AMPKγ3 is dispensable for skeletal muscle hypertrophy induced by functional overload

Abstract

Mechanisms regulating skeletal muscle growth involve a balance between the activity of serine/threonine protein kinases, including the mammalian target of rapamycin (mTOR) and 5′-AMP-activated protein kinase (AMPK). The contribution of different AMPK subunits to the regulation of cell growth size remains inadequately characterized. Using AMPKγ3 mutant-overexpressing transgenic Tg-Prkag3^225Q and AMPKγ3-knockout (Prkag3^−/−) mice, we investigated the requirement for the AMPKγ3 isoform in functional overload-induced muscle hypertrophy. Although the genetic disruption of the γ3 isoform did not impair muscle growth, control sham-operated AMPKγ3-transgenic mice displayed heavier plantaris muscles in response to overload hypertrophy and underwent smaller mass gain and lower Igf1 expression compared with wild-type littermates. The mTOR signaling pathway was upregulated with functional overload but unchanged between genetically modified animals and wild-type littermates. Differences in AMPK-related signaling pathways between transgenic, knockout, and wild-type mice did not impact muscle hypertrophy. Glycogen content was increased following overload in wild-type mice. In conclusion, our functional, transcriptional, and signaling data provide evidence against the involvement of the AMPKγ3 isoform in the regulation of skeletal muscle hypertrophy. Thus, the AMPKγ3 isoform is dispensable for functional overload-induced muscle growth. Mechanical loading can override signaling pathways that act as negative effectors of mTOR signaling and consequently promote skeletal muscle hypertrophy.

specific organ plasticity is vital for health and disease and allows the organism to respond to a wide range of stressors. Skeletal muscle, a highly malleable and responsive tissue, is highly adaptive to energy substrate availability, contractile activity, and exercise as well as various other environmental stimuli. Lifestyle interventions such as endurance- and resistance-based exercise programs improve whole body insulin sensitivity in metabolic disease by targeting skeletal muscle (9). Alterations in mRNA expression, protein abundance, and enzyme activity resulting from the repeated bouts of exercise or muscle contraction lead to chronic functional adaptations of skeletal muscle to support greater energy demands (12). In response to resistance-based exercise training, these adaptations include increased muscle fiber size, muscle strength, and power as well as myofibrillar protein synthesis, but the molecular machinery responsible for these changes is complex and incompletely understood.

5′-AMP-activated protein kinase (AMPK) and the mammalian target of rapamycin (mTOR) are energy-sensing protein kinases that govern metabolic processes to preserve cellular homeostasis. In response to metabolic stressors that decrease ATP levels, such as starvation and exercise, AMPK acts as a molecular switch to restore ATP levels (15). Upon activation, AMPK activates energy-producing pathways such as glucose transport and lipid oxidation and blunts energy-consuming processes such as protein and lipid synthesis. AMPK consists of a catalytic α-subunit, a regulatory β-subunit, and a regulatory γ-subunit. The combination of different isoforms (α1/α2, β1/β2, γ1/γ2/γ3) of the subunits leads to a possibility of 12 different heterotrimers. Whereas the AMPKα1 subunit plays a role in skeletal muscle hypertrophy (29), the AMPKα2 subunit primarily regulates metabolism (8, 26, 29). The γ3 isoform is encoded by the PRKAG3 gene and constitutes the predominate isoform expressed in glycolytic skeletal muscle fibers (24). Genetic approaches have provided evidence that the AMPKγ3 isoform plays a key role in carbohydrate and lipid metabolism (5), but its role in modulating both growth and hypertrophic responses is elusive.

Protein translation and skeletal muscle growth are mediated by mTOR activation in response to resistance exercise, nutrient availability, or stimulation with growth factors (7). mTOR is downstream of the IGF-I/PI3K/Akt pathway and constitutes part of the mTOR complex 1. mTOR activation is regulated by tuberous sclerosis complex 2 (TSC2), a protein complex downstream of Akt, and by regulatory-associated protein of mTOR (Raptor) (14, 19). AMPK inactivates mTOR by phosphorylating TSC2 (19) and Raptor (14). Thus, the balance between mTOR and AMPK signals modulates metabolic and growth responses at the cellular level. For example, 5-aminoimidazole-4-carboxamide-1-β-d-ribofuranoside (AICAR)-mediated AMPK activation inhibits protein synthesis (8), whereas AMPK repression increases protein synthesis (23), indicating that AMPK activation can impinge on skeletal muscle growth by modulating mTOR signaling. Furthermore, genetic deletion of the α1-subunit in mouse skeletal muscle augments plantaris muscle hypertrophy following functional overload (29).

Given the role for AMPK in attenuating cell growth, we sought to examine the role of the AMPKγ3 regulatory subunit in the control of skeletal muscle in response to hypertrophic stimuli using transgenic mice overexpressing the gain-of-function R225Q mutation in the AMPKγ3 subunit or AMPKγ3 subunit knockout mice. Skeletal muscle-specific overexpression of the γ3-subunit via the R225Q polymorphism constitutively activates AMPK (5) and consequently may have a feedback inhibition on skeletal muscle growth and hypertrophy. Thus, we tested the hypothesis that constitutive activation of AMPK will attenuate and AMPKγ3 subunit knockout will augment functional overload-induced hypertrophy in skeletal muscle and the involvement of mTOR signaling and related pathways. We also examined whether the composition of mitochondria and the abundance of the Ca²⁺ pump, major ATP producers and consumers, respectively, in resting skeletal muscle, are altered in the experimental models and conditions tested.

