Graphical Abstract Highlights d Mitochondrial ATP synthesis is reduced in Alzheimer's disease cell models d Lower mitochondrial Ca 2+ signal and pyruvate uptake impair cell bioenergetics d GSK-3b reduces HK1-mitochondria association, destabilizing MPC complexes d The defective mitochondrial pyruvate flux alters neuronal function SUMMARYMitochondria are key organelles for brain health. Mitochondrial alterations have been reported in several neurodegenerative disorders, including Alzheimer's disease (AD), and the comprehension of the underlying mechanisms appears crucial to understand their relationship with the pathology. Using multiple genetic, pharmacological, imaging, and biochemical approaches, we demonstrate that, in different familial AD cell models, mitochondrial ATP synthesis is affected. The defect depends on reduced mitochondrial pyruvate oxidation, due to both lower Ca 2+ -mediated stimulation of the Krebs cycle and dampened mitochondrial pyruvate uptake. Importantly, this latter event is linked to glycogen-synthase-kinase-3b (GSK-3b) hyper-activation, leading, in turn, to impaired recruitment of hexokinase 1 (HK1) to mitochondria, destabilization of mitochondrial-pyruvate-carrier (MPC) complexes, and decreased MPC2 protein levels. Remarkably, pharmacological GSK-3b inhibition in AD cells rescues MPC2 expression and improves mitochondrial ATP synthesis and respiration. The defective mitochondrial bioenergetics influences glutamate-induced neuronal excitotoxicity, thus representing a possible target for future therapeutic interventions. (A) Total cellular ATP levels (see STAR Methods) in control (pcDNA3) and PS2-T122R-expressing SH-SY5Y cells, grown either in glucose (Glu)-or galactose (Gal)containing medium. n = 42-90 wells from 4 independent experiments. (B) Representative images (YFP and CFP channels, left) and mean FRET % values (proportional to [ATP]) of nuclear (Nuc) and mitochondrial (Mit) ATeam1.03 probes in control, ER-b11-, PS2 WT-, PS2-T122R-, or PS2-N141I-expressing SH-SY5Y cells, grown in galactose-containing medium. n = 41-65 cells, 15-19 coverslips from 3 independent experiments. Scale bar, 10 mm. (C) Representative images and traces of ATP dynamics in primary cortical neurons of WT and PS2-N141I-tg mice (days in vitro [DIV] 6), transfected with both Nuc and Mit ATeam1.03. Where indicated, gramicidin (0.3 mM) was added. On the right, bars represent the mean decrease rate of Nuc-and Mit-ATP measured for 3 min, after 15 min from gramicidin addition (dotted boxes on traces). n = 30 cells, 24-26 coverslips from 3 independent experiments. Scale bar, 10 mm. (D) Left: representative traces of Ca 2+ dynamics measured with Fura-2 in primary hippocampal neurons of WT and PS2-N141I-tg mice (DIV 10-12), exposed to KCl. 20 min after KCl exposure, [Ca 2+ ] cyt recovery was evaluated, grouping cells for their 340/380 ratio (right; see STAR Methods). n = 3 independent experiments. (E) Left: representative traces of TMRM fluorescence measured in primary hippocampal neurons of WT and PS2-N141I-tg mice (DIV ...

No abstract

No abstract