INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The hepatitis B virus (HBV) is a 3.2-kb, noncytopathic DNA virus that induces chronic hepatitis in approximately 5% of the global population and is the most common infectious cause of hepatocellular carcinoma (15, 26). Infection with the closely related woodchuck hepatitis virus (WHV) in the eastern American woodchuck (Marmota monax) has been validated as a valuable virologic and pathogenic model of HBV infection (17, 37). Although liver inflammation is the paramount pathological consequence of HBV and WHV infections and hepatocytes are the site of the most active hepadnaviral replication, both viruses also invade the host lymphatic system. HBV DNA, including replicative intermediates and covalently closed circular DNA (cccDNA), as well as virus RNA transcripts and DNA sequences integrated with the cellular genome, have been detected in peripheral blood mononuclear cells (PBMC) from patients with active chronic hepatitis B (13, 29, 30, 34). Small amounts of HBV genomes have also been found in circulating lymphoid cells years after recovery from an acute episode of hepatitis B (23, 32). Studies in WHV-infected woodchucks provided even more persuasive evidence that the lymphatic system is invariably involved from the earliest stages of either symptomatic or serologically undetectable (silent) infection and is the site of life-long persistence of small amounts of pathogenic virus after resolution of hepatitis (4, 20, 22; reviewed in reference 18). It has also been documented that under certain natural conditions, i.e., in offspring born to mothers recovered from WHV hepatitis (4) or in particular experimental situations, e.g., infection of adult animals with low WHV doses (22; C. S. Coffin, P. M. Mulrooney, and T. I. Michalak, unpublished data), virus expression can be essentially confined to the lymphatic system. Although the importance of this extrahepatic site of replication for the long-term persistence of hepadnavirus becomes increasingly evident (reviewed in reference 18), the intrinsic ability of lymphotropic WHV to infect host hepatocytes and to induce liver disease as well as to invade virus-naive lymphoid cells has not yet been examined. These issues are of direct clinical relevance, since there is well-documented evidence of extrahepatic HBV persistence and prompt reinfection of donor hepatic tissue after liver transplantation in chronically infected individuals (5, 7, 12, 14, 16, 28). Furthermore, HBV infection of lymphoid cells may have a detrimental effect on a variety of host immune responses and contribute to both the establishment and the continuance of virus persistence. In this regard, we have recently shown that chronic WHV infection is associated with profound inhibition of class I major histocompatibility complex (MHC) presentation not only on hepatocytes but also on lymphoid cells (20). Impairment of class I MHC-dependent functions of the immune system could be one of the major mechanisms by which hepadnavirus compromises antiviral immunological surveillance and persists in the host.

In the present study, we investigated hepato- and lymphotropism and tested the pathobiological characteristics of WHV produced by naturally infected lymphoid cells. For this purpose, woodchuck hepatocyte and lymphoid cell cultures susceptible to WHV and lymphoid cell-hepatocyte and lymphoid cell-lymphoid cell coculture systems were established. In addition, PCR assays discriminating between the total pool of WHV DNA and that of covalently closed viral DNA were adopted to assess viral expression and replication in the infected cells. Viral cccDNA was examined as an indicator of active virus replication based on the postulate that its formation is the initial step in the hepadnaviral propagation cycle and, therefore, a marker of actually progressing replication (27, 36, 40).

The WHV DNA amplification assays were done using DNA extracted from cells subjected to DNase and limited proteolytic digestions to eliminate possible cell surface-adsorbed extracellular viral sequences. Production of infectious WHV by lymphoid cells and its pathogenic potential were examined by serial passages of the virus in cultured hepatocytes, coculture of the infected cells with virus-naive hepatocytes or lymphoid cells, and inoculation of naive animals. Our results provide direct in vitro and in vivo evidence that WHV propagating in naturally infected lymphoid cells is infectious to both hepatocytes and lymphoid cells and is able to transmit infection to virus-naive animals. In susceptible woodchucks, the lymphoid cell-derived virus can induce classical viral hepatitis, although at small doses it causes serologically undetectable infection that involves the lymphatic system but not the liver.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Cell isolation and cultures. Splenocytes, containing mostly lymphocytes, and PBMC from woodchucks with serologically and histologically confirmed chronic WHV hepatitis were isolated on Ficoll-Paque (Pharmacia Biotech, Baie d'Urfé, Qubec, Canada) gradients by methods described previously (8, 19). In all instances, the cells were depleted of residual erythrocytes with 0.83% NH₄Cl, extensively washed with sterile phosphate-buffered saline (PBS, pH 7.4) with 1 mM EDTA, suspended in Hanks' balanced salt solution (HBSS), and digested with DNase, trypsin, and again with DNase to eliminate WHV particles and free viral DNA fragments possibly attached to their surface. This digestion (referred to as DNase/trypsin/DNase treatment or surface enzymatic treatment) was done according to a previously published procedure (4, 22) with modifications. For this purpose, approximately 3 × 10⁷ cells in 2 ml of HBSS were supplemented with 200 µl of DNase digestion buffer (100 mM MgCl₂ in 500 mM Tris-HCl [pH 8.0]) and 20 µl of DNase (1 µg/µl; activity, 2 U/µg; DNase I; type IV from bovine pancreas; Sigma Chemical Co., St. Louis, Mo.) and incubated at 37°C for 1 h in a tissue culture incubator. Subsequently, the cells were treated with 10 µl of trypsin (10 mg/ml; activity, 7,300 U/mg; type XI from bovine pancreas; Sigma Chemical Co.) in the presence of 20 µl of 0.1 M CaCl₂ for 30 min on ice. The reaction was terminated with 20 µl of trypsin inhibitor (10 µg/ml; type II-O from chicken egg white; Sigma Chemical Co.). In the following step, DNase treatment was repeated under the conditions described above to eliminate WHV DNA possibly released from extracellular virions. The cells were washed twice with a total of 20 ml of PBS with 1 mM EDTA by centrifugation at 330 × g for 5 min. The final wash and 2 × 10⁶ treated cells were saved for WHV DNA analysis by specific PCR and Southern blot hybridization. The remaining cells were immediately suspended in hepatocyte culture medium (described below) and cultured in a humidified atmosphere of 5% CO₂. It was found that the lymphoid cells can be maintained in a hepatocyte culture medium for up to 4 days without noticeable changes in their viability and WHV content. PBMC were also isolated from WHV-naive, healthy woodchucks and used in coculture infection experiments (see below).

