Immunosuppression reactivates viral replication long after resolution of woodchuck hepatitis virus infection


  • Stephan Menne,

    Corresponding author
    • Gastrointestinal Unit, Department of Clinical Sciences, College of Veterinary Medicine, Room C-2005 VMC, Cornell University, Ithaca, NY 14853
    Search for more papers by this author
    • fax: 607-253-3289

  • Paul J. Cote,

  • Scott D. Butler,

  • Ilia A. Toshkov,

  • John L. Gerin,

  • Bud C. Tennant

  • Potential conflict of interest: Nothing to report.


Resolution of hepatitis B virus (HBV) infection is characterized by coordinated humoral and cellular immune responses. Immunity is durable over decades, protecting the host from reinfection and potential activation of residual HBV. Woodchucks infected at birth with woodchuck hepatitis virus (WHV) cleared viremia and developed antibodies to surface antigen (anti-WHs). Woodchucks became seronegative for anti-WHs 3-6 years later, but in some, WHV DNA was detected in serum, liver, and/or peripheral blood mononuclear cells (PBMCs). Those with WHV DNA had increased in vitro cellular immune responses to viral antigens, CD4 and CD8 markers, and Th1-type cytokines, suggesting active WHV-specific T lymphocytes. Immunosuppression for 12 weeks using cyclosporine A in such woodchucks resulted in transient reactivation of WHV replication. Serum of 1 woodchuck that became positive for WHV DNA during immunosuppression was inoculated into WHV-susceptible woodchucks, and a productive infection was demonstrated. The results indicate that after infection durable cellular immunity to WHV is essential for the long-term control of viral replication and is probably maintained by continuous priming from residual virus. Conclusion: These experimental observations demonstrate the potential of immunosuppression to reactivate HBV after resolution of infection. (HEPATOLOGY 2007;45:614–622.)

Infection of adult humans with HBV results characteristically in self-limited hepatic disease and resolution of the infection based on serological and clinical parameters. The self-limited outcome of HBV infection is determined by humoral and cellular immune responses.1, 2 Resolution of HBV infection requires induction of virus-specific cell-mediated immunity (CMI) involving Th cells, cytotoxic T lymphocytes, and Th1-type cytokines that are associated with acute hepatitis and an apparently complete serological and clinical recovery.3, 4

The humoral and CMI responses that are essential for resolution of HBV infection may not completely eradicate HBV. Residual viral DNA can persist in the liver and the periphery after resolution of infection, even in the presence of virus-neutralizing antibodies to HBV surface antigen.5–9 HBV-specific T cells that display surface markers of activated phenotypes can be detected in patients after resolution6, 7 and appear to be critical for long-term control of HBV replication and progression of liver disease. Clinical reports have shown that treatment with cytotoxic or immunosuppressive drugs can reactivate HBV replication long after resolution of infection has occurred.10, 11

The woodchuck hepatitis virus (WHV) is a hepadnavirus of the Eastern woodchuck (Marmota monax) with genomic organization, biological properties, and replicative strategy essentially identical to HBV. Experimental infection of woodchucks with WHV is a model for studies of the pathogenesis of HBV infection, and for preclinical testing of vaccines and drugs for prevention of HBV disease sequelae.12, 13

In the current study, woodchucks that had been experimentally infected with WHV as neonates and in which long-term resolution of infection was established were used to assess the durability of virus-specific CMI and the effects of its modulation by cyclosporine A (CsA) immunosuppression, a drug that blocks IL-2 gene expression and inhibits virus-specific, antigen-driven, clonal expansion of Th cells.14 Immunosuppression induced transient reactivation of WHV replication, demonstrating that maintenance of a durable, WHV antigen-specific CMI is essential for the long-term control of WHV replication.


cccDNA, covalently-closed circular DNA; CMI, cell-mediated immunity; CsA, cyclosporine A; PBMC, peripheral blood mononuclear cell; SDH, sorbitol dehydrogenase; SI, stimulation index; WHcAg, WHV core antigen; WHsAg, WHV surface antigen; WHV, woodchuck hepatitis virus; WHVge, WHV genomic equivalents.

