INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

The Ras family of GTPases are important regulators of intracellular signal transduction pathways that control cellular proliferation, transformation, and differentiation. Multiple Ras effectors, including Raf/MEK/MAPK,¹ Rac/Rho and RalGDS, contribute to the ability of oncogenic Ras to induce cellular transformation (1-7), and it is becoming increasingly apparent that the ability of oncogenic Ras to inhibit skeletal muscle differentiation also depends on signaling through a number of Ras effectors (8, 9). Interestingly, the set of Ras effectors that impact differentiation events appears to be distinct from the set which induces cellular transformation since Ras proteins defective for transformation efficiently block the morphological and biochemical differentiation of myoblasts in culture (8, 9).

In previous studies, we have demonstrated that two transformation-defective, effector-specific Ras variants inhibit MyoD-induced skeletal myogenesis (9). H-Ras G12V,T35S binds to Raf-1 and activates MAPK, whereas H-Ras G12V,Y40C does not (10). The ability of these Ras molecules to inhibit skeletal muscle differentiation does not rely on their individual abilities to activate either MEK/MAPK or Rac/Rho activity (9). Furthermore, these Ras effector-loop variants fail to function synergistically to inhibit myogenesis. Taken together, these data suggest that a potentially novel, Ras-activated signaling pathway is important for the inhibition of differentiation in this model system.

In this current study, we demonstrate that a third Ras effector-loop variant, H-Ras G12V,E37G, which specifically interacts with the Ral guanine nucleotide dissociation stimulator (RalGDS) (6, 11), also inhibits the MyoD-induced differentiation of C3H10T1/2 fibroblasts. Using a reporter gene whose muscle-specific expression relies on both SRF and the muscle regulatory factors, we show that H-Ras G12V,E37G, RalGDS, and the membrane-localized RalGDS CAAX protein inhibit muscle-specific gene expression, whereas a RalGDS protein modified by deletion of the Ras binding domain (RalGDS-rbd) does not. Additionally, coexpression of H-Ras G12V,E37G and RalGDS or RalGDS CAAX enhances this effect, suggesting that H-Ras G12V,E37G and RalGDS function through distinct pathways. In examining the potential role of SRF in this observed inhibition, we demonstrate that H-Ras G12V,E37G, but not RalGDS, RalGDS CAAX, or RalA G23V, activates a reporter gene regulated by a single SRF site. Furthermore, RalGDS, RalGDS CAAX, and RalA G23V inhibit H-Ras G12V,E37G-stimulated transactivation of the SRF reporter gene, and RalGDS CAAX actually functions to inhibit the basal level of SRF activity observed in these cells. The opposing actions of the RalGDS pathway and H-Ras G12V,E37G toward SRF in C3H10T1/2 cells suggest that the observed impact of RalGDS on the expression of certain muscle-specific genes is linked to its negative impact on SRF. In support of this, we show that H-Ras G12V,E37G blocks activation of troponin I-Luc, an SRF-independent, muscle-specific reporter gene, whereas RalGDS and RalGDS CAAX do not. Taken together, these studies suggest a minor role for RalGDS in the Ras-mediated inhibition of myogenesis and provide strong evidence that a novel Ras-activated molecule(s) is the major effector operating to repress biochemical and morphological differentiation in this model system.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Expression Constructs-- -actin-Luc, a luciferase reporter gene controlled by the muscle-specific human cardiac -actin promoter, and pSRF2-Luc, a luciferase reporter gene regulated by a single SRF binding site from the c-fos promoter, have been described previously (12, 13). Troponin I-Luc is a luciferase reporter gene controlled by the muscle-specific quail troponin I enhancer and has been described previously (9). pG5T-Luc (Gal4₅-Luc), a luciferase reporter gene containing five, tandem Gal4 DNA binding sites was obtained from Dr. J. D. Gralla, UCLA. The expression plasmid pEMc11S (pEM-MyoD) (14) contains the mouse MyoD cDNA while Gal4-Elk1 (15) encodes a fusion protein consisting of the DNA binding domain of the yeast Gal4 protein and the transcription activation domain of Elk1. Human H-ras cDNAs altered by site-directed mutagenesis were constructed as described (10) and are expressed as hemagglutinin-tagged (HA) fusion proteins using the CMV expression cassette in pDCR. The mouse RalGDS cDNA and its derivative, RalGDS CAAX, as well as RalA G23V, are in the mammalian expression vector pCEP4 and have been described previously (6). pCEP4 RalGDS-rbd is truncated at the PstI site of mouse RalGDS, removing the carboxyl-terminal 130 amino acid residues. The pZIP vector, pZIP Rac1 17N, which encodes dominant negative Rac1, and pZIP H-Ras 61L, which encodes the constitutively active human H-Ras 61L protein, have been described previously (1, 9).