Lymphoid cell- and serum-derived WHV inocula. Three types of WHV inocula were prepared and tested in this study. They were derived from naturally infected PBMC or splenocytes or from sera of WHV surface antigen (WHsAg)-positive animals with histologically evident chronic viral hepatitis (Table 1). WHV originating from lymphoid cells was prepared by culture of PBMC or splenocytes at 2 × 10⁶ cells/ml for 72 h in hepatocyte medium and passage of the resulting culture supernatants through 0.2-µm-pore-size syringe filters (Gelman Sciences, Ann Arbor, Mich.). Serum-derived WHV was obtained by diluting sera with hepatocyte culture medium and filtration through 0.2-µm filters. The final filtrates were used as sources of virus for infection of either WCM-260 hepatocytes or virus-naive lymphoid cells.

Infection of cultured hepatocytes with WHV. WCM-260 hepatocytes were seeded 24 h prior to infection at 1.2 × 10⁴ cells/cm² onto gelatin-coated 25-cm² tissue culture flasks (Corning Costar Corp., Cambridge, Mass.) in 5 ml of hepatocyte culture medium. In the next step, medium was replaced with WHV inocula at a volume of 200 µl/cm² of vessel surface and incubated with the hepatocytes for 24 h at 37°C in a humidified CO₂ atmosphere. The culture medium with inoculum was then removed, hepatocytes were washed twice with HBSS, and 5 ml of fresh hepatocyte culture medium was added. The cells were allowed to grow for 3 more days and then harvested by trypsinization. In all experiments, hepatocytes destined for DNA isolation were treated with DNase/trypsin/DNase (as described for lymphoid cells) and washed in two 10-ml changes of HBSS. The final wash was saved for WHV DNA analysis by PCR and Southern blot hybridization to assess whether extracellular virus material was removed from the surface of the cultured cells analyzed. In some experiments, the collected cells were divided into two equal parts. One part was washed, surface treated with enzymes, and used for DNA extraction. The second part was examined for expression of WHV antigens by flow cytometry. In selected cases, supernatant recovered from hepatocyte cultures at day 3 postinoculation (p.i.) was ultracentrifuged at 200,000 × g for 18 h in an SW50.1 rotor (Beckman Instruments Inc., Palo Alto, Calif.), and the pellet was suspended in 100 µl of sterile PBS. Then, DNA was extracted and assayed for WHV DNA, as presented below.

Multiple passages of WHV from lymphoid cells in hepatocyte cultures. WCM-260 cells were exposed under conditions outlined above to WHV at 3 × 10⁴ virus genome equivalents (VGE)/ml produced by naturally infected PBMC (experiment 9 in Table 1) or 2 × 10⁷ VGE/ml from splenocytes (experiment 14 in Table 1). Subsequently, 5-ml volumes of hepatocyte culture supernatants collected at day 3 p.i. from these hepatocyte cultures were transferred to naive WCM-260 cells seeded 24 h prior to passage. After 24 h of incubation, medium was removed, and hepatocytes were washed with HBSS and supplemented with 5 ml of fresh hepatocyte culture medium. The cells were cultured for an additional 3 days under standard conditions. The resulting culture supernatant was collected, hepatocytes were harvested, washed, and surface treated with enzymes, and DNA was extracted. The recovered culture supernatant was transferred to newly established 24-h WCM-260 cell culture using the approach described above. This procedure was repeated five consecutive times. The culture media after the final (sixth) hepatocyte passage from each of the experiments (i.e., experiments 9A and 14A) were separately spun down at 200,000 × g for 18 h in a Beckman SW50.1 rotor, and each pellet was suspended in 1 ml of sterile PBS. One hundred microliters of each suspension was assayed for WHV VGE content, and the remaining portion was injected intravenously into a WHV-naive woodchuck (see below).

Infection of cultured lymphoid cells with lymphoid cell-derived WHV. For in vitro infection of lymphoid cells, approx. 10⁷ PBMC from a healthy woodchuck were incubated with 5 ml of culture supernatant containing 2 × 10⁷ WHV VGE/ml derived from naturally infected splenocytes. The inoculum was removed after 24 h by centrifugation at 330 × g for 10 min. The target cells were washed with 10 ml of HBSS, supplemented with fresh medium, and cultured for an additional 72 h. At day 4 p.i., lymphoid cells were harvested, washed, and surface treated with enzymes, and DNA was extracted and analyzed for WHV DNA and cccDNA expression.

Hepatocyte and lymphoid cell cocultures. To determine whether WHV from lymphoid cells can directly infect homologous hepatocytes or lymphoid cells, lymphoid cell-hepatocyte and lymphoid cell-lymphoid cell coculture systems were established. The cocultures were set up in 12-well plates with permeable 0.4-µm-pore-size membrane inserts (Becton Dickinson Labware, Bedford, Mass.). All cocultured cells were maintained in hepatocyte culture medium, permitting measurements of WHV infectivity under comparable conditions. For cocultures of WHV-infected splenocytes with virus-naive hepatocytes, WCM-260 cells were seeded 24 h before the experiment at 1.2 × 10⁴ cells/cm² in gelatin-coated wells. When virus-naive lymphoid cells were used as targets, PBMC from a healthy woodchuck were placed at 10⁶ cells/well in the 12-well plates 2 h prior to coculture. Then, 10⁶ surface DNase/trypsin/DNase-treated splenocytes from a chronic WHV carrier in 1 ml of hepatocyte culture medium were placed in each permeable insert submerged in 1 ml of hepatocyte medium covering the target cells seeded in the companion well. As negative controls, WCM-260 cells or lymphoid cells in 2 ml of hepatocyte medium which were not exposed to WHV were cultured in parallel. Each coculture and control cells were set up in triplicate and incubated for 4 days. Then, the splenocytes were collected and inserts were removed. Target cells were harvested by either trypsinization (hepatocytes) or gentle scraping (lymphoid cells), washed in HBSS, and surface enzyme treated prior to DNA extraction.

