DRP1-mediated mitochondrial shape controls calcium homeostasis and muscle mass

Abstract

Mitochondrial quality control is essential in highly structured cells such as neurons and muscles. In skeletal muscle the mitochondrial fission proteins are reduced in different physiopathological conditions including ageing sarcopenia, cancer cachexia and chemotherapy-induced muscle wasting. However, whether mitochondrial fission is essential for muscle homeostasis is still unclear. Here we show that muscle-specific loss of the pro-fission dynamin related protein (DRP) 1 induces muscle wasting and weakness. Constitutive Drp1 ablation in muscles reduces growth and causes animal death while inducible deletion results in atrophy and degeneration. Drp1 deficient mitochondria are morphologically bigger and functionally abnormal. The dysfunctional mitochondria signals to the nucleus to induce the ubiquitin-proteasome system and an Unfolded Protein Response while the change of mitochondrial volume results in an increase of mitochondrial Ca²⁺ uptake and myofiber death. Our findings reveal that morphology of mitochondrial network is critical for several biological processes that control nuclear programs and Ca²⁺ handling.

Fig. 1

DRP1 deletion impairs muscle growth and causes a lethal phenotype. Drp1 mRNA (a) and DRP1 protein levels (b) are downregulated only in skeletal muscle and not in other tissues in DRP1-null mice. Each condition represents the average of at least three independent experiments ± SEM. c Kaplan–Meier survival curve of Drp1^fl/fl and Drp1^−/− littermates (WT, n = 43; KO, n = 14), indicates that muscle-specific DRP1 deletion results in lethality within postnatal day 30. d DRP1-deficient mice show an impairment of total body weight. e Upper panel: representative photograph of control and KO littermates at postnatal day 12 showing that Drp1^−/− mice are smaller than controls. Lower panel: haematoxilin–eosin (HE) staining of hindlimb cross-section. The cross-sectional area analysis of fibers (g) and the total number of different fiber types in hindlimb muscles (f) indicate a reduction of DRP1 KO fiber size without hypoplasia. The data represent mean ± SEM (g, n = 3 and f, n = 3). h Representative immunoblot analysis of muscle homogenates of four independent experiments. MyoD and Myogenin were normalized to GAPDH expression levels. The data represent average ± SEM. Two-tailed unpaired Student’s t test was used. Statistical significance: *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001

Fig. 2

Muscle DRP1-chronic inhibition alters mitochondrial morphology, RCS organization and complex-I- and complex-II-dependent respiration. a Representative EM cross-sectional images of mitochondrial distribution and morphology in gastrocnemius from WT and Drp1^−/− mice. In cross-sections, the different morphology of mitochondria in Drp1^−/− from WT is often evident: in Drp1^−/− fibers mitochondria are more often larger in size (scale bar: 1 μm). b Mitochondrial DNA copy number quantification. mtDNA was amplified by RT-PCR from the total DNA of gastrocnemius muscles of the indicated genotype. The data are normalized to control and represent average ± SEM of five independent experiments. c Representative immunoblots of four independent experiments of muscle homogenates showing no difference of porin/GAPDH ratio. d RT-PCR analysis of mitochondria dynamics transcripts of muscles of WT and Drp1^−/−. DRP1 deletion induces PGC1α, OPA1, and Mfn1 transcript levels. The data represent average ± SEM (n = 8) and are normalized for GAPDH and expressed as fold increase of controls. e Succinate dehydrogenase (SDH) staining of Drp1^fl/fl and Drp1^−/− muscles show a different distribution of mitochondrial network in KO mice. f Respiratory control ratio (RCR) of muscle-isolated mitochondria energized with 5 mM/2.5 mM GLU/MAL or 10 mM SUCC. The data represent average ± SEM of three independent experiments. g DRP1 deficiency leads to a significant reduction of CIII and CIV assembly in respiratory chain supercomplexes (RCS). Representative Blue Native PAGE analysis showing RCS developed and normalized for individual respiratory chain complexes. Left panel: CI subunit NDUFB8, middle panel: CIII-Core2 protein 2 and right panel: CIV subunit COXI. h Representative immunoblot analysis of OPA1 different isoforms show an increase in OPA1 cleavage in Drp1-null muscles (n = 4). Two-tailed unpaired Student’s t test was used. Statistical significance: *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001

