Allogeneic stem cell-based transplants may be limited by allograft rejection, as is seen with conventional organ transplantation. One way to avert such a response is to use autologous stem cells, but that may carry the risk of recurrence of the original disease, particularly in the context of a genetic defect. We investigated the potential for gene modification of autologous stem cells to avoid both problems, using recombinant adenoassociated virus vector expressing human α1-antitrypsin in murine liver progenitor cells. We showed that recombinant adenoassociated virus 1 was the most efficient vector for liver progenitor cell transduction among five different serotypes of recombinant adenoassociated virus vectors. Ex vivo infected green fluorescent protein-positive liver progenitor cells from C57BL/6 mice with recombinant adenoassociated virus 1-vector-expressing human α1 antitrypsin were transplanted into the liver of monocrotaline-treated and partial-hepatectomized C57BL/6 recipients. Using green fluorescent protein as a donor marker, we were able to determine that at 18 weeks after transplantation, approximately 40% to 50% of the regenerated liver was green fluorescent protein positive. In addition, transgene expression (serum human α1-antitrypsin) was sustained for the length of the study (18 weeks after transplantation). Immunostaining revealed approximately 5% to 10% of repopulating liver cells expressing human α1-antitrypsin. In conclusion, this study demonstrated the feasibility of long-term engraftment and stability of transgene expression form genetically modified liver progenitor cells with a recombinant adenoassociated virus vector and implies a novel approach to gene therapy for treatment of liver diseases, such as α1-antitrypsin deficiency. (Hepatology 2004;40:918–924).