Effect of Antiviral Treatment with Entecavir on Ag
http://www.100md.com
病菌学杂志 2005年第9期
School of Molecular and Biomedical Science, University of Adelaide, North Terrace, Adelaide SA 5005, Australia
Infectious Diseases Laboratories, Institute of Medical and Veterinary Science, Frome Road, Adelaide SA 5000, Australia
Bristol-Myers Squibb Pharmaceutical Research Institute, Research Parkway, Wallingford, Connecticut 06492
ABSTRACT
Entecavir (ETV), a potent inhibitor of the hepadnaviral polymerases, prevented the development of persistent infection when administered in the early stages of duck hepatitis B virus (DHBV) infection. In a preliminary experiment, ETV treatment commenced 24 h before infection showed no significant advantage over simultaneous ETV treatment and infection. In two further experiments 14-day-old ducks were inoculated with DHBV-positive serum containing 104, 106, 108, or 5 x 108 viral genomes (vge) and were treated orally with 1.0 mg/kg of body weight/day of ETV for 14 or 49 days. A relationship between virus dose and infection outcome was seen: non-ETV-treated ducks inoculated with 104 vge had transient infection, while ducks inoculated with higher doses developed persistent infection. ETV treatment for 49 days did not prevent initial infection of the liver but restricted the spread of infection more than 1,000-fold, a difference which persisted throughout treatment and for up to 49 days after withdrawal. Ultimately, three of seven ETV-treated ducks resolved their DHBV infection, while the remaining ducks developed viremia and persistent infection after a lag period of at least 63 days. ETV treatment for 14 days also restricted the spread of infection, leading to marked and sustained reductions in the number of DHBV-positive hepatocytes in 7 out of 10 ducks. In conclusion, short-term suppression with ETV provides opportunity for the immune response to successfully control DHBV infection. Since DHBV infection of ducks provides a good model system for HBV infection in humans, it seems likely that ETV may be useful in postexposure therapy for HBV infection aimed at preventing the development of persistent infection.
INTRODUCTION
Hepatitis B virus (HBV) is a noncytopathic virus that primarily replicates in hepatocytes. HBV infection of humans results in either a transient infection with the development of immunity to reinfection or a persistent infection, which in adults can lead to chronic active hepatitis and, ultimately, liver failure. Both types of HBV infection often result in extensive infection of the liver, with high titers of infectious virus and noninfectious surface antigen particles circulating in the bloodstream. In people with transient HBV infection, the immune response is able to clear virus from infected cells and maintain immunity to reinfection.
The likelihood of persistent HBV infection varies with age, with >90% of neonatally infected individuals and only 5 to 10% of adults progressing to persistent infection. Vaccines are available for the prevention of HBV infection, and while universal vaccination of children is now being attempted in many parts of the world, adults are typically not vaccinated. For unvaccinated adults exposed to HBV, the most widely used treatment is with anti-HBV immunoglobulin, which is efficacious but expensive and not widely available outside major medical centers. Lamivudine, used for the treatment of chronic infections, has not been evaluated for prevention of primary infections in healthy adults, nor have other drugs that are currently under development.
One antiviral drug that is currently under clinical trial for the treatment of persistent HBV infection is entecavir (ETV; formerly known as BMS-200475). ETV is a cyclopentyl 2'-deoxyguanosine nucleoside that is highly inhibitory against hepadnavirus replication (1, 4, 6, 12). The triphosphate form of ETV directly inhibits hepadnaviral polymerases and has been shown to inhibit HBV replication in vitro, as well as duck HBV (DHBV) replication in ducks and woodchuck hepatitis virus replication in woodchucks (1, 4).
The ability of ETV to inhibit DHBV replication was previously examined in short-term studies in DHBV-infected primary duck hepatocyte cultures (12), and the antiviral activity of ETV (50% effective concentration, 0.13 nM) was shown to be >1,000-fold greater than lamivudine (50% effective concentration, 138 nM). Treatment of persistently DHBV-infected ducks for 21 days with 1 mg/kg/day of ETV resulted in a mean reduction of log10 3.1 in serum DHBV DNA levels, while 0.1 mg/kg/day resulted in a reduction of log10 2.1 (12). A long-term study (4) showed that treatment of persistently infected ducks with 0.1 mg/kg/day of ETV resulted in a rapid 4-log drop in serum DHBV DNA and a slower 2- to 3-log drop in serum DHBV surface antigen (DHBsAg). ETV treatment reduced both the intensity of antigen staining and the percentage of antigen-positive cells in the liver. However, a rapid rebound of levels of DHBV DNA and antigens in serum and liver to pretreatment levels was observed 42 to 49 days after ETV withdrawal, and DNA vaccination had no additional effect on this course of infection.
Like HBV-infected humans, DHBV-infected ducks exhibit age-related outcomes of infection, with the development of persistent infection in young ducks and transient infection in adults. These different outcomes of DHBV infection are viral dose dependent, with persistent infection developing in young ducks inoculated with higher doses of virus (8). Accordingly, these studies were performed using young ducks inoculated with various doses of DHBV and were undertaken to determine if postexposure ETV treatment could alter the outcome of DHBV infection in young animals by delaying virus spread until the developing immune system could gain control of the infection.
MATERIALS AND METHODS
Animals. One-day-old DHBV-negative ducks (Anas domesticus platyrhyncos) were obtained from a commercial hatchery and held in the animal house facilities of the Institute of Medical and Veterinary Science (IMVS), Adelaide. All animal handling procedures were approved by both the IMVS and University of Adelaide Animal Ethics Committees and followed the guidelines of the National Health and Medical Research Council of Australia.
Inoculum preparation. The virus inoculum used in all experiments was derived from pooled serum prepared from ducks congenitally infected with the Australian strain of DHBV (16) containing 9.5 x 109 viral genomes (vge)/ml (7). The viral inoculum was prepared by diluting DHBV-infected serum with uninfected duck serum to obtain the appropriate dose of DHBV in 100 μl and was inoculated intravenously (i.v.) via the jugular vein.
Drug preparation and administration. ETV was synthesized and supplied in a powdered form by Bristol-Myers Squibb. A 1.0-mg/ml stock solution of ETV was prepared by adding 10 mg of ETV to 10 ml of distilled water, followed by sonication in a 65°C water bath. Stock solutions of ETV were stored at 4°C for up to 3 days and allowed to reach room temperature before use. ETV or water was administered daily to ducks in each experiment by using a 2.0 cuffless oral/nasal tube attached to a 3-ml syringe.
Experiment 1: preliminary experiment to test the effect of 24-h ETV pretreatment. Three groups of 14-day-old ducks were inoculated i.v. with 108 vge of DHBV. One group was given a placebo of 2 ml/day of water, while the second group commenced a 7-day treatment with 1.0 mg/kg/day of ETV on the day of inoculation with DHBV. The third group was given a 1.0-mg/kg dose of ETV 24 h prior to virus inoculation, followed by daily treatment with 1.0 mg/kg of ETV for 7 days. All ducks were autopsied after 7 days and examined for the presence of DHBsAg-positive hepatocytes and DHBV DNA in the liver.
Experiment 2: 49-day ETV treatment. Three groups of eight 14-day-old ducks were inoculated i.v. with either 104, 106, or 108 vge. On the day of inoculation four ducks from each group were simultaneously started on daily oral treatment with 1.0 mg/kg of ETV for 49 days, while the remaining ducks were given a placebo of 2 ml/day of water. The groups are designated as 104 + ETV, 106 + ETV, 108 + ETV, 104 + water, 106 + water, and 108 + water. The experimental plan is summarized in Fig. 1A.
Experiment 3: 14-day ETV treatment. One group of 10 14-day-old ducks and a second group of 9 14-day-old ducks were inoculated with 108 and 5 x 108 vge, respectively. On the same day, five ducks from each group were started on daily treatment with 1.0 mg/kg ETV for 14 days, and the remaining ducks received a placebo of 2 ml/day of water. Groups are designated as 108 + ETV, 5 x 108 + ETV, 108 + water, and 5 x 108 + water. The experimental plan is summarized in Fig. 1B.
Analysis of serum samples. Blood samples were collected from the jugular vein, and serum was collected following incubation of the blood samples at 37°C for 90 min and then stored at –20°C. Serum samples were tested on the day of collection for levels of alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), and aspartate aminotransferase (AST), using an automatic analyzer in the Diagnostic Services Laboratories of the IMVS. The normal ranges for the enzymes in uninfected duck serum (mean ± standard deviation) were 26.6 ± 7.7 U/liter for ALT, 2.3 ± 1.2 U/liter for GGT, and 15.9 ± 5.9 U/liter for AST (4). Serum samples from experiments 2 and 3 were analyzed by quantitative enzyme-linked immunosorbent assay (ELISA) for DHBsAg and anticore and antisurface antibodies (4, 14).
Analysis of liver tissue. Biopsy samples of liver and autopsy samples of liver, kidney, pancreas, and spleen (from experiments 1, 2, and 3) were fixed in ethanol-acetic acid, wax embedded, sectioned, and examined for DHBsAg by immunoperoxidase staining as a marker of infected cells (8, 14). The nuclei of hepatocytes that stained positive for DHBsAg were counted in liver biopsy tissue with the aid of an eyepiece graticule in 200 250- by 250-μm grid fields (containing 100,000 hepatocytes) and were expressed as a percentage of the total hepatocyte nuclei. The minimum sensitivity of detection in liver biopsy tissue was 0.001%. In autopsy liver samples, 2,000 250- by 250-μm grid fields (containing 1,000,000 hepatocytes) were counted for each duck, resulting in a minimum sensitivity of detection of 0.0001%. Tissues were also fixed in 10% formaldehyde in phosphate-buffered saline, wax embedded, sectioned, and stained with hematoxylin and eosin for histological analysis and with Congo Red for detection of amyloid deposits in experiment 2. Formalin-fixed liver sections were examined under code for the presence and degree of inflammation and amyloid.
Autopsy liver samples from all ducks in experiment 1 and biopsy liver samples from ducks W33/34 and W45/46 in experiment 2 and from all ETV-treated ducks in experiment 3 were snap-frozen in liquid nitrogen and then stored at –80°C. Aliquots of 25 mg of each sample were extracted using a DNeasy tissue kit (QIAGEN), and DNA was eluted in 100 μl.
