To assess the fetal intestine as a site for gene therapy, we have explored a xenograft model in which fetal rat intestine is grafted subcutaneously into nu/nu mice. Prior to grafting, the tissue was exposed to a replication-deficient retroviral vector bearing the neo gene. Transduction efficiency was assessed by quantitative polymerase chain reaction (PCR) of neo in DNA recovered from the grafts. Three methods of infection were employed: (i) simple flushing of the fetal intestine with the vector; (ii) incubation with the vector for 2 hr; and (iii) a combination of both. The first method gave the highest transduction efficiencies in terms of both the proportion of samples that were neo-positive and the number of neo-positive cells per sample. Using this approach, the time course of persistence of neo-positive cells was analyzed by collecting grafts at 1 versus 3 weeks post-infection. The results showed approximately five-fold more positive cells at the earlier time point than at the later, suggesting loss of transduced cells due to cell turnover. Nevertheless, the persistence of a portion of the positive cells for at least 3 weeks is encouraging for future studies with fetal intestine.