Calorie restriction and exercise differentially regulate AMP-activated protein kinase across subcellular compartments in skeletal muscle from older male rats Authors: Haiyan Wang, Jiwei Hao, Jun-Hyun Bae, Gregory D Cartee Published: 2026 ABSTRACT AbstractAMP-activated protein kinase (AMPK) is a crucial energy sensor that regulates a wide range of important processes in skeletal muscle. AMPK is present in several subcellular compartments (including the cytosol, nucleus, and mitochondria). However, the influence of physiologically relevant interventions on AMPK’s localization in skeletal muscle is not well understood, especially during older age. Accordingly, this study examined AMPK signaling in skeletal muscle from aged male rats (24–25-months-old) subjected to either eight-weeks of calorie restriction (CR; consuming 65% of ad libitum intake) or a single swim-exercise session. Phosphorylation of AMPK and its substrate acetyl-CoA carboxylase (ACC), as well as abundance of AMPK subunits (α1, α2, β1, β2, γ1, γ3), were assessed by immunoblotting in whole muscle lysates and cytosolic, nuclear, and mitochondrial-enriched fractions obtained by differential centrifugation. CR increased phosphorylation of AMPK and ACC in whole muscle lysates, but not in the three subcellular fractions that were tested, suggesting AMPK-activation occurs in other, currently unidentified compartments. In contrast, exercise significantly increased AMPK phosphorylation in the cytosolic fraction and ACC phosphorylation in whole lysates and all three subcellular fractions. AMPK-γ1 abundance was greater in the mitochondrial-enriched fraction of CR versus ad libitum muscles. These findings revealed strikingly different patterns of AMPK activation within key subcellular compartments in response to two important physiological interventions. This study substantially advances current knowledge and provides a foundation for future research on AMPK’s compartment-specific roles in skeletal muscle physiology and aging.Keywords: Mitochondria, Nucleus, Cytosol, Acetyl-CoA carboxylase, AMP-activated protein kinase Introduction AMP-activated protein kinase (AMPK) is a heterotrimeric serine/threonine kinase consisting of a catalytic subunit (α1 or α2 isoform) and two regulatory subunits (β1 or β2, and γ1, γ2, or γ3 isoforms). AMPK serves as a cellular energy sensor and plays key roles in regulating signaling proteins, enzymes, cell growth, and transcription (Trefts and Shaw, 2021). Studies with cultured cells have shown that AMPK is associated with multiple subcellular compartments, including the cytosol, nucleus, and mitochondria (Salt et al., 1998; Suzuki et al., 2007; Okamoto et al., 2019; Kim et al., 2014; Kodiha et al., 2007; Zong et al., 2019; Schmitt et al., 2022). Additionally, chemical activators of AMPK (such as AICAR, phenformin, resveratrol, and leptin) can promote its movement from the cytosol to the nucleus in isolated cells (Suzuki et al., 2007; Kodiha et al., 2011). These findings indicate that AMPK’s diverse functions likely depend, at least in part, on its distribution across various subcellular compartments (Miyamoto et al., 2015). While there is clear evidence that energetic challenges can regulate AMPK compartmentalization in cultured cells, our understanding of how physiologically relevant interventions affect AMPK localization in skeletal muscle, either in humans or animals, remains limited. One such intervention is chronic calorie restriction (CR; consuming ~20–40% less than ad libitum, AL, intake). CR produces significant and wide-ranging effects on skeletal muscle metabolism and function (Yang et al., 2016; Aspnes et al., 1997; Rhoads et al., 2020; Bevilacqua et al., 2004; Drew et al., 2003; Wetter et al., 1999; Cartee et al., 1994). It is likely that some of these effects are related to the activation of AMPK in skeletal muscle by CR (Wang et al., 2012; Zheng et al., 2023; Al-Regaiey et al., 2007). However, prior research has not addressed whether CR alters the subcellular localization of AMPK. Accordingly, the current study was the first examination of the effect of CR on the location of AMPK protein and AMPK phosphorylation in the cytosol, mitochondria, and nucleus of skeletal muscle. Activation of AMPK, measured in whole skeletal muscle lysates, is also a well-known consequence of exercise (Kjobsted et al., 2018). The impact of exercise on AMPK’s subcellular localization was first described in a seminal study showing that a single exercise session increased the association of the catalytic α2 subunit of AMPK with the nuclear fraction of skeletal muscle from young men (McGee et al., 2003). To date, the effects of exercise on the nuclear localization of the β or γ subunits of AMPK have not been reported. Furthermore, little is known about how exercise influences the phosphorylation of nuclear-localized AMPK in skeletal muscle. Therefore, the current study was the first assessment of the effects of exercise on the phosphorylation of AMPK in the nucleus of skeletal muscle. Recent research has found that AMPK is also associated with skeletal muscle mitochondria in young mice, and that exercise can increase the activation of mitochondrial-associated AMPK in muscle (Drake et al., 2021). Calorie restriction (CR) and exercise are two interventions that can each help counteract a number of age-related deficits in skeletal muscle. However, no studies have yet explored how either CR or exercise affect the spatial distribution of AMPK within skeletal muscle at an advanced age. To address this gap, the current study examined skeletal muscle AMPK in older, male rats (24–25-months) following either CR (consuming 65% of ad libitum intake for 8 weeks) or acute exercise (90 min of swimming). Although chronic CR and acute exercise can each activate AMPK in skeletal muscle, they are distinct physiological stimuli with many differences, including duration, energetic demand, hormonal environment, and mechanical input. Therefore, we did not expect these interventions to produce identical effects on AMPK abundance or activation across the subcellular fractions examined. Rather, evaluating these two distinct physiological challenges allowed a broader assessment of the compartment-specific AMPK responses in skeletal muscle from older rats. Key proteins were assessed by immunoblotting in whole skeletal muscle lysates, as well as in cytosolic-, nuclear-, and mitochondrial-enriched fractions. These compartments were selected for evaluation because: 1) AMPK is associated with each compartment in cell models; 2) crucial regulatory and metabolic events are localized in each compartment; and 3) the limited available evidence, exclusively in muscles from younger humans or