Materials and Methods


Woodchucks were inoculated at 3 days of age with the standardized WHV7P1 inoculum.15 Woodchucks that resolved WHV infection were used as described in Results. Age-matched woodchucks, negative for WHV, or chronically WHV-infected, provided negative and positive control samples of serum and tissues. Two WHV-negative woodchucks were used to test the infectivity of WHV DNA in serum from CsA-treated woodchucks. Animal experiments were approved by the Cornell University Institutional Animal Care and Use Committee.

Experimental Design.

Ten woodchucks resolved neonatal WHV infection by 6 months of age. Resolution was based on (1) clearance of serum WHV surface antigen (WHsAg) determined by enzyme-linked immunosorbent assay (sensitivity ≥ 150-250 ng/ml), (2) clearance of serum WHV DNA based on real-time PCR [sensitivity ≥1 × 103 WHV genomic equivalents (WHVge)/ml], and (3) development of antibodies against WHsAg (anti-WHs). Before CsA immunosuppression, eight 3.5-year-old to 6.5-year-old resolved woodchucks were selected, classified as having or not having residual WHV DNA in serum or tissues as determined by nested PCR (sensitivity ≥50 WHVge/assay), and with or without WHV antigen-specific CMI recall responses determined by peripheral blood mononuclear cell (PBMC) proliferation and mRNA expression assays, and separated into 2 groups. All 4 woodchucks of group 1 had residual WHV covalently closed circular DNA (cccDNA) in liver and recall CMI. Woodchucks of group 2 lacked intrahepatic WHV cccDNA, but two of the four woodchucks had residual WHV DNA in PBMCs and had either weak or undetectable recall CMI. Woodchucks of both groups received CsA (Sandimmune, Novartis Pharmaceuticals Corp., East Hanover, NJ) intraperitoneally, once daily for 12 weeks. CsA was administered at a dose of 30 mg/kg/day during the initial 2 weeks of treatment, which then was reduced to 20 mg/kg/day for the remaining 10 weeks of treatment. After cessation of CsA, woodchucks were followed for an additional 12 weeks. Blood samples were obtained 2 weeks before CsA administration, on the first day of treatment before CsA administration (“week 0”), and then biweekly during CsA treatment and follow-up.

To test the infectivity of serum WHV DNA obtained from woodchuck F3776 during week 6 of CsA treatment, 2 adult, WHV-negative woodchucks were inoculated experimentally via the sublingual vein. Both woodchucks received 0.2 ml of a 1:50 dilution of serum representing 3.4 × 107 WHVge and were monitored for 24 weeks after inoculation. Blood samples were obtained weekly.

Biochemical Testing.

Serum biochemical measurements included ALT, sorbitol dehydrogenase (SDH), and γ-glutamyltranspeptidase.16

Detection of Residual WHV DNA.

Before immunosuppression, WHV DNA in serum, PBMCs, and liver was determined by nested PCR (sensitivity ≥50 WHVge/assay) using the methods and conditions described recently.17 Briefly, WHV core and surface gene sequences or WHV cccDNA were amplified by nested PCR, followed by Southern blot hybridization using a 32P-labeled WHV recombinant DNA plasmid (pWHV8).15 For the elimination of viral particles and free WHV DNA fragments possibly attached to PBMCs, cells were subjected to DNase/trypsin/DNase treatment.17–19 For differentiating between WHV particles and free WHV DNA in serum, samples were digested with DNase.17 For differentiating between WHV cccDNA and relaxed circular WHV DNA, DNA isolated from liver was digested with mung bean nuclease to eliminate single-stranded and relaxed circular WHV DNA.17

Virological Assays.

Serum containing WHV DNA above 1 × 107 WHVge/ml was quantified by dot-blot hybridization.15 Serum with lower WHV DNA concentrations was quantified by real-time PCR (sensitivity ≥1 × 103 WHVge/ml).20 WHsAg (sensitivity ≥150-250 ng/ml), anti-WHs, and antibodies against WHV core antigen (anti-WHc) were determined by enzyme-linked immunosorbent assay.21

Histopathological and Immunohistochemical Analyses.

