Constriction of the mitochondrial inner compartment is a priming event for mitochondrial division

Abstract

Mitochondrial division is critical for the maintenance and regulation of mitochondrial function, quality and distribution. This process is controlled by cytosolic actin-based constriction machinery and dynamin-related protein 1 (Drp1) on mitochondrial outer membrane (OMM). Although mitochondrial physiology, including oxidative phosphorylation, is also important for efficient mitochondrial division, morphological alterations of the mitochondrial inner-membrane (IMM) have not been clearly elucidated. Here we report spontaneous and repetitive constriction of mitochondrial inner compartment (CoMIC) associated with subsequent division in neurons. Although CoMIC is potentiated by inhibition of Drp1 and occurs at the potential division spots contacting the endoplasmic reticulum, it appears on IMM independently of OMM. Intra-mitochondrial influx of Ca²⁺ induces and potentiates CoMIC, and leads to K⁺-mediated mitochondrial bulging and depolarization. Synergistically, optic atrophy 1 (Opa1) also regulates CoMIC via controlling Mic60-mediated OMM-IMM tethering. Therefore, we propose that CoMIC is a priming event for efficient mitochondrial division.

Figure 1. CoMIC appears in elongated mitochondria before mitochondrial division.

(a) Time-lapse images of mitochondrial matrices (DsRed-mito) in neuronal process (GFP). The white arrowheads and the arrow indicate sites of mitochondrial bulging and division, respectively. Scale bars, 10 and 5 μm in the lower and higher magnification images, respectively. (b) Kymograph of DsRed-mito in a. The lower graph is the intensity profile at two CoMIC-showing time points (1, 2), and arrow heads indicate conserved bulging site. Two illustrations on the right indicate the pattern of CoMIC and occurrence of division. CoMIC pattern and division are labelled by spikes and an arrow, respectively. (c) Kymograph of DsRed-mito in a neuronal process. The upper numbers indicate mitochondria showing CoMIC. The lower panels show the pattern of CoMIC and occurrence of division in mitochondria labelled with the numbers above. (d) Probability of CoMIC in mitochondria from young (DIV4; <5 μm=530 mitochondria, >5 μm=70 mitochondria) and mature neurons (DIV10; <5 μm=114 mitochondria, >5 μm=111 mitochondria) during a 10-min time period (***P<0.01 by the Mann–Whitney rank sum test). Error bars indicate s.e. (e) Proportion of mitochondria undergoing subsequent division among mitochondria showing no or prior CoMIC (***P<0.01 by the χ²-test with Yates's correction for continuity). (f) Proportion of mitochondria showing prior CoMIC among non-dividing or dividing mitochondria (***P<0.01 by χ²-test with Yates's correction for continuity).

Figure 2. CoMIC is potentiated by inhibition of Drp1, but does not depend on Drp1.

(a) Time-lapse images of the mitochondrial matrix and neuronal processes in DN-Drp1-expressing neurons (DIV4). Arrowheads indicate sites of mitochondrial bulging. Scale bars, 10 and 5 μm in the lower and higher magnification images, respectively. (b) Probability of CoMIC in elongated mitochondria (>5 μm) as compared with control (70 mitochondria) and DN-Drp1-expressing neurons (114 mitochondria). Error bar indicate s.e. (c) Representative kymographic analysis of a mitochondrion showing CoMIC in a DN-Drp1-expressing neuron. Arrow heads indicate conserved bulging site on two CoMIC-showing time points (1, 2). (d) Quantification of the total duration, frequency and mean interval of CoMIC in control and DN-Drp1-expressing neurons (***P<0.01 by Mann–Whitney rank sum test). (e) Straightened images of DsRed-mito showing minimal bulging (cyan signal) and maximal bulging (red signal) in control and DN-Drp1-expressing neurons. (f) Bulging ratio in foci during CoMIC in control and DN-Drp1-expressing neurons (***P<0.01 by Mann–Whitney rank sum test). (g) CoMIC in Drp1 knockout mouse embryonic fibroblasts. The time-lapse panels are enlarged in the green and blue boxes. Arrowheads indicate the sites of mitochondrial bulging. Scale bars, 10 and 5 μm in the lower and higher magnification images, respectively.

