The duration of glucocorticoid treatment alters the anabolic response to high-force muscle contractions

Abstract

Glucocorticoids induce a myopathy that includes loss of muscle mass and strength. Resistance exercise may reverse the muscle loss because it induces an anabolic response characterized by increases in muscle protein synthesis and potentially suppressing protein breakdown. Whether resistance exercise induces an anabolic response in glucocorticoid myopathic muscle is unknown, which is a problem because long-term glucocorticoid exposure alters the expression of genes that may prevent an anabolic response by limiting activation of pathways such as the mechanistic target of rapamycin in complex 1 (mTORC1). The purpose of this study was to assess whether high-force contractions initiate an anabolic response in glucocorticoid myopathic muscle. The anabolic response was analyzed by treating female mice with dexamethasone (DEX) for 7 days or 15 days. After treatment, the left tibialis anterior muscle of all mice was contracted via electrical stimulation of the sciatic nerve. Muscles were harvested 4 h after contractions. Rates of muscle protein synthesis were estimated using the SUnSET method. After 7 days of treatment, high-force contractions increased protein synthesis and mTORC1 signaling in both groups. After 15 days of treatment, high-force contractions activated mTORC1 signaling equally in both groups, but protein synthesis was only increased in control mice. The failure to increase protein synthesis may be because baseline synthetic rates were elevated in DEX-treated mice. The LC3 II/I ratio marker of autophagy was decreased by contractions regardless of treatment duration. These data show duration of glucocorticoid treatment alters the anabolic response to high-force contractions.

NEW & NOTEWORTHY Glucocorticoid myopathy is the most common, toxic, noninflammatory myopathy. Our work shows that high-force contractions increase protein synthesis in skeletal muscle following short-term glucocorticoid treatment. However, longer duration glucocorticoid treatment results in anabolic resistance to high-force contractions despite activation of the mechanistic target of rapamycin in complex 1 (mTORC1) signaling pathway. This work defines potential limits for high-force contractions to activate the processes that would restore lost muscle mass in glucocorticoid myopathic patients.

INTRODUCTION

Glucocorticoid myopathy is the most common, toxic, noninflammatory myopathy affecting ∼60% of individuals with elevated glucocorticoids (1, 2). Various populations are at risk of developing glucocorticoid myopathy including the elderly and those with age-related diseases (e.g., Alzheimer’s) as endogenous glucocorticoid production can increase up to 150% (2–4). Moreover, prescription glucocorticoids for conditions such as cancer and chronic inflammatory diseases also increases risk of developing myopathy (5, 6). Glucocorticoid myopathy is typically diagnosed by muscle atrophy and weakness (2, 7). The loss of muscle mass following short-term glucocorticoid treatment is due in part to a decreased rates of muscle protein synthesis and increased muscle protein breakdown (8). After longer-term treatment when the mass has more or less stabilized at a lower value, rates of muscle protein synthesis are no longer suppressed and may even be elevated above control levels despite continued exposure to glucocorticoids and persistent lower muscle mass (8).

Mechanical overload (e.g., resistance exercise) may be a viable therapy to treat muscle atrophy by minimizing the decrease in muscle protein synthesis following short-term treatment with glucocorticoids and/or increasing muscle protein synthesis following longer-term treatment. Moreover, resistance exercise may also limit the activation of protein degradation pathways induced by glucocorticoids, such as limiting an increase in autophagy. However, there are very limited data assessing whether the length of glucocorticoid treatment affects the ability for high-force contractions to increase muscle protein synthesis and decrease markers of protein degradation. There are at least two reasons it is important to understand the anabolic potential of glucocorticoid myopathic muscle if resistance exercise is going to be used to treat glucocorticoid myopathy. First, patients are generally unaware that they are vulnerable to glucocorticoid myopathy until they present in the clinic with lower muscle mass and muscle weakness (2, 7), meaning prescription of resistance exercise as a therapy may not occur until after myopathy has begun to or is fully developed. Second, prolonged glucocorticoid treatment can alter expression of genes such as Sestrin2, branched-chain amino acid transferase 2 (Bcat2), regulated in development and DNA damage 1 (Redd1), and dual specificity phosphatase 1 (Dusp1) that may limit activation of pathways such as the mechanistic target of rapamycin in complex 1 (mTORC1) and extracellular regulated kinase 1/2 (ERK1/2) that increase muscle anabolism and subsequent muscle growth in response to resistance exercise. Increased signaling through mTORC1 contributes to initiation of muscle protein synthesis by phosphorylating downstream substrates like 70 kD ribosomal protein S6 kinase 1 (p70S6K1) and Eukaryotic initiation factor 4E binding protein 1 (4EBP1; 9–18). Increased signaling through mTORC1 also relieves the inhibitory phosphorylation on Eukaryotic elongation factor 2 (eEF2) thereby promoting mRNA translation elongation (19). Although ERK1/2 can activate mTORC1 (20), ERK1/2 can also increase protein synthesis independent of mTORC1 by phosphorylating proteins like eIF4E (21). In addition, mTORC1 can also limit activation of autophagy-mediated protein degradation (22). Hence, glucocorticoids could limit the therapeutic potential of resistance exercise to treat glucocorticoid myopathy if the exercise fails to activate/suppress those processes.