Infection of woodchucks with lymphoid cell-derived WHV. The in vivo infectivity of WHV secreted by lymphoid cells was examined in WHV-naive woodchucks in two independent trials. In the first trial, WHV produced by naturally infected PBMC or splenocytes during a 72-h culture and then serially passaged in WCM-260 hepatocytes (experiments 9A and 14A, respectively) was intravenously injected at 1.3 × 10³ VGE and 7.8 × 10³ VGE into animals WM.1 and WM.2, respectively. Serum and PBMC samples were collected prior to inoculation, at days 3 and 7 p.i., and then weekly up to 20 weeks p.i. for animal WM.1 and 10 wks p.i. for WM.2. In the next step, both animals were challenged by intravenous injection with 1.1 × 10¹⁰ DNase-protected WHV VGE, exactly as described in our previous studies (4, 22). Sera and PBMC were collected at days 3 and 7, weekly until 6 weeks after challenge, and then biweekly. Follow-up of these animals ceased at 5.5 months after challenge with WHV. Sera were assayed for WHsAg, antibodies to WHsAg (anti-WHs) and antibodies to WHV core antigen (anti-WHc) using immunoassays described previously (21, 22). WHV DNA was tested in sera and PBMC by dot-blot hybridization or, if required, by PCR and Southern blot hybridization assays applying methods outlined below. Liver biopsies were obtained by laparotomy at 6 weeks p.i. and at 6 weeks after WHV challenge. Hepatic tissue was evaluated for WHV genome expression and by histology.

Detection of WHV DNA and cccDNA. WHV DNA content in sera, lymphoid cell, hepatocyte culture supernatants, and cell washes was determined, in the first instance, by a dot-blot hybridization assay using ³²P-labeled complete, recombinant, linearized WHV (rWHV) DNA as a probe (29). This assay was carried out according to a previously published method (4) and was followed by densitometric quantitation of hybridization signals using twofold serial dilutions of rWHV DNA as the reference and a Cyclone phosphorimaging system (Canberra-Packard Canada Ltd., Mississauga, Ontario, Canada). The estimated sensitivity of this assay was 2 × 10⁶ WHV VGE/ml. In the case of a negative result, WHV DNA was examined by direct or, if required, nested PCR followed by Southern blot analysis of the amplified products (see below).

Detection of WHV antigens. Approximately 10⁶ WCM-260 hepatocytes which had been exposed to test WHV inocula and harvested at day 4 p.i. were exposed to rabbit anti-WHs or rabbit anti-WHc antibodies for 20 min on ice. Cells were washed in HBSS by pelleting at 45 × g for 5 min at 4°C and stained for 20 min on ice with goat anti-rabbit immunoglobulin G (IgG; heavy and light chains) conjugated to fluorescein isothiocyanate (Organon Teknika Cappel, Durham, N.C.). After washing, hepatocytes were fixed in 1% (wt/vol) paraformaldehyde and analyzed with a FACStar-Plus flow cytometer (Becton Dickinson, Palo Alto, Calif.). Virus-naive and WHV-exposed WCM-260 cells, unstained, exposed to secondary antibody alone or to rabbit preimmune sera, and then stained with secondary antibody, were used as controls. Flow cytometry results were analyzed with CellQuest software (Becton Dickinson). Histograms of the test hepatocytes were overlaid on those of negative controls (i.e., naive hepatocytes stained with the same antibodies), and the number of events under the curve in non-overlapping areas was gated, counted, and expressed as a percentage of the total number of events detected in the test sample. Values equal to or greater than 10% were considered positive.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Features of WHV infection in primary lymphoid cell cultures. Splenic and peripheral lymphoid cells isolated from animals with serum WHsAg-positive, chronic viral hepatitis consistently displayed WHV DNA and cccDNA using PCR-based detection methods. Analyzed splenocytes demonstrated 10- to 100-fold-greater levels of virus DNA and cccDNA than those encountered in comparable numbers of peripheral lymphoid cells. These naturally infected cells survived in hepatocyte culture medium for up 4 days without alterations in their viability and intracellular WHV content. The levels of WHV genome expression remained constant during this time period and comparable to those detected in the freshly isolated cells (data not shown). To exclude the possibility that detection of cellular WHV-specific nucleic acid signals in freshly isolated or cultured lymphoid cells might be compromised by DNA from surface-attached virions or virus DNA fragments, the lymphoid cells were treated with DNase/trypsin/DNase and extensively washed prior to DNA extraction. Examination of the last cell washes did not reveal WHV DNA reactivity, implying the absence or complete removal of the possible extracellular viral material. It was further observed that WHV genome expression in lymphoid cells decreased if the culture progressed for longer than 4 days. This event was associated with a rapid decrease in cell survival and was neither virus nor medium related, since both virus-naive and WHV-infected PBMC survived to the same extent in either hepatocyte culture medium or RPMI 1640 medium that was used for lymphoid cell cultures in our previous studies (4, 8). It was also noticed that the level of WHV expression and the rate of lymphoid cell survival in hepatocyte medium were not enhanced in the presence of mitogens, such as phytohemaglutinin (PHA; 5 µg/ml) or lipopolysaccharide (LPS; 10µg/ml) (data not shown). Overall, the results revealed that conditions designed for culture of woodchuck hepatocytes were also adequate for maintenance of lymphoid cells and that they supported the transiently steady level of WHV replication in these cells.