rodents, has linked exercise effects on AMPK in each compartment. Measurements included the phosphorylation of AMPK-α1/2 subunits at Thr-172, and the Ser-79 phosphorylation of acetyl-CoA carboxylase (ACC), an AMPK substrate that is frequently used as a marker for AMPK activation (Wang et al., 2019; Hardie, 1992; Scott et al., 2002). The CR and exercise effects on subcellular localization of multiple AMPK subunits and on the abundance and phosphorylation of ACC were also evaluated for the first time. Materials Chemicals were obtained from Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Hanover Park, IL) unless otherwise noted. Reagents and apparatus for SDS-PAGE and immunoblotting were obtained from Bio-Rad Laboratories (Hercules, CA). Anti-phospho-ACC Ser79 that detects both pACC1 Ser79 and pACC2 Ser212 (p-ACC Ser79; #3661), anti-ACC that detects isoforms ACC1 and ACC2 (#3676), anti-phospho-AMPKα Thr172 that detected both α1 and α2 isoforms (p-AMPK Thr172; #50081), anti-AMPK-α that recognizes both α1 and α2 isoforms (#5831), anti-AMPK-α1 (#2795), anti-AMPK-α2 (#2757), anti-AMPK-β1 (#4178), anti-AMPK-β2 (#4148), anti-VDAC (voltage-dependent anion channel; #4661), anti-α-Tubulin (#2125), anti-Histone H3 (#4499), anti-Rab5 (#3547), anti-LAMP2A (lysosome-associated membrane protein 2A; #81197), anti-mouse (#7076), and anti-rabbit IgG horseradish peroxidase conjugate (#7074) were from Cell Signaling Technology (Danvers, MA). Anti-AMP-activated protein kinase-γ3 (AMPK-γ3) was provided by Dr. David Thomson (Brigham Young University) (Hardman et al., 2014). Anti-AMP-activated protein kinase-γ1 (AMPK-γ1, #ab32508) was from Abcam (Cambridge, MA). Anti-AMPK-α1 (#NBP2–22127) was purchased from Novus Biologicals (Centennial, CO). Anti-Calnexin (#10427–2-AP) was from Proteintech (Rosemont, IL). EDTA free Protease inhibitor cocktail tablet (#S8830) was from MilliporeSigma (Burlington, MA). Animal treatment Animal studies were conducted in accordance with the guidelines from the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and with approval of University of Michigan Committee on Use and Care of Animals (protocol numbers 11809 and 12196). Calorie restriction protocol Male Fischer-344 × Brown Norway (FBN) rats were obtained at 22-months of age from the National Institute of Aging (NIA). Animals were individually housed in cages and maintained on a 12:12 h light-dark cycle (lights out at 1700 h) in specific pathogen-free conditions. Rats were given free access to food (Laboratory Diet no. 5L0D; Lab Diet, St. Louis, MO, USA) and water for a one-week acclimation period before baseline food intake assessment in the University of Michigan facility. After 1 week of baseline food intake assessment (food provided minus uneaten food), the rats were randomly assigned to AL (n = 10) or CR (n = 10) groups. The CR group was provided with rodent chow once daily at 16:00 h, in an amount equal to 65% of the average daily intake of the AL group for 8 weeks. Each week, the food consumption of the AL rats was measured by subtracting the remaining uneaten chow from the total amount provided. The AL group’s average food intake was then used to calculate the CR group’s daily allotment for the following week. Daily inspections of the CR cages at 16:00 h confirmed their food allotment from the previous day had been consumed. The rationale for providing food to CR rats at 16:00 h (one hour before the dark-cycle) was that AL rats usually consume a significant portion of their daily intake between this time and the initial hours of the dark-cycle. All rats were weighed weekly between 15:00 and 16:00 h. After the 8-week CR period, both AL and CR rats (aged 24 months) were deeply anesthetized by an intraperitoneal injection of ketamine/xylazine cocktail (50 mg/kg ketamine and 5 mg/kg xylazine). Under deep anesthesia, both epitrochlearis muscles from each rat were dissected out and used to prepare whole muscle lysate, cytosolic-, nuclear-, and mitochondrial-enriched fractions (experimental design in Fig. 1A). After removal of the epitrochlearis muscles, the left retroperitoneal fat pad was dissected and subsequently weighed. Euthanasia was performed by bilateral thoracotomy after the removal of the fat pad while the rats were deeply anesthetized. A: Experimental design for CR protocol. Animals (male Fischer-344 Brown Norway rats; initially 22-months-old) were randomly assigned to the ad libitum (AL) or calorie restriction (CR; 65% of AL food intake for 8 weeks) diet groups. Epitrochlearis muscles from each rat were rapidly dissected out to process total lysate, cytosolic-, mitochondrial-, nuclear-enriched fraction. B: Experimental design for acute exercise protocol. Animals (male Fischer-344 Brown Norway rats; ~25-months-old) were used in exercise study. The exercise protocol included 9 × 10-min swimming bouts, each separated by a 10-min rest period. Immediately postexercise, epitrochlearis muscles from exercised and time-matched sedentary rats were rapidly dissected out to obtain total lysate and cytosolic-, mitochondrial-, nuclear-enriched fraction. The figures were created with BioRender.com. Acute exercise protocol A separate cohort of male FBN rats (n = 16) aged 23-months were obtained from the National Institute of Aging (NIA). Animals were housed under a 12:12 h light-dark cycle (lights off at 17:00 h) with ad libitum access to standard rodent chow and water for 8–9 weeks in the University of Michigan facility. On the day of the experiment, at approximately 09:00 h, groups of four exercised rats (aged ~25-months) underwent swimming sessions in a barrel containing water maintained at 35 °C and a depth of ~45 cm. The exercise protocol included nine 10-min swimming bouts, each separated by a 10-min rest period. Earlier research demonstrated that this exercise protocol can be completed by older rats, and that it resulted in a postexercise increase in insulin-stimulated glucose uptake by the epitrochlearis muscle (Oki et al., 2020). Immediately postexercise, rats were anesthetized with an intraperitoneal injection of ketamine/xylazine cocktail (50 mg/kg ketamine and 5 mg/kg xylazine) with time-matched sedentary rats, one epitrochlearis muscle was rapidly dissected and processed for whole muscle lysate, as well as cytosolic-, nuclear-, and mitochondrial-enriched fractions (experimental design in Fig. 1B). We selected the epitrochlearis muscle for this study because previous research has shown the following: 1) its muscle fiber type composition closely reflects that of rat skeletal muscle overall (Castorena et al., 2011; Delp and Duan, 1996); 2) the chronic calorie restriction (CR) protocol used here increases phosphorylated AMPK levels in the epitrochlearis of older