Given those potential issues, the purpose of this study was to determine the extent to which high-force muscle contractions increase muscle protein synthesis in muscle following short-term and longer-term glucocorticoid treatment. A secondary purpose was to determine the extent to which high-force muscle contractions decrease the ratio of the lipidated to nonlipidated form of microtubule-associated protein 1 light chain 3 β (LC3B), a marker of global autophagy. We show that high-force muscle contractions increase muscle protein synthesis following 7 days of glucocorticoid treatment in mice, but 15 days of glucocorticoid treatment to mice results in anabolic resistance to high-force muscle contractions. Despite the differential response in protein synthesis, high-force muscle contractions decrease the LC3 II/I marker of autophagy regardless of glucocorticoid treatment duration. Overall, these data provide information about the utility of resistance exercise as a therapy to mitigate glucocorticoid myopathy.

METHODS

Animals

Female C57Bl/6J mice were purchased from Jackson Laboratories (Bar Harbor, ME) at 14 wk of age and housed in a temperature- (25°C) and light- (12 h/12 h light-dark) controlled environment within the vivarium at Florida State University. Mice were provided standard diet 5001 rodent chow (LabDiet, St. Louis, MO) and water ad libitum. All animals were acclimated to the vivarium for at least 10 days before beginning the experimental procedures described in this methods section. All glucocorticoid and control treatments began at 16 wk of age (physical maturity) to ensure that glucocorticoid treatments did not stop muscle growth but instead induced a loss of muscle mass, and females were chosen as they are more vulnerable to glucocorticoids than males (2). The Institutional Animal Care and Use Committee of Florida State University approved the animal facilities and all experimental procedures.

Experimental Design

Experiment 1: the responsiveness of muscle to a single bout of high-force contractions following short-term glucocorticoid treatment.

The design of the short-term experiment is shown in Fig. 1. Mice were singly housed, and baseline body composition was assessed via Echo MRI. Animals were then randomized into two groups of equal body mass (n = 13 or 14 mice/group) with one group randomized to receive daily intraperitoneal injections of veterinary grade dexamethasone (DEX: 1 mg/kg/day; NDC No. 13985-533-03, VetOne) diluted in saline (NDC No. 0409-4888-02, Hospira, Lake Forest, IL) to initiate myopathy whereas the other group received saline only as a control (8). All injections were administered within the first hour of the light cycle to maximize the negative effects of the glucocorticoid (23). Body composition was reassessed on day 6. On day 7, all mice received a final DEX or saline injection immediately before undergoing unilateral, high-force muscle contractions to produce a mechanical overload stimulus as previously described (24). The rationale for the injection immediately before the contractions was to assess the anabolic response to contractions in the presence of the glucocorticoid. We reasoned this approach would maximize any negative effects of DEX in the muscle tissue and therefore test the anabolic potential when conditions are least favorable.

Figure 1.

Experimental timeline for Experiment 1 that is a short-term experiment intended to correspond to the initial development of glucocorticoid myopathy. DEX, dexamethasone.

High-force eccentric contractions were induced in the tibialis anterior (TA) muscle via unilateral stimulation of the sciatic nerve under deep isoflurane anesthesia. Contractions were initiated via two stainless steel bipolar electrodes inserted subcutaneously near the sciatic nerve. A constant current stimulator (Aurora Scientific, Ontario, Canada) was used to activate the nerve at a frequency of 100 Hz using a ∼10 mAmp constant current. Each stimulus consisted of 300 pulses that were 1 ms in duration. The entire protocol consisted of 10 sets of 6 contractions. Each stimulus within a set was separated by a 10-s rest period, and all sets were separated by a 60-s rest period. Contractions were performed between 0800 and 1100 h (beginning of the light cycle) to mimic exercise in the early evening for humans and to allow mice ample time to consume most of their daily food during the previous dark cycle so that prolonged fasting did not interfere with outcome measures.

After the contractions, all mice received a subcutaneous injection of warm saline (500 µL) and were returned to their cages with free access to water but not food until euthanasia 4 h later. Thirty minutes before tissue harvest, all mice received an intraperitoneal injection (0.04 µmol/g body mass) of puromycin (Cat. No. P-1033, AG Scientific, San Diego, CA) diluted in PBS to estimate muscle protein synthetic rates via the SUnSET method (25). At euthanasia, the TA muscles were rapidly harvested, snap-frozen in liquid nitrogen, and stored at −80°C until analysis. Muscle harvest occurred under deep isoflurane anesthesia and all mice were euthanized via cardiectomy. Tibias were removed, and tibia length was assessed with calipers. We focused on the 4 h postcontractions time point because glucocorticoids transcribe several genes that suppress mTORC1 signaling, mTORC1 signaling peaks at this postcontractions time point, and the increase in muscle protein synthesis at this time point is largely dependent on the rapamycin-sensitive portion of mTORC1 signaling (26–32).

Experiment 2: the responsiveness of muscle to a single bout of high-force contractions following a longer-term, clinically relevant glucocorticoid dosing schedule.

The design of the longer-term experiment is shown in Fig. 2. Mice were singly housed, and then on day 0, baseline forelimb grip strength was established followed by assessment of body composition via Echo MRI. Mice were then randomized into two groups of equal body mass (n = 11 mice/group). One group continued consuming regular drinking water as a control, whereas the other started consuming drinking water containing water-soluble DEX (CAS No. 50-02-2, Sigma-Aldrich; St. Louis, MO). The dose of DEX started at ∼1.5 mg/kg/day (a moderately high therapeutic dose in humans) before gradually tapering across the 15-day treatment to a final dose of ∼0.2 mg/kg/day. The taper in dose was designed to mimic a clinical dosing regimen while still maintaining a final dose that is sufficient to induce glucocorticoid myopathy (28). Body mass, food intake, and water intake were monitored daily. Water + DEX was replenished every other day and adjusted to meet the desired dose. On day 14, final forelimb grip strength was tested at the beginning of the dark cycle. Final body composition was also assessed on day 14 after grip strength testing.

