While the chimpanzee remains the only animal that closely models human hepatitis C virus (HCV) infection, transgenic and immunodeficient mice in which human liver can be engrafted serve as a partial solution to the need for a small animal model for HCV infection. The established system that was based on mice carrying a transgene for urokinase-type plasminogen activator (uPA) gene under the control of the human albumin promoter has proved to be useful for studies of virus infectivity and for testing antiviral drug agents. However, the current Alb-uPA transgenic model with a humanized liver has practical limitations due to the inability to maintain non-engrafted mice as dizygotes for the transgene, poor engraftment of hemizygotes, high neonatal and experimental death rates of dizygous mice and a very short time window for hepatocyte engraftment. To improve the model, we crossed transgenic mice carrying the uPA gene driven by the major urinary protein promoter onto a SCID/Beige background (MUP-uPA SCID/Bg). These transgenic mice are healthy relative to Alb-uPA mice and provide a long window from about age 4 to 12 months for engraftment with human hepatocytes and infection with hepatitis C or hepatitis B (HBV) viruses. We have demonstrated engraftment of human hepatocytes by immunohistochemistry staining for human albumin (30-80% engraftment) and observed a correlation between the number of human hepatocytes inoculated and the level of the concentration of human albumin in the serum. We have shown that these mice support the replication of both HBV and all six major HCV genotypes. Using HBV and HCV inocula that had been previously tittered in chimpanzees, we showed that the mice had approximately the same sensitivity for infection as chimpanzees. These mice should be useful for isolating non-cell culture adapted viruses as well as testing of antiviral drugs, antibody neutralization studies and examination of phenotypic changes in viral mutants.