Glycogen Synthase Kinase 3β Suppresses Myogenic Differentiation through Negative Regulation of NFATc3*

Skeletal muscle atrophy is a prominent and disabling feature in many chronic diseases. Prevention or reversal of muscle atrophy by stimulation of skeletal muscle growth could be an important therapeutic strategy. Glycogen synthase kinase 3β (GSK-3β) has been implicated in the negative regulation of skeletal muscle growth. Since myogenic differentiation is an essential part of muscle growth, we investigated if inhibition of GSK-3β is sufficient to stimulate myogenic differentiation and whether this depended on regulation of the transcription factor nuclear factor of activated T-cells (NFAT). In both myogenically converted mouse embryonic fibroblasts and C2C12 myoblasts, deficiency of GSK-3β protein (activity) resulted in enhanced myotube formation and muscle-specific gene expression during differentiation, which was reversed by reintroduction of wild type but not kinase-inactive (K85R) GSK-3β. In addition, GSK-3β inhibition restored myogenic differentiation following calcineurin blockade, which suggested the involvement of NFAT. GSK-3β-deficient mouse embryonic fibroblasts or myoblasts displayed enhanced nuclear translocation of NFATc3 and elevated NFAT-sensitive promoter transactivation, which was reduced by reintroducing wild type, but not K85R GSK-3β. Overexpression of NFATc3 increased muscle gene promoter transactivation, which was abolished by co-expression of wild type GSK-3β. Finally, stimulation of muscle gene expression observed following GSK-3β inhibition was strongly attenuated in NFATc3-deficient myoblasts, indicating that this response requires NFATc3. Collectively, our data demonstrate negative regulation of myogenic differentiation by GSK-3β through a transcriptional mechanism that depends on NFATc3. Inhibition of GSK-3β may be a potential strategy in prevention or treatment of muscle atrophy.

Maintenance of muscle mass is critical for health, and loss of skeletal muscle mass compromises human physical condition and survival in chronic diseases, such as chronic obstructive pulmonary disease (1,2). Restoring lost muscle mass is important for improving quality of life and ultimately disease prognosis (3). To restore muscle mass and improve muscle function in various diseases conditions, a better understanding of the molecular mechanisms of skeletal muscle (re)growth is required. Skeletal muscle differentiation is a critical element of certain types of postnatal growth of the skeletal musculature and is mainly dependent on satellite cells (quiescent myoblasts), which upon activation proliferate, differentiate, and fuse with existing muscle fibers or with each other to form new myofibers (4). Myoblast fusion allows additional muscle growth by myonuclear accretion, beyond the limitations imposed by the myonuclear domain (i.e. the maximal cytoplasm/nucleus ratio) (5).
The protein kinase GSK-3 2 is ubiquitously expressed, and, although it was originally identified as a suppressor of glycogen synthase (6), GSK-3 has been implicated in a myriad of metabolic and signaling pathways (7). Recent studies have identified GSK-3␤ as a negative regulator of both cardiac and skeletal muscle hypertrophy (8,9) as well as muscle differentiation (10).
Regulation of the transcriptional regulator nuclear factor of activated T-cells (NFAT) depends on the balance between inhibitory phosphorylation by GSK-3␤ and stimulatory dephosphorylation by the calcium-dependent serine/threonine phosphatase calcineurin. Activated calcineurin exposes the nuclear localization signal of NFAT, resulting in its nuclear translocation (11). Conversely, GSK-3␤-dependent phosphorylation masks the nuclear localization signal, resulting in nuclear export of NFAT and termination of calcineurin-induced gene transcription (12,13).
In this study, we hypothesized that inhibition of GSK-3␤ increases myogenic differentiation by promoting NFATc3 nuclear localization and transcriptional activity. Two different myogenic models were employed to examine the effects of genetic modulation of GSK-3␤ on skeletal muscle differentiation and NFAT regulation. The results revealed negative regulation of myogenic differentiation by GSK-3␤ through suppression of NFATc3 nuclear localization and subsequent inhibition of NFATc3-mediated muscle gene transcription.

