AMPK controls exercise endurance, mitochondrial oxidative capacity, and skeletal muscle integrity

The FASEB Journal • Research Communication

AMPK controls exercise endurance, mitochondrial oxidative capacity, and skeletal muscle integrity Louise Lantier,*,†,‡,1 Joachim Fentz,§ Rémi Mounier,*,†,‡,1 Jocelyne Leclerc,*,†,‡ Jonas T. Treebak,§,1 Christian Pehmøller,§,1 Nieves Sanz,*,†,‡ Iori Sakakibara,*,†,‡ Emmanuelle Saint-Amand,储 Stéphanie Rimbaud,¶ Pascal Maire,*,†,‡ André Marette,储 Renée Ventura-Clapier,¶ Arnaud Ferry,‡,# Jørgen F. P. Wojtaszewski,§ Marc Foretz,*,†,‡ and Benoit Viollet*,†,‡,2 *Institut National de la Santé et de la Recherche Médicale (INSERM), Unité (U)1016, Institut Cochin, Paris, France; †Centre National de la Recherche Scientifique (CNRS), Unité Mixte de Recherche (UMR) 8104, Paris, France; ‡Université Paris Descartes, Sorbonne Paris Cité, Paris, France; §Section of Molecular Physiology, The August Krogh Centre, Department of Nutrition, Exercise, and Sports, University of Copenhagen, Copenhagen, Denmark; 储Laval University, Ste-Foy, Quebec, Canada; ¶INSERM, U769, Université Paris Sud, Châtenay-Malabry, France; and #Institut de Myologie, INSERM, U974, CNRS UMR 7215, Université Pierre et Marie Curie, Paris, France AMP-activated protein kinase (AMPK) is a sensor of cellular energy status that plays a central role in skeletal muscle metabolism. We used skeletal muscle-specific AMPK␣1␣2 double-knockout (mdKO) mice to provide direct genetic evidence of the physiological importance of AMPK in regulating muscle exercise capacity, mitochondrial function, and contractionstimulated glucose uptake. Exercise performance was significantly reduced in the mdKO mice, with a reduction in maximal force production and fatigue resistance. An increase in the proportion of myofibers with centralized nuclei was noted, as well as an elevated expression of interleukin 6 (IL-6) mRNA, possibly consistent with mild skeletal muscle injury. Notably, we found that AMPK␣1 and AMPK␣2 isoforms are dispensable for contraction-induced skeletal muscle glucose transport, except for male soleus muscle. However, the lack of skeletal muscle AMPK diminished maximal ADP-stimulated mitochondrial respiration, showing an impairment at complex I. This effect was not accompanied by changes in mitochondrial number, indicating that AMPK regulates muscle metabolic adaptation through the regulation of muscle mitochondrial oxidative capacity and mitochondrial substrate utilization but not baseline mitochondrial muscle content. Together, these results demonstrate that skeletal mus-

ABSTRACT

Abbreviations: ACC, acetyl-CoA carboxylase; AICAR, 5aminoimidazole-4-carboxamide-1-␤-d-ribofuranoside; AMPK, AMP-activated protein kinase; BNIP, BCL2/adenovirus E1B interacting protein; CSA, cross-sectional area; DTT, dithiothreitol; EDL, extensor digitorum longus; H&E, hematoxylin and eosin; HSA, human skeletal actin; IL-6, interleukin 6; KD, kinase dead; KO, knockout; LKB1, liver kinase B1; mdKO, muscle-specific AMPK␣1␣2 double-knockout; mtTFA, mitochondrial transcription factor A; MuRF1, muscle RING finger 1; MyHC, myosin heavy chain; oxphos, oxidative phosphorylation; PGC, proliferator-activated receptor ␥ coactivator; TNF-␣, tumor necrosis factor ␣; WT, wild-type 0892-6638/14/0028-3211 © FASEB

cle AMPK has an unexpected role in the regulation of mitochondrial oxidative phosphorylation that contributes to the energy demands of the exercising muscle.—Lantier, L., Fentz, J., Mounier, R., Leclerc, J., Treebak, J. T., Pehmøller, C., Sanz, N., Sakakibara, I., Saint-Amand, E., Rimbaud, S., Maire, P., Marette, A., Ventura-Clapier, R., Ferry, A., Wojtaszewski, J. F. P., Foretz, M., Viollet, B. AMPK controls exercise endurance, mitochondrial oxidative capacity, and skeletal muscle integrity. FASEB J. 28, 3211–3224 (2014). www.fasebj.org Key Words: glucose transport 䡠 force production To sustain metabolism, intracellular ATP concentration must be maintained within an appropriate range, and AMP-activated protein kinase (AMPK), a phylogenetically conserved Ser/Thr protein kinase, is a key enzyme in this regulation. Indeed, AMPK has been described as a fuel sensor and is activated by changes in 1 Current address: L.L., Molecular Physiology and Biophysics, Vanderbilt University Medical Center, Nashville, TN, USA; R.M., Centre de Génétique et de Physiologie Moléculaires et Cellulaires, UMR CNRS 5534, Université Claude Bernard Lyon 1, Lyon, France; J.T.T., The Novo Nordisk Foundation Center for Basic Metabolic Research, Section of Integrative Physiology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; C.P., Cardiovascular, Metabolic, and Endocrine Diseases (CVMED), Pfizer Global Research and Development, Cambridge, MA, USA 2 Correspondence: Dpt. Endocrinologie, Métabolisme et Diabète, Institut Cochin, INSERM U1016, 24, Rue du Faubourg St.-Jacques, 75014 Paris, France. E-mail: benoit. [email protected] doi: 10.1096/fj.14-250449 This article includes supplemental data. Please visit http:// www.fasebj.org to obtain this information.

3211

adenine nucleotide ratios (1). Mammalian AMPK is a heterotrimeric complex consisting of a catalytic (␣) subunit and 2 regulatory (␤ and ␥) subunits, encoded by different genes (␣1, ␣2, ␤1, ␤2, ␥1, ␥2, and ␥3). An increase in the cellular AMP/ATP and ADP/ATP ratios leads to allosteric AMPK activation by AMP and ADP, as well as by phosphorylation of Thr172 situated within the activation loop segment of the ␣ subunit. This residue can be phosphorylated by several upstream AMPK kinases, one of which is liver kinase B1 (LKB1). Another canonical activation mechanism involves the phosphorylation of Thr172 by Ca2⫹/calmodulin-dependent protein kinase ␤ (CaMKK␤) in response to a rise in intracellular Ca2⫹. This mechanism can occur in the absence of any change in cellular nucleotides, although the two canonical mechanisms (AMP and Ca2⫹) can also synergize (2). Once activated by a falling energy charge, AMPK elicits major cellular metabolic changes by stimulating energy production via catabolic pathways while decreasing nonessential energy-consuming pathways to restore cellular energy stores. In skeletal muscle, AMPK activation plays a role in regulating physiological events including glucose transport, mitochondrial function such as fatty acid oxidation, and mitochondrial biogenesis (reviewed in refs. 3, 4). Exercise and muscle contraction are potent stimuli for AMPK, and its activation is induced by changes in muscular energy charge. In skeletal muscle, this activation is dependent on exercise intensity and duration, with intense muscle exercise preferentially activating AMPK␣2 but not AMPK␣1 (5). To explain these differences in the activation profile of AMPK␣1- and AMPK␣2-containing complexes, it has been postulated that every AMPK heterotrimer combination displays a distinct spectrum of intracellular localization, activation mechanism, and biochemical properties (6). Notably, among the 12 different combinations of AMPK heterotrimers, 5 have been identified in mouse skeletal muscle (7), whereas only the 3 complexes ␣2␤2␥1, ␣1␤2␥1, and ␣2␤2␥3 have been observed in human skeletal muscle. Only the ␣2␤2␥3 and ␣2␤2␥1 complexes are known to be activated during exercise and seem to target differential substrates in the signaling pathway, leading to glucose uptake by the muscle (5, 8, 9). Several studies have examined the isoform-specific contributions to AMPK function during exercise by use of transgenic mice with altered expression of AMPK␣ [AMPK␣1- and AMPK␣2-knockout (KO) mice and mice overexpressing kinase-dead AMPK␣2 (AMPK␣2KD)], AMPK␤ (AMPK␤1- and AMPK␤2-KO and muscle-specific AMPK␤1␤2 double-KO mice), and AMPK␥ (AMPK␥1R70Q and AMPK␥3 transgenic mice and AMPK␥3-KO mice) subunits. Although impaired AMPK activation is clearly associated with exercise intolerance, a direct role for skeletal muscle AMPK in the metabolic adaptation to exercise remains poorly understood. Of note, although several models of AMPK deficiency display significant decreases in exercise tol3212

Vol. 28

July 2014

erance, contraction-induced glucose uptake in skeletal muscle, measured in vitro, in situ, or in vivo, can be AMPK-dependent or -independent (10 –16). This contradiction makes it difficult to assess the contribution of reduced muscle glucose uptake to impaired exercise performance (17). Thus, the role of AMPK in metabolic response and substrate metabolism at rest and during exercise still has to be elucidated. In the current study, we tested the hypothesis that AMPK in skeletal muscle is crucial for proper muscle function, by integrating contraction-induced regulatory signals and subsequently adjusting muscle mitochondrial respiration capacity to meet the energy demands of the exercising muscle. We showed that mice specifically lacking both AMPK␣1 and AMPK␣2 in skeletal muscle [musclespecific AMPK␣1␣2 double-KO (mdKO) mice] display no impairment in contraction-stimulated glucose uptake in vitro in extensor digitorum longus (EDL) muscle of male and female mice. In contrast, AMPK␣ subunits are necessary in the soleus of male mice but not in that of female mice. In addition, our data demonstrated that skeletal muscle of mdKO mice exhibits marked reduction of muscle mitochondrial oxidative capacity that manifests as reduced exercise tolerance.

