Effects of high K⁺ on mitochondrial dynamics and morphology in neurons. (A) Neurons (10 DIV) were transfected with pDsRed2-Mito to label mitochondria. Neurons (at 11 DIV) were then treated with 45 mM K⁺ for 15 min, and mitochondrial fluorescence was analyzed by time-lapse imaging. Arrows show ringlike mitochondria formation in dendrites. Bar, 10 μm. (B) Higher magnification in dendrites show details of ringlike mitochondrial formation induced by treatment with 45 mM K⁺. Arrows show mitochondrial fission. Time in minutes after application of high K⁺ is indicated at the bottom of each panel. Bar, 1 μm. (C) Ultrastructure of mitochondria analyzed by electron microscopy. Panels i and ii show examples of mitochondria from control neurons. The remaining panels (iii–vii) show examples of mitochondria in neurons treated with 45 mM K⁺ for 15 min. (ii) Mitochondria in dendrites from control neurons were rod shaped, and their cristae were clear. (iii–vii) Mitochondria from neurons treated with 45 mM K⁺ formed clusters that exhibited less electron-dense matrices and contained less cristae. (v–vii) 45-mM K⁺ treatment caused mitochondrial fission. Arrows in v–vii show connections between dividing mitochondria. The dividing mitochondria shared continuous outer membrane with separate inner membranes (arrowhead in vi). Bars: (i and iii)1 μm; (ii and iv–vii) 200 nm.

VDCC-associated Ca²⁺ signaling is involved in the effects of high K⁺ on mitochondria. (A) Neurons (10 DIV) transfected with pDsRed2-Mito were analyzed by time-lapse imaging. Neurons (at 11 DIV) were then preincubated with 10 μM nifedipine (L-type blocker), 1 μM Cono-MVIIC (N and P/Q blocker), 10 μM UO126 (MAPK inhibitor), 20 μM KN93 (CaMK inhibitor), 20 μM KN92 (control for KN93), and 1 μM FK506 and were incubated in 45 mM K⁺ for 15 min. Bars, 10 μm. (B) Mitochondrial length was measured through analysis of the elongation index (calculated by the square of longest length divided by the area). The elongation index after 15 min of high K⁺ stimulation was normalized to the value before stimulation. Mitochondrial movement was measured by time-lapse fluorescence microscopy. The movement events after high K⁺ stimulation were normalized to the value before stimulation. Neurons were either untreated (No sti), incubated with 45 mM K⁺ for 15 min (K⁺), or preincubated with various pharmacological agents before incubation with 45 mM K⁺ as in A and also including 1 μM Cono-GVIA (N-type blocker), 0.1 μM Aga-IVA (P/Q-type blocker), or Cono-GVIA plus Aga-IVA. The results shown are from five experiments and represent the mean ± SEM (n = 5) in each group. Values were analyzed using one-way ANOVA followed by post-Tukey test. *, P < 0.05 versus K⁺ stimulation.

 CaMKIα phosphorylates recombinant Drp1 and endogenous Drp1 in cells. (A) Recombinant Drp1 was incubated with either CaMKIα or CaMKII as indicated in the presence of [³²P]ATP. Proteins were separated by SDS-PAGE and analyzed by autoradiography and Coomassie staining. (B) Wild-type Drp1 or mutant Drp1 in which Ser600 was replaced by alanine was incubated with CaMKIα and [³²P]ATP, and proteins were separated by SDS-PAGE and analyzed by autoradiography. (C) Recombinant Drp1 was incubated with CaMKIα in the absence or presence of ATP. Proteins were separated by SDS-PAGE and analyzed by immunoblotting using a phosphospecific antibody selective for phospho-Ser600. (D) HeLa cells were transfected without (control siRNA) or with siRNA selective for CaMKIα. HeLa cells were incubated without (−) or with 45 mM K⁺ for 15 min (+). Proteins were separated by SDS-PAGE and analyzed by immunoblotting with phospho-Ser600 or Drp1 antibodies. (E) Neurons were incubated for the indicated times with 45 mM K⁺ in the absence or presence of KN93 (20 μM added 30 min before high K⁺). Proteins were separated by SDS-PAGE and analyzed by immunoblotting with phospho-Ser600 or Drp1 antibodies as indicated. (F) Neurons were untreated (control) or treated with 45 mM K⁺ for 15 min. Proteins were separated by SDS-PAGE and analyzed by immunoblotting with phospho-Ser600 or total Drp1 antibodies as indicated (bottom). Bar graphs show cumulative data from three independent experiments for each condition. Error bars indicate SEM in each group. *, P < 0.05.

