Loss of REDD1 augments the rate of the overload-induced increase in muscle mass

Abstract

The overload-induced increase in muscle mass is accompanied by protein accretion; however, the initiating events are poorly understood. Regulated in Development and DNA Damage 1 (REDD1), a repressor of the mechanistic target of rapamycin in complex 1 (mTORC1), blunts the elevation in protein synthesis induced by acute muscle contractions. Therefore, this study was designed to determine whether REDD1 alters the rate of the overload-induced increase in muscle mass. Wild-type (WT) and REDD1-null mice underwent unilateral functional overload (OV) of the plantaris, while the contralateral sham leg served as a control. After 3 and 5 days of OV, puromycin incorporation was used as a measurement of protein synthesis. The percent increase in plantaris wet weight and protein content was greater in REDD1-null mice after 3, 5, and 10 days OV. The overload-stimulated rate of protein synthesis in the plantaris was similar between genotypes after 3 days OV, but translational capacity was lower in REDD1-null mice, indicating elevated translational efficiency. This was likely due to elevated absolute mTORC1 signaling [phosphorylation of p70S6K1 (Thr-389) and 4E-BP1 (Ser-65)]. By 5 days of OV, the rate of protein synthesis in REDD1-null mice was lower than WT mice with no difference in absolute mTORC1 signaling. Additionally, markers of autophagy (LC3II/I ratio and p62 protein) were decreased to a greater absolute extent after 3 days OV in REDD1-null mice. These data suggest that loss of REDD1 augments the rate of the OV-induced increase in muscle mass by altering multiple protein balance pathways.

the increase in muscle mass following long-term overload is due, at least in part, to accretion of muscle protein. This can be achieved by enhancing the rate of protein synthesis relative to protein breakdown, decreasing the rate of protein breakdown relative to protein synthesis, or a combination of the two (30). It is well accepted that an increase in the rate of protein synthesis contributes considerably to muscle protein accretion and the subsequent increase in muscle mass (2, 12), but there is also evidence that reduced protein breakdown via autophagy (lysosome-mediated protein breakdown) may also contribute (11). Thus, a better understanding of the factors that alter protein balance following muscle overload is merited, as overload fails to increase muscle mass to the same magnitude in populations such as the aged.

Enhanced rates of protein synthesis and muscle mass following overload requires signaling through the mechanistic target of rapamycin in complex 1 (mTORC1) pathway. Indeed, pharmacological inhibition of this pathway using rapamycin negates both the acute enhancement in the rate of protein synthesis, as well as the long-term increase in muscle mass induced by overload (2, 8, 12). Activated mTORC1 signaling enhances the rate of protein synthesis through phosphorylation of at least two known downstream targets, the 70-kDa ribosomal protein S6 kinase 1 (p70S6K1) and the eIF4E binding protein 1 (4E-BP1) (15). Phosphorylation of these targets increases mRNA translation initiation and elongation, as well as stimulation of ribosome biogenesis, resulting in a subsequent increase in translational capacity (15, 28). Additionally, activation of mTORC1 affects protein balance through inhibition of protein degradation by phosphorylating and inhibiting proteins required for initiation of autophagy, such as uncoordinated-51 like autophagy activating kinase 1 (ULK1), as well as other substrates (10, 37). Indeed, Fry et al. (11) showed that the activation state of autophagy, as assessed by the ratio of LC3 II to LC3 I, was reduced within 3 h following a bout of resistance exercise in humans, and the reduction persisted for up to 24 h postexercise. Thus, muscle overload impacts protein balance by enhancing protein synthesis and likely through suppressing autophagy.

Regulated in Development and DNA Damage 1 (REDD1; aka DDIT4, RTP801, dig2) is a known repressor of the mTORC1 signaling pathway (5, 6, 16). We recently reported that the maximal activation state of mTORC1 and the maximal rate of protein synthesis were lower following a single bout of muscle contractions in wild-type (WT) mice relative to mice in which the REDD1 gene was disrupted (16). Additionally, it has been shown that REDD1 expression is required for full activation of autophagy (31). Overall, the available data would suggest that REDD1 expression limits the rate of the overload-induced increase in muscle mass and protein accretion by altering protein synthesis, as well as autophagy. Therefore, the objective of the present study was to determine whether REDD1 affects the rate of the overload-induced increase in muscle mass and protein accretion following synergistic ablation-induced muscle overload in association with changes in protein synthesis and markers of autophagy.

METHODS

Animals.

