To assess the dynamics and confounding variables that influence transgene expression in neural stem cells (NSCs), we generated distinct NSC clones from the same pool of cells, carrying the same reporter gene transcribed from the same promoter, transduced by the same retroviral vector, and transplanted similarly at the same differentiation state, at the same time and location, into the brains of newborn mouse littermates, and monitored in parallel for over a year in vivo (without immunosuppression). Therefore, the sole variables were transgene chromosomal insertion site and copy number. We then adapted and optimized a technique that tests, at the single cell level, persistence of stem cell-mediated transgene expression in vivo based on correlating the presence of the transgene in a given NSC's nucleus (by fluorescence in situ hybridization [FISH]) with the frequency of that transgene's product within the same cell (by combined immunohistochemistry [IHC]). Under the above-stated conditions, insertion site is likely the most contributory variable dictating transgene downregulation in an NSC after 3 months in vivo. We also observed that this obstacle could be effectively and safely counteracted by simple serial infections (as few as three) inserting redundant copies of the transgene into the prospective donor NSC. (The preservation of normal growth control mechanisms and an absence of tumorigenic potential can be readily screened and ensured ex vivo prior to transplantation.) The combined FISH/IHC strategy employed here for monitoring the dynamics of transgene expression at the single cell level in vivo may be used for other types of therapeutic and housekeeping genes in endogenous and exogenous stem cells of many organs and lineages. J. Comp. Neurol. 515:83–92, 2009. © 2009 Wiley-Liss, Inc.