We are testing the hypothesis that exercise is neuroprotective in animal models of the dopamine (DA) deficiency in Parkinson's disease. Our studies include mice or rats provided access to a running wheel and subsequently treated with MPTP (mice) or 6-hydroxydopamine (rats) and monkeys provided access to a treadmill and subsequently treated with MPTP. Typically, the exercise occurs for 3 months prior to the toxin treatment and often for 1-2 months thereafter. Our findings indicate that exercise reduces the behavioral impairments elicited by the dopaminergic neurotoxins as well as the loss of DA neurons as assessed by PET imaging and biochemical or histochemical assessment of tissue samples. Our studies are focused on one of several possible explanations for the beneficial effects of exercise: an exercise-induced increase in the expression of neurotrophic factors, particularly GDNF. Our observations indicate that GDNF can reduce the vulnerability of DA neurons, in part due to the activation of key intracellular cascades. We also find that mild cellular stress itself can provide protection against more intensive stress, a form of preconditioning. We conclude that dopamine neurons have the capacity to respond to intracellular and extracellular signals by triggering endogenous neuroprotective mechanisms. This raises the possibility that some individuals with Parkinson's disease suffer from a reduction of these neuroprotective mechanisms, and that treatments that boost these mechanisms - including exercise - may provide therapeutic benefit.