Muscular dystrophies are inherited myogenic disorders accompanied by progressive skeletal muscle weakness and degeneration. We previously showed that resveratrol (3,5,4′-trihydroxy-trans-stilbene), an antioxidant and activator of the NAD+-dependent protein deacetylase SIRT1, delays the progression of heart failure and prolongs the lifespan of δ-sarcoglycan-deficient hamsters. Because a defect of dystroglycan complex causes muscular dystrophies, and δ-sarcoglycan is a component of this complex, we hypothesized that resveratrol might be a new therapeutic tool for muscular dystrophies. Here, we examined resveratrol's effect in mdx mice, an animal model of Duchenne muscular dystrophy. mdx mice that received resveratrol in the diet for 32 weeks (4 g/kg diet) showed significantly less muscle mass loss and nonmuscle interstitial tissue in the biceps femoris compared with mdx mice fed a control diet. In the muscles of these mice, resveratrol significantly decreased oxidative damage shown by the immunostaining of nitrotyrosine and 8-hydroxy-2′-deoxyguanosine and suppressed the up-regulation of NADPH oxidase subunits Nox4, Duox1, and p47phox. Resveratrol also reduced the number of α-smooth muscle actin (α-SMA)+ myofibroblast cells and endomysial fibrosis in the biceps femoris, although the infiltration of CD45+ inflammatory cells and increase in transforming growth factor-β1 (TGF-β1) were still observed. In C2C12 myoblast cells, resveratrol pretreatment suppressed the TGF-β1-induced increase in reactive oxygen species, fibronectin production, and expression of α-SMA, and SIRT1 knockdown blocked these inhibitory effects. SIRT1 small interfering RNA also increased the expression of Nox4, p47phox, and α-SMA in C2C12 cells. Taken together, these findings indicate that SIRT1 activation may be a useful strategy for treating muscular dystrophies.
Muscular dystrophies are hereditary diseases involving progressive muscular weakness and the degeneration of muscle cells (Emery, 2002). Loss-of-function mutations in genes encoding components of the dystroglycan complex, such as dystrophin and sarcoglycans, result in increased fragility of the sarcolemma. The mechanism leading to muscular injury remains unknown. Duchenne muscular dystrophy (DMD), caused by dystrophin deficiency, is the most prevalent lethal X-linked myopathy, and it is severe. The dystrophin-deficient mdx mouse is used as a DMD model, although it retains a normal lifespan and the resultant myopathology is much less severe compared with the human disease course. Several pharmacological treatments for DMD have been studied, including exon skipping (Alter et al., 2006) and activation of the nitric oxide-cGMP signaling pathway by a phosphodiesterase 5 inhibitor (Asai et al., 2007). However, only corticosteroids are currently used as an established therapy.
Because oxidative damage is found in the muscle even before the onset of muscle necrosis (Disatnik et al., 1998), reactive oxygen species (ROS) have been implicated in the pathogenesis of DMD (Rodriguez and Tarnopolsky, 2003). Administration of the antioxidant N-acetyl cysteine to mdx mice decreases ROS levels, significantly ameliorates stretch-induced muscle damage, and reduces the number of centrally nucleated premature muscle fibers (Whitehead et al., 2008). Inflammatory cells, such as neutrophils and macrophages, may be one of the sources of excessive ROS in the dystrophic muscle. Myofiber necrosis promotes the invasion of inflammatory cells, releases cytokines such as transforming growth factor β (TGF-β), and enhances myofiber degeneration. TGF-β induces the transformation of fibroblasts into myofibroblasts, which extensively promote fibrosis by secreting collagens and fibronectin (Barnes and Gorin, 2011). Inhibition of the TGF-β signal and inflammatory response is a therapeutic strategy for DMD (Huang et al., 2009; Taniguti et al., 2010).
