Background
Skeletal muscle atrophy may be caused by prolonged immobilization, which has a significant impact on the duration and intensity of rehabilitation [
1]. It is characterized by increased muscle fatigability associated with changes in muscle fiber size, reduced muscle protein synthesis [
2], and enhanced muscle protein ubiquitination and degradation [
3,
4]. The pro-inflammatory cytokine TNFα has been reported to be involved in the process of muscle wasting by causing myofibrillar protein degradation via the ubiquitin proteasome pathway [
5]. By activating transcription factor NFκB, TNFα plays a central role in regulating muscle protein catabolism in vitro and in vivo [
6‐
8]. Mourkioti et al. have shown that muscle genetically depleted of NFκB exhibits increased strength, diminished protein degradation under atrophic conditions, and enhanced regeneration in response to injury [
9]. Moreover, it has recently been reported that muscle atrophy caused by NFκB activation is associated with the induction of MuRF-1 [
8,
10] and Atrogin/MAFbx [
11], two markers of atrophy [
12,
13]. Interestingly, we recently showed that immobilization induced-atrophy is characterized by an up-regulation of the pro-inflammatory cytokines TNFα, IL-1, and IL-6 concomitantly with an up-regulation of MuRF-1 and Atrogin/MAFbx [
14]. The idea that inflammation may play a significant role in inducing skeletal muscle atrophy has also been proposed by Hirose et al. [
15] and Andrianjafiniony et al. [
16].
Inactive NFκB is retained in the cytoplasm through its association with IκBα. However, when stimulated with pro-inflammatory cytokines (TNFα, IL-1, IL-6), IκBα is phosphorylated, ubiquitinated, and degraded through the proteasome machinery, allowing free NFκB to translocate to the nucleus to transactivate its target genes [
17]. The inhibition of the proteasome machinery prevents the degradation of IκBα, which maintains NFκB in its inactive state [
18], thus preventing the up-regulation of MuRF-1 and Atrogin/MAFbx. The present study suggested that the NFκB canonical pathway plays a central role in immobilization-induced skeletal muscle atrophy and that proteasome inhibitors prevent muscle atrophy by maintaining NFκB in an inactive state.
Interestingly, proteasome machinery inhibitors reduce skeletal muscle proteolysis in vitro [
19‐
21]. Proteasome inhibitors (Velcade™ and MG132) also prevented muscle mass loss in an in vivo rat model of skeletal muscle wasting induced by denervation and cast immobilization of the hindlimb [
22,
23]. However, it is unclear whether the prevention of muscle mass loss by proteasome inhibitors can lead to functional.
To explore the effect of proteasome inhibitors on preventing muscle atrophy following hindlimb immobilization, we investigated the mechanism by which the proteasome inhibitor MG132 affects the NFκB canonical pathway in the C2C12 myogenic cell line. We also explored the effect of MG132 in an in vivo mouse model of skeletal muscle wasting induced by immobilization in which the tibialis anterior (TA) muscle undergoes rapid atrophy. Time to exhaustion experiments showed that proteasome inhibition markedly accelerates the rate of rehabilitation following immobilization. In summary, the effect of MG132 indicated that proteasome inhibitors show promise as pharmacological agents for preventing muscle atrophy and for favoring the rehabilitation of patients following severe trauma or surgery.
Methods
Animals and muscle atrophy-induced procedure
Sixteen-week-old male CD1 mice (Charles River, Montreal, QC, Canada) were used for all our experiments. Thirty minutes before the immobilization procedure, 0.1 mg/kg of buprenorphine was administrated IP. The mice were then anesthetized using isoflurane. The right hindlimb was immobilized as previously described [
14]. Briefly, the hindlimb was immobilized 7 days by stapling the foot exploiting normal dorso-tibial flexion using an Autosuture Royal 35W skin stapler (Tyco Healthcare, Pointe Claire, QC, Canada). One tine was inserted close to the toe at the plantar portion of the foot while the other was inserted in the distal portion of the gastrocnemius. The other hindlimb was used as a control. During the immobilization period, the mice were injected subcutaneously with MG132 (7.5 mg/kg/dose) or vehicle (DMSO) twice daily. DMSO containing or not MG132 was diluted in sterile pure corn oil (1:100, injected volume 150 μL). After 7 days, the tibialis anterior (TA) muscles of immobilized and non-immobilized hindlimbs from MG132- and DMSO-treated mice were either harvested or unstapled for remobilization studies. MG132 or DMSO was given only during the immobilization period.