Cell Culture and Transfections-- C3H10T1/2 mouse fibroblasts were maintained in growth medium consisting of basal modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum (BioWhittaker), penicillin (100 units/ml), and streptomycin (100 µg/ml). To induce myogenesis, transfected cultures were exposed to differentiation medium (DM) composed of Dulbecco's low glucose modified Eagle's medium supplemented with 2% horse serum, penicillin (100 units/ml), and streptomycin (100 µg/ml) for 48 h. Transient transfections were carried out using calcium phosphate/DNA precipitation as described previously (9) with the amounts of plasmid DNA indicated in each of the Fig. legends. Cells were exposed to DNA precipitates for 4 h, after which the cultures were treated with 20% glycerol in basal modified Eagle's medium for 2 min, maintained in growth medium for 18 h and then treated with DM for 48 h. For gene expression studies, cells were lysed in 25 mM Tris, pH 7.8, 4 mM EDTA, 1% Triton X-100, 10% glycerol. Following normalization to total protein content (Bio-Rad Protein Assay), cell extracts were analyzed using the Luciferase Assay Kit (Promega). Values are expressed as RLU (relative light units/µg of protein). A minimum of three independent transfections were performed for each experimental group. Parallel cultures were processed for immunocytochemistry and for Western blot analysis as described below.

Immunocytochemistry-- Transfected C3H10T1/2 cells in DM were fixed with 70% ethanol:formalin:acetic acid (20:2:1) for 60 s, permeabilized with 0.1% Nonidet P-40 in 10 mM Tris, pH 8.0, 150 mM NaCl for 10 min, and blocked using 2% horse serum in PBS for 30 min. Washes between treatments were performed with PBS. The cells were incubated with MF-20, a monoclonal antibody specific for the myosin heavy chain protein (16), and reactive complexes were visualized using a secondary antibody and Vectastain Kit reagents (Vector Laboratories). The number of myosin positive cells was averaged following the examination of ten, randomly chosen microscope fields. In all instances, the efficiency of myofiber formation is expressed as a percent of the efficiency observed in control cultures transfected with pEM-MyoD alone.

Western Blot Analysis-- Transfected C3H10T1/2 cells were harvested in 4× SDS-polyacrylamide gel electrophoresis sample buffer (200 mM Tris, pH 6.8, 400 mM dithiothreitol, 8% SDS, 0.4% bromphenol blue, and 40% glycerol) and normalized for protein content by Coomassie Blue staining. Equal amounts of protein, along with low molecular weight standards (Bio-Rad), were separated by 12% SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose filter paper as described (17). Nonspecific binding sites were blocked by incubation with 5% non-fat dry milk in TBST (10 mM Tris, pH 8.0, 150 mM NaCl, 0.1% Tween 20), and MyoD was detected by incubation in 5% non-fat dry milk in TBST containing a 1:300 dilution of anti-MyoD antibody (C-20, Santa Cruz Biotechnology). Following several washes in TBST, reactive complexes were visualized using a peroxidase-conjugated secondary antibody and enhanced chemiluminescence (ECL kit, Amersham Pharmacia Biotech).

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

H-Ras G12V,E37G Inhibits Myogenesis-- To determine if expression of the transformation-defective Ras effector-loop variant, H-Ras G12V,E37G, inhibits skeletal myogenesis, we tested its ability to block MyoD-induced expression of skeletal myosin and activation of a muscle-specific reporter gene, -actin-Luc. C3H10T1/2 fibroblasts were transiently transfected with an expression vector for mouse MyoD (pEM-MyoD) and either pDCR vector DNA or pDCR H-Ras G12V,E37G. Twenty-two hours following transfection, the cultures were treated with DM and incubated for 48 h prior to myosin heavy chain immunostaining. As shown in Fig. 1A, H-Ras G12V,E37G inhibits morphological differentiation by greater than 70%. C3H10T1/2 cells transfected with -actin-Luc, pEM-MyoD and either pDCR vector DNA, pDCR H-Ras G12V or pDCR H-Ras G12V,E37G were cultured as described above, and cell lysates were assayed for reporter gene expression. As shown in Fig. 1B, H-Ras G12V and H-Ras G12V,E37G also inhibit -actin-Luc activity by 75 and 68%, respectively. The repression of myogenesis in these assays is not due to a decrease in MyoD protein since Western blot analysis revealed that equivalent levels of MyoD protein accumulate in the control and in the experimental groups (Fig. 1C). Since H-Ras G12V,E37G interacts specifically with RalGDS (6, 11), these data suggest that RalGDS has a role in the repression of myogenesis by oncogenic Ras.