Liver biopsies were obtained before CsA immunosuppression. Additional biopsies were obtained at weeks 6 and 12 of CsA treatment and at weeks 6 and 12 after CsA withdrawal. From the woodchucks that were inoculated with serum from a CsA-treated woodchuck, liver biopsies were obtained before inoculation, and thereafter at 4-week intervals until week 24. Aliquots of hepatic biopsy specimens were used for histological analysis of portal and lobular hepatitis, bile duct proliferation, steatosis, liver cell dysplasia, and apoptotic hepatocytes.16 Sections of liver tissues also were stained for intrahepatic WHV core antigen (WHcAg) and WHsAg expression and for CD3-positive cells.16, 22

Analyses of CMI Responses.

PBMC recall responses of woodchucks were determined in vitro using tritiated adenine to measure proliferation of 5 × 104 cells per 200 μl AIM-V medium (Gibco, Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum (Sigma, St. Louis, MO) and 5 × 10−5 mM 2-mercaptoethanol (Sigma) after stimulation for 5 days with polyclonal activators [concanavalin A (8 μg/ml), phytohemagglutinin (2.4 μg/ml), and lipopolysaccharide (0.5 μg/ml); Sigma], viral antigens [WHcAg (1 μg/ml) and WHsAg (2 μg/ml)], and selected antigen-specific, synthetic peptides (10 μg/ml; Sigma-Genosys, The Woodlands, TX).23, 24 WHcAg-specific (C) peptides corresponded to amino acid residues 1-20, 100-119, and 112-131 of this antigen.23 WHsAg-specific (S) peptides corresponded to amino acid residues 131-150, 226-245, 341-360, and 411-431 of this antigen, starting at the N-terminus of the large surface protein.25 Negative control stimulators consisted of bovine serum albumin (1 and 2 μg/ml; Sigma) and a WHV-unrelated synthetic peptide (10 μg/ml). PBMCs also were cultured in the presence of WHcAg (1 μg/ml) or WHsAg (2 μg/ml) and collected after 2.5 days of culture. PBMCs were lysed using the RNeasy Kit from Quiagen (Valencia, CA) and preserved at −70°C for subsequent measurement of mRNA expression of woodchuck leukocyte surface markers (CD3, 4, and 8) and Th1-type cytokines (IL-2, tumor necrosis factor-alpha, and interferon-gamma) by real-time reverse transcription PCR.26

Counts per minute (cpm; proliferation assay) or copy numbers/ng cellular RNA (mRNA expression assay) of replicate cultures were averaged and expressed as a stimulation index (SI) or fold-increase by dividing the average cpm or copy number in the presence of stimulator by that in the absence of stimulator. An SI or fold-increase value of ≥3.1 represents a positive response, whereas a value in the range of ≥2.1 and <3.1 represents a weak or borderline response. No response was defined as an SI or fold-increase value <2.1 than the average SI or fold-increase from three uninfected control woodchucks.


WHV DNA Persists Long After Resolution of Neonatal Infection.

Ten woodchucks at 3.5 to 6.5 years of age were used in this study. The woodchucks were inoculated with WHV at 3 days of age, became productively infected, and then resolved WHV infection based on the clearance of WHV DNA and WHsAg and development of anti-WHs in serum. Anti-WHs was detected at 6 months of age and subsequently during the first year of life in all woodchucks; thereafter, anti-WHs titers waned, and detection was variable in individual woodchucks. Immediately before CsA immunosuppression, anti-WHs was undetectable in 9 and was borderline positive in one (F3753) (Table 1). Anti-WHc was detected in all woodchucks at 6 months of age and always thereafter.

