfection of pUTdelta1.7 into HuH7 cells, antigenomic HDV RNA was produced, processed and could enter into the replicative cycle of HDV. However, after transfection of an antigenomic ribozyme mutant (pUTdelta1.7(AGR)) constructed on the same model, plasmid DNA dependent production of genomic HDV RNA was observed, especially in COS7 cells. It seems that a promoter within vector sequences downstream from the HDV insert may initiate counter-clockwise transcription of the plasmid. The presence of two genomic ribozymes in the insert permits the excision of a genome length genomic HDV RNA from this counter-clockwise transcript. In order to allow quantitative analysis of HDV replication, this problem was eliminated by removing the second genomic ribozyme from the insert to give the vector pUTdelta1.5. This vector can be used conveniently for transfection experiments to explore HDV biology.

Direct cloning and expression of PCR amplified DNA and RNA sequences: application to the hepadnaviruses nucleocapsid proteins.

Gene amplification may benefit from the construction of primers that augments the speed at which cloning and protein expression proceeds. Such primers include EcoRI or HindIII linkers as well as an in phase initiation or termination codon. PCR was carried out directly from viral particles of human hepatitis B virus (HBV) and woodchuck hepatitis virus (WHV) without DNA purification and from RNA extracted from WHV infected liver. Amplified products were directly cloned in the pKK223-3 expression vector under the control of the tac promoter. The characterization of the recombinant clones expressing the nucleocapsid protein (C protein) was done by direct incubation of the filter with 125I-labelled anti-HBc and confirmed by radioimmunoassay and Western-blot analysis. This procedure allows easy selection of recombinant clones expressing a given protein and could be applied to many other genes.

Comparative study of DHBV DNA levels and endogenous DNA polymerase activity in naturally infected ducklings in France.

Duck hepatitis B virus (DHBV) was found in the serum of 1-6% of Pekin ducklings originated from French commercial flocks. The viremia was followed in the serum of 5 ducklings over a span of 3 mth by monitoring the levels of DHBV DNA and the endogenous DNA polymerase (DNAp) activity. The DHBV DNA levels in serum were quantified either by the DNA dot hybridization technique including counting of retained radioactivity, or by successive dilutions of each serum sample followed by DNA hybridization. The counting of the retained radioactivity was plotted on a curve and its evolution compared with that of viral DNAp activity. DHBV DNA levels in serum, estimated by both methods paralleled those of the DNAp activity, which peaked at the 4th or 5th week posthatch to decrease and fluctuate thereafter. Occasional discordance between DHBV DNA levels and the endogenous DNAp activity was observed, which could be correlated with the degree of repair of the single stranded gap of serum DHBV DNA. Parallel follow up studies comparing quantitative estimations of serum viral DNA and of DNAp activity, as presented here, may provide some clues for the understanding of the mechanisms involved in the establishment of the HEPA DNA virus carrier state. Such comparative studies may also be crucial for optimal monitoring of antiviral drugs in both human clinical trials and animal experimental studies.

Identification and detection of long incubation non-A, non-B hepatitis virus and associated antigens or antibodies.

Three distinct antigen/antibody systems supposedly associated with an HBV-like virus of non-A, non-B hepatitis have been identified. Because of previously demonstrated cross-reactivity with HBe/3 and HBc antigens and other analogies the following terminology is tentatively used. 1. The previously reported serum antigen has been redesignated non-A, non-B e antigen, since it is equivalent to HBe/3 Ag and cross-reacts with it. Non-A, non-BeAg or Ab were detected in 51/62 post-transfusion and 11/56 sporadic acute non-A, non-B hepatitis cases, and in 12/14 cases affecting staff members. In non-A, non-B chronic persistent or active hepatitis and cryptogenic cirrhosis, the prevalence was similarly high: 14/18, 22/48 and 12/18 respectively. Ten out of 26 implicated blood donors were found positive for non-A, non-BeAg accounting for 7 out of 8 post-transfusion cases. A high prevalence of non-A, non-BeAg was also found in haemophiliacs (11/48) and haemodialysed patients (6/42), whereas anti-non-A, non-Be was respectively detected in 4/48 and 6/42 of these cases. 2. Using immunofluorescence, a second antigen termed non-A, non-BcAg has been identified in liver biopsies from 55/84 non-A, non-B chronic hepatitis or cryptogenic cirrhosis cases. All 8 positive biopsies examined by electron microscopy revealed clusters of 22--25 nm intranuclear particles identical to those described in chimpanzees. Anti-non-A, non-Bc detectable by counter-electrophoresis and indirect immunofluorescence was found in the serum of all patients of which biopsy was positive for non-A, non-BcAg. Anti-non-A, non-Bc was also detected in 5/5 non-A, non-BeAg positive cases of post-transfusion hepatitis, 2--6 weeks after onset end remained positive for the 6 month follow-up period. 3. A third antigen, tentatively designated non-A, non-BsAg, has been found less frequently than non-A, non-BeAg in serum. However, it was detectable in 3/18 and 2/12 washed ultracentrifugation pellets of sera positive for non-A, non-BeAg or anti-non-A, non-Be, respectively.

Monitoring of early events of experimental woodchuck hepatitis infection: studies of peripheral blood mononuclear cells by cytofluorometry and PCR.

The peripheral blood mononuclear cells (PBMC) of woodchucks experimentally infected by woodchuck hepatitis virus (WHV) were examined simultaneously for the presence of membrane associated WHV antigens by cytofluorometry, and for WHV DNA and RNA sequences by the polymerase chain reaction (PCR). Four woodchucks were inoculated: two with a well-defined infectious inoculum and two with an inoculum obtained from an animal at the late incubation phase, which was positive for WHV DNA by PCR but still devoid of WHV markers. Infection was demonstrated in all four inoculated woodchucks by the appearance at different times of WHV DNA and WHV antigens in both leucocytes and serum. WHV DNA was first detected by PCR either in the serum (two cases) or in leucocytes (two cases). The mean percentage of cells positive for membrane associated WHsAg or WHcAg detected by cytofluorometry were 37% +/- 25 and 17% +/- 15 respectively. After 8 weeks, all inoculated animals were WHsAg positive in serum. These data suggest that PBMC are involved in the early events of hepadnavirus infection. They also show that sera which are positive by PCR for WHV DNA may transmit viral infection even while still seronegative for WHV markers and for WHV DNA by dot blot.

Design, synthesis and structure relationships of new N,N',N",N"'-tetrakis (omega-amino alkyl) tetraazamacrocycles.

A number of N,N',N",N"'-tetrakis (omega-aminoalkyl) tetraazamacrocycles and related compounds were synthesized and evaluated for their inhibitory effects on human immunodeficiency virus type 1 (HIV-1) and duck hepatitis B virus (DHBV) replication. The activity of these compounds was found to be highly dependent upon structural features: (i) the length of the alkyl linker connecting the nitrogen atoms of the macrocyclic ring to the exocyclic nitrogen atoms of the terminal amino groups (five methylenes favoured antiviral activity); (ii) substitution of the terminal amino groups of the linker reduced antiviral activity; and (iii) the size of the tetraazamacrocyclic ring (14 or 15 atoms) did not markedly affect the antiviral activity. Some analogues were potent inhibitors of HIV-1 replication, with anti-HIV activity similar to that of biscyclam (JM 2763). In contrast, other analogues were found to be highly toxic in duck hepatocyte primary culture, the 2.2.15 cell line and to a lesser extent in MT-4 cells. Structural parameters, macrocyclic ring size and metal-chelating ability have been used to develop a structure-activity relationship model in order to aid the design of antiviral molecules derived from N,N',N",N"'-tetrakis (omega-aminoalkyl) tetraazamacrocycles.

Persistence of the hepatitis B virus covalently closed circular DNA in HepaRG human hepatocyte-like cells.

The recently described hepatic cell line HepaRG is the sole hepatoma cell line susceptible to hepatitis B virus (HBV) infection. It provides a unique tool for investigating some unresolved issues of the virus' biology, particularly the formation of the viral mini-chromosome believed to be responsible for the persistence of infection. In this study, we characterized the main features of HBV infection: it is restricted to a subpopulation of differentiated hepatocyte-like cells that express albumin as a functional marker and represents around 10 % of all differentiated HepaRG cells. Infection may persist for more than 100 days in cells maintained at the differentiated state. Even though infected cells continued to produce infectious viral particles, very limited or no spreading of infection was observed. Low genetic variation was also observed in the viral DNA from viruses found in the supernatant of infected cells, although this cannot explain the lack of reinfection. HBV infection of HepaRG cells appears to be a very slow process: viral replication starts at around day 8 post-infection and reaches a maximum at day 13. Analysis of viral DNA showed slow and inefficient conversion of the input relaxed circular DNA into covalently closed circular (CCC) DNA, but no further amplification. Continuous lamivudine treatment inhibited viral replication, but neither prevented viral infection nor initial formation of CCC DNA. In conclusion, HBV infection in differentiated HepaRG cells is characterized by long-term persistence without a key feature of hepadnaviruses, the so-called 'CCC DNA amplification' described in the duck hepatitis B model.

Initiation of hepatitis B virus genome replication and production of infectious virus following delivery in HepG2 cells by novel recombinant baculovirus vector.