MATERIALS AND METHODS Animals To obtain skeletal muscle AMPK-deficient mice [AMPK␣1fl/fl ␣2fl/fl human skeletal actin (HSA)-Cre⫹ mice on a C57Bl6129Sv mixed background], AMPK␣1fl/fl␣2fl/fl mice were interbred with transgenic mice expressing Cre recombinase under the control of the HSA promoter, which is expressed in differentiated multinucleated skeletal fibers (18). The mice were housed in a facility on a 12-h light-dark cycle with free access to standard rodent chow and water. All procedures were performed under a French authorization (75– 886) to experiment on vertebrates, in accordance with the principles and guidelines established by the European Convention for the Protection of Laboratory Animals, and were approved by the ethics committee from the University Paris-Descartes (CEEA34.BV.157.12). Voluntary wheel-running exercise Mice were individually housed in cages equipped with a wheel for running and were allowed free access to exercise for a period of 7 d. Each wheel was equipped with a magnetic switch and a counter. Daily (24 h) running distance and time spent running were logged every morning. Motor coordination test To assess overall motor coordination, mice were tested on an accelerating rotarod. They were observed for time (latency) until falling off the rotarod, which accelerated from 4 to 40 rpm in 300 s. Each mouse underwent 4 consecutive trials separated by 15 min intervals.

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

Treadmill-running test All mice were familiarized with a small-animal treadmill at low speed (8 cm·s⫺1, 0° incline) for 10 min before performing the exercise test. Exercise testing consisted of a submaximal run to exhaustion with running at 15 cm/s for the first 10 min (10° incline), followed by 20 or 35 cm/s. Exhaustion was defined as when the mouse was no longer able to maintain its normal running position and/or showed frequent contact with the shock grid (ⱕ0.2 mA) at the rear of the treadmill. Force measurements The isometric contractile properties of soleus muscles were studied in vitro, as described elsewhere (19). Muscles were soaked in an oxygenated Krebs solution (95% O2 and 5% CO2), containing 58.5 mM NaCl, 24 mM NaHCO3, 5.4 mM KCl, 1.2 mM KH2PO4, 1.8 mM CaCl2, 1 mM MgSO4, and 10 mM glucose (pH 7.4), and were maintained at a temperature of 22°C. One of the muscle tendons was attached to a force transducer (Harvard Apparatus, Holliston, MA, USA). After equilibration (30 min), electrical stimulation was delivered through electrodes running parallel to the muscle. Pulses of 0.1 ms were generated by a high-power stimulator (701B; Aurora Scientific, Aurora, ON, Canada). Absolute maximal isometric force was measured during isometric contractions in response to electrical stimulation (frequency, 50 –125 Hz; train of stimulation, 1500 ms). All measurements were made at optimal muscle length (L0, length at which absolute maximal isometric force was obtained). Fatigue resistance was determined after a 5 min rest period. The muscles were stimulated at 75 Hz during 500 ms every 1.6 s for 3 min. The time taken for initial force to decline by 30% was determined. The soleus muscles were weighed, and specific maximal force was calculated by dividing absolute maximal force by the estimated cross-sectional area (CSA) of the muscle. Assuming that muscles have a cylindrical shape and a density of 1.06 mg/mm3, the CSA corresponds to the volume of the muscle divided by fiber length (Lf). The Lf to L0 ratio of 0.70 was used to calculate Lf. The isometric contractile properties of tibialis anterior muscle were studied in situ, as described previously (20). The knee and foot were fixed with clamps and pins. The distal tendon of the muscles was attached to a force transducer (Harvard Apparatus). The sciatic nerve (proximally crushed) was stimulated by a bipolar silver electrode using a supramaximal square-wave pulse of 0.1 ms duration. Absolute maximal isometric force was measured during isometric contractions in response to electrical stimulation (frequency, 75–150 Hz; train of stimulation, 500 ms). Fatigue resistance was determined after a 5 min rest period. The muscles were stimulated at 100 Hz for 500 ms, every 2 s for 3 min. The time taken for initial force to decrease by 30% was determined. The muscles were weighed, and specific maximal force was calculated by dividing absolute maximal force by muscle weight. Mitochondrial bioenergetics in permeabilized fibers Respiratory parameters of the total mitochondrial population were studied in situ, as described previously (21). Fiber bundles obtained from white gastrocnemius (⬃2 mg) were separated with fine forceps under a binocular dissecting microscope in BIOPS buffer containing 2.77 mM CaK2EGTA, 7.23 mM K2EGTA, 5.77 mM Na2ATP, 6.56 mM MgCl2·6H2O, 15 mM Na2 phosphocreatine, 20 mM imidazole, 20 mM taurine, 0.5 mM dithiothreitol (DTT), and 50 mM K-MES. After separation, the fiber bundles were placed in BIOPS AMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

containing 40 ␮g/ml saponin, agitated for 30 min, and then washed in respiration buffer R containing 2.77 mM CaK2EGTA, 7.23 mM K2EGTA, 1.38 mM MgCl2·6H2O, 3 mM K2HPO4·3H2O, 20 mM imidazole, 20 mM taurine, 0.5 mM DTT, 90 mM K-MES, 10 mM Na-MES, and 1 mg/ml fatty acid-free BSA. The fibers were left in cold buffer R until respiration analysis. Mitochondrial respiration was measured with a Clark electrode (Strathkelvin Instruments, Glasgow, UK) in a water-jacketed oxygraphic cell containing 3 ml of respiration solution at 22°C, with continuous stirring. After the study, the bundle was dried, and the results were normalized to dry weight. Two different experimental protocols were used, based on substrate utilization pathways (22). The first protocol determined consumption of oxygen related to complex I-linked substrate (5 mM glutamate and 2 mM malate) and complex II-linked substrate (10 mM succinate and 2 mM malate) in the presence of 2 mM ADP. The second protocol was aimed at determining the sensitivity of mitochondrial respiration to carbohydrates and fatty acid by monitoring the oxygen consumption in the presence of 2 mM ADP and 4 mM malate and by successive addition of 1 mM carnitine plus 0.1 mM palmitoyl-CoA, 0.15 mM palmitoyl-carnitine, 0.4 mM octanoate, 4 mM glycerol-3-phosphate, and 10 mM glutamate. Rates of respiration are given in micromoles of oxygen per minute per gram of dry weight. In vitro measurement of contraction-induced glucose uptake Soleus and EDL muscles were excised from anesthetized (pentobarbital; 5 mg/kg), fed mice (mean age, 23.8⫾4.8 wk); mounted in chambers (Multi Myograph System, Danish MyoTechnology, Aarhus, Denmark); and incubated in KrebsRinger buffer (7). After 40 min of preincubation, the muscles were electrically stimulated to contract (MultiStim SystemD330; Harvard Apparatus) by applying 1 of 2 protocols (1 train/15 s, ⬃30 – 40 V, each train 1 s long for protocol 1 or 2 s long for protocol 2, consisting of 0.2 ms pulses delivered at 100 Hz) for 10 min. The uptake of 2-deoxyglucose during contraction was measured by adding 1 mM [3H]2-deoxyglucose and 7 mM [14C]mannitol (Perkin Elmer, Allerød, Denmark) to the incubation medium. Insulin-stimulated glucose uptake was measured (by addition of tracers as above) during the last 10 min of a 40 min period of insulin stimulation (concentrations as indicated in figures; Actrapid, Novo Nordisk, Bagsvaerd, Denmark). After incubation, the muscles were harvested, stored, and analyzed for glucose uptake as described previously (7). Muscles used for Western blot analyses were treated the same, but without adding tracers. Muscle ATP levels and mitochondrial enzymatic activities Muscle ATP levels were determined using the luminescent CellTiter-Glo reagent (Promega, Charbonnières-les-Bains, France). Frozen tissue samples from gastrocnemius muscles were weighed and homogenized into ice-cold buffer (50 mg/ml) containing 5 mM HEPES (pH 8.7), 1 mM EGTA, 0.1% Triton X-100, and 1 mM DTT. Citrate synthase activity was assayed at 30°C (pH 7.5), using coupled enzyme systems (21). Muscle processing and Western blot analyses Muscles were homogenized in ice-cold lysis buffer containing 50 mM Tris, (pH 7.4), 1% Triton X-100, 150 mM NaCl, 10% glycerol, 50 mM NaF, 5 mM sodium pyrophosphate, 1 mM 3213