Table 1. Virological and Histological Parameters of Adult Woodchucks with Long-term Resolution of Experimental Neonatal WHV Infection Before Immunosuppression With Cyclosporine A
Woodchuck (ID no.)Time After Resolution of WHV Infection (years)WHV cccDNA LiverWHV DNA*Anti-WHsLiver Examination
SerumPBMCsHistopathologyImmunohistochemistryCD3-Positive Lymphocytes
  • NOTE. All woodchucks were positive for anti-WHc in serum at the time of the liver biopsy, and all had normal serum activities of liver enzymes except for F3753, in which γ-glutamyltranspeptidase activity was elevated (57 IU/l).

  • *

    Detection of DNA sequences of the WHV core and/or surface genes.

  • Cutoff for a positive anti-WHs response was ≥0.05 optical density units. Anti-WHs in serum of woodchuck F3753 was borderline positive at 0.06 optical density units.

  • Lesions of hepatitis described as grade 2 to 3 hepatitis on a 0-4 scale.

  • §

    A few scattered apoptotic bodies detected within the liver.

  • Expression of WHcAg (c) or cytoplasmic WHsAg (s) in a few hepatocytes.

  • Cell numbers were ≥2.1-fold higher than the average number of 763 (657 and 869) detected in 2 WHV-negative control woodchucks.

  • Abbreviations: ID no., identification number; M, male; F, female; Pos, positive; Neg, negative.

F37225PosNegNegNegH, Ap§Elevated
F37963NegNegPosNegc, sElevated

Before immunosuppression, the liver, serum, and PBMCs were analyzed for WHV DNA by nested PCR17 (Table 1). In the livers of 6 woodchucks, WHV cccDNA was detected (Fig. 1; Table 1), and tissues also were positive for DNA sequences of the WHV core and surface genes (data not shown). In 3 of the 6 woodchucks with intrahepatic WHV DNA, WHV DNA also was detectable in serum and/or in PBMCs. Intrahepatic WHV DNA was undetectable in 4 woodchucks, but in 2, WHV DNA was detected in PBMCs.

Figure 1.

Detection of intrahepatic WHV cccDNA in woodchucks with long-term resolution of WHV infection. DNA isolated from hepatic biopsy specimens obtained before immunosuppression was tested for the presence of WHV cccDNA by nested PCR. The amplicons were detected by Southern blot hybridization to a WHV recombinant DNA plasmid. Liver DNA from WHV-negative woodchuck M3118 and from WHV carrier woodchuck F5854 were included as negative or positive controls, respectively. Hybridization signals showed the expected 655-bp amplicon. WHV −Ctrl., WHV cccDNA-negative control; WHV +Ctrl., WHV cccDNA-positive control.

In all 10 woodchucks, serum markers of liver injury (e.g., ALT and SDH) were within normal limits. In one (F3753), serum γ-glutamyltranspeptidase activity was elevated. In 9, no histopathological abnormalities were detected in liver, but in one (F3722), a moderate degree of portal hepatitis was evident (Table 1). In woodchuck F3796, there was limited immunohistochemical evidence of intrahepatic WHcAg and WHsAg expression, but the other 9 were negative (Table 1). Intrahepatic lymphoid cells that stained positive for CD3 were detected in all 10 woodchucks, and in 5 the numbers were increased compared with livers of 2 WHV-negative control woodchucks (Table 1).

Recall of WHV Antigen-Specific CMI Responses in Resolved Woodchucks Correlates With Detection of Residual WHV DNA.

In vitro PBMC recall responses to WHcAg, WHsAg, and to selected synthetic peptides of these antigens were measured before CsA immunosuppression (Tables 2 and 3). Half of the resolved woodchucks had WHcAg-specific responses, and 3 of these 5 also exhibited WHsAg-specific responses. Another woodchuck responded only to WHsAg and to one WHs peptide. Other woodchucks had no WHV antigen-specific responses and were similar to 3 WHV-negative control woodchucks. The variability in antigen-specific responses was not a result of individual variation in the PBMC response, generally because PBMC proliferation to polyclonal activators was similar among resolved and control woodchucks (data not shown).

