Department of Biological Sciences, Purdue University, West Lafayette, Indiana, IN 47907-1392, USA

Correspondence to: E J Taparowsky, Department of Biological Sciences, Purdue University, West Lafayette, Indiana, IN 47907-1392, USA

^aCurrent address: RHeoGene 706 Forest Street Charlottesville, VA 22903, USA

^bN Mitin and A Kudla to be treated as co-first authors

Oncogenic Ras (H-Ras G12V) inhibits skeletal myogenesis through multiple signaling pathways. Previously, we demonstrated that the major downstream effectors of Ras (i.e., MEK/MAPK, RalGDS and Rac/Rho) play a minor, if any, role in the differentiation-defective phenotype of Ras myoblasts. Recently, NFB, another Ras signaling target, has been shown to inhibit myogenesis presumably by stimulating cyclin D1 accumulation and cell cycle progression. In this study, we address the involvement of NFB activation in the Ras-induced inhibition of myogenesis. Using H-Ras G12V and three G12V effector-loop variants, we detect high levels of NFB transcriptional activity in C3H10T1/2-MyoD cells treated with differentiation medium. Myogenesis is blocked by all Ras proteins tested, yet only in the case of H-Ras G12V are cyclin D1 levels increased and cell cycle progression maintained. Expression of IB SR, an inhibitor of NFB, does not reverse the differentiation-defective phenotype of Ras expressing cultures, but does induce differentiation in cultures treated with tumor necrosis factor (TNF) or in cultures expressing the RelA/p65 subunit of NFB. These data confirm that NFB is a target of Ras and suggest that the cellular actions of NFB require additional signals that are discriminated by the Ras effector-loop variants. Results with IB SR convincingly demonstrate that H-Ras G12V does not rely on NFB activity to block myogenesis, an observation that continues to implicate another unidentified signaling pathway(s) in the inhibition of skeletal myogenesis by Ras. Oncogene (2001) 20, 1276-1286.

Ras; myogenesis; NFB; signal transduction

Introduction

Ras proteins are membrane-localized, small molecular weight GTPases that play a pivotal role in cellular proliferation, transformation and differentiation. Ras GTPases are the nodal point in an intracellular signaling network that relays key growth signals from receptor protein tyrosine kinases to downstream signaling intermediates. Ras function is essential for cellular proliferation and progression through the G1 phase of the cell cycle (Feramisco et al., 1984; Mulcahy et al., 1985; Dobrowolski et al., 1994; Aktas et al., 1997). A number of studies also have shown that Ras activity exerts dramatic and varied effects on developmental decisions. For example, signaling through Ras and its downstream effectors is essential for neurite outgrowth in PC12 cells (Bar-Sagi and Feramisco, 1985; Robbins et al., 1992; Thomas et al., 1992; Wood et al., 1992), while Ras activation blocks both the biochemical and morphological differentiation of skeletal muscle (Olson et al., 1987; Gossett et al., 1988; Konieczny et al., 1989; Lassar et al., 1989; Kong et al., 1995). The inhibition of myogenesis by oncogenic Ras has proven to be a particularly useful model for investigating the Ras effector pathways in muscle that influence individual cell growth, differentiation and transformation functions (Ramocki et al., 1997, 1998; Weyman et al., 1997).

Previous studies have demonstrated that oncogenic Ras (H-Ras G12V) and a panel of transformation-deficient Ras effector-loop variants inhibit MyoD-induced myogenesis in C3H10T1/2 (10T1/2) cells (Ramocki et al., 1997, 1998). Ras effector-loop variants are useful for examining Ras signaling pathways since each interacts with a distinct set of downstream Ras effector molecules. The H-Ras G12V, T35S variant retains Raf-1 binding and thus selectively activates the MEK/MAPK pathway (White et al., 1995). H-Ras G12V,E37G does not bind Raf-1 and signals predominantly through the guanine nucleotide exchange activity of RalGDS (White et al., 1996; Miller et al., 1997). Ras G12V, Y40C leads to the activation of the Rac/Rho GTPases and PI3 kinase (Joneson et al., 1996; Rodriguez-Viciana et al., 1997). Interestingly, each of these three Ras effector-loop variants inhibits myogenesis in a manner that is independent of their characterized downstream signaling targets. This strongly suggests that other Ras effectors and/or pathways not discriminated by these variants are critical to the Ras-induced inhibition of skeletal myogenesis.