Cultured woodchuck hepatocytes support a complete cycle of WHV replication. WCM-260 hepatocytes exposed to WHV inocula and harvested daily prior to reaching confluence expressed WHV DNA from the moment of the completion of the 24-h incubation with virus (i.e., day 0 p.i.) and viral cccDNA typically from day 2 p.i. (Fig. 1). DNA from WCM-260 cells not exposed to virus, and the final washes from the infected, surface-treated WCM-260 did not show WHV DNA reactivity. Densitometric analysis of hybridization signals demonstrated that the levels of intracellular WHV DNA remained essentially constant between days 0 and 3 p.i. in hepatocytes exposed to 10⁷ VGE/ml, whereas the virus cccDNA level was highest at day 2 p.i. and then decreased. It was also found that the degree of WHV DNA and cccDNA expression in WCM-260 cultures depended to some extent on the amount of virus in the inoculum. Thus, the viral DNA level progressively increased when the WHV content of the inoculum was raised to ~10⁷ VGE/ml (Fig. 2). WHV amounts greater than 10⁷ VGE/ml usually did not further enhance the intensity of virus DNA signals (data not shown). It was also observed that exposure of WCM-260 cells to inocula with 10⁷ VGE/ml or greater led with the highest frequency to infection accompanied by cccDNA (Fig. 2 and Table 1). The appearance of cccDNA sequences in a time-dependent and virus dose-related manner provided strong evidence that cccDNA had been synthesized in WCM-260 and specifically detected by the assay employed.

View larger version (29K): [in this window] [in a new window]   FIG. 1.   Time course of appearance and expression of WHV DNA and cccDNA in cultured woodchuck hepatocytes infected in vitro with WHV. WCM-260 hepatocytes were incubated for 24 h with woodchuck serum containing 107 WHV VGE/ml. Cells were harvested at the indicated days postinoculation (d.p.i.), washed, surface treated with DNase/trypsin/DNase, and washed again. The extracted DNA was digested with mung bean nuclease and analyzed for WHV DNA and cccDNA by a differential PCR protocol described under Materials and Methods. rWHV DNA was included as a positive control, and mung bean nuclease-digested DNA from virus-naive WCM-260 hepatocytes was used as a negative control. The amplified direct PCR products were detected by Southern blot hybridization to an rWHV DNA probe. Positive samples show the expected sizes of nucleotide fragments noted by arrowheads.

View larger version (25K): [in this window] [in a new window]   FIG. 2.   Effect of various amounts of WHV in inoculum on WHV DNA and cccDNA expression in in vitro-infected woodchuck hepatocytes. WCM-260 cells were exposed for 24 h to culture supernatant containing the indicated amounts of WHV obtained after a 72-h culture of splenocytes isolated from a chronic WHV carrier. After removal of inoculum, hepatocytes were cultured for an additional 3 days and harvested. DNA extracted from hepatocytes was digested with mung bean nuclease and assayed for WHV DNA and cccDNA by direct and nested PCR, respectively, and the amplified products were analyzed by Southern blot hybridization to an rWHV DNA probe. WHV DNA was included as a positive control, and mung bean nuclease-treated DNA from virus-naive WCM-260 hepatocytes and a mock sample extracted and treated in parallel with test samples were used as negative PCR controls. Positive samples showed nucleotide fragments of the expected sizes, as noted by arrowheads. Numbers under the upper panel represent 103 WHV genome copies (VGE) per µg of total DNA in infected hepatocytes, estimated by densitometric analysis of hybridization signals.

Susceptibility of woodchuck hepatocytes to serum- and lymphoid cell-derived WHV. Assessment of WHV DNA in inocula prepared from either lymphoid cells or sera and subjected to extensive DNase digestion revealed that the vast majority (~80%) of WHV DNA-reactive molecules resisted the enzyme treatment. This implied that they were encased within a protective envelope, as would occur in virions. In consequence, it was accepted that the estimated VGE levels reflected the virion contents with a high degree of accuracy. The same DNase treatment of plasmid-derived DNA completely eliminated WHV DNA reactivity, as determined by Southern blot hybridization of PCR products (data not shown). To ascertain the specificity of WHV cccDNA determination, DNA from infectious WHV that was secreted in culture by in vivo-infected splenocytes (experiment 14, Table 1) was digested with mung bean nuclease and then amplified. The nuclease-treated DNA cannot be detected with the nick-spanning primers but was readily identifiable by PCR with primers specific for the core gene, as illustrated in Fig. 4A. The equivalent amount of the nuclease-treated WHV DNA derived from the liver of a chronically infected animal showed evident signals for both cccDNA and the core gene. This demonstrated that cccDNA can be detected in the intracellular WHV genome material but not in the extracellular virions.

View larger version (20K): [in this window] [in a new window]   FIG. 3.   Susceptibility of woodchuck hepatocytes to infection with serum WHV. WCM-260 cells were exposed to WHV at 1.2 × 109 VGE/ml derived from WHV-positive woodchuck serum as described under Materials and Methods. (A) Expression of WHV DNA and cccDNA. DNA was extracted at day 3 p.i. from hepatocytes inoculated with virus (WHV:WCM260) and from hepatocytes cultured in the absence of virus (a negative control; WCM260). DNA was digested with mung bean nuclease and analyzed for WHV DNA and cccDNA by direct and nested PCR, respectively, and then Southern blot hybridization. DNA from the liver of a chronic WHV carrier treated under exactly the same conditions was included as a positive control. (B) Fluorescence histograms of WHsAg expression. Hepatocytes exposed to WHV inoculum (solid line) and virus-naive cells (a negative control; dotted line) were probed with rabbit anti-WHs antibodies and secondary goat antibodies to rabbit IgG and then analyzed by flow cytometry. Cell numbers (counts) are plotted against the log of fluorescence intensity (FL1-Height). The percentage of hepatocytes reactive for WHsAg is indicated inside the panel.

