PoWeR elicits intracellular signaling, mitochondrial adaptations, and hypertrophy in multiple muscles consistent with endurance and resistance exercise training

Abstract

Physical activity guidelines recommend both endurance and resistance exercise to improve and maintain overall health. Recently, progressive weighted wheel running (PoWeR), a voluntary, progressive, and high-volume exercise paradigm, was posited as a singular prototype of combined endurance and resistance exercise in mice as evident by enhanced capillarization and hypertrophy of select plantar flexor muscles. Despite growing interest in this model, it remains incompletely characterized if PoWeR resembles the acute and chronic responses to resistance and/or endurance exercise in humans. Therefore, the purpose of this study was to assess canonical signaling events, mitochondrial bioenergetics, and cellular adaptations across multiple extensor and flexor muscles of the fore- and hindlimbs that may be conducive for whole-body functional improvements as traditionally observed in humans. 8-weeks of PoWeR (~8km/day) improved glucose metabolism, exercise capacity, body composition, and bone mineral density as well as increased mass, myofiber CSA, and oxidative myofiber type distribution in the soleus, plantaris, and FDL. Using two ex-vivo high-resolution flourorespirometry protocols that model in vivo physiological conditions, PoWeR decreased mitochondrial ADP sensitivity which was accompanied by greater mitochondrial H₂O₂ emissions, respiration, conductance, and protein content in the vastus lateralis, gastrocnemius, and triceps in muscle-specific fashion. Three days of short-term PoWeR stimulated mTORC1 and AMPK signaling in soleus, plantaris and/or FDL in line with the hypertrophic and metabolic adaptations observed with long-term training. Collectively, these data support PoWeR as a suitable paradigm in mice to model the acute signaling and chronic adaptations associated with endurance and resistance exercise in humans.

New and Noteworthy

Using PoWeR, we evaluated skeletal muscle mitochondrial and hypertrophic adaptions revealing muscle specific adaptations across fore and hind limbs consistent with endurance and resistance exercise in humans. We present a short-term PoWeR paradigm that identifies muscle specific signaling responses thought to support long term adaptions to PoWeR. These data provide further support for PoWeR as a model to resemble the metabolic and anabolic adaptions to endurance and resistance exercise in humans.

Graphical Abstract

Introduction

Exercise is essential to improve or maintain physiological function and decrease the risk of developing multi-morbidity and mortality (1–4). The American College of Sports Medicine (ACSM) guidelines suggest adhering to both endurance and resistance activity most days of the week to combat sarcopenia and cardiometabolic disease (5). Exercise studies in humans can be limited by tissue availability and the requisite invasiveness to comprehensively evaluate the cellular and molecular mechanisms that that confer tissue specific and whole-body health benefits of exercise. Therefore, there is a strong need for exercise paradigms in animals that resemble adaptations to both endurance and resistance exercise observed in humans.

Numerous approaches have been utilized in rodents to model whole-body and tissue specific adaptions to endurance or resistance exercise, each with strengths and limitations as previously detailed (6,7). Forced treadmill and voluntary, unweighted wheel running are the most common modes of endurance exercise. Electrical stimulation and surgical models of overload have been traditionally used to induce muscle hypertrophy in rodents. More recently, weighted cart pulling, magnetically driven continuous resistance wheel running, and other paradigms have also been developed as models of resistance exercise (8,9). While each modality provides a spectrum in the magnitude of adaptations and potential translatability to humans, there are several key limitations such as high entry costs for specialized equipment, need for circadian re-alignment, involuntary or stress stimuli, activation of only 1–2 muscles to non-physiologically induce rapid muscle growth, and/or substantial investigator time and oversight (10–13). Further, these models are largely specific to adaptations associated with endurance or resistance exercise, but rarely model adaptations commonly observed after concomitant endurance and resistance exercise as recommended by the ACSM physical activity guidelines.

Progressive weighted wheel running (PoWeR) circumvents many of these limitations since it is a relatively simple, low-cost approach and a voluntary, full body, high volume exercise with progressive loading that recapitulates select adaptions commonly associated with both resistance (myonuclear accretion, muscle mass, myofiber hypertrophy) and endurance training (capillarization, cardiac mass and left ventricular thickness, oxidative fiber type shift) in adult and/or aged mice (10,11,14,15). Despite the growing interest of using PoWeR to evaluate the cellular and molecular responses to exercise, seminal studies focused largely on growth of plantar flexor muscles, the plantaris, soleus and gastrocnemius (10–12,14–17). It remains unknown if PoWeR can stimulate muscle hypertrophy and mitochondrial adaptations in flexor and extensor muscles across fore and hind limb advantageous for whole-body metabolic and functional performance. Therefore, to expand our understanding of adaptations to PoWeR, the purpose of this investigation was to examine whole muscle, cellular, and mitochondrial adaptations across multiple muscles that accompany physical and metabolic functional improvements. Further we adapted a short-term PoWeR approach to measure canonical signaling events associated with long-term skeletal muscle adaptations after exercise training. Given the seminal studies predominantly used female mice due to consistent running behavior, we completed this study in female mice. The main findings from this investigation suggest that PoWeR is a robust tool to elicit systemic improvements to whole-body glucose metabolism, physical function, and body composition as well as stimulate intra-cellular signaling putatively involved in mitochondrial and hypertrophic adaptations across multiple muscles and in a muscle-specific fashion. These data provide an initial resource to assist skeletal muscle biologists and exercise physiologists with future experimental designs to interrogate mechanisms regulating adaptations akin to endurance and resistance exercise training in humans.

Methods

Ethical Approval:

All animal procedures were performed in conformance with institutional guidelines and were approved by the Institutional Animal Care and Use Committee of the William S. Middleton Memorial Veterans Hospital and the University of Wisconsin-Madison. Female C57BL/6J were procured from the Jackson Laboratory (000664) at 16–18 weeks of age. Mice were acclimated to the animal research facility for a minimum of 2-weeks before entering studies. All mice were singly housed, provided Purina Rodent Chow (5001) ad libitum, and provided enrichment to minimize stress. The animal room was maintained with a 12:12-hr light-dark cycle at 23°C.