Cast immobilization was performed according to an adapted procedure described by Frimel et al. [
24]. Briefly, hindlimbs were immobilized in the neutral position (160-180°). The ankle was fixed at 90° to keep the TA resting. The plaster of Paris cast encompassed both hindlimbs as in a hip spica. A thin layer of padding was placed underneath the cast in order to prevent abrasions. With the ankles at a straight angle, the cast did not have to be tight to prevent the mice from getting out of the cast. Care was taken to minimize the weight of the cast.
All animal experiments were approved by the Animal Ethics Committee of Université de Sherbrooke and were performed in accordance with Canadian Council on Animal Care guidelines (protocol #133-06B).
Cell culture and treatment
C2C12 cells were cultured in complete proliferation medium consisting of DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 IU/ml of penicillin, and 100 μg/ml of streptomycin. Differentiation was initiated by replacing the cell growth medium with differentiation medium (DMEM supplemented with 5% horse serum and antibiotics). After 96 h, the C2C12 cells had differentiated into myotubes and were treated with 40 μM MG132 in DMSO or vehicle (DMSO). After a 1-h treatment, the cells were stimulated or not with TNFα (50 ng/ml) in presence or absence of MG132.
Measurement of NFκB transcriptional activity
C2C12 cells were cultured in six-well tissue culture plates (3 × 105 cells/well) and transiently co-transfected with an NFκB-luciferase reporter plasmid (pNFκB-luc, 1 μg, a generous gift from Dr. T. E Gunter, University of Rochester, NY, USA) and a control Renilla luciferase plasmid (pRL-CMV, 30 ng, Promega, Madison WI, USA) in Lipofectamine 2000 transfection reagent (Invitrogen, Burlington, ON, Canada). Twenty-four hours after the transfection, the growth medium was replaced by differentiation medium, and the incubation was continued for an additional 96 h. Differentiated cells were then treated with MG132 (40 μM) or vehicle (DMSO). After a 1-h treatment, the cells were stimulated or not with TNFα (50 ng/ml). Six hours later, the cells were washed twice with PBS and lysed to determine transcriptional activity using a reporter assay system (Promega, Madison, WI, USA) and a SIRIUS luminometer (Berthold Detection Systems, Pforzhein, Germany).
Electrophoresis and immunoblotting
Protein extracts from C2C12 myotubes were solubilized in RIPA buffer (0.5% NP-40, 0.1% sodium deoxycholate, 150 mM NaCl, 50 mM Tris-HCl, pH 7.5) supplemented with Complete™ Protease Inhibitor Cocktail (Roche Molecular Biochemicals, Laval, QC, Canada). The homogenates were centrifuged at 14,000 rpm for 10 min at 4°C and the supernatants were collected. The protein concentrations in the supernatants were measured using Bradford's method (BioRad). The protein extracts (50 μg) were separated on a 12% polyacrylamide gel and electrotransferred to a polyvinylidene fluoride membrane (PVDF) (Millipore, Bedford, MA, USA). Blotted membranes were incubated overnight at 4°C in PBS-T (8 mM NaH
2PO
4, 1.5 mM KH
2PO
4, 3.5 mM KCl, 137 mM NaCl) containing 0.1% Tween-20 with anti-IκBα (1:1,500) or anti-GAPDH antibody (1:1,000) (Santa Cruz Biotechnology, Santa-Cruz, CA, USA). After extensive washing with PBS-T, the blots were incubated for 1 h at room temperature with a peroxidase-conjugated secondary antibody. After extensive washing with PBS-T, the immunostained bands were revealed with ECL Plus according to the manufacturer's instructions on a BioMax ML film (Kodak, Rochester, NY, USA). The autoradiograms were digitized and the bands were quantified by densitometric measurements using ImageJ software [
25].
Quantitative PCR
Total RNA was extracted from flash-frozen crushed TA muscle and C2C12 myotubes using TRIzol
® reagent according to the manufacturer's instructions. The RNA was reverse-transcribed using Reverse Transcriptase Superscript II. qPCR assays were performed using 50 ng of template cDNA. The conditions for all the reactions were as follows: an initial 5 min denaturation step at 95°C, followed by forty 40-s cycles at 95°C, 56°C, and 72°C. The qPCR assays were performed using a Rotor-Gene 6000 (Corbett Robotics, Australia) and iQSYBR Green Supermix (BioRad). Results were calculated using the 2
-ΔΔCT relative quantification method normalized to the HPRT1 gene. The primer sets are shown in Table
1.