Fig. 3

DRP1 absence leads to downregulation of protein synthesis and activation of the Ubiquitin–proteasome and autophagy–lysosome pathways. a In vivo SUnSET technique showed a significant reduction of protein synthesis in DRP1-ablated mice muscles. Representative western blot and quantification of the puromycin-labeled peptides, expressed as percentage of the values obtained in the control group, are depicted. The data represent average ± SEM of eight independent experiments. b The total protein extracts from newborns muscles were immunoblotted with the indicated antibodies. A representative immunoblot of four independent experiments is shown. Statistical significance of specified densitometric ratios is indicated on the right. The data are normalized to GAPDH (n = 8). Representative immunoblot (c) and RT-PCR (d) showing the activation of the unfolding protein response (UPR) pathway in DRP1 KO muscles. The data represent average ± SEM of seven independent experiments. e RT-PCR analysis of transcriptional levels of muscle FGF21 (n = 3). f Quantification of blood levels of FGF21 shows an increase in DRP1 KO mice (n = 3), leading to a decrease in blood glucose (g, n = 4) and to a significant decrease in IGF1 class II mRNA level in the liver (h, n = 4). The data represent average ± SEM. i Quantification of IGF1 in plasma from Drp1^−/− and control mice shows a significant decrease of IGF1 circulating levels in the absence of DRP1. RT-PCR analysis of FoxO-dependent transcripts shows an increase in FoxO3 levels (j), in the muscle-specific ubiquitin ligases Atrogin1 and MuRF1 (k) and in the novel Ubiquitin–ligase MUSA1 (l). The data represent average ± SEM (n = 6 per condition). m Representative blots and densitometric analysis of at least four independent experiments of total muscle extracts immunoblotted for anti-Ubiquitin (Lys48) and for anti-Ubiquitin (Lys63) normalized to GAPDH. The data represent average ± SEM. n Quantitative PCR analysis of autophagy-related transcripts showing a significant induction of autophagy markers in Drp1^−/−. The data represent average ± SEM (n = 8 per condition). o Representative immunoblots of four independent experiments of autophagy-related proteins. The data represent average ± SEM (WT, n = 4; KO, n = 6). Two-tailed unpaired Student’s t test was used. Statistical significance: *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001

Fig. 4

DRP1 loss in adult animals causes body weight loss, muscle atrophy and muscle weakness. a DRP1 protein levels are downregulated in different types of skeletal muscles in HSADRP1-null mice. O.D. levels represent the average of at least three independent experiments ± SEM. b Growth curve of control and Drp1^−/− littermates during and after tamoxifen treatment. KO mice start to lose body weight after 7 weeks of tamoxifen treatment and during the following weeks. The data represent average ± SEM (WT, n = 10; KO, n = 11). c Dissected gastrocnemius muscles from the control and DRP1-null mice show an important muscle atrophy after DRP1 deletion in adult animals. d Quantification of the cross-sectional area of myofibers indicates a significant reduction in DRP1-ablated muscles. Values represent average ± SEM (WT, n = 5; KO 70 days, n = 5; KO 180 days, n = 3). e Representative haematoxilin–eosin staining of Tibialis Anterior muscle showing signs of myofiber degeneration/regeneration in Drp1^−/− after 180 days of treatment. f In Drp1^−/− muscles after 180 days of treatment, there are 10 and 25% of myofibers are centrally nucleated, in tibialis anterior muscle and gastrocnemius muscle, respectively (n = 3 mice each condition). g Force measurements performed in vivo on gastrocnemius muscles. Absence of DRP1 leads to a significant decrease in both absolute force and maximal specific force generated during tetanic contraction. The data represent average ± SEM (n = 3). Two-tailed unpaired Student’s t test and two-way analysis of variant (ANOVA) were used. Statistical significance: *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001

Fig. 5

Ablation of DRP1 in adulthood modifies mitochondrial shape and reduces RCS assembly, Complex-I-, Complex-II-, and complex-IV-dependent respiration. a SDH staining indicating the presence of bigger mitochondria in KO muscles compared with control. b Representative electron micrographs of EDL muscles of controls and Drp1^−/−. Left panel: in WT mitochondria (pointed by black arrowheads) are placed in proximity of Z lines and usually exhibit an electron-dense matrix (inset). Center panel: in Drp1^−/− mice mitochondria are often larger in size, oriented longitudinally (large arrows), and damaged (white arrowhead and inset). Right panel: a Drp1^−/− fiber presenting spindle-shaped regions (asterisks) which are surrounded by a membrane (inset, white arrow) and that contains vacuoles. c Absence of DRP1 induces mitochondrial depolarization. Isolated adult fibers were loaded with TMRM. Oligomycin and the protonophore FCCP were added at the indicated time points. TMRM staining was monitored in 18 fibers for WT, 52 fibers for KO. Lower panel: representative images of adult myofibers showing altered mitochondrial distribution in DRP1-null muscles. Myofibers were loaded with the potentiometric die TMRM. d Respiratory control ratio (RCR) of muscle isolated mitochondria energized with 5 mM/2.5 mM GLU/MAL or 2 mM rotenone/10 mM succinate or 3 mM ascorbate/10 mM TMPD. The data represent average ± SEM (GLUT/MAL, n = 5 each condition; succinate/rotenone: WT, n = 2; KO, n = 7; ascorbate/TMPD, n = 7 each condition). e The assembly of CIII and CIV in RCS is significantly reduced in mitochondria of DRP1-deficient muscles. Representative Blue Native PAGE analysis showing RCS developed and normalized for individual respiratory chain complexes. CI subunit NDUFB8 (left panel), CIII-Core2 protein 2 (middle panel), and CIV subunit COXI (right panel). f Mitochondrial DNA copy number quantification in controls and KO muscles. mtDNA was amplified by RT-PCR from the total DNA of gastrocnemius muscles of the indicated genotype. The data are normalized to controls and represent the average ± SEM (WT, n = 4; KO, n = 5). g Mitochondrial mass revealed by TOM20 and porin was not affected by DRP1 deletion (n = 4). Two-tailed unpaired Student’s t test was used. Statistical significance: *p ≤ 0.05; ***p ≤ 0.001

