The optimal amount of endurance exercise required to elevate proteins involved in neuroplasticity during stroke rehabilitation is not known. This study compared the effects of varying intensities and durations of endurance exercise using both motorized and voluntary running wheels after endothelin-I-induced focal ischemia in rats. Hippocampal levels of brain-derived neurotrophic factor, insulin-like growth factor I and synapsin-I were elevated in the ischemic hemisphere even in sedentary animals suggesting an intrinsic restorative response 2 weeks after ischemia. In the sensorimotor cortex and the hippocampus of the intact hemisphere, one episode of moderate walking exercise, but not more intense running, resulted in the greatest increases in levels of brain-derived neurotrophic factor and synapsin-I. Exercise did not increase brain-derived neurotrophic factor, insulin-like growth factor I or synapsin-I in the ischemic hemisphere. In voluntary running animals, both brain and serum insulin-like growth factor I appeared to be intensity dependent and were associated with decreasing serum levels of insulin-like growth factor I and increasing hippocampal levels of insulin-like growth factor I in the ischemic hemisphere. This supports the notion that exercise facilitates the movement of insulin-like growth factor I across the blood–brain barrier. Serum corticosterone levels were elevated by all exercise regimens and were highest in rats exposed to motorized running of greater speed or duration. The elevation of corticosterone did not seem to alter the expression of the proteins measured, however, graduated exercise protocols may be indicated early after stroke. These findings suggest that relatively modest exercise intervention can increase proteins involved in synaptic plasticity in areas of the brain that likely subserve motor relearning after stroke.