Sirtuins are mammalian homologues of Sir2 (silent mating type information regulation 2), which is implicated in the longevity of yeast, worms, and flies (Horio et al., 2011). Of the seven mammalian sirtuins, SIRT1 is an NAD+-dependent histone/protein deacetylase that plays crucial roles in cell survival, oxidative stress reduction, inflammation inhibition, metabolism, differentiation, and DNA repair (Horio et al., 2011). Resveratrol (3,5,4′-trihydroxy-trans-stilbene), a naturally occurring polyphenol found in grapes and red wine, has been found to activate SIRT1 (Howitz et al., 2003). We found that SIRT1 is a nucleocytoplasmic shuttling protein (Tanno et al., 2007). Its translocation into the nucleus induces superoxide dismutase 2 (Sod2/Mn-Sod), decreases ROS levels, and inhibits oxidative stress-induced cell death in C2C12 myoblast cells (Tanno et al., 2010). The administration of resveratrol to TO-2 hamsters, an animal model of congenital heart failure, increases the expression level of cardiac Sod2, decreases fibrosis, ameliorates cardiac dysfunction, and extends the lifespan (Tanno et al., 2010). Dilated cardiomyopathy develops in TO-2 hamsters because of a mutation in δ-sarcoglycan, a component of the dystroglycan complex.
Collectively, these findings raised the possibility that the activation of SIRT1 might also be effective in treating muscular dystrophies. In addition, SIRT1 deacetylates and activates peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) (Rodgers et al., 2005), which inhibits muscular atrophy induced by denervation and is involved in muscle fiber-type switching from glycolytic fast fibers to oxidative slow fibers (Sandri et al., 2006; Schuler et al., 2006). In addition, PGC-1α induces Sod2 and reduces ROS levels (St-Pierre et al., 2006). The overexpression of PGC-1α in the muscle cells of mdx mice reduces muscle fiber damage, increases running performance, and decreases the serum creatine kinase (CK) level (Handschin et al., 2007), indicating that SIRT1's activation by resveratrol might also provide an advantage via the activation of PGC-1α.
In this study, we investigated the effect of resveratrol on muscle pathology in the dystrophin-deficient mdx mouse. We found that long-term treatment (32 weeks) with resveratrol decreased ROS levels and attenuated the muscle loss, myofibroblast differentiation, and abnormal endomysial fibrosis in the mdx mice. Our data suggest that SIRT1 activation may be a useful approach for patients with muscular dystrophies.
Materials and Methods
All in vivo experiments were conducted in strict accordance with The Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996) and approved by the Animal Use Committee of Sapporo Medical University. Male C57BL/10ScSn-Dmdmdx/J mice (mdx mice) and their control C57BL/10 mice were purchased from Oriental Yeast Co. Ltd. (Tokyo, Japan). Resveratrol was mixed with powdered meal (4 g/kg meal) and orally administered ad libitum to mice for 32 weeks from the age of 9 weeks. The mice were then sacrificed, and the biceps femoris muscles were examined.
Reagents and Antibodies.
Resveratrol, TGF-β1, and Hoechst 33342 were purchased from Wako Pure Chemicals (Osaka, Japan). Phalloidin was from Sigma-Aldrich (St. Louis, MO). The antibodies used were anti-acetyl-histone H3 rabbit polyclonal (Calbiochem, San Diego, CA), anti-SIRT1 rabbit polyclonal (Sakamoto et al., 2004), anti-nitrotyrosine mouse monoclonal, anti-Sod2 rabbit polyclonal (Millipore, Billerica, MA), anti-8-hydroxy-2′-deoxyguanosine (8-OHdG) mouse monoclonal (OXIS Research, Inc., Portland, OR), anti-CD45 rabbit polyclonal, anti-p47phox rabbit polyclonal, anti-NADPH oxidase 4 (Nox4) goat polyclonal (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), anti-fibronectin sheep polyclonal (Serotec, Oxford, UK), anti-collagen type III rabbit polyclonal (LSL, Tokyo, Japan), anti-α-smooth muscle actin (α-SMA) mouse monoclonal, anti-GAPDH mouse monoclonal, and anti-α-tubulin mouse monoclonal (Sigma-Aldrich) antibodies. Other reagents were from Wako Pure Chemicals or Sigma-Aldrich.
Immunostaining was performed as described previously (Tanno et al., 2010). After fixation with 4% paraformaldehyde, samples were blocked with phosphate-buffered saline (PBS) containing 3% bovine serum albumin, 1% goat serum, and 0.1% Triton X-100 for 30 min. The samples were then incubated with antibodies against nitrotyrosine (3-nitro-l-tyrosine; 1:200 dilution), 8-OHdG (1:50 dilution), CD45 (1:50 dilution), α-SMA (1:100 dilution), fibronectin (1: 500 dilution), collagen type III (1:100 dilution), Sod2 (1:500 dilution), acetyl-histone H3 (1:500 dilution), or SIRT1 (1:500 dilution) overnight at 4°C. The samples were then washed four times with PBS and incubated with secondary antibodies (1:1000 dilution) overnight at 4°C. Hoechst 33342 (1:1000 dilution) and phalloidin (1:50 dilution) were used to stain the nucleus and F-actin, respectively. After washing with PBS, the samples were mounted with VECTASHIELD (Vector Laboratories, Burlingame, CA). To compare the expression levels, the fluorescence intensity of an image was quantified by Image J software (National Institutes of Health, Bethesda, MD) and compared among experimental groups. Eight independent images of sections of biceps femoris muscles from each mouse were examined, and the data from three mice in each group were compared.
Quantitative Reverse Transcription-Polymerase Chain Reaction.
Total RNA was isolated with the RNeasy Fibrous Tissue Mini Kit for muscle tissues or the RNeasy Mini Kit for cultured cells (QIAGEN, Valencia, CA). First-strand cDNA was synthesized using SuperScript III (Invitrogen, Carlsbad, CA). DNA amplification was performed in StepOne (Applied Biosystems, Foster City, CA) using TaqMan PCR kits for SIRT1, Sod2, β-actin, and GAPDH or the Power SYBR Green Master Mix for other genes (Applied Biosystems). The primer sequences are listed in Supplemental Table 1.
Samples were homogenized in CelLytic M Cell Lysis Reagent (Sigma-Aldrich) with 1% protease inhibitor cocktail (Nacalai Tesque, Kyoto, Japan) and centrifuged at 10,000g for 10 min at 4°C. The protein concentration of the supernatant was measured using the Protein Quantification Kit-Rapid (Dojindo, Kumamoto, Japan). Supernatant fractions of equal protein concentration were analyzed by Western blotting as described previously (Tanno et al., 2010). The antibodies used were anti-p47phox (1:500), anti-Nox4 (1:200), anti-GAPDH (1:10,000), anti-SIRT1 (1:500), antiacetyl-histone H3 (1:1000), and anti-α-SMA (1:400).
Creatine Kinase and Lactate Dehydrogenase Activities.
Blood samples obtained by cardiac puncture before the removal of muscular tissues were incubated at room temperature for 1 to 2 h to allow clotting. The samples were then centrifuged at 1000g for 10 min, and the supernatants (serum) were stored at −80°C until use. The CK and lactate dehydrogenase (LDH) activities were assayed by the Nagahama Life Science Laboratory of the Oriental Yeast Co. Ltd. (Shiga, Japan), and the results were expressed as units/liter.
C2C12 myoblast cells and L929 fibroblast cells were cultured with Dulbecco's modified Eagle's medium (Wako Pure Chemicals) supplemented with 1% antibiotic-antimycotic mixed stock solution (Nacalai Tesque) and 10% fetal bovine serum (MP Biomedicals, Solon, OH). Cells were seeded on glass slides for immunocytochemical studies and ROS measurement.
Electroporation of Small Interfering RNA.
siRNAs against SIRT1 and control siRNAs were obtained from Sigma Genosys (Cambridge, UK). The siRNAs (100 nM) were electroporated into C2C12 cells twice, 24 h apart, using a Nucleofector kit (Lonza, Basel, Switzerland). Twenty-four hours after the second transfection, the cells were used for experiments.
Effects of Resveratrol on Cells.
Cells were pretreated with resveratrol (30 μM) or vehicle dimethyl sulfoxide for 3 h without serum, then TGF-β1 (10 ng/ml) was added to the medium and the cells were further incubated for 24 h. After washing with PBS, the cells were harvested for quantitative reverse transcription-polymerase chain reaction (qRT-PCR). For immunocytochemistry, the cells were fixed with 4% paraformaldehyde for 10 min.
Analysis of Intracellular ROS Levels.
To measure the intracellular ROS levels, cells were incubated with 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate acetyl ester (CM-H2DCFDA) (Invitrogen) for 30 min at 37°C and washed twice with PBS. The green fluorescence intensity of the oxidized DCF probe per cell was quantified by Image J software and compared among experimental groups. Eight fields of C2C12 cells cultured on a glass disk were examined, and the data from three independent experiments were compared.
Oil Red O Staining.
Frozen sections were washed with PBS twice and fixed with 10% paraformaldehyde for 10 min. After washing with PBS, the sections were washed with 60% isopropyl alcohol and stained with freshly prepared 0.18% Oil Red O (1-[[2,5-dimethyl-4-[(2-methylphenyl)azo]phenyl]azo]-2-naphthalenol) in 60% isopropyl alcohol for 20 min. The sections were then rinsed with 60% isopropanol and then with PBS.