Table 1
Primer sets used for quantitative PCR analyses
Hprt1
| NM_013556.2 | 5'- gcaaactttgctttccctgg -3' | 5'- acttcgagaggtccttttcacc -3' | 85 |
MuRF-1
| DQ229108 | 5'-tgcctggagatgtttaccaagc-3' | 5'-aaacgacctccagacatggaca-3' | 143 |
Atrogin-1/MAFbx
| AF441120 | 5'-aaggctgttggagctgatagca-3' | 5'-cacccacatgttaatgttgccc-3' | 223 |
Il-1β
| NM_008361 | 5'-gcccatcctctgtgactcat-3' | 5'-aggccacaggtattttgtcg-3' | 230 |
Il-6
| NM_031168 | 5'-agttgccttcttgggactga-3' | 5'-tccacgatttcccagagaac-3' | 159 |
TNFα
| NM_013693 | 5'-ccgatgggttgtaccttgtc-3' | 5'-tggaagactcctcccaggta-3' | 217 |
Histology
TA muscles were excised, fixed in formalin, and embedded in paraffin. Sections (5 μm) were stained with hematoxylin and eosin (H&E). The H&E-stained sections were used for the cross-sectional area (CSA) analyses. Over 40 myofibers/field from at least six different fields were examined (20× magnification) using an optical microscope (TE-2000-S; Nikon, Mississauga, ON, Canada) and the myofiber cross-sectional areas were measured using ImageJ software [
25].
Functional measurement of locomotor ability - Running time
The locomotor ability (running time) of mice that had both hindlimbs immobilized for 7 days before the staples were removed was measured using a treadmill (Panlab/Harvard Apparatus, Barcelona, Spain). The mice were divided into groups of 9-10 animals each. The groups received either twice-daily subcutaneous injections of MG132 in canola oil (7.5 mg/kg/dose) or vehicle (DMSO in canola oil) immediately after the immobilization procedure and then for the duration of the 7-day immobilization period. We compared the two groups to a group of non-hindlimb-immobilized age-matched mice. The running time experiment was performed using a two-lane motorized treadmill equipped with shocker plates. The treadmill was run at an inclination of +25° at 8 m/min for 5 min, after which the speed was increased to 10 m/min for 2 min and then 20 m/min. The test was stopped when the mouse remained on the shocker plate for more than 15 s without attempting to reengage the treadmill. The running time was determined from the beginning of the test. The running time experiments were performed 4 and 11 days after removing the staples and on days 0 and 7 for the non-hindlimb-immobilized age-matched control mice.
Statistics
All data are expressed as means ± SEM. Paired t- tests were used to assess the statistical significance of differences between treated muscles and untreated contralateral muscles. Unpaired t- tests were used to compare two groups of mice or two different time points. A p < 0.05 was considered statistically significant. Statistical values were calculated using GraphPad Prism 5.0 software™.
Discussion
Skeletal muscle atrophy caused by immobilization is a major challenge for patient rehabilitation. The development of therapies designed to prevent or attenuate skeletal muscle atrophy will fill an important medical need. Many therapeutic trials focusing on the TNFα-NFκB pathway have been undertaken with the purpose of reducing the impacts of muscle diseases. For example, anti-TNFα antibodies (Remicade™) have been show to interfere with TNFα activity and consequently reduce the breakdown of dystrophic muscles [
33]. A systemic treatment with curcumin, an NFκB inhibitor, has been shown to stimulate muscle regeneration after traumatic injury [
34]. The use of a synthetic double-stranded oligodeoxynucleotide as a cis-element to block the binding of NFκB to promoter regions has been shown to inhibit cachexia in a mouse tumor model [
35]. Another strategy to interfere with the NFκB signaling pathway involves the use of proteasome inhibitors. Proteasome inhibitors have been shown to inhibit IκBα degradation and prevent NFκB activation [
27,
28]. Since the activity of the NFκB canonical pathway relies on the proteasome machinery, we sought to determine whether the use of the proteasome inhibitor MG132 during immobilization-induced stimulus would enhance physical performance.