Figure 2.

Experimental timeline for Experiment 2 that is a long-term experiment intended to correspond with a more severe stage of glucocorticoid myopathy. DEX, dexamethasone.

Mice were subjected to the muscle contractions protocol described for Experiment 1 at the beginning of the light cycle on day 16. After contractions, all mice were returned to their cages with free access to regular drinking water (without DEX) but not food until euthanasia 4 h later. Thirty minutes before euthanasia, all mice received an intraperitoneal injection (0.04 µmol/g body mass) of puromycin diluted in PBS to estimate muscle protein synthetic rates. At euthanasia, the TA muscles were rapidly harvested, snap-frozen in liquid nitrogen, and stored at −80°C until analysis. Muscle harvest occurred under deep isoflurane anesthesia and all mice were euthanized via cardiectomy. Tibias were removed, and tibia length was assessed with calipers. Administering the DEX by water was chosen to minimize stress from daily intraperitoneal injections and to administer the hormone orally as done in humans.

Forelimb Grip Strength Testing

Forelimb grip strength was assessed on a Chatillon force gauge (Ametek Inc., Largo, FL; 33). Animals were allowed to grab onto the bar with both forelimbs and then mice were pulled away from the bar until they released. The mean peak force (Newtons, N) generated from three successful trials was used for the analysis. A trial was deemed successful if the mouse used both forelimbs with a pronated grip to generate the force. Acclimation to the device occurred across the 3 days before baseline testing and again before postintervention testing. The same researcher acclimated and performed grip strength testing on all animals. The mean coefficient of variation for the three trials was 0.12 ± 0.10 N for control animals and 0.14 ± 0.10 N for DEX-treated animals.

Western Blot Analysis

Western blot analysis was performed as previously described (24). Protein was extracted from TA muscle samples via glass-on-glass homogenization in 10 volumes of buffer (10 µL/mg muscle) consisting of 50 mM HEPES (pH 7.4), 0.1% Triton X-100, 4 mM EGTA, 10 mM EDTA, 50 mM Na₄P₂O₇, 100 mM β-glycerophosphate, 25 mM NaF, 5 mM Na₃VO₄, and 10 µL/mL of protease inhibitor cocktail (Cat. No. P-8340; Sigma-Aldrich). Muscle protein extract was centrifuged for 10 min at 10,000 g at 4°C. The soluble supernatant fraction was quantified via the Bradford method. Once quantified, all soluble proteins were diluted in 2× Laemmli buffer. Approximately 20–60 µg of protein was fractionated on 4%–20% Bio-Rad Tris-Glycine Criterion precast gels (Hercules, CA) and transferred to polyvinylidene difluoride (PVDF) membranes. Ponceau-S staining was used to determine the efficacy of transfer and ensure equal protein loading. PVDF membranes were blocked in 5%-nonfat dried milk in Tris-buffered saline + 0.1% Tween 20 (TBST), and membranes were incubated with primary antibodies diluted in TBST overnight at 4°C. Antibodies against REDD1 (1:500 dilution, Cat. No. 10638-1-AP), SESTRIN2 (1:1,000 dilution, Cat. No. 10795-1-AP), and BCAT2 (1:1,000 dilution, Cat. No. 16417-1-AP) were obtained from Proteintech (Rosemont, IL). Antibodies against p70S6K1 Thr389 (Cat. No. 9205), 4EBP1 Ser65 (Cat. No. 9451), ERK 1/2 Thr202/Tyr204 (Cat. No. 9101), total ERK 1/2 (Cat. No. 9102), eEF2 Thr56 (Cat. No. 2331), total eEF2 (Cat. No. 2332), and LC3B (Cat. No. 43566) were obtained from Cell Signaling Technology (Danvers, MA) and diluted 1:1,000 for use. Antibodies against total p70S6K1 and total 4EBP1 were custom made by Bethyl Laboratories (Montgomery, TX), were generously provided by Dr. Scot Kimball (Pennsylvania State University College of Medicine), and were diluted 1:10,000 for use. Antibodies against puromycin (Cat. No. MABE343) were obtained from Millipore Sigma (Burlington, MA) and were diluted 1:1,000 for use. Following incubation with secondary antibodies (Cat. No. A-120-101P or A-90-116P, Bethyl Laboratories both diluted 1:10,000), antigen-antibodies complexes were visualized via chemiluminescence using Clarity reagent (Cat. No. 1705061; Bio-Rad) on a Bio-Rad ChemiDoc Touch imaging system. All images were obtained in less than 10 min of exposure.

Puromycin and the Ponceau-S stain were analyzed using ImageLab software (Bio-Rad) because the quantification boxes can be adjusted to follow the curvature of the lanes. The pixel density of all other protein targets was determined using ImageJ software (National Institutes of Health, Bethesda, MD). The pixel density of each target was quantified using the same software. The pixel density for total protein blots (e.g., puromycin) was quantified as the ratio of total protein to the 25–75 kDa section of the Ponceau-S stain (24). The relative phosphorylation of a protein was assessed by the ratio of the phosphorylated protein to the corresponding total protein.