Results are presented as means ± S.E.M. and analyzed by Student's two-tailed t test for comparisons between two groups or one-way analysis of variance and the Student-Neuman-Keuls post hoc test for comparisons of three groups. The difference was considered significant if the probability value was <0.05. All analyses were carried out using SigmaStat (Systat Software, Inc., San Jose, CA).
Preservation of Muscle Mass by Resveratrol.
Actin staining by phalloidin of the biceps femoris muscles showed a loss of muscle fibers and enlarged interstitial spaces in untreated mdx mice compared with those of control C57BL/10 mice (Fig. 1, A and B). The administration of resveratrol significantly suppressed the myofiber loss and decreased the nonmuscle interstitial tissue compared with those of untreated mdx mice (Fig. 1, A and B). qRT-PCR showed that the mRNA levels of myosin heavy chain (MHC) 1, MHC2b, and troponins in the biceps femoris were up-regulated by the administration of resveratrol (Fig. 1, C and D). Skeletal muscles can be divided into type I slow and type II fast fibers. Type I fibers, consisting of proteins such as MHC1 and slow troponin, are able to generate ATP efficiently by oxidative phosphorylation and appear red, whereas glycolytic type II fibers, consisting of proteins such as MHC2b and fast troponin, are white in appearance. The mRNA levels of MHC1 and slow troponin in the muscle from mdx mice fed resveratrol were 5- and 4-fold higher than those in untreated mdx mice (Fig. 1, C and D). On the other hand, the increase in the mRNA levels of MHC2b and fast troponin were only approximately 2-fold in the resveratrol-treated mdx mice compared with control mdx mice (Fig. 1, C and D), suggesting that resveratrol affects fiber-type composition of muscles. Consistent with this result, the biceps femoris from the resveratrol- treated mdx mice appeared more red than that of the control mdx mice (data not shown). However, the number of myofibers in the biceps femoris muscles with central nuclei, indicating newly generated myofibers, was similar in the resveratrol-treated and untreated mdx mice (see Fig. 6A and Supplemental Fig. S1A). In addition, the increases in serum CK and LDH levels in the mdx mice were not suppressed by resveratrol treatment (Supplemental Fig. S1, B and C).
Inhibition of Oxidative Stress and Fibrosis by Resveratrol.
The biceps femoris of untreated mdx mice displayed high immunoreactivity against antinitrotyrosine and anti-8-OHdG antibodies in the interstitial spaces and nuclei, respectively (Fig. 1, E–H), indicating there were high levels of oxidative stress in the muscle tissue. Resveratrol significantly reduced the nitrotyrosine and 8-OHdG immunoreactivities (Fig. 1, E–H).
We next examined the effect of resveratrol on the fibrosis in mdx muscles. The biceps femoris from untreated mdx mice typically showed increased fibrosis in the interstitial spaces of the muscle, indicated by high immunoreactivities against anti-collagen type III and anti-fibronectin antibodies (Fig. 2, A–D). In contrast, resveratrol treatment significantly suppressed the fibrosis and reduced the interstitial spaces (Fig. 2, A–D). The fibronectin, collagen type Ia1, and collagen type Ia2 mRNA levels were significantly higher in the muscles of the mdx mice than in those of control mice. Resveratrol significantly inhibited the increases in these mRNA levels (Fig. 2, E–G). Myofibroblasts, which may be derived from fibroblasts, are principally involved in the progression of fibrotic events (Barnes and Gorin, 2011). Immunostaining for the myofibroblast marker, α-SMA, showed that α-SMA was expressed only in the muscular blood vessels of control C57BL/10 mice, whereas it was also highly expressed in the interstitial tissues of muscles of mdx mice, suggesting that the myofibroblasts promoted the interstitial fibrosis in the mdx mice (Fig. 2, H and I). Resveratrol significantly decreased the immunoreactivity against the α-SMA antibody in the muscles of the mdx mice (Fig. 2, H and I). In addition, pseudo-hypertrophy of the muscle by adipocyte proliferation is often found in DMD, but no fat deposition was detected by Oil Red O staining in the biceps femoris of the treated or untreated mdx mice (Supplemental Fig. S1D).
Effect of Resveratrol on Inflammatory Responses.