Long-Term PoWeR Training:

Progressively weighted running wheels (PoWeR) were used similarly to previously published work (10) and as shown in Fig 1A. Briefly, 20-week-old female C57BL/6J mice (n=20) were acclimated to unweighted running wheels for 7 days. Following 7 days, mice were allocated to 7 weeks of sedentary control (Sed) without wheel access or 7 additional weeks of exercise training (PoWeR) (n=10/group), based on propensity to run. 1-gram magnets were affixed to one side to asymmetrically load the running wheel (11cm diameter). Weights were progressively added each week starting at 2g until 6g as previously performed (10). Wheel running was monitored for running distance (kilometers/day) using ClockLab Software (Actimetrics, Wilmette, IL).

Figure 1 –

Schematic for (A) long- and (B) short- term progressive weighted wheel running (PoWeR). GTT, glucose tolerance test; PTT, pyruvate tolerance test; GXT, graded exercise test; CSA, cross sectional area. Figure created in BioRender.com.

Graded Exercise Test:

Graded exercise tests were performed on a Columbus 3/6 treadmill (Columbus Instruments, Columbus OH) before (Pre) and after 7 weeks of PoWeR. Mice were acclimated for 2 consecutive days prior to testing. Acclimation consisted of mice sitting on a stationary treadmill for 3 minutes with the shock grid activated to (3hz and 1.5mA). Next, the treadmill was inclined to 10 degrees and speed of 6m/min for 5 minutes then progressively increased to 12m/min for 5 more minutes for a total of 10 minutes of treadmill movement. A graded exercise test to exhaustion was performed following a previously published method for measuring VO₂ max in mice (18). Exhaustion was defined as spending 5 consecutive seconds in contact or repeated contact with the shock grid demonstrating an inability to reengage treadmill running. For post-training testing, mice completed one acclimation day prior to graded exercise testing. All acclimations and graded exercise tests were performed in the early dark phase approximately 30–90 minutes after light-dark transition.

Glucose Tolerance Test:

Glucose tolerance tests (GTT) were performed after a 4 hour fast and 2 hours after the start of the light phase before (Pre) and after (Post) 6.5 weeks of PoWeR. Running wheels were locked for 24 hours prior to the GTT. Glucose (2 g/kg) in a 30% saline solution was intraperitoneally injected and blood glucose was measured via a tail nick before (0 minutes) and 15, 30, 45, 60, 90, and 120 minutes following injection with a Bayer Contour Blood Glucose Meter (Bayer). Area of the blood glucose curve (AOC) was quantified as the area above baseline blood glucose established as the 0 min blood glucose reading for each mouse.

Pyruvate Tolerance Test:

Post PoWeR pyruvate tolerance testes (PTT) were performed after a 6 hour fast and 2 hours after the start of the light phase after 5.5 weeks of PoWeR. Running wheels were locked for 24 hours prior to the tolerance test. Sodium Pyruvate (2 g/kg) in a 30% saline solution was intraperitoneally injected and blood glucose was measured via a tail nick before (0 minutes) and 15, 30, 45, 60, 90, and 120 minutes following injection with a Bayer Contour Blood Glucose Meter (Bayer). AOC was quantified as the area above baseline blood glucose established as the 0 min blood glucose reading for each mouse.

Body Composition:

Body composition using an EchoMRI 3-in-1 body composition analyzer was completed on all mice after the 7^th week of the intervention. Mice were placed into a restraint tube (no anesthesia) for <2 min of total time generating duplicate measures of fat free mass (FFM) and fat mass. Duplicate measures were averaged for final data use.

Grip Strength:

Grip strength was measured by the peak force of all limbs (fore and hind) using a Grip Strength Meter (Columbus Instruments, Columbus, OH) before (Pre) and after (Post) 7.5 weeks of PoWeR. Mice were placed on the horizontal grid connected to a force transducer. Mice were held by the base of the tail and when the mouse fully gripped the grid, the mouse was pulled horizontally at a consistent speed until the grip released. This test was repeated 3 times with 15 minutes between tests. Grip strength was measured as highest grams of force across the 3 attempts.

Animal Dissections:

Following the 8-week intervention, running wheels were locked for 24-hours and mice were euthanatized via cervical dislocation. Euthanasia was performed in the morning approximately 2 hours after lights on following a 2 hour fast. The gastrocnemius, quadriceps, triceps brachii, soleus, plantaris, flexor digitorum longus (FDL) were collected, weighed, and prepared for mitochondrial bioenergetics and immunohistochemistry within 6 minutes of euthanasia. Then the tibialis anterior (TA) and extensor digitorum longus (EDL) were excised from the left leg and all muscle collections, weighing procedures, and snap freezing in liquid nitrogen were repeated on the right leg. Subsequently, the liver and heart, were excised and weighed. Muscle weights were measured in duplicate with an average 6% coefficient of variability across all muscles. Data are expressed as averaged weight of duplicates relative to bodyweight. All dissections were completed by the same investigator to limit variability.

Muscle Immunohistochemistry and Analysis:

Soleus, Plantaris, and FDL from the left leg were weighed, measured for muscle length, and submerged in optimum cutting temperature compound (OCT, Tissue-Tek; Sakura Finetek, The Netherlands) at resting length and frozen in liquid nitrogen chilled isopentane as previously performed (8,19). Mid-Belly cross sections (10 μm thick) from OCT frozen muscles were fixed for 10 min with −20°C acetone for myofiber typing. Fixed muscle sections were washed in PBS at room temperature for 15 minutes and then blocked for 20 minutes in blocking buffer (0.5% Triton X-100, 0.5% BSA dissolved in PBS). Slides were incubated in primary antibodies containing rabbit anti-laminin (1:500, #L9393, MilliporeSigma, Burlington, MA, USA), mouse IgG2b anti-Type I MHC (1:100, # BA-D5-s, Developmental Studies Hybridoma Bank, Iowa City, IA, USA), mouse IgG1 anti-Type IIA MHC (1:100, #SC-71-s, Developmental Studies Hybridoma Bank, Iowa City, IA, USA), and mouse IgM anti- Type IIB MHC (1:10, #BF-F3-s, Developmental Studies Hybridoma Bank, Iowa City, IA, USA) for 1 hour at room temperature. Slides were washed 3 times for 5 minutes with PBS then incubated in secondary antibodies containing Alexa 568 goat anti-rabbit IgG (1:5000, #A11011, Invitrogen, Waltham, MA, USA), Alexa 647 goat anti-mouse IgG2b (1:100, #115–605-207, Jackson Immunoresearch, West Grove, PA, USA), Alexa 488 goat anti-mouse IgG1 (1:3000, #115–545-205, Jackson Immunoresearch, West Grove, PA, USA), and Alexa 350 goat anti- mouse IgM (1:500, #A-31552, Invitrogen, Waltham, MA, USA) for 1 hour at room temperature. Slides were washed with PBS 3 times for 5 minutes before mounting and cover slipping with ProLong Gold antifade mountant (Invitrogen, Waltham, MA, USA). Whole muscle cross-sections were imaged with a 10X objective on a BZ-X700 Keyence microscope with 4 different filters (DAPI, GFP, TRITC, CY5) (Keyence, Itasca, IL, USA). Myofiber type cross sectional area measurements from whole muscle cross sections were determined using a previously published CellProfiler pipeline (8).

