Short Article
ATP Citrate Lyase Improves Mitochondrial Function in Skeletal Muscle Graphical Abstract
Authors Suman Das, Frederic Morvan, ..., David J. Glass, Mara Fornaro
Correspondence [email protected] (D.J.G.), [email protected] (M.F.)
In Brief Mitochondrial function is an important determinant of skeletal muscle homeostasis. Das et al. report that activation of ATP citrate lyase (ACL) by IGF1 in skeletal muscle increases cardiolipin content and mitochondrial complex activity and improves mitochondrial function, revealing how the IGF1/ACL pathway combines anabolic signaling with energy production.
Highlights d
ACL regulates mitochondrial function in skeletal muscle
d
IGF1 induces activation of ACL in an AKT-dependent fashion
d
ACL phosphorylation is impaired in skeletal muscle of sarcopenic mice
d
ACL activity is associated with increased oxygen consumption and ATP levels
Das et al., 2015, Cell Metabolism 21, 868–876 June 2, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.cmet.2015.05.006
Cell Metabolism
Short Article ATP Citrate Lyase Improves Mitochondrial Function in Skeletal Muscle Suman Das,1 Frederic Morvan,1 Benjamin Jourde,1 Viktor Meier,1 Peter Kahle,1 Pascale Brebbia,1 Gauthier Toussaint,1 David J. Glass,2,* and Mara Fornaro1,* 1Novartis
Institutes for Biomedical Research, Forum 1, Novartis Campus, 4056 Basel, Switzerland Institutes for Biomedical Research, 100 Technology Square, Cambridge, MA 02139, USA *Correspondence: [email protected] (D.J.G.), [email protected] (M.F.) http://dx.doi.org/10.1016/j.cmet.2015.05.006 2Novartis
SUMMARY
Mitochondrial dysfunction is associated with skeletal muscle pathology, including cachexia, sarcopenia, and the muscular dystrophies. ATP citrate lyase (ACL) is a cytosolic enzyme that catalyzes mitochondria-derived citrate into oxaloacetate and acetylCoA. Here we report that activation of ACL in skeletal muscle results in improved mitochondrial function. IGF1 induces activation of ACL in an AKT-dependent fashion. This results in an increase in cardiolipin, thus increasing critical mitochondrial complexes and supercomplex activity, and a resultant increase in oxygen consumption and cellular ATP levels. Conversely, knockdown of ACL in myotubes not only reduces mitochondrial complex I, IV, and V activity but also blocks IGF1-induced increases in oxygen consumption. In vivo, ACL activity is associated with increased ATP. Activation of this IGF1/ ACL/cardiolipin pathway combines anabolic signaling with induction of mechanisms needed to provide required ATP.
INTRODUCTION Dysregulation in mitochondrial oxidative phosphorylation (OXPHOS) as well as mitochondrial complex activity is associated with various skeletal muscle pathologies, including cachexia, sarcopenia, and the muscular dystrophies (Julienne et al., 2012; Marzetti et al., 2013; Szendroedi et al., 2008; Wang et al., 2010; White et al., 2011). Mitochondrial complexes are organized into several supercomplexes that facilitate the flow of electrons and thus make the system more efficient, minimizing ROS generation (Lapuente-Brun et al., 2013). It was also speculated that complex stability is influenced by the organization of supercomplexes. Recently, Ikeda et al. showed that mitochondrial activity is impaired and ATP content decreased by lack of supercomplex formation (Ikeda et al., 2013). Anabolic stimuli like exercise have a positive impact on skeletal muscle strength, endurance, and aerobic capacity by increasing protein synthesis, in addition to mitochondrial and contractile function of skeletal muscle (Egan and Zierath, 2013; Egerman and Glass, 2014; Riedl et al., 2010). The positive effects 868 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
of exercise extend to diseased populations and to the elderly (Fiatarone et al., 1990; Hansen et al., 2010; Jeppesen et al., 2006). Some of these effects are elicited in response to contractile activity and mediated by anabolic factors such as insulin growth factor 1 (IGF1) (Schiaffino et al., 2013). IGF1 is an anabolic growth factor that induces hypertrophy and blocks atrophy in skeletal muscle by activating the PI3K/ AKT/mTOR pathway (Egerman and Glass, 2014). In addition to its anabolic and anti-catabolic effects, IGF1 stimulates fatty acid (FA) uptake and glucose metabolism (Clemmons, 2012). Whereas both FAs and glucose can be used to generate ATP, glucose can also support de novo lipid synthesis via cytosolic citrate metabolism (Vander Heiden et al., 2009), which is an important step to meet the demand of membrane expansion during skeletal muscle growth. Cardiolipin is a phospholipid specific to the inner mitochondrial membrane, where it plays an important role in structural organization and function of these membranes (Paradies et al., 2014; Ren et al., 2014). ATP citrate lyase (ACL) is a cytosolic enzyme that catalyzes the conversion of mitochondria-derived citrate into oxaloacetate and acetyl-CoA, which are further utilized as building blocks for lipid synthesis and acetylation. Acetyl-CoA is vital for acetylation reactions that modify proteins including histones, affecting their functions (Lee et al., 2014; Wellen et al., 2009). In the current study, we investigated the role of ACL in regulating lipid metabolism and mitochondrial function in skeletal muscle and whether metabolic effects orchestrated by IGF1 are regulated by ACL.
RESULTS IGF1 Induces De Novo Lipid Synthesis through PI3K/AKT/ACL Conversion of glucose to lipids requires several steps, terminating with mitochondrial export of citrate into the cytosol. ACL then catalyzes conversion of cytosolic citrate to oxaloacetate and acetyl-CoA (Chypre et al., 2012). Phosphorylation of ACL at S455 was shown to be regulated by PI3K/AKT, and this phosphorylation has been reported to stimulate ACL enzymatic activity (Berwick et al., 2002; Potapova et al., 2000). Since IGF1 also activates the PI3K/AKT pathway, we asked if IGF1 could increase ACL phosphorylation. IGF1 stimulated ACL phosphorylation at S455 in a PI3K-dependent manner, since the PI3K inhibitor LY294002 significantly reduced it (Figure 1A). IGF1 also induced ACL phosphorylation on two different sites, T447/S451, and this
effect was also inhibited by LY294002 (Figure S1A). However, rapamycin treatment did not affect pACL S455 or T447/S451, suggesting that mTORC1 signaling was not required downstream of IGF1 treatment (Figure S1B). Overexpressing dominant-negative AKT (dnAKT) reduced ACL phosphorylation following IGF1 treatment (Figure 1B), indicating that AKT regulates ACL phosphorylation. To determine if AKT was sufficient to induce ACL phosphorylation, we overexpressed a constitutively active mutant, myristoylated AKT (myr-AKT), in primary human skeletal myotubes. Increased ACL phosphorylation induced by myr-AKT demonstrated that AKT activity is not only necessary (Figure 1C) but sufficient for phosphorylation of ACL at S455 and at T447/S451 (Figures 1C and S1C). To add physiological context to the in vitro data, the phosphorylation status of ACL was analyzed in muscles of aged mice. As shown in Figure 1D, the phosphorylation of ACL at T447/S451 was significantly downregulated in the tibialis (TA) muscle of 24month-old mice compared to younger animals. Furthermore, signaling upstream of ACL was also downregulated in old mice. Transcript levels of IGF1 were significantly decreased (Figure S1D), as were total protein levels of AKT and AKT phosphorylation on S473 in the tibialis muscle of aged mice (Figure 1D). IGF1 regulates glucose uptake via the PI3K/AKT pathway (Clemmons, 2012). To test if the glucose to lipid conversion in response to IGF1 was regulated by ACL, myotubes were incubated with C14-pyruvate, and the rate of lipid synthesis was analyzed. Similar to what has been demonstrated in other cell types (Hatzivassiliou et al., 2005), lipid synthesis was significantly downregulated (33%) after ACL knockdown (Figure 1E). The rate of pyruvate incorporation into lipid was significantly increased by IGF1 treatment; ACL knockdown and the PI3K inhibitor LY294002 reduced IGF1-mediated stimulation significantly (Figures 1E and 1F). Delta 6 desaturase expression, an important enzyme for the metabolism of polyunsaturated FAs (Kro¨ger and Schulze, 2012), was increased by both ACL overexpression and IGF1 treatment (Figure S1E). Collectively, the data show that the IGF1/PI3K/AKT pathway stimulates ACL phosphorylation and de novo lipid synthesis via ACL in skeletal muscle. Also, ACL phosphorylation is reduced in tibialis muscle of old versus young mice; this effect could be a consequence of decreased IGF1 signaling in aged mice. ACL Regulates Cardiolipin Synthesis and Content in Skeletal Muscle Since ACL activity is necessary for lipid synthesis, we asked whether ACL activity might also influence the synthesis and/or content of cardiolipin in skeletal muscle, thus regulating OXPHOS activity as well as mitochondrial complex activities. To examine the effect of ACL in vivo, adeno-associated viruses (AAVs) expressing wild-type ACL (AAV-ACL) or GFP (AAV-GFP) were injected into the TA muscles, and the expression of ACL was confirmed at the mRNA level (Figure S2A). In vivo expression of ACL for 4 weeks caused a significant increase of cardiolipin (37%) in the mitochondria (Figure 2A). The increase in cardiolipin induced by ACL was not accompanied by an increase in the expression of genes involved in cardiolipin synthesis such as Crls1, Taz, and Ptpmt1 (Figure 2B). However, when we used short hairpin RNA (AAV-shACL) to downregulate ACL
expression in vivo (Figure S2A), we observed a significant decrease of Crls1 and Taz but not of Ptpmt1 mRNA levels (Figure 2B). Given its positive effect on lipid synthesis, we tested whether IGF1 was able to perturb cardiolipin levels in mitochondria isolated from human myotubes treated with IGF1 for 30 hr. As shown in Figure 2C, IGF1 treatment increased cardiolipin content in a PI3K-dependent manner. Not surprisingly, insulin induced similar effects (Figure S2B). Together these results demonstrate that ACL expression is required to maintain cardiolipin-relevant gene expression, and activation of ACL in vivo can increase cardiolipin without perturbing gene expression. Interestingly, we found that cardiolipin content and transcript levels of Taz and Ptpmt1, enzymes that are required for cardiolipin formation, are reduced in TA of old compared to young mice (Figures 2D and 2E), whereas exercise resulted in a significant increase in cardiolipin content and expression of Crls1, Taz, and Ptpmt1 (Figures 2F and 2G). ACL Regulates Mitochondrial Supercomplex and Complex Levels and Activity Cardiolipin aids in binding the complexes together into a supercomplex (Mejia et al., 2014; Paradies et al., 2014; Ren et al., 2014). Since cardiolipin was significantly increased in tibialis (TA) muscle by ACL overexpression, we analyzed the content and activity of mitochondrial complexes as well as supercomplexes in the mitochondria isolated from TA muscle of mice infected with AAV-ACL or AAV-GFP. Coomassie staining revealed a significant increase in supercomplex formation containing complex I following overexpression of ACL in TA (Figures 3A and 3B). Interestingly, ACL overexpression also significantly increased the abundance of complex II, III, IV, and V (Figures 3A and 3C). These data support the model that supercomplex formation could result in stabilization of individual complexes as well. To support the results shown in Figures 3A–3C, we analyzed the protein levels of specific components of mitochondrial complexes by immunoblotting and observed significantly increased protein content for NDUFB8 (complex I) and SDHB (complex II) (Figure 3D). Complex I is known to assemble in supercomplexes containing complex III as well as complex IV, which stabilizes complex I as well as increases its activity. Accordingly, we observed significant increase in complex I activity as well as activity of supercomplex I+III2+IV, as assessed by in-gel activity assay performed after native PAGE of mitochondria obtained from AAV-ACL-infected TA (Figure 3E). Finally, ACL overexpression resulted in increased activity of complex II, IV, and V, but not of complex III (Figure 3F). Thus, increased complex content and composition accounts for the overall enhanced mitochondrial complex activity observed in the skeletal muscle overexpressing ACL. IGF1 Induces Mitochondrial Respiration through ACL Mitochondria generate ATP through the process of respiration and OXPHOS (Stephenson and Hawley, 2014). We found that IGF1 induces mitochondrial respiration in myotubes, as determined using a Seahorse XF-96 flux analyzer (Figure 4A). However, knocking down ACL significantly decreased OXPHOS activity (Figure 4A). From basal conditions, IGF1 induced oxygen consumption by 80%, whereas ACL downregulation not only decreased the oxygen consumption by 52% but also blocked Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 869
Figure 1. IGF1 Promotes De Novo Lipid Synthesis via Activation of ACL (A) Differentiated human skeletal muscle cells (HuSkMCs) were treated with vehicle or 30 nM IGF1 as indicated ± 1 mM LY294002. pACL (S455) and total ACL were analyzed. Graphs show densitometry of 3–5 independent experiments, and a representative blot is shown. ANOVA followed by Dunnett’s multiple comparison test was used. (legend continued on next page)
870 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
the effect of IGF1, indicating IGF1 was acting via ACL (Figure 4A). IGF1 increased maximal respiration by 85% and reserve capacity of the myotubes by 90%. Knocking down ACL not only led to a decrease of 53%, and 53% maximal respiration and reserve capacity, but also decreased the IGF1-induced effect by 59% and 60%, respectively (Figure 4A). In addition, IGF1 increased the coupling efficiency of the myotubes by 104%, whereas downregulation of ACL decreased the same by 90% and by 65% in the presence of IGF1 (Figure 4A). In order to understand the mechanism by which ACL enhances mitochondrial respiration, we analyzed mitochondrial number in myotubes after knocking down or overexpressing ACL. There was no significant change in mitochondrial DNA after normalizing to nuclear DNA (Figure S3A). Thus, the altered mitochondrial function of myotubes by ACL is not a consequence of modified mitochondrial number. Next, we analyzed individual mitochondrial complex activities in isolated mitochondria from myotubes after overexpressing or downregulating ACL. Overexpressing ACL induced an increase of 67%, 38%, and 93% in complex I, IV, and V, respectively, whereas knockdown of ACL led to a decrease of 42%, 25%, and 70% in complex I, IV, and V, respectively (Figure 4B). Complex II activities showed similar trends; however, the differences failed to reach significance in three independent experiments (Figure 4B). Because mitochondria produce reactive oxygen species (ROS) as a natural byproduct of electron transport chain activity, and supercomplexes control ROS generation by the respiratory chain (Sullivan and Chandel, 2014), we asked if ACL overexpression perturbs ROS generation. As shown in Figure S3B, ACL overexpression in myotubes did not affect ROS. In addition to increased mitochondrial OXPHOS activity, cellular ATP content was increased by 43% by IGF1 treatment (Figure 4C). Altered OXPHOS activity and mitochondrial complex activity by ACL knockdown or overexpression resulted in 35% decrease or 75% increase in cellular ATP content, respectively (Figure 4C), also demonstrating that IGF1-mediated increases in ATP require ACL activity. In order to verify the significance of ACL activity in maintaining cellular ATP content in vivo, we overexpressed ACL in the TA of mice (Figure S2A). In accord with the in vitro findings, overexpression of ACL increased cellular ATP content by 30% (Figure 4D). Because IGF1 is a strong anabolic stimulus for skeletal muscle (Egerman and Glass, 2014), we knocked down ACL by siRNA in human primary myotubes, subjected myotubes to IGF1 treatment, and measured protein synthesis. As shown in Figure S4A, ACL silencing significantly reduced the increase in protein synthesis and in MHC protein levels mediated by IGF1 without perturbing basal levels. In addition, ACL overexpression in human primary myotubes was suffi-
cient to increase myotube size and MHC protein levels (Figure S4B). Interestingly, ACL phosphorylation on T447/S451 and total ACL levels were decreased in the quadriceps of mice displaying prominent atrophy induced by dexamethasone treatment for 7 days (Figure S4C). DISCUSSION ACL has been appreciated for its role in fatty acid (FA) biosynthesis and its requirement in the production of acetyl-CoA (Wellen et al., 2009; Choudhary et al., 2014). However, ACL’s role in other metabolic processes, particularly in skeletal muscle, has not been extensively studied. Recently, Lee et al. reported that AKT phosphorylates and increases the activity of ACL via a PI3K-dependent pathway in cancer cells (Lee et al., 2014). This finding gave context to our observation that IGF1 was capable of increasing phospho-ACL levels and de novo lipid synthesis in skeletal muscle in an ACL-dependent manner. Further exploration demonstrated that ACL was capable of increasing levels of cardiolipin, a critical component of the mitochondrial membrane that constitutes about 20% of the organelle’s FA composition (Paradies et al., 2014; Ren et al., 2014), opening up a new arena for ACLrequired biology. ACL was found to be required for maintenance of cardiolipin, which in turn modulated mitochondrial complex formation. ACL activity is thus limiting for this critical process— critical because complex formation modulates ATP levels. It was previously demonstrated that IGF1 is an anabolic agent for skeletal muscle; it induces protein synthesis via the AKT/ mTOR pathway and blocks protein breakdown by inhibiting FOXO induction of the E3 ligases MuRF1 and MAFbx, which mediate muscle protein degradation (Egerman and Glass, 2014). What hadn’t been understood previously was that IGF1 could also perturb mitochondrial metabolism, as shown in this study. It seems especially noteworthy that a growth factor that induces protein synthesis is also able to induce the required ATP necessary for the production of new proteins, as well as to facilitate correct protein folding and subsequent protein function. In contrast, in many settings mitochondrial stimulation has seemed to be at odds with anabolic protein synthesis. For example, PGC1a, a transcriptional co-activator that stimulates mitochondriogenesis, is itself induced by AMPK activation (Lee et al., 2006), which at the same time actually inhibits mTOR signaling. Also, anabolic stimuli by mechanisms such as myostatin inhibition can induce hypertrophy without increasing mitochondrial function, and this sort of effect in part might explain why some anabolic stimuli result in more fatigable muscle (LeBrasseur et al., 2009; Mouisel et al., 2014). However, this study gives at least one physiologic mechanism in which both
(B) Myotubes were infected with AdEV or Ad dn-AKT. Forty-eight hours post-infection, cells were treated with 30 nM IGF1 for 30 min. pACL (S455) and total ACL and AKT were analyzed. (C) Myotubes were infected with AdEV or Ad myr-AKT. pACL (S455) and total ACL and AKT were analyzed 24, 48, and 72 hr post-infection by immunoblotting. In (B) and (C), a blot representative of three independent experiments is shown. Graphs show densitometric analysis of three independent experiments; unpaired t test. (D) pACL (T447/S451), pAKT (S473 and T308), and total AKT levels were analyzed in tibialis muscle of young and old mice by immunoblotting. A Mann-Whitney test was used; n = 8 mice/group. (E and F) Differentiated myotubes were subjected to siRNA to ACL and treated with vehicle or 30 nM IGF1 for 24 hr with or without 3 mM LY294002 (F) concurrently with 14C-Na-pyruvate during the last 5 hr of the 24 hr culture, 3 days post-transfection. 14C counts were determined in extracted lipids and normalized to protein concentration. Protein extracts were analyzed for ACL expression 4 days post-transfection. (E) Data are expressed as % change of control. Shown are means ± SEM; n = 3–5; ANOVA followed by Dunnett’s multiple comparison test was used. (F) Shown are mean ± SEM; ANOVA followed by Tukey’s multiple comparison test was used. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 versus control; $ p < 0.05 versus IGF1.
