Initiating aerobic exercise with low glycogen content reduces markers of myogenesis but not mTORC1 signaling

Twelve healthy, non-obese, recreationally active men between the ages of 18–39 years were enrolled to participate in this randomized, crossover study after providing informed, written consent. Individuals were excluded from the study if they were not in good health (metabolic or cardiovascular abnormalities, gastrointestinal disorders such as kidney disease, diabetes, and cardiovascular disease or were taking medications, such as stains, corticosteroids or diabetes medication that may affect macronutrient metabolism), refused to abstain from alcohol, nicotine, and dietary supplements during the study, had musculoskeletal injuries that compromised their ability to exercise, or donated blood within 8 weeks of beginning the study. Treatment order randomization was done using a random number generator. Participants in this study were part of a larger investigation that also aimed to assess the impact of glycogen content on microRNA expression and rates of exogenous glucose oxidation during steady-state aerobic exercise (Trial registration: Carbohydrate Availability and microRNA Expression, registered 15 August 2017, https://​clinicaltrials.​gov/​ct2/​show/​NCT03250234). Results from this separate objective are reported elsewhere [12]. Data are reported on 11 of the 12 enrolled participants because we were unable to collect a baseline muscle biopsy on one participant. This study was approved by the Institutional Review Board at the US Army Medical Research and Development Command (MRDC, Fort Detrick, MD) and data collection took place at the US Army Research Institute of Environmental Medicine (USARIEM, Natick, MA), between August 2017 to May 2018.

Pre-study participant characteristics, including height (Seritex, Inc., Carlstadt, NJ, USA), body mass (WB-110A, Tanita, Tokyo, Japan), and body composition (dual energy x-ray absorptiometry, DPX-IQ, GE Lunar Corporation, Madison, WI, USA), are reported in Table 1. Peak oxygen uptake (V̇O_2peak) was determined using a progressive-intensity cycle ergometer (Lode, BV, Netherlands) test and an indirect, open circuit respiratory system (True Max 2400, Parvomedics, Sandy, Utah, USA). Exercise intensities for protocol days were based on volunteers V̇O_2peak.

To normalize muscle glycogen between study arms, participants completed a glycogen depletion (Lode, BV, Netherlands) protocol 48-h prior to testing (Fig. 1), as previously described [12]. Participants completed 2-min of high-intensity cycling (work period) at 90% V̇O₂peak, followed by a 2-min recovery period cycling at 50% V̇O_2peak [2]. This work-to-recovery ratio was maintained until the participant was no longer able to complete 2-min of cycling at 90% V̇O_2peak. Cycling intensity during the work period was progressively lowered to 80%, 70%, and 60% V̇O_2peak when the participant was unable to complete 2-min of cycling at the given workload. Once the participant could not complete 2-min of cycling at 60% V̇O_2peak, exercise was stopped. The recovery period was maintained at 50% V̇O_2peak. Participants performed two practice sessions to ensure they were familiar with the protocol before testing.

Following glycogen depletion, dietary intake was controlled to normalize glycogen status before experimental days. All food and beverages (except water, which was allowed ad libitum) were provided during this glycogen normalization phase by study dietitians, who designed and prepared meals, derived from Meals, Ready-to-Eat (MRE; Ameriqual, Evansville, IN, USA) combat rations and commercially available food items. Participants returned all food and beverage wrappers and containers to dietitians to confirm intake. During the glycogen normalization phases, intakes were the same for LOW and AD; averaging 5.7 ± 0.6 g/kg/d carbohydrate, 1.2 ± 0.1 g/kg/d protein, and 1.0 ± 0.1 g/kg/d fat.