Myonuclei staining was completed on mid-belly cross sections (10 μm thick ) from the OCT frozen FDL muscle. Cross sections were fixed in 1% paraformaldehyde dissolved in PBS for 10 minutes at room temperature. Following fixation samples were washed with PBS followed by blocking (0.5% Triton X-100, 0.5% BSA, dissolved in PBS) for 20 minutes at room temperature. Samples were then incubated with mouse anti-dystrophin IgG1 (1:20 #NCL-DYS3, Leica, Buffalo Grove, IL, USA) for 1 hour at room temperature. The sections were washed with PBS 3 times for 15 min with PBS before incubation with Alexa 488 goat anti-mouse IgG1 (1:500, #115–545-205, Jackson Immunoresearch, West Grove, PA, USA) for 1 hour at room temperature. Samples were washed again with PBS followed by incubation with Hoechst (1ug/mL, #D1306, Invitrogen, Waltham, MA, USA) for 5 minutes at room temperature before coverslipping with ProLong Gold antifade mountant (Invitrogen, Waltham, MA, USA).

For myofiber typing and myonuclei staining whole muscle cross-sections were imaged with a 10X objective on a BZ-X700 Keyence microscope with 4 different filters for myofiber typing (DAPI, GFP, TRITC, CY5) or 2 filters for myonuclei (DAPI, GFP) (Keyence, Itasca, IL, USA). Myofiber type cross sectional area and myonuclei measurements from whole muscle cross sections were determined using a previously published CellProfiler pipeline (8).

Mitochondrial Bioenergetics:

The middle third of the lateral head of the gastrocnemius, middle third of the vastus lateralis, and half of the anterior triceps were preserved in buffer X containing (in mM) 7.23 K₂EGTA, 2.77 CaK₂EGTA, 20 imidazole, 20 taurine, 5.7 ATP, 14.3 phosphocreatine, 6.56 MgCl₂·6H2O, and 50 K-MES (pH 7.1) on ice until permeabilization. Muscles were mechanically and chemically permeabilized by separating into small muscle bundles with fine tipped forceps and then incubated in saponin (50μg/mL) for 30 minutes on ice. Permeabilized muscle fibers (PMFs) were washed with Buffer Z (in mM) 105 K-MES, 30 KCl, 10 KH₂PO4, 5 MgCl₂ 1 EGTA, BSA (0.5g/L) (pH 7.2 ) and blebbistatin (25μM) for 15 mins, blotted on filter paper to remove excess liquid, weighed (~1.5–2.5mg), then placed back in buffer Z and blebbistatin until placed into the Oroboros O2k chambers at 37°C with a stir bar set to 700rpm. Two protocols were performed on PMF under hyperoxygenated (250uM to 425uM) conditions.

Protocol 1 Flourorespirometry:

High resolution flourorespirometry was used to measure mitochondrial oxygen consumption and hydrogen peroxide emissions (mtH₂O₂ emissions) simultaneously (20). Amplex Ultra Red (10μM), Horseradish peroxidase (HRP 1 U/mL), and superoxide dismutase (SOD (5 U/mL) were added to Buffer Z in each chamber. Subsequently a hydrogen peroxide calibration (0–1uM) was performed by sequential injections. PMF bundles were energized with substrates; succinate (10mM), pyruvate (2mM), glutamate (2mM), and malate (2mM) to induce maximal mtH₂O₂ emissions, before ADP was titrated from 0–6.2mM. Cytochrome C (10μM) was used to test membrane integrity and any change greater than 20% was deemed a loss of membrane integrity and removed from subsequent analysis. Oxygen consumption and mtH₂O₂ emissions were normalized to PMF wet weight and the average of both chambers was used for further data analyses.

Protocol 2 Creatine Kinase Clamp:

The creatine kinase clamp assay was devised to assess mitochondrial bioenergetics under conditions of physiologically relevant energetic demand that circumvent non-physiological concentrations of ADP stimulated respiration. The current protocol is adapted from previously published methods where the enzymatic activity of creatine kinase is leveraged with exogenous ATP and Phosphocreatine (PCr) to control the energetic demand measured as Gibbs free energy of ATP hydrolysis (ΔG_ATP) (21). PMFs were placed into the assay with buffer Z supplemented with 5mM ATP, 20U/mL creatine kinase, 5mM Creatine, and 1mM PCr set to a gibbs free energy of ATP hydrolysis of −12.74kcal/mol calculated by an online resource https://dmpio.github.io/bioenergetic-calculators/ck_clamp/. PMFs were energized with substrates pyruvate (2mM), glutamate (2mM), and malate (0.4mM ) to measure complex I-linked respiration. Then succinate (10 mM) was added to measure complex I & II-linked respiration followed by sequential titrations of PCr from 1–21mM to shift the (ΔG_ATP) from −12.74kcal/mol to −14.7kcal/mol. Conductance, a measurement of resistance within the energy transduction pathway, is calculated by plotting mitochondrial respiration rate (pmol O₂ s⁻¹ mg⁻¹) against (ΔG_ATP) during the phosphocreatine titrations. Respiratory conductance is measured by the slope reflecting the relative flux of electrons through the respiratory chain (21). A greater slope is indicative of increased sensitivity to respective substrates during changing energetic demands.

μCT scans:

Following euthanasia and muscle collection, hind limbs were removed from the mouse at the femoral head and fixed in formalin for 72 hours until long term storage in 70% ethanol. Hind limbs were scanned submerged in 70% ethanol using an MI-Labs U-SPECT/CT system, achieving 10-micron resolution. Analyses were then performed by a blinded evaluator. In ImageJ, 1mm ROIs were selected along the femoral and tibial mid-shaft. Scans were manually thresholded for bone and the mean gray value within the thresholded region was measured to produce a relative mineral density measurement versus control. Mean cortical thickness was then measured using the BoneJ plugin. Two samples in the PoWeR group were excluded due to movement during scans.