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 871
Figure 2. ACL Regulates Cardiolipin Content in Skeletal Muscle (A and B) AAVs expressing ACL (AAV-ACL; A and B), GFP (AAV-GFP; A and B), AAV-shCtrl (B), or AAV-shACL (B) were injected into the tibialis. (A) Cardiolipin content was analyzed by HPTLC 30 days post-injection. Bands were quantified using ImageJ; shown are means ± SEM; n = 6–8 mice/group; Mann-Whitney test. (B) Transcripts of Crls1, Taz, and Ptpmt1 were analyzed by qRTPCR; data are expressed as fold increase of AAVGFP (n = 6–8 mice/group); Wilcoxon test. (C) Myotubes treated with vehicle or 30 nM IGF1 for 30 hr ± 3 mM LY294002; cardiolipin content was analyzed by HPTLC. Bands were quantified using ImageJ. Data are expressed as means ± SEM of four independent experiments; ANOVA followed by Tukey’s multiple comparison test was used. (D–G) Cardiolipin content (D and F) and mRNA expression of Taz, Ptpmt1, and Crls1 (E and G) were analyzed by HPTLC and qRT-PCR, respectively, in tibialis of young and old (D and E) and young exercised or sedentary (F and G) mice. (D and F) Bands were quantified using ImageJ. Shown are means ± SEM; n = 8 mice/group; MannWhitney test. In (E) and (G), data are expressed as fold increase of young or young/sedentary, respectively. (E) n = 5–6 mice/group; (G) n = 8 mice/ group; Mann-Whitney test, *p < 0.05, **p < 0.01, ***p < 0.001.
anabolic stimuli and mitochondria might be stimulated—not by increasing mitochondrial number, but by increasing mitochondrial efficiency. 872 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
When skeletal muscle ages, there is an increase in frailty, which is coincident with a decrease in muscle mass and function—this condition is called sarcopenia (Cruz-Jentoft et al., 2010). It has been shown that one of the major pathways downregulated at the onset of sarcopenia is the mitochondrial pathway, such as those genes involved in OXPHOS and the TCA cycle (Ibebunjo et al., 2013; Marzetti et al., 2013). A decline in concentrations of circulating IGF1 in sarcopenia has been described in humans and rodents (O’Connor et al., 1998; Velasco et al., 1998). This study suggests the possibility that re-activation of ACL in settings of sarcopenia might be helpful in restoring mitochondrial competence. Of interest, in this study it was observed that phosphorylated ACL is dramatically reduced in old (24 months) versus young mice, coincident with a decrease in levels of AKT phosphorylation with age. Therefore, the goal would be to reactivate ACL in settings of sarcopenia, perhaps by restoring IGF1 signaling to physiologic levels in this setting. It remains to be seen whether genetic manipulation of ACL in old mice affects mitochondrial and skeletal muscle function. Homozygous ACL knockout mice die early in development (Beigneux et al., 2004); therefore, the
Figure 3. ACL Increases Mitochondrial Complex Content and Activity (A–C) Mitochondrial proteins extracted from tibialis injected with AAV-GFP or AAV-ACL 30 days post-infection, analyzed for mitochondrial complex and supercomplex content by Coomassie native PAGE. (D) Levels of respiratory chain subunits analyzed by immunoblotting (D). (E and F) In-gel activity assay was carried out for complex I (E) and complex II–V (F). Graphs show densitometric analysis and data are shown as means ± SEM; *p < 0.05, **p < 0.01; n = 6–7, Mann-Whitney test.
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 873
Figure 4. IGF1 Increases Oxygen Consumption and ATP Content via ACL (A and B) Myotubes given siRNA to ACL (A and B) or infected with AdEV or AdACL (B). (A) Three days post-transfection, cells were treated for 24 hr with vehicle or 30 nM IGF1. Oxygen consumption rates (OCR) ± 1.6 mM oligomycin, 0.5 mM fluoro-carbonyl cyanide phenylhydrazone (FCCP), or 1 mM antimycin + 1.5 mM rotenone were determined using XFe96 extracellular flux analyzer. Graphs show data from three independent experiments; results are expressed as means ± SEM of the area under the curve (AUC). (B) Three days post-transfection or 2 days post-infection, mitochondria were isolated and complex activity was measured by ELISA. Activity is expressed as change in absorbance per minute (mOD/min) per mg of cell lysate. Data are shown as means ± SEM; n = 3. (C) Human skeletal muscle myotubes were transfected with siCtrl or siACL as described in (A) and (B). Three days post-transfection, myotubes were treated with 30 nM IGF1 for 36 hr, and cellular ATP was analyzed. In parallel, myotubes were transduced with AdEV or AdACL and analyzed for ATP concentration 48 hr postinfection. Shown are means ± SEM; n = 3. (D) ATP concentration determined in tibialis of AAV-GFP- or AAV-ACL-injected mice. Shown are means ± SEM; ANOVA followed by Tukey’s multiple comparison test (A–C), Mann-Whitney test (D); n = 6–7 mice per group, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.
generation of a skeletal muscle-specific knockout, or perhaps transgenic mice expressing activated ACL, would be needed to address this question. The finding here that ACL induced cardiolipin production, which has been shown to be required for regulated ATP production, provides an important new node in the understanding of energy metabolism in skeletal muscle. This study opens up future work aimed at understanding distinct signaling steps that might be able to induce the required phosphorylation of ACL, monitoring cardiolipin levels as a biomarker of this ACL/cardiolipin pathway. IGF1/AKT/ACL/cardiolipin/ATP constitutes an important signaling pathway that is apparently critical for providing the energy to match the anabolism of IGF1-induced protein synthesis. 874 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
EXPERIMENTAL PROCEDURES De Novo Lipid Synthesis Assay Primary human skeletal muscle myotubes were incubated with 14C-Na-pyruvate (2 mCi/ml) (PerkinElmer) for 5 hr. The cells were washed four times with cold PBS and lysed in cold PBS by ultrasound. Lipids were extracted by chloroform:methanol (1:1). Radioactivity was measured using scintillation cocktail (UltimaGold-XR, PerkinElmer). The values were normalized by protein concentration determined by BCA (Thermo Scientific). High-Performance Thin-Layer Chromatography The extracted lipids were dissolved in chloroform:methanol (2:1) and analyzed by HPTLC LiChrospher silica gel 60F254S (Millipore) using chloroform:acetone:methanol:acetic acid:water (6:8:2:2:1) as mobile phase. Cardiolipin sodium salt (Sigma) was dissolved in chloroform:methanol (2:1) and used as cardiolipin standard/marker. The HPTLC plate was derivatized by spraying
0.18% copper sulfate in orthophosphoric acid:water (40:460) and heating at 200 C for 20 min. Lipid bands were quantified by ImageJ. Mitochondria Isolation Mitochondria were isolated either from primary human myotubes or tibialis muscle using Dounce homogenizer with mitochondria isolation buffer (70 mM sucrose, 220 mM mannitol, 5 mM KH2PO4, 5 mM MgCl2, 2 mM HEPES, 1 mM EGTA [pH 7.2], and 0.2% FA-free BSA) supplemented with Halt protease and phosphatase inhibitors, trichostatin, and nicotinamide at 4 C. Cell or tissue homogenates were centrifuged at 1,0003 g for 10 min at 4 C. The supernatants were subsequently centrifuged at 12,0003 g for 20 min at 4 C, and pellets were washed in mitochondria isolation buffer and stored at 80 C until further processing. Statistical Analysis Statistical significance was evaluated with one-way analysis of variance (ANOVA) for multiple groups, followed by Bonferroni’s or Tukey’s post hoc test to evaluate differences among groups. Mann-Whitney-Wilcoxon test or unpaired t test was used to compare two groups. All error bars indicate SEM. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, and one table and can be found with this article online at http:// dx.doi.org/10.1016/j.cmet.2015.05.006. AUTHOR CONTRIBUTIONS S.D. and M.F. conceived and designed the experiments and analyzed the data; F.M., G.T., and P.B. contributed to the design, execution, and analysis of animal experiments; B.J., V.M., and P.K. performed the in vitro experiments; M.F. supervised the project; D.J.G. contributed to the design of the experiments; S.D., M.F., and D.J.G. co-wrote the manuscript. S.D. is the recipient of a NIBR postdoctoral fellowship. ACKNOWLEDGMENTS We thank the Muscle Diseases group at the Novartis Institutes for Biomedical Research (NIBR) and Dr. J. Auwerx (Ecole Polytechnique Federale de Lausanne) for their support. All authors are employees of the Novartis Institutes of Biomedical Research. We would like to thank Mark Fishman and Michaela Kneissel for their support and the Novartis Postdoctoral program for its support of author S.D. Received: November 4, 2014 Revised: March 12, 2015 Accepted: April 29, 2015 Published: June 2, 2015 REFERENCES Beigneux, A.P., Kosinski, C., Gavino, B., Horton, J.D., Skarnes, W.C., and Young, S.G. (2004). ATP-citrate lyase deficiency in the mouse. J. Biol. Chem. 279, 9557–9564. Berwick, D.C., Hers, I., Heesom, K.J., Moule, S.K., and Tavare, J.M. (2002). The identification of ATP-citrate lyase as a protein kinase B (Akt) substrate in primary adipocytes. J. Biol. Chem. 277, 33895–33900. Choudhary, C., Weinert, B.T., Nishida, Y., Verdin, E., and Mann, M. (2014). The growing landscape of lysine acetylation links metabolism and cell signalling. Nat. Rev. Mol. Cell Biol. 15, 536–550.