rats (Zheng et al., 2023); 3) CR enhances insulin-stimulated glucose uptake in the epitrochlearis of older rats (Wang et al., 2025; Wang et al., 2017; Sequea et al., 2012); and 4) the exercise protocol also elevates insulin-stimulated glucose uptake in the epitrochlearis of older rats (Oki et al., 2020). Whole muscle lysate preparation Each epitrochlearis muscle was rapidly rinsed with cold phosphate buffered saline (PBS), and a small portion of muscle (~20 mg) was cut and freeze-clamped with liquid N2. The larger, unfrozen portion of muscle was used to produce the cytosolic-, nuclear-, and mitochondrial-enriched fractions. The smaller portion was subsequently processed for the whole muscle lysate. Frozen muscles were homogenized in ice-cold lysis buffer using a glass pestle attached to a motorized homogenizer as described (Wang et al., 2022). The lysis buffer contained T-PER Tissue Protein Extraction Reagent (#PI-78510; Thermo Scientific, Rockford, IL) supplemented with 1 mM EDTA, 1 mM EGTA, 2.5 mM sodium pyrophosphate (NaPP), 1 mM sodium vanadate (Na3VO4), 1 mM ß-glycerophosphate, 1 μg/mL leupeptin, and 1 mM PMSF. Homogenates were rotated for 1 h at 4 °C prior to centrifugation (15,000 xg for 15 min at 4 °C). The supernatants were transferred to microfuge tubes and stored at −80 °C for subsequent analyses. Cytosolic-, nuclear-, and mitochondrial-enriched fraction preparation The method for fractions preparation was modified from the protocol in an earlier study (Dimauro et al., 2012). The larger portion of epitrochlearis muscle (~120 mg) was cut into two similarly sized portions (60 mg each) and then cut into smaller pieces. Throughout the procedure, the muscle tissue was kept on ice on a glass Petri dish. Each muscle underwent two rounds of homogenization. For each homogenization, 750 μL of Sucrose-Tris-MgCl2 (STM) homogenization buffer (250 mM sucrose, 50 mM Tris–HCl, pH 7.5, 5 mM MgCl2, 10 mM NaF, 10 mM β-glycerophosphate, 2.5 mM NAPP, 1 mM Na3VO4, EDTA free protease inhibitor cocktail) was added, and muscle was homogenized for 1 min on ice using a Teflon pestle attached to Caframo homogenizer (speed setting of 2). Each homogenate was centrifuged at 800 xg for 15 min. The supernatants from both rounds of homogenization were pooled into a single centrifuge tube. The supernatant (S0) was used to isolate mitochondrial and cytosolic fractions, and the pellet (P0) was saved for nuclear fraction isolation (Fig. 2). Immunoblotting with antibodies against appropriate marker proteins for each subcellular compartment was used to evaluate the cytosolic-enriched (α-Tubulin), mitochondrial-enriched (VDAC), and nuclear-enriched (Histone H3) fractions (Dias et al., 2020). Schematic representation of the protocol for isolating cytosolic, mitochondrial, and nuclear-enriched fractions. Nuclear-enriched fraction The P0 pellet was resuspended in 750 μl STM washing buffer (250 mM sucrose, 50 mM Tris-HCl, pH 7.5, 5 mM MgCl2, 10 mM NaF, 20 mM β-glycerophosphate, 20 mM NAPP, 2 mM Na3VO4 and EDTA free protease inhibitor cocktail), vortexed for 15 s at maximum speed, and then centrifuged at 500 ×g for 15 min. The supernatant was discarded, and the pellet was resuspended again in 750 μL STM washing buffer, vortexed for 15 s, and centrifuged at 1000 g for 10 min. This step was repeated three times. The final pellet (P1) was resuspended in NET buffer (20 mM HEPES pH 7.9, 1.5 mM MgCl2, 250 mM NaCl, 1% Triton X-100, 10 mM NaF, 20 mM β-glycerophosphate, 20 mM NAPP, 2 mM Na3VO4 and EDTA free protease inhibitor cocktail), vortexed for 15 s, and kept on ice for ~1 h. The suspension was centrifuged at 20,000 ×g for 30 min, and the supernatant was collected as the nuclear-enriched fraction (Fig. 2). Mitochondrial-enriched fraction The initial S0 supernatant was centrifuged at 800 ×g for 15 min, and the resulting supernatant was transferred to a new tube. This process was repeated twice, and the supernatant (S1) was saved. The S1 supernatant was subsequently centrifuged at 11,000 ×g for 10 min to obtain the supernatant (S2) and pellet the mitochondrial (P2). The supernatant (S2) was saved for cytosolic fraction processing. The P2 pellet was resuspended in 750 μL STM washing buffer and centrifugation at 11,000 ×g for 10 min. After centrifuge, supernatant was discarded and pellet was saved, with this washing step repeated twice. The mitochondrial pellet (P3) from last centrifuge was resuspended in SOL buffer (50 mM Tris-HCl pH 6.8, 1 mM EDTA, 0.5% Triton X-100, 10 mM NaF, 20 mM β-glycerophosphate, 20 mM NAPP, 2 mM Na3VO4 and EDTA free protease inhibitor cocktail) and labelled as mitochondrial fraction (Fig. 2). Cytosolic-enriched fraction The supernatant from the mitochondrial spin (S2) was centrifuged again at 11,000 xg for 10 min (repeated twice). The resulting supernatant (S3) was mixed with 1 mL pre-chilled acetone and incubated at −20 °C for ~1 h to precipitate cytosolic proteins. The sample was then centrifuged at 12,000 xg for 5 min, and the pellet was resuspended in STM washing buffer to obtain the cytosolic fraction (Fig. 2). Immunoblotting Protein concentration from each fraction was determined using the bicinchoninic acid procedure. Immunoblotting procedures were performed as described (Wang et al., 2022). Equal amounts of protein from each fraction were mixed with 6× Laemmli buffer, boiled for 5 min, and then separated by SDS-PAGE. The protein was then transferred to polyvinylidene difluoride membranes. After completion of transfer, the membranes were stained with MemCode to quantify the protein loading of each sample. Membranes were blocked in TBST (Tris-buffered saline, pH 7.5 plus 0.1% Tween-20) that was supplemented with 5% nonfat milk or BSA for 1 h at room temperature. Following blocking, the membranes were incubated with the appropriate primary and secondary antibodies, subjected to enhanced chemiluminescence, and quantified by densitometry (AlphaView; ProteinSimple, San Jose, CA). The values of specific proteins were normalized for protein loading determined by the MemCode protein stain. Results were expressed relative to the normalized average of all the samples on the blot. Statistics Comparisons between two groups were performed using Student’s t-test. Values are presented as means ± SD. Pearson correlation analysis was used to assess correlations between variables with GraphPad Prism 10.0. P-values ≤0.05 were considered statistically significant. Results CR effects on the body mass, fat pad mass, epitrochlearis mass, and food intake All of the rats in AL (n = 10) and CR groups (n = 10) completed the dietary protocol. Initial body masses were not significantly different between AL and CR diet groups (CR = 575.8 ± 25.0 g; AL = 549.3 ± 41.1 