Abstract
Duchenne muscular dystrophy (DMD) is the most severe form of muscular dystrophy which leads to progressive muscle degeneration and inflammation. The receptor activator of nuclear factor NF-κB ligand (RANKL) and its receptor (RANK), which are expressed in bone and skeletal and cardiac muscles, form a signaling network upstream from nuclear factor-kappa B (NF-κB). We thus hypothesized that prolonged silencing RANKL/RANK signaling would significantly improve DMD. We showed that RANK and RANKL protein levels were increased in the microenvironment of myofibers of 5-month-old utrophin haploinsufficient mdx (mdx/utrn+/−) mice and that a 4 mg/kg dose of anti-RANKL antibody every 3 d for 28 days is optimal and more effective than 1 mg/kg every 3 d for improving the ex vivo maximum specific force (sP0) of dystrophic EDL muscles from mdx/utrn+/− mice. This functional improvement was associated with a reduction in muscle edema, damage, and fibrosis and a marked reduction in serum CK levels. The anti-RANKL treatment inhibited the NF-κB pathway, increased the proportion of anti-inflammatory and non-cytotoxic M2 macrophages, and reduced the number of centrally-nucleated myofibers and the frequency of small myofibers, suggesting that anti-RANKL inhibits the cycle of degeneration/regeneration in dystrophic mice. A three-point bending test showed that a 28-d anti-RANKL treatment increases the mechanical properties of bone in mdx/utrn+/− dystrophic mice. In conclusion, the anti-RANKL treatment protected against skeletal muscle dysfunctions while enhancing bone mechanical properties, filling two needs with one deed in the context of muscular dystrophy.
Introduction
Duchenne muscular dystrophy (DMD), an X-linked recessive disorder, is the most common form of muscular dystrophy, with an incidence of 1 in 5000 male births (1). Patients with DMD suffer from progressive muscle degeneration and weakness, and die prematurely due to respiratory and/or cardiac failure (2). DMD is caused by lack of functional dystrophin, which is essential for myofiber integrity and signaling (3). The dystrophin-deficient mdx mouse is the most common animal model for DMD. The mice exhibit increased susceptibility to mechanical stress, leading to progressive muscle degeneration, chronic inflammatory cell recruitment (e.g. T lymphocytes and macrophages), release of oxygen-free radicals and proteases, and ultimately the formation of fibrotic and non-functional tissues (4,5,6,7). Glucocorticoids, which inhibit inflammation, are generally prescribed for DMD to theoretically break the vicious cycle of damage and scarring in dystrophic muscles.
The expression of nuclear factor-kappa B (NF-κB), a key transcription factor required for the induction of a large number of pro-inflammatory genes, is upregulated in muscular dystrophy. The specific inhibition of NF-κB activity reduces damage, inflammation, and fibrosis in dystrophic muscles (8,9,10,11). The receptor activator of nuclear factor NF-κB ligand (RANKL) and its receptor (RANK) are one of the upstream signaling pathways for NF-κB. In bone, the RANKL/RANK interaction activates NF-κB, which induces the formation of multinucleated mature osteoclasts and causes bone resorption (12). Interestingly, RANK and/or RANKL are also expressed in skeletal and cardiac muscles (13,14,15,16). Our overarching hypothesis is that the RANK/RANKL/OPG triad is synchronously involved in bone-muscle deconditioning or diseases. We showed that muscle RANK plays a role in Ca2+ storage and muscle performance during denervation (17). In addition, the genetic deletion of muscle RANK in dystrophic muscles and a short-term treatment to selectively inhibit RANKL [anti-RANKL 1 mg/kg/3d for 10d] significantly improve the force and integrity of the muscles of young dystrophic mdx mice (16). We also showed that a pharmacologic treatment of young dystrophic mdx mice with recombinant full-length osteoprotegerin fused with Fc [OPG-Fc 1 mg/kg/d for 10d], a decoy receptor for RANKL, mitigated the loss of muscle force, markedly increased the expression and activity of a key sarco/endoplasmic reticulum Ca2+ ATPase (SERCA), and preserved muscle integrity, particularly in fast-twitch skeletal muscles (15). The upregulation of the expression of the pro-inflammatory cytokines TNF-α, IL-1α, and IL-1β caused by a cardiac muscle pressure overload induced by NF-κB was significantly reduced following an anti-RANKL treatment (18). Most importantly, a very recent study showed that denosumab, a human anti-RANKL monoclonal antibody, given for 3 years improves the appendicular lean mass and handgrip strength of osteoporotic women; that mice overexpressing RANKL present muscle atrophy, fat infiltration, inflammation, necrosis, and lower force production; and that a truncated-OPG-Fc treatment increases muscle mass in osteo/sarcopenic mice (19). All of these results strongly indicate that the RANK/RANKL/OPG pathway is not limited to bone, organogenesis, immunity, and cancer, but also plays a very important role in cardiac and skeletal muscle diseases and aging-associated diseases.
Since denosumab has already been approved as a treatment for osteoporosis and bone metastases in adults and has undergone testing as a treatment for osteogenesis imperfecta in children (20,21,22), the present study aimed to determine the most effective dose of an anti-RANKL treatment in adult dystrophic mice. We tested the effectiveness of various doses of anti-RANKL on mdx mice with haploinsufficiency of utrophin (mdx/utrn+/−). Our findings showed that [4 mg/kg/3d] of anti-RANKL was the optimal dose and that this dose significantly improved the specific force of dystrophic extensor digitorum longus (EDL) and soleus (Sol) muscles relative to control PBS-treated mdx/utrn+/− mice. However, the treatments did not protect dystrophic muscles from repeated eccentric contraction-induced force loss. The anti-RANKL treatments also significantly reduced the area affected by muscle damage, serum creatine kinase activity, fibrotic areas, and muscle inflammation and promoted muscle regeneration. Anti-RANKL may thus be an effective single-agent treatment option for protecting the bone and skeletal muscles of patients with DMD.
Results
RANK/RANKL levels were elevated in the microenvironment of dystrophic skeletal muscles
We previously showed that RANK mRNA levels were 5.5-fold higher in EDL muscles from dystrophic mdx mice compared with WT mice (16). To investigate the contribution of the RANK pathway to dystrophic skeletal muscle contractile dysfunction, we first evaluated RANK and RANKL levels in dystrophic muscles. Western blotting analyses showed that RANK (Fig. 1A) and RANKL (Fig. 1B) protein levels are 2-fold higher in EDL muscles isolated from dystrophic mdx/utrn+/− mice compared with WT mice. RANK and RANKL protein levels were also significantly higher in the fast-twitch TA and slow-twitch soleus (Sol) muscles from mdx/utrn+/− mice compared with WT mice (data not shown). We next assessed RANK and RANKL expression in the muscle membrane. The immunohistochemistry results showed that RANK and RANKL were present on the myofiber cell membranes of EDL muscles (Fig. 1C) and translocate from the cytoplasm to the muscle membrane in mdx/utrn+/− mice.
RANK/RANKL expression is elevated in the microenvironment of dystrophic muscles. Muscles from 5-month-old WT C57BL/10 J and dystrophic mice with one functional allele for the utrophin gene (mdx/utrn+/−) were used. RANK (A) and RANKL (B) expression in fast-twitch EDL muscles from WT and dystrophic mdx/utrn+/− mice was analyzed by Western blotting. RANK and RANKL levels were normalized to GAPDH levels. RANK (green), RANKL (red), and DAPI (blue) staining of EDL muscle sections from dystrophic mdx/utrn+/− and WT mice (C). RANK and, to a lesser extent, RANKL are highly expressed in the cell membrane of dystrophic EDL muscles. Data are expressed as means ± SEM. *p < 0.05 indicates a significant difference between WT and dystrophic mdx/utrn+/− mice; T test analysis (n = 5). Scale bar = 20 μm.
The anti-RANKL treatment improved the function of dystrophic muscles
To investigate the long-term effects of the anti-RANKL treatment on dystrophic muscle pathology, 4-month-old mdx/utrn+/− adult mice were given i.p. injections of anti-mouse RANKL [1 or 4 mg/kg/3d] for 28 days. We first assessed the ex vivo contractile properties of the muscles, which is the gold standard for evaluating isolated muscle function. The anti-RANKL treatment [4 mg/kg/3d] had the tendency to increase the absolute tetanic forces of dystrophic EDL muscles (#p = 0.08; Table 1) but significantly improved the maximum specific force of dystrophic EDL (14.1 ± 0.5 vs. 10.9 ± 0.6 N/cm2; Fig. 2A) and dystrophic Sol muscles (17.0 ± 1.2 vs. 20.9 ± 0.6 N/cm2; Fig. 2D) compared with PBS-treated mdx/utrn+/− mice. There were no differences in the maximum specific force of muscles from mdx/utrn+/− mice treated with a low dose of anti-RANKL [1 mg/kg/3d] compared with PBS-treated mice. The muscles were then subjected to repeated eccentric contractions, which are known to decrease the isometric force, particularly of fast twitch muscles. As expected, fast-twitch EDL muscles from PBS-treated mdx/utrn+/− mice lost up to 68% of their initial force while muscles from WT mice only lost 10% of their initial force (Fig. 2B). The anti-RANKL treatment did not prevent the eccentric contraction-induced loss of force of EDL muscles (Fig. 2E). As previously reported (23) eccentric contractions do not cause a loss of force of slow-twitch Sol muscles. The muscles were then weighed, and the wet mass was normalized to the body weight to evaluate the effect of the anti-RANKL treatment on muscle mass. As expected, the EDL and Sol muscles of PBS-injected 5-month-old mdx/utrn+/− mice had higher muscle mass to body weight ratios (65%; Fig. 2C, and 48%; Fig. 2F, respectively). Interestingly, a 28-day treatment of dystrophic mice with [4 mg/kg/3d] but not [1 mg/kg/3d] of anti-RANKL significantly decreased this ratio by 10% in the EDL muscles (Fig. 2C) and by 8% in the Sol muscles (Fig. 2F) compared with PBS-treated mdx/utrn+/− mice. To determine whether the anti-RANKL treatment reduced muscle edema, the EDL and Sol muscles were dried, and their water content was determined. The dry masses of the EDL and Sol muscles of the anti-RANKL-treated and PBS-treated dystrophic mice were similar. However, the water content increased by 6% in the EDL muscles and 4% in the Sol muscles of PBS-treated mdx/utrn+/− compared with the WT mice, and the higher dose of anti-RANKL [4 mg/kg/3d] reduced the water content by 3% in the EDL muscles (Table 2).
Contractile properties of EDL and Sol muscles
| Groups | EDL | Sol | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | |
| C57BL/10 J PBS | 12.7 ± 0.1 | 8.5 ± 0.3 | 39.4 ± 1.3 (#) | 20.5 ± 2.1 | 24.8 ± 2.2 | 11.5 ± 0.1 | 5.5 ± 0.6 | 27.4 ± 1.9 | 54.3 ± 4.1 | 49.2 ± 3.1 |
| mdx/utrn+/−PBS | 12.5 ± 0.1 | 9.4 ± 0.5 | 31.7 ±2.3 | 24.4 ± 1.7 | 24.2 ± 1.0 | 10.9 ± 0.3 | 5.2 ± 0.3 | 26.1 ± 1.2 | 57.5 ± 3.3 | 50.6 ± 4.8 |
| mdx/utrn+/−1 mg/kg/3d | 13.0 ± 0.2 | 11.1 ± 1.1 | 31.6 ± 2.0 | 24.3 ± 0.7 | 24.0 ± 0.5 | 11.1 ± 0.3 | 5.5 ± 0.3 | 27.2 ± 1.7 | 47.8 ± 3.4 | 44.6 ± 3.1 |
| mdx/utrn+/−4 mg/kg/3d | 13.2 ± 0.1 | 10.2 ± 0.8 | 36.8 ± 2.0 (#) | 22.0 ± 2.3 | 24.9 ± 2.1 | 11.8 ± 0.2 | 6.4 ± 0.5 | 26.4 ± 0.9 | 56.5 ± 4.2 | 50.0 ± 3.6 |
| Groups | EDL | Sol | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | |
| C57BL/10 J PBS | 12.7 ± 0.1 | 8.5 ± 0.3 | 39.4 ± 1.3 (#) | 20.5 ± 2.1 | 24.8 ± 2.2 | 11.5 ± 0.1 | 5.5 ± 0.6 | 27.4 ± 1.9 | 54.3 ± 4.1 | 49.2 ± 3.1 |
| mdx/utrn+/−PBS | 12.5 ± 0.1 | 9.4 ± 0.5 | 31.7 ±2.3 | 24.4 ± 1.7 | 24.2 ± 1.0 | 10.9 ± 0.3 | 5.2 ± 0.3 | 26.1 ± 1.2 | 57.5 ± 3.3 | 50.6 ± 4.8 |
| mdx/utrn+/−1 mg/kg/3d | 13.0 ± 0.2 | 11.1 ± 1.1 | 31.6 ± 2.0 | 24.3 ± 0.7 | 24.0 ± 0.5 | 11.1 ± 0.3 | 5.5 ± 0.3 | 27.2 ± 1.7 | 47.8 ± 3.4 | 44.6 ± 3.1 |
| mdx/utrn+/−4 mg/kg/3d | 13.2 ± 0.1 | 10.2 ± 0.8 | 36.8 ± 2.0 (#) | 22.0 ± 2.3 | 24.9 ± 2.1 | 11.8 ± 0.2 | 6.4 ± 0.5 | 26.4 ± 0.9 | 56.5 ± 4.2 | 50.0 ± 3.6 |
The EDL and Sol muscles were incubated ex vi vo and their L0 were determined. They were then electrically stimulated, and their maximal absolute force (P0), twitch tension (Pt), half-relaxation time (½ RT), and time to peak tension (TPT) were measured. The absolute force of the EDL muscles following a [4 mg/kg/3d] anti-RANKL treatment tended to be higher than that of the PBS-treated EDL muscles (#p = 0.08). No differences in other muscle contraction parameters were observed among the three groups of mice. Data are expressed as means ± SEM. Analysis of variance with Bonferroni's multiple comparisons test (n = 5–12).