Cruz-Jentoft, A.J., Baeyens, J.P., Bauer, J.M., Boirie, Y., Cederholm, T., Landi, F., Martin, F.C., Michel, J.P., Rolland, Y., Schneider, S.M., et al.; European Working Group on Sarcopenia in Older People (2010). Sarcopenia: European consensus on definition and diagnosis: Report of the European Working Group on Sarcopenia in Older People. Age Ageing 39, 412–423. Egan, B., and Zierath, J.R. (2013). Exercise metabolism and the molecular regulation of skeletal muscle adaptation. Cell Metab. 17, 162–184. Egerman, M.A., and Glass, D.J. (2014). Signaling pathways controlling skeletal muscle mass. Crit. Rev. Biochem. Mol. Biol. 49, 59–68. Fiatarone, M.A., Marks, E.C., Ryan, N.D., Meredith, C.N., Lipsitz, L.A., and Evans, W.J. (1990). High-intensity strength training in nonagenarians. Effects on skeletal muscle. JAMA 263, 3029–3034. Hansen, D., Dendale, P., van Loon, L.J., and Meeusen, R. (2010). The impact of training modalities on the clinical benefits of exercise intervention in patients with cardiovascular disease risk or type 2 diabetes mellitus. Sports Med. 40, 921–940. Hatzivassiliou, G., Zhao, F., Bauer, D.E., Andreadis, C., Shaw, A.N., Dhanak, D., Hingorani, S.R., Tuveson, D.A., and Thompson, C.B. (2005). ATP citrate lyase inhibition can suppress tumor cell growth. Cancer Cell 8, 311–321. Ibebunjo, C., Chick, J.M., Kendall, T., Eash, J.K., Li, C., Zhang, Y., Vickers, C., Wu, Z., Clarke, B.A., Shi, J., et al. (2013). Genomic and proteomic profiling reveals reduced mitochondrial function and disruption of the neuromuscular junction driving rat sarcopenia. Mol. Cell. Biol. 33, 194–212. Ikeda, K., Shiba, S., Horie-Inoue, K., Shimokata, K., and Inoue, S. (2013). A stabilizing factor for mitochondrial respiratory supercomplex assembly regulates energy metabolism in muscle. Nat. Commun. 4, 2147. Jeppesen, T.D., Schwartz, M., Olsen, D.B., Wibrand, F., Krag, T., Dunø, M., Hauerslev, S., and Vissing, J. (2006). Aerobic training is safe and improves exercise capacity in patients with mitochondrial myopathy. Brain 129, 3402– 3412. Julienne, C.M., Dumas, J.F., Goupille, C., Pinault, M., Berri, C., Collin, A., Tesseraud, S., Couet, C., and Servais, S. (2012). Cancer cachexia is associated with a decrease in skeletal muscle mitochondrial oxidative capacities without alteration of ATP production efficiency. J Cachexia Sarcopenia Muscle 3, 265–275. Kro¨ger, J., and Schulze, M.B. (2012). Recent insights into the relation of D5 desaturase and D6 desaturase activity to the development of type 2 diabetes. Curr. Opin. Lipidol. 23, 4–10. Lapuente-Brun, E., Moreno-Loshuertos, R., Acı´n-Pe´rez, R., Latorre-Pellicer, A., Cola´s, C., Balsa, E., Perales-Clemente, E., Quiro´s, P.M., Calvo, E., Rodrı´guez-Herna´ndez, M.A., et al. (2013). Supercomplex assembly determines electron flux in the mitochondrial electron transport chain. Science 340, 1567–1570. LeBrasseur, N.K., Schelhorn, T.M., Bernardo, B.L., Cosgrove, P.G., Loria, P.M., and Brown, T.A. (2009). Myostatin inhibition enhances the effects of exercise on performance and metabolic outcomes in aged mice. J. Gerontol. A Biol. Sci. Med. Sci. 64, 940–948. Lee, W.J., Kim, M., Park, H.S., Kim, H.S., Jeon, M.J., Oh, K.S., Koh, E.H., Won, J.C., Kim, M.S., Oh, G.T., et al. (2006). AMPK activation increases fatty acid oxidation in skeletal muscle by activating PPARalpha and PGC-1. Biochem. Biophys. Res. Commun. 340, 291–295. Lee, J.V., Carrer, A., Shah, S., Snyder, N.W., Wei, S., Venneti, S., Worth, A.J., Yuan, Z.F., Lim, H.W., Liu, S., et al. (2014). Akt-dependent metabolic reprogramming regulates tumor cell histone acetylation. Cell Metab. 20, 306–319. Marzetti, E., Calvani, R., Cesari, M., Buford, T.W., Lorenzi, M., Behnke, B.J., and Leeuwenburgh, C. (2013). Mitochondrial dysfunction and sarcopenia of aging: from signaling pathways to clinical trials. Int. J. Biochem. Cell Biol. 45, 2288–2301.
Chypre, M., Zaidi, N., and Smans, K. (2012). ATP-citrate lyase: a mini-review. Biochem. Biophys. Res. Commun. 422, 1–4.
Mejia, E.M., Cole, L.K., and Hatch, G.M. (2014). Cardiolipin metabolism and the role it plays in heart failure and mitochondrial supercomplex formation. Cardiovasc. Hematol. Disord. Drug Targets 14, 98–106.
Clemmons, D.R. (2012). Metabolic actions of insulin-like growth factor-I in normal physiology and diabetes. Endocrinol. Metab. Clin. North Am. 41, 425–443, vii–viii.
Mouisel, E., Relizani, K., Mille-Hamard, L., Denis, R., Hourde´, C., Agbulut, O., Patel, K., Arandel, L., Morales-Gonzalez, S., Vignaud, A., et al. (2014). Myostatin is a key mediator between energy metabolism and endurance
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 875
capacity of skeletal muscle. Am. J. Physiol. Regul. Integr. Comp. Physiol. 307, R444–R454. O’Connor, K.G., Tobin, J.D., Harman, S.M., Plato, C.C., Roy, T.A., Sherman, S.S., and Blackman, M.R. (1998). Serum levels of insulin-like growth factor-I are related to age and not to body composition in healthy women and men. J. Gerontol. A Biol. Sci. Med. Sci. 53, M176–M182. Paradies, G., Paradies, V., De Benedictis, V., Ruggiero, F.M., and Petrosillo, G. (2014). Functional role of cardiolipin in mitochondrial bioenergetics. Biochim. Biophys. Acta 1837, 408–417. Potapova, I.A., El-Maghrabi, M.R., Doronin, S.V., and Benjamin, W.B. (2000). Phosphorylation of recombinant human ATP:citrate lyase by cAMP-dependent protein kinase abolishes homotropic allosteric regulation of the enzyme by citrate and increases the enzyme activity. Allosteric activation of ATP:citrate lyase by phosphorylated sugars. Biochemistry 39, 1169–1179. Ren, M., Phoon, C.K., and Schlame, M. (2014). Metabolism and function of mitochondrial cardiolipin. Prog. Lipid Res. 55, 1–16. Riedl, I., Yoshioka, M., Nishida, Y., Tobina, T., Paradis, R., Shono, N., Tanaka, H., and St-Amand, J. (2010). Regulation of skeletal muscle transcriptome in elderly men after 6 weeks of endurance training at lactate threshold intensity. Exp. Gerontol. 45, 896–903. Schiaffino, S., Dyar, K.A., Ciciliot, S., Blaauw, B., and Sandri, M. (2013). Mechanisms regulating skeletal muscle growth and atrophy. FEBS J. 280, 4294–4314.
876 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
Stephenson, E.J., and Hawley, J.A. (2014). Mitochondrial function in metabolic health: a genetic and environmental tug of war. Biochim. Biophys. Acta 1840, 1285–1294. Sullivan, L.B., and Chandel, N.S. (2014). Mitochondrial reactive oxygen species and cancer. Cancer Metab 2, 17. Szendroedi, J., Zwettler, E., Schmid, A.I., Chmelik, M., Pacini, G., Kacerovsky, G., Smekal, G., Nowotny, P., Wagner, O., Schnack, C., et al. (2008). Reduced basal ATP synthetic flux of skeletal muscle in patients with previous acromegaly. PLoS ONE 3, e3958. Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009). Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029–1033. Velasco, B., Cacicedo, L., Escalada, J., Lopez-Fernandez, J., and SanchezFranco, F. (1998). Growth hormone gene expression and secretion in aging rats is age dependent and not age-associated weight increase related. Endocrinology 139, 1314–1320. Wang, M., Wang, X.C., Zhang, Z.Y., Mou, B., and Hu, R.M. (2010). Impaired mitochondrial oxidative phosphorylation in multiple insulin-sensitive tissues of humans with type 2 diabetes mellitus. J. Int. Med. Res. 38, 769–781. Wellen, K.E., Hatzivassiliou, G., Sachdeva, U.M., Bui, T.V., Cross, J.R., and Thompson, C.B. (2009). ATP-citrate lyase links cellular metabolism to histone acetylation. Science 324, 1076–1080. White, J.P., Baltgalvis, K.A., Puppa, M.J., Sato, S., Baynes, J.W., and Carson, J.A. (2011). Muscle oxidative capacity during IL-6-dependent cancer cachexia. Am. J. Physiol. Regul. Integr. Comp. Physiol. 300, R201–R211.
