当前位置: 首页 > 医学版 > 期刊论文 > 内科学 > 内分泌学杂志 > 2005年 > 第8期 > 正文
编号:11168730
Myostatin Inhibits Myogenesis and Promotes Adipogenesis in C3H 10T(1/2) Mesenchymal Multipotent Cells
     Division of Endocrinology Metabolism and Molecular Medicine (J.N.A., S.B., S.R.-P., R.S., N.F.G.-C.), RCMI DNA Molecular Core, Charles R. Drew University of Medicine and Science, Los Angeles, California 90059; Department of Urology (T.R.M., N.F.G.-C.), University of California, Los Angeles School of Medicine, LABioMed Research Institute at Harbor-UCLA Medical Center, Torrance, California 90502; and School of Biological and Molecular Sciences (N.P.G., M.M.F.), Oxford Brookes University, Headington Campus, Oxford OX3 0BP, United Kingdom

    Address all correspondence and requests for reprints to: Jorge N. Artaza, Ph.D., Division of Endocrinology, Metabolism, and Molecular Medicine, Charles R. Drew University of Medicine and Science, 1731 East 120th Street, Los Angeles, California 90059. E-mail: joartaza@cdrewu.edu.

    Abstract

    Inactivating mutations of the mammalian myostatin gene are associated with increased muscle mass and decreased fat mass; conversely, myostatin transgenic mice that overexpress myostatin in the skeletal muscle have decreased muscle mass and increased fat mass. We investigated the effects of recombinant myostatin protein and antimyostatin antibody on myogenic and adipogenic differentiation of mesenchymal multipotent cells. Accordingly, 10T(1/2) cells were incubated with 5'-azacytidine for 3 d to induce differentiation and then treated with a recombinant protein for myostatin (Mst) carboxy terminal 113 amino acids or a polyclonal anti-Mst antibody for 3, 7, and 14 d. Cells were also cotransfected with a Mst cDNA plasmid expressing the full-length 375-amino acid protein (pcDNA-Mst375) and the silencer RNAs for either Mst (pSil-Mst) or a random sequence (pSil-RS) for 3 or 7 d, and Mst expression was determined. Adipogenesis was evaluated by quantitative image analysis of fat cells before and after oil-red-O staining, immunocytochemistry of adiponectin, and Western blot for CCAAT/enhancer binding protein-. Myogenesis was estimated by quantitative image analysis-immunocytochemistry for MyoD (Myo differentiation protein), myogenin, and myosin heavy chain type II, or by Western blot for myogenin. 5'-azacytidine-mediated differentiation induced endogenous full-length Mst expression. Recombinant Mst carboxy terminal 113 amino acids inhibited both early and late markers of myogenesis and stimulated both early and late markers of adipogenesis, whereas the antibody against Mst exerted the reverse effects. Myogenin levels at 7 d after transfection of pcDNA-Mst375 were reduced as expected and elevated by pSil-Mst, which blocked efficiently Mst375 expression. In conclusion, myostatin promotes the differentiation of multipotent mesenchymal cells into the adipogenic lineage and inhibits myogenesis.

    Introduction

    TRANSGENIC MICE THAT overexpress recombinant myostatin (Mst) protein in the skeletal muscle have lower skeletal muscle mass and higher whole body fat mass, in comparison with wild-type controls (1). Similarly, the mice implanted with cells engineered to overexpress Mst experience significant cachexia (2). In general, Mst expression is higher under conditions in which the content of muscle mass is reduced (3). Conversely, null mutations of Mst gene in knockout mice and in double-muscle cattle are associated with hypermuscularity and decreased fat mass (4, 5, 6). Thus, Mst gene expression is an important modulator of body composition in experimental animals.

    A recent report of a child with an inactivating mutation of the Mst gene that was associated with hypermuscularity and decreased fat mass further supports the role of Mst in the regulation of body composition (7). Significantly, in this child, as well as in the Mst knockout mice, fat mass, as measured by the weight or size of fat pads, was reduced in association with inactivating mutations of the Mst gene; in other experimental paradigms, fat mass is increased when Mst is overexpressed (5, 8, 9, 10).

    The mechanisms by which Mst regulates body composition are not fully understood. Both the full-length and the 110-amino acid carboxy terminal Mst peptide have been shown to inhibit protein synthesis and satellite cell proliferation, in the latter case by blocking satellite cell entry into the cell cycle (11). However, these effects of Mst on satellite cells do not explain easily the observed changes in fat mass in Mst null and transgenic mice (1, 4, 5, 6). It is also difficult to reconcile these effects with a putative mechanism of preadipocyte activation, particularly because Mst has been reported to inhibit in vitro the adipogenic conversion of 3T3 preadipocytes (9) and of C3H 10T(1/2) cells undergoing bone morphogenic protein (BMP)-induced adipogenesis (12). Therefore, we considered the alternative interpretation that Mst acts upstream of the committed cell stage, e.g. on multipotent cell differentiation into myogenic and adipogenic lineages.

    Multipotent cells have the ability to undergo in vitro differentiation into multiple cell lineages according to the type of incubation medium and/or the factors that are added to induce specific differentiation programs, in contrast to progenitor cells that are already committed to a single cell lineage or terminally differentiated cells (13, 14, 15). One of the multipotent cell lines widely used to study myogenic, and to a certain extent adipogenic differentiation, is the C3H 10T(1/2) cell, a mesenchymal fibroblast-like cell line of embryonic origin that upon incubation with 5'-azacytidine can form in regular Dulbecco’s medium myotubes and adipocytes (16, 17) and in more specialized media can evolve into osteogenic or myofibroblast cell lineages (18). Transfection of the untreated C3H 10T(1/2) with certain key lineage factors, such as BMP (19), TGF? (20), or MyoD (17), can trigger highly efficient specific differentiation programs selected by the forced expression of these proteins. We used C3H 10T(1/2) cells as our experimental model because these cells have been used widely to study the mechanisms that modulate myogenic and adipogenic differentiation and the molecular pathways of lineage determination.

    We hypothesized that Mst promotes the commitment and differentiation of mesenchymal, multipotent cells into the adipogenic lineage and inhibits their differentiation into the myogenic lineage; we used several complementary approaches to test this hypothesis. In the present study, we investigated first, the effects of recombinant Mst or Mst antibodies on the differentiation of 5'-azacytidine-treated C3H 10T(1/2) cells into the myogenic and adipogenic lineages. In addition, we determined the effects of transfected Mst cDNA or a cDNA construct encoding silencer (si) Mst RNA.

    Materials and Methods

    Tissue culture

    Mouse C3H 10T(1/2) cells, grown in DMEM with 10% fetal bovine serum at 37 C, were treated with or without 20 μM 5'-azacytidine (AZCT) for 3 d to induce differentiation (21), split in a 1:2 ratio, allowed to recover for 2 d, seeded at 60–70% confluence in six-well plates or eight-well chamber slides, and incubated with 4–8 μg/ml of the recombinant Mst protein 113 amino acid-Mst protein, (22) or 3.3–33 μg/ml of our custom made polyclonal anti-Mst antibody (23) in DMEM-10% serum for 3, 7, and 14 d.

    Immunocytochemical analyses of MyoD, myogenin, myosin heavy chain type II (MHC-II), adiponectin, and Mst

    Cells grown in eight-well chamber slides were fixed in 2% paraformaldehyde, quenched with H2O2, blocked with normal goat or horse serum, and incubated with specific antibodies: MyoD (1:500), myogenin (1:500) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), MHC II (1:40) (Novocastra Laboratories Ltd., Newcastle, UK) (24), and adiponectin (1:500) (Phoenix Pharmaceuticals Inc., Belmont, CA) (25). Detection was based on a secondary biotinylated antibody (1:200), followed by the addition of the streptavidin-horseradish peroxidase ABC complex (1:100; Vectastain Elite ABC system, Vector Laboratories, Burlingame, CA) and 3,3'-diaminobenzidine (Sigma, St. Louis, MO) (9). In the case of the Mst monoclonal antibody generated by our group against the Mst carboxy terminal 113 amino acids (Mst-113) (mature fraction), we used the mouse on mouse (M.O.M.; Vector Laboratories) immunodetection system, which provides a significant reduction of background staining on tissues or cells from mouse origin. The cells were counterstained with Meyer’s hematoxylin. In negative controls, we either omitted the first antibody or used a rabbit nonspecific IgG.

    The cyto (immuno) chemical staining was quantitated by densitometry using the Image-Pro Plus software (Media Cybernetics, Silver Spring, MD). The number of MyoD and myogenin positive cells was counted against the total number of cells determined by counterstaining. The area of MHC+ and adiponectin+ cells was computed per field and averaged more than 10 fields (21).

    Fusion index

    The fusion index was calculated by dividing the number of nuclei present in MHC-II-positive myotubes by the total number of nuclei counted (hematoxylin counterstain). Approximately 1000 nuclei were counted per C3H 10T(1/2) cell culture. Each fusion index represents the mean value of three independent cultures and is expressed as mean ± SEM (26).

    Immunocytochemical analyses of CD34, CD45, and stem cell antigen or ataxin-1 (Sca-1) surface markers

    Cells grown in eight-well chamber slides were fixed in 2% paraformaldehyde, blocked with fetal bovine serum, and incubated with biotinylated rat antimouse antibodies: CD34, CD45, and Sca-1 (BD PharMingen, San Jose, CA) (27). Detection was based on a secondary antibody conjugated with streptavidin-fluorescein isothiocyanate or Texas red (Vector Laboratories). The cells were counterstained with 4',6'-diamino-2-phenylindole. In negative controls, we either omitted the first antibody or used a rabbit nonspecific IgG.

    Oil-red-O staining and adipocyte counting

    C3H 10T(1/2) cells were washed with PBS, fixed in 2% paraformaldehyde, and stained with 0.3% oil-red-O. The number of adipocytes was counted under a bright-field microscope in 10 x100 fields and averaged (21).

    Western blot and densitometric analysis

    Cell lysates (50–100 μg of protein) were subjected to Western blot analyses by 7.5 or 12% gel electrophoresis, using 1:200 mouse monoclonal anti MHC-II antibody (fast, Novocastra Laboratories), 1:500 antimyogenin, 1:300 anti-CCAAT/enhancer binding protein (C/EBP)- (Santa Cruz Biotechnology), our monoclonal antibody for Mst (1/500), and 1:10,000 anti-glyceraldehide-3-phosphate dehydrogenase (GAPDH) monoclonal antibody (Chemicon International, Temecula, CA). The washed membranes were incubated with 1:1000 dilution or 1/3000 for Mst of secondary antibody linked to horseradish peroxidase. Immunoreactive bands were visualized by using the ECL Plus Western blotting chemiluminescence detection system (Amersham, Chicago, IL) (23). The densitometric analysis of the bands were done with the Scion Image software beta 4.0.2 (Scion Corp., Frederick, MD)

    Mst recombinant proteins

    Mst recombinant protein produced in Escherichia coli is a 16 kDa protein containing 113-amino acid residues of the human Mst protein, linked to an N-terminal fusion of a six histidine residues tag, and it was purified in a three-step procedure using affinity Ni-NTA chromatography and size exclusion chromatography before and after refolding. The recombinant protein was tested for its biological activity in C2C12 cells (22). Dr. Vladimir Kolar (BioVendor Laboratory Medicine, Inc., Brno, Czech Republic) provided the Mst recombinant protein. Throughout the text, the term Mst-113 is used to refer to this recombinant Mst protein and Mst-110 refers to the carboxy-terminal fragment of the human endogenous Mst protein that has been shown to originate by cleavage from the 375 amino acids precursor protein in cells transfected with a Mst cDNA encoding the full-length protein and expressing furin proteases (4). Although the mouse homologous protein has 111 amino acids (2), for the sake of uniformity, the term Mst-110 is used to refer generically to this fragment.

    RT-PCR

    Total RNA, extracted by using the Trizol-reagent (Invitrogen, Carlsbad, CA) was reverse transcribed, and cDNA was amplified for 35 cycles by PCR at 94 C for 30 sec, primer annealing at 58 C for 30 sec, and extension at 72 C for 1 min. PCR products were analyzed in 1.5% agarose gels. The sequences of the forward/reverse PCR primers are as follows: Mst (167 bp) forward primer, 5'-CAGCCTGAATCCAACTTAGG and reverse primer, 5'-TCGCAGTCAAGCCCAAAGTC. Primer sequences for GAPDH (152 bp) forward primer, 5'-ATCACTGCCACCCAGAAGACT and reverse primer, 5'-CATGCCAGTGAGCTTCCCGTT.

    siRNA Mst

    To prepare the silencer RNA construct (28, 29) for Mst, we analyzed the mouse Mst gene sequence (accession no. NM_010834) using the Web-based siRNA target finder and design tool provided on the Ambion Web site (Ambion, Inc., Austin, TX). Five regions were initially targeted for likely inhibitory activity by siRNAs. Double-stranded RNAs were transcribed in vitro using the Silencer siRNA construction kit. Each siRNA was tested for inhibitory activity at 1, 10, and 100 nM concentrations by cotransfection of pCDNA3.1-Mst into HEK293 cell cultures using lipofectamine 2000. After 48 h, cell lysates were collected in M-PER (Pierce Biotechnology, Inc, Rockford, IL) and Western blots were performed. One siRNA (GDF8 siRNA26, 5'-AAGATGACGATTATCACGCTA-3', position 428–446) was found to inhibit more than 95% of Mst gene expression (data not shown). Based on a Blast search, this sequence has homology not only to mouse but also to human, rat, rabbit, cow, macaque, and baboon. A short hairpin DNA sequence (5' sense strand: 5'-GATCCGATGACGATTATCACGCTATTCAAGAGATAGCG- TGATAATCGTCATCTTTTTTGGAAA-3') was synthesized and cloned into the pSilencer 2.1-U6 neo (Ambion) according to the manufacturer’s instructions. The DNA sequence consists of a BamHI DNA restriction site, sense strand, nine-nucleotide loop, antisense strand, RNA polymerase III terminator, and HindIII DNA restriction site 5' to 3'. The pSilencer 2.1-U6 neo-GDF8 siRNA plasmid construct was cotransfected with pCDNA3.1-Mst (each 1 μg/well plate, six-well plate) into C3H 10T(1/2) cell cultures using lipofectamine 2000 as before.

    Statistical analysis

    All data are presented as mean ± SEM; between-group differences were analyzed by using ANOVA. If overall ANOVA revealed significant differences, then pair-wise comparisons between groups were performed using Newman-Keuls multiple comparison test. All comparisons were two-tailed, and P < 0.05 was considered statistically significant. The experiments were repeated two to three times, and data from representative experiments are shown.

    Results

    Effects of Mst and Mst antibody on myogenesis

    The C3H 10T(1/2) cells selected for the current work are multipotent, have the ability to undergo differentiation into several cell lineages, but surprisingly no characterization of stem cell marker expression has been published so far. Therefore, cells untreated with AZCT were subjected to immunocytofluorescence, and detection for three of the usual stem cell markers, namely CD34+ and CD45+, and Sca-1, was performed with their respective negative controls (only one is shown as an example). Figure 1 shows that these cells express CD34+ and CD45+, but they do not express Sca-1 in the absence of AZCT.

    FIG. 1. Immunofluorescence detection of stem cell markers on C3H 10T(1/2) cells not treated with AZCT. Cells were stained with the biotinylated primary antibody and the streptavidin secondary antibody linked to the fluorescent dye, as indicated. In negative controls we either omitted the first antibody or used a rabbit nonspecific IgG. In every case cell nuclei were counterstained with 4',6'-diamino-2-phenylindole. DAPI, 4'-6-diamino-2-phenylindole; FITC, fluorescein isothiocyanate.

    To determine whether the C3H 10T(1/2) cells express Mst, cells were incubated for 1 or 2 wk in the absence and presence of AZCT. This agent triggers the differentiation of C3H 10T(1/2) cells into myogenic and adipogenic lineages. The formation of myotube was evidenced by morphology and the expression of MHC-II, detected by immunocytochemistry, Western blot, and real-time PCR (21). Immunocytochemistry with a custom-made monoclonal antibody against Mst that we have validated for the present work, combined with hematoxylin counterstain, demonstrated that in the absence of AZCT, C3H 10T(1/2) cells did not express detectable amounts of Mst either at 1 week (Fig. 2A, left panel) or 2 wk (not shown).

    FIG. 2. Mst expression in differentiating C3H 10T(1/2) cells. Cells were incubated with or without AZCT5 for 3 d and after changing the medium, they were allowed to continue for 1 or 2 wk. A, Immunocytochemical staining of cells, using anti-Mst antibody. B, Immunocytochemical staining of polynucleated cells, using anti-MHC-II antibody, incubated with and without AZCT for 2 wk. C, Total RNA was isolated and used for RT-PCR, using primers specific for Mst. D, Cell extracts were used for Western blot analysis, using monoclonal, anti-Mst antibody. M, molecular marker; C, mouse skeletal muscle RNA as a positive control.

    However, when cells were incubated with AZCT to induce differentiation, Mst expression was detected at 1 wk in most cells, in the cytoplasm (Fig. 2A, middle panel). At 2 wk Mst expression was localized in multinucleated myotubes (Fig. 2A, right panel). The differentiation into myotubes in the 2-wk experiment was shown by immunodetection of MHC-II in the same cells (Fig. 2B). Parallel incubations were used to isolate either total RNA or protein extracts. RT/PCR of RNA from cells in duplicate wells (Fig. 2C) confirmed the immunocytochemistry results. We were unable to detect the 167-bp Mst band by RT-PCR in cultures carried out in the absence of AZCT, whereas the band was visible in the AZCT-treated cells at 1 and 2 wk. In all cases, expression of housekeeping GAPDH mRNA remained constant. Western blot analysis using a custom-made monoclonal antibody (Fig. 2D) was in agreement with the immunohistochemistry and RT/PCR determinations. We observed a single 52-kDa band, presumably corresponding to the unprocessed 375-amino acid full-length Mst monomeric protein (22, 23, 24, 30) in cells treated with AZCT for 1 or 2 wk. In addition, it should be considered that cells are exposed throughout the incubation to small amounts of exogenous Mst present in the serum (23).

    The results of the Western blot analysis indicate that the cells undergoing myogenic differentiation from the C3H 10T(1/2) cells are unable to proteolytically process the Mst precursor protein. This was confirmed by validating the monoclonal antibody used to detect the Mst protein, as shown in Fig. 3. This antibody was generated against a recombinant, 113-amino acid, human Mst carboxy-terminal sequence (Mst-113), which is virtually identical in the human, mouse, and rat. The anti-Mst monoclonal antibody detected the recombinant, 113 amino acid, carboxy-terminal fragment of Mst protein that was used to elicit the antibody, as a single 14- to 15-kDa band. This antibody also detected the 32-kDa band corresponding to the processed dimer in wild-type mouse gastrocnemius and tibialis muscle homogenates (Fig. 3, left panel). In the latter case, a faint 52-kDa band is probably the unprocessed whole length 375 amino acid protein. These bands were absent in the corresponding tissue homogenates from the Mst knockout mice (4), thus confirming their Mst identity. No 13- to 15-kDa band was detected in any of the wild-type mouse muscle homogenates. The Pounceau Red staining of the membrane (Fig. 3, right panel) shows that protein loading and transfer was uniform among lanes. These results with the monoclonal antibody are similar to the ones reported for the polyclonal (3, 22, 23), except that the immunoblots with the monoclonal antibody are much cleaner in cells or muscle extracts.

    FIG. 3. Validation of the custom-made monoclonal antibody against recombinant Mst protein and skeletal muscle proteins. The antibody was elicited against recombinant Mst-113 protein and tested by Western blot analysis (left panel) using Mst-113 (1 ng) or the skeletal muscle extracts (30 μg protein). The membrane was subjected to Pounceau Red protein staining (right panel). WT, Wild-type mouse; KO, Mst knockout mouse; G, gastrocnemius; T, tibialis anterior; M, molecular marker.

    C3H 10T(1/2) cells were then incubated in triplicate wells with increasing concentrations of recombinant Mst-113 (0–8 μg/ml) for 3 d after AZCT-induced differentiation, a stage at which endogenous Mst expression was very low (see below and Fig. 7), and with anti-Mst polyclonal antibody (0.1 mg/ml). This 3-d period was chosen to detect changes in expression of early myogenic markers on the assumption that the addition of recombinant Mst may inhibit this expression, whereas the Mst antibody may block the effects produced by exogenous Mst, even if endogenous Mst were negligible. Immunocytochemistry staining for the early myogenic marker, MyoD (Fig. 4A) showed the characteristic nuclear staining in some of the mononucleated cells. No MyoD staining was observed when the primary antibody was omitted for the immunodetection. Incubation with graded concentrations of Mst-113 reduced the number of MyoD+ cells. Quantitative image analysis (Fig. 4B) showed a down-regulation by Mst-113 of the number of MyoD+ cells, with 72% inhibition at 8 μg/ml. In contrast to the recombinant Mst-113 protein, incubation of C3H 10T1/2 cells for 72 h with anti-Mst antibody did not exert any significant effect on the number of MyoD+ cells, probably due to an insufficient neutralizing ability at the dilution of the antibody chosen for this experiment. Therefore, the incubation with the antibody against Mst was repeated in a separate experiment with a higher concentration of this antibody (330 μg/ml). The number of MyoD+nuclei was higher in cells treated with 330 μg/ml anti-Mst antibody, compared with the control cells (Fig. 4C). Quantitative image analysis showed a significant 2.4-fold increase in the number of MyoD+ nuclei (Fig. 4D).

    FIG. 7. Effect of two different myostatin siRNA constructs or a random siRNA construct on Mst protein expression and myogenic differentiation in differentiating C3H 10T(1/2) cells. Cells were transfected with pcDNA-Mst-375 cDNA with or without cotransfection with siRNA, as indicated, to examine the effects of siRNA on recombinant Mst expression after a 3- and 7-d incubation. A, Luminol detection of Western blots with monoclonal antibody against Mst at 3 d after transfection. B, Densitometric evaluation of band intensities from A expressed as Mst to GAPDH ratios (mean of two wells). C, Luminol detection of Western blots with antibody against myogenin after transfection with pcDNAMst-375, Mst siRNA (pSil-Mst no. 26, including a parallel incubation with RMst-113 (8 μg/ml). Control, Cells incubated with medium alone; pSil Mst no. 4 and no. 26, siRNAs for Mst, sequence no. 4 and no. 26; pSil RS, silencer RNA random sequence.

    FIG. 4. Inhibition of myogenic differentiation in differentiating C3H 10T(1/2) cells by recombinant Mst-113 protein. AZCT-treated cells were transferred to fresh medium without serum and incubated for 4 h with recombinant Mst-113 protein or with polyclonal anti-Mst antibody at the indicated concentrations. Serum was then added and incubation was continued for 3 d. A, Immunocytochemical staining of cells treated with medium alone (control) or the indicated concentrations of recombinant Mst-113 protein, using anti-MyoD antibodies. Immunocytochemistry was performed with or without the first antibody (no 1st Ab) as an internal control. B, Effects of graded concentrations of Mst-113 or anti-Mst antibody (100 μg/ml) on the number of MyoD+ nuclei per 100 cells, using quantitative image analysis. C, Representative immunocytochemical staining of a different batch of cells incubated in a separate experiment with or without anti-Mst antibody at a higher concentration (330 μg/ml). D, Quantitative image analysis for C. C, Control cells treated with medium alone; Ab, cells treated with the indicated concentrations of anti-Mst antibody. P values are expressed in comparison with the control incubation without Mst protein or the anti-Mst antibody. Magnification, x400.

    Incubation of cells with Mst-113 protein for 1 wk, a period during which endogenous Mst expression was detectable (see Fig. 2), down-regulated the expression of myogenin, a marker of myogenic conversion that also shows nuclear staining (Fig. 5A). Image analysis (Fig. 5B) confirmed the down-regulation of the number of myogenin+ cells that reached 78% inhibition at 8 μg/ml. Incubation of cells with Mst polyclonal antibody had little effect on myogenin+ cells, and the incubation with the higher concentration was not performed.

    FIG. 5. Effect of recombinant Mst-113 protein and anti-Mst antibody on myogenesis in differentiating C3H 10T(1/2), assessed by using myogenin immunocytochemical staining. Cells were incubated with medium alone (C, control), graded concentrations of recombinant Mst-113 or anti-Mst antibody (Mst Ab) for 1 wk. A, Immunocytochemical staining using an antibody against myogenin; quantitative image analysis of stained cells is shown in B. Asterisks denote the P values for the statistical comparison of treatment group against medium control. No 1st Ab, Cells not treated with first antibody. Magnification, x400

    When incubation with recombinant Mst-113 was extended to 2 wk, a period during which endogenous Mst expression was more evident than at 1 wk (see Fig. 2), there was a significant down-regulation of expression of MHC-II, a late myogenic marker. The area of MHC-II+ myotubes decreased in wells treated with graded concentrations of recombinant Mst-113 protein (Fig. 6A). Coincubation of cells with anti-Mst antibody was associated with significantly higher MHC-II+ myotube area, compared with cells treated with medium alone. Image analysis (Fig. 6B, left panel) confirmed the down-regulation of MHC-II+ myotube area by Mst-113 protein (about 50% inhibition at 8 μg/ml) and up-regulation by the anti-Mst polyclonal antibody (120% up-regulation). The determination of the fusion index (Fig. 6B, right panel), expressed as the ratio between the number of myonuclei present in MHC-II-positive myotubes divided by the total number of nuclei, was in good agreement with the area of MHC-II myotubes.

    FIG. 6. Effect of recombinant Mst-113 protein and anti-Mst antibody on myogenesis in differentiating C3H 10T(1/2), assessed by using anti-MHC-II antibody. Cells were incubated with medium alone (C, control), graded concentrations of recombinant Mst-113, or anti-Mst antibody (Mst Ab) for 2 wk. A, Immunocytochemical staining using an antibody against MHC-II; quantitative image analysis of stained cells is shown in B (left panel). B (right panel), Fusion index based on MHC-II-positive nuclei divided by the total number of nuclei in each field. Asterisks denote the P values for the statistical comparison of treatment group against medium control. No 1st Ab, Cells not treated with first antibody. Magnification, x200.

    Effects of Mst overexpression or blockade on myogenesis

    To determine whether Mst produced endogenously during C3H 10T(1/2) cell differentiation would exert the same effects as exogenous recombinant Mst, we transfected these cells with a Mst cDNA construct. For this purpose, C3H 10T(1/2) cells were differentiated with AZCT and then transiently transfected with the cDNA-expressing mouse Mst full-length 375-amino acid sequence under a cytomegalovirus promoter (pcDNA-Mst-375) (22) or with the same vector expressing green fluorescent protein as a reporter gene and incubated for 3 and 7 d. Transfection efficiency at 3 d was about 40%, as judged by the number of green fluorescent protein+ cells (not shown), and Mst expression assessed by Western blot (Fig. 7A) was negligible in the C3H 10T(1/2) cells at this early period, but it was considerably enhanced by the mouse Mst-375 construct. Only a 52-kDa band was visualized with our monoclonal antibody, with no evidence of proteolytic processing to the Mst-113 protein (14- or 28- to 32-kDa bands). The lack of processing was also evident in separate transfections with this Mst-375 construct in HEK293 cells (not shown), similarly to what we had reported previously with the human Mst-375 construct (22).

    In the same experiment, we cotransfected C3H 10T(1/2) cells with the Mst-375 cDNA construct with and without constructs for a Mst siRNA or a random siRNA construct. All siRNA constructs had been tested previously in HEK293 cells (data not shown), and two Mst siRNAs that were the most effective in inhibiting Mst expression were selected for these experiment in C3H 10T(1/2) cells (Fig. 7A). These constructs were very effective in blocking Mst-375 cDNA expression in the C3H 10T(1/2) cells (Fig. 7A). Densitometric analysis of the respective bands showed that the siRNAs down-regulated Mst expression by 83 and 93%, respectively, whereas the random siRNA did not affect significantly Mst expression (Fig. 7B). When the phenotypic effects of Mst-375 overexpression on myogenesis were examined by Western blot 7 d after transfection, the expression of the Mst-375 cDNA was still detectable and the siRNA was active in inhibiting endogenous Mst expression (Fig. 7C, left panel). The Mst-375 cDNA inhibited the expression of the intermediate stage marker, myogenin at a level comparable with the one exerted by the Mst-113 recombinant protein, and the Mst siRNA stimulated this expression as expected (Fig. 7C, right panel).

    The effects of Mst on adipogenesis

    We also investigated the effects of recombinant Mst on adipogenic differentiation of C3H 10T(1/2) cell differentiation by using an early marker, C/EBP, and an intermediate marker, adiponectin. C3H 10T(1/2) cells were incubated in the regular medium used for myogenesis with graded concentrations of recombinant Mst-113 protein for 4 d, and C/EBP protein was detected by Western blot analysis (Fig. 8A, left panel). The selection of an early period of incubation was based on the same considerations as those used for the experiment in Fig. 4. In contrast to the myogenic markers, Mst-113 protein up-regulated dose-dependently the expression of the 42-kDa C/EBP band, and coincubation of cells with anti-Mst antibody in the absence of added Mst-113, only slightly down-regulated the expression of C/EBP protein, as determined by densitometry (Fig. 8A, right panel). The very little effect of the antibody is probably due to the low level of endogenous Mst produced at the 4-d period. Equal loading and transfer was verified by GAPDH. The up-regulation of adipogenesis by Mst-113 was confirmed at 1 wk by immunocytochemistry for adiponectin and quantitative image analysis (Fig. 8B). Control cells, when stained with antiadiponectin antibody, showed a somewhat diffuse staining, with occasional cells exhibiting intensive staining. However, incubation with Mst-113 increased the number dose-dependently of adiponectin+ cells. In contrast, anti-Mst polyclonal antibody at this longer period of incubation decreased the intensity of adiponectin staining and the number of adiponectin+ cells.

    FIG. 8. Effect of recombinant Mst-113 protein and anti-Mst antibody on adipogenic differentiation C3H 10T(1/2), assessed by measurements of C/EBP protein by Western blot analysis. A, Immunohistochemical staining, using anti-adiponectin antibody. B, Cells were incubated for 4 d (C/EBP) and 1 wk (adiponectin) with graded concentrations of recombinant Mst-113 or anti-Mst antibody, and plates were divided for either trypsination and Western blot on the cell extracts (30 μg/lane) with an antibody against C/EBP- (A) or immunocytochemistry using an anti-adiponectin antibody (B). No 1st Ab, Cells not treated with first antibody.

    Although the preceding experiment suggested that Mst up-regulated adipogenic differentiation the more conclusive confirmation was obtained after a 2-wk incubation and direct detection of adipocytes using the oil-red-O staining (Fig. 9). Only a few control cells treated with medium alone showed the typical intensively stained rosette of fat vacuoles (Fig. 9A); the number of oil-red-O+ adipocytes was significantly higher in wells incubated with Mst-113 protein and lower in wells treated with the polyclonal Mst antibody at a higher concentration than the one used for the previous experiments (0.3 mg/ml). Cell counting (Fig. 9B) showed that the number of adipocytes in the cell population in the control cultures increased dose-dependently with Mst-113 addition almost to 4-fold at 8 μg/ml. In another experiment, incubation of C3H 10T(1/2) cells with the anti-Mst antibody at the maximum concentration tested (0.3 mg/ml) reduced the number of adipocytes to one third the control wells. Because Mst-113 or Mst antibody induce opposite effects on myogenesis and adipogenesis, it is unlikely that the observed effects are due to toxic effects or other artifacts.

    FIG. 9. Effect of recombinant Mst-113 protein and anti-Mst antibody on adipogenic differentiation of C3H 10T(1/2), assessed by oil-red-O staining, A, Counting the number of adipocytes. B, Cells were incubated with medium alone (control), graded concentrations of recombinant Mst-113, or anti-Mst antibody at the indicated dilution for 2 wk. Cells were stained with oil-red-O A, and quantitative image analysis was applied B. P values for the statistical comparison with control wells are shown.

    Discussion

    To our knowledge, this is the first demonstration that Mst, a well-characterized negative regulator of skeletal muscle mass, can modulate the differentiation of mesenchymal multipotent cells by inhibiting myogenesis and stimulating adipogenesis. Recombinant Mst protein inhibited the differentiation of these multipotent mesenchymal cells into the myogenic lineage, as indicated by inhibition of the formation of MHC-II+ myotubes and down-regulation of key myogenic differentiation markers, MyoD, and myogenin as well as the decrease in the fusion index. Conversely, recombinant Mst promoted the differentiation of these multipotent cells into the adipogenic lineage, up-regulating C/EBP and adiponectin, two key adipogenic markers, and increasing the number of oil-red-O+ adipocytes. Anti-Mst antibody had the reverse effect; it up-regulated myogenic differentiation and down-regulated adipogenesis at a late stage. The observations that Mst modulates the differentiation and/or commitment of mesenchymal multipotent cells, inhibiting myogenesis and promoting adipogenesis, provide a unifying explanation for the observed reciprocal effects of Mst on muscle and fat mass in Mst hyperexpressing transgenic and Mst-null knockout mice.

    The C3H 10T(1/2) cell line we used in the present work is known to undergo differentiation into many cell lineages (16, 17, 18, 19, 20, 21) and has been extensively validated and used to elucidate the signaling cascades that regulate myogenesis. We demonstrated that these cells express before AZCT addition at least two characteristic multipotent cell surface markers CD34+ and CD45+, although in vivo the myogenically committed satellite cells in the skeletal muscle are also positive for CD34+ (13, 14, 15). On the contrary, CD45 is not expressed in a fraction of skeletal muscle stem cells obtained by sequential plating but is expressed in other fractions (31, 32). Remarkably, Sca-1, a marker used to select a special type of stem cells from the skeletal muscle named side-population cells (33), was not detected in C3H 10T(1/2) cells. This is not surprising, because this phosphatidylinositol-anchored protein is a feature of mouse hematopoietic stem cells in the bone marrow, and to our knowledge, the C3H 10T(1/2) cells have not been reported to undergo hematopoietic differentiation, as the side-population cells do. The C3H 10T(1/2) cell is a good model for multilineage differentiation, although it is not known whether the progeny from a single cell can really undergo divergent commitments or the cell line is a heterogeneous mix of progenitor cells that are activated by AZCT to continue in their predetermined cell lineages (34). We recognize that no cell line can fully recapitulate the complexity of the human organism.

    Mst has been shown previously to inhibit mouse C2C12 myoblast and satellite cell replication and MyoD activation stimulated by autocrine or paracrine-positive regulators of myogenesis (e.g. IGF-I, fibroblast growth factor, and others) as well as differentiation into myotubes (11, 22, 35). Mst could therefore reduce the availability of myoblast precursors in addition to down-regulating MyoD and myogenin expression (8, 36), and possibly MHCII, as a homeostatic counterbalance to the myogenic stimulators (37). The data presented in this manuscript suggest that, in addition to the previously reported effects on satellite cell activation, Mst may also affect a stage more upstream in the cell lineage cascade, e.g. on myogenic and adipogenic commitment itself. Whether the observed decrease in the fusion index by Mst is a consequence of the inhibition of multipotent stem cell myogenic commitment or of satellite cell activation remains to be determined.

    The stimulation of adipogenesis by recombinant Mst protein, and its inhibition by anti-Mst antibody, supports the hypothesis that Mst may act at the lineage selection point rather than at a later stage because these effects are opposite those reported previously in committed preadipocytes, such as the 3T3-L1 cells in adipogenic medium (9). Further investigation is needed to clarify why the putative inhibitory effects of Mst on preadipocyte differentiation postulated in this earlier paper (9) cannot counteract the earlier stage stimulation that we propose in the current work so that the overall result is a dramatic stimulation of adipogenesis in AZCT-treated C3H 10T(1/2) cells. The effects of growth factors on cell differentiation can vary significantly, depending on external factors such as cell type, density of plating, length of treatment, and medium composition (12). Our data do not allow us to discriminate between the alternative possibilities that Mst triggers lineage commitment in the C3H 10T(1/2) cells or whether it modulates this process after its initiation by other factors. However, we speculate that the modulatory role of Mst is more likely.

    The effects of anti-Mst antibody in the absence of Mst-113 were likely due to the neutralization of endogenous Mst produced by the AZCT-treated cells, as shown by Western blot. We do not know yet whether the full-length protein made by the transfected Mst-375 cDNA is active per se [as in the case of cell replication (22)] or requires cleavage. However, the results at 7 d with both the Mst-375 cDNA and the Mst siRNA were in agreement with the effects of recombinant Mst-113 and polyclonal antibody against Mst, suggesting that either some recombinant protein cleavage may take place with longer time or that the full-length protein is less biologically active than the 110-amino acid carboxy-terminal, Mst protein (22).

    Similarly, we do not know whether there is some endogenous Mst cleavage in the C3H 10T(1/2) cells, but we have been unable to detect it in untransfected HEK and C2C12 cells that express low levels of Mst (22, 24) but do not appear to have furin proteases that cleave Mst (38, 39). The elucidation of this question requires further investigation. In the case of other members of the TGF? family, like BMPs and Mst itself, their effects on C3H 10T(1/2) cells not treated with AZCT have been studied with the recombinant proteins (12), but the relative efficacy of posttranslational processing and autocrine or paracrine modulation mechanisms is unclear.

    However, the presence of the Mst receptor, e.g. the ActRIIB receptor, has been demonstrated specifically in undifferentiated C3H 10T(1/2) cells, using RT-PCR analysis (12). This approach also showed all the relevant receptors for signaling, including the type II receptors T?RII, ActRII, and BMPRII, and the type I receptors activin receptor-like kinase (ALK)2, ALK3, ALK4, and ALK5. The same study demonstrated that Mst efficiently binds to ActRII and forms a heteromeric complex with ALK4 or ALK5, inducing phosphorylation of Smad 2 and 3 to activate a TGF? signaling pathway, and that Mst antagonizes BMP7-induced differentiation in these cells through its binding to ActIIB receptor. The fact that the effects of Mst were investigated in C3H 10T(1/2) cells treated with BMP2 or BMP7, in which the proteins are competing for the same receptor, may explain the discrepancies with our current results in C3H 10T(1/2) cells that were differentiated with AZCT.

    We speculate that our in vitro findings on the C3H 10T(1/2) cells have physiological relevance for the maintenance of postnatal body composition in humans, specifically in the skeletal muscle. We recognize that the data generated in multipotent cell line cannot be extrapolated directly to humans. Further studies are needed to determine the effects of Mst on primary cultures of native stem cells isolated from postnatal skeletal muscle (40, 41, 42). The effects of Mst on multipotent cell differentiation are diametrically opposite the previously reported effects of testosterone and other androgens in this cell line (21, 43).

    Interestingly, androgen response elements have been found in the promoter of the human Mst gene (44), although it is not clear whether androgens reduce Mst expression, in contrast to glucocorticoids that up-regulate Mst levels (24, 44, 45).

    Our results in the C3H 10T(1/2) cell line agree with the fact that skeletal muscle is considerably increased, whereas fat tissue is reduced, when the Mst gene is inactivated in mice (5, 6) and humans (7). Conversely, Mst overexpression in transgenic mice is associated with decrease of muscle mass and increase of fat mass (1, 2). The hypothesis that Mst regulates body composition by modulating the commitment and/or differentiation of mesenchymal multipotent cells provides a unifying explanation for the reciprocal changes in muscle and fat mass in these strains.

    Acknowledgments

    We thank Dr. Vladimir Kolar (BioVendor Laboratory Medicine, Inc., Brna, Czech Republic) for providing the Mst recombinant protein.

    References

    Reisz-Porszasz S, Bhasin S, Artaza JN, Shen R, Sinha-Hikim I, Hogue A, Fielder TJ, Gonzalez-Cadavid NF 2003 Lower skeletal muscle mass in male transgenic mice with muscle-specific over-expression of myostatin. Am J Physiol Endocrinol Metab 285:E876–E888

    Zimmers TA, Davies MV, Koniaris LG, Haynes P, Esquela AF, Tomkinson KN, McPherron AC, Wolfman NM, Lee SJ 2002 Induction of cachexia in mice by systemically administered myostatin. Science 296:1486–1488

    Gonzalez-Cadavid NF, Bhasin S 2004 Role of myostatin in metabolism. Curr Opin Clin Nutr Metab Care 7:451–457

    McPherron AC, Lawler AM, Lee SJ 1997 Regulation of skeletal muscle mass in mice by a new TGF-? superfamily member. Nature 387:83–90

    Lin J, Arnold HB, Della-Fera MA, Azain MJ, Hartzell DL, Baile CA 2002 Myostatin knockout in mice increases myogenesis and decreases adipogenesis. Biochem Biophys Res Commun 291:701–706

    McPherron AC, Lee SJ 2002 Suppression of body fat accumulation in myostatin-deficient mice. J Clin Invest 109:595–601

    Schuelke M, Wagner KR, Stolz LE, Hubner C, Riebel T, Komen W, Braun T, Tobin JF, Lee SJ 2004 Myostatin mutation associated with gross muscle hypertrophy in a child. N Engl J Med 350:2682–2688[Free Full Text]

    Langley B, Thomas M, Bishop A, Sharma M, Gilmour S, Kambadur R 2002 Myostatin inhibits myoblast differentiation by down-regulating MyoD expression. J Biol Chem 277:49831–49840

    Kim HS, Liang L, Dean RG, Hausman DB, Hartzell DL, Baile CA 2001 Inhibition of pre-adipocyte differentiation by myostatin treatment in 3T3–L1 cultures. Biochem Biophys Res Commun 281:902–906

    Lee SJ 2004 Regulation of muscle mass by myostatin. Annu Rev Cell Dev Biol 20:61–86 (Review)

    McCroskery S, Thomas M, Maxwell L, Sharma M, Kambadur R 2003 Myostatin negatively regulates satellite cell activation and self-renewal. J Cell Biol 162:1135–1147

    Rebbapragada A, Benchabane H, Wrana JL, Celeste AJ, Attisano L 2003 Myostatin signals through a transforming growth factor ?-like signaling pathway to block adipogenesis. Mol Cell Biol 23:7230–7242

    Asakura A, Komaki M, Rudnicki M 2001 Muscle satellite cells are multipotential stem cells that exhibit myogenic, osteogenic, and adipogenic differentiation. Differentiation 68:245–253

    Asakura A 2003 Stem cells in adult skeletal muscle. Trends Cardiovasc Med 13:123–128

    Seale P, Rudnicki MA 2000 A new look at the origin, function, and stem-cell status of muscle satellite cells. Dev Biol 218:115–124 (Review)

    Taylor SM, Jones PA 1979 Multiple new phenotypes induced in 10T1/2 and 2T3 cells treated with azacytidine. Cell 17:769–779

    Lassar AB, Paterson BM, Weintraub H 1986 Transfection of a DNA locus that mediates the conversion of 10T1/2 fibroblast to myoblasts. Cell 47:649–656

    Fisher I, Boland G, Tuan RS 2002 Wnt-3A enhances bone morphogenetic protein-2-mediated chondrogenesis of murine C3H 10T1/2 mesenchymal cells. J Biol Chem 277:30870–30878

    Lou J, Xu F, Merkel K, Manske P 1999 Gene therapy: adenovirus-mediated human bone morphogenetic protein-2 gene transfer induces mesenchymal progenitor cell proliferation and differentiation in vitro and bone formation in vivo. J Orthop Res 17:43–50

    Li Y, Foster W, Deasy BM, Chan Y, Prisk V, Tang Y, Cummins J, Huard J 2004 Transforming growth factor-? 1 induces the differentiation of myogenic cells into fibrotic cells in injured skeletal muscle: a key event in muscle fibrogenesis. Am J Pathol 164:1007–1019

    Singh R, Artaza JN, Taylor WE, Gonzalez-Cadavid NF, Bhasin S 2003 Androgens stimulate myogenic differentiation and inhibit adipogenesis in C3H 10T1/2 pluripotent cells through an androgen receptor-mediated pathway. Endocrinology 144:5081–5088

    Taylor WE, Bhasin S, Artaza J, Byhower F, Azam M, Willard Jr DH, Kull Jr FC, Gonzalez-Cadavid N 2001 Myostatin inhibits cell proliferation and protein synthesis in C2C12 muscle cells. Am J Physiol Endocrinol Metab 280:E221–E228

    Gonzalez-Cadavid NF, Taylor WE, Yarasheski K, Sinha-Hikim I, Ma K, Ezzat S, Shen R, Lalani R, Asa S, Mamita M, Nair G, Arver S, Bhasin S 1998 Organization of the human myostatin gene and expression in healthy men and HIV-infected men with muscle wasting. Proc Natl Acad Sci USA 95:14938–14943

    Artaza JN, Bhasin S, Mallidis C, Taylor W, Ma K, Gonzalez-Cadavid NF 2002 Endogenous expression and localization of myostatin and its relation to myosin heavy chain distribution in C2C12 skeletal muscle cells. J Cell Physiol 190:170–179

    Matsuda M, Shimomura I, Sata M, Arita Y, Nishida M, Maeda N, Kumada M, Okamoto Y, Nagaretani H, Nishizawa H, Kishida K, Komuro R, Ouchi N, Kihara S, Nagai R, Funahashi T, Matsuzawa Y 2002 Role of adiponectin in preventing vascular stenosis. The missing link of adipo-vascular axis. J Biol Chem 277:37487–37491

    Zeschnigk M, Kozian D, Kuch C, Schmoll M, Starzinski-Powitz A 1995 Involvement of M-cadherin in terminal differentiation of skeletal muscle cells. J Cell Sci 108:2973–2981

    Dreyfus PA, Chretien F, Chazaud B, Kirova Y, Caramelle P, Garcia L, Butler-Browne, Gherardi RK 2004 Adult bone marrow-derived stem cells in muscle connective tissue and satellite cell niches. Am J Pathol 164:773–779

    Brummelkamp TR, Bernards R, Agami R 2002 A system for stable expression of short interfering RNAs in mammalian cells. Science 296:550–553

    Sui G, Soohoo C, Affar EB, Gay F, Shi Y, Forrester WC, Shi Y 2002 A DNA vector-based RNAi technology to suppress gene expression in mammalian cells. Proc Natl Acad Sci USA 99:5515–5520

    Sharma M, Kambadur R, Matthews K, Somers W, Devlin G, Conaglen J, Fowke P, Bass J 1999 Myostatin transforming growth factor-? superfamily member is expressed in heart muscle and is up-regulated in cardiomyocytes after infarct. J Cell Physiol 180:1–9

    Poleskaya A, Seale P, Rudnicki MA 2003 Wnt signaling induces the myogenic specification of resident CD45+ adult stem cells during muscle regeneration. Cell 113:841–852

    Cao B, Zheng B, Jankowski RJ, Kimura S, Ikezawa M, Deasy B, Cummins J, Epperly M, Qu-Petersen Z, Huard J 2003 Muscle stem cells differentiate into haematopoietic lineages but retain myogenic potential. Nat Cell Biol 5:640–646

    Fukada S, Higuchi S, Segawa M, Koda K, Yamamoto Y, Tsujikawa K, Kohama Y, Uezumi A, Imamura M, Miyagoe-Suzuku Y, Takeda S, Yamamoto H 2004 Purification and cell-surface marker characterization of quiescent satellite cells from murine skeletal muscle by a novel monoclonal antibody. Exp Cell Res 296:245–255

    Kehr J 2003 Single cell technology. Curr Opin Plant Biol 6:617–621

    Rios R, Carneiro I, Arce VM, Devesa J 2001 Myostatin regulates cell survival during C2C12 myogenesis. Biochem Biophys Res Commun 280:561–566

    Rios R, Carneiro I, Arce VM, Devesa J 2002 Myostatin is an inhibitor of myogenic differentiation. Am J Physiol Cell Physiol 282:C993–C999

    Sinha-Hikim I, Artaza J, Woodhouse L, Gonzalez-Cadavid N, Singh AB, Lee MI, Storer TW, Casaburi R, Shen R, Bhasin S 2002 Testosterone-induced increase in muscle size in healthy young men is associated with muscle fiber hypertrophy. Am J Physiol Endocrinol Metab 283:E154–E164

    Takahashi K, Fujita T, Takeuchi T 1995 Production of bioactive enkephalin from the nonendocrine cell lines COS-7, NIH3T3, Ltk-, and C2C12. Peptides 16:933–938

    Gros L, Riu E, Montoliu L, Ontiveros M, Lebrigand L, Bosh F 1999 Insulin production by engineered muscle cells. Hum Gene Ther 10:1207–1217

    Seale P, Ishibashi J, Holterman C, Rudnicki MA 2004 Muscle satellite cell-specific genes identified by genetic profiling of MyoD-deficient myogenic cell. Dev Biol 275:287–300

    Beresford JN 1989 Osteogenic stem cells and the stromal system of bone and marrow. Clin Orthop Relat Res 240:270–280

    Ashton BA, Allen TD, Howlett CR, Eaglesom CC, Hattori A, Owen M 1980 Formation of bone and cartilage by marrow stromal cells in diffusion chambers in vivo. Clin Orthop Relat Res 151:294–307

    Jasuja R, Ramaraj P, Mac RP, Singh AB, Storer TW, Artaza J, Miller A, Singh R, Taylor WE, Lee MI, Davidson T, Sinha-Hikim I, Gonzalez-Cadavid N, Bhasin S 2005 -4-Androstene-3, 17-dione (4-androstenedione) binds androgen receptor, promotes myogenesis in vitro, and increases serum testosterone levels, fat-free mass, and muscle strength in hypogonadal men. J Clin Endocrinol Metab 90:855–863

    Ma K, Mallidis C, Artaza J, Taylor W, Gonzalez-Cadavid N, Bhasin S 2001 Characterization of 5'-regulatory region of human myostatin gene: regulation by dexamethasone in vitro. Am J Physiol Endocrinol Metab 281:E1128–E1136

    Ma K, Mallidis C, Bhasin S, Mahabadi V, Artaza J, Gonzalez-Cadavid N, Arias J, Salehian B 2003 Glucocorticoid-induced skeletal muscle atrophy is associated with up-regulation of myostatin gene expression. Am J Physiol Endocrinol Metab 285:E363–E371(Jorge N. Artaza, Shalende)