INSULIN-LIKE growth factors (IGFs) are key modulators

Transcription

1 /98/$03.00/0 Vol. 139, No. 4 Endocrinology Printed in U.S.A. Copyright 1998 by The Endocrine Society Mitogen-Activated Protein Kinase Kinase (MEK) Activity Is Required for Inhibition of Skeletal Muscle Differentiation by Insulin-Like Growth Factor 1 or Fibroblast Growth Factor 2* CRYSTAL M. WEYMAN AND ALAN WOLFMAN Department of Cell Biology, Cleveland Clinic Foundation, Cleveland, Ohio ABSTRACT Both insulin-like growth factor 1 (IGF-1) and fibroblast growth factor 2 (FGF-2) are key modulators of skeletal myoblast differentiation. The critical signaling pathways used by either IGF-1 or FGF-2 to inhibit differentiation have not been determined. In this study, we show that both IGF-1 and FGF-2 inhibit the differentiation of 23A2 myoblasts and that both stimulate signaling through mitogen-activated protein kinase (MAPK) kinase (MEK) to MAPK roughly 8-fold in 23A2 myoblasts. We used the selective chemical inhibitor of MEK, PD , to determine if signaling by MEK is required by IGF-1 or FGF-2 to inhibit differentiation. PD did not affect the ability of 23A2 myoblasts to differentiate. Addition of PD to the culture medium 10 min before the addition of IGF-1 or FGF-2 completely blocked the signal from MEK to MAPK and restored the ability INSULIN-LIKE growth factors (IGFs) are key modulators of skeletal muscle differentiation (1). In L6 and C2C12 myoblasts, low concentrations of IGFs enhance their differentiation, whereas high concentrations inhibit their differentiation (2, 3). Skeletal myoblast differentiation is also inhibited by fibroblast growth factor 2 (FGF-2), transforming growth factor- (TGF ), or by the expression of oncogenic Ras (4). Pharmacological inhibitors of the PI 3-kinase pathway block the differentiation of L6 myoblasts (5), and signaling through the PI 3-kinase/p70S6K pathway is required for the IGF-1-induced enhancement of L6 myoblast differentiation (6). The signaling pathway used by IGF-1, or FGF-2, to inhibit skeletal myoblast differentiation has not been determined. Signaling through mitogen-activated protein kinase (MAPK) kinase (MEK) to MAPK plays a key role in many biological processes (6 9). Both these biological processes and the signal from MEK to MAPK are initiated by the binding of certain extracellular growth factors to their cognate transmembrane receptors. Ligand-induced activation of the receptor s intrinsic tyrosine kinase activity creates docking sites for adapter proteins such as SHC or GRB2, thus facilitating the localization of msos (the guanine nucleotide exchange factor for Ras) near Ras (10). Activation of Ras via Received August 26, Address all correspondence and requests for reprints to: Crystal M. Weyman, Cleveland Clinic Foundation, Department of Cell Biology, Cleveland, Ohio weymanc@cesmtp.ccf.org. * This work was supported by NIH Grant GM and AHA Grant (awarded to A. W.). of the 23A2 myoblasts to differentiate in the presence of either IGF-1 or FGF-2. The peak of signaling through MEK to MAPK in response to either IGF-1 or FGF-2 occurred within the first hour with maximal activation observed after 10 min. This signal remained elevated (at roughly 70% above basal) for at least 48 h. PD was added to the culture 60 min after IGF-1 or FGF-2 to test whether this initial peak of signaling was sufficient for the inhibition of differentiation. The restoration of myogenic potential seen when cells were preincubated with PD was essentially identical to that seen when PD was added to cultures after the initial peak of signaling from MEK to MAPK, suggesting that persistent signaling through MEK is required for the inhibition of differentiation by either IGF-1 or FGF-2. (Endocrinology 139: , 1998) msos enhanced exchange of GDP for GTP leads to the activation of the Raf family of serine kinases, which in turn phosphorylate and activate MEK. Activated MEK then phosphorylates and activates MAPK (11). Transmission of the extracellular signal through the cytoplasm via the MAPK phosphorylation cascade culminates with the