View larger version (18K): [in this window] [in a new window]   FIG. 4.   Analysis of WHV inoculum produced by naturally infected splenic lymphoid cells and susceptibility of woodchuck hepatocytes to this inoculum. (A) WHV at 2 × 107 VGE secreted during a 72-h culture of splenocytes isolated from a chronic WHV carrier (Spl WHV) and a comparable amount of WHV DNA from chronically infected liver (L WHV) were digested with mung bean nuclease and analyzed for WHV DNA and cccDNA by a differential PCR protocol. Mung bean nuclease-treated DNA from virus-naive WCM-260 cells (WCM260) was used as a negative PCR control. (B and C) WCM-260 hepatocytes were exposed to the same splenocyte-derived inoculum at 2 × 107 VGE/ml and analyzed as described in Materials and Methods. (B) Expression of WHV DNA and cccDNA. DNA was extracted from hepatocytes incubated with virus (WHV:WCM260) and from virus-naive hepatocytes cultured under the same conditions (WCM260). DNA was amplified and analyzed by Southern blot hybridization to detect WHV DNA and cccDNA. rWHV DNA was included as a positive control. (C) Flow cytometry histograms of WHcAg expression. Hepatocytes exposed to WHV inoculum (solid line) and virus-naive control cells (a negative control; dotted line) were stained with rabbit anti-WHc antibodies and secondary goat antibodies to rabbit IgG. Further details are provided in the legend to Fig. 3.

Infectivity of lymphoid cell- and serum-derived inocula with equal WHV content. It was found that identical numbers of splenic lymphoid cells and PBMC isolated from the same chronic carriers released distinctively different quantities of WHV. Hence, while 72-h culture supernatant from 2 × 10⁸/ml splenocytes normally contained more than 10⁷ WHV VGE/ml, the same number of PBMC secreted no more than 3 × 10⁴ VGE/ml. This finding was not entirely surprising considering the already mentioned fact that freshly isolated splenocytes expressed between 10- and 100-fold-greater amounts of WHV than PBMC. Nevertheless, it was of interest to assess whether WHV originating from splenic and peripheral lymphoid cells and from serum of the same host may vary in its potential to induce infection in hepatocytes. To test this, WCM-260 cells were exposed to identical VGE amounts of the virus secreted by splenocytes or PBMC or that present in serum of the same chronically infected woodchucks. In one of the trials, hepatocytes were exposed to WHV derived from splenocytes, PBMC, or serum at 10⁴ VGE/ml. The results revealed that while WHV DNA was expressed in hepatocytes incubated with all three types of inocula, WHV cccDNA signal was detected only in the cells exposed to virus derived from splenocytes (data not shown). In another experiment, WCM-260 hepatocytes were incubated with inocula prepared from splenocytes or serum of the same chronic carrier containing 10⁷ VGE/ml (Fig. 5). Analysis of DNA from day 3 p.i. hepatocyte cultures showed comparable levels of intracellular WHV DNA. However, cccDNA was detected only in the cells exposed to WHV produced by splenocytes (Fig. 5). Identical results were obtained when these inocula were tested on two different occasions. These observations raised the possibility that WHV inocula derived from splenocytes might somehow have a greater intrinsic ability to establish active replication in hepatocytes than that from serum or circulating lymphoid cells.

View larger version (34K): [in this window] [in a new window]   FIG. 5.   Comparison of in vitro infectivity of inocula with equal WHV content derived from serum and splenocytes of the same chronic WHV carrier. One-day WCM-260 hepatocyte cultures were exposed for 24 h to 107 DNase-protected WHV VGE/ml obtained from either serum (Ser WHV:WCM260) or splenocytes (Spl WHV:WCM260) of the same chronic WHV carrier and cultured under conditions described in Materials and Methods. DNA extracted from day 3 p.i. hepatocyte cultures was digested with mung bean nuclease and amplified with primers specific for WHV core gene by direct PCR or the virus nick region by nested PCR. As a positive control, mung bean nuclease-digested DNA from the liver of a chronic WHV carrier was used. DNA from virus-naive WCM-260 hepatocytes treated with mung bean nuclease and a mock sample treated exactly like the test samples were included as negative controls. Positive samples showed hybridization signals of the expected molecular sizes, as noted by arrowheads.

Serial passage of WHV from lymphoid cells in hepatocyte cultures. WCM-260 hepatocytes exposed to 3 × 10⁴ VGE/ml of WHV released during 72-h culture by naturally infected PBMC (experiment 9, Table 1) became virus DNA positive, and the pellet of their 3-day culture supernatant was reactive for viral DNA (Fig. 6A). Further, 19% of the hepatocytes were found to be reactive for WHcAg by flow cytometry, while WHsAg was not detected (Table 1). The culture medium harvested after the first passage in WCM-260 cells contained an estimated 1.3 × 10⁴ WHV VGE/ml. It was determined that hepatocytes collected after each of the five consecutive passages (experiment 9A) expressed WHV at comparable levels of about 2 × 10² VGE/µg of cellular DNA (Fig. 6B). The culture medium after the final 6th virus passage in WCM-260 cells contained approximately 2.2 × 10³ VGE, of which 1.3 × 10³ VGE was injected into woodchuck WM.1. These results showed that WHV released by naturally infected peripheral lymphoid cells was infectious to homologous hepatocytes even after several passages and that the virus can be maintained in vitro by serial propagation.

View larger version (31K): [in this window] [in a new window]   FIG. 6.   Serial passage of WHV secreted by naturally infected lymphoid cells in woodchuck hepatocyte cultures. WCM-260 cells were exposed to supernatant containing 3 × 104 WHV VGE/ml from cultured PBMC isolated from a chronic WHV carrier (experiment 9, Table 1). The virus produced by hepatocytes after day 3 p.i. of culture was passaged five consecutive times in WCM-260 cell cultures as described under Materials and Methods. DNA was extracted (A) from hepatocytes after the first WHV passage (WHV:WCM260) and from 100 µl of the cell supernatant (c.s.) or the suspended pellet obtained after ultracentrifugation of the supernatant (c.s.*) and (B) from hepatocytes after each of the five subsequent virus passages (passages 2 to 6) and 100 µl of the ultracentrifugation concentrated culture supernatant collected after passage 6 (6 × c.s.*). WHV DNA was assayed by direct PCR (A) or nested PCR (B) with WHV core gene-specific primers and detected by Southern blot hybridization to the rWHV DNA probe. Mung bean nuclease-digested DNA from a chronically infected liver or rWHV DNA was included as a positive control, and DNA from virus-naive WCM-260 hepatocytes was used as a negative control.

