p38 mitogen-activated protein kinase pathway promotes skeletal muscle differentiation. Participation of the Mef2c transcription factor.

Differentiation of muscle cells is regulated by extracellular growth factors that transmit largely unknown signals into the cells. Some of these growth factors induce mitogen-activated protein kinase (MAPK) cascades within muscle cells. In this work we show that the kinase activity of p38 MAPK is induced early during terminal differentiation of L8 cells. Addition of a specific p38 inhibitor SB 203580 to myoblasts blocked their fusion to multinucleated myotubes and prevented the expression of MyoD and MEF2 family members and myosin light chain 2. The expression of MKK6, a direct activator of p38, or of p38 itself enhanced the activity of MyoD in converting 10T1/2 fibroblasts to muscle, whereas treatment with SB 203580 inhibited MyoD. Several lines of evidence suggesting that the involvement of p38 in MyoD activity is mediated via its co-activator MEF2C, a known substrate of p38, are presented. In these experiments we show that MEF2C protein and MEF2-binding sites are necessary for the p38 MAPK pathway to regulate the transcription of muscle creatine kinase reporter gene. Our results indicate that the p38 MAPK pathway promotes skeletal muscle differentiation at least in part via activation of MEF2C.


Differentiation of muscle cells is regulated by extracellular growth factors that transmit largely unknown signals into the cells. Some of these growth factors induce mitogen-activated protein kinase (MAPK) cascades within muscle cells.
In this work we show that the kinase activity of p38 MAPK is induced early during terminal differentiation of L8 cells. Addition of a specific p38 inhibitor SB 203580 to myoblasts blocked their fusion to multinucleated myotubes and prevented the expression of MyoD and MEF2 family members and myosin light chain 2. The expression of MKK6, a direct activator of p38, or of p38 itself enhanced the activity of MyoD in converting 10T1/2 fibroblasts to muscle, whereas treatment with SB 203580 inhibited MyoD. Several lines of evidence suggesting that the involvement of p38 in MyoD activity is mediated via its co-activator MEF2C, a known substrate of p38, are presented. In these experiments we show that MEF2C protein and MEF2-binding sites are necessary for the p38 MAPK pathway to regulate the transcription of muscle creatine kinase reporter gene. Our results indicate that the p38 MAPK pathway promotes skeletal muscle differentiation at least in part via activation of MEF2C.
The development of skeletal muscle is a multistep process in which pluripotent mesodermal cells give rise to myoblasts that subsequently withdraw from the cell cycle and differentiate into myotubes. These stages are driven by the expression and activity of myogenic transcription factors from the MyoD family (1). Two members of the MyoD family, Myf5 and MyoD, are expressed in dividing myoblasts that are committed to the myogenic lineage. However, these proteins are not functional in myoblasts, and their activities are induced only at a subsequent stage to allow withdrawal from the cell cycle and terminal differentiation (2). Induction of the activity of MyoD and Myf5 proteins may be a result of the activity of extracellular growth factors like insulin and insulin-like growth factors known to promote terminal differentiation of myoblasts (3,4). Insulin and insulin-like growth factors are involved in the activation of phosphatidylinositol 3-kinases and mitogen-activated protein kinases (MAPK) 1 via tyrosine kinase receptors within many cell types including muscle (5,6). One MAPK induced by insulin is p38 MAPK (6). p38 is also activated by exposure of cells to environmental stress or by treatment of cells with pro-inflammatory cytokines (7,8). The affectors regulating p38 are only partly explored. However, the direct intracellular activators of p38 are MKK3 and MKK6 (MAPK kinase) (9,10). The role of p38 in skeletal muscle differentiation is not known. Several recent findings suggest that p38 MAPK may be involved in skeletal muscle differentiation: 1) Transcripts of MKK6 gene are most abundant in skeletal muscle (10). 2) Insulin that promotes skeletal muscle differentiation also induces p38 activity (6). 3) p38 pathway regulates the expression of glucose transporters in skeletal muscle cells (11). 4) p38 activates directly the transcription factor MEF2C in inflammation (12). Together with the MyoD family, members of the MEF2 family are necessary for the differentiation of myoblasts (13). 5) p38 pathway plays a role in specific gene expression and cell hypertrophy of the related cardiac muscle lineage (14). In this report we show that 1) the p38 pathway plays an essential role in the in vitro differentiation of myoblasts and 2) that this role may be mediated via the MEF2C transcription factor.