Figure 3. CoMIC is spatially associated with the ER.

(a,c) 3D images of mitochondria and ER (GFP-Sec61b) during off-phase, CoMIC and mitochondrial division in control (a) and DN-Drp1-expressing neurons (c). Scale bar, 5 μm. White arrowheads and arrows indicate the spots of CoMIC and division, respectively. Yellow arrows indicate the spots that are enclosed by the ER. (b,d) Intensity profiles of the mitochondrial matrix and the ER in the dotted boxes in a,c, respectively. (e,g) Representative images of mitochondria, GFP-Mff (e), Twinkle–GFP (g) during off-phase, CoMIC and mitochondrial division in control and DN-Drp1-expressing neurons. Scale bar, 5 μm, and white arrowheads and arrows indicate the spots of CoMIC and division, respectively. (f,h) Intensity profiles of the mitochondrial matrix, GFP-Mff and Twinkle–GFP in the dotted boxes in e,g, respectively. (i) A hypothetical model of mitochondrial association with the ER during CoMIC.

Figure 4. CoMIC is OMM-independent event accompanying rearrangement on IMM–OMM contact.

(a,c) Time-lapse images of the mitochondrial matrix and OMM (Tom20-GFP) in control (a) and DN-Drp1-expressing neurons (c). Arrowheads and arrows indicate the bulging sites of mitochondria during CoMIC and mitochondrial division, respectively. Yellow asterisks indicate synchronized bulging of the OMM and matrix. Scale bar, 5 μm. (b,d) Intensity profiles (upper graphs) and schematic diagrams (lower graphs) of the mitochondrial matrix (magenta lines) and OMM (green lines) in the dotted insects shown in a,c, respectively. (e) Probability of CoMIC (114 mitochondria) and OMM constriction (81 mitochondria) in elongated mitochondria from DN-Drp1-expressing neurons (***P<0.01 by Mann–Whitney rank sum test). Error bar indicate s.e. (f) Straightened images of DsRed-mito and Tom20-GFP showing minimal (cyan signal) and maximal bulging (red signal) in a DN-Drp1-expressing neuron. (g) Bulging ratio in foci of DsRed-mito and Tom20-GFP during CoMIC in DN-Drp1-expressing neurons (***P<0.01 by Mann–Whitney rank sum test). (h) Electron microscopy of mitochondria in neuron infected by shDrp1-virus. Scale bar, 500 nm. Green and magenta shades indicate area marked by OMM and IMM, respectively. (i) Time-lapse images for FRET-based proximity analysis between OMM and mitochondrial matrix, which are labelled by Tom20-GFP and mito-BFP, in a DN-Drp1-expressing neuron. Scale bar, 5 μm. FRET ratio (FRET/Tom20-GFP) is pseudo-coloured. (j) Straightened images of FRET/Tom20-GFP during off-phase and CoMIC of mitochondria, and their intensity profiles (lower panel). Arrows indicate hot spots of FRET ratio. (k) Kymographies of mito-BFP and FRET/Tom20-GFP of mitochondria on dotted line of i.

Figure 5. Intra-mitochondrial Ca²⁺ regulates CoMIC.

(a) Time-lapse images of DsRed-mito and CEPIA3mt in DN-Drp1-expressing neuron. Arrow heads indicate mitochondrial bulging. Scale bar, 5 μm, and CEPIA3mt/DsRed-mito is pseudo-coloured. (b) Kymographies of DsRed-mito and CEPIA3mt/DsRed-mito of mitochondria in dotted box of a. (c) Temporal CEPIA3mt/DsRed-mito fluctuation of mitochondria of dotted boxes (#1, #2) of b. Upper red boxes indicate CoMIC duration of mitochondrion labelled by #2. (d) Representative kymographies of mito-GFP and patterns of CoMIC (right diagram) in DN-Drp1-expressing neurons treated with DMSO and 2-APB (1 μM) for 1 h. (e) Probability of CoMIC in elongated mitochondria treated with DMSO (103 mitochondria) and 2-APB (94 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (***P<0.01 and *P<0.05 by Mann–Whitney rank sum test). Error bar indicate s.e. (f–i) Representative kymographies of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1-expressing neurons treated with DMSO (f), Ru360 (10 μM) (g), thapsigargin (TG, 1 μM) (h) and TG with 30 min pretreatment with Ru360 (Ru360+TG) (i) for 1 h. (j) Probability of CoMIC in elongated mitochondria treated with DMSO (100 mitochondria), Ru360 (95 mitochondria), TG (88 mitochondria) or Ru360+TG (95 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (***P<0.01, **P<0.02 and *P<0.05 by Mann–Whitney rank sum test). (k) Representative kymographies of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1+shMock- and DN-Drp1+shMcu-expressing neurons. (l) Probability of CoMIC in elongated mitochondria from DN-Drp1+shMock- (105 mitochondria) and DN-Drp1+shMcu-expressing neurons (118 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (**P<0.02 and *P<0.05 by Mann–Whitney rank sum test).

Figure 6. Intra-mitochondrial Ca²⁺ contributes to efficient mitochondrial division.

(a) Time-lapse images of DsRed-mito and CEPIA3mt in control neuron. Scale bar, 5 μm, and CEPIA3mt/DsRed-mito is pseudo-coloured. Arrow heads and arrow indicate mitochondrial bulging and division, respectively. (b) Kymographies of DsRed-mito and CEPIA3mt/DsRed-mito of mitochondria in dotted region of a. (c) Temporal CEPIA3mt/DsRed-mito fluctuation of a mitochondrion of dotted region of a. Upper red box and arrow indicate CoMIC duration and mitochondrial division, respectively. (d) Representative morphology of mitochondria in neurons treated with DMSO and Ru360 with or without TG. The lower, enlarged panels are straightened images from the boxes in the upper panels. Scale bar, 5 μm. (e) Average mitochondrial length following treatment with DMSO (2,415 mitochondria), Ru360 (635 mitochondria), TG (1,438 mitochondria) and TG+Ru360 (706 mitochondria). (***P<0.01 and **P<0.02 by Mann–Whitney rank sum test). Error bar indicate s.e. (f) Representative morphology of mitochondria in shMock- and shMcu-expressing neurons. The enlarged right panels are straightened images from the dotted boxes. Scale bar, 5 μm. (g) Average mitochondrial length following treatment with shMock- (1,034 mitochondria) and shMcu-expressing neurons (1,130 mitochondria). (***P<0.01 by Mann–Whitney rank sum test).

Figure 7. Intra-mitochondrial K⁺ influx is involved in mitochondrial bulging and depolarization during CoMIC.

(a,b) Representative kymographies of DsRed-mito and patterns of CoMIC (right) in DN-Drp1-expressing neurons after treatment with paxilline (Pax, 10 μM) (a) and TG with 30 min pretreatment with Pax (Pax+TG) for 1 h (b). (c) Probability of CoMIC in elongated mitochondria treated with DMSO (103 mitochondria), Pax (110 mitochondria), TG (100 mitochondria) and Pax+TG (117 mitochondria) (*P<0.05 by Mann–Whitney rank sum test). Error bar indicate s.e. (d) Straightened images of DsRed-mito showing minimal (cyan signal) and maximal bulging (red signal). (e) Quantification of bulging ratio in foci stained with DsRed-mito (***P<0.01 by Mann–Whitney rank sum test). (f) Time-lapse images of mitochondria labelled by mito-GFP and TMRE in DN-Drp1-expressing neurons without or with Pax treatment. Scale bar, 5 μm, and TMRE/mito-GFP is pseudo-coloured. (g,h) Temporal fluctuations of TMRE and mito-GFP in dotted box of f. Gray boxes indicate CoMIC duration of mitochondria.