Transmissibility of WHV from lymphoid cells to virus-naive hepatocytes or lymphoid cells in cocultures. To investigate whether WHV secreted by lymphoid cells can directly infect virus-naive hepatocytes or lymphoid cells, splenocytes isolated from a chronic WHV carrier were cocultured with WCM-260 cells or with freshly isolated PBMC obtained from healthy woodchucks. In splenocyte-hepatocyte cocultures, intracellular WHV DNA and cccDNA signals become readily detectable in WCM-260 cells after 4 days in culture. It was estimated that there were approximately 3 × 10⁴ copies of WHV DNA and 5 × 10³ copies of cccDNA per µg of total hepatocellular DNA. Assuming that at least one copy of WHV DNA or cccDNA existed in the infected cell, one in five hepatocytes contained a copy of virus DNA and 1 in 40 contained a copy of cccDNA. Therefore, hepatocytes infected in cocultures with splenocyte-derived virus had five- to sixfold-greater WHV levels than those exposed to culture supernatant obtained from the same splenocytes (data not shown).

View larger version (31K): [in this window] [in a new window]   FIG. 7.   In vitro infection of woodchuck lymphoid cells with WHV from chronically infected splenocytes. (A) PBMC (107) isolated from a healthy woodchuck were incubated for 24 h with 108 VGE of WHV produced by splenocytes prepared from a chronic WHV carrier. After a 24-h icubation, the inoculum was removed, and cells were washed and cultured for an additional 72 h. DNA was extracted from surface DNase/trypsin/DNase-treated cells (SplWHV:PBMC), digested with mung bean nuclease, and assayed for WHV DNA and cccDNA by nested PCR and Southern blot hybridization as described under Materials and Methods. (B) Virus-naive PBMC (106) from a healthy woodchuck were cocultured for 4 days with 106 DNase/trypsin/DNase-treated splenocytes from the same chronic carrier as in panel A. DNA was extracted from both infected (SplWHV:PBMC) and lymphoid cells (splenocytes) that supplied the virus. DNA treated with mung bean nuclease was analyzed for WHV DNA and cccDNA expression as in panel A. DNA from the liver of a chronic WHV carrier treated under the same conditions as the test samples was used as a positive control, and similarly treated DNA from virus-naive PBMC and mock-infected samples were included as negative controls.

In vivo infectivity of lymphoid cell-derived WHV. Analysis of serum samples from woodchucks WM.1 and WM.2, inoculated with low doses (~10³ VGE) of WHV produced by naturally infected PBMC or splenocytes and than serially passaged in WCM-260 hepatocyte cultures, showed the appearance of WHV DNA beginning at day 7 p.i. (Fig. 8A). The virus DNA persisted in serum at approximate levels of 10² VGE/ml until the animals were challenged with a massive WHV dose (1.1 × 10¹⁰ DNase-protected VGE) at 10 or 20 weeks after the initial inoculation. Prior to WHV challenge, small amounts of WHV DNA and, occasionally, WHV cccDNA were detectable in serial PBMC samples, but not in liver tissue collected at week 6 p.i. Interestingly, although WHV infection was evident at the molecular level, serum remained negative for WHsAg and anti-WHc antibodies. After challenge with a massive WHV dose, both animals developed a typical, histologically evident acute hepatitis accompanied by WHsAg and anti-WHc in serum (Fig. 8A). The disease was self-limiting and superseded by serological recovery, although WHV DNA persisted at the low levels in serum, PBMC, and liver until the end of the 5-month observation period.

View larger version (41K): [in this window] [in a new window]   FIG. 8.   Serological and molecular profiles of WHV infection in woodchucks inoculated with WHV that was secreted by naturally infected lymphoid cells and then serially passaged in WCM-260 hepatocytes or inoculated with equal doses of WHV obtained from splenocytes or serum of the same chronic WHV carrier. (A) WHV produced by PBMC or splenic lymphoid cells isolated from chronic WHV carriers were passaged six consecutive times in WCM-260 cells (WHV/PBMC/6×WCM260 and WHV/Spleen/6×WCM260), and the final culture supernatants containing the indicated amounts of WHV were intravenously injected into virus-naive woodchucks WM.1 and WM.2, respectively. (B) Equivalent amounts of DNase-protected WHV secreted by cultured splenocytes or obtained from serum of the chronic WHV carrier that provided the splenocytes were intravenously injected into virus-naive woodchucks WF.3 and WM.4, respectively. The graphs display the appearance and the span of duration of serum WHsAg, anti-WHs, and anti-WHc antibodies and expression of WHV DNA in serum, PBMC, and liver tissue samples. WHV DNA in serum and PBMC samples was detected by direct (dark grey bar) and nested (light grey bar) PCR followed by Southern blot analysis of the amplified sequences. The level of WHV genomes in liver tissue was assessed by direct PCR followed by Southern blot analysis of amplicons and is presented as values of 0, 1+, 2+, and 3+, corresponding to 0, 2 to 20, 2 × 103, and >2 × 106 WHV VGE per 104 cells, respectively, which were estimated as reported previously (22). Liver histology: N, normal; AH, acute hepatitis; MIN, minimal (residual) hepatitis.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Lymphoid cells isolated from animals infected with WHV and depleted of surface-attached virus genomic material displayed virus DNA and the replicative precursor cccDNA and released viral particles infectious to cultured woodchuck hepatocytes and lymphoid cells and to susceptible animals. These naturally infected lymphoid cells, of either peripheral or splenic tissue origin, survived for up to 4 days under conditions designed for woodchuck hepatocyte culture without appreciable changes in their viability and intracellular WHV content. This transiently steady expression of WHV indicated that the cells support productive virus replication in vitro and can provide lymphotropic virus for in vivo and in vitro experimentation. It was also discovered that the efficiency of WHV replication in lymphoid cells maintained in hepatocyte culture conditions remained unchanged upon stimulation with mitogens, such as LPS and PHA. This finding was in contrast to the data reported previously showing that culture of lymphoid cells from WHsAg-positive chronic or occult WHV infection in the presence of bacterial LPS, PHA, or concanavalin A enhanced synthesis of viral transcripts and secretion of infectious virus (4, 11, 22). This discrepancy was most likely due to the presence of dexamethasone in the hepatocyte medium used for culture of lymphoid cells in the present study. This glucocorticoid has been shown to upregulate HBV DNA and protein expression in HBV-transfected and -infected hepatoblastoma cell lines (2, 33, 39). Therefore, it is likely that WHV replication in lymphoid cells in our study was unresponsive to mitogen stimulation because it was already upregulated by the glucocorticoid. This was supported by our other observations that WHV-infected PBMC cultured for 72 h in RPMI 1640 medium (4, 8) with 10 µM dexamethasone displayed WHV DNA signals at twofold-greater intensity than PBMC cultured without the hormone and that these signals were comparable in strength to those seen in equivalent numbers of cells cultured in RPMI 1640 medium containing LPS mitogen (data not shown).