All experiments were performed on male mice 5–7 mo of age. Mice with a genetic disruption in the REDD1 gene (referred to hereafter as REDD1-null mice) were generated by Lexicon Genetics (The Woodlands, TX), and permission to use them was generously granted by Dr. Elena Feinstein (Quark Pharmaceuticals). Breeding pairs on a B6/129F1 background were obtained from Dr. David Williamson (SUNY Buffalo) and bred in The Pennsylvania State University College of Medicine animal facility. Male wild-type (WT), age-matched B6/129F1 control mice were obtained from Taconic (Hudson, NY). All mice were housed in a temperature- (25°C) and light (12:12-h light-dark)-controlled environment on corn cob bedding within the vivarium at the Pennsylvania State University College of Medicine. Mice were provided rodent chow (Harlan-Teklad 2018, Indianapolis, IN) and water ad libitum. The Institutional Animal Care and Use Committee of The Pennsylvania State University College of Medicine approved the animal facilities and all experimental protocols, and all animal work was performed within this facility. Details pertaining to the phenotype of REDD1-null mice have been described elsewhere (4).

Experimental design and functional overload.

All mice were housed 2–5 per cage until the day of the experiment. Unilateral functional overload of the plantaris was induced by partial removal of the synergist muscles (soleus and gastrocnemius) taking care not to affect blood flow or innervation of the plantaris. Under deep isoflurane anesthesia (∼3%) and using sterile surgical technique, the rear hind limbs were shaved, cleaned, and then disinfected with iodine solution. A small incision was made through the skin on both legs to expose the distal tendons of the triceps surae muscle complex. On the left leg, the tendons of the gastrocnemius and soleus were separated from the plantaris tendon, and the distal ∼1/3 of the gastrocnemius and soleus were removed leaving the plantaris intact. The contralateral leg underwent the same procedure, except that the gastrocnemius and soleus were left intact. The incision site was closed using 7-0 silk suture. All mice were given a subcutaneous injection of buprenorphine at a dose of 0.05 mg/kg in 1 ml of sterile saline immediately following surgery and again 5 h into recovery. No adverse effects were observed with this volume of subcutaneous injection. All mice were singly housed during the recovery for the remainder of the experimental period (1, 3, 5, or 10 days).

Protein synthesis.

Protein synthesis was assessed using the SUnSET method (13) with antibodies against puromycin (Kerafast, Boston, MA). Following a 17-h overnight fast, mice were administered an intraperitoneal injection of 0.04 μmol/g body wt of puromycin (AG Scientific; San Diego, CA) dissolved in PBS 30 min prior to the removal of the plantaris muscles. Western blot analysis procedures were performed to visualize puromycin incorporation into the protein of the plantaris, and measurements were expressed relative to total protein loaded from the Ponceau-S-stained membrane, as previously described (32). The anti-mouse secondary antibody used in this study reacts nonspecifically with proteins in mouse muscle with molecular weights of ∼25 kDa and ∼50 kDa. Consequently, quantification of the puromycin Western blots excluded these bands.

Western blot analysis.

Whole muscle protein was extracted by glass on glass homogenization in buffer consisting of 50 mM HEPES (pH 7.4), 0.1% Triton-X 100, 4 mM EGTA, 10 mM EDTA, 15 mM Na₄P₂O₇, 100 mM β-glycerophosphate, 25 mM NaF, 5 mM Na₃VO₄, and 10 μl/ml protease inhibitor cocktail (Sigma Aldrich no. P8340; St. Louis, MO). The muscle extract was centrifuged at 10,000 g for 10 min, and the protein content of the supernatant fraction was quantified using the Bradford method (3). For analysis, proteins in the supernatant were fractionated on Bio-Rad (Hercules, CA) Criterion precast gels and were transferred to PVDF membranes (Pall Lifesciences, Port Washington, NY), as previously described (14). Membranes were stained with Ponceau-S to confirm effective transfer and equal protein loading. Membranes were incubated with appropriate antibodies overnight at 4°C. Antibodies against phospho-p70S6K1 (Thr-389) (cat. no. 9205), phospho-4E-BP1 (Ser-65) (cat. no.9451), LC3B (cat. no. 2775), p62 (cat. no. 5114), phospho-ULK1 (Ser-757) (cat. no. 6888), and total ULK1 (cat. no. 8054) were obtained from Cell Signaling Technology (Danvers, MA). Antibodies against total p70S6K1 and total 4E-BP1 were custom made by Bethyl Laboratories (Montgomery, TX). Antibodies against REDD1 (cat. no. 10638-1-AP) were obtained from ProteinTech (Chicago, IL). Monoclonal antibodies against puromycin were produced in-house and are available through Kerafast (Boston, MA). Antibodies against ubiquitin were a kind gift from Dr. Vincent Chau (Pennsylvania State College of Medicine). Following incubation with appropriate secondary antibodies (Bethyl Laboratories, Montgomery, TX) (cat. no. A120-101P, A90-116P), the antigen-antibody complex was visualized by enhanced chemiluminescence using Clarity Reagent (Bio-Rad) and a ProteinSimple Fluorchem M imaging system (Santa Clara, CA). All blots were quantified using ImageJ software (National Institutes of Health, Bethesda, MD).