All antibodies used in this study have been validated by our laboratory or by others for their response to rapamycin (mTORC1 substrates), response to mechanical overload (ERK1/2), response to fasting/refeeding (LC3B), recognition of over-expressed protein (SESTRIN2), or lack of signal in knockout/knockdown animals/cells (SESTRIN1, BCAT2, and REDD1; 34–39). Although our IgG secondary antibody was raised in a mouse, preliminary experiments showed that there was little to no signal from the endogenous IgG heavy and light chain within the TA muscle extracts of our mice at the exposure time needed to quantify puromycin labeling. Therefore, we did not exclude the areas of the blot containing heavy and light chain bands from the puromycin quantification.

Experiments 1 and 2 consisted of multiple replicates to generate the overall data. Each replicate contained control mice and DEX-treated mice. The quantification of a specific target within each replicate consisted of normalizing each target to the noncontracted control muscle of the control mice. Once all experimental replicates were collected, the data from each replicate were averaged together to generate the collective data for that overarching experiment.

Statistics

Data are reported as individual data points superimposed on means ± SD or individual data points connected with a line to illustrate changes in a target within the same animal (i.e., noncontracted muscle vs. contracted muscle). Two-way ANOVA with repeated measures was used to evaluate pre- to posttreatment variables, whereas two-way ANOVA was used to assess variables using DEX and contractions as the two factors. Sidak’s multiple comparison test was used post hoc if a significant interaction was observed. Otherwise, only main effects are reported. Student’s t test was used to evaluate difference between two groups. No main effects were noted for puromycin incorporation in Experiment 2, which is likely due to a lack of power to detect differences by two-way ANOVA as Experiments 1 and 2 were powered to detect changes in puromycin between the noncontracted and contracted muscle in control mice using paired Student’s t test. The results for puromycin by paired Student’s t tests are shown for Experiment 2. Because there was a visual appearance of higher puromycin incorporation into the noncontracted muscle of DEX-treated animals compared with the noncontracted muscle of control animals in Experiment 2, puromycin labeling was compared in the noncontracted muscles of control mice and DEX-treated mice using a Mann–Whitney test to estimate basal protein synthetic rates across groups. All analyses were performed using GraphPad Prism Software (La Jolla, CA). Significance was set at P ≤ 0.05 for all analyses.

RESULTS

Experiment 1: High-Force Contractions Increase Muscle Protein Synthesis Following Short-Term Glucocorticoid Treatment

Mean daily food and water intakes for the short-term experiment are shown in Table 1. There were no differences in daily food intake between groups, whereas the DEX-treated mice consumed more water each day (P = 019). The DEX-treated mice received a total of 0.17 ± 0.01 mg of DEX throughout the experiment. In control mice, lean mass and fat mass did not change across the treatment period (Table 1). Conversely, body mass in DEX-treated mice decreased by ∼4% across the treatment period (Fig. 3A, Tables 1 and 3, P = 0.0004). The loss of body mass was largely due to ∼7% decrease in lean mass in DEX-treated animals (Fig. 3B, Table 1, P < 0.0001) as fat mass increased across the treatment period in DEX-treated animals (Fig. 3C, Table 1, P = 0.002). TA mass was modestly, but significantly, lower in DEX-treated mice compared with control values (Fig. 3D, P = 0.007). Gastrocnemius and plantaris masses were also lower in DEX-treated mice compared with control values without an effect on soleus or heart masses (Table 1, P = 0.0013, 0.03, 0.11, and 0.053, respectively). The lack of changes in the soleus is likely because the glucocorticoid receptor is expressed at lower levels in the soleus muscle compared with the gastrocnemius or TA (30), and lack of changes to the heart is likely because glucocorticoids preferentially target Type II fibers (40). Consistent with glucocorticoids inducing apoptosis of splenocytes (8), spleen mass was lower in the DEX-treated mice compared with control values (Fig. 3E, P < 0.0001). The lower muscle and spleen masses observed in DEX-treated mice were not due to differences in body size, as tibia lengths were similar between groups (Table 1, P = 0.09).

Table 1.

Morphological data, food/water intake, and tissue mass for Experiment 1

	Initial Measures		Final Measures
	CON	DEX	CON	DEX	P Value
Body mass, g	22.5 ± 1.5	22.2 ± 1.5	22.2 ± 1.3	21.2 ± 1.3$	0.0004
Lean mass, g	18.2 ± 1.0	18.2 ± 0.9	18.3 ± 1.0	17.0 ± 1.0$	<0.0001
Fat mass, g	2.2 ± 0.9	1.8 ± 0.5	2.0 ± 0.80	2.3 ± 0.6$	0.0020
Gastrocnemius, mg			103.3 ± 8.3	94.5 ± 4.9†	0.0013
Plantaris, mg			13.9 ± 1.2	12.9 ± 1.3†	0.0266
Soleus, mg			7.1 ± 0.95	6.6 ± 0.9	0.11
Heart, mg			102.2 ± 6.2	98.1 ± 6.6	0.053
Spleen, mg			72.1 ± 12.9	43.0 ± 4.1†	<0.0001
Tibia, mm			17.3 ± 0.5	17.4 ± 0.4	0.09
Daily food intake, g			4.18 ± 1.6	3.94 ± 0.44	0.60
Daily water intake, mL			4.81 ± 1.08	6.07 ± 0.77†	0.0019

Values are means ± SD. $Significantly different than initial value by Sidak’s multiple comparison post hoc test. †Significantly different than saline-treated control by unpaired Student’s t test. CON, control; DEX, dexamethasone.

Figure 3.