MATERIALS AND METHODS

AMPK-γ3^225Q transgenic and AMPKγ3-knockout mice.

The creation and general metabolic characteristics of skeletal muscle-specific transgenic mice that express a mutant form of the AMPKγ3 subunit (Tg-Prkag3^225Q) and AMPKγ3 gene (Prakg3) knockout mice (Prkag3^−/−) have been described previously (5). In all experiments, Tg-Prkag3^225Q and Prkag3^−/− mice were compared with respective wild-type littermates. Mice were maintained in a temperature- and light-controlled environment and had free access to water and standard rodent diet. All procedures were approved by the Stockholm North Animal Ethics Committee and conducted in agreement with the regulations for protection of laboratory animals.

Surgical procedure and muscle collection.

Male Tg-Prkag3^225Q, Prkag3^−/−, and corresponding wild-type littermates (13–14 wk of age) underwent functional overload of the plantaris muscle. Functional overload was performed by surgical bilateral removal of the soleus and gastrocnemius muscles (3, 6). Mice were anesthetized with isoflurane, and the skin on the side of the leg was opened from the ankle to the knee. Using forceps and scissors, the soleus and gastrocnemius muscles were dissected out and the skin of the leg was sutured back together. Sham-operated mice in which the plantaris, soleus, and gastrocnemius muscles were gently separated from each other with a blunt-end forceps were used as a control. Mice were injected with the analgesic buprenorphine (Temgesic, 0.05–0.1 mg/kg; RB Pharmaceuticals, Slough, UK) during and following the surgical procedure. After 14 days, fed mice were anaesthetized with avertin (0.02 ml/, 2.5% solution of 99% 2,2,2-tribromoethanol and tertiary amyl alcohol), and plantaris muscles were collected, weighed, snap-frozen in liquid nitrogen, and stored at −80°C until further processing. Lean and fat mass were determined by magnetic resonance imaging (EchoMRI-100; EchoMRI, Houston, TX) in the morning on the day of the intervention and euthanization.

RNA extraction and mRNA expression measurement.

Total RNA was extracted from plantaris muscle using Trizol reagent (Invitrogen, Carlsbad, CA), and cDNA was synthesized using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Forster City, CA) according to the manufacturer's instructions. Gene expression was measured by quantitative real-time PCR (qRT-PCR) with the StepOnePlus Real-Time PCR System (Applied Biosystems). Experiments were performed in duplicate in 96-well plates in a total volume of 10 μl. Primers sequences were CACATCCCTGAGTGGCATC (forward) and CACATCCCTGAGTGGCATC (reverse) for F-box protein 32 (Mafbx), TGGATGCTCTTCAGTTCGTG (forward) and GCAACACTCATCCACAATGC (reverse) for insulin-like growth factor I (Igf1), CAACGTTGCTAGGAGAACACC (forward) and GCTCTTGCCACTCATGTTCA (reverse) for myoregulin (Mrln), ACGCTACCACGGAAACAATC (forward) and AAAAGCAACATTTGGGCTTG (reverse) for myostatin (Mstn), CACTGTGACGATCACCGAAG (forward) and TTTCCATTATGCCAGGAAGG (reverse) for phospholamban (Pln), CCTTGTACCCTTCACCAATGAC (forward) and TCCTTCACCTGGTGGCTATT (reverse) for TATA box-binding protein (Tbp), and GCAAAGCATCTTCCAAGGAC (forward) and TCCTTCACCTGGTGGCTATT (reverse) for tripartite motif-containing 63 (Murf1). Taqman assays for sarcolipin (Sln) and Tbp were purchased from Applied Biosystems. Relative gene expression was calculated using the comparative CT method and normalized to a selected housekeeping gene (Tbp) for internal control.

Protein extraction and protein abundance measurement.