Following a 10 h overnight fast, a baseline (BL) muscle biopsy was taken from the vastus lateralis after glycogen normalization. Participants then completed the same glycogen depletion protocol as performed during the normalization phase. Glycogen depletion exercise time (LOW: 84 ± 25, AD: 88 ± 24 min) and intensity (mean power; LOW: 164 ± 26, AD: 161 ± 25 watts) were similar between treatments [12]. After glycogen depletion (GD), a second muscle biopsy was taken to confirm reductions in glycogen stores using an endpoint colorimetric assay (Cat# MAK016; Sigma-Aldrich, St. Louis, MO, USA) as reported elsewhere [12]. There was no difference in glycogen content at BL (LOW; 467 ± 95, AD; 472 ± 109 µmol/kg muscle dry weight) or after GD (LOW; 207 ± 99, AD; 210 ± 145 µmol/kg muscle dry weight) between LOW and AD [12]. Participants were then fed an isocaloric diet for the remainder of the day to elicit LOW (3081 ± 374 kcal/d, 1.5 ± 0.1 g/kg/d carbohydrate, 1.3 ± 0.5 g/kg/d protein, and 3.0 ± 0.5 g/kg/d fat) or AD (3086 ± 347 kcal/d, 6.0 ± 0.2 g/kg/d carbohydrate, 1.2 ± 0.5 g/kg/d protein, and 1.0 ± 0.5 g/kg/d fat) glycogen stores.

Participants returned to the laboratory the following day after a 10 h overnight fast. Resting metabolic rate (RMR) was measured to assess rested/fasted substrate oxidation before exercise (PRE) using an open-circuit indirect calorimetry (Parvo Medics). Participants rested in the supine position for ~ 30 min before expired air was collected using an acrylic hood. The test was discontinued when 20 min of steady-state V̇O₂ and V̇CO₂ were recorded. After the RMR testing protocol, a pre-exercise (PRE) muscle biopsy was taken. Participants then consumed 550 mL of the carbohydrate drink immediately before starting the exercise bout. Participants then began cycling for 80 min at their target V̇O₂ (LOW; 65 ± 4, AD; 64 ± 3% V̇O_2peak)_. Participants consumed 300 mL of the carbohydrate drink at 20, 40, and 60-min during exercise. Total carbohydrate ingested was 146 g (95 g glucose + 51 g fructose), consumed at an average ingestion rate of 1.8 g/min. The carbohydrate drink was prepared by the Combat Feeding Directorate (Natick, MA, USA) and contained corn-derived crystalline fructose (KRYSTAR® 30 0, Tate and Lyle Sugars, London, UK), maltodextrin (MALTRIN QD® M500, Grain Processing Corporation, Muscatine, IA, USA) and dextrose (CERELOSE®, Ingredion, Westchester, IL, USA). Nutrient content was confirmed before use (Eurofins Food Chemistry Testing Madison, Inc, Madison, WI, USA). During exercise, respiratory gas exchange (Parvo Medics) was measured at 0, 15, 30, 45, 60, and 75-min to assess substrate oxidation. A final biopsy was taken at the end of exercise (POST). Per study design [12], PRE and POST glycogen was lower in LOW than AD (Table 2). Following a minimum 7 day washout period, participants returned to the laboratory to complete the second arm of the study.

Resting carbohydrate and fat oxidation was calculated as [22]:

Exercise carbohydrate and fat oxidation were calculated [23]:

Substrate oxidation data during exercise were previously reported [12], but are briefly presented in Table 2 of this report to highlight differences in fuel use during exercise between the two treatments. Comparison of resting substrate oxidation data between LOW and AD has not been previously published.

Percutaneous muscle biopsies were conducted on the vastus lateralis using a 5-mm Bergstrom needle with manual suction while the participant was under local anesthesia (1% lidocaine). Muscle biopies were conducted immediately before and after the glycogen depletion protocol from a single incision made in a randomly selected leg. Muscle biopsies were also conducted before and after steady-state cylcing in the contralateral leg from a single incision. The average muscle sample weight was 110 mg. Immediately after being weighed, muscle samples were snap frozen in liquid nitrogen. At the conclusion of study muscle samples were cut under liquid nitrogen and aliquoted for assessment of muscle glycogen content, activity assays, intracellcular signaling, and mRNA expression for the current and our previously published manuscript [12].

Phosphorylation status and total protein content were determined using Western blotting. Muscle samples (~ 20 mg) were homogenized in ice-cold homogenization buffer (1:10 w/v) that contained 50 mM Tris–HCl (pH 7.5), 5 mM Na-pyrophosphate, 50 mM NaF, 1 mM.

EDTA, 1 mM EGTA, 10% glycerol (v/v), 1% Triton X-100, 1 mM DTT, 1 mM benz-amidine, 1 mM PMSF, 10 mg/mL trypsin inhibitor, and 2 mg/mL aprotinin. Samples were homogenized using a TissueLyser II with a 5-mm steel bead (Qiagen, Valencia, CA, USA). Homogenates were centrifuged for 15 min at 10,000×g at 4 °C. Supernatant (lysate) was collected and protein concentrations were determined using 660 nm Protein Assay (ThermoFisher, Waltham, MA, USA). Muscle lysates were solubilized in Laemmli buffer, with equal amounts of total protein loaded (15 µg) and separated by SDS-PAGE using precast Tris–HCl gels (Bio-Rad, Hercules, CA, USA). Proteins were transferred to polyvinylidene fluoride membranes and exposed to commercially available primary antibodies specific to p-AMPK^Thr72, p-AKT^Thr473, p-mTOR^Ser2448, p-p70S6K^Thr424/421, p-rpS6^Ser235/236, AMPK, AKT, mTOR, p70S6K, and rpS6 (Cell Signaling Technology, Danvers, MA) overnight at 4 °C. Secondary antibody (anti‐rabbit IgG conjugate with horseradish peroxidase; Cell Signaling Technology) and chemiluminescent reagent (Pierce Biotechnology, Rockford, IL) were applied to label primary antibodies. Blots were quantified using a phosphoimager (ChemiDoc XRS; Bio‐Rad) and Image Lab software (Bio‐Rad). Heat shock protein 90 (HSP90) was used to confirm equal amounts of protein were loaded per well. Phosphorylation status was normalized to its total protein. Data presented as fold change relative to BL phosphorylation for each group.

Total RNA was isolated in ~ 20 mg muscle samples using TRIzol reagent (Thermo Fisher). RNA quantity and quality were assessed using a Nanodrop ND-2000 spectrophotometer (Nanodrop, Wilmington, DE, USA). For mRNA analysis, equal amounts of total RNA (500 ng) were reverse-transcribed using the high-capacity cDNA reverse transcription (RT) kit (Applied Biosystems, Foster City, CA, USA). Reverse transcription was conducted in a T100™ Thermal Cycler (Bio-Rad). Amplifications were performed using a StepOnePlus Real-Time PCR System (Applied Biosystems). Samples were run in 10 µL reactions in duplicate using TaqMan® fast advanced master mix and commercially available TaqMan® probes (PAX7, MYOD, MYOGENIN; Applied Biosystems). All mRNA were normalized to B2M. Fold change for mRNA PRE and POST steady-state exercise were calculated using the ΔΔ cycle threshold (ΔΔCT) method [24] and expressed relative to individual BL values.

Independent of time, p-AMPK^Thr172 was higher (P < 0.05) and p-AKT^Thr473 was lower (P < 0.05) in LOW than AD (Fig. 2A, B).Regardless of treatment, p-AMPK^Thr172 was higher (P < 0.05) at PRE and POST compared to BL. Independent of treatment, POST p-mTOR^Ser2448, p-p70S6K^Thr424/421, and p-rpS6^Ser235/236 were higher (P < 0.05) than BL and PRE (Fig. 2C–E).

Independent of time, PAX7 was lower (P < 0.05) in LOW than AD (Fig. 3A). PRE MYOD was lower (P < 0.05) in LOW than AD (Fig. 3B). POST MYOD in LOW and AD were higher (P < 0.05) than PRE, with no difference between group. Independent of time, MYOGENIN was lower (P < 0.05) in LOW than AD (Fig. 3C).