g). At the end of the dietary protocol, the body mass (CR = 463.0 ± 25.5 g versus AL = 539.8 ± 42.9 g, P < 0.001), retroperitoneal fat pad mass (CR = 3.7 ± 0.87 g versus AL = 4.8 ± 1.97 g, P < 0.01), retroperitoneal fat pad/body mass ratio (CR = 7.9 ± 1.5 mg/g versus AL = 9.8 ± 1.8 mg/g, P < 0.05), and epitrochlearis mass (CR = 129.9 ± 11.7 mg versus AL = 144.7 ± 11.4 mg, P < 0.05) were significantly lower for CR versus AL animals. As intended, the food intake of CR rats was ~65% of the value for AL rats (CR = 11.8 ± 0.24 g versus AL = 18.2 ± 0.33 g, P < 0.05). Purity of the cytosolic-enriched, mitochondrial-enriched, and nuclear-enriched fractions using protein markers To assess the purity of each fraction, we used established markers for the cytosolic (α-Tubulin), mitochondrial-enriched (VDAC), and nuclear-enriched (Histone H3) compartments. As expected, the cytosolic-enriched fractions from both the CR and exercise experiments contained high levels of the cytosolic marker α-Tubulin, with undetectable levels of VDAC and Histone H3 (Supplementary Fig. 1). We also analyzed cytosolic-enriched fractions from the CR and AL groups for markers of lysosomes (LAMP2A), the endoplasmic reticulum and sarcoplasmic reticulum (calnexin), and endosomes (Rab5). In whole muscle lysates, LAMP2A levels were higher in CR than in AL samples, whereas calnexin and Rab5 did not differ by diet (Supplementary Fig. 2). No diet-related differences were detected for these proteins in the cytosolic-enriched fraction, and none were highly expressed in this fraction (Supplementary Fig. 3). The mitochondrial-enriched fractions showed high levels of the mitochondrial marker VDAC, while α-Tubulin was barely detectable and Histone H3 was undetectable. Similarly, the nuclear-enriched fractions exhibited high levels of the nuclear marker Histone H3, with undetectable levels of both α-Tubulin and VDAC. These results confirm that cross-contamination between the cytosolic-enriched, mitochondrial-enriched, and nuclear-enriched fractions was minimal. CR effects on AMPK phosphorylation, ACC phosphorylation and abundance of AMPK subunits Phosphorylated proteins values were expressed as a ratio of phosphorylated protein to total protein (determined for each sample using a separate immunoblot with a primary antibody for the relevant total protein). Whole muscle lysate In whole muscle lysate, consistent with earlier study (Zheng et al., 2023), CR rats showed significantly greater phosphorylation of AMPK-α1/2 on Thr172 and ACC on Ser79 (P < 0.05, Fig. 3A and C) compared to AL rats. The abundance of AMPK-β2 was slightly lower in CR rats versus AL rats (P < 0.05, Fig. 3H). No significant diet-related differences were observed for abundance of ACC, AMPK-α1/2, AMPK-α1, −α2, −β1, −γ1, or −γ3 (Fig. 3). AMPK-γ2 expression was not evaluated because earlier research indicated that this isoform was not detectable in rodent skeletal muscle (Bultot et al., 2016). Whole muscle lysates from epitrochlearis muscles in AL and CR rats. A: Phosphorylated AMPKα1/2Thr172/AMPK-α1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-J: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1, AMPK-γ3 abundance. K: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 5). *P < 0.05, CR versus AL group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 6 per group. AL, ad libitum; CR, calorie restriction. Cytosolic-enriched fraction In the cytosolic-enriched fraction, there were no significant diet effects on the phosphorylation of either AMPK-α1/2 on Thr172 or ACC on Ser79 (Fig. 4A and C). In addition, the abundance of ACC, AMPK-α1/2 and AMPK subunits (α1, α2, β1, β2, γ1, and γ3) were not significantly altered by diet (Fig. 4). Cytosolic-enriched fraction from epitrochlearis muscles in AL and CR rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-J: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1, AMPK-γ3 abundance. K: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 6). Data was analyzed using Student’s t-test. Values are means ± SD; n = 10 per group. AL, ad libitum; CR, calorie restriction. Mitochondrial-enriched fraction In the mitochondrial-enriched fraction, phosphorylation of AMPK-α1/2 on Thr172 was moderately decreased (~30%) after CR intervention (P < 0.05, Fig. 5A). CR significantly increased abundance of AMPK-γ1 compared to AL rats (P < 0.05, Fig. 5I). There was a non-significant trend for increased abundance of AMPK-α1/2 (P = 0.057, Fig. 5B) and AMPK-β1 (P = 0.077, Fig. 5G) in CR group. In addition, there were no CR effects on abundance of ACC, AMPK subunits α1, α2, or β2 (Fig. 5). AMPK-γ3 was not detectable in the mitochondrial-enriched fraction. Mitochondrial-enriched fraction from epitrochlearis muscles in AL and CR rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-I: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1 abundance. J: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 7). *P < 0.05, CR versus AL group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 7–10 per group. AL, ad libitum; CR, calorie restriction. Nuclear-enriched fraction In the nuclear-enriched fraction, neither the phosphorylation of AMPK α1/2 on Thr172 nor ACC on Ser79 was affected by the CR (Fig. 6A and C). In addition, the abundance of ACC, AMPK-α1/2 or AMPK subunits (α1, α2, β1, β2, γ1) did not differ between CR and AL groups (Fig. 6). AMPK-γ3 was undetectable in the nuclear-enriched fraction. Nuclear-enriched fraction from epitrochlearis muscles in AL and CR rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-I: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1 abundance. J: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 8). Data was analyzed using Student’s t-test. Values are means ± SD; n = 10 per group. AL, ad libitum; CR, calorie restriction. Acute exercise effects on AMPK phosphorylation, ACC phosphorylation and abundance of AMPK subunits Body weights of the rats did not differ between the sedentary (524.2 ± 41.7 g) and exercise (540.2 ± 40.2 g) groups. There was no significant difference in epitrochlearis mass for sedentary (152.4 ± 18.7 mg) versus exercise (133.4 ± 7.3 mg) groups. Whole muscle lysate In the whole muscle lysate, exercise significantly increased phosphorylation of ACC on Ser79 (P < 0.05, Fig. 7C). There was non-significant trend for greater phosphorylation of AMPK α1/2 on Thr172 (P = 0.0998, Fig. 7A). The abundance of AMPK-α2 (P < 0.05, Fig. 7F) and −β1 (P < 0.05, Fig. 7G) were slightly decreased in exercised versus sedentary rats. Exercise did not affect the abundance of ACC, AMPK-α1/2, AMPK-α1, – β2, – γ1, or −γ3 (Fig. 7). Whole muscle lysates from epitrochlearis muscles in SED and IPEX rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-J: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1, AMPK-γ3 abundance. K: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 9). *P < 0.05, IPEX versus SED group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 8 per group. SED, sedentary; IPEX, immediately postexercise. Cytosolic-enriched fraction In the cytosolic-enriched fraction, there was an increase in phosphorylation of AMPK-α1/2 on Thr172 (P < 0.05, Fig. 8A) and ACC on Ser79 (P < 0.05, Fig. 8C) immediately after exercise. The abundance of ACC, AMPK-α1/2, AMPK-α1, −α2, −β1, −β2, −γ1 or −γ3 were unaltered by exercise (Fig. 8). Cytosolic-enriched fraction from epitrochlearis muscles in SED and IPEX rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-J: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1, AMPK-γ3 abundance. K: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 10). *P < 0.05, IPEX versus SED group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 8 per group. SED, sedentary; IPEX, immediately postexercise. Mitochondrial-enriched fraction In the mitochondrial-enriched fraction, similar to results in whole muscle lysate, exercise resulted in increased phosphorylation of ACC on Ser79 (P < 0.05, Fig. 9C), while phosphorylation of AMPK-α1/2 on Thr172 remained unchanged (Fig. 9A). There were no exercise effects on the abundance of ACC, AMPK-α1/2, AMPK-α1, −α2, −β2, or −γ1 (Fig. 9). Consistent with observations in the mitochondrial fraction from AL and CR rats, AMPK-γ3 was not detectable. The amount of protein in the mitochondria-enriched fraction was insufficient to assess AMPK-β1. Mitochondrial-enriched fraction from epitrochlearis muscles in SED and IPEX rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-H: Total AMPK-α1, AMPK-α2, AMPK-β2, AMPK-γ1 abundance. I: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 11). *P < 0.05, IPEX versus SED group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 4–8 per group. SED, sedentary; IPEX, immediately postexercise. Nuclear-enriched fraction In the nuclear-enriched fraction, exercise induced a greater phosphorylation of ACC on Ser79 (P < 0.05, Fig. 10C), without affecting the phosphorylation of AMPK-α1/2 on Thr172 (Fig. 10A). The abundance of ACC was slightly lower (~18%) in exercised rats compared to sedentary rats (P < 0.05, Fig. 10D). No changes in the abundance of either AMPK-α1/2, or AMPK subunits (α1, α2, β1, β2, or γ1) were observed after exercise (Fig. 10). AMPK-γ3 was not detectable in the nuclear-enriched fraction. Nuclear-enriched fraction from epitrochlearis muscles in SED and IPEX rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Total AMPKα1/2 abundance. C: Phosphorylated ACCSer79/ACC. D: Total ACC abundance; E-I: Total AMPK-α1, AMPK-α2, AMPK-β1, AMPK-β2, AMPK-γ1 abundance. J: Representative immunoblots (all of the immunoblots for this figure are shown in Supplementary Fig. 12). *P < 0.05, IPEX versus SED group. Data was analyzed using Student’s t-test. Values are means ± SD; n = 8 per group. SED, sedentary; IPEX, immediately postexercise. Correlation analysis of AMPK phosphorylation, ACC phosphorylation, and α-tubulin abundance between either cytosolic-enriched versus mitochondrial-enriched fractions or cytosolic-enriched versus nuclear-enriched fractions Phosphorylation of AMPK-α1/2 was highly and positively correlated between cytosolic-enriched and mitochondrial-enriched fractions in both the sedentary (R2 = 0.876, P < 0.05) and exercised groups (R2 = 0.909, P < 0.05, Fig. 11A). In contrast, no significant correlation was detected for ACC phosphorylation or for α-Tubulin abundance (Fig. 11B and C). A-C: Correlation analysis between cytosolic and mitochondrial fractions from SED and IPEX rats. A: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. B: Phosphorylated ACCSer79/ACC. C: α-Tubulin abundance. D-F: Correlation analysis between cytosolic and nuclear fractions from SED and IPEX rats. D: Phosphorylated AMPKα1/2Thr172/AMPKα1/2. E: Phosphorylated ACCSer79/ACC. F: α-Tubulin abundance. Data was analyzed using Pearson correlation. P ≤ 0.05 was considered statistically significant. n = 8 per group. SED, sedentary; IPEX, immediately postexercise. Phosphorylation of AMPK-α1/2 on Thr172, ACC on Ser79 and α-Tubulin abundance were not significantly correlated between cytosolic-enriched versus nuclear-enriched fractions in either the sedentary or exercised group (Fig. 11D–11F). Correlation analysis of LAMP2A abundance versus AMPK phosphorylation within either whole muscle lysates or cytosolic-enriched fractions LAMP2A protein levels were significantly higher in whole muscle lysates from CR versus AL rats, paralleling the CR-related increase in AMPK phosphorylation in the same samples. These findings prompted correlational analyses, which showed that AMPK-α1/2 phosphorylation was significantly and positively associated with LAMP2A abundance in whole muscle lysates from the CR group (R2 = 0.730, P < 0.05), but not from the AL group (Supplementary Fig. 4 A). In cytosolic-enriched fractions, AMPK-α1/2 phosphorylation was not significantly correlated with LAMP2A abundance in either diet group (Supplementary Fig. 4B). Discussion This study provides important new insights into how CR and exercise differentially influence the subcellular localization and activation of AMPK in the skeletal muscle of older rats. Notably, it is the first study to report the effects of CR on the localization and phosphorylation of AMPK and its substrate ACC within cytosolic-enriched, mitochondrial-enriched, and nuclear-enriched skeletal muscle fractions. Additionally, it is the first study to examine the impact of acute exercise on the phosphorylation of AMPK and ACC in nuclear-enriched fractions, as well as the influence of exercise on AMPK’s distribution across three subcellular compartments (cytosol, mitochondria, and nucleus) collected from the same muscle samples. Finally, this research uniquely evaluates the effects of each intervention on the localization and phosphorylation of AMPK and ACC in the skeletal muscle of older rats. The current results confirmed earlier research that CR increased phosphorylation of both AMPK and its substrate ACC in whole muscle lysates (Zheng et al., 2023), and also provided the novel finding that CR did not induce increased phosphorylation in the cytosolic, mitochondrial, or nuclear-enriched fractions. This suggests that CR may activate AMPK in other, as yet unidentified, subcellular compartments within skeletal muscle. Furthermore, given the central roles of the cytosol, mitochondria, and nucleus in mediating various effects of CR, these findings indicate that some CR-mediated outcomes may depend on a combination of AMPK-independent processes within these compartments and AMPK-dependent signaling originating outside them. It is intriguing that the only significant CR-effect on AMPK phosphorylation in the three compartments was a decrease in the mitochondrial-enriched fraction. This decline occurred together with a CR-related increase in AMPK-γ1 abundance and a non-significant trend for greater AMPK-α1/2 abundance. The increased mitochondrial AMPK-γ1 under CR may reflect enhanced recruitment or stabilization of γ1-containing AMPK complexes. The explanation for the lower AMPK phosphorylation with CR is uncertain. A highly speculative idea is that AMP/ATP levels may be lower in proximity to the mitochondria of muscles from CR rats, which in turn may account for the lower AMPK phosphorylation. In contrast to CR, acute exercise robustly increased AMPK phosphorylation in the cytosolic-enriched fraction and led to elevated ACC phosphorylation in the whole lysate as well as in cytosolic, mitochondrial, and nuclear-enriched fractions. These findings represented novel evidence that exercise activates AMPK signaling across multiple subcellular compartments in the skeletal muscle of older rats, highlighting its broad impact on muscle metabolism in aged animals. Importantly, both interventions employed the same muscle fractionation protocol. Therefore, the distinct compartmental responses observed cannot be attributed to methodological artifacts but rather reflect physiological differences in AMPK regulation in response to CR and exercise. This underscores the complexity of AMPK regulation in skeletal muscle. LAMP2A protein abundance was significantly greater for whole muscle lysates from CR compared to AL rats. Earlier research reported that older CR rats versus AL rats had greater LAMP2A mRNA levels in skeletal muscle (Wohlgemuth et al., 2010). LAMP2A is a recognized protein marker for lysosomes that is an essential component in the chaperone-mediated autophagy (CMA) pathway. The CMA pathway declines with advancing age, and this deficit contributes to aging related outcomes (Endicott, 2024). Lysosomal LAMP2A protein levels and CMA activation were greater in the liver of older CR rats and mice compared to age-matched AL controls (Jafari et al., 2024). Given these findings, which subcellular compartments should future CR-related research focus on? Multiple lines of evidence point to the lysosome as a promising target for exploring CR-induced activation of AMPK in skeletal muscle. In whole muscle lysates from the CR group, LAMP2A abundance was positively correlated with AMPK phosphorylation. This association provides indirect support for the idea that CR may regulate lysosome-localized AMPK. These results should be very cautiously interpreted because the measurements were made in whole muscle lysates rather than a lysosome-enriched fraction. The CR protocol used in this study did not change muscle levels of the primary allosteric regulators of AMPK (ATP or AMP) in older rats (Zheng et al., 2023). Importantly, lysosomal AMPK activation can occur via a mechanism that is independent of ATP or AMP concentrations (Zhang et al., 2017; Li et al., 2023). This mechanism involves the glycolytic metabolite fructose-1,6-bisphosphate (FBP): when FBP levels decline, less FBP binds to aldolase, facilitating the assembly of a lysosomal protein complex that triggers increased AMPK phosphorylation. Recent research also showed that inhibiting aldolase can activate lysosomal AMPK and reproduce several beneficial effects of CR, including lifespan extension in nematodes and mice (Zhang et al., 2022). Additionally, the endoplasmic reticulum (ER) is another important compartment to consider. AMPK regulates key lipid metabolism proteins associated with the ER, such as sterol regulatory-element binding protein 1c (SREBP-1c) and 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGCR) (Trefts and Shaw, 2021). Accordingly, future research on CR’s effects should pay particular attention to the lysosome and ER, as both are potentially critical sites for AMPK activity in skeletal muscle. Several studies have shown that acute exercise can alter the localization or phosphorylation of AMPK and/or ACC in the cytosol, mitochondria, and nucleus of skeletal muscle in young humans or rodents (McGee et al., 2003; Drake et al., 2021; Steinberg et al., 2006; Little et al., 2010). However, no previous studies provided information regarding the influence of exercise on AMPK’s localization in the muscles of older humans or rodents. Our study fills this gap by showing that exercise led to increased phosphorylation of both AMPK and ACC in the cytosolic-enriched fraction. However, in the mitochondrial- and nuclear-enriched fractions, exercise increased ACC phosphorylation even in the absence of elevated AMPK phosphorylation. One possible explanation for this apparent uncoupling is that there was a transient increase in AMPK phosphorylation that was reversed by the end of the exercise protocol (9 × 10-min exercise bouts that were separated by 10-min rest periods). It is also relevant to recognize that, although increased AMPK phosphorylation on α-Thr172 is often considered essential for increasing AMPK activity, there are exceptions to this relationship. For instance, treatment of cells with the allosteric AMPK activator A769662, in combination with AMP, can markedly increase AMPK activity without raising α-Thr172 phosphorylation (Scott et al., 2014; Ovens et al., 2021). Additionally, muscle contractile activity increases AMPK activity and ACC phosphorylation without a corresponding rise in α-Thr172 phosphorylation in skeletal muscle-specific LKB1-knockout mice (Koh et al., 2006). Therefore, although seeing increased phosphorylation of AMPK’s substrate ACC without a matching increase in AMPK phosphorylation is relatively uncommon, it is not unprecedented. A small amount of the cytosolic marker protein α-Tubulin was detected in the mitochondrial-enriched fraction, but its abundance was much lower than in the cytosolic-enriched fraction. Importantly, the levels of α-Tubulin in the cytosolic-enriched and mitochondrial-enriched fractions were not significantly correlated in either sedentary or exercised rats. Similarly, there was no significant correlation in ACC phosphorylation between the cytosolic-enriched and mitochondrial-enriched fractions in both groups. If the exercise-induced increase in ACC phosphorylation observed in the mitochondrial-enriched fraction were merely an artifact due to cytosolic contamination, significant correlations between these measurements would be expected. Therefore, these findings indicate that the minor presence of cytosolic proteins in the mitochondrial-enriched fraction is unlikely to account for the exercise-induced increase in ACC phosphorylation observed in the mitochondrial fraction. AMPK phosphorylation levels were positively correlated between the cytosolic-enriched and mitochondrial-enriched fractions in both sedentary and exercised groups. This association suggests the possibility of a common regulatory mechanism acting in both compartments. Because AMPK is localized to the outer mitochondrial membrane (Drake et al., 2021), it is exposed to cytosolic AMP/ATP levels. Thus, it seems possible that shifts in cytosolic AMP/ATP may drive parallel AMPK phosphorylation in both the cytosolic- and mitochondrial-enriched fractions. A previous study found that several AMPK subunits (α1, α2, γ1, and β2) were present in the mitochondrial-enriched fractions of skeletal muscle from young adult mice (gastrocnemius and tibialis anterior) and humans (muscles obtained during knee or shoulder surgery) (Drake et al., 2021). In that study, AMPK-γ3 was not detected in mitochondrial fractions from either species. Our findings in muscle from older rats closely align with their earlier results: we also identified AMPK-α1, −α2, −γ1, and −β2 in the mitochondrial-enriched fraction, while AMPK-γ3 was undetectable. In addition, we found that exercise did not alter the abundance of α1, α2, γ1, or β2 subunits in the mitochondrial-enriched fraction. These results are notable because the effects of exercise on AMPK subunits were not previously reported. Earlier research also showed that treadmill exercise (90 min at progressively increasing speed) in young male mice (13–15 weeks old) led to greater activation of AMPK at the outer mitochondrial membrane, as measured by an AMPK biosensor (Drake et al., 2021). Using a different approach, we likewise observed increased AMPK activation in mitochondria from older male rats, as indicated by enhanced phosphorylation of ACC, an AMPK substrate, within the mitochondrial-enriched fraction. Together, these findings demonstrate that several AMPK subunits are associated with mitochondrial fractions across multiple species (mice, humans, and rats). Moreover, distinct exercise protocols can activate mitochondrially localized AMPK in skeletal muscle from both young mice and older rats. Consistent with our findings from mitochondrial-enriched samples, no significant correlations were observed for either α-Tubulin or ACC phosphorylation between the cytosolic-enriched and nuclear-enriched fractions. However, unlike mitochondrial-enriched samples, there was also no significant correlation for AMPK phosphorylation between the cytosolic- and nuclear-enriched fractions. There was no significant correlation between AMPK phosphorylation in the cytosolic-enriched versus nuclear-enriched fractions from exercised muscle. Because an elevated cytosolic AMP/ATP ratio is a well-established trigger for increased AMPK phosphorylation, this disconnect suggests that cytosolic and nuclear signals may not be tightly coupled under the conditions studied. During exercise, ATP turnover is highest near the actin-myosin apparatus and would be expected to increase the local AMP/ATP ratio relative to other regions of the myocyte. Although ATP and AMP can diffuse between the cytosol and nucleus through the nuclear pore complex (NPC) (Talcott and Moore, 1999), their concentrations likely do not fully equilibrate across all intracellular compartments within the timescale examined. In addition to small molecules such as nucleotides, some proteins can enter the nucleus by passive diffusion through the NPC, with the upper size limit for passive diffusion proposed to be ~40–60 kDa (Kumeta, 2025). In contrast, much larger proteins can enter the nucleus via facilitated transport that requires protein binding to nuclear transport receptors (NTRs), which in turn, enable protein entry via the NPC (Kumeta, 2025). Immunohisto-chemical studies have demonstrated that both AMPK and ACC can be localized and phosphorylated within the nucleus (Kodiha et al., 2011; Wakamiya et al., 2014). Consistent with this, cultured cell experiments suggest that AMPK can shuttle into and out of the nucleus (Afinanisa et al., 2021). It has been proposed that AMPK is likely to enter the nucleus by an NTR-mediated mechanism (Afinanisa et al., 2021). Whether exercise influences protein movement between the cytosol and nucleus has not been directly tested. McGee et al. (McGee et al., 2003) reported that acute exercise by young men resulted in a significant increase in nuclear levels of AMPK-α2 in the vastus lateralis at 60 min after exercise (60 min of cycle ergometry at 70% of VO2max), but not at 10 min postexercise. They proposed that exercise may promote the translocation of AMPK into the nucleus (McGee et al., 2003). In contrast, our current study found no difference in nuclear levels of any evaluated AMPK subunits, including AMPK-α2, between sedentary muscles and muscles sampled immediately after exercise. Several factors may explain these differing outcomes, including differences in species and ages of the subjects, exercise protocols, and the specific muscles analyzed. It also may be important to examine nuclear levels of AMPK-α2 and other AMPK subunits in skeletal muscles of older animals at later time points following exercise. While earlier research focused solely on nuclear-localized AMPK-α2 following exercise (McGee et al., 2003), our research also examined other AMPK subunits (α1, β1, β2, and γ1) and found no exercise-induced changes in their levels within the nuclear-enriched fraction. Furthermore, AMPK-γ3 was undetectable in the nucleus, regardless of exercise status. Which subcellular compartments should be prioritized in future exercise-related research? As discussed above, AMPK is present and functionally significant in both the lysosome and ER, making these compartments compelling targets for exercise studies. AMPK-γ3 is the only subunit that is selectively expressed primarily in skeletal muscle, and this isoform is crucial for several important metabolic outcomes of exercise (Wang et al., 2022; Kido et al., 2023; Smiles et al., 2024; Barnes et al., 2002). Additionally, immunofluorescence staining has shown that the AMPK-γ3 protein colocalizes with t-tubules and the SR in mouse skeletal muscle (Pinter et al., 2013). During exercise, calcium pumping is increased in these structures, likely raising local AMP/ATP ratios and thereby promoting AMPK activation. Therefore, examining how exercise influences AMPK activation and localization with the t-tubules and SR represents a promising opportunity to advance our understanding of skeletal muscle physiology. AMPK has been linked to a wide range of functional and health-related outcomes associated with chronic CR or acute exercise, such as enhanced insulin sensitivity, changes in mitochondrial function, regulation of cell signaling, and transcription (Trefts and Shaw, 2021; Wang et al., 2012; Kjobsted et al., 2018; Reisman et al., 2024). Although the current study did not directly examine the relationship between these specific outcomes and patterns of AMPK activation, previous work by our group has shown that similar CR and exercise protocols increase insulin-stimulated glucose uptake in the epitrochlearis muscle of older rats (Oki et al., 2020; Wang et al., 2016). Future studies should determine how AMPK activation in distinct subcellular compartments contributes to the beneficial effects of CR and/or exercise on insulin sensitivity and other important outcomes. For example, under unstimulated conditions, the insulin-responsive GLUT4 glucose transporter is largely sequestered in GLUT4 storage vesicles (GSVs) and is redistributed to cell surface membranes, including the sarcolemma and t-tubules, in response to insulin (Cartee, 2015). Key insulin-signaling proteins that regulate GLUT4 translocation, including the AMPK substrate TBC1D4, can be localized in the cytosol or associated with GSVs (Mafakheri et al., 2018). Therefore, assessing AMPK localization and activation in relation to these compartments could provide important insight into mechanisms underlying improved insulin sensitivity. AMPK has also been implicated in several other skeletal muscle processes that may be influenced by CR and/or exercise, including lysosome-associated autophagy, mitochondrial ATP synthesis, and nuclear transcriptional regulation (Trefts and Shaw, 2021; Kjobsted et al., 2018; Reisman et al., 2024). Accordingly, future research should investigate mechanistic links between compartment-specific AMPK activation and these cellular functions. The current study has several limitations. First, only male rats were included, so the effects of CR or exercise on nuclear and mitochondrial AMPK abundance and phosphorylation in female rats remain unknown. Second, the study focused exclusively on older animals, thereby limiting insight into how CR or exercise may influence AMPK subcellular localization across the lifespan. Third, AMPK activation was measured only immediately after exercise. It is possible that peak activation in some cellular compartments occurs at other time points, which were not assessed. Fourth, the study investigated the influence of acute exercise but did not examine the effects of chronic exercise training on AMPK localization. Additionally, CR and exercise effects were analyzed only in the epitrochlearis muscle. While this muscle’s fiber type composition is representative of rat skeletal muscle as a whole (Castorena et al., 2011; Delp and Duan, 1996), it would be beneficial to investigate muscles with different fiber type profiles. Our muscle preparation protocol successfully yielded fractions enriched for cytosolic, mitochondrial, or nuclear proteins, with relatively modest cross-contamination among compartments. Nevertheless, subcellular fractionation of skeletal muscle by homogenization and differential centrifugation has inherent limitations for definitively establishing spatial localization. These include disruption of cellular structures during homogenization, loss of native spatial organization, and incomplete compartmental purity. Therefore, future studies should use cellular imaging approaches, such as immunofluorescence microscopy, to more precisely define AMPK’s subcellular localization. This study revealed strikingly distinct patterns of AMPK regulation induced by two important physiological interventions that can improve both the health and function of the skeletal muscle in older rats. While CR increased AMPK and ACC phosphorylation in whole muscle lysates without affecting these outcomes in three functionally important subcellular fractions, exercise broadly activated AMPK signaling across cytosolic-enriched, mitochondrial-enriched, and nuclear-enriched compartments. The distinctive compartment-specific responses to chronic CR and acute exercise likely reflect the distinct biological contexts of these interventions. CR represents a chronic adaptation to reduced energy availability, whereas the acute exercise protocol imposed a relatively brief, but intense contractile and metabolic challenge. Therefore, although both interventions can activate AMPK in the whole muscle lysate, they are clearly not biologically equivalent stimuli. Our findings underscore the complexity and context-dependent nature of AMPK regulation in aging muscle. Furthermore, they point to several specific intracellular compartments that should be high priority targets for future research to advance our understanding of how AMPK integrates metabolic signals to optimize muscle function and health during aging. Ultimately, continued research in this area has the potential to inform the development of therapeutic strategies aimed at preventing age-related muscle decline. Supplementary Material Acknowledgments The authors thank Dr. David Thomson (Brigham Young University) for generously providing the γ3-AMPK antibody. We are grateful to Gabriela Rodriguez and Emily Ma for their valuable assistance with the exercise protocol. Funding This research was supported by the National Institutes of Health (R21AG084931). Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.exger.2026.113212. Footnotes CRediT authorship contribution statement Haiyan Wang: Writing – review & editing, Writing – original draft, Visualization, Supervision, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Jiwei Hao: Writing – review & editing, Investigation. Jun-Hyun Bae: Writing – review & editing, Investigation. Gregory D. Cartee: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Conceptualization. Ethics approval Animal studies were conducted in accordance with the guidelines from the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and with approval of University of Michigan Committee on Use and Care of Animals. Declaration of Generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used a generative AI-based language model (U-M GPT-5.2) to assist with improving the clarity of the text. After using this tool, the authors reviewed and edited the content as needed, and they take full responsibility for the content of the published article. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Haiyan Wang, Jiwei Hao, Jun-Hyun Bae, and Gregory D. Cartee report financial support was provided by National Institutes of Health (R21AG084931). Data availability Data will be made available on request. References Associated Data Supplementary Materials Data Availability Statement Data will be made available on request.