Na3VO4, 25 mM sodium-␤-glycerophosphate, 1 mM DTT, 0.5 mM PMSF, and protease inhibitors (Complete Protease Inhibitor Cocktail; Roche Diagnostics, Meylan France) in a ball-bearing homogenizer (Retsch, Haan, Germany). The homogenate was centrifuged for 10 min at 10,000 g at 4°C, and the supernatants were removed for determination of total protein content. Protein (50 ␮g) from the supernatant was separated on 7.5 or 10% SDS-polyacrylamide gels and transferred to nitrocellulose membranes. The membranes were blocked for 30 min at 37°C with Tris-buffered saline supplemented with 0.05% Nonidet P-40 and 5% nonfat dry milk. Membranes stained with Red Ponceau S (Sigma-Aldrich, Lyon, France) were used as loading controls. Immunoblot analysis was performed according to standard procedures, and the signals were detected by chemiluminescence reagents (ThermoScientific, Villebon-sur-Yvette, France). Primary antibodies were directed against total AMPK␣ (2532), AMPK␣ phosphorylated at Thr172 (2531), total acetyl-CoA carboxylase (ACC; 3676), and ACC phosphorylated at Ser212 (3661) (all from Cell Signaling Technologies, Danvers, MA, USA). Quantitative PCR analysis Total RNA from muscle was extracted with Trizol (Invitrogen, Cergy Pontoise, France), and single-stranded cDNA was synthesized from 5 ␮g of total RNA with random hexamer primers (Applied Biosystems, Courtaboeuf, France) and Superscript II (Invitrogen). Real-time RT-PCRs were performed with Lithos qPCR Mastermix (Eurogenetec, Seraing, Belgium) in a final volume of 20 ␮l, containing 250 ng of reverse-transcribed total RNA, 500 nM of prim-

ers, 10 ␮l of 2⫻ PCR mix, and 0.5 ␮l of SYBR Green. The reactions were performed in capillaries in a LightCycler (Roche Diagnostics) for 40 cycles. We determined the relative amounts of the mRNAs studied by means of the second-derivative maximal method, with LightCycler 3.5 analysis software, and cyclophilin RNA as the invariant control for all studies. The sense and antisense PCR primers are shown in Table 1. Immunohistochemistry Soleus and gastrocnemius muscles were embedded in Cryomatrix (ThermoScientific) and quickly frozen in isopentane cooled with liquid nitrogen. Cryostat sections (10 ␮m) were washed in PBS, permeabilized with 0.1% Triton X-100, and left for 1 h in blocking solution (1⫻ PBS, 1.5% goat serum, and 0.1% Triton X-100). Rabbit polyclonal antibodies directed against laminin (1:100 dilution; Z0097; DakoCytomation, Trappes, France) and monoclonal antibodies against myosin heavy chain (MyHC)-I (M8421, clone NOQ7.5.4D; Sigma-Aldrich) and MyHC-IIA (SC-71; Developmental Studies Hybridoma Bank, Iowa City, IA, USA) were applied overnight at 4°C to the treated sections. The next day, after 3 washes with 1⫻ PBS containing 0.05% Tween-20, the cryosections were incubated for 1 h with appropriate fluorescent secondary antibodies (Alexa Fluor 488 goat anti-rabbit IgG, A-11008, 1:1000 dilution, and Alexa Fluor 594 goat antimouse IgG, A-11005, 1:1000 dilution; Invitrogen). After 3 washes with 1⫻ PBS containing 0.05% Tween 20, the samples were mounted in Vectashield medium (Vector Laboratories, Burlingame, CA, USA).

TABLE 1. Primers used for quantitative real-time PCR Gene

PGC1a mtTFA Cytochrome c ALAS COX1 COX4 Citrate synthase TNFa IL6 MuRF1 FoxO3 BNIP3 Cyclophilin Mitochondrial DNA Nuclear DNA

3214

Vol. 28

July 2014

Forward/reverse primers

5=-ATACCGCAAAGAGCACGAGAAG 5=-CTCAAGAGCAGCGAAAGCGTCACAG 5=-GCTGATGGGTATGGAGAAG 5=-GAGCCGAATCATCCTTTGC 5=-CCAAATCTCCACGGTCTGTTC 5=-ATCAGGGTATCCTCTCCCCAG 5=-TTTGTGGACGAGGTCCATGCAGTA 5=-GCATTCAGCTGACGAATGTGGCTT 5=-CACTAATAATCGGAGCCCCA 5=-TTCATCCTGTTCCTGCTCCT 5=-TGGGAGTGTTGTGAAGAGTGA 5=-GCAGTGAAGCCGATGAAGAAC 5=-CATCACAGCCCTCAACAGT 5=-CACTTACATTGCCACCGTC 5=-AATGGCCTCCCTCTCATCAGTT 5=-CGAATTTTGAGAAGATGATCTGAGTGT 5=-CTTCAACCAAGAGGTAAAAGATTTA 5=-TAGGAGAGCATTGGAAATTGGGGTAGGAAGG 5=-ACCTGCTGGTGGAAAACATC 5=-CTTCGTGTTCCTTGCACATC 5=-ACCTTCGTCTCTGAACTCCTTG 5=-CTGTGGCTGAGTGAGTCTGAAG 5=-TTCCACTAGCACCTTCTGATGA 5=-GAACACCGCATTTACAGAACAA 5=-ATGGCACTGGCGGCAGGTCC 5=-TTGCCATTCCTGGACCCAAA 5=-CCGCAAGGGAAAGATGAAAGA 5=-TCGTTTGGTTTCGGGGTTTC 5=-GCCAGCCTCTCCTGATGT 5=-GGGAACACAAAAGACCTCTTCTGG

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

Transmission electron microscopy The muscles were fixed in 4% glutaraldehyde and 0.1 M sodium cacodylate buffer (pH 7.4) for 24 h at 4°C, postfixed with 1% osmium tetroxide, dehydrated with 100% ethanol, and embedded in epoxy resin. For ultrastructure analysis, ultrathin slices (70 to 100 nm thick) were cut from the resin blocks with an ultramicrotome (Ultracut S; Reichert Technologies, DePew, NY, USA), stained with lead citrate and uranyl acetate, and examined in a transmission electron microscope (model 1011; JEOL, Ltd., Akishima, Japan) at the Cochin Institute electron microscopy facility. Statistics Data are expressed as means ⫾ sem. Two-group comparisons were performed with Student’s unpaired t test. Multiplegroup comparisons were performed with a 2-way ANOVA, with or without repeated measures, as appropriate. When a difference was present, the Student-Newman-Keuls post hoc test was applied. Data were analyzed in SigmaPlot 12.5 (Systat, San Jose, CA, USA) or Prism software (GraphPad, San Diego, CA, USA), with the level of statistical significance set at P ⬍ 0.05.

RESULTS Generation and characterization of mdKO mice To investigate the specific function of AMPK␣1 and AMPK␣2 catalytic subunits in skeletal muscle, we generated skeletal muscle-specific deletion of both AMPK␣1 and AMPK␣2 genes in mdKO mice. Mice carrying the 2 floxed alleles AMPK␣1 fl/fl and AMPK␣2fl/fl (23, 24) were mated with mice that

express Cre recombinase under the HSA promoter (18). Western blot analyses showed the significant loss of AMPK␣1 and AMPK␣2 proteins in both slowand fast-twitch skeletal muscles (plantaris, EDL, tibialis anterior, gastrocnemius, and soleus) from the mdKO mice (Fig. 1A), but not in other organs, such as the heart (Fig. 1A) and the liver (data not shown). Residual expression of AMPK␣1 in some AMPKdeficient muscles was most likely due to the predominant expression of AMPK␣1 isoform in nontargeted cells, such as endothelial cells, fibroblasts, adipocytes, satellite cells, and red blood cells, which are also present within skeletal muscle tissue. Intraperitoneal injection of 5-aminoimidazole-4-carboxamide1-␤-d-ribofuranoside (AICAR), a synthetic AMPK activator, failed to increase levels of pAMPK (Thr172) and pACC (Ser212) in mdKO gastrocnemius muscle, in contrast to its effects in control mice (Fig. 1B). In the liver, as expected, the induction of pAMPK and pACC following AICAR injection was similar in mdKO and control mice (data not shown). In addition, we monitored the effect of in vitro contraction (induced by electrical stimulation) and showed the absence of AMPK-Thr172 and ACC-Ser212 phosphorylation in contracting muscle from the mdKO mice (Fig. 1C and Supplemental Fig. S1). The mdKO mice had normal survival, and neither body weight nor food intake was affected, as compared with that of the controls (data not shown). Tibia bone length of the mdKO mice was not different from that of the controls (data not shown), indicating no alteration in linear growth. Finally, glucose homeostasis in these mice was normal, as were plasma insulin levels, as

Figure 1. Characterization of muscle-specific deletion of AMPK␣1 and AMPK␣2 catalytic subunits. A) Expression of AMPK␣1 and AMPK␣2 in plantaris, soleus, EDL, tibialis anterior, gastrocnemius (gastroc), and cardiac muscles from WT and mdKO mice. B) AMPK-Thr172 and ACC-Ser212 phosphorylation in gastrocnemius muscle 40 min after injection of saline or AICAR (500 mg/kg) into WT and mdKO mice. C) AMPK-Thr172 phosphorylation in resting (Bas) and electrically stimulated (Con) soleus muscle. Ponceau staining was used as the loading control (loading). AMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