Plantaris muscles were homogenized in ice-cold homogenization buffer [20 mM Tris, pH 7.8, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl₂, 0.5 mM Na₃VO₄, 1% Triton X-100, 10% glycerol, 10 mM NaF, 0.2 mM phenylmethylsulfonyl fluoride, 1 mM EDTA, 5 mM Na₄P₂O₇, and 1% (vol/vol) Protease Inhibitor Cocktail (Calbiochem, Darmstadt, Germany)] using a mortar and pestle and TissueLyzer II (Qiagen). Muscle lysates were subsequently rotated for 1 h at 4°C and subjected to centrifugation at 12,000 g for 10 min at 4°C. The supernatant was then collected and protein concentration determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL). Protein lysates were subsequently diluted into Laemmli buffer and heated for 20 min at 56°C. Equal amounts of protein were separated on precast Criterion SDS-PAGE gradient gels (Bio-Rad, Hercules, CA) and transferred to PVDF membranes (Immobilion, Merck Millipore, Billerica, MA). Membranes were blocked in 7.5% milk in TBS-T for 1 h and incubated overnight with a primary antibody at 4°C. Membranes were washed and incubated with the appropriate horseradish peroxidase-conjugated secondary antibody (Bio-Rad) in 5% milk for 1 h at room temperature. Proteins were visualized using enhanced chemiluminescence Western blot detection reagents from GE Healthcare (Waukesha, WI). Optical density of the bands was quantified using the Quantity One imaging system (Bio-Rad). Antibodies against phospho-acetyl-CoA carboxylase (ACC) Ser⁷⁹ (no. 3661), ACC (no. 3662), phospho-Akt Ser⁴⁷³ (no. 9271), Akt (no. 9272), phospho-AMPKα Thr¹⁷² (no. 2531), AMPKα (no. 2532), phospho-mTOR Ser²⁴⁴⁸ (no. 5536), mTOR (no. 2983), phospho-p70 S6 kinase Thr³⁸⁹ (no. 9205), p70 S6 kinase (no. 9202), phospho-S6 ribosomal protein Ser^235/236 (no. 2211), S6 ribosomal protein (no. 2217), phospho-4E-BP1 Thr^37/46 (no. 9459), 4E-BP1 (no. 9644), phospho-TSC2 Thr¹⁴⁶² (no. 3617), and phospho-TSC2 Thr¹³⁸⁷ (no. 5584) were from Cell Signaling Technology (Beverly, MA). AMPKα1 antibody was a kind gift from Dr. Graham Hardie, and AMPKα2 (no. 07-363) antibody was from Merck Millipore, whereas antibody cocktail for the mitochondrial respiratory chain (ab110413), antibody against succinate dehydrogenase flavoprotein subunit (ab137040), and antibodies against sarcoplasmic reticulum Ca²⁺-ATPase (SERCA)1 and -2 (ab2819 and ab2861) were from Abcam (Nordic Biosite). Antibodies targeting p62 (no. P0067) and LC3 (no. L8918) were from Sigma-Aldrich (St. Louis, MO). Equal loading of protein on the gels was ensured using Ponceau staining.

AMPK activity.

AMPK activity was determined as described (4). AMPK kinase was immunoprecipitated from plantaris muscle lysate (300 μg) using a mixture (2.5 μg/sample each) of sheep antibodies against AMPKα1 and -α2 subunits (Prof. D. G. Hardie) and incubated for 30 min (30°C) in a total volume of 30 μl containing 60 mM of HEPES, pH = 7.8, 140 mM NaCl, 833 μM DTT, 200 μM AMP, 200 μM SAMS peptide (Abcam, Nordic Biosite), 5 mM MgCl₂, 200 μM ATP, and 2 μCi [γ-³²P]ATP. The reaction was terminated by spotting 25 μl of the mixture onto P81 filter paper and washing four times for 15 min in 1% phosphoric acid. The dried filter paper was analyzed by liquid scintillation.

Glycogen content.

Glycogen content in plantaris muscle was determined using a colorimetric commercially available kit (no. ab65620, Abcam) according to the manufacturer's instructions. Skeletal muscle (10 mg) was homogenized in deionized water and subsequently boiled for 5 min to inactivate enzymes. Thereafter, lysates were subjected to centrifugation at 13,000 rpm for 5 min. An aliquot of the supernatant (2.5 to 5 μl) was transferred to the assay plate, and glucoamylase was added to hydrolyze glycogen into glucose. Samples were incubated for 30 min in the presence of OxiRed, and optical density was determined at a wavelength of 570 nm.

Statistical analyses.

Differences between means were analyzed using two-way ANOVA. Pairwise post hoc comparisons were determined using Student's t-test and Bonferroni correction to control for type 1 errors. Significance was accepted at P < 0.05. Statistical analysis was performed using GraphPad. Data are presented as means ± SE.

RESULTS

Skeletal muscle hypertrophy in Tg-Prkag3^225Q and Prkag3^−/− mice.

Skeletal muscle hypertrophy was induced in the plantaris by removal of the synergist soleus and gastrocnemius (3, 6, 40). Compared with control sham-operated mice, Tg-Prkag3^225Q and Prkag3^−/− mice and wild-type littermates displayed a robust increase in absolute and relative plantaris weight and thus underwent hypertrophy following 14-day functional overload (Fig. 1). Plantaris muscles of control Tg-Prkag3^225Q mice were slightly heavier compared with wild-type control mice (P < 0.001; Fig. 1, A, C, and E). This difference was not preserved following functional overload. Therefore, wet weight gain in the plantaris muscle of Tg-Prkag3^225Q mice was blunted compared with wild-type counterparts (P < 0.01; Fig. 1G). Tg-Prkag3^225Q and wild-type littermates had slightly decreased body weight following overload compared with control sham-operated mice, which can be attributed to decreased lean mass, whereas fat mass was unchanged (Table 1). Absolute and relative plantaris weight muscle mass between Prkag3^−/− and wild-type mice was unaltered (Fig. 1, B, D, F, and H). Body weight and lean mass tended to be decreased following the overload intervention in Prkag3^−/− mice compared with their respective controls (not significant; Table 2). Control sham-operated Prkag3^−/− mice show increased fat mass compared with wild-type littermates (Table 2).

Fig. 1.