Inflammatory cells are often found in the muscle of patients with DMD (Emery, 2002). To examine whether the infiltration of inflammatory cells was influenced by resveratrol treatment, we examined the biceps femoris muscles with an antibody against CD45, a leukocyte common antigen. Abundant CD45+ cells were present in the muscles of resveratrol-treated and untreated mdx mice compared with those of control C57BL/10 mice (Fig. 3A). The CD45 mRNA level was also higher in the biceps femoris muscle of the mdx mice than in that of control C57BL/10 mice, and resveratrol treatment did not reduce the CD45 mRNA expression (Fig. 3B). In addition, the mRNA levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), TGF-β1, and TGF-β2 were up-regulated in the muscle of mdx mice compared with that of control mice, and resveratrol treatment failed to reduce these cytokines (Fig. 3, C–F). The expression level of TGF-β3 was similar among these three groups (Fig. 3G). Thus, resveratrol did not block the infiltration of leukocytes or cytokine release in the mdx muscle.
Effect of Resveratrol on ROS-Producing and Detoxifying Enzymes.
Nox is an enzyme complex, consisting of a transmembrane catalytic protein Nox or Duox and its regulatory subunits such as p22phox, that generates ROS by transferring an electron from NADPH to molecular O2 (Lambeth et al., 2007). The Nox family has been shown to be involved in the ROS production in the muscle of mdx mice (Whitehead et al., 2010). qRT-PCR showed that the Nox1, Nox4, Duox1, p22phox, and p47phox mRNA levels were up-regulated in the biceps femoris of mdx mice compared with those of control mice. Resveratrol significantly suppressed the induction of the Nox4, Duox1, and p47phox mRNAs (Fig. 4), but not that of the Nox1 and p22phox mRNAs (Fig. 4, A and F). The Duox2 mRNA level was not changed among these three groups (Fig. 5E), and Nox3 expression was not detected in the biceps femoris of any group (data not shown).
Previously, we found that the ROS-detoxifying enzyme Sod2 (Mn-Sod) was induced by resveratrol in the cardiomyocytes from newborn rats and in the myocardium of TO-2 hamsters (Tanno et al., 2010). We examined the expression levels of the related Sod enzymes in the biceps femoris, and although the Sod1 mRNA level was higher in the mdx mice treated with resveratrol, the levels of Sod2 nor Sod3 mRNA were not (Fig. 5, A–C). Immunostaining also showed that the level of Sod2 protein in the biceps femoris was similar among these three groups (Fig. 5, D and E). We also examined the mRNA levels of other ROS-detoxifying enzymes heme oxygenase 1 (Hmox1) and NAD(P)H quinone oxidoreductase 1 (Nqo1), and components of the rate-limiting enzyme for glutathione biosynthesis; glutamylcysteine ligase modulator (Gclm) and glutamylcysteine ligase catalytic subunit (Gclc). These proteins are induced by various types of cell stress via the activation of transcription factor Nrf2 signaling. The levels of Hmox1 and Nqo1 mRNA were similar in the muscles of control and mdx mice, and resveratrol did not affect their expression (Fig. 5, F and G). Gclm and Gclc were down-regulated in the mdx mice compared with control mice, and resveratrol did not induce their expression (Fig. 5, H and I).
Effect of Resveratrol on SIRT1 and Acetyl Histone H3.
Resveratrol is an activator of SIRT1 (Howitz et al., 2003). We next examined the localization of SIRT1 in the biceps femoris, because the nuclear localization of SIRT1 is necessary for resveratrol's cell-protective function (Tanno et al., 2010). Sarcolemmal membranes of the control and mdx mouse muscles showed high immunoreactivity against the anti-SIRT1 antibody, and no significant change was induced by the administration of resveratrol (Fig. 6A). In addition, the nuclear SIRT1 expression levels in muscle fibers with central nuclei (Fig. 6A, arrowheads) were much lower than those with peripheral nuclei (Fig. 6A, arrows). The mRNA level of SIRT1 as well as SIRT3, another sirtuin located mainly in mitochondria, was similar among these groups (Fig. 6, C and D). Because SIRT1 is a histone deacetylase, we also examined the acetylation level of histone H3 at lysine 9 and lysine 14 by immunostaining of the biceps femoris. The nuclei of mdx mice showed higher acetyl histone H3 levels compared with those of control mice, and the administration of resveratrol significantly reduced the acetylation level (Fig. 6B), suggesting that SIRT1 was activated by resveratrol in the muscle of the mdx mice.
Involvement of SIRT1 in the Function of Resveratrol.
Resveratrol decreased the ROS levels, fibrosis, and myofibroblastic changes, but it did not suppress the infiltration of inflammatory cells or TGF-β1 production. In C2C12 myoblast cells, TGF-β1 up-regulated the ROS levels, fibronectin synthesis, and α-SMA expression, indicating that TGF-β1 gave rise to oxidative damage and induced fibrotic changes in these cells (Fig. 7). Pretreatment with resveratrol reduced the ROS levels (Fig. 7, A and B) and suppressed fibronectin synthesis (Fig. 7, E and F) and α-SMA expression (Fig. 7, G and H) triggered by TGF-β1 in C2C12 cells. The up-regulation of fibronectin mRNA by TGF-β1 in C2C12 and L929 fibroblast cells was also suppressed by resveratrol (Fig. 7, C and D).
To verify the role of SIRT1, we used SIRT1 siRNA. Among three siRNAs for SIRT1, SIRT1-si(5675) effectively reduced the expression level of SIRT1 mRNA (Supplemental Fig. S2) and was used in the following experiments. Expression of the SIRT1 siRNA elevated the ROS levels and blocked resveratrol's suppressive function on ROS production in C2C12 cells (Fig. 7, A and B), indicating that SIRT1 was involved in the suppression of ROS levels and that resveratrol decreased the oxidative damage via SIRT1. The knockdown of SIRT1 also cancelled the reduction of fibronectin and α-SMA expression by resveratrol (Fig. 7, E–H). Thus, the inhibition of the myofibroblastic change and fibronectin synthesis by resveratrol also seemed to be mediated by SIRT1 in C2C12 cells.
In the biceps femoris of mdx mice, the α-SMA (Fig. 2, H and I) and acetyl histone H3 levels (Fig. 6B) increased and Nox4 and p47phox were up-regulated (Fig. 4). We found that SIRT1 knockdown with SIRT1 siRNA in C2C12 cells significantly increased the acetyl histone H3 levels, indicating that SIRT1 promoted the deacetylation of histone H3 in these cells (Fig. 8). Treatment with SIRT1 siRNA also increased the Nox4, p47phox, and α-SMA levels, demonstrating that SIRT1 down-regulated Nox4, p47phox, and α-SMA expression (Fig. 8).
Long-term treatment of mdx mice with resveratrol reduced myofiber loss (Fig. 1, A–D) accompanied by decreased levels of ROS (Fig. 1, E–H), myofibroblast cells (Fig. 2, H and I), and fibrosis (Fig. 2, A–G) in the muscles. Resveratrol treatment did not affect the infiltration of inflammatory cells or TGF-β1 cytokine release (Fig. 3). This is the first report to demonstrate the significant therapeutic efficacy of a SIRT1 activator, resveratrol, in dystrophin-deficient mdx mice. Our results also suggest that targeting SIRT1 with pharmacological activators might improve the muscle pathology in patients with muscular dystrophies.
In the biceps femoris of mdx mice, the nitrotyrosine and 8-OHdG levels were high compared with those of control mice, and the administration of resveratrol significantly reduced the oxidative damage (Fig. 2). In addition, the expression levels of some of the subunits of NADPH oxidase, which is involved in ROS production, increased in the biceps femoris of mdx mice, and resveratrol significantly suppressed the levels of Nox4, Duox1, and p47phox (Fig. 4). Previous studies reported the up-regulation of Nox2 in the tibialis anterior muscle (Whitehead et al., 2010) and Nox4 in the cardiac muscle (Spurney et al., 2008) of mdx mice. The discrepancy in the increased Nox subunits might be caused by differences in the muscle examined and the age of the mice. Duox1 is expressed in the thyroid gland (Lambeth et al., 2007), and its function in the skeletal muscle is unknown. Nox4 is able to produce ROS without cytosolic regulatory subunits (Lambeth et al., 2007), suggesting that its high expression level may directly increase the ROS level in muscle. Because Nox4 is reported to be involved in several diseases, including heart failure (Kuroda et al., 2010), ischemic stroke (Kleinschnitz et al., 2010), and atherosclerosis (Tong et al., 2010), the decrease in ROS levels by resveratrol treatment may be caused in part by the reduction of Nox4 expression in these muscle tissues. Because p47phox is a regulator of Nox2 activity (Lambeth et al., 2007), the inhibition of p47phox up-regulation might also contribute to decreased ROS production in muscle. SIRT1 knockdown increased the expression levels of Nox4 and p47phox (Fig. 8A), indicating that resveratrol decreases the Nox4 and p47phox levels via SIRT1 activation in the biceps femoris of mdx mice.