3215

indicated by similar overnight fasting and fed blood glucose levels in the mdKO and control mice (Supplemental Fig. S2A). Reduced exercise capacity and impaired muscle function in mdKO To study the effect of AMPK␣1 and AMPK␣2 deficiency on the overall performance of skeletal muscle, we assessed the ability of mdKO and control male mice to perform physical exercise. Voluntary wheel running was strongly reduced in mdKO mice, with a shorter total distance run per day (⬃55% of that of control mice), but average speed did not differ (Fig. 2A). To determine whether differences in voluntary wheel running between mdKO and control mice could be due to defects in motor coordination and balance, we evaluated these animals in a rotarod test. No differences in the latency before falling and the speed at falling were observed, indicating similar locomotor behavior in the mdKO and control mice (Fig. 2B). We then challenged the endurance of age-matched mdKO and control littermates by treadmill running to exhaustion, as an indicator of exercise tolerance. We observed a dramatic reduction

in endurance in the mdKO mice. Time to exhaustion during submaximal endurance exercise on a treadmill moving at 2 different speeds (20 and 35 cm/s) decreased by 78 and 77% in the mdKO mice and the control mice, respectively (Fig. 2C). To examine the contraction properties of the muscle, force measurements were assessed ex vivo in isolated soleus muscle in response to electrical stimulation. Consistent with a defect in endurance exercise capacity, mdKO soleus muscle showed a moderate reduction in the specific maximal force and fatigue resistance compared to the control muscle (Fig. 2D, F). In addition, we evaluated force production by examining in situ isometric contractile properties in tibialis anterior muscle and found a lower specific maximal force in the mdKO compared with that in the control mice (Fig. 2E). However, muscle fatigue resistance was similar between the mdKO and control tibialis anterior muscles (Fig. 2F). These contractile defects are not associated with alterations in muscle weight, as shown by a little increase in the masses of mdKO tibialis anterior and soleus muscles (Fig. 2G). Our data therefore show that AMPK is necessary for normal endurance exercise performance, muscular fatigue resistance, and maximal force production.

Figure 2. Exercise performance and muscle function in WT and AMPK␣ mdKO mice. Voluntary wheel running (A). rotarod test (B), forced running (C), soleus ex vivo specific maximal force (D), tibialis in situ specific maximal force (E), tibialis and soleus fatigue resistance (F), and tibialis and soleus muscle weight (G). Results are presented as means ⫾ sem (A, n⫽12; B, n⫽11–13; C, n⫽12–13; D, n⫽14 –18; E, n⫽12; F, n⫽12–18; G, n⫽12–18). *P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.005 vs. WT. 3216

Vol. 28

July 2014

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

Damaged muscle fibers in mdKO To examine whether diminished exercise tolerance in mdKO mice is associated with a change in the overall architecture of skeletal muscle, we examined hematoxylin and eosin (H&E)-stained cross sections of tibialis anterior and gastrocnemius muscles (Fig. 3A). We found a higher number of muscle fibers with centrally located nuclei, indicative of ongoing degeneration and regeneration (25), in the mdKO tibialis and gastrocnemius muscles compared with that in the control muscles (Fig. 3B). In support of this observation, mdKO gastrocnemius muscle visualized by transmission electron microscopy clearly revealed the presence of ultrastructural deterioration within the fibers, associated with centrally located nuclei and signs of inflammatory reaction (Fig. 3C). The muscle-specific ubiquitin ligase muscle RING finger 1 (MuRF1) is viewed as an archetypal marker for muscle wasting, in that it is upregulated in skeletal muscle that is undergoing atrophy (26). We found that

expression of the atrogene MuRF1 and the atrophyrelated gene BCL2/adenovirus E1B interacting protein 3 (BNIP-3) were significantly reduced in the gastrocnemius of the mdKO mice (Fig. 3D). Furthermore, mRNA levels of FoxO3, a regulator of the atrophy program, were similar in the mdKO and control gastrocnemius muscles (Fig. 3D). These data indicate that the defects in the mdKO skeletal muscle were not due to muscle wasting. Notably, mdKO muscle tissue showed increased expression of the cytokine IL-6 but not tumor necrosis factor ␣ (TNF-␣; Fig. 3E). Loss of AMPK seems to exert a stress on myofibers and skeletal muscles, and this stress is associated with energy deficiency and low-grade muscle damage. mdKO skeletal muscles exhibit impaired mitochondrial function Given the important role of AMPK in muscle adaptation and the regulation of mitochondrial biogenesis, we examined the expression of peroxisome proliferator-

Figure 3. Altered integrity in AMPK␣ mdKO muscles, compared with that in WT. A) H&E-stained transverse sections of tibialis anterior and gastrocnemius muscles. Regenerated muscle fibers with centrally located nuclei are shown (arrows). B) Percentage of centrally nucleated fibers in tibialis anterior and gastrocnemius muscles. C) Electron micrographs of longitudinal sections of soleus and gastrocnemius muscles. D, E) Relative mRNA levels of MuRF1, BNIP-3, and FoxO3 (D) and of IL-6 and TNF-␣ (E) in the gastrocnemius, as determined by quantitative real-time RT-PCR. Results are expressed as means ⫾ sem (B, n⫽6; D, n⫽5– 6; E, n⫽6). *P ⬍ 0,05; ***P ⬍ 0.005 vs. WT. AMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

3217

activated receptor ␥ coactivator 1␣ (PGC-1␣), a key regulator of mitochondrial biogenesis in skeletal muscle. Transcript levels for PGC-1␣, as well as mitochondrial transcription factor A (mtTFA), cytochrome c (Cyt c), and ␦-aminolevulinate synthase (ALAS), were significantly decreased in the mdKO gastrocnemius (Fig. 4A). Expression of the mitochondrial markers

citrate synthase, COX1, and COX4 also diminished in the mdKO muscle (Fig. 4B). However, mitochondrialto-nuclear DNA levels unexpectedly remained unperturbed in mdKO soleus and gastrocnemius muscle (Fig. 4C). Consistently, citrate synthase activity was similar in the mdKO and control gastrocnemius muscles (Fig. 4D). Furthermore, no differences in mitochondrial

Figure 4. Mitochondrial biogenesis and respiratory function in WT and AMPK␣ mdKO mice. A) Levels of mRNAs based on quantitative real-time RT-PCR performed on RNA isolated from the gastrocnemius. B) Relative mRNA levels of mitochondrial gene expression in the gastrocnemius, as determined by quantitative real-time RT-PCR. C) Quantification of mitochondrial-tonuclear DNA in soleus and gastrocnemius muscles. D) Citrate synthase activity in the gastrocnemius. E) Mitochondrial respiratory capacity measured with the OXPHOS substrates glutamate, malate, and succinate in the presence of saturating ADP in permeabilized gastrocnemius fibers. F) ATP levels in resting gastrocnemius muscle. G) Respiration rates measured during the cumulative addition of substrates (malate, carnitine⫹palmiltoyl CoA, palmitoylcarnitine, octanoate, glycerol 3-phosphate and glutamate) in permeabilized gastrocnemius fibers. Results are expressed as the mean ⫾ se (A, B, n⫽6; C, D, n⫽5– 6; E, n⫽10; F, G, n⫽5– 6). ***P ⬍ 0.005, significantly different from WT. 3218

Vol. 28

July 2014

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

number, size, or structure were revealed by electron microscopy in soleus and gastrocnemius muscle fibers (Fig. 3C). To study the effects of the absence of AMPK on mitochondrial oxidative capacity, we measured the respiration rates of in situ mitochondria in permeabilized gastrocnemius muscle fibers. The basal respiration (V0) measured in the absence of ADP was slightly but not significantly lower in AMPK-deficient permeabilized fibers than in control fibers (Table 2). However, maximal ADP-stimulated mitochondrial oxygen consumption (state 3), supported by complex I substrates (glutamate⫹malate), was significantly reduced (40%) in AMPK-deficient fibers (Fig. 4E and Table 2). The acceptor control ratio (ACR), calculated as VCI/V0, was significantly affected in the mdKO muscle (Table 2), indicating a defect in the functional coupling between oxidation and phosphorylation. Consistent with altered oxidative capacity, the mdKO gastrocnemius had lower levels of ATP than the control muscle had (Fig. 4F). Also, with dual electron input to complex I and II by addition of a complex II-supported substrate (succinate), maximal respiration rates were significantly decreased (25%) in the mdKO compared with the control fibers (Fig. 4E and Table 2). However, when complex I was inhibited by amobarbital and complex II-supported respiration was measured, the respiration rates were similar in the mdKO and control fibers (Fig. 4E and Table 2). Notably, the substrate control ratio (SCR) for complex II (calculated as VCI⫹CII/VCI) was increased by 25% in AMPK-deficient fibers, suggesting a compensatory effect through complex II (Table 2). Furthermore, staining for succinate dehydrogenase (SDH) activity, which is a marker of mitochondrial complex II content, revealed that the lack of AMPK specifically increased the percentage of fibers with high mitochondrial complex II content in both gastrocnemius and tibialis anterior muscle (Supplemental Fig. S3B). To evaluate whether these changes in the metabolic properties of mitochondria from oxidative fibers affect the structural properties of mdKO muscle, we conducted a quantitative immunohistochemical analysis of fiber type distribution. We found an increase in the number of oxidative type I (MyHC I-positive) muscle fibers in mdKO gastrocnemius and soleus muscles, with no change in the proportion of MyHC IIA-positive fibers when comTABLE 2. Respiration rates of control and mdKO gastrocnemius permeabilized fibers Parameter

V0 VCI ACR, VCI/V0 VCI⫹II VCII SCR, VCI⫹II/VCI

Control

mdKO

3.35 ⫾ 0.34 6.02 ⫾ 0.51 2.01 ⫾ 0.24 8.21 ⫾ 0.72 3.06 ⫾ 0.50 1.37 ⫾ 0.06

2.84 ⫾ 0.25 3.66 ⫾ 0.28** 1.34 ⫾ 0.09* 6.15 ⫾ 0.30*** 3.68 ⫾ 0.25 1.82 ⫾ 0.15*

Results are expressed as means ⫾ sem (␮mol O2/min/g dry weight; n⫽10). ACR, acceptor control ratio; SCR, substrate control ratio. *P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.005 vs. WT.

AMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

pared to control muscle fibers (Supplemental Fig. S3C, D). Thus, AMPK-deficient muscles showed a marked uncoupling of metabolic characteristics, with an altered oxidative program and an increase in type I fibers. To better understand the effect of AMPK deletion on substrate utilization by the mitochondria, we measured the respiratory activity of permeabilized gastrocnemius fibers in the presence of various substrates (carbohydrates and fatty acids) added sequentially. Notably, the oxygen consumption rate stimulated by malate and ADP (complex I-supported respiration) was altered in the mdKO fibers compared with that in the controls, whereas maximal oxidative phosphorylation (oxphos) capacity in the presence of the substrates glycerol 3-phosphate, palmitoyl-carnitine, or octanoylcarnitine (a medium-chain fatty acid), was not modified (Fig. 4G). These results indicate that lack of AMPK induces a defect in mitochondrial respiration that appears mainly limited to complex I. As a result, AMPK-deficient skeletal muscle is unable to meet the energy demand required to maintain normal contractile activity. AMPK is not essential for contraction-induced glucose transport glucose uptake The need for AMPK␣2 in AICAR-induced glucose uptake is well established (12, 15, 27). On the other hand, studies investigating contraction-induced glucose uptake suggest only a partial or no role of AMPK␣2. It has been suggested that compensatory signaling through AMPK␣1 rescues contraction-induced glucose transport in the AMPK␣2-deficient muscle. To address the question of the role of AMPK in contraction-induced glucose uptake, we measured in vitro contraction-induced glucose uptake in soleus and EDL muscles from control and mdKO male and female mice. Basal glucose uptake was similar in the control and mdKO muscles (both soleus and EDL) from the female mice (Fig. 5A, B). When muscles were contracted for 1 s every 15 s, glucose uptake increased comparably in the female control and mdKO muscles (Fig. 5A, B). Even when challenging the muscles with a more intense contraction protocol (2 s of contractions every 15 s), no differences between genotypes were observed in glucose uptake in muscle of female animals (Supplemental Fig. S4A, B). Similar observations were made in the EDL muscle of the male mice (Fig. 5D and Supplemental Fig. S4D). However, basal glucose uptake in soleus from the male mdKO mice was slightly lower than their control littermates. Indeed, both the mild and moderate contraction protocols elicited an increase in glucose uptake in control and mdKO soleus, but the level reached in the mdKO muscle was lower than in the control (Fig. 5C and Supplemental Fig. S4C). Independent of the mouse’s sex, recordings of force production during the 10 min of electrical stimulation showed no differences between control and mdKO muscles when contracted for 1 s every 15 s (data not 3219

Figure 5. In vitro contraction-induced muscle glucose uptake. AMPK␣ WT and mdKO soleus and EDL muscles were removed from anesthetized mice and incubated in Krebs-Ringer solution. After preincubation, muscles either rested or were subjected to electrically stimulated contraction (mild protocol: 1 s of contractions per 15 s) for 10 min. During contractions, 2-deoxyglucose uptake was measured in female soleus (A) and EDL (B) muscles and in male soleus (C) and EDL (D) muscles. Results are expressed as means ⫾ sem (A, B, n⫽12; C, D, n⫽20 –29). ***P ⬍ 0.001 vs. basal within genotype; #P ⬍ 0.05; ###P ⬍ 0.001 vs. WT within intervention.

shown). When the muscles were stimulated for 2 s every 15 s, similar force production was recorded in EDL muscles from the control and mdKO mice, but force production tended to be lower in the mdKO soleus muscle (data not shown). Both basal and contractioninduced AMPK signaling was greatly reduced in the mdKO muscles, as reflected by measurements of both AMPK-Thr172 and ACC-Ser212 phosphorylation (Supplemental Fig. S1). At submaximal and maximal insulin concentrations, glucose transport was stimulated similarly in control and mdKO muscles, independent of gender and muscle type (Fig. 6). These data indicate that insulinregulated glucose uptake in muscles of mdKO mice is not affected by the deletion of the AMPK␣1 and -␣2 subunits, which further is in line with the finding that whole-body glucose tolerance and insulin sensitivity is preserved in mdKO mice (Supplemental Fig. S2B, C).

DISCUSSION AMPK is a crucial regulator of skeletal muscle energy metabolism through the regulation of glucose uptake, lipid oxidation, and mitochondrial biogenesis (3). Several studies have established AMPK as an essential mediator of the metabolic response to exercise by using AMPK␣2-KD mice (10, 12, 28, 29) or mice lacking AMPK␤1 and AMPK␤2 in both skeletal and heart muscles (13). Considering the known role of AMPK in heart energy metabolism, it remains possible that the reduction in the exercise capacity of these mice reflects 3220

Vol. 28

July 2014

impairments in cardiac muscle rather than in skeletal muscle function (30). However, expression of inactive AMPK␣2 in the heart is not sufficient to alter cardiac function during physiological exercise conditions (10). An alternative explanation for limited exercise capacity is a reduction in blood flow and muscle capillary density (10). In the current study, we have shown that mice lacking both AMPK␣1 and AMPK␣2, specifically in skeletal muscle, are intolerant to exercise, as illustrated by marked reductions in voluntary wheel activity and endurance during treadmill running. Our data underline the idea that skeletal muscle AMPK is required during metabolic adaptation to acute exercise (10, 13). In addition, many studies have documented a potential role for AMPK in the phenotypic changes occurring in skeletal muscle with endurance training, such as mitochondrial biogenesis (13, 31). Indeed, the effects of the AMPK-activating compounds AICAR and resveratrol, as well as those of the phosphocreatinedepleting agent ␤-guanadinopropionic acid (␤-GPA), on muscle mitochondrial content and gene expression are blunted in AMPK␣2-KD and AMPK␣2-KO mice (31–33). Conversely, mitochondrial markers are increased in mouse models with muscle-specific expression of gain-of-function AMPK␥1R70Q and ␥3R225Q or ␥3R200Q mutants (34 –36). The data from the current study, however, suggest that mitochondrial function rather than the number of mitochondria is altered in mdKO muscles. This result is consistent with similar findings reported for skeletal and cardiac muscle of AMPK␣2-KO (22, 37) and AMPK␣2-KD (10) mice.

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

Figure 6. In vitro insulin-induced muscle glucose uptake. Soleus and EDL muscles from anesthetized AMPK␣ WT and mdKO mice were removed and incubated in a Krebs-Ringer solution. Muscles were preincubated with or without insulin in different concentrations: low submaximal insulin concentration, 100/200 ␮U/ml (females/males); high submaximal insulin concentration, 200/400 ␮U/ml (females/males); maximal insulin concentration, 10 mU/ml. After preincubation, 2-deoxyglucose uptake was measured by addition of 1 mM [3H]2-deoxyglucose and 7 mM [14C]mannitol to the incubation medium. Results are expressed as means ⫾ se (n⫽6 –10). ††P ⬍ 0.01, †††P ⬍ 0.001 vs. basal; §§P ⬍ 0.01, §§§P ⬍ 0.001 vs. low submaximal insulin concentration; ¤¤¤P ⬍ 0.001 vs. high submaximal insulin concentration; (§)0.05 ⱕ P ⬍ 0.10.