Functional overload of plantaris muscle in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. A and B: plantaris wet muscle mass. C and D: plantaris wet muscle mass normalized to body weight. E and F: plantaris wet muscle mass normalized to lean mass. G and H: plantaris muscle mass gain following 14-day functional overload. Results are means ± SE; n = 9–12/genotype/intervention. Individual values for plantaris muscle mass gain: ●, wild-type littermates; ■, genetically modified mouse model (Tg-Prkag3^225Q in G and Prkag3^−/− in H) *P < 0.05 for Tg-Prkag3^225Q vs. wild-type mice; #P < 0.05 for control (CTL) vs. overload (OVL).

Table 1.

Body composition characteristics of Tg-Prkag3^225Q mice and wild-type littermates before and after the surgical procedure

	Wild Type		Tg-Prkag3225Q
	Control	Overload	Control	Overload
Age, wk	15.1 ± 0.2	15.6 ± 0.2	14.9 ± 0.2	15.7 ± 0.1
Body weight, g
Presurgery	27.0 ± 0.8	28.5 ± 0.6	28.7 ± 0.4	29.3 ± 0.5
Postsurgery	27.5 ± 0.6	27.5 ± 0.6*	28.9 ± 0.4	28.5 ± 0.5*
Lean mass, g
Presurgery	23.5 ± 0.7	25.4 ± 0.7	24.9 ± 0.5	26.1 ± 0.5
Postsurgery	24.1 ± 0.7	24.7 ± 0.6*	25.3 ± 0.4	25.6 ± 0.5*
Fat mass, g
Presurgery	2.4 ± 0.2	2.0 ± 0.2	2.3 ± 0.2	2.1 ± 0.2
Postsurgery	2.2 ± 0.2	1.9 ± 0.1	2.4 ± 0.2	2.2 ± 0.3

Results are means ± SE; n = 10–11/genotype/intervention.

Tg, transgenic.

^*P ≤ 0.05 for intervention effect.

Table 2.

Body composition characteristics of Prkag3^−/− mice and wild-type littermates before and after the surgical procedure

	Wild Type		Prkag3−/−
	Control	Overload	Control	Overload
Age, wk	15.7 ± 0.3	16.2 ± 0.4	15.1 ± 0.1	15.7 ± 0.4
Body weight, g
Presurgery	28.5 ± 0.9	30.3 ± 0.7	31.4 ± 1.4	30.2 ± 0.7
Postsurgery	28.5 ± 0.9	29.8 ± 0.6	32.8 ± 1.8*	29.4 ± 0.9*
Lean mass, g
Presurgery	25.0 ± 0.7	25.7 ± 0.5	24.7 ± 0.6	24.9 ± 0.5
Postsurgery	24.3 ± 0.7	25.5 ± 0.4	25.3 ± 0.6	24.4 ± 0.5*
Fat mass, g
Presurgery	2.6 ± 0.3	3.6 ± 0.5	5.0 ± 1.6	4.2 ± 0.9
Postsurgery	2.8 ± 0.3	3.3 ± 0.3	6.0 ± 1.9*	4.0 ± 1.0

Results are means ± SE; n= 9–12/genotype/intervention.

^*P ≤ 0.05 for intervention effect.

Transcriptional response to skeletal muscle hypertrophy.

To test whether the expression of trophic markers followed the same response as the wet muscle mass, mRNA expression of the anabolic gene Igf1 and proatrophic genes Mstn, Murf1, and Mafbx was determined (Fig. 2, A and B). Anabolic markers were increased and proatrophic markers decreased following functional overload. mRNA expression of Igf1 was decreased in plantaris muscle of Tg-Prkag3^225Q control mice compared with wild-type littermates (P < 0.01). In all groups, there was no genotype difference in Mstn, Murf1, and Mafbx mRNA expression at baseline or following overload.

Fig. 2.

Transcriptional response to 14-day OVL in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. A: mRNA expression in plantaris muscle of Tg-Prkag3^225Q mice. B: mRNA expression in plantaris muscle of Prkag3^−/− mice. Results are normalized to TATA box-binding protein (Tbp) mRNA and presented as means ± SE; n = 6–9/genotype/treatment. *P < 0.05 for Tg-Prkag3^225Q vs. wild-type; #P < 0.05 for CTL vs. OVL. AU, arbitrary units; Igf1, insulin-like growth factor I; Mstn, myostatin; Murf1, tripartite motif-containing 63; Mafbx, F-box protein 32.

mTOR signaling is similar between Tg-Prkag3^225Q and Prkag3^−/−.

The IGF-I/phosphatidylinositol 3-kinase (PI3K)/Akt pathway conveys signal transduction to increased protein synthesis and muscle hypertrophy (13). Phosphorylation and total abundance of phospho-Akt Ser⁴⁷³ and Akt (Figs. 3, A and B, and 4, A and B), phospho-mTOR Ser²⁴⁴⁸ and mTOR (Figs. 3, C and D, and 4, C and D), phospho-P70S6K Thr³⁸⁹ and P70S6K (Figs. 3, E and F, and 4, E and F), phospho-S6 ribosomal protein Ser^235/236 and S6 ribosomal protein (Figs. 3, G and H, and 4, G and H), and phospho-4E-BP1 (eukaryotic initiation factor 4E-binding protein 1) Thr^37/46 and 4E-BP1 (Figs. 3, I and J, and 4, I and J) were increased with functional overload in plantaris muscle but remained unchanged between transgenic or knockout mice and their respective wild-type littermates.

Fig. 3.

Mammalian target of rapamycin (mTOR) signaling pathway in response to 14-day OVL in Tg-Prkag3^225Q mice vs. wild-type littermates. Phospho (p)-Akt Ser⁴⁷³ (A), Akt (B), p-mTOR Ser²⁴⁴⁸ (C), mTOR (D), p-P70S6k Thr³⁸⁹ (E), P70S6k (F), p-S6 ribosomal protein Ser^235/6 (G), S6 ribosomal protein (H), p-4E-BP1 (eukaryotic initiation factor 4E-binding protein) Thr^37/46 (I), and 4E-BP1 (J) in plantaris muscle. Results are means ± SE; n = 6–9/genotype/treatment. #P < 0.05 for CTL vs. OVL.

Fig. 4.

mTOR signaling pathway in response to 14-day OVL in Prkag3^−/− mice vs. wild-type littermates. p-Akt Ser⁴⁷³ (A), Akt (B), p-mTOR Ser²⁴⁴⁸ (C), mTOR (D), p-P70S6k Thr³⁸⁹ (E), P70S6k (F), p-S6 ribosomal protein Ser^235/236 (G), S6 ribosomal protein (H), p-4E-BP1 Thr^37/46 (I), and 4E-BP1 in plantaris muscle (J). Results are means ± SE; n = 6–9/genotype/treatment. #P < 0.05 for CTL vs. OVL.

AMPK signaling in plantaris muscle of Tg-Prkag3^225Q and Prkag3^−/− mice.

The gain-of-function R225Q mutation in Tg-Prkag3^225Q mice results in increased basal AMPK activity in extensor digitorum longus (EDL) muscle (5). Here, we report that phosphorylation on the Thr¹⁷² site of the α-catalytic subunit of AMPK was robustly decreased following overload in Tg-Prkag3^225Q and wild-type animals, irrespective of genotype (Fig. 5A). Total AMPKα, AMPKα1, and AMPKα2 protein abundance was increased with overload in both Tg-Prkag3^225Q and wild-type mice (Fig. 5, B–D). Phosphorylation of ACC at Ser⁷⁹ and total ACC abundance were increased with overload (Fig. 5, E and F). Phosphorylation of TSC2 at sites Thr¹⁴⁶² and Ser¹³⁸⁷ was decreased and increased, respectively, with overload, irrespective of genotype (Fig. 5, G and H).

Fig. 5.

AMP-activated protein kinase (AMPK) signaling and signaling intermediates in response to 14-day OVL in Tg-Prkag3^225Q mice vs. wild-type littermates. p-AMPK Thr¹⁷² (A), AMPKα (B), AMPKα1 (C), AMPKα2 (D), p-ACC (acetyl-CoA carboxylase) Ser⁷⁹ (E), ACC (F), p-TSC2 (tuberous sclerosis complex 2) Thr¹⁴⁶² (G), and p-TSC2 Ser¹³⁸⁷ (H) in plantaris muscle. Results are means ± SE; n = 6–9/genotype/treatment. #P < 0.05 for CTL vs. OVL.

Following overload, phosphorylation of AMPKα on Thr¹⁷² tended to increase in plantaris muscle from wild-type mice (P = 0.09) but was unchanged in Prkag3^−/− mice (Fig. 6A). Abundance of AMPKα was unaltered (Fig. 6, B and D), whereas AMPKα1 was increased in both genotypes following overload (Fig. 6C). Following overload, phosphorylation of ACC at Ser⁷⁹ was unchanged in Prkag3^−/− and wild-type mice, whereas total ACC was increased in Prkag3^−/− but not wild-type mice (Fig. 6, E and F). Finally, whereas phosphorylation of TSC2 at Thr¹⁴⁶² was unchanged by the absence of the AMPKγ3 isoform or overload, phosphorylation at Ser¹³⁸⁷ was increased with overload (Fig. 6, G and H).

Fig. 6.

AMPK signaling and signaling intermediates in response to 14-day OVL in Prkag3^−/− mice vs. wild-type littermates. p-AMPK Thr¹⁷² (A), AMPKα (B), AMPKα1 (C), AMPKα2 (D), p-ACC Ser⁷⁹ (E), ACC (F), p-TSC2 Thr¹⁴⁶² (G), and p-TSC2 Ser¹³⁸⁷ (H) in plantaris muscle. Results are mean ± SE; n = 6–9/genotype/treatment. *P < 0.05 for Prkag3^−/− vs. wild-type mice; #P < 0.05 for CTL vs. OVL.

Total AMPK activity was increased in plantaris muscle from Tg-Prkag3^225Q mice vs. wild-type littermates under both control and overload conditions (Fig. 7). AMPK activity was increased in wild-type and Tg-Prkag3^225Q mice following overload. In Prkag3^−/− mice, AMPK activity was decreased under control conditions compared with their respective wild-type littermates (Fig. 7). Importantly, basal AMPK activity was higher in skeletal muscle from the wild-type littermates for the Prkag3^−/− mice compared with the wild-type littermates for the Tg-Prkag3^225Q mice (Fig. 7). Following overload, AMPK activity was unchanged in skeletal muscle from the Prkag3^−/− mice compared with their respective wild-type littermates.

