Transduction of primary human hepatocytes with amphotropic and xenotropic retroviral vectors.

Experiments in animal models suggest that it is feasible to consider hepatic gene therapy using a strategy in which hepatocytes would be isolated by partial hepatectomy, transduced with recombinant retroviral vectors containing genes of therapeutic importance, and then transplanted back into the patient by autologous hepatocellular transplantation. The application of this strategy in clinical trials will require adapting these methods to human cells. We describe the transduction of primary human hepatocytes with two forms of retroviral vectors: amphotropic vectors, which have been used previously in clinical trials, and xenotropic vectors, which have a different host range. Human hepatocytes were harvested from organs preserved in Belzer's solution and were cultivated in a serum-free, tyrosine-free, hormonally defined medium. These cells proliferated for 3-5 days in culture, exhibited characteristic hepatocyte morphology, and expressed liver-specific functions, including phenylalanine hydroxylase, alpha 1-antitrypsin, and glutamine synthase. Transduction with an amphotropic LNL6 retroviral vector resulted in stable incorporation of the provirus into 1% of the cells as estimated by semiquantitative PCR. Consistently higher transduction efficiencies (as much as 10% of the cells) were observed with a xenotropic N2 vector. These data support the feasibility of using LNL6 as a marker gene in clinical trials of hepatocellular transplantation. These data also suggest that the efficiency of transducing hepatocytes with amphotropic vectors in animal models may not accurately reflect the utility of these vectors for human applications. Consideration should be given to the use of xenotropic vectors for optimizing the efficiency of transduction for human applications.