Oxidant stress evoked by pacemaking in dopaminergic neurons is attenuated by DJ-1

Parkinson’s disease is a pervasive, ageing-relatedneurodegenerative disease the cardinal motor symptoms of which reflect the loss of a small groupof neurons, the dopaminergic neurons in the substantia nigra pars compacta (SNc). Mitochondrial oxidant stress is widely viewed as being responsible for this loss, but why these particular neurons should be stressed is a mystery. Here we show, using transgenic mice that expressed a redox-sensitive variant of green fluorescent protein targeted to the mitochondrial matrix, that the engagement of plasma membrane L-type calcium channels during normal autonomous pacemaking created an oxidant stress that was specific to vulnerable SNc dopaminergic neurons. The oxidant stress engaged defences that induced transient, mild mitochondrial depolarization or uncoupling. The mild uncoupling was not affected bydeletionof cyclophilinD,which is a componentof thepermeability transition pore, but was attenuated by genipin and purine nucleotides,whichare antagonists of cloneduncouplingproteins.Knocking out DJ-1 (also known as PARK7 in humans and Park7 in mice), which is a gene associated with an early-onset form of Parkinson’s disease, downregulated the expression of two uncoupling proteins (UCP4 (SLC25A27) andUCP5 (SLC25A14)), compromised calciuminduced uncoupling and increased oxidation of matrix proteins specifically in SNc dopaminergic neurons. Because drugs approved forhumanuse canantagonize calciumentry throughL-type channels, these results point to a novel neuroprotective strategy for both idiopathic and familial forms of Parkinson’s disease. Calciumentry throughL-type channels inSNcdopaminergicneurons occurs throughout thepacemaking cycle, contrasting themwithneighbouring dopaminergic neurons in the ventral tegmental area (VTA), which are much less affected in Parkinson’s disease(Fig. 1a, b). Although prominent, this influx is not necessary for pacemaking, because treatment with the dihydropyridine L-type channel antagonist isradipine eliminates cytosolic calcium oscillations but leaves pacemaking intact (Fig. 1a). Calcium entry during pacemaking comes at a metabolic cost, as it must be extruded by ATP-dependent processes. This demand is met primarily by mitochondria through oxidative phosphorylation. Superoxide and reactive oxygen species are by-products of oxidative phosphorylation, raising the possibility that calcium entry creates mitochondrial oxidant stress. To determine whether this was the case, we generated transgenic mice expressing a redox-sensitive variant of green fluorescent protein (roGFP) with a mitochondrial-matrixtargeting sequence (mito-roGFP). To limit expression tomonoaminergic neurons,we expressedmito-roGFPunder the control of the tyrosine hydroxylase promoter (TH-mito-roGFP; Fig. 1c). Dopaminergic neurons in the SNc and the adjacent VTA from these mice robustly expressed mito-roGFP that co-localized with mitochondrial markers (Fig. 1c–e and Supplementary Fig. 1), providing a reversible, quantitative means of monitoring the oxidation of mitochondrial matrix proteins. Because the expression of mito-roGFP was restricted to a small set of neurons, itwas possible tomonitor themitochondrial redox state in individual neurons deep in brain slices from young adult mice using two-photon laser scanning microscopy (2PLSM). In VTA dopaminergic neurons, the basal oxidation of mito-roGFP was very low (Fig. 1f). In contrast, the oxidation of mito-roGFP was significantly higher in neighbouring SNc dopaminergic neurons (Fig. 1g, h). In juvenile SNc dopaminergic neurons, where pacemaking is similar to that of VTA neurons, mitochondrial oxidant stress also was low (Fig. 1h). To verify that this stress was not an artefact of brain slicing, we generated a transgenic mouse that expressed mito-roGFP under the control of the cytomegalovirus promoter, yielding robust neuronal expression in the cerebral cortex, striatumand hippocampus. In brain slices from these mice, principal neurons in each of these regions were devoid of any significant mitochondrial oxidation (Supplementary Fig. 2), demonstrating that slicing per se did not create oxidant stress. What did contribute significantly to the mitochondrial oxidant stress was calcium influx through plasma membrane L-type channels. Antagonizing L-type channels drastically lowered the extent of mitoroGFP oxidation (Fig. 1g, h), as did slowing pacemaking by cooling (Fig. 1i). L-type channel antagonists had no effect on the oxidation of matrix proteins in neighbouringVTAdopaminergic neurons (Fig. 1h). Blocking calcium entry into mitochondria from the cytoplasm with Ru360 diminished roGFP oxidation (without affecting pacemaking) (Fig. 1h), suggesting that it helped to drive oxidative phosphorylation. Loss-of-function mutations in DJ-1 are linked to an autosomal recessive, early-onset form of Parkinson’s disease. Although DJ-1 is not an antioxidant enzyme itself, it is redox sensitive and participates in signalling cascades made active by mitochondrial superoxide generation. To examine its role in SNc dopaminergic neurons,DJ-1 knockout mice were crossed with the TH-mito-roGFP mice. SNc dopaminergic neurons from these mice had normal pacemaking and oscillations in intracellular calcium concentration (Fig. 2a). However, basal mitoroGFP oxidation was nearly complete at physiological temperatures in these neurons, sowe re-examined cells at a lower temperature. These studies confirmed the robust difference in oxidation betweenwild-type and DJ-1 knockout neurons seen at higher temperatures (Fig. 2b, c). This differencewas virtually abolished by antagonismof L-type calcium channels (Fig. 2b, c). In contrast, the mitochondria in neighbouring VTAdopaminergic neurons were unaffected byDJ-1 deletion (Fig. 2d). A clue about the role of DJ-1 in attenuating mitochondrial oxidant stress came frommeasurements of the inner mitochondrial membrane (IMM) potential with the cationic dye tetramethylrhodaminemethylester (TMRM) (Fig. 3a and Supplementary Movie 1). In VTA dopaminergic neurons, TMRM fluorescence was robust and stable for long periods (Fig. 3b). In contrast, mitochondrial TMRM fluorescence in neighbouring SNc dopaminergic neurons repeatedly fell and then rose back to peak values, indicating that mitochondria were transiently depolarizing (Fig. 3b and Supplementary Movie 1). This ‘flickering’ was stable for long periods (.60min) and was peculiar to SNc dopaminergic neurons, suggesting that itwas not a product of thepreparation