Real-time PCR was then performed using the Roche Lightcycler on extracted samples from experiments 1 and 2 using 100 ng of extracted DNA, quantitated using Hoechst dye. A head-to-tail dimer of DHBV DNA inserted into pBluescript II KS(+) (p4.8BLx2) (16) was used as a standard. The dimer was diluted to 90 ng/μl, giving 1010 copies of dimer per μl (2 x 1010 DHBV genomes/μl), and then diluted with DNA extracted from an uninfected normal duck liver to produce a standard curve of 108 to 101 copies of DHBV dimer in 10 μl, each supplemented with 100 ng of normal duck liver DNA. Each 20-μl reaction mixture contained 100 ng of DNA, 2 μl of Roche FastStart Master SYBR Green 1, 2.4 μl of MgCl2 (final concentration of 4 mM), and 0.5 μl of 10 μM primers and was made up to 20 μl with water. The forward primer for detection of total DHBV DNA was P3 at nucleotide (nt) 1316 (5'-AGCTGGCCTAATCGGATTAC-3'), and the reverse primer was P4 at nt 1584 (5'-TGTCCGTCAGATACAGCAAG-3'). The forward primer for detection of covalently closed circular DNA (cccDNA) was CC2 at nt 2462 (5'-CCTGATTGGACGGCTCTTAC-3'), and the reverse primer was R2 at nt 52 (5'-CCCGATCCAATGATTCCTCAT-3'). The positions of each primer are numbered within the Australian strain DHBV genome as described by Triyatni et al. (16). The PCR protocol required an initial incubation for 10 min at 95°C and then 40 cycles of 5 s at 95°C, 10 s at 55°C, and 15 s at 72°C. The DNA concentration was calculated by comparing the fluorescence of SYBR Green from samples, in the log-linear phase of amplification, to the standard curve of diluted dimer, and each dilution was run in duplicate. At the end of 40 cycles, a melting curve was generated showing the fluorescence from 60 to 95°C, allowing the specificity of each product to be checked.
Liver and serum samples from experiment 3 were also tested by PCR for the presence of DHBV DNA. Liver samples were extracted as above and 0.5 μl was used in a PCR, while 200 μl of serum was extracted using a Highpure viral DNA kit (Roche), eluted in 50 μl and 10 μl of the eluate put into a 50-μl PCR mixture. The PCR mixture also contained 5 μl of 10x reaction buffer (Geneworks, Adelaide), 3 μl of 25 mM MgCl2, 1 μl of deoxynucleoside triphosphates, 0.5 μl of Taq polymerase (Geneworks, Adelaide), and 1 μl of both forward and reverse primers and made up to 50 μl with water. The forward primer used was 2554 (5'-TTCGGAGCTGCCTGCCAAGG-3'), and the reverse primer was 269c (5'-GGAGCACCTGAGCTTGGATC-3'). The PCR protocol consisted of 10 min at 94°C, followed by 30 cycles of 20 s at 95°C, 30 s at 58°C, and 30 s at 72°C, and then 5 min at 72°C. Twenty microliters of product was then run on a 1.5% agarose gel and stained with ethidium bromide, and DHBV DNA was scored as positive or negative. The sensitivity of this assay was 103 copies of DHBV DNA in each reaction mixture.
RESULTS
Experiment 1: effect of 24-h pretreatment with ETV on DHBV infection in 14-day-old ducks. Because detailed kinetic data on the time to achieve ETV antiviral efficacy in ducks were not available, a preliminary experiment was carried out to determine if ETV treatment could be initiated at the time of infection and whether greater efficacy was obtained by initiating treatment 24 h prior to infection. Non-ETV-treated control ducks inoculated with 108 vge had DHBsAg expression in 35, 93, and >95% of hepatocytes on day 7 postinfection (p.i.) (Fig. 2A and B; Table 1). In contrast, ducks with ETV treatment that was commenced simultaneously with (Fig. 2C and D) or 24 h prior to (Fig. 2E and F) DHBV infection had 0.0004 to 0.011% and 0.0058 to 0.029% DHBsAg-positive cells, respectively, on day 7 p.i. (Table 1). Clearly, commencing ETV treatment 24 h prior to DHBV infection showed no significant advantage over ETV treatment commenced at the same time as infection. This finding suggests that ETV treatment administered either simultaneously or 24 h prior to DHBV infection was unable to prevent the conversion of the input relaxed circular (RC) DNA genome into cccDNA followed by expression of viral antigens from the cccDNA template. However, given the rapid phosphorylation of ETV and a possible lag time in conversion of RC to cccDNA, we cannot rule out that ETV treatment may partially block conversion of RC to cccDNA, thus restricting initial infection of the liver and contributing to the lower numbers of DHBsAg-positive hepatocytes observed on day 7 p.i. In an attempt to measure the inhibition of DHBV replication by ETV and to determine if cells expressing DHBsAg were replicating DHBV or expressing DHBsAg from a cccDNA template in the absence of DHBV replication, quantitative PCR was performed to determine the relative amounts of DHBV total DNA and cccDNA present in the samples collected on day 7 p.i. As can be seen in Table 1, the non-ETV-treated control ducks had ratios of total DHBV DNA to cccDNA of 29 to 67, similar to those observed in other studies of DHBV infection (9). In contrast the ETV-treated ducks had ratios of total DHBV to cccDNA of 1.5 to 10, indicating that ETV treatment inhibited virus replication, resulting in reduced levels of replicative intermediates, and reduced ratios of total DHBV DNA to cccDNA. The reduced ratios of total DHBV DNA to cccDNA in the ETV-treated ducks also suggest that cells containing cccDNA and expressing detectable DHBsAg may not be supporting viral replication.
Experiment 2: 49-day ETV treatment. Dose-related outcomes of DHBV infection were seen in control ducks in experiment 2. As expected from previous work (8), 14-day-old ducks inoculated with 104 vge had transient DHBV infections. In liver tissue collected on day 7 p.i., DHBV infection was detected in 0.02 to 0.11% of hepatocytes (average of 0.06%) (Table 2) by immunoperoxidase staining of DHBsAg present in the cytoplasm of infected cells (Fig. 3A). DHBV-infected cells were no longer detected on day 35 p.i. (Fig. 3B; Table 2) or on days 98 or 182 p.i., indicating resolution of DHBV infection. This group of ducks also had no detectable DHBsAg in the serum (Fig. 4A) but developed detectable levels of antisurface and anticore antibodies by day 11 to 20 p.i. (Fig. 5A and 6A). Titers of antisurface and anticore antibodies peaked shortly after infection, followed by a general decrease to relatively stable levels throughout the rest of the experiment.
Also in agreement with previous findings (8), in 14-day-old ducks inoculated with 106 or 108 vge DHBV rapidly spread throughout the liver, resulting in persistent DHBV infection. In liver tissue collected on day 7 p.i., DHBV infection was detected in 0.32 to 6.0% of hepatocytes (average, 2.93%) (Fig. 3C; Table 2) following inoculation of 106 vge. By day 35 p.i., DHBV infection had spread throughout the entire liver, resulting in infection of >95% of hepatocytes (Fig. 3D; Table 2). Ducks inoculated with 108 vge had >95% of hepatocytes infected on day 7 p.i., a level which was maintained to day 35 p.i. (Fig. 3E and F; Table 2).
The number of cells infected on day 7 p.i. after inoculation with each dose of virus was within the expected range. This is based on estimates of the number of cells per gram of duck liver (7 x 108) (9) and the liver weight of 14-day-old ducks (measured as 20.1 g), giving an estimated total number of 1.4 x 1010 liver cells. Accordingly, we could expect that i.v. inoculation of 104, 106, and 108 vge would infect 0.000071, 0.0071, and 0.71% of total liver cells, provided all the inoculum was delivered to the liver in each case. Cell-to-cell spread of infection throughout the liver with a reported doubling time of 16 h (7) should result in infection of an estimated 0.098, 9.8, and >98% of hepatocytes on day 7 p.i. The observed percentages of infected cells on day 7 p.i. in ducks inoculated with 104, 106, and 108 vge were within this range, with DHBsAg detection in an average of 0.06, 2.93, and >95% of hepatocytes.
Infection of the liver also resulted in the appearance of DHBsAg in the serum of infected ducks, first detected on day 11 p.i. in the 106 vge group (Fig. 4C) and day 7 p.i. in the 108 vge group (Fig. 4E). Levels of DHBsAg ranged from 0.4 to 190 μg/ml. The difference in the timing of appearance of DHBsAg reflects the 100-fold difference in the viral dose and the extent of infection observed in the liver. As previously reported (8, 13), levels of DHBsAg in the serum showed a biphasic pattern, with a short-term appearance then disappearance before DHBsAg became continuously detectable for the remainder of the study.
Serum DHBsAg persisted throughout the experiment in six out of eight ducks but became undetectable in two ducks (W33/34 and W45/46) after day 60 and 98 p.i., respectively. Both of these ducks showed an absence of DHBsAg-positive hepatocytes in the liver when tested on day 182 p.i. (Table 2). Liver DHBV DNA levels determined by real-time PCR (data not shown) in duck W33/34 also showed a >3-log decrease from day 35 to day 172 p.i. In duck W45/46, liver DHBV DNA levels decreased from 1,900,000 to 5,500 copies per 87,500 cells (21.7 to 0.062 copies per cell) (data not shown), representing a 3-log reduction from day 35 to day 98 p.i. The large decrease in markers of infection seen in these two ducks indicates spontaneous clearance of DHBV infection, although the markers examined showed no characteristics that allowed these ducks to be distinguished from those that maintained a high-level persistent infection.
Antisurface and anticore antibodies were detected at variable levels in the serum of the ducks inoculated with 106 and 108 vge from day 11 to 20 p.i. and persisted throughout the study (Fig. 5C and E and 6C and E). Antisurface antibodies were in some cases detected in the presence of DHBsAg, presumably reflecting the presence of immune complexes. Levels were variable, both between individual animals and within individuals over time. Anticore antibodies were seen in two out of four ducks inoculated with 104 vge, with levels variable in the early part of infection (Fig. 6A). Anticore responses were seen in all ducks in the 106 and 108 groups (Fig. 6C and E). A general trend for increasing titers over time was seen in the 106 + water group, while a more variable response was seen in the 108 + water group both between individuals and within individuals over time.
ETV treatment restricted the spread of virus infection in the liver and delayed the rebound of DHBV infection. ETV treatment did not prevent initial DHBV infection of the liver but resulted in marked reductions in the number of DHBsAg-positive hepatocytes detected in the liver on day 7 p.i. No DHBsAg-positive hepatocytes were detected in the liver of ducks inoculated with 104 vge at any time after inoculation (Table 2). These ducks had no detectable DHBsAg in the serum (Fig. 4B) but developed low levels of antisurface and anticore antibodies from 70 to 120 days p.i. (Fig. 5B and 6B). The observed delay in detectable levels of antisurface and anticore antibodies may reflect the initial restriction in the spread of DHBV infection and antigen expression in the liver, followed by further replication of virus and expression of viral antigens after the removal of drug, which stimulates the production of antibodies to detectable levels and results in immune-mediated clearance of infected cells.
Similarly, in ducks inoculated with 106 vge the number of DHBsAg-positive hepatocytes was markedly reduced compared to the non-ETV-treated ducks, with DHBsAg expression on day 7 p.i. detected in two out of four ETV-treated ducks (W9/10 with 0.002% and W11/12 with 0.001% of hepatocytes) (Table 2). The percentage of DHBsAg-positive hepatocytes detected in the liver on day 7 p.i. was thus reduced from an average of 2.93% in the non-ETV-treated control ducks to 0.0015% of hepatocytes detected in two out of four ETV-treated ducks, representing a greater-than-1,000-fold reduction in the percentage of infected cells.
Ducks inoculated with 108 vge and treated with ETV had an average of 0.046% of liver cells (range, 0.013 to 0.08%) (Table 2) infected on day 7 p.i. (Fig. 7A and E) compared to >95% of cells in the non-ETV-treated ducks (Fig. 3E), representing a 2,000-fold reduction in the percentage of infected cells.
Based on the calculations presented above for non-ETV-treated control ducks, we predicted that i.v. inoculation of doses of 104, 106, and 108 vge would initially infect 0.000071, 0.0071, and 0.71% of total liver cells on the day of inoculation. The observed numbers of DHBsAg-positive cells in the ETV-treated ducks on day 7 p.i. was at least 10-fold lower than these estimates, with DHBsAg detection in <0.001% (all four ducks), 0.002% and 0.001% (two of the four ducks), and an average of 0.046% of cells (range, 0.013 to 0.080% in all four ducks) in the three dose groups. This observation highlights the ability of ETV to restrict the cell-to-cell spread of virus infection in the liver by inhibiting virus replication during the 7 days after virus inoculation. Although direct evidence was not obtained in experiment 1 for the ability of ETV to inhibit conversion of the RC DNA genome into cccDNA during the initiation of infection, we cannot rule out that ETV may restrict initial infection of the liver and that this might also contribute to the reduced levels of DHBsAg-positive hepatocytes on day 7 p.i.
The percentage of DHBsAg-positive cells remained low at day 35 p.i. during ETV treatment (Fig. 7B and F), but DHBV-positive cells were still detected, indicating that DHBV-infected cells containing the transcriptional template of the virus, cccDNA, persisted in the liver in the presence of ETV treatment.
Forty-nine days after the withdrawal of ETV, at day 98 p.i., DHBsAg-positive cells were detected in five out of seven ETV-treated ducks inoculated with 106 or 108 vge, with four out of five animals having less than 0.1% DHBsAg-positive hepatocytes (Fig. 7C and G; Table 2). The exception was duck W11/12, inoculated with 106 vge, in which 2.17% of hepatocytes were DHBsAg positive. Unfortunately, this duck died of unknown causes prior to day 182 p.i. However, based on the percentage of infected liver cells at day 98 p.i., it is likely that this duck would have developed a persistent DHBV infection. By autopsy at day 182 p.i., no DHBsAg-positive hepatocytes were detected in ducks W13/14 and W15/16 from the 106 + ETV group or duck W23/24 from the 108 + ETV group (Fig. 7D; Table 2). These three ducks had evidently cleared their DHBV infection.
In contrast, in one out of three ducks (W9/10) from the 106 + ETV group and in two out of three ducks (W17/18 and W19/20) from the 108 + ETV group, DHBV infection had spread throughout the liver to infect >95% of hepatocytes (Fig. 7H; Table 2).
The different outcomes of DHBV infection are illustrated in Fig. 7, where two ducks inoculated with 108 vge and treated with ETV (W23/24 and W19/20) had DHBsAg-positive hepatocytes detectable in the liver on days 7, 35, and 98 p.i., shown as individual cells or, in one case, as pairs of DHBsAg-positive cells (Fig. 7A to C and E to G). However, while cells expressing DHBsAg were no longer detected in the liver of duck W23/24 at autopsy on day 182 p.i. (Fig. 7D), in duck W19/20 DHBV infection had spread to infect the entire hepatocyte population (Fig. 7H). In summary, three out of seven ducks infected with 106 and 108 vge and simultaneously treated with ETV apparently cleared their DHBV infection, while four out of seven ducks, including duck W11/12 that died unexpectedly prior to day 182 p.i., developed persistent DHBV infection. The reasons for these different outcomes of infection are not known but suggest that infection was not efficiently controlled by the immune response.
ETV treatment prevented the development of detectable levels of serum DHBsAg during treatment and for an extended period after ETV withdrawal. As expected from the low levels of virus infection in the liver observed during ETV treatment, serum DHBsAg was not detected in any ETV-treated ducks during the course of ETV treatment (Fig. 4B, D, and F) and was only detected in three out of nine ETV-treated ducks after ETV treatment was withdrawn, coinciding with increases in the number of DHBV-infected cells in the liver. In one of three ducks (W19/20, inoculated with 108 vge), serum DHBsAg was first detected on day 112 p.i., 63 days after ETV treatment was withdrawn. A second duck from the 108 + ETV group (W17/18) had DHBsAg detectable in serum from day 126 p.i., 77 days after ETV withdrawal, until the end of the study, and duck W9/10 from the 106 + ETV group had detectable serum DHBsAg on day 147 p.i., 98 days after ETV treatment (Fig. 4D and F).
Antiviral antibodies were delayed in ETV-treated ducks. Antisurface antibodies were not detected in any ETV-treated ducks during the course of treatment (Fig. 5B, D, and F). However, from day 70 p.i., 21 days after ETV treatment was withdrawn, low titers of antisurface antibodies were detected in two out of four ducks inoculated with 104 vge (W1/2 and W7/8) and one out of four ducks inoculated with 108 vge (W23/24). Antisurface antibodies developed in all the remaining ETV-treated ducks during the course of the experiment, with antibodies appearing earlier in the ducks that had no DHBsAg evident in the serum (Fig. 5B, D, and F). In the three ducks that developed viremia (W9/10, W17/18, and W19/20), antisurface antibodies were detected from the time of DHBsAg detection or 7 days later.
Anticore antibodies were detected in one ETV-treated duck (W19/20, 108 + ETV) in the last week of ETV treatment, and this duck went on to develop persistent DHBV infection (Fig. 6F). Anticore responses were low in the remaining ETV-treated ducks (Fig. 6B, D, and F) and first appeared on day 98 p.i., 49 days after the withdrawal of ETV in the 104 + ETV group, and on day 126 p.i., 77 days or more after ETV withdrawal in the 106 and 108 + ETV groups. Not all ducks developed anticore antibodies, and they were detectable at only one to three time points in those that did produce any. The late appearance of antisurface and anticore antibodies, some time after cessation of ETV treatment, suggests that higher levels of antigen, produced by spread of infection after ETV treatment, are required to induce a measurable humorable immune response.
Liver enzyme levels were unaffected by ETV treatment. Levels of the liver function enzymes ALT, GGT, and AST (Fig. 8A, B, and C, respectively) were similar for all ducks, with some individual variation seen, especially in the GGT levels. Two ducks from the 108 + water group had 40-fold-higher-than-normal levels of GGT at day 133 p.i., and 35-fold-higher levels were seen in a duck from the 106 + ETV group at day 161 p.i. (Fig. 8B). These increases occurred after the ducks were 20 weeks old, when egg laying commences, and similar flares have been seen in noninfected and/or nontreated ducks of that age (3, 4). The average values for liver enzymes, excluding the extreme GGT values, were as follows: ALT, 25.8 ± 10 U/liter; GGT, 2.2 ± 1.2 U/liter; AST, 15.5 ± 6.5 U/liter. These values were all within previously determined normal limits for uninfected ducks (4).
One duck from the 108 + ETV group (W17/18) that developed persistent DHBV infection showed a fourfold increase in ALT levels in the last 40 days of the experiment (Fig. 8A). This ALT flare occurred around the time that DHBsAg and DHBV DNA first appeared in the serum. This duck also showed signs of amyloidosis at days 98 and 172 p.i., and slight inflammation was observed at day 172 p.i. No other animals showed signs of amyloidosis. Rates of weight gain showed no difference between the ETV-treated and control ducks (Fig. 8D).
Liver histology and duck health. All DHBV-infected ducks showed mild liver inflammation, and this was unaffected by ETV treatment. During the experiment three ducks died, one from each ETV treatment group. Duck W21/22 was euthanized on day 22 due to congenital leg deformities. Duck W3/4 was euthanized on day 26 due to severe bacterial septicemia. Duck W11/12 was found dead in its cage from an undetermined cause at day 105 p.i., 56 days after ETV treatment was withdrawn. None of the deaths were directly related to ETV therapy. The detection of DHBsAg-positive hepatocytes in these three ducks has been included in Table 2, but serological profiles have not been presented.
Experiment 3: 14-day ETV treatment. We wished to determine if a shorter treatment with ETV could lead to resolution of infection in a similar percentage of ETV-treated ducks. Fourteen-day-old ducks were infected with 108 and 5 x 108 vge of DHBV and were used as untreated control ducks or were treated simultaneously with 1.0 mg/kg of ETV for 14 days. As observed in experiment 2, DHBV infection rapidly spread throughout the liver of the control ducks injected with either 108 or 5 x 108 vge to infect >95% of hepatocytes by day 14, a level maintained to day 70 p.i. (Table 3). In contrast, although treatment with ETV for 14 days did not prevent the establishment of a low-level DHBV infection, it did restrict the spread of DHBV infection in the liver. Liver tissue collected at the end of 14 days of treatment from ducks infected with 108 or 5 x 108 vge had 0.0031 to 0.010% and 0.014 to 0.14% DHBsAg-positive hepatocytes, respectively (Table 3). This difference was thought to result from the fivefold difference in the size of the viral inoculum.
Fifty-six days after ETV treatment was withdrawn (day 70 p.i.), 7 out of 10 ducks had either undetectable (<0.0001%) or low numbers (0.0006 to 0.024%) of DHBsAg-positive hepatocytes (Table 3). In the remaining 3 out of 10 ducks, DHBV infection had spread to involve the entire hepatocyte population.
Serum DHBsAg and DHBV DNA were detected in the three ETV-treated ducks that developed a persistent infection (B13, B15, and B16) at day 49 p.i. and were detected in one of these ducks until day 63 p.i. (Table 3). The serum of a fourth ETV-treated duck (B12) also contained serum DHBsAg and DHBV DNA at day 49 p.i., but these markers were not detected at subsequent time points (Table 3). This duck had no detectable DHBsAg-positive cells in the liver at autopsy and appeared to have had a short-term DHBV infection that rebounded after drug treatment was withdrawn but that was subsequently cleared from the liver (Table 3). No other ETV-treated ducks had detectable DHBsAg or DHBV DNA in their serum during the experiment.
DISCUSSION
As expected from previous work (8), 14-day-old ducks infected with DHBV showed different outcomes of DHBV infection depending on the size of the viral inoculum. Control ducks inoculated with 104 vge had a transient infection, while ducks inoculated with 106, 108, or 5 x 108 vge developed persistent DHBV infections. The reason for the spontaneous decline in markers of infection in two of the infected ducks (W33/34 and W45/46) was not evident from the data obtained but was consistent with observations in patients infected with HBV. Spontaneous clearance of chronic HBV infections in humans is seen in a low percentage of cases, but the cause is unknown (2), although it has been found to be associated with age (10). Several previous DHBV infection experiments in this laboratory using the same age and dose relationship were carried out for less than 60 days, and no decreases in markers of infection were observed (8).
ETV treatment resulted in an altered course of infection, with an obvious delay in the development of markers of DHBV infection and/or detectable antibody responses. ETV treatment commencing at the time of DHBV infection was not effective in preventing DHBV infection of hepatocytes but led to significant suppression of viral spread. ETV treatment extended the period of low-level infection, which may have allowed for natural clearance of the virus in some individuals and could have complemented or facilitated the use of other methods designed to enhance immune responsiveness to DHBV and, possibly, HBV.
ETV is rapidly absorbed and converted to the triphosphate form (20), with peak plasma levels in ducks occurring 30 min after a 5-mg/kg dose (198.05 ng/ml) and with a half-life in plasma of 8.73 h (4). Although ETV is absorbed and converted rapidly to the triphosphate form, simultaneous treatment of ducks with ETV did not prevent conversion of the input RC DHBV genome into cccDNA with resulting expression of DHBsAg within hepatocytes. The ability of ETV to prevent conversion of RC DNA to cccDNA was further tested in experiment 1 by the pretreatment of ducks with ETV 24 h prior to DHBV infection, but this did not significantly alter the number of cells infected at day 7 p.i.
Since ETV acts via a competitive inhibition with dGTP, its ability to bind and block the conversion of RC DHBV genomes to cccDNA is likely to depend on the size of the gap in the positive strand of the RC DNA genome. In DHBV, the 3' end of the positive strand terminates immediately upstream of the DR2 sequence, where the RNA primer binds (11). In the Australian strain of DHBV (16) used in these experiments, there is one G base at position 2531 of the RNA primer, but if the termination occurs further upstream there are G bases at positions 2529, 2528, and 2526, to which ETV could bind and block the formation of cccDNA. In support of our data, treatment with another guanosine analogue, 2'-deoxyguanosine, for 24 h prior to DHBV infection of primary duck hepatocytes was also unable to block conversion of RC to cccDNA (5). ETV may be more effective in blocking cccDNA formation in HBV or woodchuck hepatitis virus infection, as those genomes have larger incomplete regions in the positive strand than the DHBV genome.
DHBV DNA remained present throughout ETV treatment and is likely to be dormant or nonreplicating, as it appeared to be insensitive to inhibition. cccDNA forms a stable template for hepadnavirus replication, making it difficult to eliminate through treatment with antiviral drugs (1, 19, 21), and would provide a mechanism for the virus to evade antiviral drug effects. Clearance of cccDNA from cells through cell division, cell death, or possibly noncytolytically via cytokine-induced mechanisms would be required to achieve viral clearance and stop the rebound commonly seen after the cessation of antiviral therapy.
In ETV-treated ducks, the absence of detectable antibodies during treatment and the lower-than-control levels after ETV withdrawal indicate their humoral immune responses were not well developed, possibly because the amount of antigen present was too low during the ETV treatment phase. Virus replication, if it did occur during drug treatment, was not at levels sufficient to stimulate an effective humoral immune response, as evidenced by the increase in infected cells after drug withdrawal. It seems possible that commencing treatment with ETV at the time of infection might have either suppressed the rapid replication of the virus, thereby allowing immune responses to control the infection, or to have held the virus at low, nonimmunogenic levels until the ducks developed a more mature immune system, so that an effective immune response could be mounted once the drug was withdrawn. Either of these outcomes would have improved the likelihood of viral clearance.
The four ETV-treated ducks that progressed to persistent infection in experiment 2 had low levels of DHBsAg-positive cells in the liver at days 35 and 98 p.i. This raises the questions of whether and how these cells evaded the immune response or whether infected cells were actually being removed and replaced by others, due to a continuing low rate of new infection of hepatocytes. The individual DHBsAg-positive cells had levels of cytoplasmic DHBsAg comparable to the controls, but it is not known if this expression was generated from low copy numbers of cccDNA, or integrated DNA, or if the virus was replicating within these cells. In situ hybridization could answer this question, but currently assays are not adequate to provide a clear answer. PCR did not provide evidence of replicating DNA in excess of total cccDNA in the liver.
The presence of both DHBsAg and antisurface antibodies may have caused some of the variability seen in levels of DHBsAg and antisurface antibodies, due to the varying amounts present in the form of antigen-antibody complexes. The presence of circulating antisurface antibodies is generally thought to be protective, and in HBV it is recognized as a marker of immunity and of the resolution of disease. However, complexes of HBsAg and immunoglobulin have been detected in acute HBV infection (17) and in healthy carriers and in patients with symptoms of chronic infection (18). The presence of these complexes has been related to HBV infection outcomes, with an appearance time of 1 to 27 weeks seen in acute infection and long-term persistence of complexes in chronically infected individuals (15).
In conclusion, ETV shows promise as a postexposure therapy for HBV infection. Although ETV treatment alone was not sufficient to lead to DHBV clearance in all ducks, significant inhibition of viral replication occurred, leading to different infection outcomes. This study showed that ETV is a safe, potent suppressor of DHBV replication, although it is unable to stop the development of low-level infection of hepatocytes when given simultaneously with or 24 h prior to DHBV. These results have provided a basis for future work on immune therapies designed to develop a therapeutic protocol for treatment of patients chronically infected with HBV.
ACKNOWLEDGMENTS
This research was supported by the National Health and Medical Research Council of Australia and by a research grant from Bristol-Myers Squibb Pharmaceutical Research Institute.
We appreciate the work of the IMVS animal care staff for the day-to-day care of the ducks involved in this study. We are also grateful to Christopher Burrell for helpful suggestions and to William Mason for critical reading of the manuscript.
REFERENCES
Colonno, R. J., E. V. Genovesi, I. Medina, L. Lamb, S. Durham, M. L. Huang, L. Corey, M. Littlejohn, S. Locarnini, B. C. Tennant, B. Rose, and J. M. Clark. 2001. Long-term entecavir treatment results in sustained antiviral efficacy and prolonged life span in the woodchuck model of chronic hepatitis infection. J. Infect. Dis. 184:1236-1245.
Da Silva, L. C., C. L. Madruga, F. J. Carrilho, J. R. Pinho, A. Saez-Alquezar, C. Santos, L. Bassit, C. Barreto, L. E. Fonseca, V. A. Alves, R., Leitao, R. Vianna, R. A. Cardoso, A. V. Franca, and L. C. Gayotto. 1996. Spontaneous hepatitis B surface antigen clearance in a long-term follow-up study of patients with chronic type B hepatitis. Lack of correlation with hepatitis C and D virus superinfection. J. Gastroenterol. 31:696-701.
Fairbrother, A., M. A. Craig, K. Walker, and D. O'Loughlin. 1990. Changes in mallard (Anas platyrhynchos) serum chemistry due to age, sex, and reproductive condition. J. Wildlife Dis. 26:67-77.
Foster, W. K., D. S. Miller, P. L. Marion, R. J. Colonno, I. Kotlarski, and A. R. Jilbert. 2003. Entecavir therapy combined with DNA vaccination for persistent duck hepatitis B virus infection. Antimicrob. Agents Chemother. 47:2624-2635.
Fourel, I., J. Saputelli, P. Schaffer, and W. S. Mason. 1994. The carbocyclic analog of 2'-deoxyguanosine induces a prolonged inhibition of duck hepatitis B virus DNA synthesis in primary hepatocyte cultures and in the liver. J. Virol. 68:1059-1065.
Genovesi, E. V., L. Lamb, I. Medina, D. Taylor, M. Seifer, S. Innaimo, R. J. Colonno, D. N. Standring, and J. M. Clark. 1998. Efficacy of the carbocyclic 2'-deoxyguanosine nucleoside BMS-200475 in the woodchuck model of hepatitis B virus infection. Antimicrob. Agents Chemother. 42:3209-3217.
Jilbert, A. R., D. S. Miller, C. A. Scougall, H. Turnbull, and C. J. Burrell. 1996. Kinetics of duck hepatitis B virus infection following low dose virus inoculation: one virus DNA genome is infectious in neonatal ducks. Virology 226:338-345.
Jilbert, A. R., J. A. Botten, D. S. Miller, E. M. Bertram, P. Hall, I. Kotlarski, and C. J. Burrell. 1998. Characterization of age- and dose-related outcomes of duck hepatitis B virus infection. Virology 243:273-282.
Jilbert, A. R., T. T. Wu, J. M. England, P. de la M. Hall, N. Z. Carp, A. P. O'Connell, and W. S. Mason. 1992. Rapid resolution of duck hepatitis B virus infections occurs after massive hepatocellular involvement. J. Virol. 66:1377-1388.
Kato, Y., K. Nakao, K. Hamasaki, H. Kato, K. Nakata, Y. Kusumoto, and K. Eguchi. 2000. Spontaneous loss of hepatitis B surface antigen in chronic carriers, based on a long-term follow-up study in Goto Islands, Japan. J. Gastroenterol. 35:201-205.
Lien, J. M., D. J. Petcu, C. E. Aldrich, and W. S. Mason. 1987. Initiation and termination of duck hepatitis B virus DNA synthesis during virus maturation. J. Virol. 61:3832-3840.
Marion, P. L., F. H. Salazar, M. A. Winters, and R. J. Colonno. 2001. Potent efficacy observed with entecavir (BMS-200475) in duck model of hepatitis B virus replication. Antimicrob. Agents Chemother. 46:82-88.
Meier, P., C. A. Scougall, H. Will, C. J. Burrell, and A. R. Jilbert. 2003. A duck hepatitis B virus strain with a knock out mutation in the putative X ORF shows similar infectivity and growth characteristics to wild type. Virology 317:291-298.
Miller, D. S., E. M. Bertram, C. A. Scougall, I. Kotlarski, and A. R. Jilbert. 2004. Studying host immune responses against duck hepatitis B virus infection. Methods Mol. Med. 96:3-28.
Surelia, P., and E. H. Boxall. 1990. Hepatitis B virus infection: detection of circulating HBsAg/IgM antibody immune complexes. Med. Lab. Sci. 47:204-209.
Triyatni, M., P. L. Ey, H. Tran, M. F. Le Mire, M. Qiao, C. J. Burrell, and A. R. Jilbert. 2001. Sequence comparison of an Australian duck hepatitis B virus strain with other avian hepadnaviruses. J. Gen. Virol. 82:373-378.
Tsai, J. F., H. S. Margolis, J. E. Jeng, M. S. Ho, W. Y. Chang, M. Y. Hsieh, Z. Y. Lin, and J. H. Tsai. 1996. Hepatitis B surface antigen- and immunoglobulin-specific circulating immune complexes in acute hepatitis B virus infection. Clin. Immunol. Immunopathol. 80:278-282.
Tsai, J. F., H. S. Margolis, J. E. Jeng, M. S. Ho, W. Y. Chang, M. Y. Hsieh, Z. Y. Lin, and J. H. Tsai. 1998. Immunoglobulin- and hepatitis B surface antigen-specific circulating immune complexes in chronic hepatitis B virus infection. Clin. Immunol. Immunopathol. 86:246-251.
Werle-Lapostolle, B., S. Bowden, S. Locarnini, K. Wursthorn, J. Petersen, G. Lau, C. Trepo, P. Marcellin, Z. Goodman, W. E. Delaney, S. Xiong, C. L. Brosgart, S. S. Chen, C. S. Gibbs, F. Zoulim. 2004. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 126:1750-1758.
Yamanaka, G., T. Wilson, S. Innaimo, G. S. Bisacchi, P. Egli, J. K. Rinehart, R. Zahler, and R. J. Colonno. 1999. Metabolic studies on BMS-200475, a new antiviral compound active against hepatitis B virus. Antimicrob. Agents Chemother. 43:190-193.
Zhu, Y., T. Yamamoto, J. Cullen, J. Saputelli, C. E. Aldrich, D. S. Miller, S. Litwin, P. A. Furman, A. R. Jilbert, and W. S. Mason. 2001. Kinetics of hepadnavirus loss from the liver during inhibition of viral DNA synthesis. J. Virol. 75:311-322.(Wendy K. Foster, Darren S)
Infectious Diseases Laboratories, Institute of Medical and Veterinary Science, Frome Road, Adelaide SA 5000, Australia
Bristol-Myers Squibb Pharmaceutical Research Institute, Research Parkway, Wallingford, Connecticut 06492
ABSTRACT
Entecavir (ETV), a potent inhibitor of the hepadnaviral polymerases, prevented the development of persistent infection when administered in the early stages of duck hepatitis B virus (DHBV) infection. In a preliminary experiment, ETV treatment commenced 24 h before infection showed no significant advantage over simultaneous ETV treatment and infection. In two further experiments 14-day-old ducks were inoculated with DHBV-positive serum containing 104, 106, 108, or 5 x 108 viral genomes (vge) and were treated orally with 1.0 mg/kg of body weight/day of ETV for 14 or 49 days. A relationship between virus dose and infection outcome was seen: non-ETV-treated ducks inoculated with 104 vge had transient infection, while ducks inoculated with higher doses developed persistent infection. ETV treatment for 49 days did not prevent initial infection of the liver but restricted the spread of infection more than 1,000-fold, a difference which persisted throughout treatment and for up to 49 days after withdrawal. Ultimately, three of seven ETV-treated ducks resolved their DHBV infection, while the remaining ducks developed viremia and persistent infection after a lag period of at least 63 days. ETV treatment for 14 days also restricted the spread of infection, leading to marked and sustained reductions in the number of DHBV-positive hepatocytes in 7 out of 10 ducks. In conclusion, short-term suppression with ETV provides opportunity for the immune response to successfully control DHBV infection. Since DHBV infection of ducks provides a good model system for HBV infection in humans, it seems likely that ETV may be useful in postexposure therapy for HBV infection aimed at preventing the development of persistent infection.
INTRODUCTION
Hepatitis B virus (HBV) is a noncytopathic virus that primarily replicates in hepatocytes. HBV infection of humans results in either a transient infection with the development of immunity to reinfection or a persistent infection, which in adults can lead to chronic active hepatitis and, ultimately, liver failure. Both types of HBV infection often result in extensive infection of the liver, with high titers of infectious virus and noninfectious surface antigen particles circulating in the bloodstream. In people with transient HBV infection, the immune response is able to clear virus from infected cells and maintain immunity to reinfection.
The likelihood of persistent HBV infection varies with age, with >90% of neonatally infected individuals and only 5 to 10% of adults progressing to persistent infection. Vaccines are available for the prevention of HBV infection, and while universal vaccination of children is now being attempted in many parts of the world, adults are typically not vaccinated. For unvaccinated adults exposed to HBV, the most widely used treatment is with anti-HBV immunoglobulin, which is efficacious but expensive and not widely available outside major medical centers. Lamivudine, used for the treatment of chronic infections, has not been evaluated for prevention of primary infections in healthy adults, nor have other drugs that are currently under development.
One antiviral drug that is currently under clinical trial for the treatment of persistent HBV infection is entecavir (ETV; formerly known as BMS-200475). ETV is a cyclopentyl 2'-deoxyguanosine nucleoside that is highly inhibitory against hepadnavirus replication (1, 4, 6, 12). The triphosphate form of ETV directly inhibits hepadnaviral polymerases and has been shown to inhibit HBV replication in vitro, as well as duck HBV (DHBV) replication in ducks and woodchuck hepatitis virus replication in woodchucks (1, 4).
The ability of ETV to inhibit DHBV replication was previously examined in short-term studies in DHBV-infected primary duck hepatocyte cultures (12), and the antiviral activity of ETV (50% effective concentration, 0.13 nM) was shown to be >1,000-fold greater than lamivudine (50% effective concentration, 138 nM). Treatment of persistently DHBV-infected ducks for 21 days with 1 mg/kg/day of ETV resulted in a mean reduction of log10 3.1 in serum DHBV DNA levels, while 0.1 mg/kg/day resulted in a reduction of log10 2.1 (12). A long-term study (4) showed that treatment of persistently infected ducks with 0.1 mg/kg/day of ETV resulted in a rapid 4-log drop in serum DHBV DNA and a slower 2- to 3-log drop in serum DHBV surface antigen (DHBsAg). ETV treatment reduced both the intensity of antigen staining and the percentage of antigen-positive cells in the liver. However, a rapid rebound of levels of DHBV DNA and antigens in serum and liver to pretreatment levels was observed 42 to 49 days after ETV withdrawal, and DNA vaccination had no additional effect on this course of infection.
Like HBV-infected humans, DHBV-infected ducks exhibit age-related outcomes of infection, with the development of persistent infection in young ducks and transient infection in adults. These different outcomes of DHBV infection are viral dose dependent, with persistent infection developing in young ducks inoculated with higher doses of virus (8). Accordingly, these studies were performed using young ducks inoculated with various doses of DHBV and were undertaken to determine if postexposure ETV treatment could alter the outcome of DHBV infection in young animals by delaying virus spread until the developing immune system could gain control of the infection.
MATERIALS AND METHODS
Animals. One-day-old DHBV-negative ducks (Anas domesticus platyrhyncos) were obtained from a commercial hatchery and held in the animal house facilities of the Institute of Medical and Veterinary Science (IMVS), Adelaide. All animal handling procedures were approved by both the IMVS and University of Adelaide Animal Ethics Committees and followed the guidelines of the National Health and Medical Research Council of Australia.
Inoculum preparation. The virus inoculum used in all experiments was derived from pooled serum prepared from ducks congenitally infected with the Australian strain of DHBV (16) containing 9.5 x 109 viral genomes (vge)/ml (7). The viral inoculum was prepared by diluting DHBV-infected serum with uninfected duck serum to obtain the appropriate dose of DHBV in 100 μl and was inoculated intravenously (i.v.) via the jugular vein.
Drug preparation and administration. ETV was synthesized and supplied in a powdered form by Bristol-Myers Squibb. A 1.0-mg/ml stock solution of ETV was prepared by adding 10 mg of ETV to 10 ml of distilled water, followed by sonication in a 65°C water bath. Stock solutions of ETV were stored at 4°C for up to 3 days and allowed to reach room temperature before use. ETV or water was administered daily to ducks in each experiment by using a 2.0 cuffless oral/nasal tube attached to a 3-ml syringe.
Experiment 1: preliminary experiment to test the effect of 24-h ETV pretreatment. Three groups of 14-day-old ducks were inoculated i.v. with 108 vge of DHBV. One group was given a placebo of 2 ml/day of water, while the second group commenced a 7-day treatment with 1.0 mg/kg/day of ETV on the day of inoculation with DHBV. The third group was given a 1.0-mg/kg dose of ETV 24 h prior to virus inoculation, followed by daily treatment with 1.0 mg/kg of ETV for 7 days. All ducks were autopsied after 7 days and examined for the presence of DHBsAg-positive hepatocytes and DHBV DNA in the liver.
Experiment 2: 49-day ETV treatment. Three groups of eight 14-day-old ducks were inoculated i.v. with either 104, 106, or 108 vge. On the day of inoculation four ducks from each group were simultaneously started on daily oral treatment with 1.0 mg/kg of ETV for 49 days, while the remaining ducks were given a placebo of 2 ml/day of water. The groups are designated as 104 + ETV, 106 + ETV, 108 + ETV, 104 + water, 106 + water, and 108 + water. The experimental plan is summarized in Fig. 1A.
Experiment 3: 14-day ETV treatment. One group of 10 14-day-old ducks and a second group of 9 14-day-old ducks were inoculated with 108 and 5 x 108 vge, respectively. On the same day, five ducks from each group were started on daily treatment with 1.0 mg/kg ETV for 14 days, and the remaining ducks received a placebo of 2 ml/day of water. Groups are designated as 108 + ETV, 5 x 108 + ETV, 108 + water, and 5 x 108 + water. The experimental plan is summarized in Fig. 1B.
Analysis of serum samples. Blood samples were collected from the jugular vein, and serum was collected following incubation of the blood samples at 37°C for 90 min and then stored at –20°C. Serum samples were tested on the day of collection for levels of alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT), and aspartate aminotransferase (AST), using an automatic analyzer in the Diagnostic Services Laboratories of the IMVS. The normal ranges for the enzymes in uninfected duck serum (mean ± standard deviation) were 26.6 ± 7.7 U/liter for ALT, 2.3 ± 1.2 U/liter for GGT, and 15.9 ± 5.9 U/liter for AST (4). Serum samples from experiments 2 and 3 were analyzed by quantitative enzyme-linked immunosorbent assay (ELISA) for DHBsAg and anticore and antisurface antibodies (4, 14).
Analysis of liver tissue. Biopsy samples of liver and autopsy samples of liver, kidney, pancreas, and spleen (from experiments 1, 2, and 3) were fixed in ethanol-acetic acid, wax embedded, sectioned, and examined for DHBsAg by immunoperoxidase staining as a marker of infected cells (8, 14). The nuclei of hepatocytes that stained positive for DHBsAg were counted in liver biopsy tissue with the aid of an eyepiece graticule in 200 250- by 250-μm grid fields (containing 100,000 hepatocytes) and were expressed as a percentage of the total hepatocyte nuclei. The minimum sensitivity of detection in liver biopsy tissue was 0.001%. In autopsy liver samples, 2,000 250- by 250-μm grid fields (containing 1,000,000 hepatocytes) were counted for each duck, resulting in a minimum sensitivity of detection of 0.0001%. Tissues were also fixed in 10% formaldehyde in phosphate-buffered saline, wax embedded, sectioned, and stained with hematoxylin and eosin for histological analysis and with Congo Red for detection of amyloid deposits in experiment 2. Formalin-fixed liver sections were examined under code for the presence and degree of inflammation and amyloid.
Autopsy liver samples from all ducks in experiment 1 and biopsy liver samples from ducks W33/34 and W45/46 in experiment 2 and from all ETV-treated ducks in experiment 3 were snap-frozen in liquid nitrogen and then stored at –80°C. Aliquots of 25 mg of each sample were extracted using a DNeasy tissue kit (QIAGEN), and DNA was eluted in 100 μl.
Real-time PCR was then performed using the Roche Lightcycler on extracted samples from experiments 1 and 2 using 100 ng of extracted DNA, quantitated using Hoechst dye. A head-to-tail dimer of DHBV DNA inserted into pBluescript II KS(+) (p4.8BLx2) (16) was used as a standard. The dimer was diluted to 90 ng/μl, giving 1010 copies of dimer per μl (2 x 1010 DHBV genomes/μl), and then diluted with DNA extracted from an uninfected normal duck liver to produce a standard curve of 108 to 101 copies of DHBV dimer in 10 μl, each supplemented with 100 ng of normal duck liver DNA. Each 20-μl reaction mixture contained 100 ng of DNA, 2 μl of Roche FastStart Master SYBR Green 1, 2.4 μl of MgCl2 (final concentration of 4 mM), and 0.5 μl of 10 μM primers and was made up to 20 μl with water. The forward primer for detection of total DHBV DNA was P3 at nucleotide (nt) 1316 (5'-AGCTGGCCTAATCGGATTAC-3'), and the reverse primer was P4 at nt 1584 (5'-TGTCCGTCAGATACAGCAAG-3'). The forward primer for detection of covalently closed circular DNA (cccDNA) was CC2 at nt 2462 (5'-CCTGATTGGACGGCTCTTAC-3'), and the reverse primer was R2 at nt 52 (5'-CCCGATCCAATGATTCCTCAT-3'). The positions of each primer are numbered within the Australian strain DHBV genome as described by Triyatni et al. (16). The PCR protocol required an initial incubation for 10 min at 95°C and then 40 cycles of 5 s at 95°C, 10 s at 55°C, and 15 s at 72°C. The DNA concentration was calculated by comparing the fluorescence of SYBR Green from samples, in the log-linear phase of amplification, to the standard curve of diluted dimer, and each dilution was run in duplicate. At the end of 40 cycles, a melting curve was generated showing the fluorescence from 60 to 95°C, allowing the specificity of each product to be checked.
Liver and serum samples from experiment 3 were also tested by PCR for the presence of DHBV DNA. Liver samples were extracted as above and 0.5 μl was used in a PCR, while 200 μl of serum was extracted using a Highpure viral DNA kit (Roche), eluted in 50 μl and 10 μl of the eluate put into a 50-μl PCR mixture. The PCR mixture also contained 5 μl of 10x reaction buffer (Geneworks, Adelaide), 3 μl of 25 mM MgCl2, 1 μl of deoxynucleoside triphosphates, 0.5 μl of Taq polymerase (Geneworks, Adelaide), and 1 μl of both forward and reverse primers and made up to 50 μl with water. The forward primer used was 2554 (5'-TTCGGAGCTGCCTGCCAAGG-3'), and the reverse primer was 269c (5'-GGAGCACCTGAGCTTGGATC-3'). The PCR protocol consisted of 10 min at 94°C, followed by 30 cycles of 20 s at 95°C, 30 s at 58°C, and 30 s at 72°C, and then 5 min at 72°C. Twenty microliters of product was then run on a 1.5% agarose gel and stained with ethidium bromide, and DHBV DNA was scored as positive or negative. The sensitivity of this assay was 103 copies of DHBV DNA in each reaction mixture.
RESULTS
Experiment 1: effect of 24-h pretreatment with ETV on DHBV infection in 14-day-old ducks. Because detailed kinetic data on the time to achieve ETV antiviral efficacy in ducks were not available, a preliminary experiment was carried out to determine if ETV treatment could be initiated at the time of infection and whether greater efficacy was obtained by initiating treatment 24 h prior to infection. Non-ETV-treated control ducks inoculated with 108 vge had DHBsAg expression in 35, 93, and >95% of hepatocytes on day 7 postinfection (p.i.) (Fig. 2A and B; Table 1). In contrast, ducks with ETV treatment that was commenced simultaneously with (Fig. 2C and D) or 24 h prior to (Fig. 2E and F) DHBV infection had 0.0004 to 0.011% and 0.0058 to 0.029% DHBsAg-positive cells, respectively, on day 7 p.i. (Table 1). Clearly, commencing ETV treatment 24 h prior to DHBV infection showed no significant advantage over ETV treatment commenced at the same time as infection. This finding suggests that ETV treatment administered either simultaneously or 24 h prior to DHBV infection was unable to prevent the conversion of the input relaxed circular (RC) DNA genome into cccDNA followed by expression of viral antigens from the cccDNA template. However, given the rapid phosphorylation of ETV and a possible lag time in conversion of RC to cccDNA, we cannot rule out that ETV treatment may partially block conversion of RC to cccDNA, thus restricting initial infection of the liver and contributing to the lower numbers of DHBsAg-positive hepatocytes observed on day 7 p.i. In an attempt to measure the inhibition of DHBV replication by ETV and to determine if cells expressing DHBsAg were replicating DHBV or expressing DHBsAg from a cccDNA template in the absence of DHBV replication, quantitative PCR was performed to determine the relative amounts of DHBV total DNA and cccDNA present in the samples collected on day 7 p.i. As can be seen in Table 1, the non-ETV-treated control ducks had ratios of total DHBV DNA to cccDNA of 29 to 67, similar to those observed in other studies of DHBV infection (9). In contrast the ETV-treated ducks had ratios of total DHBV to cccDNA of 1.5 to 10, indicating that ETV treatment inhibited virus replication, resulting in reduced levels of replicative intermediates, and reduced ratios of total DHBV DNA to cccDNA. The reduced ratios of total DHBV DNA to cccDNA in the ETV-treated ducks also suggest that cells containing cccDNA and expressing detectable DHBsAg may not be supporting viral replication.
Experiment 2: 49-day ETV treatment. Dose-related outcomes of DHBV infection were seen in control ducks in experiment 2. As expected from previous work (8), 14-day-old ducks inoculated with 104 vge had transient DHBV infections. In liver tissue collected on day 7 p.i., DHBV infection was detected in 0.02 to 0.11% of hepatocytes (average of 0.06%) (Table 2) by immunoperoxidase staining of DHBsAg present in the cytoplasm of infected cells (Fig. 3A). DHBV-infected cells were no longer detected on day 35 p.i. (Fig. 3B; Table 2) or on days 98 or 182 p.i., indicating resolution of DHBV infection. This group of ducks also had no detectable DHBsAg in the serum (Fig. 4A) but developed detectable levels of antisurface and anticore antibodies by day 11 to 20 p.i. (Fig. 5A and 6A). Titers of antisurface and anticore antibodies peaked shortly after infection, followed by a general decrease to relatively stable levels throughout the rest of the experiment.
Also in agreement with previous findings (8), in 14-day-old ducks inoculated with 106 or 108 vge DHBV rapidly spread throughout the liver, resulting in persistent DHBV infection. In liver tissue collected on day 7 p.i., DHBV infection was detected in 0.32 to 6.0% of hepatocytes (average, 2.93%) (Fig. 3C; Table 2) following inoculation of 106 vge. By day 35 p.i., DHBV infection had spread throughout the entire liver, resulting in infection of >95% of hepatocytes (Fig. 3D; Table 2). Ducks inoculated with 108 vge had >95% of hepatocytes infected on day 7 p.i., a level which was maintained to day 35 p.i. (Fig. 3E and F; Table 2).
The number of cells infected on day 7 p.i. after inoculation with each dose of virus was within the expected range. This is based on estimates of the number of cells per gram of duck liver (7 x 108) (9) and the liver weight of 14-day-old ducks (measured as 20.1 g), giving an estimated total number of 1.4 x 1010 liver cells. Accordingly, we could expect that i.v. inoculation of 104, 106, and 108 vge would infect 0.000071, 0.0071, and 0.71% of total liver cells, provided all the inoculum was delivered to the liver in each case. Cell-to-cell spread of infection throughout the liver with a reported doubling time of 16 h (7) should result in infection of an estimated 0.098, 9.8, and >98% of hepatocytes on day 7 p.i. The observed percentages of infected cells on day 7 p.i. in ducks inoculated with 104, 106, and 108 vge were within this range, with DHBsAg detection in an average of 0.06, 2.93, and >95% of hepatocytes.
Infection of the liver also resulted in the appearance of DHBsAg in the serum of infected ducks, first detected on day 11 p.i. in the 106 vge group (Fig. 4C) and day 7 p.i. in the 108 vge group (Fig. 4E). Levels of DHBsAg ranged from 0.4 to 190 μg/ml. The difference in the timing of appearance of DHBsAg reflects the 100-fold difference in the viral dose and the extent of infection observed in the liver. As previously reported (8, 13), levels of DHBsAg in the serum showed a biphasic pattern, with a short-term appearance then disappearance before DHBsAg became continuously detectable for the remainder of the study.
Serum DHBsAg persisted throughout the experiment in six out of eight ducks but became undetectable in two ducks (W33/34 and W45/46) after day 60 and 98 p.i., respectively. Both of these ducks showed an absence of DHBsAg-positive hepatocytes in the liver when tested on day 182 p.i. (Table 2). Liver DHBV DNA levels determined by real-time PCR (data not shown) in duck W33/34 also showed a >3-log decrease from day 35 to day 172 p.i. In duck W45/46, liver DHBV DNA levels decreased from 1,900,000 to 5,500 copies per 87,500 cells (21.7 to 0.062 copies per cell) (data not shown), representing a 3-log reduction from day 35 to day 98 p.i. The large decrease in markers of infection seen in these two ducks indicates spontaneous clearance of DHBV infection, although the markers examined showed no characteristics that allowed these ducks to be distinguished from those that maintained a high-level persistent infection.
Antisurface and anticore antibodies were detected at variable levels in the serum of the ducks inoculated with 106 and 108 vge from day 11 to 20 p.i. and persisted throughout the study (Fig. 5C and E and 6C and E). Antisurface antibodies were in some cases detected in the presence of DHBsAg, presumably reflecting the presence of immune complexes. Levels were variable, both between individual animals and within individuals over time. Anticore antibodies were seen in two out of four ducks inoculated with 104 vge, with levels variable in the early part of infection (Fig. 6A). Anticore responses were seen in all ducks in the 106 and 108 groups (Fig. 6C and E). A general trend for increasing titers over time was seen in the 106 + water group, while a more variable response was seen in the 108 + water group both between individuals and within individuals over time.
ETV treatment restricted the spread of virus infection in the liver and delayed the rebound of DHBV infection. ETV treatment did not prevent initial DHBV infection of the liver but resulted in marked reductions in the number of DHBsAg-positive hepatocytes detected in the liver on day 7 p.i. No DHBsAg-positive hepatocytes were detected in the liver of ducks inoculated with 104 vge at any time after inoculation (Table 2). These ducks had no detectable DHBsAg in the serum (Fig. 4B) but developed low levels of antisurface and anticore antibodies from 70 to 120 days p.i. (Fig. 5B and 6B). The observed delay in detectable levels of antisurface and anticore antibodies may reflect the initial restriction in the spread of DHBV infection and antigen expression in the liver, followed by further replication of virus and expression of viral antigens after the removal of drug, which stimulates the production of antibodies to detectable levels and results in immune-mediated clearance of infected cells.
Similarly, in ducks inoculated with 106 vge the number of DHBsAg-positive hepatocytes was markedly reduced compared to the non-ETV-treated ducks, with DHBsAg expression on day 7 p.i. detected in two out of four ETV-treated ducks (W9/10 with 0.002% and W11/12 with 0.001% of hepatocytes) (Table 2). The percentage of DHBsAg-positive hepatocytes detected in the liver on day 7 p.i. was thus reduced from an average of 2.93% in the non-ETV-treated control ducks to 0.0015% of hepatocytes detected in two out of four ETV-treated ducks, representing a greater-than-1,000-fold reduction in the percentage of infected cells.
Ducks inoculated with 108 vge and treated with ETV had an average of 0.046% of liver cells (range, 0.013 to 0.08%) (Table 2) infected on day 7 p.i. (Fig. 7A and E) compared to >95% of cells in the non-ETV-treated ducks (Fig. 3E), representing a 2,000-fold reduction in the percentage of infected cells.
Based on the calculations presented above for non-ETV-treated control ducks, we predicted that i.v. inoculation of doses of 104, 106, and 108 vge would initially infect 0.000071, 0.0071, and 0.71% of total liver cells on the day of inoculation. The observed numbers of DHBsAg-positive cells in the ETV-treated ducks on day 7 p.i. was at least 10-fold lower than these estimates, with DHBsAg detection in <0.001% (all four ducks), 0.002% and 0.001% (two of the four ducks), and an average of 0.046% of cells (range, 0.013 to 0.080% in all four ducks) in the three dose groups. This observation highlights the ability of ETV to restrict the cell-to-cell spread of virus infection in the liver by inhibiting virus replication during the 7 days after virus inoculation. Although direct evidence was not obtained in experiment 1 for the ability of ETV to inhibit conversion of the RC DNA genome into cccDNA during the initiation of infection, we cannot rule out that ETV may restrict initial infection of the liver and that this might also contribute to the reduced levels of DHBsAg-positive hepatocytes on day 7 p.i.
The percentage of DHBsAg-positive cells remained low at day 35 p.i. during ETV treatment (Fig. 7B and F), but DHBV-positive cells were still detected, indicating that DHBV-infected cells containing the transcriptional template of the virus, cccDNA, persisted in the liver in the presence of ETV treatment.
Forty-nine days after the withdrawal of ETV, at day 98 p.i., DHBsAg-positive cells were detected in five out of seven ETV-treated ducks inoculated with 106 or 108 vge, with four out of five animals having less than 0.1% DHBsAg-positive hepatocytes (Fig. 7C and G; Table 2). The exception was duck W11/12, inoculated with 106 vge, in which 2.17% of hepatocytes were DHBsAg positive. Unfortunately, this duck died of unknown causes prior to day 182 p.i. However, based on the percentage of infected liver cells at day 98 p.i., it is likely that this duck would have developed a persistent DHBV infection. By autopsy at day 182 p.i., no DHBsAg-positive hepatocytes were detected in ducks W13/14 and W15/16 from the 106 + ETV group or duck W23/24 from the 108 + ETV group (Fig. 7D; Table 2). These three ducks had evidently cleared their DHBV infection.
In contrast, in one out of three ducks (W9/10) from the 106 + ETV group and in two out of three ducks (W17/18 and W19/20) from the 108 + ETV group, DHBV infection had spread throughout the liver to infect >95% of hepatocytes (Fig. 7H; Table 2).
The different outcomes of DHBV infection are illustrated in Fig. 7, where two ducks inoculated with 108 vge and treated with ETV (W23/24 and W19/20) had DHBsAg-positive hepatocytes detectable in the liver on days 7, 35, and 98 p.i., shown as individual cells or, in one case, as pairs of DHBsAg-positive cells (Fig. 7A to C and E to G). However, while cells expressing DHBsAg were no longer detected in the liver of duck W23/24 at autopsy on day 182 p.i. (Fig. 7D), in duck W19/20 DHBV infection had spread to infect the entire hepatocyte population (Fig. 7H). In summary, three out of seven ducks infected with 106 and 108 vge and simultaneously treated with ETV apparently cleared their DHBV infection, while four out of seven ducks, including duck W11/12 that died unexpectedly prior to day 182 p.i., developed persistent DHBV infection. The reasons for these different outcomes of infection are not known but suggest that infection was not efficiently controlled by the immune response.
ETV treatment prevented the development of detectable levels of serum DHBsAg during treatment and for an extended period after ETV withdrawal. As expected from the low levels of virus infection in the liver observed during ETV treatment, serum DHBsAg was not detected in any ETV-treated ducks during the course of ETV treatment (Fig. 4B, D, and F) and was only detected in three out of nine ETV-treated ducks after ETV treatment was withdrawn, coinciding with increases in the number of DHBV-infected cells in the liver. In one of three ducks (W19/20, inoculated with 108 vge), serum DHBsAg was first detected on day 112 p.i., 63 days after ETV treatment was withdrawn. A second duck from the 108 + ETV group (W17/18) had DHBsAg detectable in serum from day 126 p.i., 77 days after ETV withdrawal, until the end of the study, and duck W9/10 from the 106 + ETV group had detectable serum DHBsAg on day 147 p.i., 98 days after ETV treatment (Fig. 4D and F).
Antiviral antibodies were delayed in ETV-treated ducks. Antisurface antibodies were not detected in any ETV-treated ducks during the course of treatment (Fig. 5B, D, and F). However, from day 70 p.i., 21 days after ETV treatment was withdrawn, low titers of antisurface antibodies were detected in two out of four ducks inoculated with 104 vge (W1/2 and W7/8) and one out of four ducks inoculated with 108 vge (W23/24). Antisurface antibodies developed in all the remaining ETV-treated ducks during the course of the experiment, with antibodies appearing earlier in the ducks that had no DHBsAg evident in the serum (Fig. 5B, D, and F). In the three ducks that developed viremia (W9/10, W17/18, and W19/20), antisurface antibodies were detected from the time of DHBsAg detection or 7 days later.
Anticore antibodies were detected in one ETV-treated duck (W19/20, 108 + ETV) in the last week of ETV treatment, and this duck went on to develop persistent DHBV infection (Fig. 6F). Anticore responses were low in the remaining ETV-treated ducks (Fig. 6B, D, and F) and first appeared on day 98 p.i., 49 days after the withdrawal of ETV in the 104 + ETV group, and on day 126 p.i., 77 days or more after ETV withdrawal in the 106 and 108 + ETV groups. Not all ducks developed anticore antibodies, and they were detectable at only one to three time points in those that did produce any. The late appearance of antisurface and anticore antibodies, some time after cessation of ETV treatment, suggests that higher levels of antigen, produced by spread of infection after ETV treatment, are required to induce a measurable humorable immune response.
Liver enzyme levels were unaffected by ETV treatment. Levels of the liver function enzymes ALT, GGT, and AST (Fig. 8A, B, and C, respectively) were similar for all ducks, with some individual variation seen, especially in the GGT levels. Two ducks from the 108 + water group had 40-fold-higher-than-normal levels of GGT at day 133 p.i., and 35-fold-higher levels were seen in a duck from the 106 + ETV group at day 161 p.i. (Fig. 8B). These increases occurred after the ducks were 20 weeks old, when egg laying commences, and similar flares have been seen in noninfected and/or nontreated ducks of that age (3, 4). The average values for liver enzymes, excluding the extreme GGT values, were as follows: ALT, 25.8 ± 10 U/liter; GGT, 2.2 ± 1.2 U/liter; AST, 15.5 ± 6.5 U/liter. These values were all within previously determined normal limits for uninfected ducks (4).
One duck from the 108 + ETV group (W17/18) that developed persistent DHBV infection showed a fourfold increase in ALT levels in the last 40 days of the experiment (Fig. 8A). This ALT flare occurred around the time that DHBsAg and DHBV DNA first appeared in the serum. This duck also showed signs of amyloidosis at days 98 and 172 p.i., and slight inflammation was observed at day 172 p.i. No other animals showed signs of amyloidosis. Rates of weight gain showed no difference between the ETV-treated and control ducks (Fig. 8D).
Liver histology and duck health. All DHBV-infected ducks showed mild liver inflammation, and this was unaffected by ETV treatment. During the experiment three ducks died, one from each ETV treatment group. Duck W21/22 was euthanized on day 22 due to congenital leg deformities. Duck W3/4 was euthanized on day 26 due to severe bacterial septicemia. Duck W11/12 was found dead in its cage from an undetermined cause at day 105 p.i., 56 days after ETV treatment was withdrawn. None of the deaths were directly related to ETV therapy. The detection of DHBsAg-positive hepatocytes in these three ducks has been included in Table 2, but serological profiles have not been presented.
Experiment 3: 14-day ETV treatment. We wished to determine if a shorter treatment with ETV could lead to resolution of infection in a similar percentage of ETV-treated ducks. Fourteen-day-old ducks were infected with 108 and 5 x 108 vge of DHBV and were used as untreated control ducks or were treated simultaneously with 1.0 mg/kg of ETV for 14 days. As observed in experiment 2, DHBV infection rapidly spread throughout the liver of the control ducks injected with either 108 or 5 x 108 vge to infect >95% of hepatocytes by day 14, a level maintained to day 70 p.i. (Table 3). In contrast, although treatment with ETV for 14 days did not prevent the establishment of a low-level DHBV infection, it did restrict the spread of DHBV infection in the liver. Liver tissue collected at the end of 14 days of treatment from ducks infected with 108 or 5 x 108 vge had 0.0031 to 0.010% and 0.014 to 0.14% DHBsAg-positive hepatocytes, respectively (Table 3). This difference was thought to result from the fivefold difference in the size of the viral inoculum.
Fifty-six days after ETV treatment was withdrawn (day 70 p.i.), 7 out of 10 ducks had either undetectable (<0.0001%) or low numbers (0.0006 to 0.024%) of DHBsAg-positive hepatocytes (Table 3). In the remaining 3 out of 10 ducks, DHBV infection had spread to involve the entire hepatocyte population.
Serum DHBsAg and DHBV DNA were detected in the three ETV-treated ducks that developed a persistent infection (B13, B15, and B16) at day 49 p.i. and were detected in one of these ducks until day 63 p.i. (Table 3). The serum of a fourth ETV-treated duck (B12) also contained serum DHBsAg and DHBV DNA at day 49 p.i., but these markers were not detected at subsequent time points (Table 3). This duck had no detectable DHBsAg-positive cells in the liver at autopsy and appeared to have had a short-term DHBV infection that rebounded after drug treatment was withdrawn but that was subsequently cleared from the liver (Table 3). No other ETV-treated ducks had detectable DHBsAg or DHBV DNA in their serum during the experiment.
DISCUSSION
As expected from previous work (8), 14-day-old ducks infected with DHBV showed different outcomes of DHBV infection depending on the size of the viral inoculum. Control ducks inoculated with 104 vge had a transient infection, while ducks inoculated with 106, 108, or 5 x 108 vge developed persistent DHBV infections. The reason for the spontaneous decline in markers of infection in two of the infected ducks (W33/34 and W45/46) was not evident from the data obtained but was consistent with observations in patients infected with HBV. Spontaneous clearance of chronic HBV infections in humans is seen in a low percentage of cases, but the cause is unknown (2), although it has been found to be associated with age (10). Several previous DHBV infection experiments in this laboratory using the same age and dose relationship were carried out for less than 60 days, and no decreases in markers of infection were observed (8).
ETV treatment resulted in an altered course of infection, with an obvious delay in the development of markers of DHBV infection and/or detectable antibody responses. ETV treatment commencing at the time of DHBV infection was not effective in preventing DHBV infection of hepatocytes but led to significant suppression of viral spread. ETV treatment extended the period of low-level infection, which may have allowed for natural clearance of the virus in some individuals and could have complemented or facilitated the use of other methods designed to enhance immune responsiveness to DHBV and, possibly, HBV.
ETV is rapidly absorbed and converted to the triphosphate form (20), with peak plasma levels in ducks occurring 30 min after a 5-mg/kg dose (198.05 ng/ml) and with a half-life in plasma of 8.73 h (4). Although ETV is absorbed and converted rapidly to the triphosphate form, simultaneous treatment of ducks with ETV did not prevent conversion of the input RC DHBV genome into cccDNA with resulting expression of DHBsAg within hepatocytes. The ability of ETV to prevent conversion of RC DNA to cccDNA was further tested in experiment 1 by the pretreatment of ducks with ETV 24 h prior to DHBV infection, but this did not significantly alter the number of cells infected at day 7 p.i.
Since ETV acts via a competitive inhibition with dGTP, its ability to bind and block the conversion of RC DHBV genomes to cccDNA is likely to depend on the size of the gap in the positive strand of the RC DNA genome. In DHBV, the 3' end of the positive strand terminates immediately upstream of the DR2 sequence, where the RNA primer binds (11). In the Australian strain of DHBV (16) used in these experiments, there is one G base at position 2531 of the RNA primer, but if the termination occurs further upstream there are G bases at positions 2529, 2528, and 2526, to which ETV could bind and block the formation of cccDNA. In support of our data, treatment with another guanosine analogue, 2'-deoxyguanosine, for 24 h prior to DHBV infection of primary duck hepatocytes was also unable to block conversion of RC to cccDNA (5). ETV may be more effective in blocking cccDNA formation in HBV or woodchuck hepatitis virus infection, as those genomes have larger incomplete regions in the positive strand than the DHBV genome.
DHBV DNA remained present throughout ETV treatment and is likely to be dormant or nonreplicating, as it appeared to be insensitive to inhibition. cccDNA forms a stable template for hepadnavirus replication, making it difficult to eliminate through treatment with antiviral drugs (1, 19, 21), and would provide a mechanism for the virus to evade antiviral drug effects. Clearance of cccDNA from cells through cell division, cell death, or possibly noncytolytically via cytokine-induced mechanisms would be required to achieve viral clearance and stop the rebound commonly seen after the cessation of antiviral therapy.
In ETV-treated ducks, the absence of detectable antibodies during treatment and the lower-than-control levels after ETV withdrawal indicate their humoral immune responses were not well developed, possibly because the amount of antigen present was too low during the ETV treatment phase. Virus replication, if it did occur during drug treatment, was not at levels sufficient to stimulate an effective humoral immune response, as evidenced by the increase in infected cells after drug withdrawal. It seems possible that commencing treatment with ETV at the time of infection might have either suppressed the rapid replication of the virus, thereby allowing immune responses to control the infection, or to have held the virus at low, nonimmunogenic levels until the ducks developed a more mature immune system, so that an effective immune response could be mounted once the drug was withdrawn. Either of these outcomes would have improved the likelihood of viral clearance.
The four ETV-treated ducks that progressed to persistent infection in experiment 2 had low levels of DHBsAg-positive cells in the liver at days 35 and 98 p.i. This raises the questions of whether and how these cells evaded the immune response or whether infected cells were actually being removed and replaced by others, due to a continuing low rate of new infection of hepatocytes. The individual DHBsAg-positive cells had levels of cytoplasmic DHBsAg comparable to the controls, but it is not known if this expression was generated from low copy numbers of cccDNA, or integrated DNA, or if the virus was replicating within these cells. In situ hybridization could answer this question, but currently assays are not adequate to provide a clear answer. PCR did not provide evidence of replicating DNA in excess of total cccDNA in the liver.
The presence of both DHBsAg and antisurface antibodies may have caused some of the variability seen in levels of DHBsAg and antisurface antibodies, due to the varying amounts present in the form of antigen-antibody complexes. The presence of circulating antisurface antibodies is generally thought to be protective, and in HBV it is recognized as a marker of immunity and of the resolution of disease. However, complexes of HBsAg and immunoglobulin have been detected in acute HBV infection (17) and in healthy carriers and in patients with symptoms of chronic infection (18). The presence of these complexes has been related to HBV infection outcomes, with an appearance time of 1 to 27 weeks seen in acute infection and long-term persistence of complexes in chronically infected individuals (15).
In conclusion, ETV shows promise as a postexposure therapy for HBV infection. Although ETV treatment alone was not sufficient to lead to DHBV clearance in all ducks, significant inhibition of viral replication occurred, leading to different infection outcomes. This study showed that ETV is a safe, potent suppressor of DHBV replication, although it is unable to stop the development of low-level infection of hepatocytes when given simultaneously with or 24 h prior to DHBV. These results have provided a basis for future work on immune therapies designed to develop a therapeutic protocol for treatment of patients chronically infected with HBV.
ACKNOWLEDGMENTS
This research was supported by the National Health and Medical Research Council of Australia and by a research grant from Bristol-Myers Squibb Pharmaceutical Research Institute.
We appreciate the work of the IMVS animal care staff for the day-to-day care of the ducks involved in this study. We are also grateful to Christopher Burrell for helpful suggestions and to William Mason for critical reading of the manuscript.
REFERENCES
Colonno, R. J., E. V. Genovesi, I. Medina, L. Lamb, S. Durham, M. L. Huang, L. Corey, M. Littlejohn, S. Locarnini, B. C. Tennant, B. Rose, and J. M. Clark. 2001. Long-term entecavir treatment results in sustained antiviral efficacy and prolonged life span in the woodchuck model of chronic hepatitis infection. J. Infect. Dis. 184:1236-1245.
Da Silva, L. C., C. L. Madruga, F. J. Carrilho, J. R. Pinho, A. Saez-Alquezar, C. Santos, L. Bassit, C. Barreto, L. E. Fonseca, V. A. Alves, R., Leitao, R. Vianna, R. A. Cardoso, A. V. Franca, and L. C. Gayotto. 1996. Spontaneous hepatitis B surface antigen clearance in a long-term follow-up study of patients with chronic type B hepatitis. Lack of correlation with hepatitis C and D virus superinfection. J. Gastroenterol. 31:696-701.
Fairbrother, A., M. A. Craig, K. Walker, and D. O'Loughlin. 1990. Changes in mallard (Anas platyrhynchos) serum chemistry due to age, sex, and reproductive condition. J. Wildlife Dis. 26:67-77.
Foster, W. K., D. S. Miller, P. L. Marion, R. J. Colonno, I. Kotlarski, and A. R. Jilbert. 2003. Entecavir therapy combined with DNA vaccination for persistent duck hepatitis B virus infection. Antimicrob. Agents Chemother. 47:2624-2635.
Fourel, I., J. Saputelli, P. Schaffer, and W. S. Mason. 1994. The carbocyclic analog of 2'-deoxyguanosine induces a prolonged inhibition of duck hepatitis B virus DNA synthesis in primary hepatocyte cultures and in the liver. J. Virol. 68:1059-1065.
Genovesi, E. V., L. Lamb, I. Medina, D. Taylor, M. Seifer, S. Innaimo, R. J. Colonno, D. N. Standring, and J. M. Clark. 1998. Efficacy of the carbocyclic 2'-deoxyguanosine nucleoside BMS-200475 in the woodchuck model of hepatitis B virus infection. Antimicrob. Agents Chemother. 42:3209-3217.
Jilbert, A. R., D. S. Miller, C. A. Scougall, H. Turnbull, and C. J. Burrell. 1996. Kinetics of duck hepatitis B virus infection following low dose virus inoculation: one virus DNA genome is infectious in neonatal ducks. Virology 226:338-345.
Jilbert, A. R., J. A. Botten, D. S. Miller, E. M. Bertram, P. Hall, I. Kotlarski, and C. J. Burrell. 1998. Characterization of age- and dose-related outcomes of duck hepatitis B virus infection. Virology 243:273-282.
Jilbert, A. R., T. T. Wu, J. M. England, P. de la M. Hall, N. Z. Carp, A. P. O'Connell, and W. S. Mason. 1992. Rapid resolution of duck hepatitis B virus infections occurs after massive hepatocellular involvement. J. Virol. 66:1377-1388.
Kato, Y., K. Nakao, K. Hamasaki, H. Kato, K. Nakata, Y. Kusumoto, and K. Eguchi. 2000. Spontaneous loss of hepatitis B surface antigen in chronic carriers, based on a long-term follow-up study in Goto Islands, Japan. J. Gastroenterol. 35:201-205.
Lien, J. M., D. J. Petcu, C. E. Aldrich, and W. S. Mason. 1987. Initiation and termination of duck hepatitis B virus DNA synthesis during virus maturation. J. Virol. 61:3832-3840.
Marion, P. L., F. H. Salazar, M. A. Winters, and R. J. Colonno. 2001. Potent efficacy observed with entecavir (BMS-200475) in duck model of hepatitis B virus replication. Antimicrob. Agents Chemother. 46:82-88.
Meier, P., C. A. Scougall, H. Will, C. J. Burrell, and A. R. Jilbert. 2003. A duck hepatitis B virus strain with a knock out mutation in the putative X ORF shows similar infectivity and growth characteristics to wild type. Virology 317:291-298.
Miller, D. S., E. M. Bertram, C. A. Scougall, I. Kotlarski, and A. R. Jilbert. 2004. Studying host immune responses against duck hepatitis B virus infection. Methods Mol. Med. 96:3-28.
Surelia, P., and E. H. Boxall. 1990. Hepatitis B virus infection: detection of circulating HBsAg/IgM antibody immune complexes. Med. Lab. Sci. 47:204-209.
Triyatni, M., P. L. Ey, H. Tran, M. F. Le Mire, M. Qiao, C. J. Burrell, and A. R. Jilbert. 2001. Sequence comparison of an Australian duck hepatitis B virus strain with other avian hepadnaviruses. J. Gen. Virol. 82:373-378.
Tsai, J. F., H. S. Margolis, J. E. Jeng, M. S. Ho, W. Y. Chang, M. Y. Hsieh, Z. Y. Lin, and J. H. Tsai. 1996. Hepatitis B surface antigen- and immunoglobulin-specific circulating immune complexes in acute hepatitis B virus infection. Clin. Immunol. Immunopathol. 80:278-282.
Tsai, J. F., H. S. Margolis, J. E. Jeng, M. S. Ho, W. Y. Chang, M. Y. Hsieh, Z. Y. Lin, and J. H. Tsai. 1998. Immunoglobulin- and hepatitis B surface antigen-specific circulating immune complexes in chronic hepatitis B virus infection. Clin. Immunol. Immunopathol. 86:246-251.
Werle-Lapostolle, B., S. Bowden, S. Locarnini, K. Wursthorn, J. Petersen, G. Lau, C. Trepo, P. Marcellin, Z. Goodman, W. E. Delaney, S. Xiong, C. L. Brosgart, S. S. Chen, C. S. Gibbs, F. Zoulim. 2004. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 126:1750-1758.
Yamanaka, G., T. Wilson, S. Innaimo, G. S. Bisacchi, P. Egli, J. K. Rinehart, R. Zahler, and R. J. Colonno. 1999. Metabolic studies on BMS-200475, a new antiviral compound active against hepatitis B virus. Antimicrob. Agents Chemother. 43:190-193.
Zhu, Y., T. Yamamoto, J. Cullen, J. Saputelli, C. E. Aldrich, D. S. Miller, S. Litwin, P. A. Furman, A. R. Jilbert, and W. S. Mason. 2001. Kinetics of hepadnavirus loss from the liver during inhibition of viral DNA synthesis. J. Virol. 75:311-322.(Wendy K. Foster, Darren S)