A key downstream mediator of the effect of AMPK activation on mitochondrial biogenesis is the critical regulator of mitochondrial biogenesis PGC-1␣, as shown by the loss of AICAR’s effect on muscle mitochondrial gene expression in the absence of PGC-1␣ (38) and the induction of PGC-1␣ mRNA expression in AMPK␣2-KD mice (27). In mdKO mice, PGC-1␣ expression is reduced by 40%, reflecting the AMPKmediated regulation of PGC-1␣ expression (38 – 40) and PGC-1␣ autoregulation on its own promoter (38). The link between AMPK and PGC-1␣ is also highlighted by the exercise intolerance observed in mdKO (current study), muscle-specific AMPK␤1␤2-KO (13), AMPK␣2-KD, (12), and PGC-1␣-KO mice (41). Nevertheless, there has been controversy regarding the precise role of PGC-1 coactivators in muscle function, as muscle-specific PGC-1␣- and PGC-1␤-KO mice have preserved locomotor activity and muscle fatigability (42), indicating that other currently unidentified factors are implicated in exercise-induced adaptations (43). Of note, the role of AMPK in the adaptations to exercise training also remains uncertain. It has been reported that changes in mitochondrial markers with exercise training were not prevented by the deletion of AMPK␣2 (33, 44). In accordance with this result, deletion of LKB1, the primary upstream AMPK␣2 kiAMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

nase in skeletal muscle, does not impair accumulation of some mitochondrial proteins after exercise training, indicating the involvement of non-AMPK␣2-dependent pathways in muscle adaptation to training (45). Mitochondrial dysfunction in mdKO skeletal muscle is the likely explanation for muscle injury and limited exercise capacity in the absence of AMPK. In our study, we found that mitochondrial respiration supported by complex I substrates was reduced in permeabilized skeletal muscle fibers of mdKO mice. The advantage of the study of mitochondrial respiration in fibers is the direct assessment of oxidative capacity of the muscle. In these conditions, mitochondria are kept in their cellular environment, and intrinsic oxidative capacities can be measured with nonlimiting amounts of substrates. We showed a significant difference in state 3 respiration with malate and ADP and also a 40% reduction of respiration with electron flux through complex I (glutamate, malate, and ADP) and a 25% reduction with electron flux through complex I⫹II (malate, glutamate, ADP and succinate) in the mdKO compared with the control mice. These findings are supported by previous studies in AMPK␣2-KD mice, in which mitochondrial ATP production and state 3 respiration were decreased in permeabilized fibers, indicative of mitochondrial dysfunction (46, 47). Thus, it may be antici3221

pated that the observed oxphos limitation is the underlying mechanism for the impaired exercise tolerance and muscle fatigability seen in mdKO mice, and our results suggest that mdKO mice are metabolically inflexible. It is important to note that reduced oxphos capacity in mdKO muscle fibers occurs without alteration of markers of muscle mitochondrial content. This observation is consistent with those in previous studies from AMPK␣2-KD mice showing impaired activities of the electron transport chain complexes I and IV, despite mitochondrial content similar to that in wild-type (WT) control mice, as determined by mitochondrial density, mitochondrial DNA, mitochondrial gene expression, and citrate synthase activity (10, 31). However, defects in mitochondrial gene and protein expression were reported in resting muscle from AMPK␣2KO, AMPK␣2-KD, and muscle-specific AMPK␤1␤2-KO mice (13, 33, 46). In the current study, we showed that AMPK is dispensable for the regulation of baseline mitochondrial muscle content and that AMPK␣ deletion is sufficient to impair mitochondrial function. This is reminiscent of recent studies in muscle-specific PGC1␣- and PGC-1␤-KO mice showing that mitochondrial oxidative capacity in muscle is clearly dissociable from mitochondrial content (42). Thus, reduction of AMPK activity may be sufficient to impair mitochondrial function per se, without altering mitochondrial biogenesis. The functional defect of mitochondrial respiration by complex I may be related to alteration in the phospholipid content at the mitochondrial inner membrane, as reported in the heart of AMPK␣2-KO mice (22). Levels of mitochondrial protein acetylation may also represent an important mechanism that regulates overall mitochondrial function (48, 49). Sirtuin 3 (Sirt3), the major mitochondrial NAD-dependent deacetylase, selectively regulates the acetylation of complex I with little or no effect on complex II (50). In addition, it has been recently demonstrated that Sirt3-mediated deacetylation of ATP synthase complex proteins is necessary to meet the energy demands of exercise (51). In this context, it should be noted that activation of AMPK causes an increase in the NAD⫹/NADH ratio, consistent with a positive feedback loop for the regulation of Sirt activity (40). However, further studies determining the mechanisms responsible for decreased mitochondrial respiration and ATP production are warranted. Another important finding in this study is the role of AMPK␣ in the regulation of acute exercise-induced skeletal muscle glucose transport. Previous studies examining the role of AMPK in the regulation of exerciseinduced skeletal muscle glucose transport have shown conflicting results. Some studies have reported normal contraction-induced glucose uptake in AMPK␣2-KD mice (10, 28), but others have found altered glucose transport in AMPK␣2-KD (11, 12, 52) and musclespecific AMPK␤1␤2-KO (13) mice. Notably, it was also demonstrated that LKB1 is not crucial for glucose uptake during physiological exercise or during moderate contractions induced by electrical stimulation (53), although LKB1 dependence has been reported after 3222

Vol. 28

July 2014

intense electrical stimulation of muscle (54). The in vitro contraction results demonstrated that AMPK␣ subunits were needed in the soleus of male but not female mice. This finding of a sex-related difference in the requirement of AMPK␣ subunits for contractioninduced glucose uptake in soleus muscle is unexpected, but may help explain some of the discrepancy in the literature on the role of AMPK in contraction-mediated glucose transport. Notably, smaller AMPK activation by exercise in women than in men has been reported, reflecting a sex difference in the maintenance of muscle cellular energy balance during exercise (55). Conversely, lack of AMPK␣ subunits did not affect contraction-induced glucose uptake in EDL muscles from either male or female mice, suggesting that the residual AMPK␣1 activity observed in previous mouse models lacking functional AMPK␣2 (12) or LKB1 in skeletal muscle (54) is not essential in maintaining a normal contraction-induced glucose uptake in the soleus muscle, despite reports showing impaired regulation of contraction-stimulated glucose uptake in AMPK␣1-deficient soleus muscle (16, 27). However, muscle-specific AMPK␤1␤2-KO mice, which also lack AMPK␣1 and AMPK␣2 subunits in skeletal muscle and exhibit a defective mitochondria phenotype, have been found to have impaired contraction-induced glucose transport, in contrast to the mdKO mice (13). The reasons for the differences between these mouse models are not clear at present, but may be explained by differences in electrical stimulation parameters and contraction time (5 min vs. 10 min). Indeed, it has been reported that the role of AMPK in contraction-induced glucose transport can be more readily observed when using lower force production and smaller contraction time (11). In summary, muscle-specific AMPK␣1␣2-deficient mice exhibit impaired endurance exercise capacity. However, our results clearly show that the AMPK␣1 and AMPK␣2 isoforms are dispensable for contraction-induced skeletal muscle glucose transport in EDL but not soleus muscle from male mice only. We conclude that the marked decrease in exercise capacity in our model was most likely due to defects in oxphos. Taken together, our results indicate that AMPK regulates muscle metabolic adaptation through the control of muscle mitochondrial oxidative capacity, but not of mitochondrial content. The authors thank G. Hardie (University of Dundee, Dundee, UK) for providing the anti-AMPK␣1 and antiAMPK␣2 antibodies, J. Melki (INSERM U788, Paris, France) for kindly providing HSA-Cre transgenic mice for breeding, and A. Schmitt (Cochin Institute electron microscopy facility) for transmission electron micrographs. This work was supported by grants from the European Commission integrated project (LSHM-CT-2004-005272/EXGENESIS) and Agence Nationale de la Recherche (ANR) to B.V., Association Française contre les Myopathies (AFM) to B.V. and P.M., Fondation pour la Recherche Médicale (FRM) to R.V.C., and the Canadian Institutes for Health Research to A.M. Region Ile de France contributed to the Cochin Institute animal care facility. This work was also performed as a part of the program of the University of Southern Denmark (UNIK) program

The FASEB Journal 䡠 www.fasebj.org

LANTIER ET AL.

Food, Fitness, and Pharma for Health and Disease (http:// www.foodfitnesspharma.ku.dk). The UNIK project is supported by the Danish Ministry of Science, Technology, and Innovation. The study was funded by the Danish Council for Independent Research Medical Sciences (FSS) and the Novo Nordisk Foundation. E.S.-A. was recipient of a Bourses Doctorales de Mobilité de l’Université (BDMU) fellowship from University Paris Descartes. J.T.T. was supported by a postdoctoral fellowship from The Danish Agency for Science, Technology, and Innovation. R.V.C. is a senior scientist at CNRS.

REFERENCES 1. 2.

3. 4.

5.

6.

7.

8. 9.

10.

11.

12.

13.

14.