TGF-β is a multifunctional cytokine that regulates diverse cellular processes, including proliferation, differentiation, apoptosis, and the immune response. We showed that resveratrol decreased ROS levels and fibrosis, but it did not reduce the TGF-β1 level in mdx mice. Our in vitro experiments showed that TGF-β1 induced ROS production, myofibroblast differentiation, and fibrotic change in C2C12 cells and resveratrol significantly suppressed these changes via SIRT1 (Fig. 7). Thus, the suppression of TGF-β1 signaling by SIRT1 may be an important mechanism for resveratrol's activity. TGF-β1 promotes the acetylation of Smad3, a critical mediator of the TGF-β1 signal, thereby increasing Smad3's activity (Inoue et al., 2007). In contrast, SIRT1 deacetylates and inhibits Smad3 (Li et al., 2010), and thus the induced deacetylation of Smad3 by resveratrol may be involved in the suppression of TGF-β1 function in mdx mice. TGF-β1 also induces acetylation of the Rel/p65 subunit of NF-κB and activates NF-κB (Ishinaga et al., 2007). Because SIRT1 inhibits NF-κB by deacetylating Rel/p65 (Yeung et al., 2004), the inhibition of NF-κB by resveratrol via SIRT1 may also have a role in suppressing the signals of TGF-β1 and other cytokines elevated in the muscle of mdx mice. In addition, the infiltration and activation of macrophages in muscle promotes ROS production and has been suggested to be critical for muscular degeneration in DMD (Desguerre et al., 2009). Because resveratrol suppresses the activity of macrophages via SIRT1 activation (Yoshizaki et al., 2010), it may also stabilize macrophages and inhibit inflammation.
Resveratrol has been shown to prevent cardiac fibrosis in cardiomyopathic hamsters (Tanno et al., 2010) and the renal fibrosis of obstructed mouse kidneys (Li et al., 2010). In the present study, treatment with resveratrol also suppressed the muscular endomysial fibrosis of mdx mice (Fig. 2). Desguerre et al. (2009) examined the relationship between the long-term outcome and initial histological alterations for myofiber atrophy, necrosis, fatty degeneration, and endomysial fibrosis in muscle biopsies from patients with DMD. It is noteworthy that they found that only endomysial fibrosis was correlated with poor motor outcome and with the age at which loss of ambulation occurred (Desguerre et al., 2009). This finding suggests that the suppression of muscle fibrosis by resveratrol may aid in maintaining the motor function of patients with dystrophy.
Differentiation into myofibroblast cells is closely related to fibrosis (Barnes and Gorin, 2011). The apparent reductions in fibrosis and α-SMA expression levels by resveratrol treatment in mdx muscle (Fig. 2) and C2C12 cells (Fig. 7) suggests that the inhibition of myofibroblast differentiation is an important mechanism for the antifibrotic action of resveratrol. Nox4 expression has been considered a critical mediator of myofibroblast differentiation and tissue fibrosis (Barnes and Gorin, 2011). Nox4 expression is induced by TGF-β via Smad3 activation (Lambeth et al., 2007; Hecker et al., 2009), whereas the knockdown of Nox4 suppresses TGF-β-induced α-SMA and fibronectin expression in cardiac and lung fibroblasts (Cucoranu et al., 2005; Hecker et al., 2009). These results indicate that the suppression of Nox4 by resveratrol contributes to reduced ROS levels, myofibroblast differentiation, and fibrosis in the muscle of mdx mice.
The expression levels of the cytoprotective ROS-detoxifying enzymes Sod1, Sod2, and Sod3 were not reduced in mdx mice compared with control C57BL/10 mice, and the administration of resveratrol only induced Sod1 (Figs. 4 and 5). In the cardiomyocytes of TO-2 hamsters, resveratrol significantly induces Sod2 (Tanno et al., 2010) (Supplemental Fig. S3); the discrepancy may be owed to the different tissues analyzed. Alternatively, down-regulation of the Sod2 gene by TGF-β (Herrera et al., 2004) may compensate the resveratrol's function on Sod2 transactivation. Previously, we found that SIRT1's cytoprotective effect against oxidative stress is derived, not from its cytoplasmic expression, but from its nuclear expression, which increases the Sod2 level (Tanno et al., 2010). Because the nuclear localization of SIRT1 is necessary for the induction of Sod2, the lower expression level of SIRT1 in centrally nucleated premature fibers compared with that in mature myocytes (Fig. 6A) may limit the efficacy of resveratrol. Severe stress seems to induce nuclear translocation of SIRT1, but the stress in mdx mice may not be sufficient for nuclear localization of SIRT1. The development and coadministration of an agent that promotes the translocation of SIRT1 from the cytoplasm to the nucleus of a cell may be important for increasing resveratrol's activity.