Materials
SB 203580 was a product of Calbiochem. Goat polyclonal antibody raised against amino acids 341-360 of mouse p38 was obtained from Santa Cruz Biotechnology. Rabbit polyclonal antibody that detects p38 MAPK only when activated by dual phosphorylation at Thr 180 and Tyr 182 was purchased from New England Biolabs. Antibodies to MEF2A and C proteins and to GAL4 (amino acids 94 -147) were purchased from Santa Cruz Biotechnology. Protein A-Sepharose was supplied by Sigma.
Plasmids pEMSV-MyoD was described by Tapscott et al. (15). The 4R-tk-CAT reporter gene was described by Weintraub et al. (16). pe ϩ AT-CAT and p( ϩ enh110)-80MCK-CAT were described by Buskin and Hauschka (17). For MEF2-80MCK-CAT, three copies of a double-stranded oligonucleotide of the MEF2 site (30 base pairs) from the MCK enhancer were inserted upstream to a minimal MCK promoter (Ϫ80 to ϩ7 relative to the transcription start site). pGEX-ATF2 plasmid was a gift from Dr. Ami Aronheim and Dr. Michael Karin. The wild type (MKK6) and activated (MKK6b(E)) alleles of MKK6 were described by Han et al. (10). The MEF2C expression vector (pCDNAI-MEF2C) was a generous gift of Dr. E. Olson (18). The inactive form of MEF2C was generated by deleting a BglII fragment that removed amino acids 202-321 of MEF2C. GAL4-MEF2C constructs were described by Han et al. (12), and GAL4-MyoD constructs were described by Weintraub et al. (19). pGEX-MEF2C was constructed by inserting a polymerase chain reaction fragment encoding amino acids 128 -467 of MEF2C into the pGEX-2T vector.

Cell Culture
L8 cells were a gift of Dr. David Yaffe (20). 10T1/2 cells were obtained from ATCC. 10T1/2 cells that expressed the MyoD-estrogen receptor (ER) chimera protein were described by Hollenberg et al. (21). Cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 15% calf serum (Hyclone), penicillin, and streptomycin (growth medium). To induce differentiation, we used Dulbecco's modified Eagle's medium supplemented with 10 g of insulin/ml and 10 g of transferrin/ml (differentiation medium, DM). Differentiation of 10T1/2 cells that expressed MyoD-ER protein was induced by the addition of DM that contained 10 Ϫ7 M estradiol.

Growth of Cells in the Presence of SB 203580
SB 203580 was dissolved in Me 2 SO to a concentration of 10 mM and was added directly to the differentiation medium to a final concentration of 10 -20 M as indicated. Control cells were incubated with the same volumes of Me 2 SO without SB 203580. The medium was replaced every 12 h with medium containing fresh SB 203580.

Transfections
Transfections were performed by calcium phosphate precipitation as described (22). Cells in 6-cm TC dishes (Corning) were transfected for 12 h with a total amount of 10 g of the following plasmid DNA: 1 g of pCMV-LacZ, 3 g of CAT reporter gene, 3 g of MyoD expression plasmid, 3 g of MEF2C expression vector, and 3 g of expression vector of wt MKK6 or activated form of MKK6b (MKK6b(E)). Following transfection, the medium was replaced with either growth medium or DM for another 24 -48 h. The efficiency of transfections were tested in soluble 5-bromo-4-chloro-3-indolyl ␤-D-galactopyranoside assays as described (23), and the amount of extracts used for the CAT assays were adjusted accordingly (24).

Immunohistochemical Staining
Cells were fixed and immunostained as described (25). The primary antibodies used were monoclonal anti-MyoD (5.8A) and polyclonal anti-MHC (Sigma). The immunochemically stained cells were viewed at ϫ200 magnification in a fluorescence microscope (Olympus, model BX50).