One pathway that has not been examined with regard to the Ras-mediated inhibition of myogenesis is the NFB pathway. NFB belongs to the Rel family of transcription factors and regulates genes involved in several cellular functions, including immune and inflammatory responses, apoptosis and cell growth (reviewed in Pahl, 1999). In vertebrates, the Rel/NFB family consists of five proteins-RelA/p65, RelB, c-Rel, p50 (NF-B1) and p52 (NF-B2)-which form homodimers or heterodimers and bind to specific DNA target sites to regulate gene expression (reviewed in Baldwin, 1996; May and Ghosh, 1997). Different combinations of NFB subunits vary in their DNA-binding and activation potentials and only RelA, RelB and c-Rel possess transcription activation domains (reviewed in Verma et al., 1995). The best characterized form of NFB is a heterodimer composed of the RelA/p65 and p50 subunits. In most unstimulated cells, this heterodimer is present as an inactive, cytoplasmic complex bound by a member of an inhibitory protein family referred to as IB (reviewed in Verma et al., 1995). Upon exposure of cells to a variety of stimuli, including mitogens and growth factors, IB becomes phosphorylated on two serine residues located within the amino terminus. In its phosphorylated state, IB is targeted for ubiquitin-dependent degradation, releasing NFB which translocates to the nucleus to impact gene expression (reviewed in Baldwin, 1996; Karin, 1999).

Although the role of the NFB transcription complex as a regulator of genes involved in immunological responses is well-defined (reviewed in Pahl, 1999), recent studies have focused on the role of NFB in promoting cell growth. Cells in which NFB activation is blocked suffer a delay in progression through the G1 phase of the cell cycle (Bargou et al., 1997; Grumont et al., 1998; Guttridge et al., 1999; Hinz et al., 1999). This cell cycle defect, in part, could be the result of the inefficient accumulation of cyclin D1 since the cyclin D1 promoter contains consensus NFB binding sites that are transcriptionally responsive following the stimulation of NFB activity in cells (Guttridge et al., 1999; Hinz et al., 1999).

NFB serves as an important downstream target for Ras-activated signal transduction pathways. NFB is required for Ras-mediated cellular transformation (Finco et al., 1997; Jo et al., 2000) and is thought to play a role in protecting Ras-transformed cells from undergoing apoptosis (Mayo et al., 1997). Ras activation results not only in the nuclear accumulation of NFB, but also in the stimulation of the transcription function ascribed to the RelA/p65 subunit of the NFB heterodimer (Finco et al., 1997; Norris and Baldwin, 1999). Furthermore, Ras signaling, like NFB activation, results in increased cyclin D1 transcription and protein levels (Albanese et al., 1995; Aktas et al., 1997; Gille and Downward, 1999; Hitomi and Stacey, 1999). Recently, Guttridge et al. (1999) showed that constitutive activation of NFB is a potent inhibitor of skeletal muscle differentiation, presumably by preventing proper cell cycle arrest in differentiation-induced cultures. This model was consistent with previous reports demonstrating that ectopic expression of cyclin D1 blocks myogenesis in a cyclin-dependent kinase (Cdk)-dependent manner (Rao et al., 1994; Skapek et al., 1995; Zhang et al., 1999). Based on these observations, it is logical to conclude that oncogenic Ras utilizes the activities of NFB to stimulate G1 phase cell cycle progression and thus to block the terminal differentiation of skeletal muscle cells.

In this study, we have tested the role of NFB in the Ras-induced inhibition of myogenesis. Our results demonstrate that while oncogenic Ras and each of the Ras effector-loop variants described above effectively activate NFB-dependent transcriptional activity in MyoD-induced 10T1/2 cells, only oncogenic Ras leads to increased cyclin D1 protein levels and continued cell cycle progression. Additionally, we demonstrate that the inhibition of differentiation by all Ras constructs tested is not relieved by the overexpression of IB SR, a potent inhibitor of NFB activity. This is in stark contrast to the IB SR-mediated reversal of the block in myogenic differentiation induced by TNF treatment or by forced expression of the RelA/p65 subunit of NFB. These data rule out a major role for NFB in the Ras-mediated inhibition of myogenesis and continue to support the existence of a yet to be characterized pathway(s) that mediates the inhibitory effects of Ras activation on skeletal muscle differentiation.