Fig. 6

Protein degradation pathways are activated with acute inhibition of DRP1 in adult muscles. a The total protein extracts from adult muscles were immunoblotted with indicated antibodies. A representative immunoblot of four independent experiments is shown. Statistical significance, of specified densitometric ratios is indicated with asterisks on the right. b In vivo SUnSET technique shows no differences in protein synthesis rate during adulthood (n = 4). c Representative immunoblot (n = 4) and RT-PCR analysis (n ≥ 5) (d) showing the activation of the unfolded protein response and the upregulation of FGF21 in DRP1 KO muscles. e Immunoblot and densitometric quantification showing the accumulation of FoxO3 in the nuclear fraction of tibialis anterior muscles from Drp1^−/− mice (n = 5 each condition). f RT-PCR analysis of transcriptional levels of the muscle-specific ubiquitin ligases Atrogin1 and MuRF1 and the novel ubiquitin–ligases MUSA1, SMART, FbxO31, TRIM37, and Itch. The data represent average ± SEM (at least n = 5 per condition). g Quantitative PCR analysis of autophagy-related transcripts showing a significant induction of autophagy markers in Drp1^−/−. h Western blot analysis of autophagy marker showing an impaired autophagy in Drp1^−/− (WT, n = 4; KO, n = 5). i Immunoblots and relative densitometric quantification from colchicine treated mice support a block in autophagic flux after Drp1 inhibition (n ≥ 5 mice per each condition). j Keima assay indicating the decreased mitophagy in DRP1-KO myofibers (WT, n = 41; KO, n = 39). Two-tailed unpaired Student’s t test was used. Statistical significance: **p ≤ 0.01; ***p ≤ 0.001

Fig. 7

Acute DRP1 inhibition alters calcium homeostasis. a Cytosolic calcium level at rest is not affected in KO muscles (40 days: WT, n = 24; KO, n = 28; 70 days, n = 19 fibers). b Amplitude of cytosolic Ca²⁺ transients induced by 0.5 Hz electrical pulses is lower in Drp1^−/− mice (40 days, WT, n = 25; KO, n = 28 fibers; 70 days, WT, n = 33; KO, n = 29 fibers). c Cytosolic calcium transients induced by caffeine–thapsigargin are strongly reduced (40 days, WT, n = 10; KO, n = 14 fibers; 70 days, WT, n = 13 fibers; KO, n = 14 fibers). d Mitochondrial matrix-free calcium concentration is not affected at rest but is higher in Drp1^−/− compared with control during a train of electrical stimulation at 60 Hz, 2 s duration (WT, n = 9; KO, n = 7 fibers; Peak, n = 6 each condition). e Immunoblot analysis of isolated mitochondria. MCU levels were normalized to TOM20 levels. f Quantitative RT-PCR analysis showing a significant decrease of miR1 levels in Drp1^−/− skeletal muscle (n = 4 per condition). The data represent average ± SEM. g Ultrastructure and geometry of CRUs in DRP1 KO muscles. Inserts A and B show T-tubule staining in WT (A) and in DRP1 KO (B); the frequency of oblique/longitudinal tubules increases (stars) when compared with control samples. Inserts C–F are representative EM pictures showing the structural geometry of a normal, an oblique, a longitudinal triad and of a dyad respectively. In inserts A and B, T-tubules are stained in black while in inserts C–F, T-tubules has been false-labeled in green. h Cytosolic calcium transients induced by caffeine–thapsigargin after the reintroduction of DRP1 (Drp1-YFP) in muscle fibers. Cytosolic calcium transients in Drp1^−/− in the presence of Drp1-YFP show no difference compared with control fibers (Drp1^fl/fl, n = 22; Drp1^fl/fl—Drp1-YFP, n = 7; Drp1^−/−, n = 22; Drp1^−/− – Drp1-YFP, n = 13). i Inhibition of mitochondria Ca²⁺ uptake via MCU knockdown restored cytosolic Ca²⁺ levels during caffeine–thapsigargin treatment (WT, n = 17 fibers; WT shMCU, n = 23 fibers; KO, n = 27 fibers; KO shMCU, n = 17 fibers). j MCU inhibition reduces centrally nucleated fibers in KO muscles. For all experiments a minimum of three mice was used for each condition. Two-tailed unpaired Student’s t test and two-way analysis of variant (ANOVA) were used. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001

Fig. 8

Scheme of the mechanisms induced by acute DRP1 inhibition