Fig. 7.

AMPK activity in plantaris muscle in response to 14-day OVL in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. Results are means ± SE; n = 6 genotype/treatment. *P < 0.05 for Tg-Prkag3^225Q vs. wild-type mice; #P < 0.05 for CTL vs. OVL; ‡P < 0.05 for Prkag3^−/− vs. wild-type for Tg-Prkag3^225Q; †P < 0.05 for Prkag3^−/− vs. wild-type mice.

Protein abundance of autophagy markers is increased following overload.

Autophagy is a process regulated by mTOR and AMPK (20). In plantaris muscle from Tg-Prkag3^225Q and Prkag3^−/− mice, p62 and LC3-II protein abundance was upregulated following overload (Fig. 8, A, B, D, and E), whereas the LC3I/LC3II ratio was unchanged (Fig. 8, C and F).

Fig. 8.

Autophagy signaling intermediates in response to 14-day OVL in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. p62 (A), LC3-II (B), and LC3-I/LC3-II ratio (C) in plantaris muscle of Tg-Prkag3^225Q and wild-type mice and p62 (D), LC3-II (E), and LC3 I/LC3 II ratio (F) in plantaris muscle of Prkag3^−/− and wild-type mice. Results are means ± SE; n = 8–9/genotype/treatment. #P < 0.05 for CTL vs. OVL.

Plantaris muscle glycogen content.

Fed and fasted Tg-Prkag3^225Q mice but not Prkag3^−/− mice have increased glycogen content in glycolytic skeletal muscle (5). Here, we confirm that glycogen content in plantaris muscle was increased 40% in control Tg-Prkag3^225Q mice compared with wild-type littermates (P < 0.05; Fig. 9). Glycogen content was increased after functional overload (P < 0.01; Fig. 9), but this difference was not observed in Tg-Prkag3^225Q mice.

Fig. 9.

Glycogen content in response to 14-day OVL in Tg-Prkag3^225Q mice vs. wild-type littermates. Results are means ± SE; n = 7–9/genotype/treatment. *P < 0.05 for Tg-Prkag3^225Q vs. wild-type mice; #P < 0.05 fo CTL vs. OVL.

Protein abundance of mitochondrial respiratory chain complexes in plantaris muscle.

Abundance of both NADH and succinate dehydrogenases (complex I and II) was reduced following functional overload Prkag^225Q and Prkag3^−/− mice (Fig. 10, A–E). Cytochrome b-c1 complex subunit 2 (complex III) and ATP synthase subunit-α (complex V) abundance was also reduced after functional overload in Prkag3^225Q and wild-type mice (Fig. 10, F and H) but unchanged in between Prkag3^−/− and wild-type mice (Fig. 10, G and I).

Fig. 10.

Components of the mitochondrial respiratory chain in response to 14-day OVL in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. A: representative Western blot of complexes I, II, III, and V of the mitochondrial respiratory chain in plantaris muscle from Tg-Prkag3^225Q and Prkag3^−/− mice. B–G: quantifications of NDUFB8 (complex I; B and C), succinate dehydrogenase complex subunit A (SDHA) (complex II; D and E), UQCRC2 (complex III; F and G), and ATP5A (complex V; H and I). Results are means ± SE; n = 6–9/genotype/treatment. *P < 0.05 for Tg-Prkag3^−/− vs. wild-type mice; #P < 0.05 for CTL vs. OVL.

Skeletal muscle abundance of calcium pump isoforms and regulatory proteins.

SERCA1 coassociates with Pln, Sln, and Mrln in oxidative and glycolytic skeletal muscle to regulate the reuptake of calcium into the sarcoplasmic reticulum (1, 21, 39). We found decreased Mrln mRNA expression and increased Sln mRNA expression after functional overload in both genetically modified mouse models and their respective wild-type littermates (Fig. 11, A and B). Pln mRNA was decreased after functional overload in wild-type mice but remained unchanged in the Prkag3^−/− mice. In contrast, mRNA of Pln was unaltered between Prkag3^225Q and wild-type mice. Following functional overload, SERCA1 protein abundance was downregulated in both Prkag3^225Q and wild-type mice (Fig. 11C) but unaltered in Prkag3^−/− and wild-type mice (Fig. 11E). Conversely, SERCA2 protein abundance was increased in both Prkag3^225Q and wild-type mice (Fig. 11D) but unchanged in Prkag3^−/− and wild-type mice (Fig. 11F).

Fig. 11.

Changes in gene expression and protein abundance of components of the calcium pump in response to 14-day OVL in Tg-Prkag3^225Q and Prkag3^−/− mice vs. wild-type littermates. A: mRNA expression in plantaris muscle of Tg-Prkag3^225Q and respective wild-type mice. B: mRNA expression in plantaris muscle of Prkag3^−/− and respective wild-type mice. Results are normalized to Tbp mRNA. C and D: protein abundance in plantaris muscle of Tg-Prkag3^225Q mice for sarcoplasmic reticulum Ca²⁺-ATPase (SERCA)1 (C) and SERCA2 (D). E and F: protein abundance in plantaris muscle of Tg-Prkag3^−/− mice for SERCA1 (E) and SERCA2 (F). Results are mean ± SE; n = 6–9/genotype/treatment. #P < 0.05 for CTL vs. OVL.