In Vitro Kinase Assays
Expression of GST Fusion Proteins-The GST-ATF2 and GST-MEF2C (128 -467) proteins were expressed from the bacterial expression vectors in BL21 strain of Escherichia coli and purified from extracts on glutathione beads as described (26).
Preparation of Cell Extracts-Cell extracts were prepared as described (25).
In Vitro Kinase Assay for p38 -Cells were extracted in lysis buffer (described above). Equal amounts of protein from each time point of differentiation were rotated with 6 l of anti-p38 antibody (Santa Cruz) for 2.5 h at 4°C. The immunoprecipitation procedure and the kinase assay were performed as described for ERK (25). However, the substrate in the present assay was GST-ATF2 (20 g).
In Vitro Kinase Activity of p38 Using GST-MEF2C as a Substrate-Equal amounts of cell extracts were collected at different times after the initiation of cell differentiation and were mixed with GSH-agarose beads to which GST or GST-MEF2C proteins were bound. The mixtures were processed, and the kinase assay was performed as described by Hibi and colleagues for the JNK kinase assay (26). Results were quantified by PhosphorImager.

Western Analysis
Cells were lyzed as described for the kinase assays, and equal amounts of extracted proteins were loaded and separated by SDS-PAGE and transferred to nitrocellulose filters. Immunoblotting was conducted with the following two antibodies: anti-p38 (Santa Cruz) 1:100 and anti-phospho-p38 antibody (New England Biolabs) 1:100. Proteins were visualized using the enhanced chemiluminescence kit of Amersham Pharmacia Biotech.
Labeling with [ 32 P]Orthophosphate-L8 cells were incubated in Dulbecco's modified Eagle's medium without sodium phosphate to which [ 32 P]orthophosphate was added at 0.75 mCi/ml for 3 h before proteins were extracted in RIPA buffer.
Immunoprecipitation-Equal amounts of labeled proteins were incubated with antibody for 2 h and then for an additional hour with protein A-Sepharose. The protein A-Sepharose beads were washed four times with RIPA containing 0.5 M NaCl and once in RIPA.

p38 Activity Is Induced during Differentiation of L8 Cells-
The in vitro kinase activity of p38 was studied using extracts of L8 muscle cells from different stages of differentiation. Kinase activity of p38 was analyzed by immunoprecipitation of p38, which was used to phosphorylate its substrate, ATF2 (Fig. 1A). The kinase activity of p38 was present in dividing myoblasts and was gradually induced during the differentiation to multinucleated myotubes. Kinase activity was specific; it was induced by UV irradiation and inhibited by the p38-specific inhibitor SB 203580 added to L8 cells (Fig. 1A). The amount of the phosphorylated form of p38 that reflects the active kinase was studied in L8 cells. An antibody that recognizes the active dual phosphorylated form of p38 was used with the same extracts in a Western analysis. Phosphorylated forms of p38 accumulated during the differentiation of cells in DM (Fig. 1B). The level of total p38 protein in differentiating cells was not constant; however, the calculated percentage of phosphorylated p38 in cells growing in DM for 48 h was 2-3-fold higher than that of dividing myoblasts (Fig. 1B). Next, we assayed p38 kinase activity with a second substrate, GST-MEF2C. p38 was recently demonstrated to associate and directly phosphorylate MEF2C (12). In addition, MEF2C is an essential transcription activator of muscle-specific genes (13). Phosphorylation of GST-MEF2C substrate was increased during the differentiation of L8 myoblasts to multinucleated myotubes (Fig. 1C). p38 was probably the kinase that phosphorylated GST-MEF2C because kinase activity was significantly inhibited in cells treated with SB 203580 (Fig. 1C, lanes 3 and 4). The increased activity of p38 during muscle differentiation suggests that it might have a role in this process. Four isoforms of p38 (␣, ␤, ␥, and ␦) are presently known. To study the relative contributions of p38 isoforms in the phosphorylation of MEF2C, each isoform was immunodepleted from muscle extracts. Differential depletion of p38 isoforms affected the phosphorylation of MEF2C (Fig. 1C, lanes 6 -11). p38␣ contributed most of p38 kinase activity (40% inhibition), whereas p38␤ and ␦ were less active (25 and 10% inhibition, respectively). Depletion of p38␥ did not inhibit the phosphorylation of MEF2C (Fig. 1C, lane 10).
