Hepatitis A virus (HAV), a member of the Picornaviridae family, causes acute hepatitis in humans (for a review, see reference 14). The mature viral particle consists of a coat of 60 copies of at least three viral proteins (VP1, VP2, and VP3), which encapsidates a positive-strand RNA genome of approximately 7,500 nucleotides. It is not clear whether the fourth capsid protein, VP4, which is found in all other picornaviruses, is also present in mature HAV particles. In general, wild-type HAV grows inefficiently in vitro but accumulates attenuating mutations during cell culture adaptation (for a review, see reference 14). HAV has been adapted to grow in cells of primate (1, 5-7, 10, 12, 28) and nonprimate (9) origin. Therefore, the cellular factors required for HAV growth are not restricted to primate cells.The pathogenesis of HAV is poorly understood, and experimentation is difficult because primates are the only animal model for this virus (8,17,21). Development of a small-animal model to study the pathogenesis of HAV is highly desirable, especially in light of recent findings showing an inverse association between HAV infection and the development of asthma (22,23). It is likely that this inverse association is mediated by the human HAV cellular receptor 1 (hhavcr-1) (11), an ortholog of Tim1 and Tim2, which have been described as asthma determinant genes in mice (24, 37). Moreover, it has recently been shown that infection with HAV may protect individuals from atopy if they carry a variant of hhavcr-1 (25).HAV does not grow efficiently in mouse cell lines (9, 35), and our initial attempts to infect mice with HAV were also unsuccessful (J. Lu, D. Feigelstock, and G. Kaplan, unpublished data). Adaptation of this virus to mouse cells may be required to develop a mouse model for HAV. We have recently shown that mouse Ltk-cells transfected with the infectious cDNA of HAV can support intracellular virus replication (19). However, the HAV grown in Ltk-cells was not capable of reinfecting naïve mouse cells. Since it is possible that liverspecific factors are required for cell entry of HAV, we attempted to adapt HAV to grow in mouse liver cells. Few immortalized mouse hepatocyte cell lines have been generated to date, and most present a transformed phenotype and loss of liver cell characteristics (3, 4). Recently, Amicone et al. (2) generated nontransformed immortalized cell lines from livers of transgenic mice carrying a truncated cytoplasmic form of the human MET gene. One of these cell lines, termed MMH-D3, was derived from a liver of a 3-day-old MET transgenic mouse and has a homogeneous epithelial morphology (34). Here, we report that HAV grew in MMH-D3 cells transfected with virion RNA but not in those infected with viral particles. Interestingly, the MMH-D3 cells gained susceptibility to HAV infection after withdrawal of epidermal growth factor (EGF) from the culture medium. Serial passages of HAV in MMH-D3 cells grown under suboptimal conditions, i.e., in the absence of growth factors and in uncoated plates, res...