translocation of activated MAPK to the nucleus and the subsequent phosphorylation and regulation of transcription factors (12), which presumably are ultimately responsible for eliciting the appropriate biological response. Treatment of C2C12 myoblasts with concentrations of IGF-1 reported to enhance differentiation results in no appreciable increase in MAPK activity (13). The activation of MAPK by concentrations of IGF-1 that would inhibit C2C12 myoblast differentiation was not examined. Treatment of C2C12 myoblasts with FGF-2 activates MAPK, but the significance of the FGF-2-induced MAPK activation to the inhibition of differentiation was not determined (13). Treatment of MM14 myoblasts with FGF-2 activates MEK, but transmission of the signal to MAPK is prevented by a dual specificity phosphatase (14, 15). The significance of the FGF-2 induced MEK activation to the FGF-2 induced inhibition of differentiation was not determined. An essential role for signaling by MEK has been demonstrated in many systems through the use of the selective chemical inhibitor PD (8, 16, 17). This inhibitor blocks the activation of MEK and, consequently, the activation of MAPK (17, 18). In this study, we show that either IGF-1 or FGF-2 inhibits the differentiation of 23A2 myoblasts and that both potently stimulate signaling through MEK to MAPK. 1794

2 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT 1795 We also demonstrate the significance of signaling through MEK by using PD to abrogate the inhibition of differentiation by either IGF-1 or FGF-2. Reagents Materials and Methods The anti-map kinase antibody (TR10) was provided by Michael Weber (University of Virginia, Charlottesville, VA), and the PD by Alan Saltiel (Parke-Davis, Ann Arbor, MI). IGF-1 and FGF-2 are from Austral Biologicals (San Ramon, CA). Cells and cell culture Undifferentiated, proliferating 23A2 myoblasts were maintained in growth medium (BME supplemented with 10% FBS and 1% P/S) at low density on gelatin coated dishes. Differentiation was induced by treating the cells with differentiation medium (BME, 1% P/S). PD was dissolved in dimethyl sulfoxide to give a final concentration of 37 mm and stored at 80 C. Dimethyl sulfoxide alone ( 0.15% vol/vol) was added to control cultures. MAP kinase activity assay Cell extracts were prepared in p21 buffer (20 mm MOPS ph 7.4, 4 mm magnesium chloride, 200 mm sucrose, 0.1 mm EDTA) supplemented with 1% CHAPS, protease, and phosphatase inhibitors. TR10 antibody precoupled to protein A sepharose was added to extracts followed by rotation for 1 h at 4 C. After sedimentation, pellets were washed three times with Tris-buffered saline (TBS)/0.1% NP40, once with p21 buffer and then incubated for 20 min at 37 C with 25 m myelin basic protein (MBP), 10 m [ - 32 P]ATP (10 Ci, 4, 500 Ci/mmol), and 10 m phosphatase inhibitors in a total volume of 50 l (19). Assays were performed within the linear range of 32 P incorporation and were terminated by the addition of 2 Laemmli buffer. Reaction products were separated by discontinuous SDS-PAGE (15%). The gel was dried and phosphorylated bands visualized and quantitated by Phosphorimager analysis. MBP sometimes appears as a phosphorylated doublet. This doublet has nothing to do with the treatment but rather the batch of purchased MBP. Please refer to Fig. 2A where MBP is a single band regardless of whether the cells are treated with FGF-2 or IGF-1 and to (see Fig. 5) where MBP is a doublet regardless of whether the cells are treated with FGF-2 or IGF-1. The results presented in Fig. 2B were simply obtained at different times with different batches of MBP. Immunoblot analysis Equal amounts of cell extracts were resolved by SDS-PAGE (8% for MHC detection and 10% for MAPK detection) and electrophoretically transferred to Immobilon P. The Immobilon P was blocked for 1 h at room temperature in TBS/0.1% NP40 with 1% calf serum and 5% dry milk. After washing with TBS/0.1% NP40, the Immobilon P was incubated for 1 h with MF20, a mouse monoclonal antibody specific for skeletal muscle myosin heavy chain (generously provided by E. J. Taparowsky, Purdue University, West Lafayette, IN) or overnight with the phospho-mapk specific rabbit polyclonal antibody (Promega, Madison, WI). After washing 5 with TBS/0.1% NP40 (10 min each), for detection of MHC, the Immobilon P was incubated for 1 h with a rabbit antimouse secondary antibody (1:1000), washed as before, and incubated for another hour with 125 I-labeled goat antirabbit (5 Ci) antibody (Amersham, Arlington Heights, IL). After washing, 125 I was visualized and quantitated by Phosphorimager analysis. For detection of MAPK, the Immobilon P was incubated for 1 h with an antirabbit horseradish peroxidase conjugated secondary antibody, followed by washing and visualization of immunoreactive bands by standard enhanced chemiluminescence techniques. Results Effect of IGF-1 or FGF-2 on 23A2 myoblast differentiation The in vitro differentiation of skeletal myoblasts is a complex process during which proliferating myoblasts exit the cell cycle, begin expressing a battery of muscle-specific genes, elongate, align, and eventually fuse into multinucleated myotubes (20). The IGFs have been reported to enhance the differentiation of L6 and C2C12 myoblasts at concentrations up to 40 ng/ml, whereas higher concentrations inhibit their differentiation (2, 3). FGF-2 inhibits the differentiation of the MM14 (21) and C2C12 (22) myoblast cell lines at roughly 0.2 and 10 ng/ml, respectively. FGF-2 also inhibits differentiation in the 23A2 myoblast cell line (22), but the effect of the IGFs on their differentiation has not been thoroughly investigated. Induction of the muscle-specific myosin heavy chain (MHC) protein and fusion of individual cells into multinucleated myotubes are commonly monitored differentiation markers (8, 22). Undifferentiated 23A2 myoblasts in growth medium (GM: BME/10% FBS) do not express MHC and are not fused (8). Culturing 23A2 myoblasts in differentiation medium (DM: BME/0% FBS) for 48 h induces their differentiation as monitored by the expression of MHC and (Fig. 1) by the ability to elongate and fuse into multinucleated myotubes (see Fig. 4B). To compare the effect of IGF-1 signaling to FGF-2 signaling on the differentiation of 23A2 myoblasts, these cells were switched from GM to DM with increasing concentrations of FGF-2 or IGF-1. Both IGF-1 at 7 ng/ml (1 nm) and FGF-2 at 10 ng/ml (0.6 nm) inhibit the differentiation of 23A2 myoblasts as monitored by the expression of MHC (Fig. 1) or by the ability to elongate and fuse into multinucleated myotubes (data not shown, d.n.s., and Fig. 4B). Signaling through MEK to MAPK by IGF-1 or FGF-2 in 23A2 cells MEK is the immediate upstream activator of MAPK (11). After switching MM14 skeletal myoblast cells to DM for 3 h, FGF-2 (10 ng/ml) induced the activation of MEK, but not MAPK. This was due to the activity of a MAPK phosphatase that declined during commitment to terminal differentiation so that FGF-2 induced the activation of MAPK only after at least 10 h in DM. Signaling by IGF-1 was not examined in MM14 myoblasts (14). In contrast, after switching C2C12 myoblasts to DM for 60 min, FGF-2 (17 ng/ml) did signal through MEK to MAPK, but the significance of this signaling to a differentiation-defective phenotype was not determined. IGF-1, at a concentration sufficient to enhance differentiation (22.5 ng/ml), induced only a slight activation of MAPK (13). A concentration of IGF-1 sufficient to inhibit C2C12 myoblast FIG. 1. Effect of IGF-1 or FGF-2 on MHC expression in 23A2 myoblasts induced by culture in differentiation medium. Equal numbers of 23A2 myoblasts were plated and then switched to differentiation medium (DM) with or without increasing concentrations of growth factor as indicated. After 48 h, cell extracts were prepared, and 50 g of total cell protein was separated by SDS-PAGE. Following electrophoretic transfer of proteins to Immobilon P, a quantitative Western analysis was performed using 125 I as described in Experimental procedures. Shown are results from one experiment that are representative of two independent experiments.

3 1796 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT Endo 1998 Vol 139 No 4 differentiation was not tested. To determine if and under what conditions the signal initiated by either IGF-1 or FGF-2 is transduced from the activation of MEK to the activation of MAPK in 23A2 myoblasts, cells were switched to DM for 3 or 18 h before the addition of growth factor. MAPK immunoprecipitated from 23A2 myoblasts cultured in DM for either 3 or 18 h was roughly 50% less active than MAPK immunoprecipitated from an equal number of cells in GM (Fig. 2A). Treatment of 23A2 myoblasts cultured in DM for either 3 or 18 h with either IGF-1 or FGF-2 for 10 min induced a 7- to 8-fold stimulation of signaling through MEK to MAPK (Fig. 2A). These results were confirmed by Western analysis with an antibody specific for activated phosphomapk (Fig. 2B). Furthermore, the signaling through MEK to MAPK induced by either IGF-1 or FGF-2 is readily detected if IGF-1 or FGF-2 is added to GM or if added simultaneously when cells are switched from GM to DM (d.n.s.). Activated MAPK could be detected after stimulation with as little as 1 ng/ml of either growth factor with maximal activation in response to 5 ng/ml (Fig. 2C). FIG. 2. Signaling through MEK to MAPK in 23A2 myoblasts induced by treatment with IGF-1 or FGF-2. Equal numbers of 23A2 myoblasts were plated and then switched to DM for either 3 or 18 h before stimulation with 10 ng/ml IGF-1 or FGF-2. After 10 min, cell extracts were prepared. MAPK was either (A) specifically immunoprecipitated and assayed for activity, or (B) 50 g of cell extract was separated by SDS-PAGE. Following electrophoretic transfer of proteins to Immobilon P, Western analysis was performed using phosphorylated (phospho-mapk)-specific polyclonal antibodies. Similar results were obtained in two additional experiments. C, Equal numbers of 23A2 myoblasts were plated and then switched to DM for 3 hours before stimulation with the indicated concentration of IGF-1 or FGF-2. After 10 min, cell extracts were prepared, MAPK was specifically immunoprecipitated and assayed for activity as described in Experimental procedures. Similar results were obtained in five independent experiments. PD blocks the IGF-1 or FGF-2-induced signal from MEK to MAPK and restores the myogenic potential to IGF-1 or FGF-2 treated myoblasts PD selectively blocks the activation of MEK, and consequently, the activation of MAPK (17, 18). Cells were incubated for 10 min with increasing concentrations of PD before the addition of IGF-1 or FGF-2 to determine the concentration of PD required for maximum abrogation of the growth factor-induced signal from MEK to MAPK. After 10 min of growth factor stimulation, the MAPK activity immunoprecipitated from cells preincubated with 50 m PD (Fig. 3A) was comparable to the MAPK activity immunoprecipitated from unstimulated control cells. PD must be stable over the 48-h time period used to monitor the induction of MHC to be a useful reagent to assess the significance of the signal from MEK to MAPK to the inhibition of differentiation by IGF-1 or FGF-2. Cells were treated with 50 m PD (or vehicle) for 10 min or 48 h before the addition of either IGF-1 or FGF-2. Transmission of the signal from MEK to MAPK was assessed after 10 min of growth factor stimulation. The PD mediated inhibition of MAPK activation after 48 h in culture medium was essentially identical to that obtained after only 10 min in culture medium (Fig. 3B). These results are consistent with previous reports from our lab (8) and others (16, 18) on the stability and efficacy of PD To use PD to examine the significance of signaling through MEK to the inhibition of 23A2 myoblast differentiation by IGF-1 or FGF-2, PD should not interfere with the normal differentiation of 23A2 myoblasts. Culturing 23A2 myoblasts in DM with or without 50 m PD for 48 h results in similar expression of MHC [(8 and Fig. 4A)] and similar fusion into multinucleated myotubes [(8 and Fig. 4B)]. Including either 10 ng/ml IGF-1 or FGF-2 in the DM inhibits their differentiation as monitored by the expression of MHC (Figs. 1 and 4A) and their ability to elongate and fuse (Fig. 4B). A 10-min preincubation with PD before the addition of IGF-1 or FGF-2 (a treatment sufficient to block the IGF-1 or FGF-2 induced signal from MEK to MAPK) also abrogates the ability of IGF-1 or FGF-2 to effectively inhibit the differentiation of 23A2 myoblasts restoring their ability to elongate and fuse (Fig. 4B) and to express MHC at levels 70 80% of that detected from 23A2 myoblasts cultured identically but without IGF-1 or FGF-2 (Fig. 4A). We did note in roughly 10% of our experiments that PD alone appeared to enhance the expression of MHC by 15 20% when compared with MHC expression from control cultures (Fig.

4 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT 1797 FIG. 3. Effect of PD on the signal from MEK to MAPK induced by IGF-1 or FGF-2. A, Equal numbers of 23A2 myoblasts were plated and then switched to DM with the indicated concentration of PD After 10 min, cells were treated for 10 min with 10 ng/ml of IGF-1 or FGF-2 and the MAPK activity was determined as in Fig. 2. Results are presented as % inhibition of MAPK activity relative to the MAPK activity observed with IGF-1 or FGF-2, respectively, in the absence of PD B, Equal numbers of 23A2 myoblasts were plated and then switched to DM without or with 50 M PD After either 10 min or 48 h, cells were stimulated for 10 min with 10 ng/ml IGF-1 or FGF-2 before lysis, and determination of MAPK activity as described in Fig. 2. Similar results were obtained with FGF-2 in three additional experiments and with IGF-1 in one additional experiment. 4A, top panel). This result was not only irreproducible, but could not possibly explain the levels of MHC obtained (70 80% of that detected in control cultures) when the FGF-2 or IGF-1 induced signals through MEK were abrogated by PD Furthermore, this slight increase in MHC expression detected by Western analysis was not observed when cell cultures were fixed and immunostained for MHC (d.n.s and Ref. 8). Persistent signaling through MEK is required for the inhibition of differentiation by IGF-1 or FGF-2 The presence of FGF-2 in the culture medium is continually required to prevent differentiation (22). In C2C12 myoblasts, FGF-2-induced activation of MAPK was observed at 2 min and returned to nearly basal levels by 10 min. A second but smaller increase in MAPK activity could be seen at 15 min, which returned to basal by 60 min (13). In 23A2 myoblasts, the peak of signaling through MEK to MAPK in response to either IGF-1 or FGF-2 occurred within 60 min with maximal activation observed after 10 min. MAPK activity remained roughly 70% above basal, however, even after 48 h (Fig. 5, A and C), an increase similar to that seen in 23A2 myoblasts transformed by oncogenic Ras (8). Western analysis was performed to demonstrate that MAPK protein levels remained unchanged after 3, 18, or 48 h in DM or after treatment for 48 h with 10 ng/ml of either FGF-2 or IGF-1 (Fig. 5B). To determine if continuous signaling through MEK to MAPK was required to inhibit differentiation, PD was added after the initial burst of MAPK activity. The restoration of myogenic potential seen when cells were preincubated with PD was similar to that seen when PD was added to cultures after the initial peak of signaling from MEK to MAPK (Fig. 6), indicating that persistent signaling through MEK is required for the inhibition of differentiation by either IGF-1 or FGF-2. We observed that although 5 ng/ml FGF-2 was not sufficient to inhibit 23A2 myoblast differentiation (Fig. 1), 5 ng/ml was sufficient for maximal signaling through MEK to MAPK (Fig. 2B). Having demonstrated that persistent signaling through MEK by these growth factors is required to inhibit differentiation, we speculated that it is not the concentration of growth factor required to maximally stimulate signaling (which occurs after 10 min) which is significant but rather that the critical size of the initial dose of FGF-2 is directly related to growth factor stability so that a persistent signal is transduced over the time period necessary for the expression of MHC. To test this hypothesis, media containing FGF-2 at 5 ng/ml and at 10 ng/ml was incubated for 24 or 48 h in the tissue culture incubator before transfer to naive 23A2 myoblasts. After 10 min, activation of the MEK/MAPK pathway was determined. Media containing 10 ng/ml FGF-2 for 24 h before transfer for 10 min to naive 23A2 myoblasts stimulated the signal from MEK to MAPK by 5-fold, whereas media containing 5 ng/ml FGF-2 for 24 h before transfer did not stimulate this signal (Fig. 7). Media containing 10 ng/ml FGF-2 for 48 h before transfer for 10 min to naive 23A2 myoblasts stimulated the signal from MEK to MAPK to MBP by 80% (d.n.s). This stimulation is similar to the persistent activation seen when cells are treated with 10 ng/ml FGF-2 for 24 or 48 h (Fig. 5, A and C) and further substantiates the observation in Fig. 6 that persistent signaling, and not just the initial burst of signal observed within the first hour of treatment (Fig. 5, A and C) is required to prevent differentiation from occurring. Furthermore, the persistent signaling through MEK to MAPK seen when cells are treated with 10 ng/ml FGF-2 for 48 h is not found when cells are treated with 5 ng/ml for 48 h (d.n.s).

5 1798 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT Endo 1998 Vol 139 No 4 FIG. 4. Effect of PD pretreatment on the inhibition of MHC expression by IGF-1 or FGF-2. Equal numbers of 23A2 myoblasts were plated and then switched to DM with or without 50 M PD as indicated. After 10 min, 10 ng/ml IGF-1 or FGF-2 was added to the culture media as indicated. A, After 48 h, cell extracts were prepared and a Western analysis for MHC was performed as described in Fig. 1. Similar results were obtained with FGF-2 in three independent experiments and with IGF-1 in two independent experiments. B, Phase contrast microscopy of parallel cultures treated as described in (A) after 48 h. Discussion IGF-1 is a key modulator of skeletal myoblast differentiation. Autocrine stimulation by endogenously secreted IGFs is necessary for differentiation in the absence of exogenously added IGFs. Consequently, exogenous addition of critical concentrations of IGF-1 enhances the differentiation of cell lines that secrete suboptimal concentrations of endogenous IGFs. Addition of concentrations of IGF-1 in excess of the level required for differentiation are sufficient to inhibit differentiation. The differing sensitivity of distinct myoblast cell lines to the effects of exogenously added IGFs is thought to be a consequence of the level of secreted endogenous IGFs (2, 3). Pharmacological inhibitors of the PI 3-kinase pathway block the differentiation of L6 myoblasts (5) and signaling through the PI 3-kinase/p70S6K pathway is required for the IGF-1-induced enhancement of L6 myoblast differentiation (6). The signaling pathways used by IGF-1 at concentrations that inhibit skeletal myoblast differentiation have not been determined. The signaling pathways used by FGF-2 to inhibit skeletal myoblast differentiation have also not been determined. In this study, we have shown that either IGF-1 or FGF-2 stimulates signaling through MEK to MAPK. Furthermore, we have used the selective chemical inhibitor of MEK, PD098059, (17, 18) to demonstrate that continual signaling through MEK is required for both the IGF-1 and the FGF-2 induced differentiation-defective phenotype in 23A2 myoblasts. Our finding that continual signaling through MEK is required by FGF-2 to inhibit skeletal myoblast differentiation is consistent with the requirement for the continual presence of FGF-2 in the differentiation medium (22). It is important to make a distinction between a requirement for MEK activation and a requirement for MAPK activation. Although MEK is the immediate upstream activator of MAPK (11), activation of MEK does not always result in MAPK activity. In the MM14 skeletal myoblast cell line, FGF-2 is a potent activator of MEK1, but this signal only leads to the activation of MAPK after the cells have been in differentiation medium for at least 10 h, a time point at which FGF-2 can no longer inhibit their differentiation (14). Although we have found that either FGF-2 or IGF-1 is capable

6 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT 1799 FIG. 5. Time course of signaling through MEK to MAPK stimulated by IGF-1 or FGF-2. A, Equal numbers of 23A2 myoblasts were plated and switched to DM for 3 h before stimulation with 10 ng/ml IGF-1 or FGF-2. After the indicated times, cells were lysed and MAPK activity determined as described in Fig. 2. Similar results were obtained in three independent experiments. B, Equal numbers of 23A2 myoblasts were plated and switched to DM. After the indicated times, cell extracts were prepared and 50 g of cell extract was separated by SDS-PAGE. Following electrophoretic transfer of proteins to Immobilon P, Western analysis was performed using MAPK-specific polyclonal antibodies. (C) Quantitation of phosphorimager analysis. Shown are averages from two independent experiments. FIG. 7. FGF-2 induced signaling in naive cells after 24 h in the tissue culture incubator. Equal numbers of 23A2 myoblasts were switched to DM for 3 h and then treated for 10 min before lysis and the determination of MAPK activity with: lane 1: no treatment, lane 2: 10 ng/ml FGF-2, lane 3: media containing 10 ng/ml FGF-2 for 24 h before transfer, lane 4: 5 ng/ml FGF-2, lane 5: media containing 5 ng/ml FGF-2 for 24 h before transfer. FIG. 6. Continuous signaling through MEK is required for the inhibition of MHC expression by IGF-1 or FGF-2. Equal numbers of 23A2 myoblasts were plated and then switched to DM with or without 50 M PD for 10 min before addition of 10 ng/ml IGF-1 or FGF-2 as indicated and described in Fig. 4. Alternatively, cells were switched to DM without PD but with 10 ng/ml IGF-1 or FGF-2 for 60 min, as indicated, before the addition of PD After 48 h, cell extracts were prepared, and a Western analysis for MHC was performed as described in Fig. 1. Similar results were obtained in two independent experiments. of signaling through MEK to MAPK 1) if added to growth medium, 2) if added simultaneously when cells are switched from growth medium to differentiation medium, or 3) if added after switching the cells to differentiation medium for either 3 or 18 h, a reagent that would selectively eliminate only MAPK signaling would be required to attribute any significance to this MAPK activation. It is also important to make a distinction between signals that are necessary and signals that are sufficient for the inhibition of skeletal myoblast differentiation. Although we have clearly demonstrated that signaling through MEK is required by FGF-2 and IGF-1 to inhibit differentiation, others have shown that signaling by constitutively active MEK is not sufficient to inhibit MyoDinduced differentiation in 10T1/2 fibroblasts (23). We report here a distinction between the signaling pathways used by oncogenic Ras and both IGF-1 and FGF-2 in that we have previously used PD to show that the differentiation-defective phenotype established by oncogenic Ras in 23A2 myoblasts is MEK independent (8). Although it might seem likely that IGF-1 and FGF-2 signal through endogenous Ras, this has not been formally proven. It is possible, therefore, that the IGF-1 and FGF-2-induced activation of MEK, and subsequent activation of MAPK, is Ras independent. If IGF-1 and FGF-2 do signal through endogenous Ras, it remains formally possible that endogenous Ras and oncogenic Ras signal through distinct, as well as common, effectors. Although signaling through MEK was not required by oncogenic Ras to inhibit 23A2 myoblast

7 1800 IGF-1 OR FGF-2 DIFFERENTIATION BLOCK IS MEK DEPENDENT Endo 1998 Vol 139 No 4 differentiation, we used PD to demonstrate that oncogenic-ras relies on signaling through MEK to permit serum-independent proliferation (8). PD also blocked the mitogenic effect of IGF-1 in L6 myoblasts (6) and the mitogenic effect of FGF-2 or IGF-1 in C2C12 myoblasts (13). FGF-2 and, to a lesser extent, IGF-1 induce serum-independent thymidine incorporation in 23A2 myoblasts, which is abrogated by the inclusion of PD (d.n.s.). This does not contribute to the restoration of the myogenic phenotype, however, because it is well documented that the ability of FGF-2, IGF-1 or oncogenic Ras to inhibit differentiation is independent of their ability to induce proliferation (1, 24, 25). Furthermore, we have previously shown that, although PD effectively abrogates the oncogenic Ras-induced serum independent proliferation of 23A2 myoblasts, PD is not able to overcome the oncogenic Ras-induced differentiation block (8). Skeletal myoblast differentiation is positively regulated by activation of the muscle regulatory factor (MRF) family of basic helix-loop helix transcription factors. To positively regulate the expression of genes necessary for differentiation, the MRFs (MyoD, myogenin, Myf-5, and MRF4) heterodimerize with ubiquitously expressed E-proteins and bind to target E-box sequences located within specific promoters or enhancers. The signaling pathways used by inhibitors of differentiation ultimately impinge on the ability of the MRFs to induce transcription (26). Oncogenic Ras inhibits the ability of the muscle regulatory transcription factors, MyoD or MRF4, to induce skeletal myoblast differentiation without altering their dimerization, DNA binding, or inherent transcriptional activities (27). FGF-2 also inhibits the ability of these MRFs to induce differentiation without altering their dimerization or DNA binding. FGF-2 does, however, inhibit their inherent transcriptional activity (27). Thus, distinctions between the signaling pathways used by oncogenic Ras and FGF-2 that we have shown here are evident in the cytoplasm also exist at the level of MRF regulation. Unfortunately, the effect of IGF-1 signaling on MRF regulation has not been explored. The challenge now is to understand how the signal initiated by IGF-1 and FGF-2 and transmitted through MEK impinges on the transcription factors responsible for skeletal muscle specific gene expression. Acknowledgments The authors thank Jane Yoder-Hill and Mark Hamilton and Janice C. Wolfman for helpful comments and discussions. References 1. Florini JR, Ewton DZ, Magri KA 1991 Hormones, growth factors, and myogenic differentiation. Annu Rev Physiol 53: Florini JR, Magri K, Ewton D, James P, Grindstaff K, Rotwein P 1991 Spontaneous differentiation of skeletal myoblasts is dependent upon autocrine secretion of insulin-like growth factor. J Biol Chem 266: Florini JR, Ewton DZ, Falen SL, Van Wyk JJ 1986 Biphasic concentration dependency of stimulation of myoblast differentiation by somatomedins. Am J Physiol 19:C771 C Florini JR, Magri KA 1989 Effects of growth factors on myogenic differentiation. Am J Physiol 256:C701 C Kaliman P, Vinals F, Testar X, Palacin M, Zorzano A 1996 Phosphatidylinositol 3-kinase inhibitors block the differentiation of skeletal muscle cells. J Biol Chem 271: Coolican S, Samuel D, Ewton D, McWade F, Florini JR 1997 The mitogenic and myogenic actions of insulin-like growth factors utilize distinct signaling pathways. J Biol Chem 272: Pagès G, Lenormand P, L Allemain G, Chambard J-C, Meloche S, Pouysségur J 1993 Mitogen-activated protein kinases p42 mapk and p44 mapk are required for fibroblast proliferation. Proc Natl Acad Sci USA 90: Weyman CM, Ramocki MB, Taparowsky EJ, Wolfman A 1997 Distinct signaling pathways regulate transformation and inhibition of differentiation by oncogenic Ras. Oncogene 14: Cowley S, Paterson H, Kemp P, Marshall CJ 1994 Activation of MAP kinase kinase is necessary and sufficient for PC12 differentiation and for transformation of NIH 3T3 cells. Cell 77: Feig LA 1993 The many roads that lead to ras. Science 260: Moodie SA, Wolfman A 1994 The 3Rs of life: Ras, Raf and growth regulation. Trends Genet 10: Lenormand P, Sardet C, Pagès G, L Allemain G, Brunet A, Pouysségur J 1993 Growth factors induce nuclear translocation of MAP kinases (p42 mapk and p44 mapk ) but not of their activator MAP kinase kinase (p45 mapkk ) in fibroblasts. J Cell Biol 122: Milasincic DJ, Calera MR, Farmer SR, Pilch PF 1996 Stimulation of C2C12 myoblast growth by basic fibroblast growth factor and insulin-like growth factor 1 can occur via mitogen-activated protein kinase-dependent and -independent pathways. Mol Cell Biol 16: Campbell JS, Wenderoth MP, Hauschka SD, Krebs EG 1995 Differential activation of mitogen-activated protein kinase in response to basic fibroblast growth factor in skeletal muscle cells. Proc Natl Acad Sci USA 92: Mourey R, Vega Q, Campbell J, Wenderoth M, Hauschka S, Krebs EG, Dixon JE 1996 A novel cytoplasmic dual specificity protein tyrosine phosphatase implicated in muscle and neuronal differentiation. J Biol Chem 271: Pang L, Sawada T, Decker SJ, Saltiel AR 1995 Inhibition of MAP kinase kinase blocks the differentiation of PC-12 cells induced by nerve growth factor. J Biol Chem 270: Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR 1995 PD is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem 270: Dudley DT, Pang L, Decker SJ, Bridges AJ, Saltiel AR 1995 A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc Natl Acad Sci USA 92: Moodie SA, Willumsen BM, Weber MJ, Wolfman A 1993 Complexes of Ras GTP with Raf-1 and mitogen-activated protein kinase kinase. Science 260: Andres V, Walsh K 1996 Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J Cell Biol 132: Clegg C, Linkhart T, Olwin B, Hauschka S 1987 Growth factor control of skeletal muscle differentiation: commitment to terminal differentiation occurs in G 1 phase and is repressed by fibroblast growth factor. J Cell Biol 105: Vaidya TB, Rhodes SJ, Taparowsky EJ, Konieczny SF 1989 Fibroblast growth factor and transforming growth factor repress transcription of the myogenic regulatory gene MyoD1. Mol Cell Biol 9: Ramocki MB, Johnson SE, White MA, Ashendel CL, Konieczny SF, Taparowsky EJ 1997 Signaling through mitogen-activated protein kinase does not duplicate the effects of activated Ras on skeletal myogenesis. Mol Cell Biol 17: Rosenthal SM, Cheng ZQ 1995 Opposing early and late effects of insulin-like growth factor I on differentiation and the cell cycle regulatory retinoblastoma protein in skeletal myoblasts. Proc Natl Acad Sci USA 92: Spizz G, Roman D, Strauss A, Olson EN 1986 Serum and fibroblast growth factor inhibit myogenic differentiation through a mechanism dependent on protein synthesis and independent of cell proliferation. J Biol Chem 261: Ludolph DC, Konieczny SF 1995 Transcription factor families: Muscling in on the myogenic program. FASEB J 9: Kong YF, Johnson SE, Taparowsky EJ, Konieczny SF 1995 Ras p21 Val inhibits myogenesis without altering the DNA binding or transcriptional activities of the myogenic basic helix-loop-helix factors. Mol Cell Biol 15: