Resistance training relieves skeletal muscle atrophy induced by hypoxia via the Akt- FoxO1-MuRF1/Atrogin-1 signaling pathway

2.1 Animals and ethics statement

The study was approved by the Animal Ethics Committee at Beijing Sport University. 40 male Sprague-Dawley rats (8 weeks of age) were housed at 22–25℃and 12-h light-dark cycles, with food and water available ad libitum. The animals were divided randomly into four groups and weighed once every day: control group (C group, n = 10), resistance training group (R group, n = 10), hypoxia group (H group, n = 10) and hypoxia resistance training group (HR group, n = 10).

2.2 Hypoxic intervention and resistance training (RT)

The rats in the H and HR groups lived in normobaric hypoxia condition that simulated the environment at a height of 4000m (12.4%O2). The C and R groups acted as controls and were housed in a normal oxygen environment. The rats in the R group trained in a normal environment while the HR group trained in a hypoxic room. We designed and made a 1.2m ladder with a 1 cm grid, which was inclined at 85° against the wall during training. The R and HR groups pre-trained for a week to familiarize them with the climbing ladder. Weights were placed in a Corning tube, with small iron balls and a hook, attached to the tail root of the rat by a rubber belt (Fig. 1B). The rats were trained to climb from the bottom to the top in 10s by gentle stimulation on their tails. The initial load was 50% of their body weight and was increased 10% every other day throughout the 4 weeks of the training period.

Table 1

Training load of R and HR groups 

2.3 Body composition, muscle wet weight and cross-sectional area (CSA)

Dual energy x-ray absorptiometry (DEXA) was used to measure the body composition of the rats. 48 h after the last training session, the animals were anesthetized with 3% pentobarbital sodium (3 ml/kg body weight). Parameters measured in the test included body weight, lean body mass, fat mass and the percentage of lean body mass. The rats were sacrificed by drawing blood from the abdominal aorta after the DEXA test. The extensor digitorum longus (EDL), soleus, gastrocnemius and bicipital muscle of forelimb were isolated quickly and weighed, followed by fixation of the muscle in 4% paraformaldehyde overnight at 4 ℃. The tissue specimens were stained with hematoxylin and eosin (HE) according to standard protocols and the CSA of each muscle analyzed using Image J software.

2.4 Cell culture experiment

Rat L6 skeletal muscle cells were purchased from ScienCell library. Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Gibco. The cells were cultured at 37℃ in high glucose DMEM (10% FBS, penicillin 100 U/ml, and streptomycin 0.1 mg/ml ) at 5% CO2. When the cells had grown to a confluence of 80–90%, the DMEM was supplemented with 2% equine serum, followed by incubation for a further 7–8 days to induce the cells to differentiate into myotubes. The differentiated myotubes were then transferred to a hypoxia chamber (Smarter 118 hypoxia system) maintained at 1% O2, 5% CO2 and 94% N2 for a further 6 h. 1, 3-Dicaffeoylquinic acid is an effective activator of the PI3K/Akt pathway, was added to the medium of the Akt-Act group and regulated to a final concentration at 50µM/L.

2.5 Immunohistochemistry and imaging

Paraffin sections of the skeletal muscle were dewaxed and repaired using citric acid antigen repair buffer. Fluorescence quenching agent was then added dropwise onto the tissue section. After washing in PBS, the sections were incubated overnight in FoxO1-S256 antibody (1:100 dilution, Cell Signaling Technology, Boston, USA). The sections were then incubated with fluorophore-conjugated secondary antibody for 1 h in a dark room. The nucleus was stained with DAPI. L6 cells were grown on glass coverslips and fixed in 4% paraformaldehyde for 15 min, followed by washing in PBS and incubation in goat serum for 30 min. The next steps were similar to those used for the tissue sections. The images were captured by laser scanning confocal microscopy and merged with NIS-Elements D image analysis software.

2.6 Western Blotting

EDL and arm biceps muscles from at least six rats tissue were dissected and frozen in a -80℃ ultra cold storage freezer. Total protein was abstracted using RIPA lysis buffer (Thermo Fisher Scientific, Massachusetts, USA) supplemented with protease inhibitor and phosphatase inhibitor (Roche, Basel, Switzerland). 30 mg of tissue was crushed using beads suspended in in homogenizer system (Eppendorf, Hamburg Germany) in 300 µl of RIPA buffer at 4℃ for 5 min. The samples were then centrifuged (12000 rpm, 10 min,4℃) and the supernatant isolated. L6 myotubes grown in six wells plates were washed with ice DPBS, followed by drop-wise addition of RIPA buffer into the wells and then left to stand for five minutes. The cell fraction was scraped from the bottom of plate situated in an ice bath and then centrifuged at 12000 rpm at 4 ℃ for 10 min. The protein concentration was measured using a bicinchonininc acid (BCA) kit (Thermo Fisher Scientific). A 20 µg aliquot of protein of each sample was subjected to SDS-PAGE (4–12% Bis-Tris gel, Thermo Fisher Scientific) and then transferred onto PVDF membrane (Thermo Fisher Scientific). The tailored membranes were then immunoblotted with primary antibodies: PI3K, (1:1000 dilution, Proteintech, Chicago, USA ); Akt (1:1000 dilution, Abcam, Cambridge, UK); FoxO1 (1:500 dilution, Abcam); FoxO1 (S256) (1:1000 dilution, Cell Signaling Technology); MuRF1 (1:2000 dilution, Abcam); Atrogin-1 (1:1000 dilution, Abcam); β-Actin (1:1000, Cell Signaling Technology), and Tubulin (1:1000, Cell Signaling Technology).

2.7 Antibodies and instruments

PI3K P110 (Alpha) antibody (20583-1-AP) was purchased from Proteintech. Anti-pan-AKT antibody (ab8805), anti-FOXO1 antibody (ab52857), anti-Fbx32 antibody (ab74023), anti-MuRF1 antibody (ab172479) and anti-fast myosin skeletal heavy chain antibody (ab91506) were purchased from Abcam. Phospho-FoxO1(Ser256) antibody (#9461), β-actin (#4970) and α-tubulin antibody (#2144) were purchased from Cell Signaling Technology. A resistance training ladder was designed and made by our group (Patent number: 201521113887.1).1, 3-dicaffeoylquinic acid was used as the Akt activator and was purchased from MCE (New Jersey, USA).

2.8 Statistical analysis

Images of the immunohistochemistry sections were analyzed using Image-Pro Plus software. The CSA analyses were carried out using Image. J. All values were expressed as mean ± SD. One- or two-way analysis of variance (ANOVA) was used to examine the effects, in the various experiments. All statistical analyses were performed using SPSS 19.0. A P value < 0.05 was considered statistically significant.

3.1 Results of rat body composition after hypoxia and exercise intervention

Table 2 shows the effect of continuous hypoxia exposure and ladder climbing resistance training on the body composition of the rats. Body weight and muscle mass in the H group were significantly lower than in the C group (341.20 ± 16.75 vs. 377.5 ± 10.75 and 226.83 ± 8.33 vs. 260.50 ± 9.35, respectively P < 0.05). Muscle mass was significantly higher in the HR group than in the H group (246.17 ± 4.71 vs. 226.83 ± 8.33, P < 0.05).

3.2 Wet weight of skeletal muscle

The wet weight of the rat soleus, gastrocnemius, EDL, and forelimb biceps after intervention are shown in Table 3, while Fig. 2 shows the morphological differences in skeletal muscle after hypoxia and resistance training. The wet weight of the EDL and forelimb biceps of the HR group were significantly higher than those in the H group by 6.7% and 7.5% respectively (153.50 ± 6.12 mg vs. 143.83 ± 7.85 mg P < 0.05, 182.17 ± 7.73 mg vs. 169.17 ± 7.28 mg, respectively, P < 0.05).

3.3 Cross section area (CSA) of skeletal muscle fibers

As shown in Table 4, the CSA measurements of the gastrocnemius, EDL and forelimb biceps in the H group were significantly lower than in the C group (11.55 ± 1.70 vs. 14.17 ± 2.23 P < 0.05, 12.40 ± 1.34 vs. 15.01 ± 2.06 P < 0.05, 12.35 ± 1.54 vs. 16.97 ± 1.66 P < 0.05, respectively). The CSA of forelimb biceps of the HR group was obviously higher than in the H group (15.02 ± 2.29 vs. 12.35 ± 1.54, P < 0.05), while the CSA of the EDL in the HR group was significantly lower than in the HR group (13.90 ± 1.96 vs. 16.00 ± 2.17 P < 0.05).

3.4 Analysis results of FoxO1 (S256) protein in soleus, EDL and forelimb biceps

As shown in Fig. 3 and Table 3, there were significant differences in the immunofluorescence reaction of the FoxO1 (S256) protein in the rat soleus, EDL and forelimb biceps among the four groups of rats. Regardless of the soleus, EDL or biceps in the HR group, the level of FoxO1 (S256) expression was higher than in the control group (P < 0.01). The level of FoxO1 (S256) expression of the EDL in the H group was significantly lower than in the C group (P < 0.05). Significant differences in the level of FoxO1 (S256) expression were observed in the EDL and biceps between the HR and H groups (P < 0.05).

^^ indicates comparison between HR group with C group, P < 0.01, ^& indicates comparison between HR group and H group, P < 0.05, ** indicates comparison between R group and H group, P < 0.01, ^§ indicates comparison between H group and C group P < 0.05, ^ααindicates comparison between HR group and R group.

3.5 Relative expression level of MCH, Akt, FoxO1, FoxO1(S256), Atrogin-1 and MuRF-1 in proteins skeletal muscle

The results of the Western blotting tests are shown in Fig. 4. The relative expression level of myosin heavy chain (MHC) protein in the EDL of H group was significantly lower than in the C group (P < 0.05). The relative expression level of FoxO1 (S256) protein in the EDL and biceps in the HR group was significantly higher than in the H group (P < 0.01 and P < 0.01, respectively). In the R group, the relative expression level of Akt in the EDL and biceps was significantly higher than that measured in the C group (P < 0.01 and P < 0.01). Akt relative expression level in the biceps of the HR group was significantly higher than in the H group (P < 0.05). The relative expression level of FoxO1 in the EDL and biceps were both significantly higher in the H group than that measured in C group (P < 0.05 and P < 0.05, respectively). Significantly lower relative expression level of FoxO1 in the EDL and biceps was observed in the HR group compared to that observed in the H group (P < 0.05 and P < 0.01, respectively). Atrogin-1 relative expression in the EDL and biceps in the H group was significantly higher than in the C group (P < 0.01 and P < 0.01, respectively). Atrogin-1 relative expression level in the EDL in the HR group was significantly lower than H group (P < 0.01). The H group had a significantly higher relative expression level of MuRF-1 in the EDL and biceps than that measured in C group (P < 0.01 and P < 0.05, respectively). The relative expression level of MuRF-1 in the EDL of the HR group was significantly lower than that observed in the H group (P < 0.01).

3.6 Activated Akt promotes FoxO1 phosphorylation and alleviated myotubes atrophy.

As shown in image (A), exposure to 1% O2 in the cultures induced rat L6 myotubes atrophy (A2). An IF image of FoxO1 (S256) in myotubes at 1% O2 showed that FoxO1 (S256) extra nuclear expression was lower than that observed at normal oxygen concentrations (A5 and A4). Akt activation alleviated myotube atrophy at 1% O2 (A3). The fluorescence intensity of FoxO1 (S256) outside the extra nucleus in Akt-Act group was higher than that observed in the H group (A6 and A5). MHC relative expression in Akt-Act cells was significantly higher than that measured in hypoxia cells (P < 0.01). The relative protein expression results showed that activation of Akt suppressed FoxO1, Atrogin-1 and MuRF1 expression during hypoxia (Figs. C-E: P < 0.05, P < 0.05 and P < 0.05). In contrast, FoxO1 (S256) relative expression in Akt-Act cells was significantly higher than in hypoxia cells (P < 0.05).

4.1 Effects of resistance training on skeletal muscle during residence in hypoxia

The morphological results of the current study showed that continuous exposure to hypoxia causd skeletal muscle loss, whereas regular resistance exercises during exposure alleviated muscle atrophy. The molecular mechanism of this process is related to an imbalance between decomposition and synthesis of muscle protein [15, 16]. Recent studies suggested that resistance ladder training blunted the skeletal muscle degradation of hindlimb immobilized rats [17] and age-induced accumulation of connective tissue [18]. There is also evidence that morphological and neuromuscular adaptions can be achieved by high intensity resistance under hypoxia [19]. This process is related to skeletal muscle remodeling. Other evidence suggests that skeletal muscle may lose volume in response to a variety of stimuli such as denervation, suspension fixation and hypoxia. This process is related to protein synthesis and degradation pathways [20, 21].