Results

H-Ras G12V and G12V effector-loop variants stimulate NFB-mediated transcription in differentiation-induced 10T1/2 cells

H-Ras G12V and G12V effector-loop variants enhance NFB transcriptional activity in proliferating cells (Finco et al., 1997), but the ability of these proteins to enhance NFB activity in cells maintained under low growth, differentiation-inducing conditions has not been examined. Using a luciferase reporter gene containing three tandem B sites (3´B-Luc) as a read-out for NFB activation, 10T1/2 cells were transfected with an expression vector for MyoD and vectors encoding one of the following proteins: oncogenic H-Ras (Ras G12V); Ras G12V, T35S; Ras G12V, E37G; Ras G12V, Y40C or the RelA/p65 subunit of NFB. Cultures were treated with differentiation-inducing medium (DM) and luciferase activities measured as described in Materials and methods. Cells co-expressing MyoD and Ras G12V or MyoD and each of the Ras effector-loop mutants exhibit a 2-3-fold increase in 3´B-Luc reporter activity when compared to the vector only control (Figure 1). A similar level of 3´B-Luc expression is obtained with the transcriptionally active p65 subunit of NFB. Activation of the 3´B-Luc reporter in this experiment is a direct result of stimulating endogenous NFB activity, since co-expression of a constitutively active IB protein (IB SR) effectively blocks luciferase accumulation in all groups (Figure 1). IB SR expression has no effect on the transcriptional activity of control reporter genes assayed in parallel (data not shown), confirming that IB SR is not functioning as a general repressor of transcription in this model system. These data suggest that for MyoD expressing 10T1/2 cells maintained in DM, Ras G12V and each of the Ras effector-loop mutants tested enhance NFB-mediated transcription.

H-Ras G12V proteins and NFB inhibit myogenesis in 10T1/2-MyoD-ER cells

Recently, it has been reported that NFB activation inhibits myogenesis (Guttridge et al., 1999). C2C12 myoblasts treated with TNF, an activator of NFB, are blocked from differentiating and forced expression of the p65/RelA subunit of NFB in 10T1/2 cells inhibits the activation of muscle-specific reporter genes by MyoD (Guttridge et al., 1999). To extend these observations, we compared the ability of NFB and Ras G12V to inhibit myogenesis in a 10T1/2-MyoD inducible cell line (10T1/2-MyoD-ER) (Hollenberg et al., 1993). 10T1/2-MyoD-ER cells commit to terminal differentiation only when the MyoD-estrogen receptor (ER) fusion protein is activated by the addition of -estradiol to DM. As shown in Figure 2, 98% of the 10T1/2-MyoD-ER cells terminally differentiate in the presence of -estradiol as assayed by the expression of troponin T. Following transfection with expression vectors for Ras G12V or the G12V effector-loop variants T35S and E37G, the myogenic program is severely compromised. Interestingly, expression of Ras G12V, Y40C also inhibits myogenesis in induced MyoD-ER cells but, in this instance, the block in differentiation is not as complete as with the other Ras forms. As expected, treatment of 10T/2-MyoD-ER cells with TNF also proves to be a potent inhibitor of terminal differentiation (Figure 2). We established that TNF functions through NFB activation in these cells by measuring the stimulation of 3´B-Luc reporter gene activity by TNF and observing the restoration of myogenesis in TNF-treated cultures following expression of IB SR (data not shown). The data in Figures 1 and 2 provide strong evidence of a direct link between Ras and NFB activities in 10T1/2-MyoD expressing cells and suggest that NFB could function as the downstream effector mediating the inhibition of myogenesis by Ras.

Effects of Ras and NFB activation on cell cycle progression in 10T1/2-MyoD-ER cells

An early event in myogenic differentiation is the irreversible withdrawal of myoblasts from the cell cycle. As a result, one mechanism proposed for the block of myogenesis by constitutive activation of NFB is the induction of cell cycle progression (Guttridge et al., 1999). Considering the role of Ras signaling in cellular proliferation and transformation, it is possible that Ras activation inhibits muscle differentiation via this same mechanism. To test this possibility, we quantified the percentage of cells that continue to synthesize DNA (i.e., proliferate) in DM/-estradiol-treated 10T1/2-MyoD-ER myoblasts blocked from differentiating by the expression of H-Ras G12V or the Ras effector-loop variants or by exposure to TNF (Figure 3). Vector transfected 10T1/2-MyoD-ER cells grown in the absence of inducer continue to proliferate as revealed by the high percentage of BrdU positive cells in the culture. Induction of MyoD activity with -estradiol rapidly results in cell cycle arrest since only 5% of the cells remain BrdU positive. Interestingly, cells treated with -estradiol and expressing Ras G12V exhibit an increase in the number of cycling cells relative to the vector control, suggesting that Ras G12V expression supports a significant level of cell proliferation in DM. Similarly, 10T1/2-MyoD-ER cells in DM supplemented with -estradiol and TNF proliferate. In striking contrast to the behavior of the H-Ras G12V transfected cells, cells transfected with the Ras effector-loop variants (Ras G12V, T35S; Ras G12V, E37G; Ras G12V, Y40C) arrest growth in DM plus -estradiol, even though the cultures do not differentiate (see Figure 2). We conclude that the inhibition of myogenesis by the Ras effector-loop variants must occur through a pathway(s) that does not impinge on signals directing proper cell cycle arrest in DM.

To investigate one potential molecular correlate of the cell cycle progression induced by H-Ras G12V and TNF, we utilized cyclin D1 antisera to detect cyclin D1 accumulation in 10T1/2-MyoD-ER cells induced to differentiate with DM and -estradiol. In a previous study, Guttridge et al. (1999) described accumulation of cyclin D1 protein accompanying the treatment of differentiating myoblasts with TNF. Whether a similar induction of cyclin D1 accompanies Ras activation in myoblasts has not been examined. As shown in Figure 4, approximately 50% of vector transfected control cells maintained in the absence of -estradiol stain positive for cyclin D1. Following induction of differentiation, the number of cyclin D1 positive cells in these cultures drops to 10%, consistent with previous reports showing that cyclin D1 mRNA and protein decline as myoblasts arrest in early G1 (Rao et al., 1994; Skapek et al., 1995). Interestingly, 10T1/2-MyoD-ER cells exposed to -estradiol and expressing the Ras G12V protein exhibit a sevenfold increase in cyclin D1 positive cells when compared to the control -estradiol group, suggesting that Ras G12V supports G1 phase progression in DM by upregulating cyclin D1 protein levels. When cells transfected with each of the Ras effector-loop variants were examined, however, the number of cyclin D1 positive cells was not significantly different than the number in the -estradiol control. Thus, the data in Figures 3 and 4 indicate that the Ras effector-loop variants block myogenesis in DM-treated 10T1/2-MyoD-ER cells by a mechanism that does not involve cyclin D1 accumulation or continued G1-S phase progression.

IB expression does not reverse Ras-induced inhibition of skeletal muscle differentiation

Our results indicate that Ras effector-loop variants are likely to impact a set of cellular targets that is distinct from the set activated by NFB to block myogenesis. This in turn suggests that TNF-induced activation of NFB is qualitatively different from Ras-induced activation of NFB in this model system, even though other studies have reported that Ras activation mediates or enhances the effects of TNF signaling in some cell types (Trent et al., 1996; Anrather et al., 1999). To address the question of whether NFB serves as an obligate target of Ras signaling to inhibit myogenesis, we began by examining if TNF-induced activation of NFB in this system relies on endogenous Ras p21 function. For these experiments we used Ras N17, a dominant-negative form of H-Ras, that competitively inhibits the guanine nucleotide exchange required to activate endogenous Ras p21 (Feig and Cooper, 1988). The biological activity of the Ras N17 protein in 10T1/2 cells first was established using epidermal growth factor (EGF)-stimulated, Ras-mediated activation of Gal4-Elk1 as a readout. As shown in Figure 5a, the (Gal4)₅-Luc reporter gene is expressed to high levels in control cells transfected with Gal4-Elk1 and stimulated with EGF, while in cells transfected with Ras N17 and treated with EGF, luciferase activity is repressed by 60%. 10T1/2 cells then were transfected with MyoD and 3´B-Luc in the presence or absence of Ras N17 and luciferase activities measured after exposure to DM containing or lacking TNF. As shown in Figure 5b, TNF stimulates a high level of 3´B-Luc expression which is not reduced in the presence of Ras N17. In addition, Ras N17 does not reverse the TNF-induced inhibition of myogenesis as measured by the MyoD-mediated activation of TnI-Luc gene expression (Figure 5c) and the formation of multinucleate myofibers in the culture (data not shown). This suggests that the effects of TNF in stimulating NFB activation and the inhibition of skeletal muscle differentiation in this model system do not rely on the activation of endogenous Ras p21.