Phenotype following short-term glucocorticoid treatment. Percent changes from baseline in body mass (A), lean mass (B), and fat mass (C). Lean and fat mass determined by EchoMRI. Tibialis anterior (TA; D) mass and spleen mass (E). Data are individual data points superimposed on means ± SD. n = 13 or 14 female mice/group generated from five experimental replicates. *Significantly different at P ≤ 0.05 by unpaired Student’s t test. CON, control; DEX, dexamethasone.

Table 3.

Comparisons of morphological data and tissue masses of DEX-treated animals between Experiments 1 and 2

	% Change from Baseline (DEX Only)			% Difference of DEX vs. CON
	Experiment 1	Experiment 2	P Value	Experiment 1	Experiment 2	P Value
Body mass, g	−4.4 ± 3.4%	−7.6 ± 4.2%††	0.026
Lean mass, g	−6.6 ± 2.4%	−9.4 ± 2.9%††	0.02
Fat mass, g	32.1 ± 30.1%	0.9 ± 23.0%††	0.005
Tibialis anterior, mg				−9.0 ± 6.2%	−14.2 ± 12.5%	0.096
Gastrocnemius, mg				−9.7 ± 4.7%	−19.3 ± 6.0%††	0.0001
Plantaris, mg				−8.4 ± 9.4%	−11.1 ± 9.5%	0.25
Soleus, mg				−6.1 ± 12.2%	1.5 + 11.6%	0.07
Heart, mg				−4.7 ± 6.4%	−0.7 ± 11.4%	0.15
Spleen, mg				−42.9 ± 5.5%	−36.2 ± 8.0%††	0.012

Values are means ± SD. ††Significantly different than Experiment 1 by unpaired Student’s t test. CON, control; DEX, dexamethasone.

There was a main effect of contractions to increase puromycin incorporation into the TA muscle (P = 0.0099) and a main effect of DEX to decrease puromycin incorporation (P < 0.0001; Fig. 4, A and H). The increased puromycin incorporation in response to contractions coincided with main effects of contractions to increase phosphorylation of the direct mTORC1 substrates p70S6K1 (Thr389) and 4EBP1 (Ser65) in both control and DEX-treated mice (Fig. 4, B, C, and H, P < 0.0001). Similarly, there was a main effect of contractions to decrease eEF2 phosphorylation in both control and DEX mice compared with the noncontracted muscle (Fig. 4, D and H, P = 0.002). The only mTORC1 substrate that was affected by DEX was 4EBP1 (Ser65), as there was a main effect of DEX to lower this measure in DEX-treated mice (Fig. 4C, P = 0.011). Although REDD1, BCAT2, and SESTRIN2 are glucocorticoid receptor target genes that suppress mTORC1 signaling (28–30), only REDD1 protein levels were modestly increased in the DEX-treated mice (Fig. 4, E–H, P = 0.0049) It is possible that the final DEX injection caused the increase in REDD1 and the modest decrease in 4E-BP1 (Ser65) phosphorylation. However, assessing the anabolic response in the presence of DEX ensures we made our measures under the most undesirable conditions. Therefore, the ability to induce an anabolic response under those conditions is favorable.

Figure 4.

High-force muscle contractions increase muscle protein synthesis following short-term glucocorticoid treatment. A: puromycin incorporation into tibialis anterior muscles of control animals and dexamethasone-treated animals was determined by Western blot analysis. The phosphorylated to total protein ratio for p70S6K1 (Thr389; B), 4EBP-1 (S65; C), and eEF2 (Thr56; D) was determined by Western blot analysis. The total protein content of REDD1 (E), BCAT2 (F), and SESTRIN2 (G) was determined by Western blot analysis. H: representative Western blots. Results are individual data points superimposed on the means with lines connecting paired points. n = 13 or 14 female mice/group generated from five experimental replicates. Within each replicate, Western blots were normalized to the noncontracted muscle of control animals, with the mean of all replicates used in the final analysis. Two-way ANOVA was used to assess differences in variables. Main effects from the two-way ANOVA are shown on the graph with significance at P ≤ 0.05. CONT, contraction; DEX, dexamethasone.

Experiment 2: Anabolic Resistance Develops following Longer-Term Glucocorticoid Treatment Despite No Impairment in mTORC1 or ERK1/2 Signaling

Mean daily food and water intakes for the longer-term experiment are shown in Table 2. Daily food and daily water intake were both lower in the DEX-treated mice (P = 0.0013 and P < 0.0001, respectively). Body mass and lean mass were not affected across the treatment period in control animals whereas fat mass increased (Table 2, P = 0.0078). In DEX-treated mice, the daily dose of DEX started at ∼1.5 mg/kg/day with the dose gradually tapering across the treatment period to a final dose of ∼0.2 mg/kg/day (Fig. 5A). The DEX-treated mice received a total of 0.25 ± 0.07 mg of DEX throughout the experiment. This treatment paradigm decreased body mass and lean mass in DEX-treated animals (Fig. 5, B and C, Table 2, P < 0.0001). Fat mass did not change across the treatment period in DEX-treated mice (Fig. 5D and Table 2). Grip strength was lower in DEX-treated mice compared with control mice after the treatment period (Fig. 5, E and F, P < 0.05), showing development of muscle weakness. Mass of the TA (Fig. 5G, P = 0.007), gastrocnemius, and plantaris muscles were also lower in the DEX-treated mice compared with control mice (Table 2, P < 0.0001 and P = 0.002). Soleus mass and heart mass were not affected by DEX treatment (Table 2, P = 0.32 and 0.48), whereas DEX treatment once again resulted in significantly lower spleen mass (Fig. 5H, P < 0.0001). There were no differences in body size as tibia lengths were similar across groups (Table 2, P = 0.57). The 15-day DEX treatment in Experiment 2 resulted in a larger percent decrease in body mass, lean mass, and gastrocnemius mass compared with the shorter 7-day treatment in Experiment 1 (Table 3, P = 0.026, 0.02, and 0.0001). Although the percent change in the TA mass and plantaris mass were also greater after the 15-day treatment compared with the 7-day treatment, the differences did not achieve significance (Table 3).