DISCUSSION

To investigate the role of the AMPKγ3 isoform in the control of skeletal muscle size, we subjected Tg-Prkag3^225Q and Prkag3^−/− mice to 2 wk of functional overload. Muscle weight, transcriptional, and signaling data provide evidence against the involvement of the AMPKγ3 isoform in the regulation of skeletal muscle hypertrophy. Plantaris muscle from both Tg-Prkag3^225Q and Prkag3^−/− mice underwent hypertrophy, with total wet muscle mass similar to their respective wild-type littermates following the 14-day functional overload. However, in control sham-operated Tg-Prkag3^225Q mice, plantaris muscle mass was slightly increased compared with wild-type mice, indicating that the subsequent hypertrophic response to overload was blunted.

Overload-induced changes in mRNA expression of myostatin as well as E3 ubiquitin ligases occurred independent of genotype. These results are consistent with our finding that the AMPKγ3 isoform is dispensable for the hypertrophic response in skeletal muscle. In Tg-Prkag3^225Q mice, the decrease in Igf1 expression could contribute to the reduced plantaris mass gain compared with wild-type mice. AMPKα1^−/− mice have decreased plantaris weight and, in response to functional overload, accelerated hypertrophy compared with wild-type littermates, concomitant with heightened activation of the mTOR pathway (29). In this study, the increase in total protein abundance and phosphorylation of mTOR signaling intermediates in both genotypes indicate that hypertrophy signaling pathways leading to protein synthesis and cell growth are similarly activated following functional overload, as indicated by similar abundance and phosphorylation of downstream targets of p70S6 kinase S6 ribosomal protein and elongation factor 4E-BP1. This finding provides further evidence that the AMPKγ3 isoform is dispensable for skeletal muscle hypertrophy. Our findings argue in favor of a ceiling effect on muscle weight gain following functional overload-induced hypertrophy.

Tg-Prkag3^225Q mice have constitutively activated basal AMPK activity but are resistant to further increases in AMPK activity induced by rising AMP levels (5). AMPK phosphorylation between Tg-Prkag3^225Q, Prkag3^−/−, and wild-type littermates is similar under basal, AICAR stimulation, or muscle contraction conditions (5). Phosphorylation on the Thr¹⁷² site of the AMPKα catalytic subunit is robustly decreased following overload in Tg-Prkag3^225Q and wild-type mice but remains unchanged in the Prkag3^−/− mice. This finding diverges from previous studies using other genetic mouse models in which increased AMPK phosphorylation following overload is attributed to increased AMPKα1 activity (26, 36, 38). However, we found AMPK phosphorylation is decreased (in Tg-Prkag3^225Q mice) or unchanged (in Prkag3^−/− mice) following overload despite an increase in AMPKα1 abundance in all genotypes. AMPKα2 activity and protein abundance following functional overload vary according to the investigated time point and are thought to have little influence on muscle growth (29), although the response to functional overload in AMPKα2-knockout mice has not been investigated. The differences in AMPK phosphorylation, AMPK activity, and rate of muscle growth could be related to the time point investigated (7-, 14-, or 21-day overload). Moreover, growth rate and body composition are unaffected in mice expressing a kinase-dead form of AMPK in skeletal muscle (30, 31). Thus, the genetic alteration of different AMPK isoforms could affect the rate at which muscle growth occurs during the hypertrophy but is likely to be without an impact on final muscle weight.

AMPK activity is increased in skeletal muscle from Tg-Prkag3^225Q mice under basal and overload-induced conditions. Interestingly, basal AMPK activity was higher in the wild-type littermates for the Prkag3^−/− mice compared with the wild-type littermates for the Tg-Prkag3^225Q mice. These findings indicate differences in AMPK activity between strains and highlight the importance of using appropriate wild-type littermates as a control group for metabolic studies. The baseline AMPK activity differences between these two mouse lines are consistent with previously published gene array data reporting the expression of the AMPKγ1, -γ2, and -β1 subunits, and the expression of the upstream AMPK kinase LKB1 is increased in skeletal muscle from the wild-type littermates for the Prkag3^−/− mice vs. the wild-type littermates of the Tg-Prkag3 mice (34).

AMPK regulates mTOR activity by signaling through TSC2. In energy-consuming conditions, when the AMP/ATP ratio increases, AMPK phosphorylates TSC2 at Ser¹³⁸⁷, activating TSC2 and downregulating mTOR activity (17, 19). Conversely, Akt upstream of mTOR and AMPK inhibits TSC2 by phosphorylating its Thr¹⁴⁶² residue. We found increased phosphorylation of TSC2 on Ser¹³⁸⁷ in all genotypes following functional overload, supporting a role for AMPK activation, despite reduced or unchanged AMPK phosphorylation. In Prkag3^225Q mice and their respective wild-type littermates, phosphorylation of TSC2 on Thr¹⁴⁶² is reduced following overload but unchanged in Prkag3^−/− mice and their respective wild-type littermates. Despite differences in AMPK signaling between Tg-Prkag3^225Q and Prkag3^−/− mice, mTOR signaling and muscle hypertrophy were unchanged between the genetically modified mouse models and their wild-type littermates.