ATP Citrate Lyase Improves Mitochondrial Function in Skeletal Muscle Graphical Abstract
Authors Suman Das, Frederic Morvan, ..., David J. Glass, Mara Fornaro
Correspondence [email protected] (D.J.G.), [email protected] (M.F.)
In Brief Mitochondrial function is an important determinant of skeletal muscle homeostasis. Das et al. report that activation of ATP citrate lyase (ACL) by IGF1 in skeletal muscle increases cardiolipin content and mitochondrial complex activity and improves mitochondrial function, revealing how the IGF1/ACL pathway combines anabolic signaling with energy production.
Highlights d
ACL regulates mitochondrial function in skeletal muscle
d
IGF1 induces activation of ACL in an AKT-dependent fashion
d
ACL phosphorylation is impaired in skeletal muscle of sarcopenic mice
d
ACL activity is associated with increased oxygen consumption and ATP levels
Das et al., 2015, Cell Metabolism 21, 868–876 June 2, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.cmet.2015.05.006
Cell Metabolism
Short Article ATP Citrate Lyase Improves Mitochondrial Function in Skeletal Muscle Suman Das,1 Frederic Morvan,1 Benjamin Jourde,1 Viktor Meier,1 Peter Kahle,1 Pascale Brebbia,1 Gauthier Toussaint,1 David J. Glass,2,* and Mara Fornaro1,* 1Novartis
Institutes for Biomedical Research, Forum 1, Novartis Campus, 4056 Basel, Switzerland Institutes for Biomedical Research, 100 Technology Square, Cambridge, MA 02139, USA *Correspondence: [email protected] (D.J.G.), [email protected] (M.F.) http://dx.doi.org/10.1016/j.cmet.2015.05.006 2Novartis
SUMMARY
Mitochondrial dysfunction is associated with skeletal muscle pathology, including cachexia, sarcopenia, and the muscular dystrophies. ATP citrate lyase (ACL) is a cytosolic enzyme that catalyzes mitochondria-derived citrate into oxaloacetate and acetylCoA. Here we report that activation of ACL in skeletal muscle results in improved mitochondrial function. IGF1 induces activation of ACL in an AKT-dependent fashion. This results in an increase in cardiolipin, thus increasing critical mitochondrial complexes and supercomplex activity, and a resultant increase in oxygen consumption and cellular ATP levels. Conversely, knockdown of ACL in myotubes not only reduces mitochondrial complex I, IV, and V activity but also blocks IGF1-induced increases in oxygen consumption. In vivo, ACL activity is associated with increased ATP. Activation of this IGF1/ ACL/cardiolipin pathway combines anabolic signaling with induction of mechanisms needed to provide required ATP.
INTRODUCTION Dysregulation in mitochondrial oxidative phosphorylation (OXPHOS) as well as mitochondrial complex activity is associated with various skeletal muscle pathologies, including cachexia, sarcopenia, and the muscular dystrophies (Julienne et al., 2012; Marzetti et al., 2013; Szendroedi et al., 2008; Wang et al., 2010; White et al., 2011). Mitochondrial complexes are organized into several supercomplexes that facilitate the flow of electrons and thus make the system more efficient, minimizing ROS generation (Lapuente-Brun et al., 2013). It was also speculated that complex stability is influenced by the organization of supercomplexes. Recently, Ikeda et al. showed that mitochondrial activity is impaired and ATP content decreased by lack of supercomplex formation (Ikeda et al., 2013). Anabolic stimuli like exercise have a positive impact on skeletal muscle strength, endurance, and aerobic capacity by increasing protein synthesis, in addition to mitochondrial and contractile function of skeletal muscle (Egan and Zierath, 2013; Egerman and Glass, 2014; Riedl et al., 2010). The positive effects 868 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
of exercise extend to diseased populations and to the elderly (Fiatarone et al., 1990; Hansen et al., 2010; Jeppesen et al., 2006). Some of these effects are elicited in response to contractile activity and mediated by anabolic factors such as insulin growth factor 1 (IGF1) (Schiaffino et al., 2013). IGF1 is an anabolic growth factor that induces hypertrophy and blocks atrophy in skeletal muscle by activating the PI3K/ AKT/mTOR pathway (Egerman and Glass, 2014). In addition to its anabolic and anti-catabolic effects, IGF1 stimulates fatty acid (FA) uptake and glucose metabolism (Clemmons, 2012). Whereas both FAs and glucose can be used to generate ATP, glucose can also support de novo lipid synthesis via cytosolic citrate metabolism (Vander Heiden et al., 2009), which is an important step to meet the demand of membrane expansion during skeletal muscle growth. Cardiolipin is a phospholipid specific to the inner mitochondrial membrane, where it plays an important role in structural organization and function of these membranes (Paradies et al., 2014; Ren et al., 2014). ATP citrate lyase (ACL) is a cytosolic enzyme that catalyzes the conversion of mitochondria-derived citrate into oxaloacetate and acetyl-CoA, which are further utilized as building blocks for lipid synthesis and acetylation. Acetyl-CoA is vital for acetylation reactions that modify proteins including histones, affecting their functions (Lee et al., 2014; Wellen et al., 2009). In the current study, we investigated the role of ACL in regulating lipid metabolism and mitochondrial function in skeletal muscle and whether metabolic effects orchestrated by IGF1 are regulated by ACL.
RESULTS IGF1 Induces De Novo Lipid Synthesis through PI3K/AKT/ACL Conversion of glucose to lipids requires several steps, terminating with mitochondrial export of citrate into the cytosol. ACL then catalyzes conversion of cytosolic citrate to oxaloacetate and acetyl-CoA (Chypre et al., 2012). Phosphorylation of ACL at S455 was shown to be regulated by PI3K/AKT, and this phosphorylation has been reported to stimulate ACL enzymatic activity (Berwick et al., 2002; Potapova et al., 2000). Since IGF1 also activates the PI3K/AKT pathway, we asked if IGF1 could increase ACL phosphorylation. IGF1 stimulated ACL phosphorylation at S455 in a PI3K-dependent manner, since the PI3K inhibitor LY294002 significantly reduced it (Figure 1A). IGF1 also induced ACL phosphorylation on two different sites, T447/S451, and this
effect was also inhibited by LY294002 (Figure S1A). However, rapamycin treatment did not affect pACL S455 or T447/S451, suggesting that mTORC1 signaling was not required downstream of IGF1 treatment (Figure S1B). Overexpressing dominant-negative AKT (dnAKT) reduced ACL phosphorylation following IGF1 treatment (Figure 1B), indicating that AKT regulates ACL phosphorylation. To determine if AKT was sufficient to induce ACL phosphorylation, we overexpressed a constitutively active mutant, myristoylated AKT (myr-AKT), in primary human skeletal myotubes. Increased ACL phosphorylation induced by myr-AKT demonstrated that AKT activity is not only necessary (Figure 1C) but sufficient for phosphorylation of ACL at S455 and at T447/S451 (Figures 1C and S1C). To add physiological context to the in vitro data, the phosphorylation status of ACL was analyzed in muscles of aged mice. As shown in Figure 1D, the phosphorylation of ACL at T447/S451 was significantly downregulated in the tibialis (TA) muscle of 24month-old mice compared to younger animals. Furthermore, signaling upstream of ACL was also downregulated in old mice. Transcript levels of IGF1 were significantly decreased (Figure S1D), as were total protein levels of AKT and AKT phosphorylation on S473 in the tibialis muscle of aged mice (Figure 1D). IGF1 regulates glucose uptake via the PI3K/AKT pathway (Clemmons, 2012). To test if the glucose to lipid conversion in response to IGF1 was regulated by ACL, myotubes were incubated with C14-pyruvate, and the rate of lipid synthesis was analyzed. Similar to what has been demonstrated in other cell types (Hatzivassiliou et al., 2005), lipid synthesis was significantly downregulated (33%) after ACL knockdown (Figure 1E). The rate of pyruvate incorporation into lipid was significantly increased by IGF1 treatment; ACL knockdown and the PI3K inhibitor LY294002 reduced IGF1-mediated stimulation significantly (Figures 1E and 1F). Delta 6 desaturase expression, an important enzyme for the metabolism of polyunsaturated FAs (Kro¨ger and Schulze, 2012), was increased by both ACL overexpression and IGF1 treatment (Figure S1E). Collectively, the data show that the IGF1/PI3K/AKT pathway stimulates ACL phosphorylation and de novo lipid synthesis via ACL in skeletal muscle. Also, ACL phosphorylation is reduced in tibialis muscle of old versus young mice; this effect could be a consequence of decreased IGF1 signaling in aged mice. ACL Regulates Cardiolipin Synthesis and Content in Skeletal Muscle Since ACL activity is necessary for lipid synthesis, we asked whether ACL activity might also influence the synthesis and/or content of cardiolipin in skeletal muscle, thus regulating OXPHOS activity as well as mitochondrial complex activities. To examine the effect of ACL in vivo, adeno-associated viruses (AAVs) expressing wild-type ACL (AAV-ACL) or GFP (AAV-GFP) were injected into the TA muscles, and the expression of ACL was confirmed at the mRNA level (Figure S2A). In vivo expression of ACL for 4 weeks caused a significant increase of cardiolipin (37%) in the mitochondria (Figure 2A). The increase in cardiolipin induced by ACL was not accompanied by an increase in the expression of genes involved in cardiolipin synthesis such as Crls1, Taz, and Ptpmt1 (Figure 2B). However, when we used short hairpin RNA (AAV-shACL) to downregulate ACL
expression in vivo (Figure S2A), we observed a significant decrease of Crls1 and Taz but not of Ptpmt1 mRNA levels (Figure 2B). Given its positive effect on lipid synthesis, we tested whether IGF1 was able to perturb cardiolipin levels in mitochondria isolated from human myotubes treated with IGF1 for 30 hr. As shown in Figure 2C, IGF1 treatment increased cardiolipin content in a PI3K-dependent manner. Not surprisingly, insulin induced similar effects (Figure S2B). Together these results demonstrate that ACL expression is required to maintain cardiolipin-relevant gene expression, and activation of ACL in vivo can increase cardiolipin without perturbing gene expression. Interestingly, we found that cardiolipin content and transcript levels of Taz and Ptpmt1, enzymes that are required for cardiolipin formation, are reduced in TA of old compared to young mice (Figures 2D and 2E), whereas exercise resulted in a significant increase in cardiolipin content and expression of Crls1, Taz, and Ptpmt1 (Figures 2F and 2G). ACL Regulates Mitochondrial Supercomplex and Complex Levels and Activity Cardiolipin aids in binding the complexes together into a supercomplex (Mejia et al., 2014; Paradies et al., 2014; Ren et al., 2014). Since cardiolipin was significantly increased in tibialis (TA) muscle by ACL overexpression, we analyzed the content and activity of mitochondrial complexes as well as supercomplexes in the mitochondria isolated from TA muscle of mice infected with AAV-ACL or AAV-GFP. Coomassie staining revealed a significant increase in supercomplex formation containing complex I following overexpression of ACL in TA (Figures 3A and 3B). Interestingly, ACL overexpression also significantly increased the abundance of complex II, III, IV, and V (Figures 3A and 3C). These data support the model that supercomplex formation could result in stabilization of individual complexes as well. To support the results shown in Figures 3A–3C, we analyzed the protein levels of specific components of mitochondrial complexes by immunoblotting and observed significantly increased protein content for NDUFB8 (complex I) and SDHB (complex II) (Figure 3D). Complex I is known to assemble in supercomplexes containing complex III as well as complex IV, which stabilizes complex I as well as increases its activity. Accordingly, we observed significant increase in complex I activity as well as activity of supercomplex I+III2+IV, as assessed by in-gel activity assay performed after native PAGE of mitochondria obtained from AAV-ACL-infected TA (Figure 3E). Finally, ACL overexpression resulted in increased activity of complex II, IV, and V, but not of complex III (Figure 3F). Thus, increased complex content and composition accounts for the overall enhanced mitochondrial complex activity observed in the skeletal muscle overexpressing ACL. IGF1 Induces Mitochondrial Respiration through ACL Mitochondria generate ATP through the process of respiration and OXPHOS (Stephenson and Hawley, 2014). We found that IGF1 induces mitochondrial respiration in myotubes, as determined using a Seahorse XF-96 flux analyzer (Figure 4A). However, knocking down ACL significantly decreased OXPHOS activity (Figure 4A). From basal conditions, IGF1 induced oxygen consumption by 80%, whereas ACL downregulation not only decreased the oxygen consumption by 52% but also blocked Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 869
Figure 1. IGF1 Promotes De Novo Lipid Synthesis via Activation of ACL (A) Differentiated human skeletal muscle cells (HuSkMCs) were treated with vehicle or 30 nM IGF1 as indicated ± 1 mM LY294002. pACL (S455) and total ACL were analyzed. Graphs show densitometry of 3–5 independent experiments, and a representative blot is shown. ANOVA followed by Dunnett’s multiple comparison test was used. (legend continued on next page)
870 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
the effect of IGF1, indicating IGF1 was acting via ACL (Figure 4A). IGF1 increased maximal respiration by 85% and reserve capacity of the myotubes by 90%. Knocking down ACL not only led to a decrease of 53%, and 53% maximal respiration and reserve capacity, but also decreased the IGF1-induced effect by 59% and 60%, respectively (Figure 4A). In addition, IGF1 increased the coupling efficiency of the myotubes by 104%, whereas downregulation of ACL decreased the same by 90% and by 65% in the presence of IGF1 (Figure 4A). In order to understand the mechanism by which ACL enhances mitochondrial respiration, we analyzed mitochondrial number in myotubes after knocking down or overexpressing ACL. There was no significant change in mitochondrial DNA after normalizing to nuclear DNA (Figure S3A). Thus, the altered mitochondrial function of myotubes by ACL is not a consequence of modified mitochondrial number. Next, we analyzed individual mitochondrial complex activities in isolated mitochondria from myotubes after overexpressing or downregulating ACL. Overexpressing ACL induced an increase of 67%, 38%, and 93% in complex I, IV, and V, respectively, whereas knockdown of ACL led to a decrease of 42%, 25%, and 70% in complex I, IV, and V, respectively (Figure 4B). Complex II activities showed similar trends; however, the differences failed to reach significance in three independent experiments (Figure 4B). Because mitochondria produce reactive oxygen species (ROS) as a natural byproduct of electron transport chain activity, and supercomplexes control ROS generation by the respiratory chain (Sullivan and Chandel, 2014), we asked if ACL overexpression perturbs ROS generation. As shown in Figure S3B, ACL overexpression in myotubes did not affect ROS. In addition to increased mitochondrial OXPHOS activity, cellular ATP content was increased by 43% by IGF1 treatment (Figure 4C). Altered OXPHOS activity and mitochondrial complex activity by ACL knockdown or overexpression resulted in 35% decrease or 75% increase in cellular ATP content, respectively (Figure 4C), also demonstrating that IGF1-mediated increases in ATP require ACL activity. In order to verify the significance of ACL activity in maintaining cellular ATP content in vivo, we overexpressed ACL in the TA of mice (Figure S2A). In accord with the in vitro findings, overexpression of ACL increased cellular ATP content by 30% (Figure 4D). Because IGF1 is a strong anabolic stimulus for skeletal muscle (Egerman and Glass, 2014), we knocked down ACL by siRNA in human primary myotubes, subjected myotubes to IGF1 treatment, and measured protein synthesis. As shown in Figure S4A, ACL silencing significantly reduced the increase in protein synthesis and in MHC protein levels mediated by IGF1 without perturbing basal levels. In addition, ACL overexpression in human primary myotubes was suffi-
cient to increase myotube size and MHC protein levels (Figure S4B). Interestingly, ACL phosphorylation on T447/S451 and total ACL levels were decreased in the quadriceps of mice displaying prominent atrophy induced by dexamethasone treatment for 7 days (Figure S4C). DISCUSSION ACL has been appreciated for its role in fatty acid (FA) biosynthesis and its requirement in the production of acetyl-CoA (Wellen et al., 2009; Choudhary et al., 2014). However, ACL’s role in other metabolic processes, particularly in skeletal muscle, has not been extensively studied. Recently, Lee et al. reported that AKT phosphorylates and increases the activity of ACL via a PI3K-dependent pathway in cancer cells (Lee et al., 2014). This finding gave context to our observation that IGF1 was capable of increasing phospho-ACL levels and de novo lipid synthesis in skeletal muscle in an ACL-dependent manner. Further exploration demonstrated that ACL was capable of increasing levels of cardiolipin, a critical component of the mitochondrial membrane that constitutes about 20% of the organelle’s FA composition (Paradies et al., 2014; Ren et al., 2014), opening up a new arena for ACLrequired biology. ACL was found to be required for maintenance of cardiolipin, which in turn modulated mitochondrial complex formation. ACL activity is thus limiting for this critical process— critical because complex formation modulates ATP levels. It was previously demonstrated that IGF1 is an anabolic agent for skeletal muscle; it induces protein synthesis via the AKT/ mTOR pathway and blocks protein breakdown by inhibiting FOXO induction of the E3 ligases MuRF1 and MAFbx, which mediate muscle protein degradation (Egerman and Glass, 2014). What hadn’t been understood previously was that IGF1 could also perturb mitochondrial metabolism, as shown in this study. It seems especially noteworthy that a growth factor that induces protein synthesis is also able to induce the required ATP necessary for the production of new proteins, as well as to facilitate correct protein folding and subsequent protein function. In contrast, in many settings mitochondrial stimulation has seemed to be at odds with anabolic protein synthesis. For example, PGC1a, a transcriptional co-activator that stimulates mitochondriogenesis, is itself induced by AMPK activation (Lee et al., 2006), which at the same time actually inhibits mTOR signaling. Also, anabolic stimuli by mechanisms such as myostatin inhibition can induce hypertrophy without increasing mitochondrial function, and this sort of effect in part might explain why some anabolic stimuli result in more fatigable muscle (LeBrasseur et al., 2009; Mouisel et al., 2014). However, this study gives at least one physiologic mechanism in which both
(B) Myotubes were infected with AdEV or Ad dn-AKT. Forty-eight hours post-infection, cells were treated with 30 nM IGF1 for 30 min. pACL (S455) and total ACL and AKT were analyzed. (C) Myotubes were infected with AdEV or Ad myr-AKT. pACL (S455) and total ACL and AKT were analyzed 24, 48, and 72 hr post-infection by immunoblotting. In (B) and (C), a blot representative of three independent experiments is shown. Graphs show densitometric analysis of three independent experiments; unpaired t test. (D) pACL (T447/S451), pAKT (S473 and T308), and total AKT levels were analyzed in tibialis muscle of young and old mice by immunoblotting. A Mann-Whitney test was used; n = 8 mice/group. (E and F) Differentiated myotubes were subjected to siRNA to ACL and treated with vehicle or 30 nM IGF1 for 24 hr with or without 3 mM LY294002 (F) concurrently with 14C-Na-pyruvate during the last 5 hr of the 24 hr culture, 3 days post-transfection. 14C counts were determined in extracted lipids and normalized to protein concentration. Protein extracts were analyzed for ACL expression 4 days post-transfection. (E) Data are expressed as % change of control. Shown are means ± SEM; n = 3–5; ANOVA followed by Dunnett’s multiple comparison test was used. (F) Shown are mean ± SEM; ANOVA followed by Tukey’s multiple comparison test was used. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 versus control; $ p < 0.05 versus IGF1.
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 871
Figure 2. ACL Regulates Cardiolipin Content in Skeletal Muscle (A and B) AAVs expressing ACL (AAV-ACL; A and B), GFP (AAV-GFP; A and B), AAV-shCtrl (B), or AAV-shACL (B) were injected into the tibialis. (A) Cardiolipin content was analyzed by HPTLC 30 days post-injection. Bands were quantified using ImageJ; shown are means ± SEM; n = 6–8 mice/group; Mann-Whitney test. (B) Transcripts of Crls1, Taz, and Ptpmt1 were analyzed by qRTPCR; data are expressed as fold increase of AAVGFP (n = 6–8 mice/group); Wilcoxon test. (C) Myotubes treated with vehicle or 30 nM IGF1 for 30 hr ± 3 mM LY294002; cardiolipin content was analyzed by HPTLC. Bands were quantified using ImageJ. Data are expressed as means ± SEM of four independent experiments; ANOVA followed by Tukey’s multiple comparison test was used. (D–G) Cardiolipin content (D and F) and mRNA expression of Taz, Ptpmt1, and Crls1 (E and G) were analyzed by HPTLC and qRT-PCR, respectively, in tibialis of young and old (D and E) and young exercised or sedentary (F and G) mice. (D and F) Bands were quantified using ImageJ. Shown are means ± SEM; n = 8 mice/group; MannWhitney test. In (E) and (G), data are expressed as fold increase of young or young/sedentary, respectively. (E) n = 5–6 mice/group; (G) n = 8 mice/ group; Mann-Whitney test, *p < 0.05, **p < 0.01, ***p < 0.001.
anabolic stimuli and mitochondria might be stimulated—not by increasing mitochondrial number, but by increasing mitochondrial efficiency. 872 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
When skeletal muscle ages, there is an increase in frailty, which is coincident with a decrease in muscle mass and function—this condition is called sarcopenia (Cruz-Jentoft et al., 2010). It has been shown that one of the major pathways downregulated at the onset of sarcopenia is the mitochondrial pathway, such as those genes involved in OXPHOS and the TCA cycle (Ibebunjo et al., 2013; Marzetti et al., 2013). A decline in concentrations of circulating IGF1 in sarcopenia has been described in humans and rodents (O’Connor et al., 1998; Velasco et al., 1998). This study suggests the possibility that re-activation of ACL in settings of sarcopenia might be helpful in restoring mitochondrial competence. Of interest, in this study it was observed that phosphorylated ACL is dramatically reduced in old (24 months) versus young mice, coincident with a decrease in levels of AKT phosphorylation with age. Therefore, the goal would be to reactivate ACL in settings of sarcopenia, perhaps by restoring IGF1 signaling to physiologic levels in this setting. It remains to be seen whether genetic manipulation of ACL in old mice affects mitochondrial and skeletal muscle function. Homozygous ACL knockout mice die early in development (Beigneux et al., 2004); therefore, the
Figure 3. ACL Increases Mitochondrial Complex Content and Activity (A–C) Mitochondrial proteins extracted from tibialis injected with AAV-GFP or AAV-ACL 30 days post-infection, analyzed for mitochondrial complex and supercomplex content by Coomassie native PAGE. (D) Levels of respiratory chain subunits analyzed by immunoblotting (D). (E and F) In-gel activity assay was carried out for complex I (E) and complex II–V (F). Graphs show densitometric analysis and data are shown as means ± SEM; *p < 0.05, **p < 0.01; n = 6–7, Mann-Whitney test.
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 873
Figure 4. IGF1 Increases Oxygen Consumption and ATP Content via ACL (A and B) Myotubes given siRNA to ACL (A and B) or infected with AdEV or AdACL (B). (A) Three days post-transfection, cells were treated for 24 hr with vehicle or 30 nM IGF1. Oxygen consumption rates (OCR) ± 1.6 mM oligomycin, 0.5 mM fluoro-carbonyl cyanide phenylhydrazone (FCCP), or 1 mM antimycin + 1.5 mM rotenone were determined using XFe96 extracellular flux analyzer. Graphs show data from three independent experiments; results are expressed as means ± SEM of the area under the curve (AUC). (B) Three days post-transfection or 2 days post-infection, mitochondria were isolated and complex activity was measured by ELISA. Activity is expressed as change in absorbance per minute (mOD/min) per mg of cell lysate. Data are shown as means ± SEM; n = 3. (C) Human skeletal muscle myotubes were transfected with siCtrl or siACL as described in (A) and (B). Three days post-transfection, myotubes were treated with 30 nM IGF1 for 36 hr, and cellular ATP was analyzed. In parallel, myotubes were transduced with AdEV or AdACL and analyzed for ATP concentration 48 hr postinfection. Shown are means ± SEM; n = 3. (D) ATP concentration determined in tibialis of AAV-GFP- or AAV-ACL-injected mice. Shown are means ± SEM; ANOVA followed by Tukey’s multiple comparison test (A–C), Mann-Whitney test (D); n = 6–7 mice per group, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.
generation of a skeletal muscle-specific knockout, or perhaps transgenic mice expressing activated ACL, would be needed to address this question. The finding here that ACL induced cardiolipin production, which has been shown to be required for regulated ATP production, provides an important new node in the understanding of energy metabolism in skeletal muscle. This study opens up future work aimed at understanding distinct signaling steps that might be able to induce the required phosphorylation of ACL, monitoring cardiolipin levels as a biomarker of this ACL/cardiolipin pathway. IGF1/AKT/ACL/cardiolipin/ATP constitutes an important signaling pathway that is apparently critical for providing the energy to match the anabolism of IGF1-induced protein synthesis. 874 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
EXPERIMENTAL PROCEDURES De Novo Lipid Synthesis Assay Primary human skeletal muscle myotubes were incubated with 14C-Na-pyruvate (2 mCi/ml) (PerkinElmer) for 5 hr. The cells were washed four times with cold PBS and lysed in cold PBS by ultrasound. Lipids were extracted by chloroform:methanol (1:1). Radioactivity was measured using scintillation cocktail (UltimaGold-XR, PerkinElmer). The values were normalized by protein concentration determined by BCA (Thermo Scientific). High-Performance Thin-Layer Chromatography The extracted lipids were dissolved in chloroform:methanol (2:1) and analyzed by HPTLC LiChrospher silica gel 60F254S (Millipore) using chloroform:acetone:methanol:acetic acid:water (6:8:2:2:1) as mobile phase. Cardiolipin sodium salt (Sigma) was dissolved in chloroform:methanol (2:1) and used as cardiolipin standard/marker. The HPTLC plate was derivatized by spraying
0.18% copper sulfate in orthophosphoric acid:water (40:460) and heating at 200 C for 20 min. Lipid bands were quantified by ImageJ. Mitochondria Isolation Mitochondria were isolated either from primary human myotubes or tibialis muscle using Dounce homogenizer with mitochondria isolation buffer (70 mM sucrose, 220 mM mannitol, 5 mM KH2PO4, 5 mM MgCl2, 2 mM HEPES, 1 mM EGTA [pH 7.2], and 0.2% FA-free BSA) supplemented with Halt protease and phosphatase inhibitors, trichostatin, and nicotinamide at 4 C. Cell or tissue homogenates were centrifuged at 1,0003 g for 10 min at 4 C. The supernatants were subsequently centrifuged at 12,0003 g for 20 min at 4 C, and pellets were washed in mitochondria isolation buffer and stored at 80 C until further processing. Statistical Analysis Statistical significance was evaluated with one-way analysis of variance (ANOVA) for multiple groups, followed by Bonferroni’s or Tukey’s post hoc test to evaluate differences among groups. Mann-Whitney-Wilcoxon test or unpaired t test was used to compare two groups. All error bars indicate SEM. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, and one table and can be found with this article online at http:// dx.doi.org/10.1016/j.cmet.2015.05.006. AUTHOR CONTRIBUTIONS S.D. and M.F. conceived and designed the experiments and analyzed the data; F.M., G.T., and P.B. contributed to the design, execution, and analysis of animal experiments; B.J., V.M., and P.K. performed the in vitro experiments; M.F. supervised the project; D.J.G. contributed to the design of the experiments; S.D., M.F., and D.J.G. co-wrote the manuscript. S.D. is the recipient of a NIBR postdoctoral fellowship. ACKNOWLEDGMENTS We thank the Muscle Diseases group at the Novartis Institutes for Biomedical Research (NIBR) and Dr. J. Auwerx (Ecole Polytechnique Federale de Lausanne) for their support. All authors are employees of the Novartis Institutes of Biomedical Research. We would like to thank Mark Fishman and Michaela Kneissel for their support and the Novartis Postdoctoral program for its support of author S.D. Received: November 4, 2014 Revised: March 12, 2015 Accepted: April 29, 2015 Published: June 2, 2015 REFERENCES Beigneux, A.P., Kosinski, C., Gavino, B., Horton, J.D., Skarnes, W.C., and Young, S.G. (2004). ATP-citrate lyase deficiency in the mouse. J. Biol. Chem. 279, 9557–9564. Berwick, D.C., Hers, I., Heesom, K.J., Moule, S.K., and Tavare, J.M. (2002). The identification of ATP-citrate lyase as a protein kinase B (Akt) substrate in primary adipocytes. J. Biol. Chem. 277, 33895–33900. Choudhary, C., Weinert, B.T., Nishida, Y., Verdin, E., and Mann, M. (2014). The growing landscape of lysine acetylation links metabolism and cell signalling. Nat. Rev. Mol. Cell Biol. 15, 536–550.