We recently showed that the inflammatory molecule TNFα is involved in skeletal muscle atrophy using an immobilized hindlimb mouse model [
14]. TNFα is known to increase the activity of the canonical NFκB pathway in conditions associated with muscle weakness, including sepsis, cancer, and ageing [
36‐
39] and to cause protein degradation in cultured myotubes [
5,
6]. In fact, the ubiquitin-proteasome pathway has been shown to be of major importance in the catabolic signaling of TNFα leading to the breakdown of muscle protein [
40]. Ubiquitinated proteins are selectively targeted for degradation through the 26S proteasome pathway by tissue-specific E3 ligases [
12,
13] that catalyze the transfer of activated ubiquitin. The muscle-specific E3 ligases MuRF-1 and Atrogin/MAFbx are part of atrophy [
13]. A review of the literature revealed slight differences in the results regarding the genes involved in the proteasome degradation pathway when it is perturbed by an increase in NFκB activity. MuRF1 mRNA, but not Atrogin/MAFbx mRNA, is upregulated in the gastrocnemius muscle of transgenic mice with constitutive NFκB activity [
8]. Judge et al. electrotransferred a dominant negative IκBα into the soleus muscle of mice that were subsequently tested in an unloading atrophy model. They showed that Atrogin/MAFbx but not MuRF-1 is regulated by the NFκB pathway in the soleus muscle [
11]. On the other hand, Bodine et al. (2001) used disuse, denervation, and hindlimb suspension models of atrophy to show that both MuRF1 and Atrogin/MAFbx are upregulated in the gastrocnemius muscle [
13]. These slight differences may be due to the different types of muscle studied, the backgrounds of the animals used, and the different models used to induce muscle atrophy.
In the present work, we showed that the proteasome inhibitor MG132 interfered with the NFκB canonical pathway of TNFα-treated C2C12 cells by protecting IκBα from degradation and thus preventing the activation of NFκB transcription. Lastly, we showed that MG132 prevents the TNFα-induced up-regulation of MuRF1 and Atrogin-1/MAFbx in vitro. Our results demonstrated that MG132 protects IκBα from degradation, which confirmed the central role of the proteasome machinery in the activation of the NFκB canonical pathway [
41,
42].
We used an immobilization model that induced skeletal muscle atrophy to show that a treatment with the proteasome inhibitor MG132 prevented the immobilization-induced atrophy of the TA muscle. These results correlated with the histological analysis showing that MG132 preserved myofiber size during immobilization. The proteasome inhibitor Velcade™ (also known as PS-341 or Bortezomib™) has been shown to partially but significantly reduce the denervation-induced atrophy of the soleus muscle [
22]. The authors concluded that the partial inhibition was due to an inadequate dose of Velcade™ or other proteolytic mechanisms that are not inhibited by the drug, such as the Ca
2+-dependent protease or lysosomal pathways, which are also known to be involved in muscle wasting [
43]. In addition to inhibiting proteasome activity, MG132 has been reported to repress certain lysosomal cysteine proteases and calpains [
44], which might also explain the total rescue of muscle mass in our study.
The immobilization-induced model is known to cause an early inflammatory process characterized by the up-regulation of TNFα, IL-1, and IL-6 mRNA, which are targets of the canonical NFκB pathway [
45‐
47]. Interestingly, treating mice with MG132 prevented the up-regulation of these pro-inflammatory genes during immobilization. Likewise, the levels of MuRF-1 and Atrogin/MAFbx mRNA, which are targets of NFκB, increased after immobilization. This effect was prevented by MG132, which completely inhibited the expression of MuRF-1 mRNA. Surprisingly, unlike the in vitro experiment, MG132 did not completely inhibit Atrogin/MAFbx mRNA expression in immobilized mice, but did significantly attenuate it. We cannot rule out the involvement of another mechanism in the increase in Atrogin/MAFbx mRNA levels in vivo. There is some evidence suggesting that disuse muscle atrophy induced by unloading is associated with the activation of an alternative NFκB pathway distinct from the pathway seen with cachexia where TNFα is responsible for muscle atrophy [
48,
49]. Despite the fact that we observed an up-regulation of the canonical NFκB pathway target genes for TNFα, IL-1, and IL-6 [
45‐
47], it is possible that there is a cross-over between with the non-canonical NFκB pathway, which is independent of IκBα. Other investigators have shown that TNFα mediates the upregulation of atrogin/MAFbx expression in the skeletal muscle via p38MAPK [
10,
50]. These conflicting findings suggest that multiple signaling pathways mediate muscle wasting, and highlight the complexity of the molecular pathways involved in the regulation of skeletal muscle atrophy.
To correlate the morphological, histological, and molecular results with the functional properties of TA muscles from MG132-treated mice, we evaluated the time to exhaustion performance of mice. Our results showed that the endurance of MG132-treated mice was significantly higher than that of vehicle-treated control mice after 4 days of remobilization. However, no significant differences were observed after 11 days of remobilization. Indeed, after 11 days, the time to exhaustion was similar to that of mice that had never had their hindlimbs immobilized.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AZC and GG designed the study and drafted the manuscript. AZC, SH, FT, EL, CG performed the experimental work and the statistical analysis. AA participated in study design. All authors have read and approved the final manuscript.