Hardie, D. G., Ross, F. A., and Hawley, S. A. (2012) AMPK: a nutrient and energy sensor that maintains energy homeostasis. Nat. Rev. Mol. Cell Biol. 13, 251–262 Fogarty, S., Hawley, S. A., Green, K. A., Saner, N., Mustard, K. J., and Hardie, D. G. (2010) Calmodulin-dependent protein kinase kinase-beta activates AMPK without forming a stable complex: synergistic effects of Ca2⫹ and AMP. Biochem. J. 426, 109 –118 O’Neill, H. M. (2013) AMPK and exercise: glucose uptake and insulin sensitivity. Diabetes Metab. J. 37, 1–21 O’Neill, H. M., Holloway, G. P., and Steinberg, G. R. (2013) AMPK regulation of fatty acid metabolism and mitochondrial biogenesis: implications for obesity. Mol. Cell. Endocrinol. 366, 135–151 Treebak, J. T., Pehmoller, C., Kristensen, J. M., Kjobsted, R., Birk, J. B., Schjerling, P., Richter, E. A., Goodyear, L. J., and Wojtaszewski, J. F. (2014) Acute exercise and physiological insulin induce distinct phosphorylation signatures on TBC1D1 and TBC1D4 in human skeletal muscle. J. Physiol. 592, 351–375 Jensen, T. E., Wojtaszewski, J. F., and Richter, E. A. (2009) AMP-activated protein kinase in contraction regulation of skeletal muscle metabolism: necessary and/or sufficient? Acta Physiol. (Oxf.) 196, 155–174 Treebak, J. T., Birk, J. B., Hansen, B. F., Olsen, G. S., and Wojtaszewski, J. F. (2009) A-769662 activates AMPK beta1containing complexes but induces glucose uptake through a PI3-kinase-dependent pathway in mouse skeletal muscle. Am. J. Physiol. Cell Physiol. 297, C1041–C1052 Birk, J. B., and Wojtaszewski, J. F. (2006) Predominant alpha2/ beta2/gamma3 AMPK activation during exercise in human skeletal muscle. J. Physiol. 577, 1021–1032 Treebak, J. T., Birk, J. B., Rose, A. J., Kiens, B., Richter, E. A., and Wojtaszewski, J. F. (2007) AS160 phosphorylation is associated with activation of alpha2beta2gamma1— but not alpha2beta2gamma3—AMPK trimeric complex in skeletal muscle during exercise in humans. Am. J. Physiol. Endocrinol. Metab. 292, E715–E722 Lee-Young, R. S., Griffee, S. R., Lynes, S. E., Bracy, D. P., Ayala, J. E., McGuinness, O. P., and Wasserman, D. H. (2009) Skeletal muscle AMP-activated protein kinase is essential for the metabolic response to exercise in vivo. J. Biol. Chem. 284, 23925– 23934 Lefort, N., St-Amand, E., Morasse, S., Cote, C. H., and Marette, A. (2008) The ␣-subunit of AMPK is essential for submaximal contraction-mediated glucose transport in skeletal muscle in vitro. Am. J. Physiol. Endocrinol. Metab. 295, E1447–E1454 Mu, J., Brozinick, J. T., Jr., Valladares, O., Bucan, M., and Birnbaum, M. J. (2001) A role for AMP-activated protein kinase in contraction- and hypoxia-regulated glucose transport in skeletal muscle. Mol. Cell 7, 1085–1094 O’Neill, H. M., Maarbjerg, S. J., Crane, J. D., Jeppesen, J., Jorgensen, S. B., Schertzer, J. D., Shyroka, O., Kiens, B., van Denderen, B. J., Tarnopolsky, M. A., Kemp, B. E., Richter, E. A., and Steinberg, G. R. (2011) AMP-activated protein kinase (AMPK) beta1beta2 muscle null mice reveal an essential role for AMPK in maintaining mitochondrial content and glucose uptake during exercise. Proc. Natl. Acad. Sci. U.S.A. 108, 16092– 16097 Abbott, M. J., Bogachus, L. D., and Turcotte, L. P. (2011) AMPKalpha2 deficiency uncovers time dependency in the regulation of contraction-induced palmitate and glucose uptake in mouse muscle. J. Appl. Physiol. (1985) 111, 125–134

AMPK IN SKELETAL MUSCLE METABOLIC ADAPTATION

15.

16.

17.

18. 19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Fujii, N., Hirshman, M. F., Kane, E. M., Ho, R. C., Peter, L. E., Seifert, M. M., and Goodyear, L. J. (2005) AMP-activated protein kinase alpha2 activity is not essential for contraction- and hyperosmolarity-induced glucose transport in skeletal muscle. J. Biol. Chem. 280, 39033–39041 Jensen, T. E., Schjerling, P., Viollet, B., Wojtaszewski, J. F., and Richter, E. A. (2008) AMPK alpha1 activation is required for stimulation of glucose uptake by twitch contraction, but not by H2O2, in mouse skeletal muscle. PLoS One 3, e2102 Friedrichsen, M., Mortensen, B., Pehmoller, C., Birk, J. B., and Wojtaszewski, J. F. (2013) Exercise-induced AMPK activity in skeletal muscle: role in glucose uptake and insulin sensitivity. Mol. Cell. Endocrinol. 366, 204 –214 Miniou, P., Tiziano, D., Frugier, T., Roblot, N., Le Meur, M., and Melki, J. (1999) Gene targeting restricted to mouse striated muscle lineage. Nucleic Acids Res. 27, e27 Hadj-Said, W., Bangratz, M., Vignaud, A., Chatonnet, A., ButlerBrowne, G., Nicole, S., Agbulut, O., and Ferry, A. (2012) Effect of locomotor training on muscle performance in the context of nerve-muscle communication dysfunction. Muscle Nerve 45, 567– 577 Joanne, P., Hourde, C., Ochala, J., Cauderan, Y., Medja, F., Vignaud, A., Mouisel, E., Hadj-Said, W., Arandel, L., Garcia, L., Goyenvalle, A., Mounier, R., Zibroba, D., Sakamato, K., ButlerBrowne, G., Agbulut, O., and Ferry, A. (2012) Impaired adaptive response to mechanical overloading in dystrophic skeletal muscle. PLoS One 7, e35346 Mettauer, B., Zoll, J., Sanchez, H., Lampert, E., Ribera, F., Veksler, V., Bigard, X., Mateo, P., Epailly, E., Lonsdorfer, J., and Ventura-Clapier, R. (2001) Oxidative capacity of skeletal muscle in heart failure patients versus sedentary or active control subjects. J. Am. Coll. Cardiol. 38, 947–954 Athea, Y., Viollet, B., Mateo, P., Rousseau, D., Novotova, M., Garnier, A., Vaulont, S., Wilding, J. R., Grynberg, A., Veksler, V., Hoerter, J., and Ventura-Clapier, R. (2007) AMP-activated protein kinase alpha2 deficiency affects cardiac cardiolipin homeostasis and mitochondrial function. Diabetes 56, 786 –794 Viollet, B., Andreelli, F., Jorgensen, S. B., Perrin, C., Geloen, A., Flamez, D., Mu, J., Lenzner, C., Baud, O., Bennoun, M., Gomas, E., Nicolas, G., Wojtaszewski, J. F., Kahn, A., Carling, D., Schuit, F. C., Birnbaum, M. J., Richter, E. A., Burcelin, R., and Vaulont, S. (2003) The AMP-activated protein kinase alpha2 catalytic subunit controls whole-body insulin sensitivity. J. Clin. Invest. 111, 91–98 Mounier, R., Theret, M., Arnold, L., Cuvellier, S., Bultot, L., Goransson, O., Sanz, N., Ferry, A., Sakamoto, K., Foretz, M., Viollet, B., and Chazaud, B. (2013) AMPKalpha1 regulates macrophage skewing at the time of resolution of inflammation during skeletal muscle regeneration. Cell Metab. 18, 251–264 Morgan, J. E., Hoffman, E. P., and Partridge, T. A. (1990) Normal myogenic cells from newborn mice restore normal histology to degenerating muscles of the mdx mouse. J. Cell Biol. 111, 2437–2449 Bodine, S. C., Latres, E., Baumhueter, S., Lai, V. K., Nunez, L., Clarke, B. A., Poueymirou, W. T., Panaro, F. J., Na, E., Dharmarajan, K., Pan, Z. Q., Valenzuela, D. M., DeChiara, T. M., Stitt, T. N., Yancopoulos, G. D., and Glass, D. J. (2001) Identification of ubiquitin ligases required for skeletal muscle atrophy. Science 294, 1704 –1708 Jorgensen, S. B., Viollet, B., Andreelli, F., Frosig, C., Birk, J. B., Schjerling, P., Vaulont, S., Richter, E. A., and Wojtaszewski, J. F. (2004) Knockout of the alpha2 but not alpha1 5=-AMP-activated protein kinase isoform abolishes 5-aminoimidazole-4-carboxamide-1-beta-4-ribofuranosidebut not contraction-induced glucose uptake in skeletal muscle. J. Biol. Chem. 279, 1070 –1079 Maarbjerg, S. J., Jorgensen, S. B., Rose, A. J., Jeppesen, J., Jensen, T. E., Treebak, J. T., Birk, J. B., Schjerling, P., Wojtaszewski, J. F., and Richter, E. A. (2009) Genetic impairment of AMPKalpha2 signaling does not reduce muscle glucose uptake during treadmill exercise in mice. Am. J. Physiol. Endocrinol. Metab. 297, E924 –E934 Lee-Young, R. S., Ayala, J. E., Fueger, P. T., Mayes, W. H., Kang, L., and Wasserman, D. H. (2011) Obesity impairs skeletal muscle AMPK signaling during exercise: role of AMPKalpha2 in the regulation of exercise capacity in vivo. Int. J. Obes. (Lond.) 35, 982–989