Short-Term Exercise Training:

As shown in Fig 1B, 16 female C57BL/6J mice (18-week-old) were acclimated to unweighted running wheels for 5 days. After the 5^th day, mice were randomized to sedentary (locked wheel) or PoWeR for three days (n=8/group) with a daily progression of 2, 3, and 4g. Daily running distance did not change with increasing resistance on the running wheel. A premeasured amount of food was provided 6 hours prior to euthanasia to mimic the food availability during long-term PoWeR and account for food consumption on nutrient signaling pathways. During the last night, mice were sacrificed via cervical dislocation 1 hour after locking wheels (PoWeR group only) and 4 hours after the start of the dark phase. Muscle tissues (in order: soleus, plantaris, FDL, gastrocnemius, quadriceps, and triceps) were collected less than 6 minutes after euthanasia from the left leg, snap frozen in liquid nitrogen and stored at −80°C for later biochemical analysis.

Immunoblotting:

Muscles from short-term and long-term PoWeR (gastroc, soleus, plantaris, FDL, triceps, and quad) were powdered in liquid nitrogen-cooled mortar and pestle and bead homogenized in RIPA lysis buffer (150mM NaCl, 0.1mM EDTA, 50mM Tris, 0.1% wt/vol deoxycholate, 0.1% wt/vol SDS, 1% vol/vol Triton X-100) with PhosSTOP™ phosphatase and Complete™ Protease inhibitor cocktails (Roche: 05892970001, PHOSS-RO). After determining protein concentration with BCA Assay (Thermo Scientific, 23225), protein lysates were mixed with β-mercaptoethanol in 2 or 4x Laemmli Sample Buffer (BioRad 1610737), then heated at 35°C for OXPHOS proteins or 95°C for other proteins. Protein lysates (7.5 μg for OXPHOS and 15 μg for other proteins) were separated on 4–15% TGX Precast Gels (BioRad, 4561083) and transferred to 0.2μm nitrocellulose membranes (BioRad). Membranes were blocked in TBS-Tween 20 (0.1%) (TBST) with 5% BSA and then incubated overnight with primary antibodies for OXPHOS (Abcam MS604, 1:1000), P-RPS6 Ser235/236 (Cell Signaling, 4858, 1:2000), RPS6 (Cell Signaling, 2217, 1:1000), P-P70S6K T389 (Cell Signaling, 9234, 1:1000), T-P70S6K T389 (Cell Signaling, 2708, 1:1000), P-AMPKα Thr172 (Cell Signaling, 2535, 1:1000), T-AMPKα (Cell Signaling, 2532, 1:1000), P-ACC S79 (Cell Signaling, 3661, 1:1000), T-ACC (Cell Signaling, 3662, 1:1000), P-AKT S473 (Cell Signaling, 4060, 1:1000), AKT (Cell Signaling, 4685, 1:1000). Primary antibody was diluted in in 5% BSA/TBST, while OXPHOS in 1% Nonfat milk/PBS (Biorad, 1706404). Membranes were then incubated with secondary antibodies: anti-rabbit (Cell Signaling, 7074, 1:5000) or anti-mouse (Abcam, 6728, 1:10 000) and exposed with SuperSignal Pico for OXPHOS or Pico/Femto 1:1 substrate mix (Fisher, PI34095, PI34577) and imaged with Biorad ChemiDoc MP. Membranes were stripped with Restore Stripping Buffer (ThermoFisher Scientific, 21059). Ponceau S stain was performed for total protein normalization in OXPHOS blots. Densitometric calculations were determined using Image Lab Software (BioRad)

Statistical Analysis:

Pre-Post analyses were completed using repeated measures 2-way ANOVA with main effects for Time (pre-post) and Group (Sedentary vs. PoWeR). When an interaction was present, Holm-Sidak multiple comparison testing was performed. Any cross-sectional comparisons were evaluated with an unpaired T-test. P<0.05 was determined as significant. The first 2 concentrations of ADP (0.025 and 0.05mM) were not used to estimate the apparent sensitivity to ADP (K_m ADP) via Michaelis-Menten Kinetics because these concentrations did not stimulate respiration and therefore introduced improper curve fitting. Data were screened for potential outliers and samples were removed based on Grubbs outlier testing (alpha = 0.05). Data were analyzed and figures created using Prism GraphPad (Version 10.3.1).

Results

PoWeR elicited numerous physiological and metabolic adaptations

Average daily running volume per week and progressive wheel weight resistance is presented in Fig 2.A. Adult female mice ran on average a cumulative distance of 460 kilometers therefore averaging 7.7 km/day over 8-weeks of PoWeR (Fig 2A & B) However, an important observation was adult female mice were averaging 9.9 km/day for the first 5 weeks of PoWeR prior to the stress from metabolic and physiological testing. After the 5^th week, when physiological testing began, running volume declined to 7.6 km/day (Fig 1A, Fig 2A). An additional observation we note is following metabolic or physiological testing it takes approximately 1–4 days for mice to return near pre-testing running behavior (data not shown).

Figure 2 –

Physiological and metabolic adaptions to PoWeR. (A) Average daily voluntary running distance per week during 8-weeks of PoWeR. (B) Cumulative and average daily running distance over the duration of PoWeR per mouse. (C) Bodyweight, (D) fat free mass, and (E) fat mass body composition measured by EchoMRI. (F) Muscle mass of soleus (SOL), plantaris (PLA), flexor digitorum longus (FDL), tibialis anterior (TA), and quadriceps (QUAD) expressed relative to Sed. (G) Heart and (H) Liver mass normalized to body mass and expressed relative to sedentary. (I) Time run during a maximal graded exercise treadmill test. (J) Grip strength expressed as absolute grams of force. (K) Glucose tolerance test and (L) Pyruvate tolerance test with respective area of the curve. (M) Femoral bone mineral density measured by μCT. N=9–10 per group. Mean plus individual data points or error bars represent mean ± SEM. 2-way ANOVA with repeated measures was used to assess main effects of group, time, and group x time interaction (B, J, K, L) otherwise unpaired t-test (C-E, F-I, M). *P<0.05, **P<0.01, ***P<0.001.

Both PoWeR and Sed mice increased body mass over time with no difference between groups (Fig 2B). However, PoWeR trained mice had greater fat-free mass (FFM) and less adiposity versus Sed control (Fig. 2C & D). The greater FFM was likely driven by greater mass (P<0.05) of the SOL, PLA, FDL, TA, Quad (P=0.07), heart and liver relative to sedentary counterparts (Fig. 2E-G). Gastroc, triceps and EDL muscle mass was not different between PoWeR and Sed control (Supp Fig. 1B). Further, PoWeR trained mice had greater femoral but not tibial bone mineral density (Fig 2M, Supplemental Fig. S1C).

The greater skeletal and cardiac mass in PoWeR trained mice was accompanied by increased time (Fig 2J) and distance (not shown) during the GxT as well as all-limb grip strength (Fig. 2K). To assess whole body glucose uptake and gluconeogenesis, we performed a glucose and pyruvate tolerance test, respectively. PoWeR improved glucose (Fig. 2L) and pyruvate tolerance (Fig. 2M) as evident by a decrease in AOC. These results support PoWeR as an exercise model to elicit whole-body functional, metabolic and body composition adaptions associated with both endurance and resistance exercise training.

PoWeR induced robust mitochondrial adaptions in representative fore and hindlimb muscles

Increased skeletal muscle mitochondrial respiratory capacity and volume are hallmark adaptations to exercise training. However, traditional approaches assess maximal mitochondrial capacity and do not interrogate how different muscles respond to exercise training using approaches that more closely resemble in vivo physiological energetic demands (22). We focused our mitochondrial bioenergetic analysis on 3 muscles, the vastus lateralis (VL), gastrocnemius (gastroc), and triceps brachii which are large, fiber type diverse, and region specific.

Compared to Sed, PoWeR trained mice had greater mitochondrial respiration rates throughout the ADP titration protocol in all 3 muscles (Fig. 3A). Maximal respiratory rates (ADP 6.2mM) were 110%, 66%, and 46% greater in the VL, gastroc, and triceps, respectively (Fig. 3A). In the gastroc and VL, but not the triceps, PoWeR trained mice had lower mitochondrial ADP sensitivity as indicated by a rightward shift in the curve for ADP driven respiration (Fig. 3B) and a greater apparent Km of ADP (Fig. 3C).

Figure 3 –

Mitochondrial oxygen consumption in the vastus lateralis (VL), Gastroc, and Triceps. (A) ADP titration with substrates, succinate, pyruvate, glutamate, and malate. (B) ADP-stimulated respiration displayed as a Michaelis-Menten kinetic curve used to estimate the (C) apparent K_m of ADP. N=9–10 per group. Points and error bars represent mean ± SEM. Groups analyzed by multiple unpaired t-test (A) or unpaired t-test (C). *P<0.05, **P<0.01, ***P<0.001

PoWeR trained mice had greater absolute maximal mtH₂O₂ emissions compared to Sed controls in the VL, gastroc, and triceps (Fig. 4A). To estimate the apparent IC₅₀ of ADP to suppress mtH₂O₂ emissions by 50%, we expressed H₂O₂ emissions relative to maximal emission rate during the ADP titration (Fig. 4B). IC₅₀ was greater in the VL and gastroc (P=0.07) from PoWeR trained mice versus Sed and lower in the triceps (Fig. 4C). To evaluate fractional electron leak (FEL) or electron transfer efficiency, mtH₂O₂ emissions were expressed relative to mitochondrial respiratory rates (23). PoWeR trained VL had lower maximal FEL and greater concentration of ADP needed to reduce maximal FEL by 50% compared to Sed with no differences between groups in gastroc or triceps (Fig. 4D-F).

Figure 4 –

Mitochondrial H₂O₂ emissions in the vastus lateralis (VL), Gastroc, and Triceps. (A) Maximal mtH₂O₂ emissions and (B) the suppression of mtH₂O₂ emissions by ADP used to calculate (C) the apparent IC₅₀ of ADP. (D) Maximal mtH₂O₂ emissions expressed per unit of oxygen (Fractional electron leak) and (E) the suppression of fraction electron leak curves to calculate (F) IC₅₀ of ADP reduce maximal fractional electron leak. N=9–10 per group. Points and error bars represent mean ± SEM. Groups analyzed by unpaired t-test (A, C, D, F). *P<0.05, **P<0.01

Next, we used the creatine kinase clamp technique to evaluate mitochondrial respiration by modeling the energetic demands during the transition between rest and exercise under physiological ADP concentrations. PoWeR trained mice had significantly greater mitochondrial respiratory rates throughout the entire assay and in all three muscles tested (Fig. 5A). We also identified that the conductance was greater in PoWeR trained versus Sed mice in the gastroc and triceps but not the VL (Fig. 5B).

Figure 5 –

Mitochondrial respiratory conductance in the vastus lateralis (VL), Gastroc, and Triceps. (A) Mitochondrial respiratory rates during creatine kinase clamp and (B) respiratory conductance. N=9–10 per group. Points and error bars represent mean ± SEM. Groups analyzed by multiple unpaired t-test (A) or unpaired t-test (B). *P<0.05, **P<0.01, ***P<0.001

To estimate mitochondrial protein content, we used an OXPHOS antibody that includes 1 subunit from each respective complex of the electron transport system. Muscle-type and complex specific differences emerged. Increased abundance of OXPHOS subunits were detected in the gastroc, quad, triceps, and plantaris (Fig. 6). The greatest increase in OXPHOS appeared in the triceps while we found no change in the FDL muscle and a decrease in complex IV for the soleus.

Figure 6 –

Western blot quantification of electron transport chain subunits across fore and hind limb whole muscle homogenates. All data is the respective PoWeR muscles expressed relative to the matched muscle from the sedentary group (dashed black line = 1). OXPHOS proteins were normalized to total protein measured via Ponceau S. N=9–10 per group. One Sed for the FDL was not included due to loss of sample quality during tissue homogenization. Data presented as mean with individual samples. Ponceau stain images are contrast adjusted for better visual clarity. Groups analyzed by unpaired t-test. *P<0.05, **P<0.01, ***P<0.001.

PoWeR increased myofiber size and myonuclear number in the FDL

Consistent with previous reports and our own muscle mass data, the lower leg muscles receive significant stimulus for growth from PoWeR (10,14). We validated these whole muscle mass gains through immunohistochemical approaches for fiber type specific CSA. We found type IIA myofiber CSA increased in the soleus (P=0.07) and plantaris with a shift towards a more oxidative profile that did not result in any mid-belly CSA changes in either PoWeR trained muscle (Supplementary Fig. S2 A-G).