RNA extraction, cDNA synthesis, and qRT-PCR.

All procedures were performed as previously described (34). Briefly, RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA). RNA quantity was determined spectrophotometrically by 260-to-280 nm ratio. cDNA was synthesized using a Superscript VILO cDNA synthesis kit (Invitrogen) from 2 μg of RNA. Quantitative RT-PCR was performed using GoTaq qPCR Master Mix (Promega, Fitchburg, WI) or PowerUp SYBR Green Master Mix (Thermo Fisher Scientific, Waltham, MA). Each sample was analyzed in triplicate for each transcript. A melting curve analysis was performed for each primer pair to ensure that a single product was efficiently amplified. Relative expression levels of the 45S pre-rRNA externally transcribed spacer (ETS) and p62 were normalized by using the ΔΔ Ct method with GAPDH as a housekeeping gene. The two sets of primer sequences used for GAPDH were the folowing: forward 5′-CCA AGT GTT CAT GCC ACG TG-3′ and reverse 5′-CGA GCG ACT GCC ACA AAA A-3′. Primer sequences for GAPDH were as follows: forward 5′-ATG TTC CAG TAT GAC TCC ACT CAC G 3′ and reverse 5′-GAA GAC ACC AGT AGA CTC CAC GAC A-3′ and forward 5′-GTT GTC TCC TGC GAC TTC A-3′ and reverse 5′-TGC TGT AGC CGT ATT CAT TG-3′. Primer sequences for p62 were as follows: forward 5′-TGT GGT GGG AAC TCG CTA TA-3′ and reverse 5′-GAG AAG CTA TCA GAG AGG TGG CC-3′.

Statistical analysis.

All data are presented as means ± SE. Percent change comparisons between wild-type mice and REDD1-null mice within a time point were performed using Student's t-test. Where appropriate, two-way ANOVA was used to determine the effect of genotype and overload on the dependent variables. An unpaired Student's t-test was used post hoc when significance was observed, as previously described (16, 17). Analysis of select relationships was performed using the Pearson product moment correlation. All analyses were performed using GraphPad Prism software (La Jolla, CA). Statistical significance was set at P ≤ 0.05 for all analyses.

RESULTS

Loss of REDD1 augments the rate of the overload-induced increase in muscle mass.

To determine whether REDD1 alters the rate of the overload-induced increase in muscle mass, unilateral functional overload of the plantaris was performed, and muscle samples were collected 1, 3, 5, and 10 days following surgery. Muscle wet weight was enhanced after 1 day of overload in WT mice (P < 0.05), but this effect was not detectable after 3 days (Table 1). A significant increase in muscle wet weight and percent change in muscle wet weight (overloaded muscle vs. contralateral leg) was observed after 3 days of overload only in REDD1-null mice (P < 0.05) (Table 1 and Fig. 1A). Importantly, this was not due to an already elevated increase in sham muscle mass in REDD1-null mice, as the mean wet weight was lower than that in the WT mice (Table 1). Although both WT and REDD1-null mice showed an increase in muscle wet weight after 5 and 10 days of overload, the percent change remained significantly greater (P < 0.05) in REDD1-null mice compared with that observed in WT mice at these time points (Table 1 and Fig. 1A). Further examination of the early time points (i.e., 3 and 5 days) showed that the percent change in muscle protein with overload was greater in REDD1-null mice relative to WT mice (P < 0.05) (Fig. 1B), suggesting that the augmented rate of the overload-induced increase in muscle mass of REDD1-null mice was at least partly due to protein accretion. In support of this idea, strong, positive correlations were observed between the percent change in muscle wet weight and the percent change in total protein in the plantaris of REDD1-null mice after both 3 and 5 days of overload (P = 0.006 and P = 0.001, respectively) (Table 2). However, others have shown that factors other than protein accretion may contribute to the increase in muscle mass during these early time points. For example, transient edema is often observed after overload surgery (7). Indeed, in the present study the increase in muscle wet weight observed in WT mice on day 1 after surgery was likely due to edema. However, there was no difference in wet weight between the overloaded and sham legs on day 3, suggesting that any edema present on day 1 had resolved by day 3. Also, while others (e.g., Ref. 1) have reported changes in protein content after 3 and 5 days that appear to conflict with the data herein, the mean muscle weight of the overloaded plantaris in WT mice after 5 and 10 days is strikingly similar to those values previously reported at these time points (∼28 and 35 mg, respectively), showing consistency with other studies (27). Overall, these data indicate that loss of REDD1 augments the rate of the overload-induced increase in muscle mass through increased accretion of muscle protein during the initial periods of overload.

