The technique of neuronal transplantation provides a powerful tool to deliver many pharmacologically active agents into the brain to address some fundamental questions regarding the basic principles of brain function. For example, grafting of embryonic tissues or cells into the central nervous system (CNS) has been used to study neuronal development (McConnell, 1985; Sotelo and Alvarado-Mallart, 1986; O’Leary and Stanfield, 1989), the factors necessary for the survival of axotomized neurons, and axon elongation from damaged neuronal populations (Blakemore and Franklin, 1991; Dunnett, 1991; Lindvall, 1991). In addition, foreign tissues and cells have been used for implantation in the CNS to replace the lost functions of neuronal populations in the damaged or diseased brain. An alternative approach to this application has focused on the genetic modifications of cells in vitro to produce specific products (e.g., growth factors) before their implantation into the brain (Gage et al., 1987,1991). In this strategy, defined populations of cells can be genetically modified to express the gene of interest, and the implantation of these cells can yield a graft consisting of phenotypically homogeneous cells. The success of this approach depends in large part on achieving efficient gene transfer and stable long-term gene expression in recipient cells. A number of efficient methods and vector systems have been described for gene transfer in immortalized and primary cells including postmitotic neurons. Although these methods have been used primarily to express reporter genes like β-galactosidase (β-gal), they are general methods that can be used to express other genes, such as growth factors. Nerve growth factor (NGF) is the only growth factor that has been expressed in cells and then implanted into the brain (Rosenberg et al., 1988; Stromberg et al., 1990; Kawaja and Gage, 1992a). However, the discovery of other growth factors and establishment of their roles and functions will facilitate their use in gene transfer.