A continuous WCM-260 hepatocyte line established from the liver of a healthy adult woodchuck was found to be susceptible to WHV infection and able to support a complete cycle of virus replication years after its founding. The active propagation of WHV in WCM-260 hepatocytes was evidenced by (i) appearance of virus DNA and cccDNA after exposure to different WHV inocula prepared from infected sera or lymphoid cells, (ii) synthesis and cell surface expression of WHV core and surface antigens, and (iii) secretion of viral particles infectious to cultured hepatocytes and to virus-naive animals. Analysis of WHV DNA and cccDNA expression within WCM-260 cells at different days p.i. showed that while virus DNA became identifiable at the end of a 24-h incubation with virus, cccDNA was first detected at day 1 or 2 p.i. This delayed appearance of cccDNA in WCM-260 hepatocytes is consistent with the previous observations on WHV and duck HBV virus (DHBV) replication in primary woodchuck or duck hepatocytes (1, 3, 31, 40, 41). It was postulated that this postponement in cccDNA formation could be a consequence of various intracellular processes, which may include inefficiencies in virus genome uncoating and/or nucleocapsid translocation to the nucleus, a requirement for accumulation of the virus products controlling cccDNA assembly, or be due to an intrinsic inefficiency of the cell cultures employed (1, 9, 31).

In our culture system, proliferating hepatocytes were exposed to WHV from lymphoid cells or sera and maintained for 3 days after inoculation, i.e., 4 days after addition of virus, to the moment when they become fully confluent. The kinetics of cccDNA expression appeared to follow the dynamics of WCM-260 cell proliferation, which achieved maximum at day 3 after seeding, as determined by a microculture tetrazolium assay (6, 25) (data not shown). This may suggest that WHV cccDNA formation could depend upon the pace of new cell formation. In this regard, it has been shown previously that the early phase of DHBV cccDNA amplification in in vitro-infected primary duck hepatocytes occurred only when cells underwent division (38). The suppression of hepatocyte division at the G₁ phase inhibited the accumulation of DHBV cccDNA, whereas restoration of cell proliferation increased cccDNA expression in these cells. The dependence of hepadnaviral cccDNA formation on the cell cycle was thought to be due to the transport of viral nucleocapsids to the nucleus, which appears to be upregulated during the G₁ phase (43). In our system, the delay in the appearance of cccDNA could be a consequence of rounds of virus replication in newly formed cells that led to progressive accumulation of cccDNA to the level that became detectable by the PCR and Southern blot hybridization assay.

The levels of WHV DNA and cccDNA expression in WCM-260 hepatocytes infected with WHV derived from lymphoid cells or sera were lower than those reported for primary woodchuck hepatocytes (1, 24). In our work, it was estimated that at least 1 in 30 and 1 in 300 WCM-260 hepatocytes carried a copy of the WHV genome and cccDNA, respectively. The levels of these nucleic acids were six- to sevenfold greater for WCM-260 hepatocytes cocultured with WHV-infected splenocytes. In comparison, for WHV-infected primary woodchuck hepatocytes, it has been reported that each cell presumably carried a copy of cccDNA (1, 24). However, there are important differences between our experimental system and that mentioned above, which need to be considered when comparisons are attempted. The unique features of our approach were (i) the use of a stable woodchuck hepatocyte line in the active phase of cell proliferation instead of primary hepatocytes seeded at confluence this provided an unvarying supply of well-characterized, proliferating hepatocytes but potentially decreased susceptibility of the cells to WHV infection because of their long-term culture; (ii) an extensive enzymatic treatment of infected hepatocytes prior to DNA extraction that eliminated carryover of extracellular viral material but caused a partial loss (by approximately 20%) of the cells available for molecular analysis; and (iii) the inclusion of the mung bean nuclease digestion step that eliminated molecules with single-stranded DNA sequences and therefore strengthened specificity of cccDNA detection but lowered viral DNA due to additional manipulations. In consequence, evaluations of virus-specific DNA and cccDNA expression were done under very stringent experimental and assay conditions, and therefore the reported levels are the most conservative estimations.

The expression of WHsAg and WHcAg varied significantly in WCM-260 hepatocytes from different experiments and was not related to the estimated intracellular levels of virus genomes. Since there is no direct relationship between hepadnavirus DNA expression and the rate of its protein synthesis in infected hepatocytes, even in an in vivo situation, and because the levels of viral genomes were almost certainly underestimated in this study due to the stringency of the detection approach used, the above observation is not unexpected. On the other hand, a variance in the display of individual WHV antigens in infected WCM-260 could be a consequence of a number of virus- or cell-dependent factors, which remain largely unidentified for infections progressing either in vivo or in cell culture conditions. Since the inocula used in this study were derived from different chronic WHV carriers and sites, genotypic diversity of the infectious virus could have influenced viral replication efficiency and antigen synthesis in individual experiments. Other potential factors include variations in the rates of viral antigen secretion from infected cells, intrinsic characteristics of WCM-260 cells, and the relative insensitivity of the detection method used. This complex issue will require further investigation.