Figure 8. Oma1-mediated cleavage of Opa1 regulates CoMIC.

(a–d) Representative kymographies of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1+shMock- (a), DN-Drp1+shOpa1- (b), TG-treated DN-Drp1+shMock- (c) or TG-treated DN-Drp1+shOpa1-expressing neurons (d). (e) Probability of CoMIC in elongated mitochondria from DN-Drp1+shMock-, DN-Drp1+shOpa1- (123 mitochondria), TG-treated DN-Drp1+shMock- (138 mitochondria) and TG-treated DN-Drp1+shOpa1-expressing neurons (105 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (***P<0.01, **P<0.02 and *P<0.05 by Mann–Whitney rank sum test). Error bar indicate s.e. (f) Representative kymography of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1+shOma1-expressing neuron. (g) Probability of CoMIC in elongated mitochondria from DN-Drp1+shMock- and DN-Drp1+shOma1-expressing neurons (134 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (**P<0.02 and *P<0.05 by Mann–Whitney rank sum test). (h–m) Representative kymographies of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1- (h), DN-Drp1+S-Opa1- (i), ruthenium Red (RuR, 10 μM)-treated DN-Drp1+S-Opa1- (j), Pax-treated DN-Drp1+S-Opa1- (k), DN-Drp1+L-Opa1- (l) and TG-treated DN-Drp1+L-Opa1-expressing neurons (m). (n) Probability of CoMIC in elongated mitochondria from DN-Drp1- (105 mitochondria), DN-Drp1+S-Opa1- (111 mitochondria), RuR-treated DN-Drp1+S-Opa1- (105 mitochondria) and Pax-treated DN-Drp1+S-Opa1-expressing neurons (104 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (***P<0.01, **P<0.02 and *P<0.05 by Mann–Whitney rank sum test). (o) Probability of CoMIC in elongated mitochondria from DN-Drp1-, TG-treated-DN-Drp1- (138 mitochondria), DN-Drp1+L-OPA1- (140 mitochondria) and TG-treated DN-Drp1+L-Opa1-expressing neurons (104 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC. (***P<0.01 and **P<0.02 by Mann–Whitney rank sum test).

Figure 9. Opa1 neutralizes Mic60-mediated tethering OMM and IMM so regulating CoMIC.

(a) Representative kymography of DsRed-mito and patterns of CoMIC (right diagram) in DN-Drp1+shMic60-expressing neurons. (b) Probability of CoMIC in elongated mitochondria from DN-Drp1+shMock- and DN-Drp1+shMic60-expressing neurons (94 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (*P<0.05 by Mann–Whitney rank sum test). Error bar indicate s.e. (c–f) Representative kymographies of mito-GFP and patterns of CoMIC (right diagram) in DN-Drp1+Mock (c), DN-Drp1+Mic60- (d), DN-Drp1+S-Opa1- (e) and DN-Drp1+S-Opa1+Mic60-expressing neurons (f). (g) Probability of CoMIC in elongated mitochondria from DN-Drp1-, DN-Drp1+Mic60- (118 mitochondria), DN-Drp1+S-Opa1- (71 mitochondria) and DN-Drp1+S-Opa1+Mic60-expressing neurons (63 mitochondria), and quantification of the total duration, frequency and average single duration of CoMIC (***P<0.01, **P<0.02 and *P<0.05 by Mann–Whitney rank sum test).

Figure 10. Schematic model of CoMIC for mitochondrial division.

Intra-mitochondrial Ca²⁺ flux triggers mitoBK_Ca-mediated mitochondrial bulging and depolarization. Synergistically, stabilized Oma1 cleaves L-Opa1 resulting accumulation of S-Opa1, which neutralizes Mic60-mediated OMM–IBM tethering. This CoMIC contributes to Drp1-mediated mitochondrial division.