Another mechanism by which AMPK may control muscle mass is by modifying autophagy, a self-digestive, tightly-regulated cellular process where cytoplasmic components are degraded to generate nutrients and energy. Nutrient deprivation, acute exercise, and thus AMPK activation promote autophagy (20, 37). Moreover, autophagy is required for exercise training-induced adaptations in skeletal muscle (37). Conversely, mTOR activation represses autophagy (20), and whether functional overload intervention impacts autophagy in skeletal muscle is unknown. Although LC3-II protein abundance or LC3-I/LC3-II ratio provides insight into autophagosome formation (28), increased levels of the LC3-binding protein p62 are indicative of a disrupted autophagy process (22). In Tg-Prkag3^225Q, Prkag3^−/−, and wild-type mice, functional overload increased protein abundance of p62 and LC3-II without altering the LC3-I/LC3-II ratio. Although these data are indicative of increased autophagosome formation and perturbed autophagy, further work is required to assess the impact of functional overload on autophagic flux in muscle.

The R225Q single-nucleotide polymorphism in the PRKAG3 gene was first discovered in Hampshire pigs (27) and later in humans (10). This mutation increases glycogen content in glycolytic skeletal muscle of both pigs and Tg-Prkag3^225Q mice (4, 5, 27). Consistent with these earlier results, glycogen content in plantaris muscle of control sham-operated Tg-Prkag3^225Q mice was increased by 40% compared with wild-type mice. Overload intervention increased glycogen content in plantaris muscle in wild-type mice, but this did not reach statistical significance in Tg-Prkag3^225Q mice. Glycogen synthase kinase-3β regulates activity of the TSC1/TSC2 complex, which could be of relevance to regulate cell growth in these genetically modified mouse models (18). Whereas the effects of endurance training on glycogen muscle content have been studied extensively (16, 33), the effects of resistance exercise are less well characterized. Glycogen can account for 0.7% of skeletal muscle weight given that 3–4 g of water is bound to each gram of glycogen (35). However, the increase in water content is insufficient to explain the 10% difference in plantaris wet mass between control sham-operated Prkag3^225Q and wild-type mice.

Functional overload intervention robustly increased sarcolipin mRNA, whereas it concomitantly decreased myoregulin mRNA. These changes occurred independent any change in SERCA protein abundance. Genetic deletion of myoregulin improves Ca²⁺ handling in skeletal muscle and enhances exercise performance (1). Conversely, sarcolipin promotes skeletal muscle thermogenesis (2) and enhances energy expenditure (25). The reciprocal changes in myoregulin and sarcolipin expression in skeletal muscle are likely to be an adaptation response to the increased workload and ability to handle Ca²⁺ fluxes in response to functional overload.

Our main finding is that the AMPKγ3 isoform is dispensable for skeletal muscle hypertrophy. Nevertheless, we cannot exclude the involvement of other pathways that have been implicated in skeletal muscle remodeling, namely the phospholipase D/phosphatidic acid pathway (41) and focal adhesion kinase (FAK) signaling (11). FAK phosphorylation and activation in response to integrin engagement or growth factors, including IGF-I, may subsequently inhibit TSC2 at Thr¹⁴⁶² (11) and ultimately influence muscle growth. The role of these signaling pathways in mediating the hypertrophic response requires further elucidation.

In conclusion, constitutive (R225Q) AMPK activation attenuates the gain of muscle mass independent of mTOR activation following functional overload. More importantly, AMPKγ3 subunit ablation does not enhance mTOR activation or skeletal muscle hypertrophy. This unexpected result suggests that AMPK heterotrimeric complexes containing the γ1- or γ2-subunit may play a role in skeletal muscle growth control. Our study supports the notion that the AMPKγ3 subunit plays a metabolic rather than a growth-enhancing role. The role of AMPK in skeletal muscle hypertrophy is far more complex than previously highlighted and suggests that mechanical loading can override signaling pathways that act as negative effectors of mTOR signaling and consequently lead to muscle growth.

GRANTS

This study was financed by grants from Swedish Research Council, the European Research Council Advanced Grant Ideas Program, the Novo Nordisk Foundation, the Swedish Research Council, the Stockholm County Council, the Swedish Foundation for Strategic Research, the Swedish Diabetes Foundation, and the Strategic Diabetes Program at Karolinska Institutet.

DISCLOSURES

The authors have no conflicts of interest, financial or otherwise, to declare.

AUTHOR CONTRIBUTIONS

I.R., M.E.O., and A.V.C. performed experiments; I.R., M.E.O., M.B., B.E., G.A.N., A.V.C., and J.R.Z. analyzed data; I.R., M.B., B.E., G.A.N., A.V.C., and J.R.Z. interpreted results of experiments; I.R. and A.V.C. prepared figures; I.R. and J.R.Z. drafted manuscript; I.R., M.E.O., M.B., B.E., G.A.N., A.V.C., and J.R.Z. edited and revised manuscript; I.R., M.E.O., M.B., B.E., G.A.N., A.V.C., and J.R.Z. approved final version of manuscript; M.E.O., B.E., G.A.N., A.V.C., and J.R.Z. conception and design of research.