1. We have developed a novel technique which causes primary human hepatocytes to proliferate by transducing them with genes that upregulate their proliferation. 2. Upcyte(®) hepatocytes did not form colonies in soft agar and are not immortalised anchorage-independent cells. Confluent cultures expressed liver-specific proteins, produced urea and stored glycogen. 3. CYP activities were low but similar to that in 5-day cultures of primary human hepatocytes. CYP1A2 and CYP3A4 were inducible; moreover, upcyte(®) hepatocytes predicted the in vivo induction potencies of known CYP3A4 inducers using the "relative induction score" prediction model. Placing cells into 3D culture increased their basal CYP2B6 and CYP3A4 basal activities and induction responses. 4. Phase 2 activities (UGTs, SULTs and GSTs) were comparable to activities in freshly isolated hepatocytes. 5. Upcyte(®) hepatocytes were markedly more sensitive to the hepatotoxin, α-amanitin, than HepG2 cells, indicating functional OATP1B3 uptake. The cytotoxicity of aflatoxin B(1), was decreased in upcyte(®) hepatocytes by co-incubation with the CYP3A4 inhibitor, ketoconazole. Upcyte(®) hepatocytes also differentiated between ten hepatotoxic and eight non-hepatotoxic compounds. 6. In conclusion, upcyte(®) hepatocyte cultures have a differentiated phenotype and exhibit functional phase 1 and 2 activities. These data support the use of upcyte(®) hepatocytes for CYP induction and cytotoxicity screening.