Recent work has demonstrated that a weight pulling model of resistance exercise elicits hypertrophy of FDL (8,19). Consistent with our goal to expand characterization of skeletal muscle in response to PoWeR, we hypothesized the FDL may also receive considerable stimulus based on the biomechanics of weighted wheel running (24). The FDL in PoWeR trained mice had greater average CSA of type I (29%), IIA (32%), and IIX (19%) fibers with a trend in IIB (11%) (Fig. 7B). The FDL also undergoes a shift in fiber type distribution as evident by an increased proportion of IIA fibers and a reciprocal decrease in IIB fibers (Fig. 7C). Plotting myofiber histograms reveals an overall rightward shift to indicate a greater number of larger myofibers in all fiber types except type IIB (Fig 7D-G). The observed trend for increased IIB myofiber CSA is driven by the greater number of fibers within 800–1000 μm². To complete our characterization of the FDL, we next measured the number of myonuclei per myofiber number (Fig. 7H-J). Similar to previous reports in the soleus and plantaris following PoWeR and in the FDL following weight pulling (8,10,11,25), we detected an increase in the ratio of myonuclei per myofiber number. Our results suggest the effects of PoWeR on myofiber adaptions in the soleus and plantaris are largely repeatable and we identify the FDL as a hypertrophic muscle in response to PoWeR.

Figure 7 –

Adaptations to FDL myofibers after PoWeR. (A) Representative region of a mid-belly cross section subjected to immunohistochemistry for fiber type identification (Type I, IIA, IIB, and IIX) and Laminin. (B) Myofiber cross sectional area (CSA) and C) distribution. (D-G) Frequency histograms displaying number of fibers per cross sectional area within the muscle cross section. (H) Representative images of mid-belly cross sections stained for dystrophin (myofiber borders) and Hoechst (nuclei). Arrows point towards myonuclei (nuclei that are within the myofiber). Entire cross sections were analyzed to determine (I) number of total nuclei and (J) number of myonuclei to number of fibers. N=9–10 per group. Data presented as mean with individual samples or points and error bars representing mean ± SEM. Groups analyzed by multiple unpaired t-test. *P<0.05, **P<0.01, ***P<0.001.

Short-Term PoWeR increased skeletal muscle anabolic and metabolic intracellular signaling

After characterizing adaptations to 8-weeks of PoWeR, we sought to determine if short-term PoWeR would induce conventional intracellular signaling events associated with long-term adaptations (Fig 1B). During the early transition from light to dark phase, PoWeR mice demonstrated consistent running behavior for 4 hours before becoming more variable in the later dark phase (Fig. 8A). These data informed us to target the early dark phase transition to minimize the impact of running behavior variability on skeletal muscle intracellular signaling responses. While groups did not differ in bodyweight, we observed that exercising mice consumed less food during the 6-hr window (Fig. 8B & C). Mice accumulated 4.0 to 5.8 km while averaging 1.2–1.4 km each hour for the last 3 hours (Fig. 8D & E). These results suggest that the voluntary running behavior was relatively consistent prior to locking the wheels.

Figure 8 –

Characterization of short-term wheel running. (A) The hourly percentage of total distance run of a representative 24 hours (n=8). The grey box indicates the dark phase while the bracket indicates a period of the most consistent running behavior between all mice (B) Body weight and (C) food consumed during the 6 hours prior to euthanasia. (D) Average hourly running distance for the 6 hours prior to euthanasia. Grey box indicates dark phase. (E) Total distance ran the hours before tissue collection. N=8 per group. Data presented as mean with individual samples or points and error bars representing mean ± SEM Groups analyzed by unpaired t-test. ***P<0.001.

Using our results from the long-term PoWeR study, we focused our attention on the muscles demonstrating significant growth (soleus, plantaris, and FDL) and mitochondrial adaptions (gastroc, quad, triceps). We measured signaling events within common intracellular pathways implicated for growth, oxidative capacity, and metabolism including the mechanistic target of rapamycin (mTOR), mitogen activated protein kinases (MAPK), and AMP activated protein kinase (AMPK). Short-term PoWeR increased downstream substrates of mTOR complex 1 (mTORC1) signaling as evident by greater phosphorylation of P70S6K (Thr 389) and RPS6 (S235/S236) in the soleus, plantaris, and FDL (Fig. 9A). However, no change to downstream mTORC1 signaling was detected in the gastroc, quad, or triceps. Short-term PoWeR increased phosphorylation of MK2 (Thr 334) in the soleus but no detectable differences in the plantaris or FDL, nor in other MAPK signaling proteins, p38 (Thr180, Tyr182) and MKK3 (S189) (Fig. 9B). AMPKα (Thr182) phosphorylation was elevated in the plantaris and FDL but was decreased in the triceps (Fig. 9C), with no change in the soleus, gastroc, or quad. ACC (S79), one downstream target of AMPK signaling, was elevated in the soleus and quad, but decreased in the plantaris and FDL. Overall, these results suggest short-term PoWeR increases canonical exercise signaling pathways, but the responses are muscle and signaling pathway specific.

Figure 9 –

Anabolic and metabolic signaling following short term wheel running. (A) mTORC1 signaling determined by phosphorylated P70S6K (T389) and RPS6 (S235/S236) (B) MAPK signaling determined phosphorylated MAPKAP2 (MK2) (Thr 334), p38 (Thr 180, Tyr 182), and MKK3 (S189) (C) AMPK signaling determined by phosphorylated AMPKα (Thr172) and ACC (S79). (D). Respective images for all signaling proteins. All phosphorylated proteins were expressed relative to their respective total proteins. Data from short-term PoWeR mice are expressed relative to Sed indicated as dashed line equal to 1. N=6–8 per group. One Gastroc sample in the Sed was not included due to loss of sample quality during tissue homogenization. No gap in the well loading between treatment groups was included for the soleus MKK3 blots. Groups analyzed by unpaired t-test. *P<0.05, **P<0.01, ***P<0.001.

Discussion

The overarching goal of this study was to use a multi-muscle approach to determine if PoWeR can be used in mice to model select signaling events, mitochondrial adaptations, and myofiber hypertrophy implicated in whole-body functional improvements that are commonly observed after endurance and resistance exercise training in humans. To accomplish this goal, we first validated similar running behavior and soleus and plantaris hypertrophy in adult female mice compared to previous studies to suggest PoWeR can be consistently adopted across laboratories and vivariums (10). Further, we significantly expanded the assessments performed in seminal studies (10,11,14–17) to include both hypertrophic and oxidative adaptations in several skeletal muscles across the fore- and hind-limb and evaluated whole-body metabolism. We add to the existing knowledge base by demonstrating that PoWeR 1) enhanced physical function and glucose metabolism, 2) improved mitochondrial bioenergetics under physiological conditions in muscles from the upper and lower hind limb and fore-limb, 3) increased muscle mass, myofiber CSA, and myonuclear number in the FDL, and 4) activated conventional anabolic and metabolic signaling associated with the aforementioned long-term adaptions.

Physical and Metabolic Adaptions

The data collected herein support the use of PoWeR to model whole body and skeletal muscle adaptations that are commonly associated with endurance and resistance exercise training in humans. In addition to less adiposity, PoWeR trained mice had greater FFM likely due to both central and peripheral adaptations such as greater bone mineral density and liver, cardiac and skeletal muscle mass. 8 weeks of PoWeR increased hind limb skeletal muscle mass (5–12%), grip strength, and maximal exercise capacity which are consistent with the changes in humans after 8–12 weeks of aerobic and/or resistance exercise (26–29). PoWeR also improved glucose metabolism as evident by a greater ability to clear circulating glucose after an i.p. glucose challenge and suppression of gluconeogenesis during a pyruvate tolerance test, again in line with findings in humans that exercise can improve hepatic and peripheral insulin sensitivity (29–32). This is the first comprehensive assessment of PoWeR across the whole-body which demonstrated robust physical and metabolic adaptions consistent with both endurance and resistance exercise training.

Mitochondrial bioenergetics

Endurance exercise is an effective stimulus to increase mitochondrial volume and respiratory capacity in skeletal muscle. However, emerging evidence indicates that resistance exercise can also enhance mitochondrial respiratory capacity although this may be specific to model, muscle, or techniques to study mitochondria (19,33). Similarly, previous work using low volume, continuous resistance wheel running has reported conflicting results on estimates of mitochondrial volume and capacity depending on muscle selected and technique used (9,11,13,34). In the present study, to more thoroughly investigate mitochondrial adaptations compared to previous work, we employed high resolution flourorespirometry utilizing two protocols to directly evaluate mitochondrial bioenergetics under conditions more reflective of the in vivo environment in three different muscles located in the forelimb (triceps), upper hind limb (VL) and lower hindlimb (Gastroc) which provide abundant tissue that could be paired with future biochemical and molecular investigation. We first used an ADP titration protocol to evaluate mitochondrial ADP sensitivity, respiration and H₂O₂ emissions ranging from physiologically relevant ADP concentrations to maximally stimulated respiratory conditions. This technique has identified differences in mitochondrial bioenergetics that may be overlooked when providing a single saturating bolus of ADP (35–37). We next employed the creatine kinase clamp under ex vivo conditions to assess PoWeR induced changes in respiratory sensitivity and resistance during shifts in energetic conditions (21,38). To our knowledge this is the first application of the CK clamp to evaluate respiration and conductance across multiple muscles from healthy mice following an exercise regimen.

Consistent with the notion that acute and chronic endurance exercise decrease ADP sensitivity to enhance redox stress signaling for mitochondrial biogenesis and respiratory performance (39), PoWeR lowered mitochondrial ADP sensitivity, while increasing absolute mitochondrial H₂O₂ emissions, mitochondrial protein content and mitochondrial respiration. Although there were greater absolute levels of mitochondrial mtH₂O₂ emissions, PoWeR decreased the fraction of electron leak to mtH₂O₂ indicating increased electron transfer efficiency and potentially less mtH₂O₂ emissions per mitochondria. In line with these findings in adult female mice after PoWeR, 12-weeks of aerobic exercise in insulin resistant women decreased mtH₂O₂ emissions per mitochondrial protein content restoring mtH₂O₂ emitting potential toward that of lean, insulin sensitive counterparts while enhancing antioxidant content, mitochondrial respiratory efficiency, and glucose tolerance (40).

PoWeR trained mice had greater CI+II linked respiration throughout the range ΔG_ATP resembling the energetic demands during the transition from rest to strenuous exercise. Conductance can be determined by the change in mitochondrial respiratory rates relative to the change in ΔG_ATP to estimate the sensitivity of mitochondria to given substrates. A steeper slope in the relationship between respiration and ΔG_ATP after PoWeR in gastroc and triceps but surprisingly not the VL, indicates greater respiratory conductance and responsiveness to the prevailing mix of carbon fuel. Previous work has found greater respiratory rates and conductance in different tissues may be driven by heightened dehydrogenase enzyme kinetics to better equip the mitochondria to respond to rapid energy perturbations which would seem advantageous to perform greater work after exercise training (21,41).

Finally, the goal of our multi muscle approach was to provide a resource into which muscles may be most appropriate to study the molecular regulation of mitochondrial adaptations and remodeling to PoWeR in mice. The VL had the largest respiratory differences in PoWeR versus Sed while the triceps demonstrated the greatest change to OXPHOS protein content. Increased triceps OXPHOS protein content after PoWeR is consistent with other reports following unweighted wheel running (42,43). However, based on our experience, the fiber orientation of the triceps made mechanical permeabilization of muscle fibers more challenging and time consuming compared to the gastroc and VL.

Muscle Hypertrophy

Previously, low volume, continuous resistance wheel running and PoWeR increased myofiber size in hindlimb muscles particularly the soleus and plantaris (9–11,16,44–46). In the current study we confirmed that PoWeR increased soleus and plantaris muscle mass in female mice while also detecting increased type IIA myofiber size. Albeit, the magnitude of changes are relatively smaller in comparison to previous studies which could be the result of a shorter weighted wheel running protocol or lower running volume during the testing phases when wheel weight was near the greatest. While we also documented greater muscle mass in the TA and Quad, we were particularly interested in the FDL because the FDL is responsible for the toes gripping, plantar flexion, and toe off during walking and running. Further, during wheel running mice shift body position to bias lower hindlimb muscles compared to over ground running (24). Comparing the changes in FDL myofiber size following 8-weeks of PoWeR versus 13-weeks of a weight pulling (WP) model of resistance exercise (PoWeR vs. WP: type I (29% vs. 21%), Type IIA (32% vs. 27%), Type IIX (19% vs. 37%) and Type IIB (11% vs. 18%) it appears PoWeR may favor greater hypertrophic effects within more oxidative fiber types (Type I, IIA) while WP may have greater hypertrophic effects on highly glycolytic fibers (Type IIX, IIB) consistent with traditional endurance versus resistance exercise in humans (26,27,47). However, these studies were not performed in parallel, in the same sex, nor in the same laboratory so any direct comparisons should be taken with caution (8). We also observed a rightward shift for a greater number of larger CSA for most fiber types, but this was most apparent for Type I and IIA fibers. A peculiar observation we noted after PoWeR, is while nearly all fiber types increased average cross-sectional area, this did not translate to greater mid-belly CSA. We believe this is likely due to a transition from the largest Type IIX an IIB fibers toward an increased distribution of relatively smaller oxidative Type I and IIA fibers.

Short-Term PoWeR

In the lower hind-limb muscles (soleus, plantaris, FDL), short-term PoWeR activated the mTORC1 signaling pathway that contributes to muscle growth, consistent with the increased mass and myofiber size following long-term PoWeR. Interestingly, we also detected increased AMPK signaling in the plantaris and FDL but not the soleus. It is generally assumed mTORC1 and AMPK negatively regulate each other due to each pathway regulating anabolic and catabolic pathways, respectively (48). In the current study, we hypothesize that short-term PoWeR may acutely stimulate mTORC1 signaling preferentially in oxidative myofibers (type I and IIA) and AMPK phosphorylation in glycolytic fibers (IIB and IIX). This is supported by 1) the preferential growth of oxidative myofibers after PoWeR in the current study and 2) previous work that demonstrated acute running increased RPS6 (S235/S236), a downstream target of mTORC1 signaling, localized to oxidative fibers in the gastroc yet no change in mixed muscle lysate (49). Further, the substantial running volume with PoWeR generates significant energetic demand which may place a greater stress on muscles with a high prevalence of glycolytic fibers such as FDL and plantaris thereby increasing AMPK signaling, while more oxidative muscles like the soleus may be relatively more resistant to the energetic stress of PoWeR.

While the mTOR signaling pathway is regarded as a key driver of muscle growth, mTOR independent pathways are being elucidated (50). The p38 MAPK pathway is implicated in regulating oxidative capacity and muscle growth after endurance and resistance exercise paradigms (8,50,51). Here we aimed to investigate whether MAPK signaling was upregulated by PoWeR in the hind-limb muscles demonstrating increased mass and myofiber CSA. We explored up and downstream proteins of p38 but surprisingly only the downstream MK2 (Thr 334) was increased in the soleus. Electrically evoked maximal intensity contractions and intense exercise contractions induce activation of several MAPK signaling events suggesting either PoWeR may have been an inefficient stimulus or perhaps the 1 hour timepoint was inappropriate to capture the activation of the MAPK pathway (50,52). Future studies in MAPK signaling with PoWeR may consider alternative timepoints or a greater resistance to provide a larger stimulus while still maintaining running volume.

Limitations and Considerations

Voluntary running volume during PoWeR was consistent with previous reports (10), however, we identified a decline in running volume during the increased stress of physiological testing performed during the last 3-weeks of PoWeR. Therefore, future studies should consider minimizing the number of tests per week and/or include 1–2 weeks of uninterrupted training to maintain running volume to alleviate any impact on primary endpoints at euthanasia. Similarly, social isolation can contribute to increased running volume as well as influence animal health. While it is the outside of the scope of the current study, exploring a means to maintain the social group housing of mice while maintaining the ability to measure individual mouse running would be highly impactful to the field. A limitation of this investigation was solely using female mice of a single strain. Previous work has demonstrated male mice of similar age run similar volumes to female mice (8–10km/day) during PoWeR while achieving muscle mass gains and myofiber hypertrophy of select plantar flexor muscles (17) It would be important for future studies to not overgeneralize the current findings to male mice and/or other strains as there are likely sex and possibly strain specific mechanisms regulating skeletal muscle and whole-body adaptations to PoWeR.

The use of wheel running to study the molecular underpinnings of acute or short-term exercise has been less adopted primarily due to the inability to provide a fixed stimulus and the nocturnal behavior of rodents. We observed consistent running behavior over the first 4 hours during the early dark phase and any subtle variability in running distance was not associated with the change in the intracellular signaling pathways measured. Our approach to habituate mice to wheel running over 5 days before allocating to sedentary or PoWeR was to avoid the first bout phenomena to a unique stimulus which may confound interpretations linking acute responses to chronic adaptions. In many cases, the majority of molecular pathways initiated following the first bout of exercise in naïve/untrained states reflects stress responses along with adaptive responses (53–55). These stress pathways are dampened and refined following repeated bouts of exercise which allow us to reveal molecular regulation from the specificity of the exercise modality (53–61). Our approach with short-term PoWeR to measure intracellular signaling events following a bout of weighted wheel running offers several advantages including exercise alignment with circadian rhythms, lack of negative stress stimuli by forced exercise, and low investigator burden prior to euthanasia (62,63). Additionally, an important consideration in our approach is the provision of a standardized amount of food for animals to eat ad-libitum. The post-prandial conditions were to reflect the nutritional status of mice during long-term PoWeR. However, it is important to acknowledge the contribution of nutrition on these nutrient/energy sensing kinases and that prior exercise can enhance the sensitivity to nutrient signaling. While the PoWeR mice consumed less food versus Sed and demonstrated differences in several signaling events, our study cannot disentangle the exercise and nutrient induced signaling through our measured kinases.

Conclusion

The results of the current study significantly expand the characterization of PoWeR and further support the use of this paradigm to model the physical and metabolic adaptions observed after endurance and resistance exercise training in humans. Using high resolution flourorespirometry to evaluate mitochondrial respiration and H₂O₂ emissions that more closely resemble physiological conditions, PoWeR trained mice had improved mitochondrial bioenergetics across muscles from the fore- and hind-limbs with muscle specific adaptions that require deeper investigation. Furthermore, we demonstrate PoWeR increased muscle mass in multiple muscles and identified robust myofiber hypertrophy in the FDL. We also demonstrated short-term PoWeR as a translational model to detect changes in select intracellular signaling pathways presumed to contribute to long-term adaptations. These data can be used to provide an initial resource for others to understand the short-term responses and long-term adaptions across multiple muscles to exercise and supports the use of PoWeR to explore the ‘molecular transducers of exercise’ in a model that reflects hypertrophic and metabolic adaptations to voluntary endurance and resistance exercise in humans.

Supplementary Material

Supplemental Figs. S1-S2: doi.org/10.6084/m9.figshare.28319648.v1