Myostatin (Mstn) is a secreted growth factor belonging to the tranforming growth factor (TGF)-β superfamily. Inactivation of murine Mstn by gene targeting, or natural mutation of bovine or human Mstn, induces the double muscling (DM) phenotype. In DM cattle, Mstn deficiency increases fast glycolytic (type IIB) fiber formation in the biceps femoris (BF) muscle. Using Mstn null (−/−) mice, we suggest a possible mechanism behind Mstn-mediated fiber-type diversity. Histological analysis revealed increased type IIB fibers with a concomitant decrease in type IIA and type I fibers in the Mstn−/− tibialis anterior and BF muscle. Functional electrical stimulation of Mstn−/− BF revealed increased fatigue susceptibility, supporting increased type IIB fiber content. Given the role of myocyte enhancer factor 2 (MEF2) in oxidative type I fiber formation, MEF2 levels in Mstn−/− tissue were quantified. Results revealed reduced MEF2C protein in Mstn−/− muscle and myoblast nuclear extracts. Reduced MEF2-DNA complex was also observed in electrophoretic mobility-shift assay using Mstn−/− nuclear extracts. Furthermore, reduced expression of MEF2 downstream target genes MLC1F and calcineurin were found in Mstn−/− muscle. Conversely, Mstn addition was sufficient to directly upregulate MLC promoter-enhancer activity in cultured myoblasts. Since high MyoD levels are seen in fast fibers, we analyzed MyoD levels in the muscle. In contrast to MEF2C, MyoD levels were increased in Mstn−/− muscle. Together, these results suggest that while Mstn positively regulates MEF2C levels, it negatively regulates MyoD expression in muscle. We propose that Mstn could regulate fiber-type composition by regulating the expression of MEF2C and MyoD during myogenesis.
- myocyte enhancer factor 2
- myosin heavy chain
myostatin (Mstn) is a secreted growth factor that belongs to the transforming growth factor (TGF)-β superfamily (21). It has been shown to be a potent negative regulator of myogenesis, since lack of functional Mstn results in heavy muscle growth due to hyperplasia (21). In contrast, increased levels of Mstn in the circulation result in cachectic muscle wasting (46). Mstn is expressed in the muscle as a 375 amino acid peptide, which is proteolytically processed to give rise to a 110-amino-acid-long mature Mstn (COOH-terminal) and an NH2-terminal latency-associated peptide (38). The secreted mature Mstn functions by binding to the activin type IIB receptor and signaling through a Smad signaling pathway (17).
Mstn expression is predominately detected in skeletal muscle (21). During muscle development, Mstn expression is first detected in the somites of E9.5 embryos and continues in the adult muscle (21). Hence, it is believed that Mstn functions both pre- and postnatally. Using a myoblast culture system, it has been shown that Mstn negatively regulates both proliferation and differentiation processes. Mstn controls the cell cycle progression of myoblasts by regulating the G1 to S phase transition via an Rb-dependent pathway and the differentiation of myoblasts by controlling MyoD activity (16, 38).
Recently, a number of published studies have clearly proposed a postnatal function for Mstn protein. In disease conditions such as HIV infection, which leads to heavy muscle wasting, increased level of Mstn was observed in circulation (12). Similarly, mice undergoing atrophy induced by unloading also showed increased levels of Mstn (39). Postnatally, Mstn functions by controlling satellite cell behavior. Recently published results from our laboratory indicate that Mstn maintains the quiescence of satellite cells (18). Hence, in the absence of Mstn there is increased activation of satellite cells. In addition, lack of Mstn also leads to increased self-renewal of satellite cells (18). As a result, when Mstn −/− mice were injured with notexin, there was more efficient regeneration of muscle compared with the wild-type muscle (19). Thus Mstn controls postnatal muscle growth by regulating satellite cell activation and number.
Another major event in postnatal myogenesis is the generation of skeletal muscle fiber-type diversity. The muscle fiber-type diversity is in part distinguished by the expression of various muscle protein isoforms [e.g., myosin heavy chain (MHC), myosin light chain (MLC), sarcoplasmic reticulum ATPase, and lactate dehydrogenase isoforms] that influence the fiber's contractile and metabolic properties (37). At present, the intracellular pathways that transduce physiological stimuli into molecular signals that alter phenotypic gene expression are largely unknown. Several reports suggest that MEF2 (a MADS box transcription factor) and calcineurin (a calcium-regulated 2B protein phosphatase) may be responsible for the formation of slow-twitch fibers (42–44). Furthermore, using gain and loss of function analysis in mice, very recently it has been shown that whereas expressing hyperactive form of MEF2 in transgenic mice promotes slow-fiber formation, inactivation of MEF2 (by specifically increasing class II HDAC proteins) leads to loss of slow-fiber formation (27). In addition to MEF2, Schiaffino's lab (20) has suggested that nerve activity via calcineurin/nuclear factor of activated T cells (NFAT) signaling pathway controls the slow-fiber formation. On the other hand, MyoD specifically accumulates in fast type IIB/IIX fiber type (14) thereby leading to a hypothesis that MyoD is required for fast-fiber formation. Consistent with this hypothesis, it is recently shown that conditional ablation of Fox01 expression (repressor of MyoD expression) in the soleus muscle leads to reduced type I fiber and concomitant increase in type II fiber formation confirming that MyoD expression is required for the fast-fiber formation (15).
In addition to controlling muscle mass, Mstn also appears to regulate muscle fiber-type composition postnatally. In cattle lacking Mstn and in Mstn−/− mice there is an increased number of fast glycolytic fibers (11, 35). However, to date the mechanism behind this change in composition is unknown. Using Mstn−/− mice and molecular analysis, here we suggest that Mstn could accomplish the fiber-type diversity by regulating the expression of both MyoD and MEF2C genes.
Mstn−/− mice (C57BL/6 background) were obtained from Se-Jin Lee (Johns Hopkins University, Baltimore, MD). Wild-type C57BL/6 mice were bred at the Ruakura Small Animal Colony. All animals were housed under the same conditions with ad libitum access to food and water. All animals were handled in accordance with the guidelines of the Ruakura animal ethics committee (AgResearch, Hamilton, New Zealand).
Six-month-old male wild-type and Mstn−/− mice were euthanized by CO2 inhalation followed by cervical dislocation. Biceps femoris (BF) and tibialis anterior (TA) muscles were dissected out and frozen in isopentane cooled in liquid nitrogen. Transverse sections (10 μm) were cut from the midbelly region of each muscle and immunostained using monoclonal antibodies specific for MHCs. Briefly, sections were blocked in 0.2% BSA and 10% normal sheep serum (NSS) in PBS-T with 0.2% Tween 20 for 1 h at room temperature before being incubated overnight at 4°C with type I (BA-D5, 1:100), IIA (SC-71, 1:50), or IIB (BF-F3, 1:100) (31) MHC antibodies in 0.2% BSA and 5% NSS in PBS-T. The sections were washed with PBS and fixed in 10% buffered Formalin for 5 min followed by an additional PBS wash. A biotinylated sheep anti-mouse immunoglobulin secondary antibody at 1:300 (Amersham Biosciences) in 0.2% BSA + 5% NSS in PBS-T was then added for 1 h at room temperature. The sections were then washed and incubated with streptavidin Alexa Fluor 488 conjugate at 1:400 (Molecular Probes) in 0.2% BSA in PBS-T for 1 h. Sections were then washed and counterstained with DAPI at 1:1,000 (Molecular Probes) in PBS before being mounted with fluorescent mounting medium (Dako) and viewing under fluorescent illumination using an Olympus BX50 microscope and SPOT-RT 4.01 camera and software. Counting of the various fiber types was performed over the entire muscle cross-sections as to establish the number of each fiber type in relation to the total fiber number. Approximately 4,000 to 8,000 fibers per section were counted for this analysis. A total of 10 BF and TA muscles collected from five wild-type and five Mstn−/− mice were analyzed.
Nuclear extracts were prepared as follows. Primary myoblasts from 6-wk-old wild-type and Mstn−/− mice were cultured according to previously published methods (2, 24, 45). After 48 h incubation, two million cells of each genotype were rinsed in PBS at room temperature and harvested by scraping after addition of buffer A [in mM: 10 HEPES (pH 7.9), 1.5 MgCl2, 10 KCl, and 0.5 dithiothreitol (DTT)]. Cells were homogenized by passing 10 times through a 26-gauge needle, followed by centrifugation at 3,000 rpm. Nuclear pellets were resuspended in buffer A, rehomogenized, and centrifuged as described above. The nuclear pellet was then resuspended in buffer B [20 mM HEPES (pH 7.9), 25% glycerol, 420 mM KCl, 0.2 mM EDTA, and 1 mM DTT] and left on ice for 20 min before being homogenized as above. Extracts were then spun at high speed for 20 min before the supernatant was recovered. Bradford Reagent was used to estimate total protein.
The enhancer fragment used to bind MEF2 in the nuclear extracts corresponded to the MEF2 binding site within the MLC-1/3 enhancer. Sense (S) and anti-sense (AS) oligonucleotides were custom synthesized (Invitrogen), and sequences were as follows: S-5′GATCCTCATCTTTTAAAAATAACTTTTCAAAAG3′ and AS-5′CTTTTGAAAAGTTA TTTTTAAAAGATGGATC3′ (13). Oligos were individually end labeled with [32P]dATP then mixed in equal volumes, heated to 95°C, and left to cool and anneal. A 5-μg sample of nuclear extract was incubated with 2 μl of 10× binding buffer [in mM: 40 KCl, 15 HEPES (pH 7.9), 1 EDTA, and 0.5 DTT], 0.5 μg of poly(dI-dC), and 0.5% fetal bovine serum (FBS). Extracts were then incubated with various concentrations of unlabeled enhancer fragment (competitor) in a final volume of 19 μl for 20 min at room temperature. After this incubation, 1 μl (10 pmol) of [32P]dATP-labeled enhancer fragment was added, and the samples were incubated for an additional 20 min at room temperature. The protein-DNA complex was resolved on 5% native polyacrylamide gel as described previously (33).
Western Blot analysis
Muscles were resuspended in 500 μl of lysis buffer [50 mM Tris (pH 7.6), 250 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40, Complete protease inhibitor (Roche Molecular Biochemicals)], and homogenized. Lysates were centrifuged to pellet cell debris, and the supernatant was recovered. Bradford reagent (Bio-Rad) was used to estimate total protein content. Muscle lysates (15 μg) and nuclear extracts (5 μg) were separated on 4–12% SDS-PAGE gels (Invitrogen) and transferred to nitrocellulose membrane (Bio-Rad) by electroblotting. The Western blot analysis was performed according to the previously published protocol (38). The following primary antibodies were used for immunoblotting: 1:10,000 diluted monoclonal mouse anti-MyoD (554130; BD), 1:400 diluted polyclonal rabbit anti-MEF2 [sc-313; Santa Cruz Biotechnology (SCBT)], 1:4,000 diluted monoclonal mouse anti-tubulin (T-9026; Sigma), and 1:10,000 diluted mouse monoclonal anti-GAPDH (RDI-TRK5G4-6C5; Research Diagnostics). The densitometric analysis of the Western blot was performed using a Bio-Rad GS-800 densitometer and Quantity One 4.4.0 software.
Northern Blot Analysis
Total RNA was extracted from BF muscle of double muscled (Belgian Blue) and normal muscled (Friesian) cattle from day 210 fetuses using TRIzol (Invitrogen) according to the manufacturer's protocol. Northern analysis was performed as previously described (30). The membranes were then exposed to autoradiographic film before densitometric analysis using a Bio-Rad GS-800 densitometer and Quantity One 4.4.0 software.
Functional Electrical Stimulation
Mice were euthanized by CO2 inhalation followed by cervical dislocation. The BF was rapidly excised and placed in a Krebs-Henseleit solution (in mmol: 118 NaCl, 4.75 KCl, 0.7 MgSO4, 24.8NaHCO3, 1.18 KH2PO4, 2.54 CaCl2, 10.0 glucose, and 20 U/l insulin) continuously aerated with carbogen (95% O2-5% CO2) at room temperature. To avoid any artefacts from nerve activity 0.025 mM of d-tubocurarine was also included.
While immersed in the physiological solution a section of the BF of ∼10 mm × 5 mm was dissected from the midbelly region with the longer length orientated with the muscle fibers, transferred to a muscle bath of volume 50 ml, and mounted longitudinally using two vessel clips. The muscle bath contained identical Krebs-Heinseleit and aerated with carbogen and kept at a constant temperature of 25°C via a water jacket, to avoid the impaired contractile function observed with temperatures in excess of 25°C, particularly in heavier muscles (5, 32).
Electrical stimulation and muscle contractile force recording.
Data acquisition and drive of the stimulation system were achieved using a personal computer configured with a Data Acquistion Card (National Instruments DAQCard-6062E). A sampling frequency of 25 kHz was chosen to be able to record the stimulation waveform simultaneously with the muscle contractile force. All systems were controlled using software written in the LabVIEW programming environment. The muscle force recording system was based on that of the Myobath I (WPI) fitted with 50 ml water-jacketed tissue baths. The myobath consists of an upper and lower stage between which the tissue sample was attached. The lower stage was held rigid while the upper stage could be raised or lowered by using a micrometer screw-thread to allow precision adjustment of passive tension and resting length of the mounted tissue sample. The contractile force was recorded using a cantilever-based force transducer (FORT 100, WPI,full-scale deflection 100 g) connected to a signal conditioner (TBM 4M, WPI). The electrical stimulator consisted of a custom-made fast power amplifier able to deliver constant voltage square waves to the isolated tissue via massive platinum electrodes spaced 8-mm apart. The stimulation system was tested and found to have a slew-rate greater than 100 V/μs.
Before the start of each experiment, the optimal length for maximum tetanic force was determined. This was achieved by performing a series of brief tetani using a unipolar pulse train of 120 Hz, high time of 80 μs and of duration 1 s, and gradually increasing the muscle resting length via the micrometer screw thread. Once the optimal length was found, a recovery period of 15 min was allowed before the experimental protocol. The stimulus voltage strength was kept constant at 35 V/cm, which had previously been found to create a supramaximal stimulus.
The fatigue stimulation protocol consisted of a series of 11 tetani of duration 5 s separated by gradually increasing resting periods of 0.25, 0.5, 1.0, 2.0, 4.0, 8.0, 16, 32, 64, and 128 s sequentially to monitor fatigue onset and recovery. The stimulation waveform consisted of unipolar square waves of frequency 120 Hz with a high-time of 80 μs. This has been shown to produce maximal contractile force in type I muscle fibers (3).
The force data were filtered using a third-order low pass Butterworth filter with a cut-off frequency of 1 kHz. The force recordings were normalized to the maximum attained peak tetanic force. The peak isometric force achieved in each tetanus contraction was used as an index of fatigue. The magnitude of fatigue was expressed as a percentage of that in the first tetanus. Additionally, comparisons of the rise-time (time taken to rise from 10% to 90% peak force) during the fatigue protocol together with the time taken for the tetanic force to fall from the peak to 50% of the maximum. As a comparison of fast and slow muscle fibers, the BF and soleus muscles from wild-type mice were compared using the same stimulation protocol.
Semiquantitative RT PCR of Calcineurin
RNA was extracted from wild-type and Mstn−/− BF muscle as described above. RT-PCR was performed to quantitatively measure calcineurin cDNA levels. First-strand cDNA was synthesized from 5 μg of total RNA using a Superscript preamplification kit (Invitrogen) according to the manufacturer's protocol. Semiquantitative PCR was performed using 2 μl of RT reaction at 94°C for 15 s, 60°C for 30 s, and 72°C for 1 min. The primers used to amplify calcineurin were 5′-GGATCCATGTCCGAGCCCAA-3′ and 5′-GGATCCTGTTTCTGATGACTTCCTTCC-3′. Primers 5′TATCATCCCTGCTTCTACTTG 3′ and 5′CTGTTGAAGTCGCAGGAGAC 3′ were used to amplify the internal control GAPDH cDNA. GAPDH cDNA was amplified using 2 μl RT reaction at 94°C for 20 s, 55°C for 40 s, and 70°C for 1 min. An 18-cycle PCR produced a linear relationship between the amount of original cDNA and the PCR-amplified calcineurin or GAPDH and therefore was used for subsequent quantitation. The PCR products were resolved on an agarose gel, transferred to a Hybond N+ membrane (Amersham), and prehybridized in a Denhardt-based hybridizing solution at 42°C for at least 30 min before hybridization with either a 32P-labeled calcineurin cDNA or GAPDH probe overnight at 42°C. The hybridized blot was washed with 2× SSC-0.5% SDS at room temperature three times for 10 min and once in 1× SSC-0.5% SDS for 10 min at 42°C. The calcineurin and GAPDH bands were visualized by exposing against audioradiographic film before densitometric analysis using a Bio-Rad GS-800 densitometer and Quantity One 4.4.0 software.
C2C12 myoblasts were grown in DMEM (Invitrogen)and buffered with 41.9 mM NaHCO3 (Sigma) and 5% gaseous CO2. Penicillin (1 × 105 IU; Sigma), streptomycin (100 mg/l; Sigma), and 10% FBS (Invitrogen) were added to media. For transfection and reporter assay analysis C2C12 myoblasts were seeded at a density of 20,000 cells/cm2.
Analysis of MLC1f Promoter-Enhancer Activity
Cloning of MLC1f promoter enhancer into a reporter vector.
The luciferase gene was mobilized as a XhoI and XbaI fragment from a pGL3-basic vector (Promega) and subcloned into the Sma1 site of a MDAF2 vector (10). The entire cassette of the MLC1f promoter-Luciferase-enhancer region was mobilized as a NotI fragment and subcloned into the NotI site of a pcDNA3 vector (Invitrogen) where the cytomegalovirus (CMV) promoter was deleted with BamH1 and Nru1 restriction enzymes before the subcloning step. This construct is designated as MLC-PE-LUC.
Regulation of the MLC1f Enhancer
To study the synergic interactions between a myogenic regulatory factor (MRF) and the MEF2C transcription factor on the regulation of the MLC1f enhancer sequence, cotransfection experiments were performed. The MLC-PE-LUC vector was cotransfected with either MyoD or MEF2C or both expression plasmids (34) into C2C12 cells using Lipofectamine 2000 (Invitrogen), according to the manufacturer's protocol. Each cotransfection was performed in triplicate. Differentiation medium (DMEM + 2% horse serum) was added after a 24-h period of transfection and left for a further 48 h before the cells were harvested. Lysis of transfected cells was performed as described previously (33). In addition, stable cell lines were selected with G418 after transfection with MLC-PE-LUC, as previously published (34). Stable cells were differentiated for 48 h and treated with 6 μg/ml of recombinant Mstn protein for a further 24 h before the cells were harvested.
The luciferase assay was performed using a Luciferase kit (Promega) according to the manufacturer's protocol. Luciferase activity in the lysates (10 μl) was measured on a Berthold luminometer, integrating light emission over 20 s. Luciferase activities were normalized to β-galactosidase expression from the transfection of the SV40-β-galactosidase control vector pCH110 (Amersham Biosciences) before treatment. β-Galactosidase assay system (Promega) was used according to the manufacturer's protocol to determine β-galactosidase activity.
Student's t-test was used to determine whether differences between two groups were significant (P < 0.05). All data are presented as means ± SE.
Percentages of fibers type are presented as means ± SE. However, data were log10 transformed for analysis of variance (general linear model, GLM) to assess the effects of the myostatin-null genotype. The pooled residual standard deviation from these analyses was used to assess particular comparisons of interest.
Mstn Regulates Fiber-Type Composition
In mammalian skeletal muscle, myofibers are mainly classified into glycolytic and oxidative fibers based on their metabolic profiles. Whereas glycolytic fibers primarily use glycolysis to generate ATP, oxidative fibers rely on oxidative phosphorylation to provide energy. In mice, fast glycolytic fibers express the type IIB MHC isoform, whereas oxidative fibers express either type I (slow fibers) or type IIA (fast) MHC. Previous reports have suggested that it is possible to transform oxidative fibers into glycolytic fibers and vice versa (25). To investigate whether Mstn has any effect on fiber-type diversity, we performed immunohistochemical and biochemical analysis on the muscle tissues from wild-type and Mstn−/− mice. As shown in Fig. 1, TA muscle from wild-type mice contained ∼6% type IIA fibers and ∼94% type IIB/X fibers, whereas in Mstn−/− mice, there was a significant increase in the percentage of type IIB fibers and concomitant decrease in type IIA fibers. To assess the effect of Mstn on slow-fiber formation, we performed immunohistochemistry on BF muscle, which not only consists of type IIA, B and X fibers, but also the slow fiber type I. As shown in Fig. 1, in Mstn−/− mice, there was a decrease in the percentage of slow fibers in the BF muscle. It is noteworthy that in the BF muscle there was also an increase in type IIB fibers and a concomitant decrease in the percentage of type IIA fibers.
MEF2C Levels are Downregulated in Mstn −/− Mice
Previously, it has been demonstrated that MEF2 proteins are selectively active in slow oxidative fibers (42). In addition, conditions that transform glycolytic fibers into oxidative fibers enhance MEF2 activity in skeletal muscle (44). The above observation that oxidative fibers were reduced in the Mstn−/− mice with a concomitant increase in glycolytic fibers prompted us to investigate the expression levels of MEF2. Western blot analysis showed MEF2C levels were significantly downregulated in the Mstn−/− muscle (Fig. 2A). Also, quantitative Western blot analysis performed on the myoblast nuclear extracts further confirmed that in the Mstn−/− mice there was reduced MEF2C level (Fig. 2B). To further prove that reduced levels of MEF2 were present in the Mstn−/− mice, we performed electrophoretic mobility shift assay (EMSA) with a MEF2 enhancer binding site and nuclear extracts isolated from wild-type and Mstn−/− primary myoblasts. The EMSA showed a single protein-DNA complex, which was competed out by the excess cold self-competitor, confirming that the observed band was specific for MEF2 (Fig. 3). Consistent with the reduced levels of MEF2C in Mstn−/− mice, a reduction in enhancer fragment binding was also observed (Fig. 3). Thus we concluded that in the absence of Mstn there was a downregulation of the MEF2 gene.
MyoD Levels are Upregulated in the Mstn −/− Mice
In recent years, the myogenic regulatory factor (MRF) family of basic helix loop helix transcription factors have been suggested to play an important role in the differentiation process of adult skeletal muscle cells through transcriptional control of phenotype-specific proteins. In rats, MyoD and myogenin mRNA have been shown to be most prevalent in fast glycolytic and slow oxidative fibers, respectively (14). Since there was an increased number of glycolytic fibers in the BF and TA muscles of Mstn−/− mice, we determined the MyoD expression in both the BF and TA muscles of wild-type and Mstn−/− mice. Consistent with the histological observation, the Western blot analysis indicated an increased MyoD protein level in the Mstn−/− mice (Fig. 4).
Reduced Calcineurin Expression in Mstn−/− Mice
The calcineurin-NFAT pathway has been shown to be important for the formation of slow oxidative fibers, since overexpression of calcineurin leads to increased oxidative fibers (42). Furthermore, increased calcineurin activity was found in slow fiber-enriched muscles such as soleus (4). Given that a lack of Mstn leads to an increased percentage of fast glycolytic fibers and a reduced percentage of slow oxidative fibers in BF muscle, we monitored calcineurin levels in the BF muscle. Semiquantitiate RT-PCR used to measure calcineurin cDNA levels showed there was a 50% reduction in the expression of calcineurin in the Mstn−/− BF compared with the wild-type BF (Fig. 5).
MLC1f, a Downstream Gene of MEF2 is Also Downregulated in the Mstn−/− Cattle
The above results clearly suggest that Mstn could regulate myofiber composition by regulating MEF2 protein expression. It is well characterized that the MLC1f promoter/enhancer is regulated by MEF2 family members (28). To confirm that a lack of Mstn, which results in MEF2C downregulation, also leads to reduced expression of a MEF2 target gene MLC1f, we performed Northern blot analysis on BF mRNA from normal muscled Friesian and heavy muscled Belgian Blue cattle. The results confirmed that in the absence of Mstn, there was a significant reduction in the expression of MLC1f (Fig. 6).
MLC1f Gene Expression is Regulated by Mstn at the Transcription Level
To further prove that Mstn positively regulates MLC gene expression, we assessed the effect of recombinant Mstn on the MLC regulatory elements. Previously, it has been shown in the transgenic mouse that both promoter and enhancer elements are required for the developmental specific expression of MLC (9). To monitor the effect of Mstn, luciferase gene was subcloned such that the MLC1f 1.2 kb promoter and 0.5 kb of the enhancer element flanked the 5′ and 3′ end of the luciferase gene. The reporter construct was transfected into C2C12 myoblasts, and stable cell lines were selected. Upon treatment of stable C2C12 cells with Mstn, there was a threefold increase in the reporter activity compared with nontreated cells (Fig. 7A). These results were consistent with Mstn being a positive regulator of MLC1f gene expression. The MLC enhancer sequence has been shown to contain several E box motifs and MEF2 binding sites (29). Although mutation analysis suggests that the E box motif and MEF2 binding sites are required for normal expression of the MLC gene, synergy between MyoD and MEF2 in regulating MLC is not shown to date. Thus to investigate whether the MLC1f enhancer would be synergistically activated by MyoD and MEF2C, the reporter construct was cotransfected with either MyoD or MEF2C, or MyoD and MEF2C together. The luciferase activity was later measured in the differentiated myotubes. As shown in Fig. 7B, while there was no significant increase in the fold activation by transfection of MyoD alone, transfection of MEF2C was able to significantly increase the activity of the MLC enhancer. In addition, when MEF2C and MyoD were cotransfected, there was a synergistic activation of the MLC enhancer, indicating that in vivo, MEF2 family members may play a critical role in the positive regulation of MLC gene expression.
Functional Electrical Stimulation
High-frequency fatigue (HFF) is described as the reduction of tetanic force during continuous stimulation. HFF has been shown to have significant differences depending on muscle fiber type. Thus HFF stimulation experiments were carried out on wild-type and Mstn−/− BF muscle to compare their fatigue resistance. Fatigue in this experiment is indicated by the length of time each subsequent tetanic contraction takes to drop to 50% of its peak force value. The results indicate that Mstn−/− BF demonstrates the lowest resistance to HFF, indicating a lower proportion of type I fibers. In contrast, wild-type BF (which has higher proportions of type I fibers) demonstrated a higher tolerance to the HFF protocol (Fig. 8A). However, measurements of the rise time of single twitches did not reveal significant differences between wild-type and Mstn−/− BF muscle. As expected, when wild-type soleus and BF were compared, BF muscle demonstrated the lowest resistance to HFF (Fig. 8B). These findings further confirm that the observed changes in BF fiber-type composition between Mstn−/− and wild-type animals have functional significance.
Mstn is a secreted growth factor that profoundly affects muscle growth. It has been shown to be a negative regulator of muscle growth, since lack of Mstn results in increased muscle mass due to hyperplasia. In this communication, we describe that in addition to the regulation of muscle mass, Mstn also regulates muscle fiber-type composition. Lack of Mstn results in a fiber-type switch that results in an increase in the fast glycolytic fibers.
Myofibers from mammals are heterogeneous with respect to their metabolic, contractile, and morphological properties (41). It has been noted that individual muscles have a characteristic proportion and distribution of fiber types (41). For example, whereas TA muscle is enriched with mainly fast glycolytic and oxidative fibers, soleus muscle is enriched with slow oxidative fibers. These specialized myofibers are highly plastic, and under several physiological and pathological circumstances, the metabolic and enzymatic nature of the fiber can be altered (25). It is thought that innervations with specific neurons and expression of specific MRFs play a dominant role in the regulation of fiber-type diversity (6, 26, 41). However, contributions from growth factors, let alone the negative regulators of myogenesis, are not well understood. Hence, we determined the role and possible molecular mechanism behind Mstn-mediated regulation of fiber-type diversity in postnatal muscle growth. We have previously shown that inactivation of Mstn leads to hyperplasia by increasing the myoblast number possibly during embryonic and or fetal myogenesis (38). One of the main questions we wanted to address was the consequences of the increased fetal myoblast numbers on the fiber-type diversity in the adult. The results indicated that in both BF and TA Mstn−/− muscle there was a significant increase in the number of fast glycolytic fibers (Fig. 1).
The observed change in the fiber-type composition in the myostatin null muscle has functional significance, as a significant decrease in resistance to fatigue was observed in Mstn−/− BF muscle (Fig. 8). One reason for the fiber-type difference in Mstn−/− mice could be due to a change in the specification of myoblasts. Alternatively, the increased number of fast glycolytic fibers could also be due to the hyperplasia observed in the Mstn−/− muscle. Hence, the proportion of different fiber types in both genotypes were derived from results normalized to 100 myofibers. The results confirmed that the increase in the proportion of type IIB fibers in both TA and BF was due to a decrease in the IIA and/or type I fibers irrespective of the fiber number (Fig. 1). Thus it could be concluded that a lack of Mstn results in an alteration in the fiber-type composition.
Potentially, Mstn could regulate the fiber-type diversity either at the fetal or postnatal stages. All myofibers are produced by the fusion of myotubes derived from myogenic precursor cells (22). The myoblasts are generated during two distinct waves in the course of fetal growth. The early wave of myoblasts (primary myoblasts) produce primary fibers that start to appear before individual muscle can be recognized. Primary fibers are slow-twitch fibers, which are mostly oxidative in nature and are only a minority of the final muscle fiber number but play a critical role in the generation of the later forming secondary fibers (36). Secondary fibers, derived from the secondary wave of myoblasts, are fast fibers in nature (36). Thus Mstn could be controlling the fiber-type diversity by regulating the number of secondary myoblasts. Indeed cell culture experiments support this hypothesis, since Mstn can regulate the cell cycle progression of myoblasts thereby controlling the cell number (38). In the Mstn−/− mice, the loss of such cell cycle control could lead to an increased proliferation of secondary myoblasts leading to an increased number of fast fibers. Furthermore, using anti-MHC antibodies, it has been shown that in the 100-day-old Belgian Blue fetus (which express mutated allele of myostatin), there is increased proliferation of secondary myoblasts leading to an increase in the accumulation of secondary fibers (8). In addition to influencing embryonic myogenesis, it is also possible that Mstn may regulate fiber-type diversity postnatally. MHC isoforms show sequential expression during development, which culminates in the adult pattern of fiber types (41). At their formation, all primary fibers express MHC embryonic, which is not found in adult muscles (40). Instead, all primary fibers that are destined to become slow fibers start expressing MHC-β, which is a slow MHC isoform. Unlike slow fibers, in fast fibers, the MHC embryonic is replaced by adult fast isoforms that are either MHC IIA, IIB, or IIX/D (41). Since different muscles have different proportions of fast fibers, regulation of MHC gene expression must entail a mechanism for generating fiber-type diversity. MRFs are important regulators of myogensis and recently they have been shown to have a role in the regulation of MHC gene expression (1). Enhancer analysis has shown that the MyoD transcription factor is important for the activation of MHC IIB gene expression. Hence, logically MyoD, along with myogenin, have been shown to influence the number of fast- and slow-twitch fibers (14). In this study, we have clearly shown that a lack of Mstn leads to the upregulation of the MyoD transcription factor (Fig. 4). Thus it is quite possible that this increased expression of MyoD would lead to the expansion of MHC IIB resulting in an increased number of fast glycolytic fibers. It is noteworthy that in Mstn−/− TA and BF muscle where increased MyoD expression is observed, we also found an increase in fast glycolytic fibers. These results further confirm that the increase in glycolytic muscle fibers seen in the Mstn−/− mice could also be due to the increased MyoD expression. This observation is further supported by recent results that showed that conditional overexpression of MyoD in soleus muscle results in the increased accumulation of type II fibers (15).
As opposed to the fast glycolytic fibers, we observed a reduction in the slow and fast oxidative fibers in the Mstn−/− mice (Fig. 1). Previously it has been shown that MEF2 family members, along with calcineurin, are involved in the formation of slow fibers, since overexpression of MEF2 isoforms and calcineurin results in an increased number of type I fibers (42). Similarly, in myotonic mice, where an increased number of slow fibers is seen, there is increased activity of MEF2 proteins (44). These results are consistent with the theory that while increased expression of MEF2 leads to the formation of slow fibers, loss of MEF2 may result in the reduction of slow fibers. Hence, we performed a molecular analysis to investigate the expression levels of MEF2 in Mstn−/− mice. Consistent with the reduction in the percentage of slow fibers, the Mstn−/− muscle also had reduced levels and activity of MEF2C and calcineurin (Figs. 2, 3, and 5). Concomitant with the loss of MEF2C, we also found that the MLC gene, a downstream target gene of MEF2, was also downregulated, suggesting that in the absence of Mstn there is reduced MEF2 activity (Fig. 6). Thus loss of MEF2 and calcineurin in the Mstn−/− mice could lead to decreased slow fibers. Indeed, a recent publication using both gain and loss of function has shown that activity levels of Mef 2 protein regulates type I fiber accumulation. Whereas loss of function of Mef2 resulted in the loss of type I fibers, overexpression of hyperactive form of Mef2 led to increased accumulation of slow fibers (27).
Even though an increase in fast twitch fibers composition was noted in the myostatin null muscle, an anomaly of a significant decrease in MLC1f gene expression (Fig. 6) was seen in the myostatin null muscles. However, this significant reduction in the MLC1 gene expression did not result in the reduction in the contraction velocity of the muscles (Fig. 8). Thus the observed reduction in MLC1 gene expression does not appear to affect the functionality of muscles. At the molecular level this decrease in MLC1f expression could be attributed to reduced levels of Mef2 protein in the myostatin null muscles (Fig. 2), since MLC1f is a downstream target gene of MEF2 protein (13).
Clearly, a lack of Mstn results in an alteration in the fiber-type composition. It is possible the Mstn acts via two different mechanisms, either by switching type I and IIA fibers to type IIB fibers during fetal and or early postnatal myogenesis, or by changing the specification of the fibers during development through altered MRF expression levels. Although at this time the exact mechanism by which Mstn influences fiber types is unclear, an earlier report supports the latter, as increased MyoD was observed during fetal development of double muscle cattle in relation to normal muscle cattle (23).
In summary, here we have shown that Mstn regulates fiber-type composition in muscle. Loss of Mstn leads to a decrease in slow and fast oxidative fibers with a concomitant increase in fast glycolytic fibers, whereas as a funcitional consequence, increases the susceptibility to HFF. Mstn may accomplish the alteration in fiber-type composition either during prenatal or postnatal myogenesis by controlling myoblast number and or regulation of MyoD and MEF2 gene expression.
We are indebted to Foundation for Research, Science and Technology, New Zealand, for financial support.
We thank Developmental Studies Hybridoma Bank for antibody.
↵* A. Hennebry and C. Berry contributed equally to this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2009 the American Physiological Society