Cytotoxic T lymphocytes (CTL) are thought to contribute to viral clearance and liver cell injury during hepatitis B virus (HBV) infection. Using a strategy involving the in vitro stimulation of peripheral blood mononuclear cells (PBMC) with HBV-derived synthetic peptides containing HLA-A2.1, -A31, and -Aw68 binding motifs, we have previously described CTL responses to several epitopes within the HBV nucleocapsid and envelope antigens in patients with acute hepatitis. In this study we define six HLA-A2-restricted CTL epitopes located in the highly conserved reverse transcriptase and RNase H domains of the viral polymerase protein, and we show that the CTL response to polymerase is polyclonal, multispecific, and mediated by CD8+ T cells in patients with acute viral hepatitis, but that it is not detectable in patients with chronic HBV infection or uninfected healthy blood donors. Importantly, the peptide-activated CTL recognize target cells that express endogenously synthesized polymerase protein, suggesting that these peptides represent naturally processed viral epitopes. DNA sequence analysis of the viruses in patients who did not respond to peptide stimulation indicated that CTL nonresponsiveness was not due to infection by viral variants that differed in sequences from the synthetic peptides. CTL specific for one of the epitopes were unable to recognize several naturally occurring viral variants, except at high peptide concentration, underlining the HBV subtype specificity of this response. Furthermore, CTL responses against polymerase, core, and envelope epitopes were detectable for more than a year after complete clinical recovery and seroconversion, reflecting either the persistence of trace amounts of virus or the presence of long lived memory CTL in the absence of viral antigen. Finally, we demonstrated that wild type viral DNA and RNA can persist indefinitely, in trace quantities, in the serum and PBMC after complete clinical and serological recovery, despite a concomitant, vigorous, and sustained polyclonal CTL response. Since viral persistence is not due to escape from CTL recognition under these conditions, the data suggest that HBV may retreat into immunologically privileged sites from which it can seed the circulation and reach CTL-inaccessible tissues, thereby maintaining the CTL response in apparently cured individuals and, perhaps, prolonging the liver disease in patients with chronic hepatitis.

Hepatocellular carcinoma (HCC) is a common complication of chronic hepatitis B virus (HBV) infection. The pathogenetic mechanisms potentially responsible for HCC during chronic HBV infection are not well defined. This study demonstrates that chronic immune-mediated liver cell injury triggers the development of HCC in the absence of viral transactivation, insertional mutagenesis, and genotoxic chemicals. These results strongly suggest that the immune response to HBV is both necessary and sufficient to cause liver cancer during chronic HBV infection, and that all other procarcinogenic events associated with HCC are probably dependent on this process.

The HLA class I-restricted cytotoxic T Iymphocyte (CTL) response is a major defense mechanism in viral infections. It has been suggested that the CTL response may contribute to viral clearance and liver cell injury during hepatitis C virus (HCV) infection. To test this hypothesis requires an understanding of the characteristics of HCV-specific cytotoxic effector cells and identification of the target antigens to which they respond. To begin this process we stimulated peripheral blood mononuclear cells (PBMC) from a group of HLA-A2 positive patients with chronic hepatitis C with a panel of 130 HCV-derived peptides containing the HLA-A2 binding motif. Effector cells were tested for their capacity to lyse HLA-A2-matched target cells that were either sensitized with peptide or infected with a vaccinia virus construct containing HCV sequences. Using this approach we have identified nine immunogenic peptides in HCV, three of which are derived from the putative core protein, three from the nonstructural (NS) 3 domain, two from NS4 and one from NS5. Selected responses were shown to be HLA-A2 restricted, mediated by CD8+ T cells and to recognize endogenously synthesized viral antigen. Unexpectedly, peptide-specific CTL responses could also be induced in seronegative individuals, suggesting in vitro activation of naive CTL precursors. The precursor frequency of peptide-specific CTL was 10 to 100-fold higher in infected patients compared to uninfected controls, and the responses were greatly diminished by removal of CD45 RO + (memory) T cells. Further quantitative studies are clearly required to establish whether a correlation exists between the HCVspecific CTL response and the clinical course of this disease. Definition of the molecular targets of the human CTL response to HCV creates this opportunity, and may also contribute to the development of a T cell-based HCV vaccine.(J. Clin. Invest. 1995. 95:521-530.)

StlmmaryThe absence of readily manipulable experimental systems to study the cytotoxic T lymphocyte (CTL) response against hepatitis B virus (HBV) antigens has thus far precluded a definitive demonstration of the role played by this response in the pathogenesis of liver cell injury and viral clearance during HBV infection. To circumvent the problem that HBV infection of human cells in vitro for production of stimulator/target systems for CTL analysis is not feasible, a panel of 22 overlapping synthetic peptides covering the entire amino acid sequence of the HBV core (HBcAg) and e (HBeAg) antigens were used to induce and to analyze the HBV nucleocapsid-specific CTL response in nine patients with acute hepatitis B, six patients with chronic active hepatitis B, and eight normal controls. By using this approach, we have identified an HLA-A2-restricted CTL epitope, located within the NH2-terminal region of the HBV core molecule, which is shared with the e antigen and is readily recognized by peripheral blood mononudear cells from patients with self-limited acute hepatitis B but less efficiently in chronic HBV infection. Our study provides the first direct evidence of HLA class I-restricted T cell cytotoxicity against HBV in humans. Furthermore, the different response in HBV-infected subjects who successfully clear the virus (acute patients) in comparison with patients who do not succeed (chronic patients) suggests a pathogenetic role for this CTL activity in the clearance of HBV infection. H epatitis B virus (HBV) 1 infection is a worldwide health problem with an estimate of about 250 million people carrying the virus who can develop chronic hepatitis, liver cirrhosis, and ultimately hepatocellular carcinoma. As in other infections with noncytopathic viruses, a HLA class I-restricted cytotoxic T cell response to endogenously synthesized HBV proteins is believed to be the major determinant of infected cell clearance (1-4).It is well known that CTLs recognize short amino acid (AA) sequences derived from the intracellular processing of viral antigens in association with HLA class I molecules on the surface of the infected cells (5-10). Since HBV does not efficiently infect human cells in vitro, the use of short synthetic peptides mimicking the processed antigen fragments can be a rational strategy to stimulate the HBV-specific CTL response (11-15). Based on this concept, a panel of short pep-1 Abbreviations used in this paper: AA, amino acid; AVH-B, acute hepatitis type B; HBcAg, Hepatitis B virus core antigen; HBV, hepatitis B virus; SGPT, serum glutamic pyruvic transaminase. tides representing the entire HBV nucleocapsid region were synthesized and used for CTL analysis.Our results demonstrate that HBV nucleocapsid-specific CTL are present in the peripheral blood of patients with acute HBV infection and that this response in the context of HLA-A2 haplotype is focused on a CTL epitope included within the 11-27 sequence of the HBV core antigen (HBcAg). The observation that patients with acute and chronic HBV infection e...

A transgenic mouse strain that expresses the hepatitis B virus (HBV) large envelope protein in the liver was used to determine the extent of oxidative DNA damage that occurs during chronic HBV infection. This mouse strain develops a chronic necroinflammatory liver disease that mimics the inflammation, cellular hyperplasia, and increased risk for cancer that is evident in human chronic active hepatitis. When perfused in situ with nitroblue tetrazolium, an indicator for superoxide formation, the liver of transgenic mice displayed intense formazan deposition in Kupffer cells, indicating oxygen radical production, and S-phase hepatocytes were commonly seen adjacent to the stained Kupffer cells. Similar changes were not observed in nontransgenic control livers. To determine whether these events were associated with oxidative DNA damage, genomic DNA from the livers of transgenic mice and nontransgenic controls was isolated and examined for 8-oxo-2'-deoxyguanosine, an oxidatively modified adduct of deoxyguanosine. Results showed a significant, sustained accumulation in steady-state 8-oxo-2'-deoxyguanosine that started early in life exdusively in the transgenic mice and increased progressively with advancing disease. The most pronounced increase occurred in livers exhibiting microscopic nodular hyperplasia, adenomas, and hepatocellular carcinoma. Thus, HBV transgenic mice with chronic active hepatitis display greatly increased hepatic oxidative DNA damage. Moreover, the DNA damage occurs in the presence of heightened hepatocellular proliferation, increasing the probability of fixation of the attendant genetic and chromosomal abnormalities and the development of hepatocellular carcinoma.Chronic inflammatory diseases affect millions of people world-wide and may be a primary factor in the development of up to one-third of all cancers (1). A major association between a persistent infection and cancer is evident in chronic active hepatitis B virus (HBV) infection (2). Epidemiological studies show that incidence of hepatocellular carcinoma (HCC) correlates strongly to geographical areas of endemic HBV infection (3). Furthermore, patients with chronic hepatitis have a much higher incidence of liver cirrhosis and HCC than individuals not infected with HBV (4). Also, HBV antigens in sera of patients are associated with HCC (5). Because the virus infects >300 million people throughout the world, HBV has been described as second only to tobacco as a known human carcinogen (6).Despite the apparent epidemiological association of chronic active hepatitis infection and cancer, the mechanisms involved are still not entirely understood. Research has been hampered due to the long latency period between the onset of infection and cancer development, and the lack of suitable animal models that mimic chronic active human hepatitis.HBV does not contain any known acutely transforming oncogenes, and only rarely do integrated HBV sequences activate cellular protooncogenes (7-10). In contrast, persistent inflammation associated with chro...

