The availability of suitable human livers for transplantation falls short of the number of potential patients. In addition, the availability of primary human hepatocytes for cell-therapy and drug development applications is significantly limited; less than 700 livers per year are available for such studies. However, the majority of these organs cannot be utilized due to pathological infections (e.g., HepB, HepC, or HIV) or excessive levels of steatosis. Thus, the number of cells needed for cell therapy applications far exceeds the number of cells available from donated livers. The ability to implant progenitor cell populations that can form liver tissue in situ, or can be differentiated in vitro would be a major advance in current cell-based therapies. In addition, and importantly for this application, the ability to utilize a non-hepatic progenitor cell to mimic hepatocytes in vitro would enable the scale-up production of cells for bioartifical liver assist devices, cell-therapy and drug discovery applications. We demonstrate the feasibility of inducing adipose-derived stromal (ASC) cells to express several features of human hepatocytes such as glycogen storage and expression of liver specific genes. Importantly, we also show that undifferentiated ASCs and ASC-derived hepatic cells engraft robustly into the liver in a mouse model of toxic injury. These data indicate a significant potential for the use of undifferentiated ASCs and ASC-derived hepatic cells as novel and valuable products for cell therapy. J. Cell. Physiol. 225: 429–436, 2010. © 2010 Wiley-Liss, Inc.