Physical exercise is a non-pharmacological intervention which has beneficial effects on the motor and non-motor symptoms of Parkinson disease (PD). The mechanisms that underlie these benefits are still unclear. Our aim is to determine whether exercise by PD patients can restore basal ganglia synaptic density and brain connectivity. We hypothesize that the 10-20% decreased baseline SV2A binding measured by [11C]UCB-J positron emission tomography (PET) in the nigra and striatum of PD patients will be restored after performance of a controlled a high-intensity aerobic exercise protocol and frontal-basal ganglia connectivity will be strengthened. For that, an in vivo study combining functional magnetic resonance imaging (fMRI)/PET imaging study is being performed in 40 PD patients. Two timepoints are being evaluated: (1) baseline scan prior to exercise protocol; and (2) 24 weeks after first imaging. 20 PD subjects are having high-intensity aerobic exercise training and 20 PD subjects are keeping to their daily life routine. The exercise protocol includes supervised training. Five training sessions per 14 days for 24 weeks (1 x continuous + 1-2 x interval training/week). [11C]UCB-J PET, a marker of synaptic SV2A protein, is being used to evaluate the brain synaptic density/plasticity. We are studying basal ganglia-cortical connectivity using BOLD fMRI. These changes will be correlated with improvements of motor and non-motor symptoms and alterations in synaptic blood markers. This is the first in vivo investigation of synaptic density changes in response to high-intensity aerobic exercise training. This work will help us understand if synaptic changes are induced by long term exercise and help rationalize this approach to combat the disease.