Convergent amelioration of mitochondrial oxidant stress in substantia nigra dopamine neurons by isradipine and rasagiline

Post hoc analysis of the Phase 2 clinical trial with isradipine suggested that there was a beneficial interaction with rasagiline in terms of UPDRS progression. Specifically, patients taking the combination of rasagiline and isradipine appeared to progress less rapidly than patients taking either drug alone over a 12 month period. We therefore sought to determine the mechanisms that might mediate a cooperative interaction between these two drugs. One potential common target is dopamine (DA). By limiting Ca2+ entry through L-type channels, isradipine slows DA synthesis (Mosharov et al., 2009). As cytosolic DA is widely viewed as toxic (Surmeier and Sulzer, 2013), isradipine might diminish this stress on substantia nigra pars compacta (SNc) DA neurons. In addition, isradipine limits mitochondrial calcium loading, thereby lessening mitochondrial oxidant stress arising from the generation of mitochondrial oxidant species (Surmeier and Schumacker, 2013). How does rasagiline protect? The neuroprotective actions of rasagiline in PD have been widely described (Weinreb et al., 2010) and are presumed to arise from its inhibition of monoamine oxidase B (MAO-B) in astrocytes. Recent work has shown that SNc DA neurons co-express MAO-B and MAO-A (Woodard et al., 2014), raising the possibility that rasagiline may also act directly on these neurons. MAO-B cleaves the Cα-H amine of DA (and other catecholamines). This oxidative reaction yields an imine and is associated with the transfer of a pair of electrons to the FAD group on MAO-B. According to the canonical model, these electrons are offloaded nonspecifically from the FAD group to O2, yielding H2O2 that may cause cell injury. However, MAO-B (and MAO-A) are anchored to the mitochondrial outer membrane by the C-terminus, placing the FAD group in proximity to the intermembrane space and the electron transport chain (ETC) in the inner membrane. We hypothesize that the electrons derived from MAO-B activity are transferred through the outer mitochondrial membrane and delivered to cytochrome c in the ETC. Such a mechanism could allow MAO-B to function without generating potentially toxic ROS in the cytosol. However, if true, then MAO activity should increase mitochondrial oxygen consumption and might elevate mitochondrial matrix oxidant stress by augmenting mitochondrial membrane potential. On the other hand, if the canonical model is correct then MAO-B activity should instead produce an increase in cytosolic ROS without affecting mitochondrial potential or matrix oxidant stress. To provide a rigorous test of these two hypotheses, two photon laser scanning microscopy of ex vivo mouse brain slices was used to monitor cytosolic and mitochondrial oxidant stress in striatal terminals of SNc DA neurons, as well as in somatic and dendritic regions. Cytosolic and mitochondrial oxidant stress were assessed using genetically encoded variants of the redox-sensitive green fluorescent protein (ro-GFP). These experiments revealed that in striatal axon terminals and in dendrites of SNc DA neurons, elevating cytosolic DA levels – either by releasing it from vesicular pools with methamphetamine or by boosting synthesis with levodopa administration ¬– increased mitochondrial but not cytosolic oxidant stress. Neither manipulation elevated oxidant stress in the somatic region where DA concentrations are low. Importantly from a therapeutic standpoint, this DA-dependent mitochondrial oxidant stress was dramatically alleviated by rasagiline. Thus, isradipine and rasagiline lower mitochondrial oxidant stress in SNc