Table 2. WHcAg-Specific Recall Cell-Mediated Immune Responses of Adult Woodchucks With Long-term Resolution of Experimental Neonatal WHV Infection Before Immunosuppression With Cyclosporine A
Woodchuck (ID no.)Stimulation Index for PBMC Proliferation to Stimulation WithFold-Increase in mRNA Expression in WHcAg-Stimulated PBMC Cultures
  1. NOTE. Values in bold were above the assay cutoff of ≥3.1 and considered positive for WHcAg-specific PBMC recall responses or WHcAg-specific mRNA recall expression of leukocyte surface markers and Th1-type cytokines in PBMC cultures. Underlined values were below the assay cutoff but were ≥2.1-fold above the average SI of PBMCs or the average fold-increase in mRNA expression of PBMC cultures of 3 WHV-negative control woodchucks. The average SI of PBMC responses to stimulation with WHcAg and WHc synthetic peptides for the control woodchucks were: WHcAg, 1.3 (1.4, 1.3, 1.3), C1-20, 1.3 (1.2, 1.3, 1.4), C100-119, 1.4 (1.4, 1.5, 1.4), and C112-131, 1.3 (1.3, 1.3, 1.2). The average cpm for unstimulated PBMCs from control woodchucks was 2245 (1816, 2396, 2528). The average fold-increase in mRNA expression of leukocyte markers and cytokines in WHcAg-stimulated PBMC cultures from control woodchucks were: CD3, 1.1 (1.2, 1.1, 1.1), CD4, 1.2 (1.3, 1.2, 1.2), CD8, 1.1 (1.2. 1.1, 1.1), IL-2, 1.1 (1.1, 1.1, 1.0), IFN-γ, 1.1 (1.3, 1.1, 1.0), and TNF-α, 1.1 (1.2, 1.1, 1.1). The average copy number/ng total RNA for unstimulated PBMC cultures from control woodchucks were: CD3, 1753 (919, 2986, 1354), CD4, 917 (641, 1265, 845), CD8, 331 (226, 449, 318), IL-2, 42 (28, 62, 36), IFN-γ, 76 (52, 109, 67), and TNF-α, 647 (483, 895, 563). Abbreviations: ID no., identification number; M, male; F, female.

Table 3. WHsAg-Specific Recall Cell-Mediated Immune Responses of Adult Woodchucks with Long-Term Resolution of Experimental Neonatal WHV Infection Before Immunosuppression With Cyclosporine A
Woodchuck (ID no.)Stimulation Index for PBMC Proliferation to Stimulation WithFold-Increase in mRNA Expression in WHsAg-Stimulated PBMC Cultures
  1. NOTE. Values in bold were above the assay cutoff of ≥3.1 and considered positive for WHsAg-specific PBMC recall responses or WHsAg-specific mRNA recall expression of leukocyte surface markers and Th1-type cytokines in PBMC cultures. Underlined values were below the assay cutoff but were ≥2.1-fold above the average SI of PBMCs or the average fold-increase in mRNA expression of PBMC cultures of 3 WHV-negative control woodchucks. The average SI of PBMC responses to stimulation with WHsAg or WHs synthetic peptides for the control woodchucks were: WHsAg, 1.2 (1.2, 1.3, 1.2), S131-150, 1.1 (1.1, 1.2, 1.1), S226-245, 1.3 (1.2, 1.3, 1.3), S341-360, 1.2 (1.2, 1.1, 1.3), and S411-431, 1.1, (1.2, 1.2, 1.1). The average fold-increase in mRNA expression of leukocyte markers and cytokines in WHsAg-stimulated PBMC cultures from the control woodchucks were: CD3, 1.0 (1.1, 1.0, 1.0), CD4, 1.1 (1.1, 1.0, 1.1), CD8, 1.1 (1.1. 1.0, 1.1), IL-2, 1.0 (1.1, 1.0, 1.0), IFN-γ, 1.2 (1.2, 1.2, 1.1), and TNF-α, 1.0 (1.1, 1.0, 1.0). For average cpm and average copy number/ng total RNA for unstimulated PBMC cultures from control woodchucks see Table 2. Abbreviations: ID no., identification number; M, male; F, female.


