Fig. 5: The MCU governs NAD(P)H dynamics during high-frequency action potential firing in pyramidal neurons.

a Left & Middle, Relative to control, blocking mitochondrial complex I with rotenone (10 µM) occludes the NAD(P)H dip and delayed overshoot evoked by a 50 Hz action potential train. Inset, Bath application of rotenone produces a significant increase in NAD(P)H fluorescence relative to baseline in an unstimulated neuron (n = 6, N = 5; paired t test, p = 0.036). Right, Dialysis of Ru360 (20 µM), to block MCU-dependent mitochondrial Ca²⁺ uptake, partially attenuates the stimulus-evoked NAD(P)H dip and largely prevents the delayed overshoot phase. Traces from each condition represent the mean and SEM of NAD(P)H measured from multiple pyramidal neurons. b Left, Rotenone, but not Ru360, significantly reduces the peak train-evoked NAD(P)H dip (control: n = 12, N = 6; rotenone: n = 6, N = 3; Ru360: n = 10, N = 5; One-way ANOVA and Dunnett’s multiple comparisons test: rotenone, p = 0.01; Ru360, p = 0.22). Middle, The NAD(P)H dip magnitude at 5 sec post-stimulus is significantly attenuated in Ru360 relative to control (control: n = 12, N = 6; rotenone: n = 6, N = 3; Ru360: n = 10, N = 5; One-way ANOVA and Dunnett’s multiple comparisons test: rotenone, p = 0.073; Ru360, p = 0.0059). Right, The evoked NAD(P)H overshoot, measured at 60-sec post-train, is significantly reduced by rotenone or Ru360 (control: n = 12, N = 6; rotenone: n = 6, N = 3; Ru360: n = 10, N = 5; One-way ANOVA and Dunnett’s multiple comparisons test: rotenone, p = 0.0086; Ru360, p = 0.0037). Summary data presented as the mean ± SEM. *P < 0.05, **P < 0.01. The number of cell replicates is shown in the graphs as well as the ‘n’ value in the text. The number of animal replicates is represented by the ‘N’ value.