Table 1.

Characteristics of REDD1+/+ and REDD1−/− mice

	BW, g (Initial)	BW, g (Sacrifice)	Tibia Length, mm	Sham Plantaris Wet Weight, mg	Overload Plantaris Wet Weight, mg
REDD1+/+ (1-day overload)	32.4 ± 1.4	30.7 ± 1.4*	—	21.3 ± 0.6	25.0 ± 0.5^
REDD1−/− (1-day overload)	37.6 ± 1.9	34.8 ± 2.0*	—	21.4 ± 0.9	23.9 ± 1.5
REDD1+/+ (3-day overload)	31.3 ± 0.5	27.8 ± 0.5*	18.7 ± 0.1	20.8 ± 0.4	21.8 ± 1.3
REDD1−/− (3-day overload)	30.1 ± 0.5	26.3 ± 0.7*	17.9 ± 0.1¥	17.2 ± 0.6	22.7 ± 0.7^§
REDD1+/+ (5-day overload)	31.7 ± 0.9	28.2 ± 0.7*	18.5 ± 0.2	19.4 ± 0.5	25.8 ± 0.8^
REDD1−/− (5-day overload)	31.6 ± 0.7	28.1 ± 0.7*	17.7 ± 0.1¥	17.8 ± 0.7	27.8 ± 2.7^
REDD1+/+ (10-day overload)	31.9 ± 0.8	—	18.7 ± 0.3	19.9 ± 0.5	34.0 ± 1.3^
REDD1−/− (10-day overload)	29.1 ± 0.8	—	17.7 ± 0.2¥	18.3 ± 1.7	44.7 ± 1.4^§

Data are expressed as means ± SE. BW, body weight. REDD1+/+, mice with REDD1 gene. REDD1−/−, mice with disrupted REDD1 gene.

^*Significantly different than Initial BW, P < 0.05.

^¥Significantly different than REDD1+/+ within time point, P < 0.05.

^{^}Significantly different than Sham within genotype, P < 0.05.

^§Genotype × Overload interaction. n = 3–8 from one or two independent experiments.

Fig. 1.

Loss of REDD1 augments the rate of the overload-induced increase in muscle mass and protein accretion following unilateral synergistic ablation-induced overload. A: percent change in muscle wet weight between the sham control muscle and overloaded muscle after 1, 3, 5, and 10 days. B: percent change in total protein between sham control muscle and the overloaded muscle after 3 and 5 days. *Significant difference between genotypes within the time point. n = 3–8 per group from 1–2 independent experiments. Significance was set at P ≤ 0.05.

Table 2.

Pearson product-moment correlation matrix for REDD1-null mice after 3 and 5 days of overload

	% Change in Total Protein	P Value
% Change in Muscle Wet Weight (3-day overload)	r = 0.89	P = 0.006
% Change in Muscle Wet Weight (5-day overload)	r = 0.97	P = 0.001

Augmented rate of increased muscle mass and protein accretion occurs independent of translational capacity.

Accretion of muscle protein can occur by a change in translational efficiency, i.e., the rate of protein synthesis per unit of RNA, translational capacity, i.e., the number of ribosomes available for protein synthesis, or both. The ribosome content of a cell has been proposed to impose an upper limit to the rate of protein synthesis (20). Ribosomes are composed of both ribosomal RNA and ribosomal proteins. Given that nearly 90% of total RNA consists of ribosomal RNA, changes in RNA content are used as an indication of a change in ribosome biogenesis (34). Indeed, it was recently shown that the magnitude of increased muscle mass following various intensities of muscle overload were strongly correlated with changes in RNA content (29). Given the importance of ribosome biogenesis for the increase in muscle mass, we determined whether an increase in translational capacity in the absence of REDD1 was an initiating event that led to the augmented rate of increased mass and protein accretion in REDD1-null mice observed by 3 days of overload. As shown in Fig. 2A, the magnitude of the increase in total RNA after 3 days of overload was the same in both WT and REDD1-null mice, such that there was no difference between genotypes after either 1 or 3 days of overload. Likewise, there was also no difference between genotypes for the overload-induced increase in the relative abundance of the ETS of the 45S pre-rRNA after either 1 or 3 days of overload (Fig. 2B), suggesting that ribosomal RNA transcription and ribosome biogenesis were similar for both genotypes. No difference between genotypes was observed in translational capacity after 1 day of overload (Fig. 2C). However, this measure was significantly lower in the overloaded plantaris of REDD1-null mice compared with WT mice after 3 days of overload (P < 0.05) (Fig. 2C). These data suggest that the augmented rate of the increase in muscle mass and protein accretion in REDD1-null mice occurred in a manner that was independent of translational capacity. Further, the reduced translational capacity in REDD1-null mice compared with WT mice was not due to impaired ribosome biogenesis, but rather an increase in muscle mass.