 CaMKIα-dependent phosphorylation regulates Drp1 intracellular distribution in neurons. (A) Neurons (10 DIV) were transfected with YFP-Drp1 or YFP-Drp1-S600A. Neurons (at 11 DIV) were untreated (left) or treated with (right) 45 mM K⁺ or preincubated with 20 μM KN93 for 30 min before high K⁺ stimulation. Bars, 10 μm. (B) Quantitative analysis of exogenous Drp1 foci. Data were obtained from at least five independent experiments for each condition shown in A. Control is YFP-Drp1 without stimulation. Error bars indicate SD in each group. *, P < 0.05. (C) Neurons were treated with 45 mM K⁺ in the absence or presence of 20 μM KN93 for various times as indicated. Proteins in mitochondrial fractions were separated by SDS-PAGE and analyzed by immunoblotting using Drp1 T-Drp1, P-Drp1, or Tom40 (for quantification of mitochondrial membrane) antibodies. (D) Quantitative analysis of Drp1 translocation. Drp1 levels were normalized to Tom40 levels at the time points indicated in Fig. 7 C. Cumulative results from three independent experiments for each condition are shown. Error bars indicate the mean ± SEM in each group.

Phosphorylation regulates Drp1 and hFis1 binding in vitro. (A) Recombinant Drp1-6His was preincubated in the absence or presence of CaMKIα (KI) and ATP and mixed with GST-hFis1 as indicated. Proteins were recovered using glutathione-Sepharose and analyzed by immunoblotting using an anti-6His antibody to detect Drp1 (top; bottom shows Drp1 input). Quantitation of the results is shown in the bar graph. The results were otained in three independent experiments. Error bars indicate SEM in each group. (B) Drp1-6His was maximally phosphorylated by CaMKI, and the sample was mixed in different ratios with nonphosphorylated Drp1-6His (values are expressed as percent P-Drp1/Drp1; see x axis). GST pull-down assays were performed as in A. The data represent means from three independent experiments. (C) The binding of mutant Drp1-S600A-6His to GST-hFis1 was compared with unphosphorylated and phosphorylated wild-type Drp1-6His. Drp1 was detected by immunoblotting as described above. Similar results were obtained from two independent experiments.

High K⁺ induces elevation in intracellular Ca²⁺ and reversible changes in mitochondrial morphology. (A) Intracellular Ca²⁺ level was measured in hippocampal neurons (12 DIV) in response to treatment with 45 mM K⁺. Graphs show normalized ratios of F340/F380 in four different neurons (thin lines) and the mean value (bold line). (B) Neurons transfected with pDsRed2-Mito were analyzed before addition of 45 mM K⁺ and 15 min after treatment with high K⁺. Arrows show ringlike mitochondria formations. After 15 min, 45 mM K⁺ was removed by exchange of media, and mitochondria were imaged after 3 h. The same result was obtained in three separate experiments (n = 3). Bars, 10 μm. (C) Neurons (12 DIV) were untreated or treated for 15 min with 500 μM glutamate or 45 mM K⁺. After the 15-min treatment, the media was replaced by conditioned medium from 10-d-old neuronal cultures. 24 h later, cell death was measured by Hoechst 33258 and MAP2 staining. Data were obtained from at least five visual fields under a 20× objective from at least three independent experiments for each group. Error bars indicate SEM in each group. Bars, 40 μm.