Cruz-Jentoft, A.J., Baeyens, J.P., Bauer, J.M., Boirie, Y., Cederholm, T., Landi, F., Martin, F.C., Michel, J.P., Rolland, Y., Schneider, S.M., et al.; European Working Group on Sarcopenia in Older People (2010). Sarcopenia: European consensus on definition and diagnosis: Report of the European Working Group on Sarcopenia in Older People. Age Ageing 39, 412–423. Egan, B., and Zierath, J.R. (2013). Exercise metabolism and the molecular regulation of skeletal muscle adaptation. Cell Metab. 17, 162–184. Egerman, M.A., and Glass, D.J. (2014). Signaling pathways controlling skeletal muscle mass. Crit. Rev. Biochem. Mol. Biol. 49, 59–68. Fiatarone, M.A., Marks, E.C., Ryan, N.D., Meredith, C.N., Lipsitz, L.A., and Evans, W.J. (1990). High-intensity strength training in nonagenarians. Effects on skeletal muscle. JAMA 263, 3029–3034. Hansen, D., Dendale, P., van Loon, L.J., and Meeusen, R. (2010). The impact of training modalities on the clinical benefits of exercise intervention in patients with cardiovascular disease risk or type 2 diabetes mellitus. Sports Med. 40, 921–940. Hatzivassiliou, G., Zhao, F., Bauer, D.E., Andreadis, C., Shaw, A.N., Dhanak, D., Hingorani, S.R., Tuveson, D.A., and Thompson, C.B. (2005). ATP citrate lyase inhibition can suppress tumor cell growth. Cancer Cell 8, 311–321. Ibebunjo, C., Chick, J.M., Kendall, T., Eash, J.K., Li, C., Zhang, Y., Vickers, C., Wu, Z., Clarke, B.A., Shi, J., et al. (2013). Genomic and proteomic profiling reveals reduced mitochondrial function and disruption of the neuromuscular junction driving rat sarcopenia. Mol. Cell. Biol. 33, 194–212. Ikeda, K., Shiba, S., Horie-Inoue, K., Shimokata, K., and Inoue, S. (2013). A stabilizing factor for mitochondrial respiratory supercomplex assembly regulates energy metabolism in muscle. Nat. Commun. 4, 2147. Jeppesen, T.D., Schwartz, M., Olsen, D.B., Wibrand, F., Krag, T., Dunø, M., Hauerslev, S., and Vissing, J. (2006). Aerobic training is safe and improves exercise capacity in patients with mitochondrial myopathy. Brain 129, 3402– 3412. Julienne, C.M., Dumas, J.F., Goupille, C., Pinault, M., Berri, C., Collin, A., Tesseraud, S., Couet, C., and Servais, S. (2012). Cancer cachexia is associated with a decrease in skeletal muscle mitochondrial oxidative capacities without alteration of ATP production efficiency. J Cachexia Sarcopenia Muscle 3, 265–275. Kro¨ger, J., and Schulze, M.B. (2012). Recent insights into the relation of D5 desaturase and D6 desaturase activity to the development of type 2 diabetes. Curr. Opin. Lipidol. 23, 4–10. Lapuente-Brun, E., Moreno-Loshuertos, R., Acı´n-Pe´rez, R., Latorre-Pellicer, A., Cola´s, C., Balsa, E., Perales-Clemente, E., Quiro´s, P.M., Calvo, E., Rodrı´guez-Herna´ndez, M.A., et al. (2013). Supercomplex assembly determines electron flux in the mitochondrial electron transport chain. Science 340, 1567–1570. LeBrasseur, N.K., Schelhorn, T.M., Bernardo, B.L., Cosgrove, P.G., Loria, P.M., and Brown, T.A. (2009). Myostatin inhibition enhances the effects of exercise on performance and metabolic outcomes in aged mice. J. Gerontol. A Biol. Sci. Med. Sci. 64, 940–948. Lee, W.J., Kim, M., Park, H.S., Kim, H.S., Jeon, M.J., Oh, K.S., Koh, E.H., Won, J.C., Kim, M.S., Oh, G.T., et al. (2006). AMPK activation increases fatty acid oxidation in skeletal muscle by activating PPARalpha and PGC-1. Biochem. Biophys. Res. Commun. 340, 291–295. Lee, J.V., Carrer, A., Shah, S., Snyder, N.W., Wei, S., Venneti, S., Worth, A.J., Yuan, Z.F., Lim, H.W., Liu, S., et al. (2014). Akt-dependent metabolic reprogramming regulates tumor cell histone acetylation. Cell Metab. 20, 306–319. Marzetti, E., Calvani, R., Cesari, M., Buford, T.W., Lorenzi, M., Behnke, B.J., and Leeuwenburgh, C. (2013). Mitochondrial dysfunction and sarcopenia of aging: from signaling pathways to clinical trials. Int. J. Biochem. Cell Biol. 45, 2288–2301.
Chypre, M., Zaidi, N., and Smans, K. (2012). ATP-citrate lyase: a mini-review. Biochem. Biophys. Res. Commun. 422, 1–4.
Mejia, E.M., Cole, L.K., and Hatch, G.M. (2014). Cardiolipin metabolism and the role it plays in heart failure and mitochondrial supercomplex formation. Cardiovasc. Hematol. Disord. Drug Targets 14, 98–106.
Clemmons, D.R. (2012). Metabolic actions of insulin-like growth factor-I in normal physiology and diabetes. Endocrinol. Metab. Clin. North Am. 41, 425–443, vii–viii.
Mouisel, E., Relizani, K., Mille-Hamard, L., Denis, R., Hourde´, C., Agbulut, O., Patel, K., Arandel, L., Morales-Gonzalez, S., Vignaud, A., et al. (2014). Myostatin is a key mediator between energy metabolism and endurance
Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc. 875
capacity of skeletal muscle. Am. J. Physiol. Regul. Integr. Comp. Physiol. 307, R444–R454. O’Connor, K.G., Tobin, J.D., Harman, S.M., Plato, C.C., Roy, T.A., Sherman, S.S., and Blackman, M.R. (1998). Serum levels of insulin-like growth factor-I are related to age and not to body composition in healthy women and men. J. Gerontol. A Biol. Sci. Med. Sci. 53, M176–M182. Paradies, G., Paradies, V., De Benedictis, V., Ruggiero, F.M., and Petrosillo, G. (2014). Functional role of cardiolipin in mitochondrial bioenergetics. Biochim. Biophys. Acta 1837, 408–417. Potapova, I.A., El-Maghrabi, M.R., Doronin, S.V., and Benjamin, W.B. (2000). Phosphorylation of recombinant human ATP:citrate lyase by cAMP-dependent protein kinase abolishes homotropic allosteric regulation of the enzyme by citrate and increases the enzyme activity. Allosteric activation of ATP:citrate lyase by phosphorylated sugars. Biochemistry 39, 1169–1179. Ren, M., Phoon, C.K., and Schlame, M. (2014). Metabolism and function of mitochondrial cardiolipin. Prog. Lipid Res. 55, 1–16. Riedl, I., Yoshioka, M., Nishida, Y., Tobina, T., Paradis, R., Shono, N., Tanaka, H., and St-Amand, J. (2010). Regulation of skeletal muscle transcriptome in elderly men after 6 weeks of endurance training at lactate threshold intensity. Exp. Gerontol. 45, 896–903. Schiaffino, S., Dyar, K.A., Ciciliot, S., Blaauw, B., and Sandri, M. (2013). Mechanisms regulating skeletal muscle growth and atrophy. FEBS J. 280, 4294–4314.
876 Cell Metabolism 21, 868–876, June 2, 2015 ª2015 Elsevier Inc.
Stephenson, E.J., and Hawley, J.A. (2014). Mitochondrial function in metabolic health: a genetic and environmental tug of war. Biochim. Biophys. Acta 1840, 1285–1294. Sullivan, L.B., and Chandel, N.S. (2014). Mitochondrial reactive oxygen species and cancer. Cancer Metab 2, 17. Szendroedi, J., Zwettler, E., Schmid, A.I., Chmelik, M., Pacini, G., Kacerovsky, G., Smekal, G., Nowotny, P., Wagner, O., Schnack, C., et al. (2008). Reduced basal ATP synthetic flux of skeletal muscle in patients with previous acromegaly. PLoS ONE 3, e3958. Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009). Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029–1033. Velasco, B., Cacicedo, L., Escalada, J., Lopez-Fernandez, J., and SanchezFranco, F. (1998). Growth hormone gene expression and secretion in aging rats is age dependent and not age-associated weight increase related. Endocrinology 139, 1314–1320. Wang, M., Wang, X.C., Zhang, Z.Y., Mou, B., and Hu, R.M. (2010). Impaired mitochondrial oxidative phosphorylation in multiple insulin-sensitive tissues of humans with type 2 diabetes mellitus. J. Int. Med. Res. 38, 769–781. Wellen, K.E., Hatzivassiliou, G., Sachdeva, U.M., Bui, T.V., Cross, J.R., and Thompson, C.B. (2009). ATP-citrate lyase links cellular metabolism to histone acetylation. Science 324, 1076–1080. White, J.P., Baltgalvis, K.A., Puppa, M.J., Sato, S., Baynes, J.W., and Carson, J.A. (2011). Muscle oxidative capacity during IL-6-dependent cancer cachexia. Am. J. Physiol. Regul. Integr. Comp. Physiol. 300, R201–R211.
