We examined the potential benefit of gene therapy in a mouse model of tuberous sclerosis complex (TSC) in which there is embryonic loss of Tsc1 (hamartin) in brain neurons. An adeno-associated virus (AAV) vector (serotype rh8) expressing a tagged form of hamartin was injected into the cerebral ventricles of newborn pups with the genotype Tsc1(cc) (homozygous for a conditional floxed Tsc1 allele) SynI-cre(+), in which Tsc1 is lost selectively in neurons starting at embryonic day 12. Vector-treated Tsc1(cc)SynIcre(+) mice showed a marked improvement in survival from a mean of 22 days in non-injected mice to 52 days in AAV hamartin vector-injected mice, with improved weight gain and motor behavior in the latter. Pathologic studies showed normalization of neuron size and a decrease in markers of mTOR activation in treated as compared to untreated mutant littermates. Hence, we show that gene replacement in the brain is an effective therapeutic approach in this mouse model of TSC1. Our strategy for gene therapy has the advantages that therapy can be achieved from a single application, as compared to repeated treatment with drugs, and that AAV vectors have been found to have minimal to no toxicity in clinical trials for other neurologic conditions. Although there are many additional issues to be addressed, our studies support gene therapy as a useful approach in TSC patients.