Isolated human hepatocytes have been shown to represent a valuable in vitro model to investigate the metabolism and cytotoxicity of xenobiotics. In addition, human hepatocyte transplantation and artificial liver support systems using isolated human hepatocytes are currently investigated as treatment for acute and chronic hepatic failure. In this regard, human hepatocyte banking by cryopreservation would be of great interest. In the present study, freshly isolated hepatocytes from resected liver biopsies of 28 separate donors (viability: 88 +/- 2%; plating efficiency: 79 +/- 5%) were cryopreserved using two different protocols, stepwise freezing (SF) or progressive freezing (PF), in combination (PF(+), SF(+)) or not (PF(-), SF(-)) with a 30 min preincubation in culture medium at 37 degrees C. Total recovery was higher after PF (38 +/- 3%) than after SF (12 +/- 2%). Preincubation prior to SF had no effect on plating efficiency of thawed hepatocytes (SF(-): 38 +/- 6% versus SF(+): 46 +/- 7%) while preincubation prior to PF increased plating efficiency of thawed hepatocytes (PF(-): 42 +/- 6% versus PF(+): 64 +/- 4%, p < 0.05). In attached cultured human cryopreserved/thawed hepatocytes (CH) from the PF(+) group, albumin production and glutathione content were not significantly different from those of the freshly isolated hepatocyte (FIH) cultures. Cells in CH monolayers appeared smaller than cells in FIH monolayers. In addition, the pattern of cytochrome P450- and UDP-glucuronosyl transferase-dependent isoenzyme activities and GST activity were different, suggesting a variability in the resistance to cryopreservation of the various liver hepatocyte populations. Taken all together, the results of the present study suggest that recovery of human hepatocytes after isolation prior to progressive freezing should allow human hepatocyte banking for use in pharmacotoxicology and cell therapy research purposes.