We investigated whether anti–transforming growth factor β (TGF-β) molecular intervention can halt the progression of liver fibrosis in rats. To block TGF-β action in a specific manner, we prepared an adenovirus expressing a truncated type II TGF-β receptor (AdTβ-TR), which specifically inhibits TGF-β signaling as a dominant-negative receptor. We also used an adenovirus expressing bacterial β-galactosidase (AdLacZ) as a control adenovirus. Rats were treated with dimethylnitrosamine (DMN) for 3 weeks; then, AdTβ-TR, AdLacZ, or saline was intravenously applied once, followed by an additional 3-week DMN treatment. The ratio between the truncated receptor and the wild-type receptor at the mRNA level was 15 at 1 week and 10 at 3 weeks after gene transfer. Immunohistostaining analysis showed that the truncated receptor was expressed mainly in septal cells including hepatic stellate cells. Liver fibrosis, as assessed by histology, hydroxyproline content, and the serum level of hyaluronic acid, progressed during the additional 3-week DMN treatment. However, in rats infected with AdTβ-TR, the fibrosis remained at the level seen in rats given DMN for only 3 weeks. All AdTβ-TR–treated rats remained alive, whereas DMN-treated rats infused with either AdLacZ or saline died of liver dysfunction. In the livers of AdTβ-TR–treated rats, electron microscopy showed: 1) less accumulation of extracellular matrix proteins in the Disse's spaces; 2) regenerated hepatocytes; and 3) fat droplet–rich “quiescent” hepatic stellate cells. Our results demonstrate that TGF-β plays a critical role in the progression of liver fibrosis, and suggest that anti–TGF-β intervention should be therapeutic in already-established fibrotic livers, not only by suppressing fibrosis, but by facilitating hepatocyte regeneration.