One of the major problems in gaining further insight into hepatitis B virus (HBV)/host-cell interactions is to improve the existing cellular models for the study of HBV replication. The first objective of this study was to improve the system based on transduction of HepG2 cells with a recombinant baculovirus to study HBV replication. A new HBV recombinant baculovirus, Bac-HBV-1.1, in which the synthesis of pre-genomic RNA is driven by a strong mammalian promoter, was generated. Transduction with this new recombinant baculovirus led to higher levels of HBV replication in HepG2 cells compared with levels obtained with previously described baculovirus vectors. The initiation of a complete HBV DNA replication cycle in Bac-HBV-1.1-transduced HepG2 cells was shown by the presence of HBV replicative intermediates, including covalently closed circular DNA (cccDNA). Only low levels of cccDNA were detected in the nucleus of infected cells. Data showed that cccDNA resulted from the recycling of newly synthesized nucleocapsids and was bound to acetylated histones in a chromatin-like structure. HBV particles released into the supernatant of transduced HepG2 cells were infectious in differentiated HepaRG cells. Several Bac-HBV-1.1 baculoviruses containing HBV strains carrying mutations conferring resistance to lamivudine and/or adefovir were constructed. Phenotypic analysis of these mutants confirmed the results obtained with the transfection procedures. In conclusion, an improved cell-culture system was established for the transduction of replication-competent HBV genomes. This will be useful for future studies of the fitness of HBV mutants.

Non-A, Non-B hepatitis virus: identification of a core antigen-antibody system that cross reacts with hepatitis B core antigen and antibody.

Using immunodiffusion, a major cross-reactivity had been previously demonstrated between hepatitis B(HBe/3 Ag) and the antigen reported in the serum on non A, non-B hepatitis patients, therefore redesignated NANBe Ag. By direct immunofluorescence a new antigen associated with, but distinct from, NANBe Ag has now been identified in the liver of 14/26 patients with NANB chronic active hepatitis. The homologous antibody was detected in the serum of these 14 patients. Behaving like HBc Ag and cross reacting with it by immunofluorescence, the new antigen was termed NANBc Ag. Anti NANBc also became detectable in serial acute phase and convalescence sera from 5/5 NANB Ag-positive posttransfusion hepatitis cases. Further characterization of NANBe and NANBc antigens achieved by fractionation of a NANB virus-infected liver showed NANBc Ag to be expressed on 22-25 nm HB core-like particles containing DNA polymerase activity. Cross-reactivity between NANBc and HBc antigens was confirmed by immunodiffusion. Liver-derived NANBe Ag identical to serum NANBe Ag exhibited the same physical properties as HBe/3 Ag and could be similarly released by disruption of the non-A, non-B, virus cores. These results indicate that hepatitis B and NANB virions not only share the same structure and DNA polymerase activity but are also antigenically related and belong to the same new class of DNA viruses.

Non-a, non-b hepatitis: identification of hepatitis-B-like virus particles in serum and liver.

Hepatitis B virus-like particles including: small spheres and filaments 15--25 nm in diameter together with a 35--40 nm Dane particle-like virion have been identified in sera of patients with non-A, non-B hepatitis. In a coded serological study, such particles were detected transiently in 3/4 acute, and persistently in 7/8 chronic cases of non-A, non-B hepatitis with non-A, non-B antigenemia. Only 2/12 similar cases without non-A, non-B antigens (Ag) in serum had detectable particles but neither patients with drugs, or type A hepatitis, nor cases of obstructive jaundice. The particles did not express hepatitis B surface (HBs) or non-A, non-B Ag at their surface but were associated, in three patients, with significant endogenous DNA polymerase activity. Furthermore, particles similar to hepatitis B cores (BHc) and also associated with DNA polymerase activity were demonstrated by sucrose gradient ultracentrifugation of a liver homogenate obtained from a patient who had died of non-A, non-B hepatitis. The non-A, non-B hepatitis virion described here appears, therefore, as a hepatitis B-like virus. The exact kinship between these two agents is currently being investigated.

[Identification of a virus similar to hepatitis B virus in non-A non-B hepatitis]. / Identification d'un virus semblable au virus de l'hépatite B dans les hépatites NON-A NON-B.

Hepatitis B virus-like particles (including DANE particles) with DNA polymerase activity but negative for HBs Ag have been identified in NON-A, NON-B hepatitis sera positive for HC Ag. Although specifically associated with the particles, HC Ag is not a surface antigen of the hepatitis C virus identified here for the first time. The relationship of this agent with HBV seems obvious, and deserves further study.

Antigenic cross-reactions between woodchuck hepatitis virus and human hepatitis B virus shown by immune electron microscopy.

Using immune electron microscopy (IEM), low-level cross-reactions could be demonstrated between the surface antigens of hepatitis B and woodchuck hepatitis. However, immune complex formation was greatly enhanced by pre-exposure of the antigens to 0.5% deoxycholate. Cross-reaction between the core antigens and e antigens of both viruses was also confirmed by IEM as well as radioimmunoassay. It appears that the woodchuck sera used in this study may well contain an anti-immunoglobulin akin to rheumatoid factor.