Nevertheless, we found that resveratrol activated SIRT1-dependent signal cascades, for example, to inhibit TGF-β1 signaling and decrease the acetyl histone H3 level (Fig. 6B) in the muscle of mdx mice. The increase in slow muscle type I fibers relative to fast-twitch fibers in the biceps femoris of resveratrol-treated mdx mice (Fig. 1, C and D) also suggests that resveratrol activated SIRT1 and promoted the deacetylation of PGC-1α, which plays central roles in muscle type switching from type II fibers to type I fibers (Handschin et al., 2007). SIRT1 affects differentiation by regulating transcription factors such as MyoD, FOXOs, and NF-κB (Fulco et al., 2003; also see review by Horio et al., 2011). Modulation of muscular differentiation and regeneration by resveratrol may also contribute to the suppression of myofiber loss (Fig. 1).
Although resveratrol increased the muscle mass (Fig. 1), the high-serum CK and LDH levels (Supplemental Fig. S1) indicated that muscle injuries still proceeded in the resveratrol-treated mdx mice. Resveratrol has multiple targets; it is known as a mitochondrial depolarizing agent and an inhibitor of tumor cell growth (Dörrie et al., 2001). The dose of resveratrol used in the present study was 4 g/kg diet, which is the same as in our previous study using the TO-2 hamster (Tanno et al., 2010). The amount of resveratrol taken in by the TO-2 hamsters with heart failure is estimated to be approximately 80 mg/kg/day (Tanno et al., 2010), whereas that of the mdx mice in our current study was approximately 500 mg/kg/day. This 6-fold increase in daily resveratrol intake in the mdx mice compared with the TO-2 hamsters may have affected the experimental comparisons. Furthermore, we used a long-term administration for 32 weeks, which may have enhanced the cytotoxic effect of resveratrol. Because a much lower dose of resveratrol (2.5 mg/kg/day) is reported to be effective for improving the insulin sensitivity of mice fed a high-fat diet (Sun et al., 2007), investigations using lower doses of resveratrol will help clarify the optimum dose for treating mdx mice.
Unfortunately, we could not analyze the functional parameters in the mdx mice. However, clinical trials investigating the safety and efficacy of resveratrol have already started in patients with metabolic syndrome and type 2 diabetes (http://clinicaltrials.gov/). Cumulative data exist for safe dosage levels, pharmacokinetics, and adverse effects of resveratrol (Patel et al., 2011). Thus, the use of resveratrol in patients with muscular dystrophies may be acceptable in the clinical setting.
Participated in research design: Hori, Kuno, Tanno, Miura, Shimamoto and Horio.
Conducted experiments: Hori, Kuno, and Hosoda.
Performed data analysis: Hori, Kuno, and Horio.
Wrote or contributed to the writing of the manuscript: Hori, Kuno, and Horio.
This work was supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan [Grant-in-Aid 22590245]; a National Project of Knowledge Cluster Initiative of Second Stage; Sapporo Biocluster Bio-S; the Program for Developing the Supporting System for Upgrading the Education and Research; and the Akiyama Life Science Foundation.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- Duchenne muscular dystrophy
- α-smooth muscle actin
- creatine kinase
- 5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate acetyl ester
- glyceraldehyde-3-phosphate dehydrogenase
- lactate dehydrogenase
- NADPH oxidase
- phosphate-buffered saline
- quantitative reverse transcription-polymerase chain reaction
- peroxisome proliferator-activated receptor-γ coactivator 1α
- reactive oxygen species
- small interfering RNA
- superoxide dismutase
- transforming growth factor-β
- tumor necrosis factor-α
- myosin heavy chain
- heme oxygenase 1
- NAD(P)H quinone oxidoreductase 1
- glutamylcysteine ligase modulator
- glutamylcysteine ligase catalytic subunit
- not significant
- nuclear factor κB
- Received April 21, 2011.
- Accepted June 6, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics