Currently the best model systems available for in vitro study of HBV include stably transfected hepatocyte-derived cell lines such as HepG2 2.2.15 (28
), derived from the human hepatoblastoma cell line HepG2, and HB611, derived from the hepatoma cell line Huh-6 (31
). Alternatively, plasmid DNA containing the HBV genome can be transiently transfected into cell lines which support HBV replication such as HepG2 or the hepatoma cell line Huh-7 (5
). Stably transfected cell lines, in particular, the 2.2.15 cell line, have been used frequently as in vitro culture systems for studying the efficacy of antiviral agents (21
). In contrast to stable cell lines whose expression is restricted to the specific copy number of integrated HBV genomes in that cell line, the HBV baculovirus-HepG2 system developed recently in our laboratory allows the manipulation of HBV expression levels over a wide range (10
). Infection with HBV baculovirus at moderate MOIs also results in overall levels of viral gene expression and replication far higher than those observed in 2.2.15 cells. Because higher levels of HBV replication can be obtained after HBV baculovirus infection, the rapid detection of intracellular DNA, in particular, non-protein-associated RC and CCC HBV DNA, is possible. The purpose of the study reported here was twofold: to explore the use of the HBV baculovirus-HepG2 system as a model for in vitro testing of antivirals and to further characterize the antiviral properties of the cytosine analog 3TC.
When testing viral mRNA synthesis and HBV antigen secretion by HBV baculovirus-infected HepG2 cells, we found no differences between treated and untreated cultures. This finding has also been reported by other investigators and is not unexpected, based on the mechanism by which 3TC acts. 3TC is phosphorylated inside cells and is subsequently incorporated into nascent viral DNA by the HBV polymerase during replication (4
). 3TC incorporation results in the termination of DNA elongation by virtue of its lack of a 3′ hydroxyl group. Therefore, the expected result would be that 3TC would not directly affect the transcription or translation of HBV gene products from nuclear DNA because it acts downstream of these events. It is interesting to note that an almost complete inhibition of the presence of extracellular HBV DNA did not result in any discernible alteration in the trafficking or secretion of HBeAg or HBsAg in HepG2 cells. This effect is also observed in patients, the majority of whom do not clear either HBeAg or HBsAg after long-term treatment with 3TC, even though their serum HBV DNA levels are markedly reduced (3
When the effects of increasing 3TC concentration and treatment time on a single level of HBV replication (at an MOI of 50 PFU/cell) were measured, we found that both HBV DNA synthesis and the secretion of virions into the medium were highly sensitive to 3TC. The production of extracellular HBV DNA was inhibited by more than 99% after 9 days of treatment with 2.0 μM 3TC. Intracellular replicative forms of the HBV genome were only slightly less sensitive to inhibition than extracellular DNA at equal 3TC concentrations and treatment times. It should be noted that many of the intracellular HBV DNA molecules detected in 3TC-treated cultures were partially double-stranded and single-stranded species; this might suggest that many replicating genomes had incorporated 3TC and were blocked from fully elongating into mature DS and RC genomes. One might expect to observe an increasing accumulation of chain-terminated SS species in 3TC-treated cells with time. We have never observed this phenomenon in the HBV baculovirus-HepG2 system. Since the amount of SS DNA present in treated cells is a function of (i) its rate of formation and (ii) its rate of removal (either by completion to DS DNA and export from the cell or by the degradation of intracellular capsids) and taking into account that extracellular HBV DNA does not increase with time in 3TC-treated cells, this data could suggest that capsids containing SS chain-terminated HBV genomes may have a relatively short intracellular half-life in HepG2 cells and thus would not continually accumulate in the cells.
The pretreatment of cells with 2.0 μM 3TC 16 h before HBV baculovirus infection resulted in a consistently greater inhibition of replicative intermediates and extracellular HBV DNA than that observed when 3TC treatment was initiated 24 h p.i. The effects of pretreatment were generally most evident at the earliest time points tested. These findings were not surprising and may simply indicate that cells which were pretreated with the drug were exposed to the drug for a slightly longer period of time by 4 days p.i. or instead may reflect an actual difference between adding 3TC prior to infection and adding it once the HBV DNA replication cycle had been initiated.
The data presented here agree well with published studies on the efficacy of 3TC in vitro. After treating HepG2 2.2.15 cells for 12 days, Doong et al. (12
) and Kruining et al. (22
) reported a 50% reduction of extracellular HBV DNA at concentrations of 0.05 and 0.02 μM 3TC, respectively. We found that a 9-day treatment with 0.02 μM 3TC resulted in a 64% reduction of extracellular HBV DNA produced by HepG2 cells infected at an MOI of 50 PFU of HBV baculovirus/cell. Analysis of results from earlier time points indicated that the decrease in extracellular HBV DNA was dependent on time; after only 3 days of treatment HBV DNA in the medium was reduced by only about 30%. Korba (20
) reported that treatment with a 0.222 μM concentration of 3TC resulted in a 90% decrease in virion DNA produced by HepG2 2.2.15 after 9 days of treatment. Similarly, we found approximately a 98% reduction in extracellular DNA after a 9-day treatment of HBV baculovirus-infected HepG2 cells with 0.2 μM 3TC. Two important distinctions must be made between the 2.2.15 cell line used by other investigators and the HBV baculovirus-HepG2 system used here. First, at an MOI of 50 PFU of HBV baculovirus/cell, HepG2 cells exhibit much higher levels of HBV expression and replication than 2.2.15 cells. Second, with the exception of the appearance of CCC DNA, the copy number of HBV transcriptional templates per culture does not increase with time after HBV baculovirus infection. This is in contrast to cell lines containing integrated HBV genomes which double the number of transcriptional templates per culture each time the cells divide. Bearing these differences in mind, the results obtained by using stable cell lines and the HBV baculovirus-HepG2 system are remarkably similar.
During hepadnaviral replication, CCC HBV DNA can be produced by two pathways: (i) the entry of exogenous Dane particles into host cells and subsequent migration of HBV cores to nuclei and (ii) the cycling of newly synthesized progeny core particles from the cytoplasms of infected cells back to the nuclei. Once a core particle reaches the nucleus by either pathway, the HBV genome gains access to the nucleus by an unknown mechanism, is repaired to form RC DNA, and is subsequently supercoiled into CCC DNA. The effects of 3TC treatment on the first pathway cannot be evaluated by using the HBV baculovirus-HepG2 system or any cell lines because HBV does not directly infect cultured hepatic cell lines. However, the effects of 3TC on the second pathway were addressed by the experiments carried out in this study. Analysis of non-protein-associated RC and CCC forms of the HBV genome revealed that in addition to replicative intermediates and extracellular HBV DNA, the amplification of CCC DNA also can be inhibited by 3TC in a dose-dependent manner. Here, we report a greater than 90% inhibition of non-protein-associated RC and CCC HBV DNA production by HepG2 cells treated with 2.0 μM 3TC. These data were similar to those reported previously (24
), showing that treatment of primary woodchuck hepatocytes with 3TC prior to infection with WHV caused an 80% inhibition of CCC WHV DNA amplification.
Our findings are consistent with the prediction that the cycling of newly synthesized HBV genomes back to the nucleus for CCC DNA amplification appears to require the completion of second-strand synthesis. The ability of 3TC to interfere with the synthesis of viral DNA effectively reduces the pool of mature core particles available to become enveloped virions or to cycle back to the nucleus to form RC and CCC DNA. It is also possible that 3TC could interfere with the nuclear repair of mature DS HBV genomes into CCC DNA. However, previous studies (19
) have suggested that this repair is most likely carried out by host polymerases which are not sensitive to the concentrations of 3TC used in our experiments. RC and particularly CCC HBV DNA were not reduced to the same extent as extracellular HBV DNA when the same 3TC protocols were used. These findings could suggest that when very few mature HBV cores are present in the cytoplasm, there is a tendency for those cores to enter the CCC amplification pathway instead of acquiring an envelope and exiting the cell. Indeed, the finding in this study that extracellular HBV DNA levels were suppressed to a greater extent than intracellular replicative intermediates provides support for this hypothesis. This finding in 3TC-treated cells would not be unlike the natural early stages of hepadnaviral replication when the initial cores produced after infection are believed to cycle back to the nucleus to allow an amplification of CCC DNA before the secretion of virions takes place (32
While studying the effects of initiating 3TC treatment on HBV baculovirus-infected HepG2 cells which had already accumulated CCC DNA, we made several interesting observations. The treatment of cultures which had already accumulated CCC DNA with 3TC did result in a reduction in the levels of CCC DNA. We believe that this reduction is occurring as a result of the block of HBV replication in the cytoplasm, which limits the number of mature genomes potentially available for cycling back to the nucleus to replenish CCC DNA pools. Not surprisingly, treating cultures with existing CCC DNA was not as effective as treating HepG2 cells prior to the initiation of HBV replication. Using the highest 3TC concentration that we tested (0.2 μM 3TC), we observed that CCC DNA was at 51% of control levels after 3 days of treatment and 23% of control levels after 6 days of treatment. The data we have obtained suggest that, at least under the conditions used, the half-life of CCC HBV DNA in HepG2 cells is roughly 3 days. This number would agree well with the previously reported 3- to 5-day half-life of duck hepatitis virus CCC in primary duck hepatocytes (7
). However, it is necessary to take into consideration that the half-life of CCC DNA in an intact liver in which the CCC HBV DNA resides in nonreplicating hepatocytes may differ substantially from that in the in vitro HBV baculovirus-HepG2 system. It is also important to note that the production of extracellular HBV DNA was consistently more sensitive to equal 3TC concentrations and treatment lengths than those of replicative intermediates and particularly CCC HBV DNA.
The initiation of 3TC treatment in HBV-positive patients receiving orthotopic liver transplants prior to transplantation may have short-term benefits. First, administering 3TC before surgery should markedly lower the level of circulating virions capable of infecting new liver tissue. Second, in new hepatocytes which do become infected, a sufficient dose of 3TC may block or at least delay the onset of CCC DNA amplification by suppressing HBV replication. Depending on the stability of CCC DNA formed following infection, it is likely that some amplification will occur, albeit at a reduced rate, in 3TC-treated cells. Although it is unlikely that 3TC alone could prevent liver reinfection, it is possible that continual treatment with sufficient doses may result in a significant delay in the accumulation of CCC DNA within new tissue. Ultimately, a cure for HBV will likely require the elucidation of a method for eliminating episomal HBV DNA in the nuclei of infected cells. Whether this can be accomplished by an exogenous agent or by the induction of an existing cellular pathway remains to be seen. Although the initial formation of CCC HBV DNA due to viral entry may not be prevented by 3TC, the amplification of CCC DNA could potentially be blocked or delayed sufficiently to increase the efficacy of other antiviral agents.
One limitation of using stable cell lines, such as HepG2 2.2.15 cells, for evaluating the efficacy of an antiviral on HBV replication is that the magnitude of virus replication is at a static level predetermined by the number of integrated HBV genome copies. This limitation does not exist when HBV replication in HepG2 cells is mediated by recombinant HBV baculovirus, because it is possible to modulate the level of production of HBV virions over several magnitudes simply by altering the baculovirus MOI. In this study, we examined the effects of 3TC on HBV replication by using input MOIs of recombinant HBV baculovirus that varied over a 16-fold range. We found that the largest reduction of extracellular HBV DNA (>99% reduction after 6 days of treatment) occurred in cells infected with 25 PFU of baculovirus, the lowest MOI tested. Cultures infected at MOIs ranging from 50 PFU/cell to as high as 400 PFU/cell showed an average reduction of extracellular HBV DNA of greater than 96% after 6 days of 3TC treatment. This finding was somewhat unexpected and clearly indicated that 0.2 μM 3TC was highly effective at inhibiting HBV replication even in the presence of large amounts of the virus. However, it is also important to note that even a 96% reduction in HBV replication still allowed high levels of HBV virions to be secreted from 3TC-treated cells which were replicating very high levels of HBV (i.e., cells infected with HBV baculovirus at a high MOI). We estimated that cells infected with 200 PFU of HBV baculovirus/cell were secreting approximately 1,070 pg of HBV DNA/60-mm-diameter plate/day at 4 days p.i. Cultures infected with 200 PFU/cell and treated with 0.2 μM 3TC for 3 days exhibited a 79.3% reduction in extracellular HBV DNA; however, even after this reduction, the cells were still secreting approximately 220 pg of HBV DNA/60-mm-diameter plate/day at this time.
We conclude from these studies that the HBV baculovirus-HepG2 system has specific advantages for drug studies and can serve as a complement to other in vitro model systems currently used for testing antiviral compounds. The results presented here agree well with previous reports of the efficacy of 3TC in reducing levels of extracellular HBV DNA in vitro. Different types of information can be obtained by using the HBV baculovirus-HepG2 system because experiments can be carried out at various levels of HBV replication, including levels significantly higher than those that can be obtained from conventional HBV-expressing cell lines. The ability to manipulate and treat cells prior to HBV infection should also aid in studying the properties and potential efficacy of antivirals as prophylactic agents. Finally, the enhanced ability to detect CCC HBV DNA in the HBV baculovirus-HepG2 system facilitates the in vitro study of a crucial form of the HBV genome, which has to be evaluated in developing any treatment protocols for curing HBV infection.