Fig 1.
Drosophila vimar affects mitochondrial morphology under normal conditions.
(A) a-e, Live imaging of the mitochondrial morphology in the flight muscle of adult flies. The mitochondria are labeled with UAS-mitoGFP driven by Mhc-Gal4 (Mhc>mitoGFP). The genotype is indicated on each micrograph. f, To quantify the mitochondrial size, the averaged mitochondrial size of the control (+/+) is set as 1, and the relative ratios of the other genotypes to the control are shown. Five thoraces from each genotype were quantified. Bar graphs throughout all figures are means ± SD. The white bar represents the control, the gray bar represents no statistical different from the control, and the black bar represents significantly different from the control. * for p<0.05; ** for p<0.01; ***for p<0.001. (B) Mitochondrial distribution and morphology in larval oenocytes. a, The mitochondria are labeled with PromE(800)> mitoGFP. b-e, The effects of Khc, Milton, Miro and vimar RNAi are shown. The dotted red lines denote the cell boundaries, which were determined by the mitoGFP background. a'-e', Enlarged view of the white box labeled area in the upper panel. f, To quantify the mitochondrial length, the averaged mitochondrial length of the control (+/+) is set as 1, and the relative ratios of the other genotypes to the control are shown. Mitochondrial length of five oenocytes was quantified per genotype and shown as means ± SD. (C) Live imaging of mitochondria in eye disc after knocking down vimar by GMR>mitoGFP. Three eye discs were analyzed for each genotype. (D) Effect of vimar on mitochondrial transport. The mitochondria are labeled with mitoGFP (CCAP>mitoGFP), and their movements in the axons were recorded and transformed into kymographs. Mitochondria motion in ten axons from five larvae was analyzed for each genotype. The quantification is shown on the bar graph. (E) Subcellular vimar protein distribution by protein fractionation. The proteins from adult thoraces (Mhc>MitoGFP) were separated into cytosolic and crude mitochondrial fractions. The vimar protein enrichment was analyzed by immunobloting with the anti-vimar antibody. The mitoGFP protein was detected by the anti-GFP antibody; and β-actin is a cytosolic protein.
Fig 2.
Interaction of vimar and Miro.
(A) Co-Immunoprecipitation of vimar and Miro. The proteins were collected from the HEK293T cells that expressed both Flag-tagged Vimar (Flag-Vimar) and HA-tagged Miro (HA-Miro). Then, the proteins were precipitated with a HA (left panel) or Flag antibody (right panel). The control IgG is shown as a negative control. The total protein input is shown as the protein loading control. (B) An example of the defective wing posture. Compared to the control, overexpression of Miro in the adult flight muscle (Mhc>Miro) resulted in an upright fly wing posture. (C) Quantification of defective wing posture in the Miro overexpression background or in the Mhc-Gal4 background. Trial N = 3, with 100–150 flies examined in each experiment. (D) Live imaging of the mitochondrial morphology in the fly flight muscle. The genotype of each fly muscle is labeled on the micrograph. Five thoraces were quantified for each genotype. (E) Quantification of defective wing posture in the Miro20V and Miro25N overexpression background. Trial N = 3, with 100–150 flies examined in each experiment. (F) Live imaging of the mitochondrial morphology in the fly flight muscle in the indicated genotypes. Five thoraces were quantified for each genotype.
Fig 3.
Vimar modulates the mitochondrial length in the high calcium condition.
(A) Effect of the Drp1 mutation on mitochondrial fission in high calcium conditions. a and b, Mitochondrial dendrites in the larval chordotonal neurons in the control flies (Appl>mitoGFP at 18°C and Appl>mitoGFP, tub-Gal80ts at 30°C). c and d, Mitochondrial morphology in the AGM and AGM/drp11 flies. a'-d' are the enlarged view from the boxed area in a-d, and the mitochondrial lengths in a'-d' were quantified. 10 to 16 chordotonal organs from each genotype were examined. (B) Effect of the vimar overexpression on the mitochondrial length after induction of AGM expression for 26 hours. 10 to 16 chordotonal organs were examined for each genotype. (C) Effect of vimar mutant (vimark16722) on the mitochondrial fragmentation of the AGM flies. 10 to 16 chordotonal organs were examined for each genotype. (D) Effect of Miro overexpression and the vimar mutant on the mitochondrial fragmentation of the AGM flies. 10 to 16 chordotonal neurons were examined for each genotype.
Fig 4.
Vimar suppresses neuronal necrosis and muscle degeneration induced by the Pink1 mutant.
(A) Effect of the Drp1 mutant on neuronal necrosis. The micrographs showed the live images from larval chordotonal neurons. The control (Appl>GFP;tub-Gal80ts) displays the cell bodies of the wild type chordotonal neurons, which form a cluster containing 6 neurons. In the AGG background, the wild type (+/+) flies showed swollen cell bodies, weakened GFP intensity and neuronal cell loss; and these defects were rescued under the Drp1 mutant (Drp11) background. The right panel shows the quantification of the cell loss. For all quantification of neuronal necrosis, trial N = 5, with 10–15 flies were examined in each trial in this figure. (B) Effect of the Drp1 mutant on the survival of the AG adult flies. For all quantification of AG lethality, trial N = 3, with 100–150 flies were examined for each trial. (C) Effect of vimar mutant on neuronal necrosis. (D) Effect of the vimar mutant on the survival of the AG flies. (E) Effect of vimar overexpression on neuronal necrosis. (F) Effect of vimar overexpression on the survival of the AG flies. (G) Effect of Miro overexpression on neuronal necrosis. The result showed that Miro overexpression enhanced neuronal necrosis; and the vimar mutant had no rescue effect on this defect. (H) Effect of Miro overexpression on the survival of the AG flies. (I) Effect of the vimar mutant (vimark16722) on PINK1 mutant induced mitochondrial defect. The live image showed the mitochondrial morphology in the PINK1 mutant (PINK15) and under the vimar mutant background. Ten thoraces were analyzed for each genotype. (J) Effect of the vimar mutant (vimark16722) on the wing posture defect of the PINK1 mutant (PINK15). Trial N = 3, with 100–150 flies were examined in each trial.
Fig 5.
The conserved role of RAP1GDS1 in mammalian cells.
(A) Effect of RAP1GDS1 knock down on the mitochondrial morphology in HEK293T cells. The mitochondria in the cells that stably express the RAP1GDS1 shRNA are labeled with a transiently transfected MitoDsred expression vector. The cells were classified as tubular-shape or punctate-shape based on differences in their mitochondrial lengths. The ratio of punctate-shape mitochondria is shown in the right panel. The result showed that RAP1GDS1 shRNA had a trend to increase the punctate-shape mitochondria (not statistically different from the control shRNA). Trial N = 3, with 100 cells were quantified in each trial. (B) Effect of RAP1GDS1 knocking down on the mitochondrial fragmentation under calcium overload stress. The HEK293T cells were treated with 20 μM calcium ionophore (A23187) for 4 hours. The result showed that RAP1GDS1 shRNA reduced fragmented mitochondria upon calcium ionophore treatment. Trial N = 3, with 100 cells were quantified in each trial. (C) Effect of the RAP1GDS1 shRNA on calcium ionophore-induced necrosis. The HEK293T control and RAP1GDS1 shRNA stable cell lines were treated with 20 μM A23187 for 14 hours. Then, the cell death was quantified by the ATP assay. The result indicated that less cell death occurred in the RAP1GDS1 shRNA expressing cells. Trial N = 3. (D) Effect of the RAP1GDS1 shRNA on calcium ionophore-induced necrosis. The PI and DAPI staining patterns are shown. The red signals indicate the PI-positive cells and the blue channel indicates the DAPI staining. Trial N = 3. (E) Effect of the Miro1 siRNA on calcium ionophore induced necrosis determined by the ATP assay. The Miro1 siRNA was transiently transfected in HEK293T cells for 48 hours. Trial N = 3. (F) Effect of the Miro1 siRNA on calcium ionophore induced necrosis determined by the PI staining assay. The PI and DAPI staining patterns are shown. The same result was observed as in E. Trial N = 3. (G) Co-Immunoprecipitation of RAP1GDS1 and Miro1. The proteins were collected from the HEK293T cells that expressed Flag-tagged RAP1GDS1 (Flag-RAP1GDS1) and HA-tagged Miro1 (HA-Miro1). The control IgG is shown as a negative control. The total protein input is shown as the protein loading control. Trial N = 3.
Fig 6.
A schematic model of Miro/vimar function on mitochondrial morphology.
In normal calcium conditions, the Miro/vimar complex promotes mitochondrial fission inhibition, and their GOF results in elongated mitochondria. Increased mitochondrial fusion is known to occur in the PINK1 mutant flies, and this defect can be rescued by LOF Miro/vimar. In the high calcium state, the Miro/vimar complex promotes mitochondrial fragmentation, which accelerates neuronal necrosis. Regardless of the intracellular calcium level, vimar enhances the function of Miro, because vimar is likely the GEF to promote Miro's GTP/GDP exchange.