CaMKIα mediates VDCC-dependent mitochondrial changes after high K⁺. (A) Neurons (10 DIV) were cotransfected with pDsRed2-Mito and GFP (negative control; top) or GFP–CaMKIα-DN. Neurons (at 11 DIV) were treated with 45 mM K⁺ for 15 min. GFP fluorescence is shown in the left panels, and mitochondrial fluorescence is shown before (Pre-K⁺) and after high K⁺ treatment in the middle and right panels. Bars, 20 μm. (B) Neurons were transfected with GFP–CaMKIα-CA and pDsRed2-Mito. GFP fluorescence is shown in the left panel, and mitochondrial fluorescence is shown in the right panel. Bar, 20 μm. (C) Mitochondrial movement or length was measured for neurons that were either untreated (no stimulation) or treated with 45 mM K⁺ as shown in A. The movement events after high K⁺ stimulation were normalized to the value obtained before stimulation. The elongation index after 15 min of high K⁺ stimulation was also normalized to the value obtained before stimulation. Data were obtained from >150 mitochondria in three to five independent experiments for each condition. Error bars indicate SEM in each group. *, P < 0.001. (D) Mitochondrial movement or elongation index was measured for neurons that were either untreated (Control; transfected with GFP) or after transfection with GFP–CaMKI-CA as in B. Mitochondrial movement was measured by time-lapse fluorescence microscopy. Mitochondrial length was measured through analysis of the elongation index. Data were obtained from at least 350 mitochondria in 7–10 neurons for each condition. Error bars indicate SEM in each group. *, P < 0.001.

Drp1-S600 phosphorylation is required for the effects of high K⁺ on mitochondrial dynamics. (A) HeLa cells were first transfected with control siRNA (siCont; Invitrogen) and siRNA for Drp1 (siDrp1) before transfection with the Drp1 plasmids. 1 d later, HeLa cells were transfected with GFP, YFP-Drp1^R, YFP-Drp1^R-S600A, or YFP-Drp1^R-S600D together with pDsRed2-Mito. Proteins were separated by SDS-PAGE and analyzed by immunoblotting using Drp1 and actin antibodies. (B) HeLa cells were treated without (No sti) or with 20 mM K⁺ (high K⁺), and mitochondrial fluorescence was analyzed by time-lapse imaging. Mitochondrial length was measured through analysis of the elongation index. Bars, 5 μm. (C) The elongation index was normalized to the value before stimulation. Values shown are the mean ± SEM (n = 5) in each group. Values were analyzed using one-way ANOVA followed by post-Tukey test. *, P < 0.01. (D) Neurons (10 DIV) were cotransfected with pDsRed2-Mito and GFP, YFP-Drp1, or YFP-Drp1-S600A as indicated. Neurons (at 11 DIV) were then treated without (No sti) or with 45 mM K⁺ for 15 min, and mitochondrial fluorescence was analyzed by time-lapse imaging. Bars, 10 μm. (E) Mitochondrial length was measured through analysis of the elongation index. Values shown are the mean ± SEM (n = 5) in each group. Values were analyzed using one-way ANOVA followed by post-Tukey test. *, P < 0.05; **, P < 0.005.

High K⁺ treatment regulates Drp1 dynamics in HeLa cells. (A) HeLa cells were transfected with control siRNA (siCont) and siRNA for Drp1 (siDrp1) before the transfection with GFP, YFP-Drp1^R, or YFP-Drp1^R-S600A and mito-DsRed as described above. HeLa cells were then either untreated or incubated with 20 mM K⁺ for 15 min. (B) Quantitative analysis of exogenous Drp1 foci. Data were obtained from the results from at least five independent experiments for each condition shown in A. Values were analyzed using one-way ANOVA followed by post-Turkey test. Control is siDrp1 without stimulation. Error bars indicate SD in each group. *, P < 0.01 versus siDrp1+YFP-Drp1^R+No stimulation. Bars, 10 μm.