Despite the extensive molecular information on serum-derived human hepatitis B viruses (HBV), liver-derived replicative HBV genomes have remained largely uninvestigated. We have examined the sequences of the entire core antigen (nucleocapsid) of liver-derived HBVs in 15 different hepatoma patients. Bona fide mutations, rather than subtype polymorphism, have been identified based on the high-frequency occurrence of structural differences from wild type at the highly evolutionarily conserved positions, instead of at the positions known to contain genetic heterogeneity among different isolates from different geographic locations. The distribution of these naturally occurring mutations of HBV core gene appears to be nonrandom and is found predominantly within three major (I, IV, and V) and four minor domains (II, III, VI, and VII). In general, domain IV mutations correlate with domain V mutations. The replicative HBV DNAs tend to accumulate a higher number of mutated core domains than the integrated HBV DNAs. At the domain level, there is no significant difference in HBV core mutation frequencies between the liver tumors and the adjacent nontumorous livers. Strikingly, domains I, III, and V coincide with three major known T cell epitopes within the core protein in acute and chronic hepatitis B patients. Furthermore, these domains coincide with HLA class II-restricted T cell epitopes, rather than with the conventional HLA class I-restricted epitopes of cytotoxic T lymphocytes. Our results support the hypothesis that HBV core antigen variants can accomplish immunoevasion via accumulated escape mutations. In addition, they also provide a potential molecular explanation for the maintenance of persistent infection of human hepatitis B virus in chronic carriers.

Hispid cotton rats were inoculated intranasally with either measles virus (MV) Edmonston, a multipassaged, tissue culture-adapted strain of MV, or with one of three clinical MV isolates that had limited passages (three to five times) in tissue culture cells. MV Edmonston was recovered from the lungs of every (n = 37) hispid cotton rat inoculated with this virus for at least 7 days after virus inoculation. Peak pulmonary titers occurred on Day +4 (3.3-4.4 log10/g lung). Scattered areas of inflammation were observed interstitially in lung sections from infected animals stained with hematoxylin and eosin, and a similar pattern of diffuse fluorescence was seen in cryostat sections stained with an indirect fluorescent antibody procedure specific for virus antigens. Fluorescent antibody and virus isolation studies on lung lavage cells both suggested that lung leukocytes were a primary target of the virus. In contrast to these findings, virus was isolated only sporadically from hispid cotton rats inoculated with any of the clinical measles virus isolates. Despite the restricted growth of MV in these animals, cotton rats may be useful for studying certain aspects of measles virus pathogenesis and for screening potential antiviral compounds in vivo.

No abstract