3223

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

3224

Russell, R. R., 3rd, Li, J., Coven, D. L., Pypaert, M., Zechner, C., Palmeri, M., Giordano, F. J., Mu, J., Birnbaum, M. J., and Young, L. H. (2004) AMP-activated protein kinase mediates ischemic glucose uptake and prevents postischemic cardiac dysfunction, apoptosis, and injury. J. Clin. Invest. 114, 495–503 Zong, H., Ren, J. M., Young, L. H., Pypaert, M., Mu, J., Birnbaum, M. J., and Shulman, G. I. (2002) AMP kinase is required for mitochondrial biogenesis in skeletal muscle in response to chronic energy deprivation. Proc. Natl. Acad. Sci. U.S.A. 99, 15983–15987 Um, J. H., Park, S. J., Kang, H., Yang, S., Foretz, M., McBurney, M. W., Kim, M. K., Viollet, B., and Chung, J. H. (2010) AMP-activated protein kinase-deficient mice are resistant to the metabolic effects of resveratrol. Diabetes 59, 554 –563 Jorgensen, S. B., Treebak, J. T., Viollet, B., Schjerling, P., Vaulont, S., Wojtaszewski, J. F., and Richter, E. A. (2007) Role of AMPKalpha2 in basal, training-, and AICAR-induced GLUT4, hexokinase II, and mitochondrial protein expression in mouse muscle. Am. J. Physiol. Endocrinol. Metab. 292, E331–E339 Barre, L., Richardson, C., Hirshman, M. F., Brozinick, J., Fiering, S., Kemp, B. E., Goodyear, L. J., and Witters, L. A. (2007) Genetic model for the chronic activation of skeletal muscle AMP-activated protein kinase leads to glycogen accumulation. Am. J. Physiol. Endocrinol. Metab. 292, E802–E811 Rockl, K. S., Hirshman, M. F., Brandauer, J., Fujii, N., Witters, L. A., and Goodyear, L. J. (2007) Skeletal muscle adaptation to exercise training: AMP-activated protein kinase mediates muscle fiber type shift. Diabetes 56, 2062–2069 Scheffler, T. L., Scheffler, J. M., Park, S., Kasten, S. C., Wu, Y., McMillan, R. P., Hulver, M. W., Frisard, M. I., and Gerrard, D. E. (2013) Fiber hypertrophy and increased oxidative capacity can occur simultaneously in pig glycolytic skeletal muscle. Am. J. Physiol. Cell Physiol. 306, C354 –C363 Klein, D. K., Pilegaard, H., Treebak, J. T., Jensen, T. E., Viollet, B., Schjerling, P., and Wojtaszewski, J. F. (2007) Lack of AMPKalpha2 enhances pyruvate dehydrogenase activity during exercise. Am. J. Physiol. Endocrinol. Metab. 293, E1242–E1249 Jager, S., Handschin, C., St-Pierre, J., and Spiegelman, B. M. (2007) AMP-activated protein kinase (AMPK) action in skeletal muscle via direct phosphorylation of PGC-1alpha. Proc. Natl. Acad. Sci. U.S.A. 104, 12017–12022 Canto, C., Gerhart-Hines, Z., Feige, J. N., Lagouge, M., Noriega, L., Milne, J. C., Elliott, P. J., Puigserver, P., and Auwerx, J. (2009) AMPK regulates energy expenditure by modulating NAD⫹ metabolism and SIRT1 activity. Nature 458, 1056 –1060 Canto, C., Jiang, L. Q., Deshmukh, A. S., Mataki, C., Coste, A., Lagouge, M., Zierath, J. R., and Auwerx, J. (2010) Interdependence of AMPK and SIRT1 for metabolic adaptation to fasting and exercise in skeletal muscle. Cell Metab. 11, 213–219 Handschin, C., Chin, S., Li, P., Liu, F., Maratos-Flier, E., Lebrasseur, N. K., Yan, Z., and Spiegelman, B. M. (2007) Skeletal muscle fiber-type switching, exercise intolerance, and myopathy in PGC-1alpha muscle-specific knock-out animals. J. Biol. Chem. 282, 30014 –30021 Rowe, G. C., Patten, I. S., Zsengeller, Z. K., El-Khoury, R., Okutsu, M., Bampoh, S., Koulisis, N., Farrell, C., Hirshman, M. F., Yan, Z., Goodyear, L. J., Rustin, P., and Arany, Z. (2013) Disconnecting mitochondrial content from respiratory chain capacity in PGC-1-deficient skeletal muscle. Cell Rep. 3, 1449 – 1456 Leick, L., Wojtaszewski, J. F., Johansen, S. T., Kiilerich, K., Comes, G., Hellsten, Y., Hidalgo, J., and Pilegaard, H. (2008) PGC-1alpha is not mandatory for exercise- and training-induced

Vol. 28

July 2014

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

adaptive gene responses in mouse skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 294, E463–E474 Jorgensen, S. B., Wojtaszewski, J. F., Viollet, B., Andreelli, F., Birk, J. B., Hellsten, Y., Schjerling, P., Vaulont, S., Neufer, P. D., Richter, E. A., and Pilegaard, H. (2005) Effects of alpha-AMPK knockout on exercise-induced gene activation in mouse skeletal muscle. FASEB J. 19, 1146 –1148 Tanner, C. B., Madsen, S. R., Hallowell, D. M., Goring, D. M., Moore, T. M., Hardman, S. E., Heninger, M. R., Atwood, D. R., and Thomson, D. M. (2013) Mitochondrial and performance adaptations to exercise training in mice lacking skeletal muscle LKB1. Am. J. Physiol. Endocrinol. Metab. 305, E1018 –E1029 Larsen, S., Kristensen, J. M., Stride, N., Wojtaszewski, J. F., Helge, J. W., and Dela, F. (2012) Skeletal muscle mitochondrial respiration in AMPKalpha2 kinase-dead mice. Acta Physiol. (Oxf.) 205, 314 –320 Kristensen, J. M., Larsen, S., Helge, J. W., Dela, F., and Wojtaszewski, J. F. (2013) Two weeks of metformin treatment enhances mitochondrial respiration in skeletal muscle of AMPK kinase dead but not wild type mice. PLoS One 8, e53533 Kim, S. C., Sprung, R., Chen, Y., Xu, Y., Ball, H., Pei, J., Cheng, T., Kho, Y., Xiao, H., Xiao, L., Grishin, N. V., White, M., Yang, X. J., and Zhao, Y. (2006) Substrate and functional diversity of lysine acetylation revealed by a proteomics survey. Mol. Cell 23, 607–618 Shinmura, K., Tamaki, K., Sano, M., Nakashima-Kamimura, N., Wolf, A. M., Amo, T., Ohta, S., Katsumata, Y., Fukuda, K., Ishiwata, K., Suematsu, M., and Adachi, T. (2011) Caloric restriction primes mitochondria for ischemic stress by deacetylating specific mitochondrial proteins of the electron transport chain. Circ. Res. 109, 396 –406 Ahn, B. H., Kim, H. S., Song, S., Lee, I. H., Liu, J., Vassilopoulos, A., Deng, C. X., and Finkel, T. (2008) A role for the mitochondrial deacetylase Sirt3 in regulating energy homeostasis. Proc. Natl. Acad. Sci. U.S.A. 105, 14447–14452 Vassilopoulos, A., Pennington, D. J., Andresson, T., Rees, D., Fearnley, I., Ham, A., Yan, Y., Flynn, C. R., Jones, K., Kim, H. S., Deng, C., Walker, J., and Gius, D. (2014) SIRT3 deacetylates ATP synthase F1 complex proteins in response to nutrient and exercise-induced stress. [E-pub ahead of print] Antioxid. Redox Signal. doi: 10.1089/ars.2013.5420 Jensen, T. E., Rose, A. J., Jorgensen, S. B., Brandt, N., Schjerling, P., Wojtaszewski, J. F., and Richter, E. A. (2007) Possible CaMKK-dependent regulation of AMPK phosphorylation and glucose uptake at the onset of mild tetanic skeletal muscle contraction. Am. J. Physiol. Endocrinol. Metab. 292, E1308 – E1317 Jeppesen, J., Maarbjerg, S. J., Jordy, A. B., Fritzen, A. M., Pehmoller, C., Sylow, L., Serup, A. K., Jessen, N., Thorsen, K., Prats, C., Qvortrup, K., Dyck, J. R., Hunter, R. W., Sakamoto, K., Thomson, D. M., Schjerling, P., Wojtaszewski, J. F., Richter, E. A., and Kiens, B. (2013) LKB1 regulates lipid oxidation during exercise independently of AMPK. Diabetes 62, 1490 –1499 Sakamoto, K., McCarthy, A., Smith, D., Green, K. A., Grahame Hardie, D., Ashworth, A., and Alessi, D. R. (2005) Deficiency of LKB1 in skeletal muscle prevents AMPK activation and glucose uptake during contraction. EMBO J. 24, 1810 –1820 Roepstorff, C., Thiele, M., Hillig, T., Pilegaard, H., Richter, E. A., Wojtaszewski, J. F., and Kiens, B. (2006) Higher skeletal muscle alpha2AMPK activation and lower energy charge and fat oxidation in men than in women during submaximal exercise. J. Physiol. 574, 125–138

The FASEB Journal 䡠 www.fasebj.org

Received for publication February 9, 2014. Accepted for publication March 10, 2014.

LANTIER ET AL.