In the present paper, we examined the effect of the adenoviral vector dosage, the role of T cells, and the influence of the presence of replication-competent adenovirus (RCA) in adenoviral vector stocks, on the efficacy of adenoviral vector-directed transgene expression in the facial nucleus of immunocompetent Wistar and athymic nude rats. A small number of motor neurons and glial cells was transduced at low dosages of viral vector (1 x 10(6) pfu) and in the absence of RCA, and transgene-expressing cells persisted throughout the 3-week period of observation. Intraparenchymal infusion of 2 x 10(7) pfu of a recombinant adenoviral vector free of RCA was required for optimal transduction of facial motor neurons. In Wistar rats, a biphasic immune response occurred at higher dosages of the vector (5 x 10(6) and 2 x 10(7) pfu) that was characterized by early infiltration of macrophages and the occurrence of T cells during the second week after injection of the vector. The immune response was associated with the loss of transduced neural cells. In nude rats, administration of an adenoviral vector free of RCA resulted in a macrophage response comparable to that in the Wistar rat and long-term survival of transduced astroglial cells. However, transduced motor neurons degenerated according to a similar time course as observed in Wistar rats. Small amounts of RCA (2 x 10(5) pfu) injected with 2 x 10(7) pfu recombinant viral vector particles resulted in an accelerated T cell response and a rapid elimination of transduced cells within 1 week in Wistar rats, whereas in nude rats transgene expression continued during this period. Taken together, these observations suggest that at the high viral vector loads necessary to achieve optimal transduction of the facial nucleus, T cells play a role in the degeneration of adenoviral vector-transduced astroglial cells. The adverse effects on neurons appear to be due to the observed inflammatory response or to direct adenoviral vector toxicity.