S and A.A. performed and supervised in vitro experiments in cell/astrocyte cultures and ex vivo analysis of brain tissue; A.B.G, C.I. and P.G.S. performed behavioral experiments and surgical procedures in mice; E.R. and M.G. provided some CB1-KO mice to the group of J.P.B.; D.A and A.P. performed electrophysiological experiments not shown in the manuscript; M.V. and F.J.K performed mouse perfusion and immunohistochemistry experiments; A.C. and L.B. produced some of the viral constructs used (e.g. Syn-mitoCAT); I.B.R, N.P., S.A. and P.G. performed and supervised electron microscopy experiments; M.L.L.R. provided pharmacological tools (HU-Biot); C.J., N.D and L.P provided specific viral constructs to modulate the MCT-2 transporter; C.J. and G.B. provided data and viral vectors regarding mouse retro-orbital injections; B.L and P.V.P. provided important conceptual ideas; A.K.B.S performed in vivo NMR experiments.

Brain mitochondrial complex I (CI) damage is associated with the loss of the dopaminergic neurons of the Substantia Nigra in Parkinson's Disease (PD) patients. However, whether CI inhibition is associated with any alteration of the mitochondrial respiratory chain (MRC) organization in PD patients is unknown. To address this issue, here we analyzed the MRC by blue native gel electrophoresis (BNGE) followed by western blotting, in mitochondria purified from fibroblasts of patients harboring PD-relevant Pink1 mutations. We found a decrease in free CI, and in free versus supercomplexes (SCs)-assembled CI in PD; however, free complex III (CIII) was only modestly affected, whereas its free versus SCs-assembled forms decreased. Interestingly, complex IV (CIV) was considerably lost in the PD samples. These results were largely confirmed in mitochondria isolated from cultured neurons from Pink1 mice, and in cultured neurons and forebrain samples from the PD-related Dj1 mice. Thus, besides CI damage, the MRC undergoes a profound structural remodeling in PD likely responsible for the energetic inefficiency and mitochondrial reactive oxygen species (mROS) over-production observed in this disease.

The assembly of complex I (CI) with complexes III (CIII) and IV (CIV) of the mitochondrial respiratory chain (MRC) to configure I-III- or I-III-IV-containing supercomplexes (SCs) regulates mitochondrial energy efficiency and reactive oxygen species (mROS) production. However, whether the occurrence of SCs impacts on CI specific activity remains unknown to our knowledge. To investigate this issue, here we determined CI activity in primary neurons and astrocytes, cultured under identical antioxidants-free medium, from two mouse strains (C57Bl/6 and CBA) and Wistar rat, i.e. three rodent species with or without the ability to assemble CIV into SCs. We found that CI activity was 6- or 1.8-fold higher in astrocytes than in neurons, respectively, from rat or CBA mouse, which can form I-III-IV SC; however, CI activity was similar in the cells from C57Bl/6 mouse, which does not form I-III-IV SC. Interestingly, CII-III activity, which was comparable in neurons and astrocytes from mice, was about 50% lower in astrocytes when compared with neurons from rat, a difference that was abolished by antioxidants- or serum-containing media. CIV and citrate synthase activities were similar under all conditions studied. Interestingly, in rat astrocytes, CI abundance in I-III-IV SC was negligible when compared with its abundance in I-III-containing SCs. Thus, CIV-containing SCs formation may determine CI specific activity in astrocytes, which is important to understand the mechanism for CI deficiency observed in Parkinson's disease.