EXPERIMENTAL PROCEDURES
Cell Culture-The murine skeletal muscle cell line C2C12 obtained from the American Type Culture Collection (ATCC number CRL1772) was cultured in growth medium (GM), composed of low glucose Dulbecco's modified Eagle's medium (DMEM) containing antibiotics (50 units/ml penicillin and 50 g/ml streptomycin) and 9% (v/v) fetal bovine serum (FBS) (all from Invitrogen), or differentiation medium (DM), which contained low glucose DMEM with 0.5% heat-inactivated FBS and antibiotics. Wild type and GSK-3␤ Ϫ/Ϫ mouse embryo fibroblast (MEF) (26) cell lines were generously provided by Dr. J. R. Woodgett (Ontario Cancer Institute, Toronto, Canada) and cultured in GM or DM, which contained high glucose DMEM with 0.5% heat-inactivated FBS and antibiotics. Both cell types were grown on Matrigel (BD Biosciences)-coated (1:50 in DMEM) dishes as described previously (27). Cells were plated at 10 4 /cm 2 and cultured in GM for 24 h before transfection or induction of differentiation. When applicable, lithium chloride (LiCl) (Sigma) or FK506 (Fujisawa Pharmaceuticals) was added directly after induction of differentiation and again 24 h later when the cells were provided with fresh DM.
RNA Interference and Retroviral Infection-Vectors expressing hairpin small interference RNAs were constructed by inserting pairs of annealed, HindIII/BglII-digested oligonucleotides containing the 19-nt target sequence into the pRetro-Super (pRS) vector (31,32), kindly provided by Dr. R. Agami (The Netherlands Cancer Institute, Amsterdam). Ecotropic retroviral supernatants were produced by transfection of phoenix packaging cells (Dr. G. Nolan, Stanford University), cultured in high glucose DMEM containing antibiotics (50 units/ml penicillin and 50 g/ml streptomycin) and 9% (v/v) FBS, with the pRS GSK-3␤ shRNA vector or the pRS control vector by calcium phosphate precipitation. 48 h post-transfection, the tissue culture medium was filtered through a 0.45-m filter, and supernatant containing the viral particles was used for infection of C2C12 cells after the addition of 4 g/ml Polybrene (Sigma). Cells were infected for at least 6 h and allowed to recover for 24 h with fresh GM. Infected cells were selected with puromycin (2.5 g/ml) for at least 96 h, and silencing efficacy was evaluated by Western blot analysis for GSK-3␤ or NFATc3, respectively, and tubulin as a loading control. Of the three target sequences tested, the pRS vector containing the 19-nt target sequence 5Ј-GTTGTATATGTATCAGCTG-3Ј demonstrated the strongest RNA silencing for GSK-3␤, whereas the targeting sequence 5Ј-TACTAGAGTCCGACTTGTA-3Ј proved most successful in NFATc3 silencing.
Total protein was assessed by the Bio-Rad DC protein assay kit (Bio-Rad) according to the manufacturer's instructions, and 5-30 g of protein was loaded per lane and separated on a 7 or 10% polyacrylamide gel (Mini Protean 3 System; Bio-Rad), followed by electroblot transfer to a 0.45-m nitrocellulose membrane (Schleicher & Schuell). The membrane was blocked for 1 h at room temperature in 5% (w/v) nonfat, dried milk. Nitrocellulose blots were washed in PBS-Tween 20 (0.05%), followed by overnight incubation (4°C) with a polyclonal antibody specific for either GSK-3␤ 1:1000 (catalog number 9315; Cell Signaling, Danvers, MA) or NFATc3 1:500 (catalog number sc-8321; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or a monoclonal antibody ␣-tubulin 1:500 (catalog number sc-8035; Santa Cruz Biotechnology). After three wash steps of 20 min each, the blots were probed with a peroxidase-conjugated secondary antibody, 1:5000 (Vector Laboratories, Burlingame, CA), and visualized by Supersignal WestPico chemiluminescent substrate (Pierce) according to the manufacturer's instructions.