The overall strength and specificity of PBMC recall responses to WHV antigens in individual resolved woodchucks was directly related to the presence of WHV DNA in tissues. Two woodchucks with WHV cccDNA in liver, and with WHV DNA in serum and PBMCs, had strong responses to WHcAg, WHsAg, and to the entire panel of WHc and WHs peptides based on high SIs. In the 3 woodchucks with WHV DNA detectable only in liver, the responses were generally weaker and occurred less frequently. One woodchuck with WHV DNA in liver and in PBMCs had responses to WHcAg, and to 2 WHc and WHs peptides that were comparable in magnitude and frequency to those of the woodchucks with WHV DNA detectable only in liver. Two woodchucks with WHV DNA detectable only in PBMCs had borderline responses to WHcAg and to WHc and WHs peptides that were comparable to the responses of control woodchucks, all of which were below the cutoff value (SI ≥3.1).23 The 2 woodchucks that lacked detectable WHV DNA in serum or tissues had no WHV antigen-specific recall responses.

The expression of mRNA for leukocyte CD markers and Th1-type cytokines in WHV antigen-stimulated PBMCs correlated with the proliferation responses (Tables 2 and 3). In 5 of 9 woodchucks tested, elevations of CD3 and CD4 mRNA were detected (≥2.1-fold) compared with uninfected controls. In 4 of the 5 woodchucks with elevations in CD3 and CD4 mRNA, elevations also were detected in CD8 mRNA. All woodchucks with elevated CD mRNA had increased mRNA expression for IL-2, interferon gamma, and tumor necrosis factor alpha. Thus, increased WHV antigen-specific PBMC responses and expression of mRNAs for key leukocyte markers and Th1-type cytokines were most evident in resolved woodchucks that harbored residual WHV DNA, indicating that WHV-specific T lymphocytes persisted years after resolution of infection. This suggests that active WHV-specific CMI is maintained long-term by continued exposure to residual virus.

CsA Immunosuppression Reactivates WHV Replication After Long-Term Resolution of Infection.

Eight of the 10 resolved woodchucks were selected for treatment with CsA for 12 weeks. Four of the eight woodchucks had residual intrahepatic WHV cccDNA and WHV antigen-specific CMI recall responses (group 1). Four others lacked intrahepatic WHV cccDNA (group 2), but 2 of the 4 had residual WHV DNA detected in PBMCs, and either had weak or undetectable recall CMI. Initial differences in residual WHV DNA and recall CMI were not a function of age or time after resolution of infection; Group 1 consisted of 3 woodchucks that were 3.5 years old and 1 that was 6.5 years old, and group 2 had 3 woodchucks that were 3.5 years old and 1 that was 4.5 years old.

Administration of CsA resulted in recrudescence of serum WHV DNA in all 4 group 1 woodchucks (Fig. 2) within 2 to 6 weeks after initiation of immunosuppression. Remarkable increases in viremia were maximal after 6 weeks of treatment and declined thereafter, even though CsA administration was continued through week 12. In 3 of the group 1 woodchucks, WHV DNA was detectable at low levels at the end of treatment. Only 2 of the 4 woodchucks of group 2 had evidence of transient viremia, with only very low WHV DNA levels detected in each woodchuck at week 6. Viremia remained unchanged, at or below detectable levels, in the other 2 woodchucks of group 2. Two weeks after cessation of CsA treatment, there was a slight rise in WHV DNA in all woodchucks of group 1 and in one woodchuck of group 2. Thereafter, WHV DNA returned to baseline levels in 2 to 4 weeks and remained there until the end of the study. CsA treatment induced parallel transient elevations in the serum WHsAg of all woodchucks of group 1, and in the 2 woodchucks of group 2 that had detectable peaks in WHV DNA (Fig. 2). None of the woodchucks from either group had a detectable boost in serum anti-WHs (data not shown).

Figure 2.

Changes in serum viremia and WHs antigenemia of woodchucks with long-term resolution of WHV infection before, during, and after CsA immunosuppression. Group 1 woodchucks had intrahepatic WHV cccDNA and WHV antigen-specific recall CMI responses before immunosuppression. Group 2 woodchucks lacked intrahepatic WHV cccDNA. Woodchucks F3746 and F3796 from group 2 had WHV DNA in PBMCs and weak or undetectable recall CMI. Horizontal bar: 12-week period of CsA administration. Black circle, WHV DNA; white circle, WHsAg; ODU, optical density units.

The pronounced viremia observed in woodchucks M3704 and F3776 from group 1 were followed by transient elevations in serum ALT and SDH between weeks 8 and 10 of CsA treatment (up to 7-fold increases compared with pretreatment), suggesting increased hepatitic activity. In a liver biopsy obtained from woodchuck M3704 at week 6 of treatment and before the peak in serum enzyme markers, transient increases in portal and lobular hepatitis were observed. Recall CMI responses were not measured during or after CsA immunosuppression. Such “flare” reactions in woodchucks are frequently associated with augmented WHV-specific CMI.25

Serum WHV Detected During Viral Recrudescence in CsA-Treated Woodchucks Is Infectious.

Serum was obtained at the peak titer of WHV DNA from woodchuck F3776 of group 1, and 3.4 × 107 WHVge was inoculated intravenously into 2 WHV-susceptible woodchucks. Productive WHV infection was demonstrated in both woodchucks (Fig. 3). One of the 2 (F6432) developed viremia and WHs antigenemia, beginning at weeks 5 to 6 after inoculation, that persisted until week 21. Both woodchucks developed significant anti-WHc levels beginning at weeks 7 or 8. Both woodchucks developed acute liver injury detected biochemically by transient elevations in serum ALT and SDH activity (up to 17-fold increases compared with preinoculation) and also developed portal and lobular hepatitis in liver biopsy specimens.

Figure 3.

Changes in serum viremia and WHs antigenemia and development of anti-WHc and anti-WHs in woodchucks inoculated with serum from woodchuck F3776 that was obtained at week 6 of CsA immunosuppression. Arrows: Serum inoculation with 3.4 × 107 WHVge. Black circle, WHV DNA; white circle, WHsAg; black triangle, anti-WHc; white triangle, anti-WHs; ODU, optical density units.


Results of this study confirm the key role of CMI in controlling hepadnaviral replication after long-term resolution of acute infection. Resolved woodchucks with waning or undetectable anti-WHs had durable WHV antigen-specific CMI responses that were associated with the presence of residual WHV DNA in liver, serum, and/or in PBMCs. CMI responses of resolved woodchucks with less residual WHV DNA in these compartments were diminished and may represent individuals with a different, perhaps more complete, state of resolution based on current assays. Upon CsA immunosuppression, transient reactivation of WHV replication was most robust in woodchucks with residual WHV DNA detectable in liver and in which stronger CMI recall responses were observed. In contrast, CsA did not induce significant viral reactivation in woodchucks that lacked both detectable WHV DNA and WHV antigen-specific recall CMI. The infectivity of virus generated during CsA-mediated viral reactivation was demonstrated by inoculation of susceptible woodchucks that developed productive WHV infection and subsequent acute hepatitis. Our results indicate that the degree of viral eradication is variable in woodchucks with serologically resolved WHV infection. Cellular immune mechanisms, however, are generally adequate and durable for the control of hepadnaviral replication long-term.

Resolution of acute HBV infection is characterized by reductions of viral DNA and antigens in the liver and peripheral blood below levels of detection by current methods, and by seroconversion to virus-neutralizing antibodies to HBV surface antigen.2 Analogous markers and patterns of resolution of viral infection are observed in the woodchucks.15, 23, 24 Residual HBV DNA is evident in patients even decades after resolution of HBV infection.5, 7–9, 27 Similar to humans, woodchucks often have residual WHV DNA in liver, serum, and PBMCs for years after resolution, even in the presence of detectable anti-WHs.18, 19 In some woodchucks of this study, viral cccDNA was detected in the liver (Fig. 1; Table 1), an observation similar to that reported by others in resolved woodchucks.18, 28 The detection of intrahepatic HBV cccDNA in patients with resolved infection has been reported,8, 9 but less is known concerning the potential for reactivation of HBV in humans in this setting.

The maintenance of anti-WHc in resolved woodchucks throughout life22 suggests sustained stimulation of the immune system by nucleocapsid protein. Constitutive antibodies, however, may not reflect trace replication of virus, because viral antigens may persist in dendritic cells for prolonged periods.29 In contrast, anti-WHs, present initially in all of the resolved woodchucks, was undetectable at the time of immunosuppression.

However, CMI recall responses to WHcAg and WHsAg were detected in approximately half of the woodchucks, especially in those harboring residual WHV DNA (Tables 2 and 3). This suggests that sustained WHV-specific CMI responses were maintained by continued antigenic stimulation by virus or viral antigen incompletely cleared during resolution. SI values for WHV-specific PBMC responses were comparable to those observed sporadically in chronic WHV carriers inoculated with WHV as neonates but lower than those observed either in neonatal or adult woodchucks with resolving infections.23–25 Two woodchucks of this study did clear WHV to levels that were below the limits of detection by the methods used, and both lacked WHV-specific CMI. In these woodchucks, immunosuppression was not associated with recrudescence of WHV replication. This particular profile seems more the exception than the norm in neonatal (Fig. 2) and adult WHV infections19 but suggests that in some neonatal infections, full resolution may be possible. Further experimental evidence for the critical role of CMI in resolution derives from studies in which CsA immunosuppression during the acute phase of WHV infection in adult woodchucks was shown to significantly increase the frequency of chronicity as an outcome of infection.30

Resolved woodchucks had in vitro CMI recall responses to WHV antigens that were most likely primed in vivo. The strength of CMI, and the increases in mRNA expression of key leukocyte markers and cytokines, were related directly to the higher levels of residual WHV DNA and to the greater degree of WHV reactivation induced by immunosuppression. The detection of WHV antigen-specific cells expressing increased CD4 and CD8 markers and Th1-type cytokines 3 to 6 years after resolution may reflect recent activation of T lymphocytes by residual virus. Similarly, HBV-specific Th cell and cytotoxic T lymphocyte responses have been observed in patients long after resolution of HBV infection has occurred.6, 7 WHV DNA present in liver, serum, and PBMCs appeared to represent the greatest immunogenic stimuli for the maintenance of WHV-specific CMI. In patients with resolved HBV infection, the stronger cytotoxic T lymphocyte response was observed when HBV DNA was present in both serum and in PBMCs.7 In resolved woodchucks, the presence of residual WHV cccDNA in liver suggests the possibility of low levels of ongoing viral RNA transcription and protein translation.18, 19

This study in woodchucks supports observations that HBV can persist for many years after apparent serological resolution of infection. In some woodchucks, however, WHV may be eradicated or functionally inactivated because, within the limits of detection of current methods, immunosuppression did not always result in measurable reactivation of WHV replication. Viral reactivation in other woodchucks during immunosuppression suggests WHV infection is held in check mainly by virus-specific CMI.

The infectivity of the serum WHV DNA from one of the immunosuppressed woodchucks demonstrated that the residual WHV remained replication-competent, and capable of causing acute hepatitis even 3 years after resolution of infection (Fig. 3). Persistence of replication-competent and pathogenic WHV in resolved woodchucks probably was not attributable to escape from recognition by host immune responses, because such responses appeared durable and stable until the immunosuppression was initiated. Alternatively, WHV may become sequestered in immunologically privileged compartments, a mechanism that may allow the virus to escape complete elimination even in the presence of humoral and cellular immune responses capable of terminating the hepatic infection.

The results of this study in woodchucks have important implications related to HBV. Patients with resolved HBV infection may have persistent viral DNA despite complete clinical, serological, and biochemical resolution of infection, and these individuals (or their organs) may be infectious to others. Caution during immunosuppression of such patients is essential because of the possibility of viral reactivation and the development of liver injury.


We thank Mary Ascenzi, Betty Baldwin, Lou Ann Graham, and Dr. Chris Bellezza of Cornell University for expert assistance.