Myostatin inhibits adipogenic differentiation of bone marrow-derived hMSC. hMSCs were incubated for 21 days in adipogenic DM with or without myostatin (myst) and anti-myostatin antibody (JA16). A and B, photomicrograph of Oil-red-O stained hMSCs incubated in DM alone (A, bar = 100 μm) or DM containing 1.0 μg/ml myostatin (B). C, hMSC distribution among different differentiation stages in DM (open bar) or in DM containing myostatin (1 μg/ml, dark bar, ^*, p < 0.05, n = 3). Stage I: elongated fibroblast-like cells without microscopically detectable lipid droplets; stage II: flattened cells without detectable lipid droplets; stage III: multiple small lipid droplets (> 12 per cell) that are only visible under high magnification (×250); stage IV: fewer but larger lipid droplets (6–12 per cell) that are detected under lower magnification (×100); stage V: fewer but larger coalescent lipid droplets (3–6 per cell) that are readily detectable at low magnification (×40); stage VI: 1–3 very large coalescent lipid droplet(s) that occupy the majority space within a cell. D, lipid synthesis rate assessed by measuring the incorporation of Bodipy-fatty acid into cellular lipids of hMSCs incubated in DM or DM containing myostatin (a>b>c, p < 0.05, n = 8). E, expression of aP2 and leptin mRNAs in hMSCs (a>b>c, p < 0.05, n = 3); open bar: DM control; dark gray bar: DM containing myostatin (1.0 μg/ml), light gray bar: DM containing myostatin (1.0 μg/ml) and anti-myostatin antibody JA16 (10 μg/ml). F, lipid content of hMSCs measured by Oil-red-O staining and quantification at 500 nm, with myostatin (0.1 μg/ml) added at each indicated time point (a>b>c, p < 0.05, n = 3). G, lipid content in hMSCs treated with DM. Myostatin (0.1 μg/ml) was added to hMSCs together with DM at time 0 and withdrawn at the indicated time points, and the incubation was continued in DM for a total of 21 days. Lipid content was measured by Oil-Red-O staining and quantification at 500 nm (a>b>c, p < 0.05, n = 3).

Myostatin inhibits adipogenic differentiation of human preadipocytes. A and B, phase-contrast photomicrographs of human preadipocytes incubated in differentiating medium (DM) alone (A) or in DM containing 1.0 μg/ml myostatin (B, bar = 100 μm) for 21 days; C, lipid content was measured by Oil-red-O staining on day 21; D, mRNA expression levels of aP2 and leptin in cells treated for 21 days with DM control (open bar) or DM containing myostatin (1.0 μg/ml, dark bar).

Myostatin inhibits adipogenic differentiation downstream from C/EBPβ. A, time course of mRNA expression of C/EBPβ, PPARγ, and C/EBPα, measured by qPCR in hMSCs incubated in basal medium (not shown), DM (open bars), or DM containing myostatin (0.1 μg/ml, dark bars). B, Western analysis of C/EBPβ, aP2, and PPARγ in hMSCs overexpressing a constitutively active C/EBPβ-LAP construct after being treated for 6 days in basal medium, DM, or DM containing myostatin (0.1 μg/ml). Nonspecific bands (NS) are shown as an indication of equal sample loading. C, hMSCs were transfected with adenovirus encoding eGFP or PPARγ, and then treated with DM (open bars) or DM containing myostatin (dark bars). mRNA levels of aP2 and C/EBPα were measured by qPCR.

Smad3 plays a pivotal role in myostatin-induced suppression of adipogenesis. A, Western analysis of Smad3 and phospho-Smad3 in hMSCs grown in DM or DM containing myostatin (0.1 μg/ml) with (DM/Alk5i) or without an Alk5 inhibitor (100 μm) for 3 h. β-Tubulin served as a loading control. B, Western analysis of C/EBPα, PPARγ, and aP2 in hMSCs treated the same as in A but with extended incubation for 48 h. Each blot in both A and B is representative of three independent experiments. C, expression of Smad3 and adipocyte marker genes in hMSCs after RNAi knockdown of Smad3. hMSCs were treated with two different RNAi oligonucleotides (RNAi-1 and RNAi-2) administered separately or in combination (1:1 mixing). Control cells were transfected with a nonspecific 21-mer oligonucleotide. All oligonucleotides were added at a final concentration of 100 nm. The cells were then incubated in DM (open bars) or DM plus myostatin (0.1 μg/ml, dark bars) for 6 days and used for qPCR analysis of Smad3, PPARγ, aP2, and C/EBPα. D, Western analysis of Smad3 and aP2 in cells treated the same as those used for PCR analysis. Two RNAi oligos were mixed in 1:1 ratio in this experiment. Results are representative of two experiments in duplicates.

β-Catenin interacts with Smad3 and acts downstream of Smad3 to mediate the inhibitory effect of myostatin on adipogenesis. A, physical interaction of Smad3 with β-catenin and TCF4. hMSCs were incubated for 3 h in DM, DM plus myostatin (0.1 μg/ml), or DM plus myostatin (0.1 μg/ml) and Alk5 inhibitor (100 μm). Nuclear extracts were immunoprecipitated using anti-Smad3 or anti-β-catenin antibody, and the resulting protein complex was separated by electrophoresis and immunoblotted for the expected binding partner of the target protein. After stripping, both membranes were re-probed for the common binding partner TCF4. Histone-1 was used to demonstrate equal sample input for the immunoprecipitation reactions. B, nuclear distribution of eGFP-tagged β-catenin in hMSCs after 3 h of incubation with DM or DM containing myostatin (0.1 μg/ml). C, Western blot analysis of nuclear and cytosolic β-catenin in hMSCs incubated with DM or DM plus myostatin (0.1 μg/ml) for 24 and 48 h. Histone-1 and β-tubulin were used as loading controls for the nuclear and cytosolic fractions, respectively. D, effect of β-catenin silencing on mRNA expression of adipocyte markers in hMSCs. β-Catenin expression was silenced by two different RNAi oligonucleotides (RNAi-1 and RNAi-2) separately or in combination. Control cells were transfected with a nonspecific 21-mer oligonucleotide. All oligonucleotides were added at a final concentration of 100 nm. hMSCs were then incubated with DM (open bars) or DM plus myostatin (0.1 μg/ml, dark bars) for 6 days and used for qPCR analysis for β-catenin, PPARγ, C/EBPα, and aP2. E, Western analysis of β-catenin and aP2 in cells treated the same as those used for PCR analysis. Two RNAi oligos were mixed in 1:1 ratio in this experiment. Results are representative of two experiments in duplicates.

Under adipogenic differentiation conditions, myostatin does not affect cell cycle progression or DNA synthesis. Confluent hMSCs were treated with basal medium, DM, and DM containing myostatin (0.1 μg/ml). A, cells were harvested after 24 h and subjected to fluorescent-activated cell sorting (FACS) analysis. The area under each phase (G₀/G₁ versus G₂/M) represents the relative number of cells residing in the corresponding phase. The majority of the cells fell in the quiescent G₀/G₁ phase, regardless of the presence of myostatin; B, DNA synthesis assessed by BrdU incorporation in hMSCs incubated in basal medium, DM, or DM containing myostatin (0.1 μg/ml) for different days, as shown.

Dominant-negative TCF4 (dn-TCF) attenuates the inhibitory effect of myostatin on adipogenesis. hMSCs were transfected with an empty control vector (Ad5-Vector), an adenovirus vector encoding a dn-TCF4, or wild-type TCF4. Cells were then induced to differentiate with or without myostatin. A, lipid accumulation in hMSCs treated with Ad5 or dn-TCF4 (final concentration of 5 × 10⁸ pfu/ml for each) after 6 days of incubation with DM or DM containing myostatin (0.1 μg/ml). The cells were stained with Texas Red and photographed under a fluorescent microscope. B, hMSCs were transfected with different doses of dn-TCF4 and then treated with DM (open bars) or DM containing myostatin (0.1 μg/ml, dark bars) for 6 days and the mRNA levels of PPARγ, C/EBPα, and aP2 were measured by qPCR. C, hMSCs transfected with Ad-5 or dn-TCF4 (5 × 10⁸ pfu/ml) were incubated for 6 days with DM or DM containing myostatin at different doses. The mRNA levels of PPARγ, C/EBPα, and aP2 were measured by qPCR. D, ratio between TCF4 reporter gene activity (TOPFLASH luciferase) and the transfection control Renilla luciferase in COS-7 cells in proportion to dn-TCF4 virus concentration (n = 3, mean ± S.E.).