Table 2.

Morphological data, food/water intake, and tissue mass for Experiment 2

	Initial Measures		Final Measures
	CON	DEX	CON	DEX	P Value
Body mass, g	23.2 ± 1.6	23.3 ± 1.7	23.6 ± 2.1	21.3 ± 1.9$	<0.0001
Lean mass, g	18.5 ± 1.0	18.2 ± 0.9	18.5 ± 1.1	16.5 ± 0.7$	<0.0001
Fat mass, g	2.4 ± 0.9	2.8 ± 1.0	3.2 ± 1.5$	2.8 ± 0.9	0.0078
Gastrocnemius, mg			104.3 ± 6.9	84.3 ± 6.5†	<0.0001
Plantaris, mg			14.6 ± 1.1	12.7 ± 1.3†	0.0020
Soleus, mg			7.1 ± 1.1	7.3 ± 0.8	0.32
Heart, mg			102.9 ± 8.0	102.4 ± 12.4	0.46
Spleen, mg			70.6 ± 9.4	45.1 ± 5.7	<0.0001
Tibia, mm			17.2 ± 0.6	17.6 ± 0.3	0.57
Daily food intake, g			5.1 ± 0.9	3.80 ± 0.7†	0.0013
Daily water intake, mL			8.05 ± 1.04	5.17 ± 1.19†	<0.0001

Values are means ± SD. $Significantly different than initial value by Sidak’s multiple comparison post hoc test. †Significantly different than saline-treated control by unpaired Student’s t test. CON, control; DEX, dexamethasone.

Figure 5.

Phenotype following a longer-term, clinically relevant dexamethasone treatment regimen. Mice were either given regular drinking water as a control or dexamethasone dissolved in the drinking water for 2 wk. A: dose of dexamethasone over the experimental period. The pre- to posttreatment percent change in body mass (B), lean mass (C), and fat mass (D). Lean and fat masses were determined by EchoMRI. Final grip strength (E) and percent change in grip strength (F) pre- to posttreatment. Tibialis anterior (TA; G) mass and spleen mass (H). Results are individual data points superimposed on means ± SD. n = 11 female mice/group generated from four experimental replicates. *Significantly different at P ≤ 0.05 by unpaired Student’s t test. CON, control; DEX, dexamethasone.

Unlike Experiment 1, there were no main effects of contraction or DEX for puromycin incorporation (Fig. 6, A and F), likely due to a lack of statistical power to detect main effects or an interaction between groups. However, a paired Student’s t test showed puromycin incorporation was higher in the contracted TA of control mice (P = 0.041) but not DEX-treated mice (P = 0.35, Fig. 6, A and F). Despite these observations for puromycin, there was a main effect of contractions to increase phosphorylation of p70S6K1 (Thr389) and 4EBP1 (Ser65) in both control and DEX-treated mice with no main effect of DEX noted (Fig. 6, B, C, and F, P = 0.0044 and 0.0004). The ability for contractions to increase 4EBP1 phosphorylation to a similar magnitude in DEX-treated animals as in control animals occurred despite total 4EBP1 protein levels being ∼40% higher in the TA muscles of DEX-treated animals (Fig. 6F, P < 0.0001). There was also a main effect of contraction to decrease phosphorylation of eEF2 in both groups with no effect of DEX observed (Fig. 6, D and F, P = 0.0003). The inability of contractions to increase puromycin incorporation in DEX-treated mice was not because phosphorylation of ERK1/2 (Thr202/Tyr204) did not increase as there was a main effect of contraction to increase ERK1/2 phosphorylation in both DEX-treated and control groups (Fig. 6, E and F, P = 0.0009). Although we realize the SUnSET method does not account for differences in the puromycin precursor pool, thereby making synthetic assessments across animals potentially problematic, we did find that puromycin incorporation into the noncontracted TA muscle of DEX-treated mice was ∼25% higher than in the noncontracted TA of control mice by Mann–Whitney test (Fig. 6, A and F, P = 0.042), suggesting basal muscle protein synthetic rates may have been elevated in DEX-treated mice. Assessing differences in puromycin incorporation across animals should be considered preliminary as the SUnSET method does not account for differences in the puromycin precursor pool across animals.

Figure 6.

Longer-term glucocorticoid treatment leads to anabolic resistance but does not inhibit the contraction-mediated activation of mTORC1 signaling or ERK1/2 phosphorylation. A: puromycin incorporation into tibialis anterior muscles of control animals and dexamethasone-treated animals was determined by Western blot analysis. The phosphorylated to total protein ratio for p70S6K1 (Thr389; B), 4EBP-1 (S65; C), eEF2 (Thr56; D), and ERK1/2 (Thr202/Tyr204; E) was determined by Western blot analysis. F: representative Western blots. Results are individual data points superimposed on the means with lines connecting paired points. n = 11 female mice/group generated from four experimental replicates. Within each replicate, Western blots were normalized to the noncontracted muscle of control animals, with the mean of all replicates used in the final analysis. Two-way ANOVA was used to assess differences in variables. Main effects from the two-way ANOVA are shown on the graph with significance at P ≤ 0.05. #Significantly different than the noncontracted muscle of control mice by paired Student’s t test. &Significantly different than the noncontracted muscle of control mice by Mann–Whitney test. CONT, contraction; DEX, dexamethasone.