Contractile properties of EDL and Sol muscles
| Groups | EDL | Sol | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | |
| C57BL/10 J PBS | 12.7 ± 0.1 | 8.5 ± 0.3 | 39.4 ± 1.3 (#) | 20.5 ± 2.1 | 24.8 ± 2.2 | 11.5 ± 0.1 | 5.5 ± 0.6 | 27.4 ± 1.9 | 54.3 ± 4.1 | 49.2 ± 3.1 |
| mdx/utrn+/−PBS | 12.5 ± 0.1 | 9.4 ± 0.5 | 31.7 ±2.3 | 24.4 ± 1.7 | 24.2 ± 1.0 | 10.9 ± 0.3 | 5.2 ± 0.3 | 26.1 ± 1.2 | 57.5 ± 3.3 | 50.6 ± 4.8 |
| mdx/utrn+/−1 mg/kg/3d | 13.0 ± 0.2 | 11.1 ± 1.1 | 31.6 ± 2.0 | 24.3 ± 0.7 | 24.0 ± 0.5 | 11.1 ± 0.3 | 5.5 ± 0.3 | 27.2 ± 1.7 | 47.8 ± 3.4 | 44.6 ± 3.1 |
| mdx/utrn+/−4 mg/kg/3d | 13.2 ± 0.1 | 10.2 ± 0.8 | 36.8 ± 2.0 (#) | 22.0 ± 2.3 | 24.9 ± 2.1 | 11.8 ± 0.2 | 6.4 ± 0.5 | 26.4 ± 0.9 | 56.5 ± 4.2 | 50.0 ± 3.6 |
| Groups | EDL | Sol | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | L0(mm) ± SEM | Pt (g) ± SEM | P0(g) ± SEM | ½ RT (ms) ± SEM | TPT (ms) ± SEM | |
| C57BL/10 J PBS | 12.7 ± 0.1 | 8.5 ± 0.3 | 39.4 ± 1.3 (#) | 20.5 ± 2.1 | 24.8 ± 2.2 | 11.5 ± 0.1 | 5.5 ± 0.6 | 27.4 ± 1.9 | 54.3 ± 4.1 | 49.2 ± 3.1 |
| mdx/utrn+/−PBS | 12.5 ± 0.1 | 9.4 ± 0.5 | 31.7 ±2.3 | 24.4 ± 1.7 | 24.2 ± 1.0 | 10.9 ± 0.3 | 5.2 ± 0.3 | 26.1 ± 1.2 | 57.5 ± 3.3 | 50.6 ± 4.8 |
| mdx/utrn+/−1 mg/kg/3d | 13.0 ± 0.2 | 11.1 ± 1.1 | 31.6 ± 2.0 | 24.3 ± 0.7 | 24.0 ± 0.5 | 11.1 ± 0.3 | 5.5 ± 0.3 | 27.2 ± 1.7 | 47.8 ± 3.4 | 44.6 ± 3.1 |
| mdx/utrn+/−4 mg/kg/3d | 13.2 ± 0.1 | 10.2 ± 0.8 | 36.8 ± 2.0 (#) | 22.0 ± 2.3 | 24.9 ± 2.1 | 11.8 ± 0.2 | 6.4 ± 0.5 | 26.4 ± 0.9 | 56.5 ± 4.2 | 50.0 ± 3.6 |
The EDL and Sol muscles were incubated ex vi vo and their L0 were determined. They were then electrically stimulated, and their maximal absolute force (P0), twitch tension (Pt), half-relaxation time (½ RT), and time to peak tension (TPT) were measured. The absolute force of the EDL muscles following a [4 mg/kg/3d] anti-RANKL treatment tended to be higher than that of the PBS-treated EDL muscles (#p = 0.08). No differences in other muscle contraction parameters were observed among the three groups of mice. Data are expressed as means ± SEM. Analysis of variance with Bonferroni's multiple comparisons test (n = 5–12).
![The anti-RANKL treatment significantly improved the specific force of dystrophic muscles but failed to prevent the loss of specific force following 7 consecutive eccentric contractions. Four-month-old dystrophic mdx/utrn+/− mice were treated with either vehicle (PBS; phosphate-buffered saline) or with [1 mg/kg/3d or 4 mg/kg/3d] of anti-RANKL for 28 days. The contractile properties of the EDL and Sol muscles were evaluated ex vivo. The [4 mg/kg/3d] of anti-RANKL treatment significantly improved the specific force (sP0) of dystrophic EDL (A) and Sol (D) muscles compared with PBS-treated mdx/utrn+/− mice. The EDL muscles were then subjected to 7 consecutive eccentric contractions. The anti-RANKL treatment did not protect dystrophic EDL and Sol muscles from repeated eccentric contraction-induced force loss (B and E). THE [4 mg/kg/3D] anti-RANKL treatment significantly decreased the muscle mass to body weight ratio of EDL (C) and Sol (F) muscles compared with PBS -treated mdx/utrn+/− mice. Muscles isolated from WT C57BL/10 J mice served as controls. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 indicate significantly different from the PBS-treated mdx/utrn+/− mice. Analysis of variance with Bonferroni's multiple comparisons test (n = 6–12).](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/hmg/28/18/10.1093_hmg_ddz124/1/m_ddz124f2.jpeg?Expires=1747897352&Signature=Ty7e1Kt9Lw8YpUqgtWncEWldjERyiW8P3iu~GRKCZEdsV7-czOK~n22UzLJBIDMwJEkghRpWNgnbWZ7g3vt19mkd0Y-FYzfc1WqSXyU0DEHALFOIOELfqByqGTG-PZIR3NrnzWmIE5aRTKjx~kVRqCBM9GuMmatuptyTMLG-7X-LWPxQAukU3FEP78c9VSh0eB~UlpQogdFfNfBaZ7CVfcaO4vYyaWv-IBJ-zVKPRt3M7wIVDE9PfuYtvZySl0xPIdPpT51ZSfT-eo7fBREs8ql3hAaf1ZN4R2hzf9Ycciti-dctJbdoNo~YshklriqucIZJVVA4zINCrdG92BxeiA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
The anti-RANKL treatment significantly improved the specific force of dystrophic muscles but failed to prevent the loss of specific force following 7 consecutive eccentric contractions. Four-month-old dystrophic mdx/utrn+/− mice were treated with either vehicle (PBS; phosphate-buffered saline) or with [1 mg/kg/3d or 4 mg/kg/3d] of anti-RANKL for 28 days. The contractile properties of the EDL and Sol muscles were evaluated ex vivo. The [4 mg/kg/3d] of anti-RANKL treatment significantly improved the specific force (sP0) of dystrophic EDL (A) and Sol (D) muscles compared with PBS-treated mdx/utrn+/− mice. The EDL muscles were then subjected to 7 consecutive eccentric contractions. The anti-RANKL treatment did not protect dystrophic EDL and Sol muscles from repeated eccentric contraction-induced force loss (B and E). THE [4 mg/kg/3D] anti-RANKL treatment significantly decreased the muscle mass to body weight ratio of EDL (C) and Sol (F) muscles compared with PBS -treated mdx/utrn+/− mice. Muscles isolated from WT C57BL/10 J mice served as controls. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 indicate significantly different from the PBS-treated mdx/utrn+/− mice. Analysis of variance with Bonferroni's multiple comparisons test (n = 6–12).
Dry and wet muscle mass of EDL and Sol muscles from anti-RANKL-treated mdx/utrn+/− mice
| Groups | EDL | SOL | ||||
|---|---|---|---|---|---|---|
| Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % of water ± SEM | Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % water ± SEM | |
| C57BL/10 JPBS | 10.4 ± 0.7 (****) | 2.5 ± 0.1 (***) | 75.4 ± 0.9 (***) | 8.6 ± 0.3 (****) | 2.0 ± 0.3 | 75.3 ± 0.6 |
| mdx/utrn+/−PBS | 17.4 ± 0.5 | 3.5 ± 0.2 | 79.6 ± 0.7 | 12.2 ± 0.3 | 2.6 ± 0.1 | 77.9 ± 0.7 |
| mdx/utrn+/−1 mg/kg/3d | 15.7 ± 0.8 | 3.3 ± 0.1 | 78.2 ± 1.1 | 11.9 ± 0.8 | 2.5 ± 0.3 | 77.4 ± 1.1 |
| mdx/utrn+/−4 mg/kg/3d | 16.0 ± 0.3 (*) | 3.6 ± 0.9 | 77.3 ± 0.2 (*) | 11.4 ± 0.3 | 2.8 ± 0.2 | 76.4 ± 1.2 |
| Groups | EDL | SOL | ||||
|---|---|---|---|---|---|---|
| Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % of water ± SEM | Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % water ± SEM | |
| C57BL/10 JPBS | 10.4 ± 0.7 (****) | 2.5 ± 0.1 (***) | 75.4 ± 0.9 (***) | 8.6 ± 0.3 (****) | 2.0 ± 0.3 | 75.3 ± 0.6 |
| mdx/utrn+/−PBS | 17.4 ± 0.5 | 3.5 ± 0.2 | 79.6 ± 0.7 | 12.2 ± 0.3 | 2.6 ± 0.1 | 77.9 ± 0.7 |
| mdx/utrn+/−1 mg/kg/3d | 15.7 ± 0.8 | 3.3 ± 0.1 | 78.2 ± 1.1 | 11.9 ± 0.8 | 2.5 ± 0.3 | 77.4 ± 1.1 |
| mdx/utrn+/−4 mg/kg/3d | 16.0 ± 0.3 (*) | 3.6 ± 0.9 | 77.3 ± 0.2 (*) | 11.4 ± 0.3 | 2.8 ± 0.2 | 76.4 ± 1.2 |
EDL and Sol muscles from mdx/utrn+/− mice have a higher muscle wet mass than those from WT mice. A [4 mg/kg/3d] anti-RANKL treatment for 28 days significantly reduced the wet mass and percentage of water in EDL muscles compared with PBS. There was no difference in the dry muscle masses of EDL and Sol muscles from mdx/utrn+/− mice. There were no differences in the Sol wet and dry muscle masses or percentages of water content between the anti-RANKL and PBS-treated mdx/utrn+/− mice. Data are expressed as means ± SEM. Significantly different from PBS-treated mdx/utrn+/− mice, *p < 0.05, ***p < 0.001, and ****p < 0.0001. Analysis of variance with Bonferroni's multiple comparisons test (n = 5–12).
Dry and wet muscle mass of EDL and Sol muscles from anti-RANKL-treated mdx/utrn+/− mice
| Groups | EDL | SOL | ||||
|---|---|---|---|---|---|---|
| Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % of water ± SEM | Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % water ± SEM | |
| C57BL/10 JPBS | 10.4 ± 0.7 (****) | 2.5 ± 0.1 (***) | 75.4 ± 0.9 (***) | 8.6 ± 0.3 (****) | 2.0 ± 0.3 | 75.3 ± 0.6 |
| mdx/utrn+/−PBS | 17.4 ± 0.5 | 3.5 ± 0.2 | 79.6 ± 0.7 | 12.2 ± 0.3 | 2.6 ± 0.1 | 77.9 ± 0.7 |
| mdx/utrn+/−1 mg/kg/3d | 15.7 ± 0.8 | 3.3 ± 0.1 | 78.2 ± 1.1 | 11.9 ± 0.8 | 2.5 ± 0.3 | 77.4 ± 1.1 |
| mdx/utrn+/−4 mg/kg/3d | 16.0 ± 0.3 (*) | 3.6 ± 0.9 | 77.3 ± 0.2 (*) | 11.4 ± 0.3 | 2.8 ± 0.2 | 76.4 ± 1.2 |
| Groups | EDL | SOL | ||||
|---|---|---|---|---|---|---|
| Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % of water ± SEM | Wet mass (mg) ± SEM | Dry mass (mg) ± SEM | % water ± SEM | |
| C57BL/10 JPBS | 10.4 ± 0.7 (****) | 2.5 ± 0.1 (***) | 75.4 ± 0.9 (***) | 8.6 ± 0.3 (****) | 2.0 ± 0.3 | 75.3 ± 0.6 |
| mdx/utrn+/−PBS | 17.4 ± 0.5 | 3.5 ± 0.2 | 79.6 ± 0.7 | 12.2 ± 0.3 | 2.6 ± 0.1 | 77.9 ± 0.7 |
| mdx/utrn+/−1 mg/kg/3d | 15.7 ± 0.8 | 3.3 ± 0.1 | 78.2 ± 1.1 | 11.9 ± 0.8 | 2.5 ± 0.3 | 77.4 ± 1.1 |
| mdx/utrn+/−4 mg/kg/3d | 16.0 ± 0.3 (*) | 3.6 ± 0.9 | 77.3 ± 0.2 (*) | 11.4 ± 0.3 | 2.8 ± 0.2 | 76.4 ± 1.2 |
EDL and Sol muscles from mdx/utrn+/− mice have a higher muscle wet mass than those from WT mice. A [4 mg/kg/3d] anti-RANKL treatment for 28 days significantly reduced the wet mass and percentage of water in EDL muscles compared with PBS. There was no difference in the dry muscle masses of EDL and Sol muscles from mdx/utrn+/− mice. There were no differences in the Sol wet and dry muscle masses or percentages of water content between the anti-RANKL and PBS-treated mdx/utrn+/− mice. Data are expressed as means ± SEM. Significantly different from PBS-treated mdx/utrn+/− mice, *p < 0.05, ***p < 0.001, and ****p < 0.0001. Analysis of variance with Bonferroni's multiple comparisons test (n = 5–12).
The anti-RANKL treatment preserved muscle integrity and reduced damage and fibrosis in dystrophic muscles
Since the anti-RANKL [4 mg/kg/3d] treatment improved the force production of dystrophic muscles, we determined whether or not the treatment had an effect on the structural and biochemical parameters of dystrophic EDL muscles. Hematoxylin and eosin staining revealed a marked reduction in the damaged areas of dystrophic EDL muscles from anti-RANKL-treated mdx/utrn+/− mice (3.1 ± 1.1%) compared with PBS-treated mdx/utrn+/− mice (12.4 ± 5.8%) (Fig. 3A and B). We then assessed muscle fibrosis by quantifying the Masson trichrome-stained area and by measuring the hydroxyproline content of the muscles. The two approaches revealed the presence of elevated fibrosis in dystrophic EDL muscles from PBS-treated mdx/utrn+/− mice compared with WT mice. The fibrotic area was significantly lower in anti-RANKL-treated mdx/utrn+/− mice compared with PBS-treated mdx/utrn+/− mice (2% vs. 7%, respectively; Fig. 3C). The hydroxyproline content also decreased by approximately 50% following the anti-RANKL treatment. (Fig. 3D). We then measured serum CK levels, an indirect indicator of muscle damage. As expected, CK levels were 6-fold higher in PBS-treated mdx/utrn+/− mice compared with WT mice. The anti-RANKL treatment significantly reduced serum CK levels by approximately 35% compared with the PBS-treated mdx/utrn+/− mice (Fig. 3E).
The anti-RANKL treatment reduced the muscle damage observed in dystrophic EDL muscles.