The establishment of the in vitro system in the present study permitting culture of hepatocytes and lymphoid cells under the same microenvironment provided a significant experimental advantage that enabled direct transfer of WHV synthesized by lymphoid cells to hepatocyte cultures and analysis of virus infectivity under uniform in vitro conditions. This also constituted a basis for the development of the lymphoid cell-hepatocyte and lymphoid cell-lymphoid cell cocultures for direct cell infection experiments. The results obtained provided definitive in vitro evidence that WHV synthesized by lymphoid cells is infectious to host hepatocytes and lymphoid cells. This fact was documented using either culture supernatants containing WHV secreted by naturally infected lymphoid cells or WHV directly permeating from the infected lymphoid cells to virus-naive cell targets in cocultures.

Although the WHV content was found to be an important determinant of the inoculum's ability to establish infection in WCM-260 hepatocytes, virus origin also appeared to have some influence on the efficiency of this process. It was noticed that when inocula with equal WHV content prepared from splenocytes, PBMC, or serum from the same chronic carriers were used to infect WCM-260 cells, virus derived from splenocytes had the highest potency to initiate replication, as evidenced by the appearance of cccDNA. This finding raised the possibility that WHV propagating within cells of lymphatic organs might somehow be more invasive toward host hepatocytes than that circulating freely or present within PBMC. At this stage, the basis of this phenomenon remains unexplained. It is conceivable that the presence in inoculum of virus genotypic and structural variants and/or replication-defective particles might modify the virus's final ability to recognize and invade permissive cells. It also is possible that the organ- or cell type-specific processes controlled, for example, by local or intracellular proteases may contribute to this process. In this regard, of note are our previous findings demonstrating that proteolytic cleavage of the WHV envelope is essential for presentation of the cell binding site recognizing woodchuck hepatocytes and lymphoid cells in a strictly host and cell type-specific manner (8). This site (designated WHV CBS1) has been identified in the N-terminal portion of the pre-S1 domain of the large WHV envelope protein and mapped to amino acids 10 to 13 of the domain. In addition, synthetic analogues of the site displayed considerably greater capacity for binding to woodchuck lymphoid cells than to hepatocytes. This ability was further augmented when lymphoid cells stimulated to cell division and blastoid formation by mitogens were used as targets (8). From these findings, we have postulated that dedifferentiated or immature lymphoid cells in the lymphatic organs, which embrace the most active lymphoid cell proliferation during adolescence, might be preferential targets and perhaps initial sites of WHV replication (8).

In an attempt to recognize whether the virus produced by lymphoid cells is as pathogenic as that existing freely in the circulation and to determine whether the observed increased potential to induce replication in cultured hepatocytes by splenocyte-derived virus has its equivalent in vivo, comparable amounts of WHV secreted by naturally infected splenic lymphoid cells or existing in serum of the same chronic WHV carrier were DNase treated and administered to virus-naive woodchucks. The results revealed that the virus originating from these two extrahepatic locations induced typical, self-limiting acute hepatitis with roughly similar profiles of WHV serological markers. This experiment revealed that virus produced by naturally infected organ lymphoid cells is infectious and pathogenic to at least the same degree as that occurring freely in the circulation. Accordingly, no persuasive evidence has been obtained so far that the splenocyte-derived virus might be better predisposed to induce more severe liver injury. In this regard, our previous study has shown that the same pool of WHV derived from peripheral lymphoid cells induced dramatically different patterns of viral hepatitis in individual animals, indicating that the host milieu may profoundly influence the infection outcome (22). This indicates that more extensive investigation in a series of WHV-naive woodchucks infected with various doses of the lymphoid cell-derived virus will be required to make conclusions with full confidence about the pathogenic vigor of hepadnavirus propagating in lymphatic organs.

WHV produced by infected circulating or splenic lymphoid cells remained infectious to virus-naive animals after several passages in hepatocyte cultures. This result demonstrates that the virus continued to be biologically competent in spite of the multistep transfer in permissive cells and that our hepatocyte culture system did not alter the virus's infectious capabilities. The virus recovered following passage in WCM-260 hepatocytes was injected at amounts not exceeding 10⁴ VGE and caused infection characterized by the presence of WHV genomes in serum and circulating lymphoid cells, but not in the liver. Moreover, this infection advanced in the absence of WHV serological markers, i.e., WHsAg, anti-WHs, and anti-WHc, and did not provide protection against challenge with a massive WHV dose (1.1 × 10¹⁰ VGE). In this context, it is important to indicate that we have previously observed exactly the same pattern of WHV infection when investigating offspring born to mothers recovered from WHV hepatitis (4), woodchucks inoculated with WHV derived from serum or lymphoid cells of these offspring (18; Coffin et al., unpublished), and a woodchuck experimentally inoculated with viable PBMC collected years after resolution of acute WHV hepatitis (22). A comparative analysis of the data from these studies and from the current work points to a common factor predisposing to the development of this occult, nonprotective, lymphatic system-restricted infection. The uniform prerequisite appears to be a small amount of the invading virus (reviewed in reference 18). At this stage, other viral factors, including virus variants with tropism for lymphoid cells and not yet recognized elements of the host milieu, need to be considered, although their contributions seems to be less decisive. Further studies, currently in progress in our laboratory, should clarify this important issue.

The results of this study provide straightforward evidence that lymphoid cells are the sites of active replication of hepadnavirus infectious to homologous hepatocytes and lymphoid cells. They substantiate a motion based on fragmented clinical and experimental observations postulating that circulating lymphoid cells can be hepadnavirus chaperons from which virus may strike the liver. The data also reinforce the concept that invasion of the lymphatic system is an unavoidable consequence of hepadnavirus infection, even when the liver remains unaffected (reviewed in reference 18). As in other viral infections that involve the lymphatic system, the presence of replicating virus in lymphoid cells may have multiple detrimental effects on the host regulatory and protective immune responses.

Among others, as we have recently shown for chronic WHV infection (20), hepadnavirus severely reduces the display of class I MHC molecules on lymphoid cells and, in consequence, may potentially alter cell-to-cell communication and deregulate immune cell functions. These virus-induced effects, although not always directly apparent, may have generalized dissenting effect on the immune system and contribute to long-term virus persistence in the lymphatic system.