Experiments in transgenic mice show that hepatocytes are the source for postnatal liver growth and do not stream

One hypothesis is that postnatal liver growth involves replication of mature hepatocytes, which have an unlimited proliferative potential. An alternative viewpoint is that only certain periportal cells can replicate extensively and that daughter cells stream slowly from the periportal to the pericentral region of the liver. Transgenic mice expressing the beta‐galactosidase (β‐gal) gene from the human α1 antitrypsin promoter were used to examine the proliferative potential of hepatocytes. Surprisingly, only 10% of hepatocytes in two different transgenic lines stain blue with X‐gal. In neonatal animals, singlets or doublets of expressing cells are randomly scattered throughout the liver. Although the overall frequency of blue cells is similar in older animals, these cells are present in much larger clusters, suggesting that individual expressing cells have replicated to form a clonally derived cluster. Expression patterns are not altered by the administration of an acute phase stimulus or by the performance a partial hepatectomy, suggesting that the expression state cannot be easily altered, and making it more likely that the expression state is indeed fixed. These results suggest that the clusters of blue cells are clonally derived in the transgenic mice. They argue that the parenchymal hepatocyte is responsible for growth in the postnatal liver and that streaming of liver cells does not occur. (H EPATOLOGY 1995; 22:160–168.)