Representative hematoxylin/eosin and Masson’s trichrome-stained histological sections of EDL muscles from WT C57BL/10 J and mdx/utrn+/− mice treated with either PBS or [4 mg/kg/3d] of anti-RANKL for 28 days (A). Compared with the control PBS treatment, the anti-RANKL treatment significantly reduced the area of muscle damage indicated by the black arrows (A and B), fibrotic areas (A and C), hydroxyproline content normalized to wet muscle (D), and serum creatine kinase activity, a biomarker of muscle damage (E). Healthy EDL muscles from age-matched WT C57BL6/10 J mice served as controls. Muscle damage was quantified using ImageJ software. Data are expressed as means ± SEM. Significantly different from PBS-treated mdx/utrn+/− mice, *p < 0.05, **p < 0.01, and ****p < 0.0001 indicate significantly different from PBS-treated mdx/utrn+/− mice. Analysis of variance with Bonferroni's multiple comparisons test (n = 5–7). Scale bar =100 μm.
The anti-RANKL treatment shifted the macrophage phenotype and reduced p-NF-κB levels in dystrophic muscles.
We hypothesized that an anti-RANKL treatment would slow down the progression of the dystrophic pathology by counteracting excessive inflammation and promoting a shift in macrophage phenotype. To test our hypothesis, we analyzed the effect of the anti-RANKL treatment on M1 and M2 macrophages by immunohistochemistry. EDL cross-sections were double-labelled with F4/80, a pan macrophage marker, and CD206, an M2-specific macrophage antibody (Fig. 4A). The anti-RANKL treatment had no effect on total number of macrophages (F4/80+ cells; Fig. 4B) but significantly reduced the number of M1-macrophages (F4/80+/CD206- cells) by 38% (Fig. 4C) and increased the number of M2-macrophages (F4/80+/CD206+ cells) by 132% compared with PBS-treated mdx/utrn+/− mice (Fig. 4D). To determine the mechanism by which anti-RANKL reduces muscle inflammation in mdx/utrn+/− mice, we studied the activation of the NF-κB pathway in dystrophic TA muscles. The phosphorylated NF-κB (p-NF-κB-p65 on Ser536) to the total NF-κB ratio was ~ 2-fold higher in PBS-treated mdx/utrn+/− mice than in WT mice, and a 28-day anti-RANKL treatment significantly reduced this ratio in dystrophic mdx/utrn+/− muscles (Fig. 4E-G).
The anti-RANKL treatment reduced inflammation in dystrophic EDL muscles.
Cross-sections of EDL muscle were labeled with anti-F4/80 (green), which binds to all macrophage phenotypes, anti-CD206 (red), which binds to M2 macrophages and satellite cells, and DAPI (blue), which binds to DNA. M1 macrophages are F4/80+/CD206- (green) and are indicated by white arrows. M2 macrophages are F4/80+/CD206+ (orange) and are indicated by yellow arrows (A). A quantification of the total number macrophages revealed no difference between PBS-treated mdx/utrn+/− mice and anti-RANKL-treated mdx/utrn+/−mice ([4 mg/kg/3d] for 28 days) (B). The number of M1 macrophages was significantly lower (C) while the number of M2 macrophages was significantly higher (D) in EDL muscles from anti-RANKL-treated mdx/utrn+/− mice. The expression of total NF-κB (NF-κB-p65) and phosphorylated NF-κB (p-NF-κB-p65) in the TA muscles from PBS-treated C57BL/10 J, PBS-treated mdx/utrn+/−, and anti-RANKL-treated mdx/utrn+/− mice was measured by Western blotting. Twenty-eight days anti-RANKL [4 mg/kg/3d] treatment markedly reduced the level of active phosphorylated p-NF-κB-p65 and slightly reduced total NF-κB-p65 protein levels compared with PBS-treated mice (E-F). The ratio of active phosphorylated p-NF-κB-p65 to the total NF-κB-p65 protein levels was significantly reduced with anti-RANKL treatment compared with PBS-treated mdx/utrn+/− mice (G). NF-KB and p-NF-KB protein levels were normalized to GAPDH content and expressed in arbitrary units. The ratio of active p-NF-κB-p65/NF-κB-p65 was calculated from normalized western blot densities. EDL muscles from WT age-matched C57BL6/10 J mice served as controls. Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, and ****p < 0.0001 indicate significantly different from PBS-treated mdx/utrn+/− mice. Analysis of variance with Bonferroni's multiple comparisons test (n = 5). Scale bar =100 μm or 20 μm.
The anti-RANKL treatment reduced satellite cell activation and promoted a more effective reconstitution of muscle fiber size.
Cross-sections of EDL muscle were labeled with anti-Pax-7 (red), which binds to satellite cells, anti-laminin (green), and DAPI (blue) (A). The arrows indicate Pax7+ cells. A [4 mg/kg/3d] anti-RANKL treatment for 28 days significantly reduced the average number of Pax7+ cells (magnification 200x) (B) and centrally nucleated fibers (C) per field compared with PBS-treated EDL muscles. The fiber cross-sectional area distribution showed a slight shift toward medium-sized fibers, with a significantly reduced number of small-sized fibers in EDL muscles from anti-RANKL-treated mdx/utrn+/− mice compared with PBS-treated mdx/utrn+/− mice (D). Variance coefficient of the muscle fiber cross-sectional area determined using the minimal ‘Feret's diameter’ method showed a reduced heterogenicity of muscle fiber size in EDL muscles from anti-RANKL-treated mdx/utrn+/− mice compared with PBS-treated mdx/utrn+/− mice (E) Muscles from WT C57BL/10 J mice served as controls. Data are expressed as means ± SEM. *p < 0.05 and **p < 0.01 indicate significantly different from the PBS-treated mdx/utrn+/− mice. Analysis of variance with Bonferroni's multiple comparisons test (n = 4–6). Scale bar = 20 μm.
The anti-RANKL treatment reduced myofiber regeneration and promoted a shift in fiber size distribution
Previous studies have shown that macrophage phenotype switching promotes muscle regeneration and reduces muscle membrane damage (24,25) . We first determined whether the anti-RANKL treatment affected the number of satellite cells responsible for the repair of damaged myofibers. EDL muscle sections from WT, PBS-treated mdx/utrn+/− and anti-RANKL-treated mdx/utrn+/− mice were labeled with anti-Pax-7, a marker for satellite cells (Fig. 5A). The number of satellite cells was significantly lower in dystrophic EDL muscles from anti-RANKL-treated mdx/utrn+/− mice compared with PBS-treated mdx/utrn+/− mice, suggesting that the cycle of muscle degeneration/regeneration is attenuated and is closer to healthy muscles following the anti-RANKL treatment (Fig. 5B). We next assessed the number of centrally nucleated myofibers as an index of muscle regeneration. Consistent with the lower number of satellite cells, the proportion of centrally nucleated myofibers was also significantly reduced in dystrophic EDL muscles from anti-RANKL-treated mdx/utrn+/− mice compared with PBS-treated mdx/utrn+/− mice (Fig. 5C). An analysis of CSA distribution frequency revealed roughly a 3-fold reduction in the number of small myofibers (<500 μm2) accompanied by a significant increase in the number of medium-sized myofibers (2500 μm2) (Fig. 5D). Differences in muscle fiber size distribution were also quantified by calculating the variance coefficient of muscle fiber cross sectional area, which indicate the heterogenicity of muscle fiber size. The variance coefficient, determined with the minimal ‘Feret's diameter’ method, was significantly increased in dystrophic EDL muscles from PBS-treated mdx/utrn+/− mice compared with WT mice. The anti-RANKL treatment significantly reduced the variance coefficient compared with PBS-treated mdx/utrn+/− mice (249 ± 15 vs 330 ± 26, respectively; Fig. 5E).