HLA class I-restricted human cytotoxic T cells recognize endogenously synthesized hepatitis B virus nucleocapsid antigen (viral pathogenesis/viral ABSTRACTKnowledge of the immune effector mechanisms responsible for clearance of hepatitis B virus (HBV)-infected cells has been severely limited by the absence of reproducible systems to selectively expand and to characterize HBV-specific cytotoxic T lymphocytes (CTLs) in the peripheral blood of patients with viral hepatitis. By using a strategy involving sequential stimulation with HBV nucleocapsid synthetic peptides followed by autologous, or HLA class I-matched, HBV nucleocapsid transfectants, we now report the existence of CTLs able to lyse target cells that express endogenously synthesized HBV nucleocapsid antigen in the peripheral blood of patients with acute viral hepatitis B. The CTL response is HLA-A2 restricted, mediated by CD8-positive T cells, and specific for a single epitope, located between amino acid residues 11 and 27 of HBV core protein; these residues are shared with the secretable precore-derived hepatitis B e antigen. Equivalent lysis of target cells that express each of these proteins suggests that their intracellular trafficking pathways may intersect. The current report provides definitive evidence that HLA class I-restricted, CD8-positive CTLs that recognize endogenously synthesized HBV nucleocapsid antigen are induced during acute HBV infection in humans and establishes a strategy that should permit a detailed analysis of the role played by HBV-specific CTLs in the immunopathogenesis of viral

Several lines of experimental evidence suggest that inclusion of core sequences in the hepatitis B vaccine may represent a feasible strategy to increase the efficacy of the vaccination. In order to identify immunodominant core epitopes, peripheral blood T cells purified from 23 patients with acute hepatitis B and different HLA haplotypes were tested with a panel of 18 short synthetic peptides (15 to 20 amino acids [AA]) covering the entire core region. All patients except one showed a strong T cell proliferative response to a single immunodominant 20 amino acid sequence located within the aminoterminal half of the core molecule. Two additional important sequences were also identified at the aminoterminal end and within the carboxyterminal half of the core molecule. These sequences were able to induce significant levels of T cell proliferation in 69 and 73% of the patients studied, respectively. T cell response to these epitopes was HLA class II restricted. The observations that (a) polyclonal T cell lines produced by PBMC stimulation with native HBcAg were specifically reactive with the relevant peptides and that (b) polyclonal T cell lines produced with synthetic peptides could be restimulated with native HBcAg, provide evidence that AA sequences contained within the synthetic peptides represent real products of the intracellular processing of the native core molecule. In conclusion, the identification of immunodominant T cell epitopes within the core molecule provides the molecular basis for the design of alternative and hopefully more immunogenic vaccines. (J. Clin. Invest. 1991.

SummaryWe have recently developed the technology to identify and characterize the human histocompatibility leukocyte antigen (HLA) class I-restricted, CD8 + cytotoxic T lymphocyte (CTL) response to hepatitis B virus (HBV)-encoded antigens in patients with acute viral hepatitis. CTL are expanded in vitro by stimulation with HBV-derived synthetic peptides and selected by restimulation with a panel of HLA-matched stable transfectants that express the corresponding HBV protein. We have recently reported the existence of an HLA-A2-restricted, CD8 + CTL response to an epitope located between residues 18 and 27 of the HBV nucleocapsid core antigen (HBcAg). We now report the discovery of a CTL epitope located between HBcAg residues 141 and 151 that completely overlaps a critical domain in the viral nucleocapsid protein that is essential for its nuclear localization and genome packaging functions as well as processing of the precore protein. The CTL response to this epitope is dually restricted by the HLA-A31 and HLA-Aw68 alleles, which, unexpectedly, appear to use a common binding motif based on the results of alanine substitution and competition analysis, and the binding properties of these two alleles predicted from their known primary sequence, and from the three-dimensional structure of HLA-Aw68. We have also demonstrated that the HBV-specific CTL response to this epitope is polyclonal during acute viral hepatitis, since these two restriction elements can present the HBcAg 141-151 epitope to independent CTL clones derived from a single patient; and that the CTL response is multispecific, since HLA-A2-restricted and HLA-Aw68-restricted CTL responses to HBcAg 18-27 and HBcAg 141-151, respectively, have been identified to coexist in another patient. The foregoing argue against the emergence of CTL escape mutants as a significant problem during HBV infection, especially at this locus, where mutations might be incompatible with viral replication. Finally, our data suggest an association between the HBV-specific CTL response and viral clearance, and they have implications for the design of immunotherapeutic strategies to terminate HBV infection in chronically infected patients.T he hepatitis B virus (HBV) 1 is a small hepatotropic enveloped virus with a double-stranded DNA genome (1) that causes acute and chronic necroinflammatory liver disease and hepatocellular carcinoma (2). The mechanisms responsible for viral clearance and liver cell injury in HBV infection are not well understood. Based on precedent in other viral 1 Abbreviations used in this paper: HBcAg, hepatitis B core antigen; HBV, hepatitis B virus. systems, and a limited analysis of the intrahepatic (3-5) and PBL (6) response to HBV-encoded antigens, it appears that HBV-specific helper and CTL responses may play an important pathogenetic role in this disease. Since the viral nucleocapsid protein is an important target of the CTL response to other viruses (7-11), we have begun to examine the HLA class I-restricted CTL response to the hepatitis B nucleocapsid c...