Fig. 2.

Loss of REDD1 alters translational capacity independent of a change in ribosome biogenesis. A: percent change in total RNA between the sham control muscle and the overloaded muscle after 1 and 3 days. B: relative abundance of the external transcribed spacer (ETS) of the 45S pre-rRNA was determined by quantitative RT-PCR in the overloaded muscle of both wild-type and REDD1-null mice after 1 and 3 days. C: translational capacity [RNA content per milligram of muscle (ng/mg)] of the overloaded muscle of wild-type and REDD1-null mice after 1 and 3 days of overload. *Significant difference between overloaded muscles of genotypes within a time point. n = 6 or 7 per group. Significance was set at P ≤ 0.05.

Overload-induced translation efficiency and mTORC1 signaling are altered in REDD1-null mice.

Since the augmented rate of overload-induced increase in muscle mass and protein accretion in REDD1-null mice occurred independent of translational capacity, we next examined changes in translational efficiency. As shown in Fig. 3, A and F, protein synthesis was increased in the overloaded leg compared with the sham leg after 3 days of overload (P < 0.05). However, there was no difference in the overload-induced percent change in the rate of synthesis or absolute value between genotypes (Fig. 3, A, E, and F). Given that the translational capacity in the overloaded plantaris of REDD1-null mice was lower at this time point (Fig. 2C), the observation that the rate of protein synthesis was maintained would suggest that translational efficiency was increased. Because of the small size of the plantaris (particularly in the sham leg), protein synthesis and RNA content could not be measured in the same animals. However, when the individual values obtained for protein synthesis in the overloaded muscle presented in Fig. 3A were normalized using the average total RNA value in the overloaded plantaris in Fig. 2C, translation efficiency was found to be 38% higher (P < 0.05) in REDD1-null mice compared with WT mice at this time point. This was likely due to an increase in the absolute overload-induced activation of mTORC1, as evidenced by increased phosphorylation of p70S6K1 (Thr-389) and 4E-BP1 (Ser-65) in the plantaris of REDD1-null mice relative to the values observed in WT mice after 3 days of overload (P < 0.05) (Fig. 3, B, C, and F). Moreover, an interaction between genotype and overload was observed for phosphorylation of p70S6K1 (Thr-389) (P < 0.05) (Fig. 3B). However, although the absolute values for p70S6K1 and 4E-BP1 phosphorylation were greater in REDD1-null compared with WT mice in response to overload, the percent change in the overload-induced increase in p70S6K1 (Thr-389) phosphorylation was not different between genotypes (Fig. 3E), and the overload-induced difference in 4E-BP1 (Ser-65) phosphorylation was less in REDD1-null compared with WT mice (Fig. 3E). The latter effect was most likely due to a higher baseline value (Fig. 3, B and C). Lastly, REDD1 expression in the overloaded plantaris of WT mice was significantly reduced to 50% (P < 0.05) of the value observed in the nonoverloaded muscle (Fig. 3D). Thus, although speculative, the loss of REDD1 enhanced translational efficiency, which was likely due to a greater absolute level of mTORC1 activation, resulting in an increase in mRNA translation initiation, elongation, or both.

Fig. 3.

Loss of REDD1 alters absolute mTORC1 signaling and translational efficiency following 3 days of overload. A: protein synthesis was determined by the SUnSET method. The phosphorylated to total protein ratio of p70S6K1 (Thr-389) (B) and 4E-BP1 (Ser-65) (C) were determined by Western blot analysis. D: REDD1 content was determined by Western blot analysis. E: percent change in indicated measures between the sham and overloaded muscle. F: representative Western blots. Lines connecting bars indicate significant differences. OV, overload; +/+, wild-type mice; −/−, REDD1-null mice; P, phosphorylated protein; T, total protein. n = 7/group from two independent experiments.

After 5 days, protein synthesis was still higher in the overloaded compared with the sham leg in both WT and REDD1-null mice (Fig. 4, A and F). However, the percent change in the overload-induced rate of protein synthesis was blunted by nearly 50% in REDD1-null mice compared with WT mice (Fig. 4E). Additionally, although phosphorylation of p70S6K1 (Thr-389) and 4E-BP1 (Ser-65) was elevated in response to overload in both WT and REDD1-null mice (Fig. 4, B, C, and F), the absolute value was no longer different between genotypes. These data suggest that a failure to sustain greater absolute overload-induced phosphorylation of mTORC1 substrates in REDD1-null mice in combination with reduced translational capacity leads to a reduction in the rate of protein synthesis. In all, increased translational efficiency may have contributed to the augmented rate of the overload-induced increase in muscle mass and protein accretion at early time points (<3 days of overload), but by itself, it would not explain the continued growth in REDD1-null mice.

Fig. 4.

Loss of REDD1 blunts the overload-induced rate of protein synthesis but does not alter mTORC1 signaling after 5 days of overload. A: protein synthesis was determined by the SUnSET method. The phosphorylated-to-total protein ratio of p70S6K1 (Thr-389) (B) and 4E-BP1 (Ser-65) (C) were determined by Western blot analysis. D: REDD1 content was determined by Western blot analysis. E: percent change in indicated measures between the sham and overloaded muscle. F: representative Western blots. Lines connecting bars indicate significant differences. OV, overload; +/+, wild-type mice; −/−, REDD1-null mice; P, phosphorylated protein; T, total protein. n = 6/8 per group from two independent experiments.

Loss of REDD1 augments the suppression of markers of autophagy following overload.

The data presented thus far indicate that the augmented rate of the overload-induced increase in muscle mass and protein accretion in REDD1-null mice compared with WT cannot be fully explained by a change in translational capacity or translational efficiency. Therefore, we investigated markers of other processes affected by muscle overload that alter protein balance. Others have shown that markers of autophagy activation are reduced following a single bout of resistance exercise in humans (11). For instance, a significant reduction in the ratio of LC3 II to LC3 I in human skeletal muscle was reported following a single bout of resistance exercise that persisted for up to 24 h (11). In addition, increased expression of p62 is another, and complementary marker used to assess repressed autophagy (25, 31). Given that REDD1 regulates autophagy (31) and that markers of autophagy are reduced following a bout of resistance exercise in humans, it is possible that loss of REDD1 would lead to a further suppression in markers of this degradative process. Consistent with this idea, although the percent reduction in the LC3 II to LC3 I ratio was not different between genotypes with overload (Fig. 5F), the absolute ratio was lower in the overloaded plantaris of REDD1-null mice (P < 0.05) relative to that of WT mice (Fig. 5, A and G). Similarly, the percent increase in p62 protein content with overload was not different between genotypes (Fig. 5F), whereas the absolute p62 protein content was greater in the overloaded plantaris of REDD1-null mice (P < 0.05) relative to that of WT mice (Fig. 5, B and G). Importantly, the changes in p62 protein content observed with both overload and genotype were not due to a corresponding alteration in the relative abundance of p62 mRNA (Fig. 5E). The observation that the LC3 II/I ratio was lower and p62 content higher in REDD1-null compared with WT mice is suggestive that autophagy is repressed to a greater absolute extent in mice lacking REDD1, although the magnitude of the repression with overload is unaffected by the protein. Additional evidence supporting a role of REDD1 in the regulation of autophagy is garnered from significant correlations observed between REDD1 expression and either the ratio of LC3 II to LC3 I (r = 0.87: P = 0.002) or p62 protein content (r = −0.52: P = 0.028) in both the sham and overloaded plantaris of WT mice. To determine the potential contribution of mTORC1 to the repression of autophagy, we analyzed the phosphorylation of ULK1 (Ser-757). Phosphorylation of ULK1 (Ser-757) was increased in response to overload in mice of both genotypes (both magnitude and the absolute value) (Fig. 5, C, F, and G), suggesting the possibility that REDD1 regulates autophagy independent of mTORC1 following overload or that it enables a more effective repression by mTORC1. Lastly, there was no genotype or overload effect observed for total protein ubiquitylation (Fig. 5, D, F, and G). These data suggest that a greater absolute suppression in protein degradation as assessed via markers of autophagy following overload in REDD1-null mice contributed to the augmented rate of the overload-induced increase in muscle mass and protein accretion.

Fig. 5.

Loss of REDD1 leads to a greater absolute suppression of autophagy markers following 3 days of overload. A: The LC3 II-to-LC3 I ratio. B: p62 protein content. C: ratio of phosphorylated ULK1 (Ser-757) to total ULK1. D: total ubiquitin content were determined by Western blot analysis. E: relative p62 mRNA content was determined by quantitative RT-PCR. F: percent change in indicated measures between the sham and overloaded muscle. G: representative Western blots. Lines connecting bars indicate significant difference. OV, overload; +/+, wild-type mice; −/−, REDD1-null mice; P, phosphorylated protein; T, total protein. n = 6/7 per group from two independent experiments.

DISCUSSION

Though numerous groups have shown a change in REDD1 mRNA and/or protein expression following a single bout of resistance exercise or induction of muscle contractions (9, 16, 18, 33), the role of the mTORC1 repressor in the anabolic response is just beginning to be realized through the use of mice with a disruption in the REDD1 gene. We previously reported that REDD1-null mice had increased absolute activation of mTORC1 signaling and absolute rates of protein synthesis relative to WT mice following a single bout of muscle contractions (16). Similarly, in the present study, translational efficiency was enhanced in REDD1-null mice compared with WT mice after 3 days of overload in conjunction with elevated mTORC1 activity. However, although translation efficiency was elevated in REDD1-null mice compared with WT mice, absolute rates of protein synthesis in the overloaded muscle were not different between genotypes at this time point, most likely because of a reduction in translational capacity in REDD1-null mice compared with WT mice. Thus, the increase in translational efficiency in muscle in the overloaded leg of REDD1-null mice was sufficient to maintain protein synthesis at values equivalent to those observed in WT mice, even though the capacity for mRNA translation was reduced in null animals. Although not directly tested, it is tempting to speculate that the increase in translational efficiency was a consequence of the greater absolute mTORC1 activation that occurs in REDD1-null mice compared with WT mice in response to overload. Although mTORC1 signaling and protein synthesis are concomitantly increased in response to both an acute bout of muscle contraction (16) and overload, the mechanisms involved may differ between the models as synergistic ablation results in a continuous, prolonged increase in the rate of synthesis, while the response to muscle contractions is transient. However, the contribution of translation efficiency to the sustained growth with longer periods of overload in REDD1-null mice remains in question, as the global rate of protein synthesis after 5 days of overload was lower in the overloaded muscle of REDD1-null mice compared with that observed in WT mice with no significant difference in the absolute value of mTORC1 signaling between genotypes.

Because changes in protein synthesis do not fully account for the augmented and sustained overload-induced rate of increased muscle mass and protein accretion in the absence of REDD1, other pathways affecting protein balance were likely altered. Indeed, others have reported that a single bout of resistance exercise in humans leads to repression of autophagy (11). Similar to a resistance-training bout in humans, the results of the present study suggest that autophagy is repressed in response to overload and that the magnitude of the repression is greater in mice lacking REDD1. Although speculative, this effect may contribute to the sustained increase in muscle mass observed after 10 days of overload in REDD1-null mice. On the basis of the newly described role of REDD1 in the regulation of autophagy (31), the apparent difference in autophagy in the overloaded muscle between genotypes would likely continue to diverge at later time points. For instance, we show that REDD1 expression is no longer reduced after 5 days of overload in WT mice, and others show that REDD1 expression is increased in the muscle at 14 days of overload (32). Thus, the ability to sustain the overload-induced suppression of autophagy would likely persist in REDD1-null mice, whereas the failure to decrease REDD1 with overload in WT mice (i.e., 5 days of overload) and an eventual increase in REDD1 expression (i.e., 14 days of overload) would blunt the ability to suppress this degradative pathway. Importantly, the overload-induced change in markers of autophagy may have been independent of mTORC1-mediated control of autophagy initiation, as there was no difference between genotypes in either the percent change in ULK1 (Ser-757) phosphorylation or its absolute level in response to overload. This finding supports the newly proposed idea that REDD1 regulates autophagy in an mTORC1-independent manner (31). It is important to note that this report did not directly measure autophagy flux; rather, it measured markers of this process. As the expression of the markers used in this report are dynamic during autophagy (e.g., degradation of LC3 II and p62), it might be useful in future studies to utilize autophagy inhibitors to gain a better understanding of the role of REDD1 in the overload-induced suppression of this process. However, these inhibitory methods are not without fault, as many have nonspecific side effects, they might disrupt the lysosome-dependent activation of mTORC1, or they may induce toxicity, which could all confound the results. As such, the recent recommendations for measurement and assessment of autophagy (25) suggest the combination of the LC3 II/I ratio together with p62 content as indirect measures of autophagy flux.

Recently, Hamilton et al. (19) proposed that certain molecular brakes act to limit the absolute magnitude of mTORC1 signaling during the overload-induced increase in muscle mass. In that report, expression of ER stress proteins (i.e., BiP and CHOP) was increased after 3 days of overload and remained so through 12 days (19). As ER stress represses mTORC1 signaling at the same time as it enhances REDD1 expression (21), it follows that REDD1 should be considered as one of the molecular brakes. However, no overload-induced change in REDD1 protein expression was reported in the Hamilton et al. paper (19). Potential explanations for the discordant results include species analyzed (rat vs. mouse) or sex of the rodents (female vs. male). Alternatively, it may be due to the fact that the overload-induced repression in REDD1 protein expression occurs in a transient manner, i.e., only prior to, or at the early stages of, the onset of the increase in muscle mass (i.e., 3 days of overload; see Fig. 3C), while its expression returns to control values after an increase in muscle mass has ensued (i.e., 5 days of overload; see Fig. 4C). Thus, in the Hamilton et al. (19) report, REDD1 expression may have transiently decreased, and subsequently returned to control levels at their earliest time point of measurement (3 days of overload), in which growth was already apparent. Interestingly, a transient change in REDD1 expression has also been reported in rats subjected to unilateral hindlimb immobilization (22–24), i.e., REDD1 expression is elevated during the initial 3 days of immobilization, but returns to control values at later times. As such, it can be proposed that changes in REDD1 expression only manifest during the initial response of muscle to overload or disuse, suggesting that REDD1 plays an important role in regulating the initiating events leading to changes in muscle mass.

The functional consequences of suppressed autophagy with overload, particularly in the absence of REDD1, remains to be determined. Recent data suggest that impaired autophagy via deletion of the Atg7 gene is detrimental to skeletal muscle, and results in muscle atrophy, accumulation of protein aggregates, appearance of abnormal mitochondria, concentric membranous structures, induction of oxidative stress, and activation of the unfolded protein response (26). This is paradoxical to the findings reported here showing that a suppression in autophagy markers occurs in parallel with an overload-induced increase in muscle mass, which is typically associated with increased muscle function and muscle health. It is possible that lysosome-mediated breakdown of certain proteins or cellular structures remain intact within an enlarging muscle. This possibility is consistent with the idea that there are two major forms of autophagy, including nonselective autophagy, which is associated with stimuli such as starvation, and selective autophagy, which is triggered by and functions to remove aggregate/misfolded proteins and damaged organelles (36). Hence, it is likely that selective autophagic removal of certain damaged proteins and/or organelles remains intact with overload despite an overall repression of autophagy.

In conclusion, the data show that REDD1 modulates the rate of the overload-induced increase in muscle mass and protein accretion. When viewed in the context of previous work (16), the results of the present study are consistent with a model in which REDD1 limits the absolute level of mTORC1 activation and translation efficiency during the early time points of muscle overload, as well as suppresses activation of autophagy. Thus, REDD1 expression affects multiple pathways that augment the rate of the overload-induced increase in muscle mass.

Perspectives and Significance

Skeletal muscle is a highly plastic organ that adapts in size and function to various external stimuli, including nutrient status and mechanical load. Although it is generally accepted that muscle mass is important for functional independence and survival in various diseased populations where muscle mass is compromised, the factors that lead to changes in muscle mass with different anabolic/catabolic stimuli are not fully defined. The work described within is a continuation of a line of research aimed at defining the role of REDD1 in the regulation of muscle metabolism. As REDD1 expression is elevated in skeletal muscle during atrophic conditions such as disuse, and its expression is decreased in skeletal muscle during hypertrophic conditions such as increased mechanical overload, REDD1 represents a potentially important regulator of muscle metabolism, particularly as it relates to changes in muscle mass.

GRANTS

This project was supported by National Institutes of Health Grants R01 DK-15658 (to L. S. Jefferson) and F32 AA023422 (to J. L. Steiner).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

B.S.G., L.S.J., and S.R.K. conception and design of research; B.S.G., C.L., J.L.S., and G.A.N. performed experiments; B.S.G., C.L., and G.A.N. analyzed data; B.S.G., L.S.J., and S.R.K. interpreted results of experiments; B.S.G. prepared figures; B.S.G. drafted manuscript; B.S.G., C.L., J.L.S., G.A.N., L.S.J., and S.R.K. edited and revised manuscript; B.S.G., C.L., J.L.S., G.A.N., L.S.J., and S.R.K. approved final version of manuscript.