The anti-RANKL treatment increased the mechanical properties of bone from dystrophic mice
Three-point bending tests were performed at the tibia and femur mid-diaphysis to determine whether the mechanical properties of the bones were affected by a 28-day anti-RANKL treatment. The ultimate loads, stiffness, and lengths of the tibial and femur bones were similar for the anti-RANKL-treated [1 mg/kg/3d] mdx/utrn+/−, PBS-treated mdx/utrn+/−, and WT mice. Although a 28-day treatment is relatively short to induce changes in the mechanical properties of bone, the anti-RANKL treatment [4 mg/kg/3d] significantly increased the stiffness of the tibial bone but not that of the femoral bone compared with PBS-treated mice and did not affect other mechanical parameters of the bone (Table 3).
The mechanical properties of tibial and femoral bones
| Groups | Tibia | Femur | ||||
|---|---|---|---|---|---|---|
| Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | |
| C57BL/10 JPBS | 17.8 ± 0.9 | 70.2 ± 4.1 | 17.9 ± 0.2 | 15.1 ± 0.7 | 64.6 ± 4.9 | 15.2 ± 0.1 |
| mdx/utrn+/−PBS | 16.1 ± 0.3 | 63.6 ± 3.1 | 17.8 ± 0.1 | 15.8 ± 0.5 | 62.1 ± 7.2 | 15.4 ± 0.1 |
| mdx/utrn+/−1 mg/kg/3d | 17.3 ± 0.7 | 68 ± 5.7 | 17.5 ± 0.3 | 17.1 ± 1.1 | 75.6 ± 9.1 | 17.5 ± 0.3 |
| mdx/utrn+/−4 mg/kg/3d | 17.6 ± 0.6 | 76.7 ± 4.8 (*) | 18.2 ± 0.1 | 17.4 ± 0.6 | 73.5 ± 6.2 | 18.2 ± 0.1 |
| Groups | Tibia | Femur | ||||
|---|---|---|---|---|---|---|
| Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | |
| C57BL/10 JPBS | 17.8 ± 0.9 | 70.2 ± 4.1 | 17.9 ± 0.2 | 15.1 ± 0.7 | 64.6 ± 4.9 | 15.2 ± 0.1 |
| mdx/utrn+/−PBS | 16.1 ± 0.3 | 63.6 ± 3.1 | 17.8 ± 0.1 | 15.8 ± 0.5 | 62.1 ± 7.2 | 15.4 ± 0.1 |
| mdx/utrn+/−1 mg/kg/3d | 17.3 ± 0.7 | 68 ± 5.7 | 17.5 ± 0.3 | 17.1 ± 1.1 | 75.6 ± 9.1 | 17.5 ± 0.3 |
| mdx/utrn+/−4 mg/kg/3d | 17.6 ± 0.6 | 76.7 ± 4.8 (*) | 18.2 ± 0.1 | 17.4 ± 0.6 | 73.5 ± 6.2 | 18.2 ± 0.1 |
The bone mechanical properties assessed using a three-point bending test at tibia and femur mid-diaphysis were similar for the PBS-treated mdx/utrn+/− and WT mice. A [4 mg/kg/3d] anti-RANKL treatment for 28 days significantly increased tibial stiffness compared with PBS-treated mice. Data are expressed as means ± SEM. Significantly different from PBS-treated mdx/utrn+/− mice, *p < 0.05. Analysis of variance with Bonferroni's multiple comparisons test (n = 7–16).
The mechanical properties of tibial and femoral bones
| Groups | Tibia | Femur | ||||
|---|---|---|---|---|---|---|
| Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | |
| C57BL/10 JPBS | 17.8 ± 0.9 | 70.2 ± 4.1 | 17.9 ± 0.2 | 15.1 ± 0.7 | 64.6 ± 4.9 | 15.2 ± 0.1 |
| mdx/utrn+/−PBS | 16.1 ± 0.3 | 63.6 ± 3.1 | 17.8 ± 0.1 | 15.8 ± 0.5 | 62.1 ± 7.2 | 15.4 ± 0.1 |
| mdx/utrn+/−1 mg/kg/3d | 17.3 ± 0.7 | 68 ± 5.7 | 17.5 ± 0.3 | 17.1 ± 1.1 | 75.6 ± 9.1 | 17.5 ± 0.3 |
| mdx/utrn+/−4 mg/kg/3d | 17.6 ± 0.6 | 76.7 ± 4.8 (*) | 18.2 ± 0.1 | 17.4 ± 0.6 | 73.5 ± 6.2 | 18.2 ± 0.1 |
| Groups | Tibia | Femur | ||||
|---|---|---|---|---|---|---|
| Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | Ultimate load (N) ± SEM | Stiffness (N/mm) ± SEM | Length (mm) ± SEM | |
| C57BL/10 JPBS | 17.8 ± 0.9 | 70.2 ± 4.1 | 17.9 ± 0.2 | 15.1 ± 0.7 | 64.6 ± 4.9 | 15.2 ± 0.1 |
| mdx/utrn+/−PBS | 16.1 ± 0.3 | 63.6 ± 3.1 | 17.8 ± 0.1 | 15.8 ± 0.5 | 62.1 ± 7.2 | 15.4 ± 0.1 |
| mdx/utrn+/−1 mg/kg/3d | 17.3 ± 0.7 | 68 ± 5.7 | 17.5 ± 0.3 | 17.1 ± 1.1 | 75.6 ± 9.1 | 17.5 ± 0.3 |
| mdx/utrn+/−4 mg/kg/3d | 17.6 ± 0.6 | 76.7 ± 4.8 (*) | 18.2 ± 0.1 | 17.4 ± 0.6 | 73.5 ± 6.2 | 18.2 ± 0.1 |
The bone mechanical properties assessed using a three-point bending test at tibia and femur mid-diaphysis were similar for the PBS-treated mdx/utrn+/− and WT mice. A [4 mg/kg/3d] anti-RANKL treatment for 28 days significantly increased tibial stiffness compared with PBS-treated mice. Data are expressed as means ± SEM. Significantly different from PBS-treated mdx/utrn+/− mice, *p < 0.05. Analysis of variance with Bonferroni's multiple comparisons test (n = 7–16).
Discussion
We previously showed that EDL muscles and fully differentiated C2C12 myotubes express the RANK protein and that RANK mRNA levels are 5.5-fold higher in EDL muscles from dystrophic mice compared with WT mice (17,16). Furthermore, a short-term treatment with a low dose of anti-RANKL [1 mg/kg/3d] improved the function of fast-twitch dystrophic muscles in 5-week-old juvenile mdx mice (16). Since the anti-RANKL antibody denosumab had already been approved and prescribed for osteoporosis and bone metastases, pre-clinical studies are essential for determining the optimal dose of anti-RANKL for new therapeutic applications such as for DMD.
In the present study, we showed that RANK and RANKL protein levels are highly elevated in the microenvironment of adult dystrophic EDL muscles compared with healthy muscles. We treated dystrophic mdx/utrn+/− mice with increasing doses of anti-RANKL antibody. Our results showed that a 28-day treatment with anti-RANKL [4 mg/kg/3d] is more effective than [1 mg/kg/3d] at improving the ex vivo maximum specific force (sP0) of EDL muscles from 5-month-old mdx/utrn+/− mice. A treatment with a higher dose of anti-RANKL [8 mg/kg/3d] did not produce any additional gain of force (data not shown). In addition, despite the pathology, there were no differences in the absolute isometric force (P0) of EDL and Sol muscles from mdx/utrn+/− and WT mice, suggesting, as reported by previous studies, that compensatory hypertrophy mechanisms could in part account for the maintenance of the absolute force of dystrophic EDL and Sol muscles (26). The prolonged treatment with anti-RANKL at any dose failed to significantly increase the absolute force (P0) of dystrophic EDL and Sol muscles or to protect dystrophic EDL muscles from a loss of force following repeated eccentric contractions. However, an anti-RANKL [4 mg/kg/3d] treatment reduced the water content of the muscles, suggesting that the gain in specific force can most likely be attributed to a reduction in muscle mass, i.e. muscle edema, fibrosis, or fat cells. These improvements were associated with a significant reduction in serum CK levels, suggesting that the integrity of the dystrophic muscles had improved. Collectively, these findings support our hypothesis that a prolonged anti-RANKL treatment is effective in preventing muscle edema and loss of integrity and in improving the specific force production of dystrophic muscles.