4.2 Akt-FoxO1-E3 signaling pathway and skeletal muscle atrophy in hypoxia

FoxO is a family of transcription factors which regulate various biological processes in mammalian cells. FoxO1 is one of these factors and regulates cell differentiation, survival, metabolism and has been reported to regulate muscle fiber type transformation and proteolysis [22] [23]. On the other hand, FoxO1 also induces skeletal muscle atrophy by upregulating the expression of the E3-ubiquitine ligase [24]. while another study found that MuRF1 and Atrogin-1 are E3 ligases specifically expressed in skeletal muscle with direct protein polyubiquitination targeting them for proteolysis by the 26S proteasome. It has been reported that atrogin-1 gene knock-out mice have reduced skeletal muscle atrophy induced by denervation [10]. Chiel C et al. showed that muscle atrophy induced by hypoxia is related to the E3 ligases MuRF1 and Atrogin-1 [25]. In addition, another study found that movement type and muscle contraction had different effects on the transcription MuRF1 and Atrogin-1 [26]. For example, the expression of Atrogin-1 in skeletal muscle was shown to be down-regulated after one 6–12 h period of resistance training [25, 27]. The current study demonstrated that for the first time that four-week exposure to a simulated 4000 m (12.4% O2) normobaric hypoxia (12.4% O2) environment induced rats skeletal muscle atrophy in rats, accompanied by up-regulation of FoxO1, Atrogin-1, and MuRF-1. Similar results were reported in a human study that showed eight weeks of muscle atrophy de-training resulted in a decrease in phospho-Akt and an increase of FoxO1 [28]. The results of our study confirmed that resistance training alleviated muscle atrophy induced by hypoxia and that muscle protein breakdown regulated by Akt-FoxO1 played an important role in this process. Yin et al. reported that muscular atrophy in patients with end-stage renal disease was also associated with up-regulation of FoxO1, Atrogin-1 and MuRF-1 [29]. Waddell et al. demonstrated using immunoprecipitation that FoxO1 binds to the MuRF1 promoter in C2C12 myotubular cells, a result that indicates that FoxO1 transcription factor plays a regulatory role in MuRF1 transcription in skeletal muscle [30].

Akt (protein kinase B) has been shown to activate mTOR (mammalian target of rapamycin) and downstream effectors, with exercise inducing Akt expression and activation in skeletal muscle [31]. In the current study, we showed that regular resistance training induced an increase in Akt expression and production of FoxO1, Atrogin-1 and MuRF-1 in skeletal muscle during hypoxia. A recent study also suggested that aerobic training and resistance training both safely alleviated skeletal muscle atrophy by activating the PI3K/Akt pathway [32]. Regardless of the type of skeletal muscle fiber, ladder resistance training affects skeletal muscle mass by modulating the Akt-FoxO3a pathway [33]. Gombos developed an overload model by removing the synergistic muscles (i.e. gastrocnemius, soleus) of the plantaris muscle in rats which resulted in stimulation of plantaris hypertrophy, associated with increased Akt and suppressed FoxO1 levels [34]. Taken together, these results indicate that Akt activation is the key to inhibiting skeletal muscle atrophy. In our in vitro study were observed using microscopy that 1% hypoxia in cultures induced myotube atrophy, while the results of Western blot tests suggested that exposure to hypoxia induced a decrease in MHC and Akt expression and an increase in FoxO1, MuRF-1 and Atrogin-1 expression in L6 myotubes. These results provide evidence that the Akt-FoxO1-E3 signaling pathway regulates skeletal muscle atrophy induced by hypoxia.

4.3 Phosphorylation of FoxO1 at S256 regulates skeletal muscle atrophy

The activity of FoxO1 is regulated by various post-translational modifications, including phosphorylation, acetylation and ubiquitination [35]. An early study suggested that S256 is an important phosphorylation site in FoxO1 which is an Akt-dependent target that suppresses FoxO1 activity [36]. The S256 site in FoxO1 is a switch to bind 14-3-3 dimers and promote nuclear export. It has been reported that Akt-overexpression in mouse islet cells stimulated FoxO1 phosphorylation at S256 and exerted beneficial effects on beta cells [36]. In the current study, it was obvious that hypoxia exposure inhibited FoxO1 phosphorylation at S256 in skeletal muscle. IF images showed that resistance training stimulated an increase in FoxO1 (S256) expression in the cytoplasm. In vitro, activation of Akt has also been shown to alleviate myotube atrophy during hypoxia, accompanied by an increase in FoxO1 (S256) expression. These results show that resistance training stimulates Akt expression and FoxO1 phosphorylation at S256 in muscle cells.