Quiescent hepatic stem cells (HSCs) can be activated when hepatocyte proliferation is compromised. Chemical injury rodent models have been widely used to study the localization, biomarkers, and signaling pathways in HSCs, but these models usually exhibit severe promiscuous toxicity and fail to distinguish damaged and non-damaged cells. Our goal is to establish new animal models to overcome these limitations, thereby providing new insights into HSC biology and application. We generated mutant mice with constitutive or inducible deletion of Damaged DNA Binding protein 1 (DDB1), an E3 ubiquitin ligase, in hepatocytes. We characterized the molecular mechanism underlying the compensatory activation and the properties of oval cells (OCs) by methods of mouse genetics, immuno-staining, cell transplantation and gene expression profiling. We show that deletion of DDB1 abolishes self-renewal capacity of mouse hepatocytes in vivo, leading to compensatory activation and proliferation of DDB1-expressing OCs. Partially restoring proliferation of DDB1-deficient hepatocytes by ablation of p21, a substrate of DDB1 E3 ligase, alleviates OC proliferation. Purified OCs express both hepatocyte and cholangiocyte markers, form colonies in vitro, and differentiate to hepatocytes after transplantation. Importantly, the DDB1 mutant mice exhibit very minor liver damage, compared to a chemical injury model. Microarray analysis reveals several previously unrecognized markers, including Reelin, enriched in oval cells. Here we report a genetic model in which irreversible inhibition of hepatocyte duplication results in HSC-driven liver regeneration. The DDB1 mutant mice can be broadly applied to studies of HSC differentiation, HSC niche and HSCs as origin of liver cancer.