Although the inflammatory and damaging roles of RANKL in dystrophic muscles are certainly not fully understood, inflammatory arthritis models have shown that the role of RANKL is not limited to the induction of osteoclastogenesis and bone resorption and indicate that RANKL may be an underestimated proinflammatory modulator (27). RANKL is a chemotactic molecule for monocytes/macrophages (28), and its presence increases the expression of pro-inflammatory cytokines and the M1 macrophage phenotype (29). The present findings show that the anti-RANKL treatment significantly reduced the number of M1 macrophages (F4/80+/CD206-) and increased the number of M2 macrophages (F4/80+/CD206+). Since M1 macrophages are associated with muscle degeneration and M2 macrophages are associated with muscle growth, repair, and regeneration (30), this shift in macrophage phenotype following the anti-RANKL treatment should prevent further muscle degeneration and/or promote muscle repair. Our results also showed that the anti-RANKL treatment significantly reduce the number of centrally-nucleated myofibers, the frequency of small myofibers < 500 μm, the heterogenicity of myofiber size and the number of satellite cells, suggesting that anti-RANKL prevents the vicious cycle of muscle degeneration/regeneration in dystrophic mice. It is interesting to note that RANKL is a member of TNF superfamily and that TNF-α, a related molecule, is an important activator of satellite cells and inhibitor of myogenic differentiation (31,32). Our observation that the anti-RANKL treatment reduced the number of satellite cells suggests that RANKL may play similar role in muscular dystrophy. Furthermore, the specific deletion of TNF-receptor associated factor-6 (TRAF-6), an essential RANK adaptor for relaying downstream pathways, including NF-kB and the c-Jun N terminal kinase, improves muscle strength, reduces fiber necrosis and the infiltration of macrophages, and suppresses the NF-κB pathway in dystrophic mdx mice, supporting a pro-inflammatory role for the RANKL/RANK interaction in muscle diseases (33). Additionally, RANKL expression increases in response to a pressure overload in a model of transverse aortic constriction (TAC) and stimulates the expression of TNF-α, IL-1α, and IL-1β in cultured cardiomyocytes (18). Interestingly, a single anti-RANKL injection is sufficient to attenuate the expression of pro-inflammatory cytokines following TAC (18). Our findings indicated that RANKL inhibits the NF-κB signaling pathway and inflammation, a validated target for DMD and a key controller of many genes involved in inflammatory cell recruitment. This observation is of the utmost importance since NF-κB is aberrantly activated and its inhibition markedly improves the functions of dystrophic skeletal and cardiac muscles (34,35,36). Taken together, these results showed that the anti-RANKL treatment favors anti-inflammatory and non-cytotoxic M2 macrophage proliferation and inhibits the NF-κB pathway and muscle damage, thus improving the function of dystrophic muscles.
Our findings showed that a 28-day anti-RANKL treatment significantly increases tibial stiffness but does not modify other bone biomechanical parameters. The limited changes in bone mechanical properties are not surprising since bone is not metabolically as active as skeletal muscle and since muscular dystrophy mainly affects skeletal muscle not bone at early ages. In humans, a prolonged 18-month anti-RANKL treatment improved lumbar bone mineral density and bone turnover markers in a glucocorticoid-treated boy with DMD (37). In addition a, 2-year treatment with anti-RANKL improved bone mineral density, the normalization of vertebral shape, and mobility, and reduced the fracture rate in children with osteogenesis imperfecta (38). The possibility that anti-RANKL may have a dual effect on skeletal muscle and bone is best exemplified by recent work by Ferrari et al., who reported that a 3-year treatment with denosumab, but not bisphosphonate, improved the appendicular lean mass and handgrip strength of osteoporotic women while mice overexpressing RANKL presented muscle atrophy, fat infiltration, inflammation, necrosis, and lower force production (19). Evidence from the present study and others thus support the notion that anti-RANKL preserves both bone and skeletal muscle functions in the context of osteoporosis and DMD.
In conclusion, we showed for the first time that RANK/RANKL protein levels are higher in the microenvironment of dystrophic muscles and that inhibiting their interaction with anti-RANKL antibody results in less muscle damage, fibrosis, edema, and inflammation and inhibits NF-κB activity. The anti-RANKL treatment was also effective in preserving muscle function and increasing bone stiffness in dystrophic mdx/utrn+/− mice. The ultimate goal of the present study was to extrapolate an effective dose of anti-RANKL for mice to humans. The extrapolation of a dose from mice to humans is based on the body surface area where 4 mg/kg/3d for mice would be divided by 12.3 to give 0.3 mg/kg/3d for humans or 3 mg/kg/month of human monoclonal anti-RANKL (39). A 1 mg/kg of body weight dose of denosumab every 12 weeks in children with osteogenesis imperfecta is safe and effective and should be helpful in guiding any dose escalation trial in DMD patients.
Materials and Methods
Animals
The animal experiments were approved by the Université Laval Research Center Animal Care and Use Committee based on Canadian Council on Animal Care guidelines. Male wild-type (WT) (C57BL/10ScSnJ) and mdx/utrn+/− mice were initially purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and subsequently bred in our animal facility. Mdx/utrn+/− mice were used because they present significant more inflammation and fibrosis than mdx mice (40). The genotypes were determined from tail biopsy specimens using real-time PCR with specific probes. The mice were housed under a 12:12-h light/dark cycle. Food and water were provided ad libitum.
The mice were injected i.p. with phosphate-buffered saline (PBS) or 1 or 4 mg/kg] of anti-mouse RANKL mAbs (IK22–5) every 3 days (41) from 16 to 20 weeks of age, i.e. for 28 days. Age-matched dystrophic and WT mice received an equivalent volume of PBS. EDL and Sol muscles were removed to assess their contractile properties. Tibialis anterior (TA) muscles were snap frozen and were stored at −80°C for further analysis. At the end of the experimental procedures, the mice were euthanized by cervical dislocation under anesthesia
Assessment of skeletal muscle contractile properties
Contractile properties were determined at the end of the 28-d treatment. The mice were injected s.c. with 0.1 mg/kg of buprenorphine to alleviate pain and were anesthetized i.p. with 50 mg/kg of pentobarbital sodium 15 min later. The EDL and Sol muscles were gently removed, and their contractile properties were measured using a 305B-LR dual-mode muscle arm system controlled by a dynamic muscle data acquisition and analysis system (Aurora Scientific Inc., Aurora, ON, Canada). The muscles were incubated in a controlled physiological environment (Krebs-Ringer supplemented with 95% O2 and 5% CO2, pH 7.4) at 25°C and were attached to an electrode and a force sensor. A single twitch contraction was obtained, and tetanic contractions were elicited at frequencies of 10, 20, 50, 80, and 100 Hz. The muscles were then subjected to seven consecutive 700-ms isometric eccentric contractions at 150 Hz separated by 1-min rest periods. The muscles were stretched for 200 ms to 110% L0 at 0.5 L0/s 500 ms after the start of each isometric contraction. Maximum specific tetanic tension sP0 (in N/cm2) values were obtained by normalizing the absolute force P0 with the cross-sectional area (CSA) using the following equation: sP0 = P0/CSA. CSA was determined by dividing the muscle mass by the product of the optimum fiber length (Lf) corresponding to the result of multiplying L0 with the fiber length ratio (0.44 for EDL muscles and 0.71 for Sol muscles) and the muscle density (1.06 mg/mm3). Muscle contractility measurements were analyzed using Dynamic Muscle Data Analysis software (Aurora Scientific Inc.). At the end of experiment, the EDL and Sol muscles were weighed to determine their wet, were dried overnight at 65°C, and were weighed once again to determine their dry mass. The percentage of water content was calculated using the following equation: 100 x (wet mass-dry mass)/wet mass.