View larger version (57K): [in this window] [in a new window]   Fig. 1.   H-Ras G12V,E37G inhibits myogenesis. A, C3H10T1/2 cells were transfected with 5.0 µg of pEM-MyoD and 10.0 µg of pDCR (control) or pDCR H-Ras G12V,E37G DNAs. Following treatment with DM for 48 h, the cells were fixed and immunostained for MHC expression. Representative microscope fields from the indicated experimental groups, photographed under bright light conditions, are shown. Ten random fields from each group, from three independent transfections, were scored for myofiber formation, and the average number of myofibers per field is indicated on the left of each panel as a percentage of the average value obtained for the control group. B, C3H10T1/2 cells were transfected with 2.0 µg of -actin-Luc, 0.5 µg of pEM-MyoD, and 1.0 µg of pDCR vector (control), pDCR H-Ras G12V, or pDCR H-Ras G12V,E37G DNAs. Following 48 h in DM, cell extracts were prepared and assayed for luciferase activity. Luciferase activity is expressed relative to the control group, for which the value is set at 100. Each value represents the average from a minimum of three independent transfections. Error bars indicate the standard errors of the means. C, C3H10T1/2 fibroblasts were transfected as described for panel A. Following treatment with DM for 48 h, cell extracts were prepared, and the expression of MyoD in each group was analyzed by Western blot hybridization as described under "Experimental Procedures."

RalGDS and RalGDS CAAX, but Not RalGDS-rbd, Inhibit -actin-Luc Reporter Gene Activity-- To determine if RalGDS and the membrane-localized variant RalGDS CAAX duplicate the effects of H-Ras G12V,E37G on muscle-specific reporter gene activity, C3H10T1/2 fibroblasts were transfected with -actin-Luc, pEM-MyoD, and pCEP4 expression vectors for RalGDS or RalGDS CAAX. The requirement for Ras interaction in the RalGDS effect was tested using pCEP4 RalGDS-rbd, which expresses a RalGDS protein deleted for the Ras binding domain. Twenty-two hours following transfection, the cells were cultured in DM, and 48 h later, cell lysates were prepared. RalGDS and RalGDS CAAX inhibit -actin-Luc activity by 52 and 48%, respectively, whereas RalGDS-rbd has no significant effect on reporter gene expression (Fig. 2A). The observed repression of reporter gene activity by RalGDS and RalGDS CAAX is not due to low levels of MyoD protein accumulation since Western blot analysis indicates that high levels of MyoD protein are found in all groups (Fig. 2B). Likewise, the inability of RalGDS-rbd to block myogenesis is not a function of protein expression levels, since all three RalGDS proteins are expressed equivalently in these cells (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   RalGDS and RalGDS CAAX inhibit -actin-Luc activity. A, C3H10T1/2 cells were transfected with 2.0 µg of -actin-Luc, 0.5 µg of pEM-MyoD, and 1.0 µg of pCEP4 vector (control), pCEP4 RalGDS, pCEP4 RalGDS CAAX, or pCEP4 RalGDS-rbd. Following 48 h in DM, cell extracts were prepared and analyzed for luciferase activity. Luciferase activity is expressed relative to the control group, which is set at 100 RLU (relative light units/µg of protein). Each value represents the average from at least three independent transfections. Error bars indicate the standard errors of the means. B, C3H10T1/2 cells were transfected with 5.0 µg of pEM-MyoD and 10.0 µg of pCEP4 vector, pCEP4 RalGDS, pCEP4 RalGDS CAAX, or pCEP4 RalGDS-rbd. Following treatment with DM for 48 h, cell extracts were prepared, and the expression of MyoD in each group was analyzed by Western blot hybridization as described under "Experimental Procedures."

RalGDS and RalGDS CAAX Enhance the Ability of H-Ras G12V,E37G to Inhibit -actin-Luc Expression-- If RalGDS mediates the inhibition of myogenesis by H-Ras G12V,E37G, it follows that coexpression of RalGDS or RalGDS CAAX with H-Ras G12V,E37G should result in an increased repression of reporter gene activity. To test this prediction, C3H10T1/2 fibroblasts were transfected with -actin-Luc, pEM-MyoD, pDCR H-Ras G12V,E37G, and either pCEP4 RalGDS or pCEP4 RalGDS CAAX. Interestingly, coexpression of RalGDS or RalGDS CAAX with H-Ras G12V,E37G generates even lower levels of -actin-Luc activity than either of the proteins alone. However, no true synergy is observed in these assays (Fig. 3). The increase in the inhibition of -actin-Luc expression by RalGDS and RalGDS CAAX in the presence of H-Ras G12V,E37G, which we have interpreted to be additive, suggests that RalGDS functions as a mediator of the Ras effect. Alternatively, RalGDS and RalGDS CAAX may function within the context of a Ras signaling network to relay a negative signal of their own. Previously, we demonstrated that the constitutive activation of the Rac/Rho signaling pathway does not duplicate the effects of activated H-Ras expression in this system (9). To investigate if the inhibition of myogenesis by H-Ras G12V,E37G relies on Rac1 activity, C3H10T1/2 cells were transfected with pEM-MyoD, -actin-Luc, pDCR H-Ras G12V,E37G, and pZIP Rac1 17N, which expresses a dominant negative form of Rac1. Results from these experiments show that expression of Rac1 17N in the presence of H-Ras G12V,E37G generates a level of reporter gene activity equivalent to that produced by H-Ras G12V,E37G alone (Fig. 3). We have confirmed by Western analysis that the Rac1 17N protein is expressed in these cells (data not shown), and therefore, we conclude that H-Ras G12V,E37G inhibits -actin-Luc activity independently of Rac1.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   RalGDS and RalGDS CAAX enhance the ability of H-Ras G12V,E37G to inhibit -actin-Luc expression. C3H10T1/2 cells were transfected with 2.0 µg of -actin-Luc, 0.5 µg of pEM-MyoD, 1.0 µg of pDCR vector or pDCR H-Ras G12V,E37G, and 2.0 µg of pCEP4 vector, pCEP4 RalGDS, pCEP4 RalGDS CAAX, or pZIP Rac1 17N (DN Rac1). Following 48 h in DM, cell extracts were prepared and analyzed for luciferase activity. Luciferase activity is expressed relative to the control which is set at 100 RLU (relative light units/µg of protein). Each value represents the average from at least three independent transfections. Error bars indicate the standard errors of the means.

H-Ras G12V,E37G Does Not Enhance MAPK-dependent Transactivation and, Unlike RalGDS and RalGDS CAAX, Does Activate SRF -- To confirm that H-Ras G12V,E37G functions as expected in C3H10T1/2 fibroblasts, we examined its influence on the transcriptional activation of two distinct reporter genes: pG5T-Luc, which is activated in a MAPK-dependent manner by Gal4-Elk1, and pSRF-Luc, which is expressed in response to SRF binding. C3H10T1/2 cells were transfected with pG5T-Luc, Gal4-Elk1, and pDCR H-Ras G12V,E37G, or with pSRF-Luc and pDCR H-Ras G12V,E37G. As expected, H-Ras G12V,E37G does not activate endogenous MAPK activity (2) and therefore does not significantly enhance Gal4-Elk1-mediated transactivation in these cells (Fig. 4A). Control groups transfected with oncogenic H-Ras alone (H-Ras 61L (Fig. 4A) or H-Ras G12V (9)) reveal that the cells are capable of producing activated MAPK as measured by this assay. H-Ras G12V,E37G does, however, activate SRF-mediated transactivation by 7.1-fold (Fig. 4B). Interestingly, the -actin-Luc reporter gene contains an SRF site that is essential for its activity (12) and RalGDS and RalGDS CAAX have been reported to act upstream of SRF (19). To determine if RalGDS and RalGDS CAAX negatively influence -actin-Luc reporter gene expression by blocking the activity of SRF, C3H10T1/2 cells were transiently transfected with pSRF-Luc and either RalGDS, RalGDS CAAX, RalA G23V (a putative downstream effector of RalGDS), or H-Ras G12V,E37G plus RalGDS, RalGDS CAAX, or RalA G23V. As shown in Fig. 4B, the level of pSRF-Luc activity measured in the presence of RalGDS CAAX (a constitutively active variant of RalGDS) is reduced significantly compared with control values. Similarly, the coexpression of RalGDS, RalGDS CAAX, or RalA G23V and H-Ras G12V,E37G in an equimolar ratio stimulates pSRF-Luc expression by only 4.0-, 2.4-, and 3.6-fold, respectively. These data suggest that, in C3H10T1/2 cells, the RalGDS signaling pathway represses -actin-Luc activity by inhibiting SRF and thus provides a possible explanation for the observed additive inhibition of -actin-Luc expression by H-Ras G12V,E37G and RalGDS or RalGDS CAAX.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   H-Ras G12V,E37G does not enhance MAPK-dependent transactivation and, unlike RalGDS, RalGDS CAAX, and RalA G23V, does activate SRF. A, C3H10T1/2 cells were transfected with 1.0 µg of pG5T-Luc, 0.5 µg of Gal4-Elk1, 0.25 µg of pEM-MyoD, and 0.5 µg of pDCR vector (control), pDCR H-Ras G12V,E37G, or pZIP H-Ras 61L. Following 48 h in DM, cell extracts were prepared and analyzed for luciferase activity. Luciferase activity is expressed relative to the control group, which is set at 1.0 RLU (relative light unit/µg of protein). Each value represents the average from at least three independent transfections. Error bars indicate the standard errors of the means. B, C3H10T1/2 fibroblasts were transfected with 2.0 µg of pSRF-Luc, 2.0 µg of control pDCR vector DNA, and 2.0 µg of pDCR H-Ras G12V,E37G, pCEP4 RalGDS, pCEP4 RalGDS CAAX, or pCEP4 RalA G23V, or 2.0 µg each of pDCR H-Ras G12V,E37G and pCEP4 Ral GDS, pCEP4 RalGDS CAAX, or pCEP4 RalA G23V. Luciferase activity was measured, and values were expressed as described in panel A.

H-Ras G12V,E37G, but Not RalGDS or RalGDS CAAX, Inhibits an SRF-independent, Muscle-specific Reporter Gene-- To further explore the observation that the RalGDS signaling pathway does not mimic the effects of H-Ras G12V,E37G in C3H10T1/2 cells, we decided to test the effects of H-Ras G12V,E37G, RalGDS, and RalGDS CAAX on a second muscle-specific reporter gene, troponin I-Luc. The troponin I-Luc (TnI-Luc) reporter gene is a muscle-specific reporter gene that is regulated independently of SRF. C3H10T1/2 fibroblasts were transfected with TnI-Luc, pEM-MyoD, and either H-Ras G12V,E37G, RalGDS, or RalGDS CAAX. The expression of H-Ras G12V,E37G inhibited TnI-Luc activity by 80% (Fig. 5). Expression of RalGDS had no effect on TnI-Luc activity, and RalGDS CAAX inhibited activity of the reporter gene by only 30% (Fig. 5). The ability of RalGDS CAAX, a constitutively active form of RalGDS, to inhibit this reporter gene to any extent suggests that the RalGDS signaling pathway may have a minor role in the Ras-mediated inhibition of myogenesis. In this regard, we also tested the ability of RalGDS and RalGDS CAAX to inhibit the morphological differentiation of C3H10T1/2 cells. Compared with control cultures, RalGDS and RalGDS CAAX did not significantly affect fiber formation as detected by myosin heavy chain immunostaining (data not shown). Taken together, these data convincingly demonstrate that signaling through RalGDS does not duplicate the effects observed with H-Ras G12V,E37G in this cell type and provide strong evidence that a novel Ras effector mediates the inhibition of myogenesis by H-Ras G12V,E37G.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   H-Ras G12V,E37G, but not RalGDS or RalGDS CAAX, inhibits the TnI-Luc reporter gene. C3H10T1/2 cells were transfected with 2.0 µg of TnI-Luc, 0.5 µg of pEM-MyoD, and 1.0 µg of pDCR vector DNA, pDCR H-Ras G12V,E37G, pCEP4 RalGDS or pCEP4 RalGDS CAAX. Following 48 h in DM, cell extracts were prepared and analyzed for luciferase activity. Luciferase activity is expressed relative to the control which is set at 100 RLU (relative light units/µg of protein). Each value represents the average from at least three independent transfections. Error bars indicate the standard errors of the means.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

It has been known for many years that oncogenic H-Ras inhibits the differentiation of muscle cells independently of their continued proliferation (8, 18). H-Ras relies on multiple effector molecules including Raf/MEK/MAPK, Rac/Rho, and RalGDS to induce cellular transformation (1-7) and presumably a subset of these effectors to block differentiation events. H-Ras G12V,E37G is a variant of oncogenic H-Ras which does not activate the Raf/MEK/MAPK signaling pathway (2) and is defective in cellular transformation (10). Here we show that H-Ras G12V,E37G activates an intracellular signaling pathway which inhibits both the morphological differentiation of myogenic competent cells and the transcriptional activity of two muscle-specific reporter genes. Using the H-Ras effector-loop variants, H-Ras G12V,T35S and H-Ras G12V,Y40C, we demonstrated previously that Ras inhibits myogenesis independently of Raf/MEK/MAPK and Rac/Rho, suggesting the involvement of novel, Ras-activated signaling intermediates in this model system. The goal of this study, therefore, was to use H-Ras G12V,E37G to investigate further the possible identity of these intermediates.

H-Ras G12V,E37G has been shown to interact specifically with RalGDS (6). RalGDS and the membrane-associated variant RalGDS CAAX inhibit -actin-Luc activity and further decrease expression from this muscle-specific reporter gene when coexpressed with H-Ras G12V,E37G. H-Ras G12V,E37G does not enhance Gal4-Elk1 activity, confirming its inability to activate MAPK in this system. H-Ras G12V,E37G activates the transactivation potential of endogenous SRF, as evidenced by its ability to increase expression of the SRF-Luc reporter gene. Interestingly, RalGDS CAAX inhibits pSRF-Luc activity when expressed alone, and RalGDS, RalGDS CAAX, and RalA G23V inhibit H-Ras G12V,E37G-stimulated pSRF-Luc activity by 43, 66, and 49%, respectively. Since the full expression of -actin-Luc depends on the activity of SRF, the ubiquitous transcription factor SP1, and a muscle-specific regulatory factor such as MyoD (12), this result strongly suggests that one manner in which RalGDS and RalGDS CAAX block -actin-Luc activity is via the negative regulation of SRF. To address this issue further, we tested whether a second muscle-specific reporter gene, TnI-Luc (17), which is activated independently of SRF, is inhibited to the same extent by H-Ras G12V,E37G, RalGDS and RalGDS CAAX. The results of this experiment demonstrate that H-Ras G12V,E37G inhibits this reporter gene by 80%, that RalGDS CAAX exerts only a modest negative effect on TnI-Luc expression, and that RalGDS has no effect at all. These data suggest that constitutive activation of the RalGDS signaling pathway does not duplicate the inhibitory effects of H-Ras G12V,E37G in this system. We also have ruled out contributions from RalA and Rac1 proteins in this signaling pathway by demonstrating that the constitutively active RalA G23V protein has no effect on TnI-Luc activity² and that coexpression of dominant negative Rac1 does not reverse the inhibition of -actin-Luc activity induced by H-Ras G12V,E37G (Fig. 3), RalGDS or RalGDS CAAX.²

A recent report has shown that RalGDS does not activate the SRF-responsive c-fos promoter in NIH3T3 cells unless coexpressed with constitutively activated Raf-1 kinase (19). Separate studies have identified Rlf, a protein possessing 30% identity to RalGDS, as a molecule that associates with H-Ras G12V,E37G in vivo and that can stimulate c-fos promoter activity, alone or synergistically with H-Ras G12V,E37G (20, 21). At this time, we are unable to reconcile these data from other labs with our observation that RalGDS inhibits SRF-mediated transcription in C3H10T1/2 cells. It is our view that these contrasting results provide additional evidence for the existence of cell-type-specific signaling pathways for activated Ras and its effectors.

To date, all of the major Ras effectors have been tested for a role in the Ras-mediated inhibition of skeletal myogenesis (8, 9, 22) and none have been able to duplicate the effects of oncogenic Ras expression in this system. H-Ras G12V,E37G most likely activates a novel Ras effector which also is activated following the expression of H-Ras G12V,T35S or H-Ras G12V,Y40C. Designing strategies to identify this novel, Ras-induced signaling pathway operating in skeletal muscle precursor cells will be essential to fully understand how Ras activation impacts cellular proliferation, transformation, and differentiation events in a wide variety of cell types.