High-Force Contractions Decrease the LC3 II/I Marker of Autophagy Regardless of Glucocorticoid Treatment Duration

Although we and others have shown that acute exposure to glucocorticoids increases the LC3 II/I marker of autophagy, which is indicative of enhanced activity of this degradation pathway, there were no main effects of DEX to increase the LC3 II/I ratio after 7 days (Fig. 7, A and C), and there was a main effect of DEX to decrease the ratio after 15 days of glucocorticoid treatment (P = 0.0113, Figure 7, B and C). High-force contractions decreased the LC3 II/I ratio in both control and DEX-treated mice after 7 days (P < 0.0001) or 15 days (P < 0.0001; Fig. 7, A–C).

Figure 7.

High-force muscle contractions decrease the LC3 II/I marker of autophagy regardless of glucocorticoid treatment duration. The LC3 II/I ratio in the tibialis anterior following 7 days of treatment (A) and 15 days of treatment (B) was determined by Western blot analysis. C: representative Western blots. Results are individual data points superimposed on the means with lines connecting paired points. n = 11–14 female mice/group generated from four to five experimental replicates. Within each replicate, Western blots were normalized to the noncontracted muscle of control animals, with the mean of all replicates used in the final analysis. Two-way ANOVA was used to assess differences in variables. Main effects from the two-way ANOVA are shown on the graph with significance at P ≤ 0.05. CONT, contraction; DEX, dexamethasone.

DISCUSSION

Despite glucocorticoid myopathy being the most common, toxic, noninflammatory myopathy (1), very little is known about the response of glucocorticoid myopathic muscle to resistance exercise. We demonstrate that the potential of skeletal muscle to increase muscle protein synthesis in response to high-force contractions is intact following short-term glucocorticoid treatment, and this coincides with the unimpaired ability to activate mTORC1. However, as the treatment duration progresses, the potential of skeletal muscle to increase muscle protein synthesis in response to high-force contractions is abolished even though the putative signals that contribute to the increase in protein synthesis (e.g., mTORC1) respond normally to the contractions. The LC3 II/I ratio decreased after high-force contractions regardless of treatment duration, meaning contractions may limit a negative shift in protein balance for both the short- and longer-term DEX treatment protocols. Overall, these data show the length of glucocorticoid treatment alters the ability of high-force contractions to induce an anabolic response therefore potentially altering the ability of resistance exercise to treat the atrophy in glucocorticoid myopathic muscle.

High-force contractions fully activated mTORC1 following both short-term and longer-term glucocorticoid treatment even though glucocorticoids have been shown to suppress mTORC1 activity. Full activation of mTORC1 is due in large part to signals that localize mTORC1 to the lysosomal/late endosomal membrane and signals that directly activate mTORC1 on membrane localization (41, 42). Although glucocorticoid-mediated induction of SESTIRN2 and BCAT2 can reduce mTORC1 localization at the lysosomal/late endosomal membrane by disrupting the Gap Activity Toward RAGS (GATOR)-RAG GTPase pathway (15, 29, 41, 43, 44), it is unlikely that glucocorticoid treatment affected mTORC1 localization via BCAT2 or SESTRIN2 as levels of both proteins were not increased in the muscle after 7 days of DEX treatment. Alternatively, glucocorticoid-mediated induction of REDD1 can reduce the GTP loading of RHEB (Ras Homolog Enriched in Brain) thereby limiting activation of any lysosomal/late endosomal localized mTORC1 (11), and REDD1 protein levels were higher in the muscle of DEX-treated mice at the 7-day time point. Mechanical activation of mTORC1 occurs largely via production of phosphatidic acid (PA) and activation of RHEB (27, 45, 46), which are downstream of REDD1. This means the sustained ability for mechanical overload to increase mTORC1 signaling despite increased REDD1 was likely because production of PA and activation of RHEB remained intact. The ability for mechanical overload to activate mTORC1 in the face of increased REDD1 levels is consistent with our previous work showing that REDD1 does not stop the mechanical activation of mTORC1, but rather, REDD1 lowers the baseline from where the induction begins (34, 47). It is unlikely that REDD1, SESTRIN2, or BCAT2 are increased following longer-term glucocorticoid treatment (i.e., 15 days) as neither the baseline nor the absolute peak induction of mTORC1 signaling was affected by DEX treatment, meaning the glucocorticoid induction of those mTORC1-inhibitory genes may diminish over time.