Assessment of Myogenic Differentiation-Myogenic differentiation of C2C12 cells was assessed biochemically via determination of muscle creatine kinase (MCK) activity and morphologically by determination of the myogenic index. For MCK activity, cells grown and differentiated in 35-mm dishes for 72 h were washed twice in cold PBS, lysed in 0.5% Triton X-100, and scraped off of the dish with a rubber policeman. Lysates were centrifuged for 2 min at 16,000 ϫ g, and the supernatant was stored in separate aliquots at Ϫ80°C for determination of protein content or MCK activity in the presence of 1% bovine serum albumin. MCK activity was measured using a spectophotometry-based (33) kit from Stanbio (Stanbio, Boerne, TX). Specific MCK activity was calculated after correction for total protein, which was assessed by the Bradford method (34).
To quantify myoblast fusion, the myogenic index, defined as the number of nuclei residing in cells containing three or more nuclei divided by the total number of nuclei, was determined in May-Grunwald Giemsa-stained cells. Cells were grown on Matrigel-coated 60-mm dishes, and after 72 h in differentiation medium, cells were washed twice in PBS (room temperature), fixed in methanol and stained in May-Grunwald Giemsa (Sigma) according to the manufacturer's instructions. The myogenic index was assessed by counting five fields/60-mm dish (n ϭ 3) at a ϫ100 magnification, and the total number of nuclei analyzed was 300 -500/field. RNA Isolation and Assessment of mRNA Abundance-Total RNA from C2C12 cells was isolated using the Totally RNA TM kit (Ambion, Austin, TX) according to the manufacturer's instructions. After isolation, RNA was dissolved in 1 mM sodium citrate (pH 6.4) and stored at Ϫ80°C. One g of RNA was reverse transcribed to cDNA using the Reverse iT First Strand Synthesis kit (ABgene, Epsom, UK) with anchored oligo(dT) primers. MCK and ␤-actin mRNA were determined by quantitative reverse transcription-PCR. Quantitative reverse transcription-PCR primers were designed using Primer Express 2.0 software (Applied Biosystems, Foster City, CA) and obtained from Sigma Genosys (Haverhill, UK). MCK was amplified using the following primers: MCK FP, 5Ј-AGGTTT-TCCGCCGCTTCT-3Ј; RP, 5Ј-CGGTGCCCAGGTTGGA-3Ј. PCRs (25 l total volume) contained 1ϫ MasterMix Plus for SYBR green I (Eurogentec, Seraing, Belgium) and primers (300 nM). Standard curves were made in duplicate by performing serial dilutions of pooled cDNA aliquots. Ct values were obtained for each sample, and the relative DNA concentrations were derived from the standard curve. The expression of the genes of interest was normalized to ␤-actin (primers obtained from Ambion). Real time PCRs were performed in an ABI PRISM TM 7700 Sequence Detector (Applied Biosystems).
Immunofluorescence-GSK-3␤ shRNA or control pRS-C2C12 myoblasts were grown on glass coverslips and were fixed and stained after 24 or 48 h of culture in DM for NFATc3 using a polyclonal antibody 1:100 (catalog number sc-8321; Santa Cruz Biotechnology) and a fluorescein isothiocyanate-fluorophore-conjugated anti-rabbit secondary antibody, 1:1000 (Molecular Probes, Leiden, The Netherlands). Nuclei were counterstained with 4Ј,6-diamidino-2phenylindole (20 g/ml). Images were taken at a ϫ400 magnification, using a fluorescent microscope connected to a digital DXM 1200F camera, both from Nikon (Kanagawa, Japan). MyoD is a useful tool to study activation of muscle-specific gene transcription (35). To address the suppressive effects of endogenous GSK-3␤ on myogenesis by genetic rather than pharmacological modulation, we compared myogenic conversion of GSK-3␤ Ϫ/Ϫ and WT MEFs by measuring transactivation of the troponin-I (TnI) promoter. MEFs lacking endogenous GSK-3␤ displayed increased TnI promoter transactivation compared with WT MEFs (Fig. 1A). Moreover, restoring GSK-3␤ expression by co-transfecting GSK-3␤ Ϫ/Ϫ MEFs with a plasmid encoding WT GSK-3␤ decreased TnI promoter transactivation, whereas expression of a kinase-dead mutant (K85R) did not show any effects (Fig. 1B). This finding demonstrates that GSK-3␤ kinase activity suppresses muscle-specific gene transcription, suggesting a negative regulatory role for GSK-3␤ in myogenic differentiation.
Inhibition of GSK-3␤ Overcomes Repression of Myogenic Differentiation by Calcineurin Blockade-The activity of the transcriptional regulator NFAT is dependent on the balance between inhibitory phosphorylation by GSK-3␤ and stimulatory dephosphorylation by calcineurin (12). Previous reports have revealed that calcineurin activity is required for muscle differentiation (22). To establish whether pharmacological inhibition of calcineurin resulted in repression of myogenic differentiation in our model, C2C12 myoblasts were cultured in DM for 72 h with or without the calcineurin inhibitor FK506, and MCK activity was assessed. MCK was significantly decreased in a dosedependent fashion in response to FK506 (Fig. 3A). In contrast, pharmacological inhibition of GSK-3␤ by LiCl resulted in increased MCK activity (Fig. 3B) and MCK mRNA abundance (Fig. 3C) (10). Simultaneous inhibition of GSK-3␤ and calcineurin completely prevented the repressive effect of calcineurin blockade on myogenic differentiation based on assessment of MCK enzyme activity (Fig. 3B) and MCK mRNA abundance (Fig. 3C). These data demonstrate that calcineurin and GSK-3␤ have opposing effects on myoblast differentiation, which may be conveyed through their mutual phosphosubstrate NFAT, which has been implicated in myogenic differentiation (24).

GSK-3␤ Suppresses NFATc3 Nuclear Localization in Skeletal
Muscle Cells-Next, NFATc3 nuclear localization was assessed in myoblasts by immunostaining during differentiation. Nuclei were counterstained with 4Ј,6-diamidino-2-phenylindole. pRS-shGSK-3␤ C2C12 cells showed a marked increase in NFATc3 nuclear localization compared with control pRS cells in which NFATc3 mainly resided in the cytoplasm (Fig. 4A). Increased NFATc3 nuclear abundance in GSK-3␤ shRNA compared with control pRS-C2C12 cells was confirmed by Western blot anal-ysis of nuclear extracts (Fig. 4B). Total NFATc3 abundance in whole cell extracts was not different between the cells (data not shown). This finding demonstrates that increased myogenic differentiation of myoblasts in absence of GSK-3␤ is associated with increased NFATc3 nuclear localization.
GSK-3␤ Controls NFATc3 Transcriptional Activation-To further examine the functional consequences of NFATc3 nuclear exclusion by GSK-3␤, NFATc3 was transiently overexpressed in GSK-3␤ Ϫ/Ϫ and WT MEFs together with a NFATsensitive promoter reporter construct. Overexpression of NFATc3 induced NFAT transcriptional activation to greater extent in GSK-3␤-deficient compared with WT MEFs (Fig. 5A). As expected, pharmacological inhibition of GSK-3␤ by LiCl further enhanced NFAT transcriptional activity in WT MEFs but not in GSK-3␤ Ϫ/Ϫ MEFs (Fig. 5A). To evaluate whether NFAT transcriptional activity is also suppressed by GSK-3␤ in skeletal muscle cells, C2C12 cells were transiently transfected with the NFAT reporter construct and NFATc3. Overexpression of NFATc3 caused a strong increase of NFAT transcriptional activation, which was further increased by inhibition of endogenous GSK-3␤ using LiCl (Fig. 5B). Co-transfection of NFATc3 with plasmids encoding WT GSK-3␤ but not kinase-inactive (K85R) GSK-3␤ decreased NFAT transcriptional activation, whereas transcriptional activity of a nonphosphorylatable mutant (ca) of NFATc3 was not affected by GSK-3␤ over-expression (Fig. 5C). Combined, these data position NFAT downstream of GSK-3␤ signaling and show a direct regulatory role of GSK-3␤ in NFATc3-mediated transcriptional activity in C2C12 cells.   NFAT-dependent Muscle-specific Gene Expression Is Inhibited by GSK-3␤-To examine whether NFATc3 positively affects myogenic differentiation, C2C12 cells were transiently transfected with the TnI promoter reporter construct to assess muscle-specific gene expression during differentiation. Co-expression of NFATc3 increased TnI promoter transactivation after 24 h (not shown) and 48 h following induction of differen-tiation (Fig. 6A). TnI promoter transactivation was further enhanced by GSK-3␤ inhibition (Fig. 6A). NFATc3 overexpression also increased TnI promoter transactivation following myogenic conversion of GSK-3␤ Ϫ/Ϫ MEFs (Fig. 6B). Reintroducing WT, but not K85R GSK-3␤, repressed NFATc3-stimulated TnI promoter transactivation. In contrast, increased TnI promoter activity in response to ca NFAT overexpression was refractory to GSK-3␤ (Fig. 6B). Importantly, TnI promoter transactivation was also increased in differentiating C2C12 myoblasts when NFATc3 was overexpressed, which was in turn suppressed by WT but not K85R GSK-3␤ (Fig. 6C). Overexpression of ca NFATc3 also stimulated muscle-specific gene expression but was insensitive to GSK-3␤ (Fig. 6C). These results show that muscle gene expression during myogenic differentiation is promoted by NFATc3, which in turn is negatively regulated by GSK-3␤.
Stimulation of Muscle Gene Expression during Myogenic Differentiation by GSK-3␤ Inhibition Requires NFATc3-Finally, we investigated whether stimulation of myogenic differentiation by inhibition of GSK-3␤ is dependent on NFATc3 signaling. Using the same RNA interference approach as for GSK-3␤ knockdown, C2C12 cell lines harboring a stably integrated NFATc3 targeting sequence (pRS-NFATc3 shRNA) were generated. Compared with control cells, NFATc3 protein abundance was efficiently reduced (ϳ75%) in pRS-NFATc3 shRNA myoblasts (Fig. 7A). Next, these and control myoblasts were transiently transfected with the TnI promoter reporter construct to assess muscle-specific gene expression during differentiation. Differentiation-induced TnI promoter transactivation was strongly attenuated in myoblasts with reduced NFATc3 levels compared with control myoblasts (2-fold versus 7-fold over GM at 48 h of DM, respectively). Importantly, inhibition of GSK-3␤ using LiCl stimulated TnI transactivation more potently in control myoblasts (4-fold at 48 h in DM) compared with pRS-NFATc3 shRNA myoblasts (2-fold at 48 h in DM). These data demonstrate that stimulation of muscle gene expression during differentiation by inhibition of GSK-3␤ is dependent on NFATc3.

DISCUSSION
Skeletal muscle growth involves different processes, including a net increase in protein synthesis as well as proliferation, differentiation, and fusion of satellite cells and simultaneous expression of muscle-specific genes (5,36). These cellular and intracellular processes are coordinated at multiple levels by insulin-like growth factor-I (IGF-I)/Akt/mammalian target of rapamycin and IGF-I/Akt/GSK-3␤ signaling pathways (9,37). Various studies have shown that GSK-3␤ is a negative regulator of both cardiac and skeletal muscle growth (9,38,39). Recently, GSK-3␤ inactivation was associated with myonuclear accretion and myogenic differentiation in skeletal muscle recovering from atrophy (10). In addition, in vitro studies have suggested that GSK-3␤ suppresses myogenesis, since pharmacological inhibition of GSK-3␤ was reported to stimulate myogenic differentiation (10,41). The results presented in the current study extend these observations, since loss of endogenous GSK-3␤ was demonstrated to stimulate muscle-specific gene expression and myotube formation during differentiation and implicate NFAT-sensitive promoter-luciferase reporter and ␤-galactosidase (0.25 g each) constructs were transfected in GSK-3␤ Ϫ/Ϫ or GSK-3␤ ϩ/ϩ (WT) MEFs (A) or C2C12 myoblasts (B) with or without a plasmid encoding NFATc3 (0.5 g). After incubation in DM for 48 h with or without LiCl (5 mM), cells were lysed to measure luciferase and ␤-galactosidase activity. Alternatively, C2C12 myoblasts were transfected with NFAT-luciferase, ␤-galactosidase, NFATc3, or ca NFAT (0.5 g) and/or WT or K85R GSK-3␤ (1.0 g)-encoding plasmids. Following incubation in DM for 48 h, cells were lysed to measure luciferase and ␤-galactosidase activity (C). Shown are representative graphs of three independent experiments (n ϭ 3 Ϯ S.E.). regulation of NFATc3 nuclear translocation and transcriptional activity as the mechanism of suppression of muscle gene expression by GSK-3␤ during differentiation.
In the present study, ablation of GSK-3␤ signaling increased muscle-specific promoter transactivation, muscle gene expression, and myotube formation in two independent models of myogenic differentiation. This is in line with the stimulatory effect of IGF-I-mediated or pharmacological inhibition of GSK-3␤ activity on myogenic differentiation reflected by increased myotube formation and enhanced expression of musclespecific genes such as TnI, slow and fast, MCK, and MyoD (10). Conversely, reintroduction of enzymatically active GSK-3␤, but not kinase-inactive GSK-3␤, reduced muscle-specific gene expression, indicating that GSK-3␤ kinase activity is required for its suppressive effects on myogenesis.
Previous reports have demonstrated that stimulation of myogenic differentiation can be accomplished via activation of calcineurin, which antagonizes the phosphorylation of certain GSK-3␤ substrates (15,17,24). Conversely, inhibition of calcineurin by cyclosporin A treatment suppressed biochemical and morphological differentiation of skeletal muscle cells (17). In line with these findings, calcineurin inhibition using cyclosporin A (data not shown) or FK506 blocked myotube formation (data not shown) and accumulation of MCK mRNA and enzyme activity.
Interestingly, pharmacological inhibition by LiCl or genetic ablation (data not shown) of GSK-3␤ kinase activity not only stimulated MCK expression and myotube formation (10) but also completely restored the adverse effect of calcineurin inhibition on myogenic differentiation, suggesting a dominant role of GSK-3␤ over calcineurin in muscle growth. In line with these observations, calcineurin-induced cardiac hypertrophy is completely prevented by simultaneous cardiac specific activation of GSK-3␤ (42). These antagonizing actions of GSK-3␤ and calcineurin on muscle growth may be conveyed through their mutual phosphosubstrate NFAT (12,13). NFAT activity is mainly determined by control of its nuclear localization, which is subject to regulatory phosphorylation of the NFAT homology region by kinases, including GSK-3␤, and dephosphorylation by calcineurin.
In the present study, C2C12 myoblasts with reduced GSK-3␤ levels revealed increased nuclear localization and transcriptional activation of NFATc3 during differentiation, indicative of negative regulation of NFATc3 by GSK-3␤. This is in line with the findings of Diehn et al., describing prolonged nuclear localization of NFAT when GSK-3␤ was inhibited (43). Moreover, our data revealed that increased muscle gene expression and myotube formation in either GSK-3␤ shRNA myoblasts (not shown) or LiCl-treated control myoblasts was insensitive to calcineurin inhibition, suggesting that NFAT nuclear translocation occurs by default in the absence of both GSK-3␤ kinase and calcineurin phosphatase activity, as was reported previously (44). Nuclear exclusion of NFATc3 by GSK-3␤ probably required phosphorylation of the NFAT homology region, since NFAT transcriptional activation by an NFATc3 mutant protein lacking this region (ca NFAT) was refractory to the inhibitory effects of GSK-3␤ in either fibroblasts or myoblasts.
The current study does not exclude regulation of other NFAT isoforms by GSK-3␤ during differentiation, since four NFAT isoforms, NFATc1 to -c3 and NFAT5, are expressed in skeletal muscle (17,18). Although NFAT5, -c2, and -c3 have all been attributed a role in the regulation of myoblast differentiation, we focused on NFATc3 in this work for a number of reasons. First, NFAT5 is not subject to calcineurin regulation, yet our data and other studies (19,24) imply that calcineurin signaling is required for basal differentiation. Conversely, stimulation of myogenesis was also observed in these studies by overexpression of ca calcineurin or calcineurin activation by increasing intracellular calcium levels. Moreover, previous studies have postulated a specific role for NFATc3 during stimulation of myoblast differentiation (17,24). In line with these reports, myoblasts with reduced NFATc3 protein levels displayed decreased TnI promoter transactivation during differentiation. Together with our data, this suggests a model of NFATc3 regulation in which the balance between inhibitory phosphorylation by GSK-3␤ and stimulatory dephosphorylation by calcineurin is in favor of GSK-3␤-mediated nuclear exclusion of NFATc3 during basal, nonstimulated differentiation. In contrast, signals increasing calcineurin activity (e.g. calcium) or decreasing GSK-3␤ activity (e.g. IGF-I) result in NFATc3 nuclear translocation and stimulation of myogenic differentiation.
In support of this model, increased NFATc3 nuclear localization coincided with up-regulation of various muscle specific mRNA transcripts after either pharmacological inhibition (10) or knockdown of GSK-3␤. This positions the effects of GSK-3␤ at the pretranslational level and suggests stimulation of musclespecific gene transcription by NFATc3. Indeed, overexpression of NFATc3 was sufficient to stimulate transactivation of the TnI promoter during muscle differentiation. Moreover, NFATc3 was subject to negative regulation by GSK-3␤ kinase activity, since simultaneous overexpression of GSK-3␤ prevented the stimulatory effects of NFATc3 on muscle gene expression. Finally, stimulation of muscle gene expression during myogenic differentiation by GSK-3␤ inhibition required NFATc3, since the induction of transcriptional activation of TnI promoter in response to LiCl was strongly attenuated in myoblasts in which NFATc3 expression was silenced.
Participation of NFAT in muscle-specific gene expression has been documented previously and occurs in association with other transcription factors, such as GATA2 (15), Myf5 (45), and MEF2 (46). In addition, NFATc3 was shown to enhance the myogenic activity of MyoD (24). Since NFATc1 and -c4 did not have this effect (24), this further supports a specific role for NFATc3 in stimulation of muscle differentiation.
NFATc3 silencing did not completely block enhanced differentiation following GSK-3␤ inhibition (Fig. 7B). This may have resulted from the presence of residual NFATc3 protein (Fig.  7A). Alternatively, additional signaling modules that are negatively regulated by GSK-3␤ may contribute to increased differentiation in the absence of GSK-3␤ kinase activity, besides NFAT-dependent transcription. For example, eukaryotic initiation factor 2B is negatively regulated by GSK-3␤ (47), and inactivation of GSK-3␤ during IGF-I-induced muscle hypertrophy is associated with increased protein synthesis (9), which . Stimulation of muscle gene expression during myogenic differentiation by GSK-3␤ inhibition requires NFATc3. RNA interference against NFATc3 was achieved in C2C12 myoblasts as described under "Experimental Procedures." pRS-control or pRS-NFATc3 shRNA myoblasts were drug-selected for 96 h and passaged when appropriate. Cells were cultured in GM for 24 h, and soluble protein (15 g) from lysates was separated by SDS-PAGE to assess NFATc3 or tubulin (loading control) abundance by Western blot analysis (A). pRS-control or pRS-NFATc3 shRNA myoblasts were transiently transfected with a TnI promoter luciferase reporter construct and a plasmid encoding ␤-galactosidase (0.25 g each). After incubation in GM or DM for the indicated times with or without LiCl (5 mM), cells were lysed to measure luciferase and ␤-galactosidase activity. TnI promoter transactivation levels were normalized to values obtained in GM for the individual cell lines (B).
also occurs during myogenic differentiation (10). In addition, ␤-catenin, a crucial major downstream affector molecule in Wnt signaling is rapidly degraded upon phosphorylation by GSK-3␤ (48). Interestingly, ␤-catenin has been shown to regulate several myogenic proteins, such as MyoD and myogenin (49), suggesting a role for ␤-catenin in myogenesis (40), which may be stimulated in the absence of GSK-3␤. Therefore, we propose that GSK-3␤ may act as a central mediator of myogenic differentiation, since differentiation-promoting stimuli, such as calcium/calcineurin, IGF-I, and Wnt, converge on GSK-3␤, which may subsequently control multiple regulatory steps of myogenic differentiation. In conclusion, this study identifies GSK-3␤ as a potential target for stimulation of myogenic differentiation via NFATc3 to enhance skeletal muscle growth or promote recovery from muscle atrophy.