Treatment of L8 Cells with SB 203580 Inhibits Their Differentiation-To evaluate the role of p38 in myogenesis, we used the p38 inhibitor SB 203580 (27) that was added to L8 cells at concentrations that blocked most of the p38 activity (Fig. 1A). The inhibitor was added concomitantly with the induction of differentiation (differentiation medium, see "Experimental Procedures"). After 36 h in that medium, cells not treated with the drug fused to form developed myotubes, whereas cells treated with the drug did not form myotubes (Fig. 2). Moreover, the expression of muscle specific genes like myogenin and MLC2 was completely inhibited in cells treated with the drug (Fig. 3A). The expression of MEF2 that participates with the MyoD family in the differentiation process was also abolished in the drug-treated cells (Fig. 3A). Induction of the cyclin/ cyclin-dependent kinase inhibitor p21 (Waf-1) participating in the exit of myoblasts from the cell cycle was inhibited in SB 203580-treated cells. Consequently, the drug blocked differentiation completely. We noticed dead cells in both untreated and treated cells grown in the serum-free medium (DM), which probably underwent programmed cell death under conditions of mitogen deprivation (Fig. 2) (28). However, the drug was most probably not toxic to cells because the percentage of living cells was not changed in drug-treated culture, and these cells recovered and differentiated following removal of the drug (Figs. 2 and 3B).
Activation of p38 with MKK6 Induces Muscle Differentiation by MyoD-The dramatic effect of p38 inhibitor on the expression of MyoD and MEF2 family members prompted us to investigate whether p38 affected also the activity of the myogenic regulator MyoD. The ectopic expression of MyoD converts 10T1/2 fibroblasts to muscle (29). The effect of p38 on MyoD activity was studied by transiently transfecting 10T1/2 fibroblasts with expression vectors of MyoD and wild type MKK6, the direct activator of p38. Most 10T1/2 fibroblasts transfected with MyoD expression vector also expressed the differentiation marker MHC when grown for 24 h in differentiation medium (Fig. 4). However, if the transfected cells were continuously grown in high serum-containing medium (15% bovine calf serum), only 28% of the cells that expressed MyoD expressed also MHC (Fig. 4). Under these conditions, cells that were co-transfected with MKK6 exhibited a significantly higher proportion of MHC staining in the cytoplasm of MyoD-expressing cells (69% of the cells) (Fig. 4). On the other hand, addition of SB 203580 reversed the effect of MKK6 and completely inhibited MyoD, i.e. only about 5% of the cells that expressed MyoD also expressed MHC. In a parallel experiment, we studied the effects FIG. 1. p38 kinase activity is induced during differentiation of L8 cells. A, the in vitro kinase activity of p38 was studied using extracts of L8 muscle cells grown in differentiation medium for different time periods as indicated. Kinase activity of p38 was analyzed by immunoprecipitation of p38. The immunocomplex was used to phosphorylate the substrate, GST-ATF2 protein, as described under "Experimental Procedures." Proteins were separated over SDS-PAGE, and the amount of phosphorylated GST-ATF2 was quantified using the Phos-phorImager and plotted in the histogram. Maximal kinase activity was set to 100 units. Average results from two independent experiments are presented. Bars represent standard errors. B, the same protein extracts from L8 cells as in A were separated using SDS-PAGE, and phosphorylated forms of p38 were identified by Western blotting with an antibody that recognized the dual phosphorylated form of p38 (upper bands). The same blot was exposed also to an antibody that recognized all forms of p38 (lower bands). The control lanes (Cont.): Ϫ, extract of C6 cells; ϩ, extract of anisomycine-treated C6 cells. The asterisk represents a nonspecific band. C, GST-MEF2C protein was phosphorylated in vitro as described under "Experimental Procedures" using extracts of L8 cells grown in differentiation medium for different time periods as indicated. In one case, L8 cells were grown for 48 h in DM in the presence of 20 M SB 203580. Isoforms of p38 were depleted from extracts in the following way: antibodies to the different isoforms of p38 were added to extracts followed by an immunoprecipitation procedure. p38-depleted extracts were used to phosphorylate GST-MEF2C. Control lane, GST protein was used in a kinase assay with extract of L8 cells grown for 48 h in DM. of MKK6 and SB 203580 on the in vitro kinase activity of p38 in extracts from transfected 10T1/2 cells (Fig. 4C). As expected, MKK6 induced whereas SB 203580 inhibited p38 kinase activity. Like MKK6, ectopic expression of p38 isoforms, mainly p38␥ and ␦, in 10T1/2 cells strongly induced the ability of MyoD to activate endogenous MHC (data not shown). Therefore we conclude that activation of p38 MAPK pathway appears to contradict the inhibitory effects of serum on the function of MyoD as judged by the activation of endogenous MHC expression.
Another approach was used to investigate whether p38 affected the activity of MyoD. We used a cell line expressing an estrogen receptor-MyoD chimera protein (21). The chimera protein remains inactive in the cytoplasm of cells. Treatment of cells with estradiol induces the chimera protein that migrates to the nucleus and activates muscle-specific transcription. In this cell line, myogenesis is initiated by the activity of the chimera protein. Therefore, inhibition of MyoD in these cells is expected to abolish myogenesis. Activity of the chimera protein was induced by the addition of estradiol to cells that were grown in the presence or absence of SB 203580. Cells were grown in the presence of estradiol for 24 h, at which time mRNA levels of several muscle-specific genes were induced to significant levels (Fig. 5, lane 3) (30). If cells were treated with SB 203580 during that period, the induction of MEF2, myogenin, and MLC2 expression was significantly inhibited, although expression of chimeric MyoD-ER remained unchanged (Fig. 5, lanes 4 and 5). Therefore, the inhibitor either inactivated the chimera MyoD protein or affected other transcription factors or co-activators that may function in concert with MyoD to initiate myogenesis.
MKK6 Augments Transcription of MCK via the MEF2-binding Site-To further analyze the effect of MKK6 on the transcriptional activity of MyoD, expression vectors of these proteins were co-transfected with a reporter gene that contained a minimal promoter and four MyoD-binding sites (4R-tk-CAT). Surprisingly, the activity of MyoD in the induction of this reporter gene was almost not affected by co-transfection of MKK6 (Fig. 6A, lanes 9 -12). However, MyoD activity was increased more significantly by MKK6 in activating the transcription of a reporter gene whose expression was driven by MCK regulatory sequences (pe ϩ AT-CAT) (Fig. 6A, lanes 1-4). This reporter gene contained two MEF2-binding sites within the MCK enhancer element that could contribute to the effect of MKK6 (12). Co-expression of MEF2C did not further potentiate the transcriptional activity of MyoD on this promoter (not shown). These results may be explained by previous results that indicated that endogenous MEF2 expression was stimulated by the expression of myogenin in fibroblast cells (31). For this reason, we did not co-transfect a MEF2C expression vector in the subsequent experiments that analyzed MCK regulatory sequences. To study the possible role of the MEF2 sites in the stimulation of MyoD activity by MKK6, a similar transfection experiment was done with an MCK reporter gene that contained the MyoD-binding sites and only one MEF2 site (p110-MCK-CAT) (17). Induction of this reporter gene by MyoD was only mildly affected by the co-expression of MKK6 (Fig. 6A,  lanes 5-8). To prove MKK6 activated MCK via the MEF2 site, we generated a reporter gene that contained multiple MEF2 sites of the MCK enhancer and a basal promoter of MCK (MEF2-MCK-CAT). Indeed, this promoter was induced by the expression of MEF2C and was further induced by MKK6 (Fig.  6A, lanes 13-15). Expression of MKK6 alone was sufficient to induce this promoter in a significant fashion (Fig. 6A, lane 16). We conclude that MEF2 sites probably play a role in the stimulation of MCK transcription observed in the presence of MKK6. Addition of the p38 inhibitor, SB 203580, significantly blocked the transcriptional activity of MyoD in the induction of the two MCK enhancer reporter genes, (pe ϩ AT-CAT and p110-MCK-CAT) (Fig. 6B, lanes 1-4), and more modestly affected its activity on the minimal promoter driven by E boxes (4R-tk-CAT) (Fig. 6B, lanes 5 and 6). The transcriptional activity of MEF2 on a basal promoter driven by MEF2 sites was blocked by SB203580 (Fig. 6B, lanes 9 and 10). All in all, these results suggest that p38 modulates the transcriptional activity of the MCK reporter gene via the MEF2-binding sites.
A Transcriptionally Inactive MEF2C Protein Abrogates the Effects of MKK6 on the MCK Enhancer-To study the role of MEF2 in mediating the effects of p38 on the transcription of the MCK reporter gene, we generated a transcriptionally inactive MEF2C protein (see "Experimental Procedures"). This protein did not contain a large part of its transactivation domain that included two phosphorylation sites of p38 (MEF2C-⌬) (12). A similar MEF2A protein that functioned as a dominant negative was recently described by Ornatsky and colleagues (32). By itself, MEF2C-⌬ retained minimal transcriptional activity compared with the wild type protein (Fig. 7A, compare lanes 2 and  6). When co-expressed with wild type MEF2C, the mutant protein was able to repress the transcriptional activity of the pe ϩ AT-CAT carried a regulatory sequence of MCK enhancer that contained two MEF2-binding sites and MCK promoter; p110 MCK CAT carried a smaller MCK enhancer that contained one MEF2-binding site and MCK promoter; 4R-tk-CAT that contained four MyoD-binding sites in the promoter region and MEF2-MCK-CAT that contained three MEF2-binding sites and MCK promoter. All reporter genes contained the CAT reading frame. Protein extracts of the transfected cells were used in a CAT assay according to the transfection efficiency that was measured as described under "Experimental Procedures." The chloramphenicol products were separated by thin layer chromatography and quantified by phosphor imaging (Fuji). For each reporter gene, the maximal CAT activity was set to 100 units. Average results from three independent experiments are presented. Error bars represent standard errors. B, the same constructs described in A were used to transfect 10T1/2 cells with the difference that some plates were treated immediately after transfection with 10 M SB 203580. Also, the expression vector of GAL4-VP16 was co-transfected with the pGAL-CAT reporter gene. For each reporter gene, the maximal CAT activity was set to 100 units. Average results from three independent experiments are presented. Error bars represent standard errors.
former protein (Fig. 7A, lanes 2-5). However, the mutant MEF2C protein did not repress activation by MyoD (lanes 8 and 9) or GAL4-VP16 (lanes 11 and 12) and therefore was specific to wild type MEF2C. To find out whether the mutant protein could block the effect of p38 on the MCK regulatory sequences, we transfected the different expression plasmids with the MCK reporter gene (pe ϩ AT-CAT). The transcriptionally inactive MEF2C protein inhibited the induced expression of MCK reporter mediated by MyoD (Fig. 7B, lanes 1-4). Moreover, it also abolished the additional activity contributed by MKK6 (Fig. 7B, lanes 5-7). The inhibition was specific to the MCK enhancer that contained the MEF2-binding site, because the inactive MEF2C protein did not inhibit MyoD activation of reporter gene containing only MyoD-binding sites (4R-tk-CAT-) (Fig. 7A, lanes 8 and 9). We conclude that MKK6 induced the transcription of MCK via MEF2 protein(s) and that this induction was blocked by competition of the transcriptionally inactive MEF2C protein. Fig. 6 suggest that the p38 pathway operates via MEF2 sites, we could still notice an effect on MyoD (Fig. 6, A, lanes 9 -12, and B,  lanes 5 and 6). This effect on MyoD may result from its association with MEF2, which is known to function as a co-activator of MyoD (18). To study the effect of p38 on MyoD independently of MEF2, we used the GAL4 activator/reporter system. The transcription activators in this system were composed of MyoD fragments that were fused to the DNA-binding domain of yeast transcription factor GAL4. The chimeric activators are expected to bind to the GAL4 DNA-binding sites of a reporter gene and activate transcription via the transactivation domain of MyoD. MEF2 was suggested to interact with the DNAbinding domain of MyoD-E12 heterodimers (18). Two MyoD proteins that were not expected to interact with MEF2 were used; one that did not contain the HLH domain (GAL-⌬HLH MyoD) and another that contained a substituted DNA-binding domain from the Drosophila acheate-scute protein (GAL-T 4 basic MyoD). The transcription mediated by GAL4-MyoD proteins was not affected by co-expression of MKK6 (Fig. 8,  lanes 7-12). However, MKK6 augmented the activity of a chimeric GAL4-MEF2C transcription factor (Fig. 8, lanes 1 and 2). The activity of two MEF2C mutants that did not contain their p38 phosphorylation sites (293, 300A and 387A) was not significantly increased by the expression of MKK6 (Fig. 8, lanes  3-6). Therefore, we conclude that the transcriptional activity of GAL4-MyoD was not affected, whereas the activity of GAL4-MEF2C was affected by MKK6.

The p38 Pathway Affects the MEF2C Protein but Not the MyoD Protein-Although the results presented in
MEF2C Is Phosphorylated by p38 in Muscle Cells-To demonstrate that GAL4-MEF2C protein is phosphorylated in cells by p38, 293 cells were transfected with expression vector of GAL4-MEF2C alone or GAL4-MEF2C with MKK6 (Fig. 9A). In a duplicate set of transfected plates, the cells were treated with SB 203580. Proteins were metabolically labeled with [ 35 S]methionine, and GAL4-MEF2C was immunoprecipitated. The mobility of the GAL4-MEF2C protein in the gel indicated its phosphorylation status. MEF2C was phosphorylated by p38 in cells because it migrated faster in cells treated with SB 203580 (Fig. 9A, compare lanes 1 and 2). Ectopic expression of MKK6 in the transfected cells resulted in a smear of slower migrating GAL-MEF2C species (Fig. 9A, lane 3). Phosphorylation of MEF2C was confirmed by treatment of the immunoprecipitates with alkaline phosphatase that compressed the smear to a tight faster migrating band (Fig. 9A, lanes 5-7). Unlike GAL4-MEF2C, the migration of transfected GAL4 protein was not affected by the co-expression of MKK6 (Fig. 9A, lanes 8 and 9). Therefore, both the expression of MKK6 and treatment with SB 203580 affected the phosphorylation state of transfected GAL4-MEF2C protein in cells.
To find out if MEF2 proteins are phosphorylated in muscle cells, we immunoprecipitated endogenous MEF2 proteins from L8 cells (Fig. 9B). MEF2 proteins were detected mainly in differentiated myotubes (Fig. 9B, lanes 1 and 2). Differentiated cells were phosphate-labeled, and MEF2 proteins were isolated to learn whether these proteins were phosphorylated in myotubes. MEF2 proteins are phosphorylated in muscle cells (Fig.  9B, lane 6). Phosphorylation of MEF2 proteins is partly inhib-ited by treatment of cells with SB 203580 (Fig. 9B, lane 8).
Consequently, MEF2 proteins are phosphorylated by p38 in L8 muscle cells. DISCUSSION In this report we suggest for the first time the possible involvement of p38 in skeletal muscle differentiation. We show that the activity of p38 was induced in L8 myoblasts during the formation of multinucleated myotubes. Interference with p38 activity by the specific inhibitor SB 203580 completely abolished muscle fusion and expression of all myogenic markers that were tested. We studied the effects of p38 MAPK on muscle-specific transcription and found that its activator, MKK6, stimulated the expression of muscle-specific genes, and the effect was mediated by MEF2C. The latter was demonstrated before to be an essential transcription activator in myogenesis (13,33,34).
Members of the MEF2 and MyoD families act within a regulatory network that establishes the differentiated phenotype of skeletal muscle (13). MyoD and MEF2 family members work in concert to activate the expression of many muscle-specific genes including their own. Therefore, inhibition of the activity of either MEF2 or MyoD family members results in complete inhibition of skeletal muscle differentiation. p38 MAPK is a potential activator of MEF2C and as part of the MEF2-MyoD circuit it is expected to control muscle differentiation.
That MEF2 family members played a part in mediating p38 function in the activation of MCK transcription and myogenesis was suggested by the following. 1) Induction of MCK reporter gene by MyoD was augmented by MKK6 only if MEF2 sites were present at the regulatory sequences. MKK6 did not significantly affect the activation of MyoD from promoters that did not contain MEF2 site (Fig. 6).
2) The effect of MKK6 was specifically abolished by the expression of a transcriptionally inactive MEF2C protein (Fig. 7). 3) The transcriptional activity of GAL4-MEF2C fusion proteins was induced, whereas the activity of GAL4-MyoD proteins was not changed by MKK6 (Fig. 8).
MEF2C is not the only member of the family that is expressed in muscle. In fact MEF2A protein accumulates before MEF2C, which is expressed later during the differentiation of myoblasts (35,36). For that reason we should expect that other members of the family may be similarly regulated by p38. Interestingly, MEF2A contains a serine at the position equivalent to Ser 387 found in MEF2C (37), suggesting that it may be regulated similarly. Indeed, recent studies suggest that MEF2A is phosphorylated by p38. 2 We noticed that inhibition of p38 in muscle cells also abolished the expression of MEF2 (Figs. 3A and 5). Inhibition of MEF2 expression may be a result of the repression of MEF2C that functions to induce its own transcription and/or the repression of other transcription factors such as MyoD or Myf5. Our results do not rule out the possibility that MyoD was directly affected by p38, because the reporter gene regulated by MyoD only, 4R-tk-CAT, was mildly affected by MKK6 or SB 203580 (Fig. 6). Other transcription factors activated by p38, like ATF2, CREB, and Elk-1 may participate in the differentiation process induced by p38.
Four isoforms of p38 are known (␣, ␤, ␥, and ␦). We suggest that p38␣ and ␤ are the major isoforms involved in differentiation of L8 cells because 1) The inhibitor, SB 203580, that is specific to the ␣ and ␤ isoforms (38 -40) inhibited the differentiation of L8 cells (Figs. 2 and 3), and 2) immunodepletion of the ␣ and ␤ but not of ␥ and ␦ isoforms from L8 extracts significantly inhibited the phosphorylation of GST-MEF2C (Fig. 1C).  4) or with [ 32 P]orthophosphate (lanes 5-9). MEF2A and MEF2C proteins were immunoprecipitated and separated over SDS-PAGE. In lanes 3 and 7, a competitive peptide against which the MEF2 antibody was made was added to extracts prior to the immunoprecipitation procedure. Lanes 4 and 9 are control lanes; MEF2 antibody was not added to the immunoprecipitation reaction.
Interestingly, our preliminary studies suggest that ␥ and ␦ isoforms are more efficient than ␣ and ␤ isoforms in the induction of muscle differentiation when transfected with MyoD. Farther studies will elucidate the role of p38 isoforms in muscle differentiation.
Recently, we reported that ERK MAPK activity was induced during the differentiation of C2 cells and that this activity promoted muscle differentiation (25). The similarities in the pattern of activities of p38 and ERK in muscle cells imply that these two distinct pathways have similar effects on muscle differentiation. However, our data suggest that the two pathways also exhibit distinct activities. Firstly, inhibition of ERK with PD 098059 only partially prevented the fusion of myoblasts and did not inhibit the expression of muscle-specific markers, whereas inhibition of p38 with SB 203580 prevented fusion of myoblasts and expression of muscle-specific markers. Secondly, the activities of ERK and p38 are differentially induced; p38 is induced earlier than ERK in cells grown in differentiation medium (Ref. 25 and present work). These differences suggest that the two pathways perform distinct functions and that their combined activity may be required for the differentiation process.
Studies of different cell lineages suggest that a distinct balance between the different MAPK pathways is essential for the survival of these cells. During maturation of HL-60 cells to the neutrophil phenotype, the activity of JNK was reduced, whereas the activity of p38 was induced (41). In rat pheochromocytoma PC-12 cells that serve as a model for neuronal differentiation, induced activity of ERK MAPK and reduced activities of JNK and p38 pathways were critical for the survival of these cells (42). The induction of p38 in myocardial cells served to protect them from apoptosis in one study (43) or to increase apoptosis in another study (44). In these cellular models, cell survival or programmed cell death happens as a result of a balance between the activities of ERK, JNK, and p38 MAPKs. Therefore, it is possible that the induction of ERK and p38 MAPKs during skeletal muscle differentiation plays a role in preventing programmed cell death as well as in the activation of muscle-specific genes.