Regardless of the mechanism by which high-force contractions activate mTORC1 in glucocorticoid myopathic muscle, protein synthesis was not increased in mice subjected to longer-term glucocorticoid treatment. Our data suggest that this anabolic resistance following long-term glucocorticoid treatment may be because basal rates of muscle protein synthesis are elevated and could not be further increased despite contraction-mediated activation of mTORC1 and ERK1/2. Therefore, those putative signals that initiate protein synthesis are likely relevant up to a certain synthetic rate with additional signals having no ability to further increase rates. Although a single bout of high-force contractions may not immediately increase synthetic rates in muscle subjected to long-term glucocorticoid treatment, continued activation of mTORC1 in response to repeated bouts of contractions may eventually overcome the anabolic resistance by increasing muscle translational capacity by changing the types of mRNAs that are translated. For instance, mTORC1 promotes translation of mRNAs with long 5′ untranslated regions, specifically mRNAs encoding ribosomal proteins that contain 5′ terminal oligopyrimidine tracts immediately downstream of the N⁷-methyl cap (TOP mRNAs; 48–51). By increasing the translation of TOP mRNAs that encode protein synthetic machinery (e.g., ribosomal genes), the muscle may ultimately increase translational capacity (i.e., ribosomal quantity), which would in turn increase the overall protein synthetic rates and enable muscle growth. Indeed, kidney transplant patients that had been taking prescription glucocorticoids for some time were able to increase muscle mass and strength in response to resistance exercise training (52). However, it should be cautioned that glucocorticoid myopathy was not properly assessed in that study as it was not possible to have a group of kidney transplant patients not taking glucocorticoids for comparison, nor was there a record of how much mass and strength was lost (if any) as a result of the glucocorticoid treatment. Regardless, mTORC1-mediated translation of TOP mRNAs may have facilitated growth in those patients.

The decrease in the LC3 II/I ratio in response to contractions suggests high-force contractions may limit activation of the autophagy pathway regardless of treatment duration. Such a result could ultimately limit the loss of muscle mass in response to glucocorticoid administration as we and others have shown that acute exposure to glucocorticoids increases the LC3 II/I ratio, suggestive of increased autophagy. Although the LC3 II/I ratio was not increased after 7 days of treatment and was decreased following 15 days of treatment (suggesting autophagy was no longer overactive or even suppressed), the static measure of LC3 at those time point may not have captured an increase in autophagy flux that could be occurring. Therefore, measures of autophagy flux in response to glucocorticoids and high-force contractions would give a more definitive assessment of changes in autophagy pathway activity.

The potential increase in basal rates of muscle protein synthesis following longer-term glucocorticoid treatment is perplexing given that glucocorticoids have been consistently shown to decrease muscle protein synthesis. Indeed, our data support a decrease in muscle protein synthesis during the initial development of the myopathy (i.e., 7-day treatment). Despite this perplexing finding, it is consistent with results from others that show rates of muscle protein synthesis are elevated in muscle where mass and/or fiber cross-sectional area is decreased or actively decreasing. For example, rates of myofibrillar protein synthesis were elevated in the muscle of aged mice even though muscle mass was actively decreasing (53–55). A similar increase in protein synthesis was also observed in the atrophied muscle that was subjected to 14 days of denervation (56). The reason(s) for the increase in protein synthesis is not clear, but protein degradation rates would have to be elevated under those conditions to maintain the smaller muscle. It is possible that degradation rates are elevated under those conditions as a quality control mechanism because the muscle is synthesizing dysfunctional proteins. Thus, the reason for the overall increase in muscle protein synthesis that could be occurring in response to longer-term glucocorticoid treatment (and in other atrophic conditions) remains perplexing.

There are a few limitations from this work that should be noted. First, we only analyzed a single time point postcontraction, meaning the duration for which synthetic rates are increased following high-force contractions may be shortened in glucocorticoid-treated animals, which would impact the therapeutic potential of resistance exercise. Moreover, there are mTORC1-independent pathways that increase muscle protein synthesis and muscle mass in response to mechanical overload, with those pathways possibly being activated despite glucocorticoid administration. Second, the analysis of protein synthesis across animals should be considered preliminary because differences in the actual rates of puromycin incorporation would require normalizing to the tracer precursor pool. Third, a more thorough investigation of protein breakdown is needed in future work as that will help define the extent to which high-force contractions may dampen overactivation of that process in glucocorticoid myopathic muscle. Finally, different dosing regimens could yield different results, and future work should assess whether a single bolus versus a consistent daily dose leads to different outcomes.

In conclusion, we show that high-force contractions can activate mTORC1 signaling and induce an anabolic response in skeletal muscle following short-term glucocorticoid treatment. This means resistance exercise training may limit the degree of atrophy that occurs if resistance exercise is prescribed during the initial development of myopathy. However, continued exposure to glucocorticoids resulted in a state of anabolic resistance to the high-force contractions despite full activation of putative anabolic pathways. Although the anabolic resistance is undesirable, the ability to activate mTORC1 and reduce markers of autophagy at later stages of myopathy may still facilitate muscle growth through preferential translation of growth promoting mRNAs and minimizing over activation of protein breakdown. Overall, these data provide support for the use of resistance exercise as a therapeutic intervention to counteract the loss of muscle at the early stages of glucocorticoid myopathy development, with more work needed to assess the therapeutic potential following longer-term treatment.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

American College of Sports Medicine Doctoral Research Grant to K.R.D., NIH Grant AG073445 (to B.S.G.), and Florida Department of Health Grant 9BC03 (to J.L.S.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.R.D., J.L.S., R.C.H., P.B.C., and B.S.G. conceived and designed research; K.R.D. and B.S.G. performed experiments; K.R.D. and B.S.G. analyzed data; K.R.D., J.L.S., R.C.H., P.B.C., and B.S.G. interpreted results of experiments; K.R.D. and B.S.G. prepared figures; K.R.D. drafted manuscript; K.R.D., J.L.S., R.C.H., P.B.C., and B.S.G. edited and revised manuscript; K.R.D., J.L.S., R.C.H., P.B.C., and B.S.G. approved final version of manuscript.