Immunofluorescence and staining
Transversal EDL muscle sections (10 μm) were cut using a refrigerated (−20°C) cryostat (Leica Microsystems CM1850). Tissue sections were stained with hematoxylin and eosin using the manufacturer’s protocol (Sigma-Aldrich) to assess muscle damage, the cross-sectional areas (CSA) of the fibers and central nucleation. The variance coefficient of CSA was calculated using Feret’s diameter method as follows: variance coefficient = (standard deviation of the muscle fiber size/mean muscle fiber size) × 1000 (42). Masson’s trichrome staining was used to access collagen infiltration. Digital photographs were acquired from at least five different sections at 400x magnification with a digital camera and were examined with an inverted microscope (Nikon). The damaged area was defined as the area occupied by infiltrating cells and devoid of normal or regenerating muscle fibers. The nuclei of the infiltrating cells were used to delineate manually the damaged area. Empty spaces devoid of infiltrating cells resulting from tissue preparation were not considered. Data are expressed as the percentage of damaged and fibrotic areas with respect to the total area using ImageJ (software version 1.41). Other sections were incubated overnight at 4°C with primary anti-RANK (Abcam, 1:200) and anti-RANKL (IK22–5, 1:10) antibodies diluted in blocking solution composed of PBS supplemented with 1% horse serum and 2% bovine serum albumin (BSA). In other preparations for double-labeling, the sections were incubated overnight at 4°C with anti-F4/80 (Bio-Rad, 1:100) and anti-CD206 (Santa Cruz, 1:50) or anti-Pax-7 (R&D systems, 1:100) and anti-laminin (Sigma-Aldrich, 1:1000) antibodies in blocking solution. The sections were washed briefly with PBS, incubated with Alexa Fluor 488 or 594-conjugated secondary antibody (Invitrogen, 1:500) for 1 h at room temperature, and washed three times for 15 min with PBS. The slides were then mounted with Fluoromount-G™, with DAPI immunofluorescent stain and were analyzed with an Axio Imager M2 microscope connected to an AxioCam camera using ZEN2 software (Zeiss, Germany).
Serum creatine kinase assay
Blood collected from the mice by cardiac puncture was allowed to clot and was centrifuged at 10000 xg for 10 min at 4°C. The supernatant was transferred to a clean tube for a second round of centrifugation. The serum was then collected and was stored at −80°C until used. Serum creatine kinase (CK) levels, an indicator of muscle damage and sarcolemma membrane fragility in dystrophic mice, were determined using a commercially available kit according to the manufacturer’s instructions (Pointe Scientific Creatine Kinase CK10 reagent, Fisher Scientific) and a modified protocol from Treat NMD_M.2.2.001. Serum CK activities were measured using a microplate reader (Infinite F200, TECAN) and were expressed as U/L.
Hydroxyproline content
Collagen content was quantified in EDL muscles using a protocol adapted from Treat-NMD SOP DMD_M.1.2.006. The muscles were dried for 16–18 h at 56°C, weighed, and hydrolyzed for 3 h with 6 N HCl at 130°C in a heat block. The hydrolysates were suspended in a mixture composed of 1 mL of distilled water, 0.1 mL of red methyl, 0.15 mL of 2.5 N NaOH, completed with distilled water up to 2 mL and were centrifuged at 125 x g for 5 min. A hydroxyproline standard curve was prepared using known concentrations of L-hydroxyproline [0 to 6 μg] (Sigma-Aldrich). Next, 0.2 mL of supernatant and standard were mixed with 0.1 mL of 0.05 M chloramine-T in citrate buffer and was oxidized at RT for 20 min. The standard mixtures were then mixed with 0.1 mL of perchloric acid at RT for 5 min. Lastly, 0.1 mL of preheated 20% p-dimethylaminobenzaldehyde was added, and the mixtures incubated for 45 min at 60°C. Hydroxyproline values (μg/μL) were determined by measuring the absorbance at 560 nm and were expressed as μg of hydroxyproline/mg muscle wet weight.
Western blotting
The muscles were frozen immediately after dissection and were homogenized (PowerGen 125 homogenizer; Fisher Scientific) for 45 s on ice in RIPA lysis buffer. The protein content was measured using a BCA protein assay kit (EMD Chemical). The homogenates were dissolved in Laemmli buffer, heated to 95°C for 2 min, separated by SDS-PAGE, and transferred to a polyvinylidene difluoride membrane (PVDF; Bio-Rad). The membrane was blocked in 5% skim milk and was incubated overnight at 4°C with anti-RANK (Abcam, 1:1000), anti-RANKL (Novus Biologicals, 1:500), and anti-NF-κB-p65 and anti-phospho-NF-κB-p65 (Ser536) (Cell Signaling, 1:1000 and 1:500, respectively) primary antibodies. The membranes were washed and were incubated with the appropriate HRP-conjugated secondary antibodies, and protein bands were visualized by chemiluminescent detection using the ECL-Plus imaging system (Perkin-Elmer). Band densities were analyzed using Quantity One software (v4.6.6, Bio-Rad) and were normalized to the GAPDH band of the appropriate sample. Protein levels were expressed in arbitrary units relative to GAPDH content. The p-NF-κB-p65 to total NF-κB-p65 ratio was calculated using band densities normalized to GAPDH.
Three-point bending testing of the tibia and femur mid-diaphysis
The functional properties of bone were assessed by performing a three-point binding test at mid-diaphysis using an MTS Bionix® servohydraulic test system (MTS Systems Corporation). The tibia and femur were collected at the end of the anti-RANKL treatment and were placed on a set of supports separated by 1 cm. The rate of displacement was 2 mm/min until failure. The load displacement curve for each bone was analyzed, and the functional properties of the bone were quantified as ultimate load and stiffness. Ultimate load was determined as the highest load obtained prior to failure, and stiffness corresponded to the highest slope along the linear portion of the curve prior to reaching the ultimate load (43).
Statistical analysis
All values are expressed as means ± SEM. Statistical analyses were performed with GraphPad Prism software (version) using a one-way ANOVA or a two-way ANOVA for fiber size frequency. All analyses of variance were followed by Bonferroni's multiple comparisons test (a posteriori) to determine significant differences between groups (Instats Graph software version 3.1). p < 0.05 was considered statistically significant.
Conflit of interest statement
The authors declare that they have no competing interests.
Acknowledgements
This work was supported by grants to JF from the Natural Sciences and Engineering Research Council of Canada (RGPIN-2016-05845) and two important foundations, Ryan’s Quest and Jesse’s Journey.