To establish if the Ras-initiated inhibition of myogenesis relies on the NFB pathway, we employed the constitutive NFB inhibitory protein, IB SR, to prevent activation of NFB in response to Ras signaling (see Figure 1). As a control, we duplicated previously reported experiments in which IB SR was used to revert the non-myogenic phenotype of 10T1/2 myoblasts transfected with the RelA/p65 subunit of NFB (Guttridge et al., 1999). 10T1/2 cells were co-transfected with the muscle-specific TnI-Luc reporter gene, a vector expressing MyoD and vectors for H-Ras G12V, each of the Ras effector-loop variants, p65 or, where indicated, IB SR. Expression of each introduced protein was confirmed by Western blot analysis (Figure 6b and data not shown). Following induction of differentiation, relative TnI-Luc activity was measured for each group, setting the MyoD-mediated activation of TnI-Luc to 100 (Figure 6a). Cells expressing H-Ras G12V, each of the effector-loop variants or RelA/p65 did not permit induction of TnI-Luc expression by MyoD. Introduction of the IB SR expression vector in the MyoD control group generates almost a twofold increase in TnI-Luc expression which is consistent with the report showing a positive influence of NFB inhibition on myogenesis (Guttridge et al., 1999). While IB SR efficiently reverses the inhibition of TnI-Luc expression by NFB, IB SR is unable to restore expression of this muscle-specific gene in groups expressing any of the Ras proteins.

To insure that the inability of the IB SR protein to restore TnI-Luc expression in Ras-inhibited cultures is an accurate reflection of other aspects of myogenic differentiation, the transfection was repeated and the differentiation-induced cultures fixed and immunostained for myosin heavy chain protein as described in Materials and methods. Myofiber formation was scored by visual inspection (Figure 6c and d) and reveals the presence of fully differentiated, myosin-positive fibers in the IB SR/RelA group, but not in the IB SR/Ras G12V group. We conclude from these data that while Ras activates NFB and NFB activation in response to other stimuli (e.g. TNF) inhibits skeletal muscle differentiation, NFB is not essential in the signaling pathway through which Ras blocks the biochemical and morphological differentiation of 10T1/2-MyoD myoblasts.

Discussion

The differentiation of skeletal muscle precursor cells in culture is marked by the transcriptional activation of muscle-specific genes and the morphological differentiation of myoblasts into multinucleate myotubes. Thus, myogenesis is an attractive model in which to dissect the signaling pathways that regulate various aspects of the process and serve to enhance, or to inhibit, this developmental progression. We have used skeletal muscle differentiation to show that the signaling pathways activated by an oncogenic Ras protein inhibit both the biochemical and the morphological differentiation of the cells (Konieczny et al., 1989; Lassar et al., 1989; Vaidya et al., 1991). Transformation-defective, Ras effector-loop variants, each of which signals predominantly through a subset of Ras effectors, also block myogenesis (Ramocki et al., 1997, 1998). This inhibition has been demonstrated in transient gene expression assays (Ramocki et al., 1997, 1998), in myoblast cell lines stably expressing Ras proteins (M Ramocki and E Taparowsky, unpublished data) and, in this study, using the 10T1/2-MyoD-ER cell line. The simplest interpretation of these data is that each of the Ras effector pathways is sufficient to block myogenesis. However, using constitutively active or dominant-negative signaling molecules and specific chemical inhibitors of Ras effector pathways, we and others have shown that none of these Ras pathways, alone or in combination, duplicate the effects of Ras in this system (Kaliman et al., 1996; Bennett and Tonks, 1997; Pinset et al., 1997; Weyman et al., 1997; Ramocki et al., 1998; Takano et al., 1998; Dorman and Johnson, 1999). For example, while constitutive activation of MAP kinase in the absence of Ras G12V effectively blocks the early stages of muscle differentiation (Bennett and Tonks, 1997; Ramocki et al., 1997; Dorman and Johnson, 1999), preventing MAP kinase activation in cells expressing Ras G12V or the T35S effector domain variant does not reverse the non-myogenic phenotype of the cells (Ramocki et al., 1997; Weyman et al., 1997). Similarly, whereas oncogenic Ras is known to activate PI3 kinase and the Rho GTPases in cells (Rodriguez-Viciana et al., 1994; Prendergast et al., 1995), exposure of myoblasts to chemical inhibitors of PI3 kinase blocks myogenesis (Kaliman et al., 1996; Pinset et al., 1997) and constitutive activation of RhoA enhances myogenesis (Ramocki et al., 1997; Takano et al., 1998; Meriane et al., 2000). Therefore, the available experimental evidence has resulted in our embracing an alternative hypothesis-that there is yet another pathway that mediates the effects of Ras in this model system.

In this study, we have investigated the role of NFB, another downstream Ras effector, in the inhibition of myogenesis. Our study was prompted by a number of reports describing the importance of NFB activation to cellular transformation by Ras (Finco et al., 1997; Jo et al., 2000), to progression through the G1 phase of cell cycle (Bargou et al., 1997; Grumont et al., 1998; Hinz et al., 1999) and to protecting cells from Ras-induced apoptosis (Mayo et al., 1997). Guttridge et al. (1999) demonstrated that constitutive activation of NFB blocks the differentiation of muscle cells, presumably by up-regulating transcription of the cyclin D1 gene and preventing the DM-induced cell cycle arrest that is a critical prerequisite to myogenic differentiation. These results supported earlier studies describing the non-myogenic phenotype of myoblasts overexpressing cyclin D1 (Rao et al., 1994; Skapek et al., 1995; Zhang et al., 1999). Our present work has extended these observations by showing that the forced expression of Ras G12V and the Ras effector-loop variants in 10T1/2-MyoD cells treated with DM results in the activation of endogenous NFB transcriptional activity. Furthermore, our studies confirm that constitutive activation of NFB, either by expression of the RelA/p65 subunit or by treatment of the cultures with TNF, blocks the biochemical and morphological differentiation of 10T1/2-MyoD cells. However, while Ras G12V and each of the Ras variants lead to the activation of NFB in DM, our results also show that only Ras G12V stimulates the accumulation of cyclin D1 protein and promotes a significant level of cell cycle progression. DM-treated cells expressing the Ras effector variants arrest growth appropriately, yet maintain a differentiation-defective Ras phenotype. The importance of NFB activation to the phenotype of Ras G12V expressing cells was investigated by co-expressing IB SR. This protein completely reversed NFB reporter gene expression stimulated by Ras, but did not restore expression of the muscle-specific reporter gene TnI-Luc or lead to any visible signs of morphological differentiation in the cultures (Figure 6). This is in stark contrast to the differentiation-defective phenotype of 10T1/2-MyoD cells expressing p65 or treated with TNF. In both of these cases, myogenesis is completely restored following expression of IB SR (Figure 6 and data not shown). We conclude from these experiments that activated Ras does not rely on the downstream activation of NFB to block myogenesis and, interestingly, that the cellular consequences of NFB activation by the Ras variants are qualitatively different from those induced by Ras G12V since they are not correlated with increased cyclin D1 accumulation or a failure to cell cycle arrest. Although we have not determined whether the lack of cyclin D1 accumulation with the variants is due to decreased cyclin D1 gene transcription or to increased cyclin D1 protein instability, we suspect the latter since our experiments with the NFB reporter gene indicate that all Ras proteins stimulate the transactivation properties of NFB appropriately. In this regard, Gille and Downward (1999) have demonstrated a synergy between individual Ras signaling pathways to activate cyclin D1 transcription, protein accumulation and protein activity during the G1 to S transition (Gille and Downward, 1999). Experiments currently underway with a cyclin D1 reporter construct will directly address these possibilities.

With a role for NFB in the inhibition of myogenesis by Ras ruled out, we next tested if activation of endogenous Ras p21 was an intermediate in the activation of NFB by TNF. By expressing the dominant-negative Ras N17 variant, which blocks full activation of MAP kinase by EGF in 10T1/2 cells (Figure 5a), we show no effect on the inhibition of myogenesis by TNF. Since signaling by TNF in other cell types is associated with Ras activation (Trent et al., 1996; Anrather et al., 1999), these results have revealed the existence of a TNF-induced, Ras-independent mechanism through which NFB can be activated to exert profound effects in cells. Interestingly, in a series of experiments designed to investigate the involvement of Ras activation in TNF-induced cell growth and apoptosis in C3H10T1/2 fibroblasts, Trent et al. (1996) showed that TNF treatment did not increase intracellular levels of GTP-bound Ras in this cell type. Our results would suggest that despite the inability to activate endogenous Ras p21, TNF treatment of 10T1/2 cells still leads to NFB activation.

We continue to be intrigued by the molecular mechanism of the Ras-induced inhibition of myogenesis, both by the identity of the components of the cytoplasmic signaling cascade activated by Ras in myoblasts and by the precise nuclear events initiated by this cascade that culminate in the inhibition of MyoD activity. Based on the data in this paper, we now have ruled out a significant contribution from NFB activation in this cascade and essentially any obligate role for continued cell cycle progression in the process. Clearly, the next step is to focus on qualitative differences between myoblasts arrested in the cell cycle in the presence or absence of the Ras effector-loop variants. The status of Cdk inhibitors such as p21 and p27 should be examined since previous reports have demonstrated a role for p21 and p27 in promoting myogenesis (Halevy et al., 1995; Skapek et al., 1995; Wang and Walsh, 1996; Zabludoff et al., 1998). Similarly, the potential regulation of the p38 MAPK pathway by Ras and the Ras variants is attractive, given the recent report that p38-directed phosphorylation events enhance MyoD-induced myogenesis (Zetser et al., 1999; Wu et al., 2000). In addition, we have initiated a library screen to identify novel Ras effectors in muscle cells. The complete delineation of the Ras signaling pathway(s) that prevents complete cell cycle withdrawal and terminal differentiation, without inducing cell cycle progression, will impact our understanding of skeletal muscle differentiation and may help to explain how early mutational activation of Ras proteins contributes to tumor initiation in vivo.

Materials and methods

Cell lines and media

C3H10T1/2 Clone 8 mouse embryo fibroblasts (10T1/2) were obtained from the American Type Culture Collection (ATCC CCL 226) and were maintained at subconfluent density in high glucose Dulbecco modified Eagle medium (DMEM-H) supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 g/ml streptomycin (GIBCO-BRL, Rockville, MD, USA). 10T1/2 MyoD-ER cells (Hollenberg et al., 1993) were maintained on gelatinized plates in DMEM-H, 10% calf serum (GIBCO-BRL) and antibiotics. Myogenesis was induced by exposure to differentiation medium (DM) consisting of DMEM-H, 2% horse serum and antibiotics. Where indicated, cultures were treated with medium supplemented with 0.1 M -estradiol (Sigma, St. Louis, MO, USA), 100 ng/ml EGF (Sigma) or 20 ng/ml TNF (Sigma).

Plasmids

pEMSV scribe2, pEM-MyoD, TnI-Luc, pGT5-Luc ((Gal4)₅-Luc), Gal4-Elk and the pDCR vectors expressing Ras G12V and the T35S, E37G and Y40C Ras variants have been described previously (Kong et al., 1995; Ramocki et al., 1997). pSVX RasN17 expressing dominant-negative Ras was obtained from S Green (University of Iowa). The 3´B-Luc reporter gene (Guttridge et al., 1999) and a plasmid expressing the RelA/p65 protein (Beg et al., 1992) were gifts of AS Baldwin (University of North Carolina). IkB SR (Brockman et al., 1995) was obtained from D Ballard (Vanderbilt University).

Transfection

10T1/2 or 10T1/2-MyoD-ER cells were seeded in 6-well plates at 1´10⁵ cells per well in growth medium. On the following day, the cells were exposed to SuperFect Transfection Reagent as instructed by the manufacturer (Qiagen, Valencia, CA, USA) in 600 l of growth medium containing the DNA amounts indicated in each Figure legend. After 3 h in SuperFect, the cells were rinsed in calcium and magnesium free (CMF)-saline (130 mM NaCl, 1.5 mM KH₂PO₄, 2.7 mM KCl, 8 mM Na₂HPO₄, 0.001% phenol red, pH 7.4) and the medium replaced by fresh growth medium for 24 h. To induce differentiation, the cultures were rinsed twice in CMF-saline and the medium switched to DM (containing -estradiol or TNF where indicated). After 24 h in DM, cell extracts were prepared and luciferase activities measured as described previously (Ramocki et al., 1997).

Immunofluorescence

10T1/2-MyoD-ER cells were plated in growth medium at a density of 2´10⁵ cells per gelatinized 60 mm dish. The following day, the cells were exposed to SuperFect reagent as described above using 1 ml of growth medium containing 2 g of the indicated plasmid DNA. Twenty-four hours after removal of SuperFect, the cultures were treated with DM and -estradiol for 48 h. Differentiated cultures were washed twice in CMF-saline, fixed with a solution of 4% formaldehyde in phosphate-buffered saline (PBS; 140 mM NaCl, 1.8 mM KH₂PO₄, 2.7 mM KCl, 10 mM Na₂HPO₄, pH 7.4) and stored at 4°C under PBS. Cells were permeabilized by incubating at 30°C for 45 min in a solution of PBS containing 1% horse serum and 0.1% Triton X-100. To detect hemagglutinin antigen (HA)-tagged Ras proteins and cyclin D1, cells were incubated first with 3 g/ml anti-HA monoclonal antibody 12CA5 (Roche, Indianapolis, IN, USA) followed by 6.7 g/ml fluorescein conjugated rabbit anti-mouse IgG_2b (American Qualex, San Clemente, CA, USA) and then with 1.3 g/ml anti-cyclin D1 monoclonal antibody 72-13G (Santa Cruz, Santa Cruz, CA, USA) followed by 5 g/ml Texas Red conjugated rabbit anti-mouse IgG1 (American Qualex). To detect HA-Ras and troponin T, cultures were incubated as above with the anti-HA primary antibody and fluorescein labeled secondary antibody and then with a 1 : 5 dilution of anti-troponin T monoclonal antibody CT3 (Developmental Studies Hybridoma Bank) followed by 5.7 g/ml Texas Red conjugated rabbit anti-mouse IgG_2a (American Qualex). Incubations were performed overnight at 4°C with each primary antibody and 75 min at room temperature with each secondary antibody. Cells were washed with PBS several times between treatments. Staining was visualized using an Olympus fluorescence microscope and photographed under 400´for analysis.

Cell cycle analysis

To approximate the per cent of cycling cells, 10T1/2 cultures were plated, transfected, differentiated and fixed as described in previous sections, except that 10 M 5-bromo-2'-deoxyuridine (BrdU; Roche) was added to the DM 18 h prior to fixation. Fixed cells were incubated with 2 N HCl for 1 h at room temperature, washed with TBE (90 mm Tris-borate, 2 mM EDTA, pH 8.0) and permeabilized as described above. Cells expressing HA-Ras proteins were detected as described above and BrdU uptake assessed in the same cultures by incubation with a 1 : 100 dilution of anti-BrdU monoclonal antibody G3G4 (Developmental Studies Hybridoma Bank) and 5 g/ml Texas Red conjugated secondary rabbit anti-mouse IgG1. Stained cells were visualized and photographed as described above.

Myofiber formation

The efficiency of myofiber formation in 10T1/2 cells transfected with MyoD was determined as described previously (Ramocki et al., 1997). Briefly, cells transfected with the amounts of plasmid DNA indicated in the Figure legend were exposed to DM for 48 h and fixed in a solution of 70% ethanol-formalin-acetic acid (20 : 2 : 1). A 1 : 5 dilution of monoclonal antibody MF-20 (Developmental Studies Hybridoma Bank) was used to detect myosin heavy chain protein which was visualized with a biotinylated secondary antibody and Vectastain kit reagents (Vector Laboratories). Immunostained cultures were viewed under bright light conditions and fibers scored and photographed at 100´magnification. For each experimental group, the average number of myofibers in five randomly chosen microscope fields is expressed relative to the MyoD control group which is set at 100.

Western blot hybridization

Cell extracts from transfected 10T1/2 cells were prepared, resolved by SDS-PAGE and transferred to nitrocellulose membrane as described previously (Ramocki et al., 1997). Blocking, hybridization and washes also were performed as described in Ramocki et al. (1997). The HA monoclonal antibody 12CA5 was used to detect HA-Ras and the Flag M2 monoclonal antibody (Sigma) was used to detect Flag-IB SR and Flag-p65 proteins. Immunoreactive complexes were visualized on the blot using peroxidase conjugated secondary antisera and enhanced chemiluminescent reagents purchased from Amersham

Acknowledgements

We thank A Baldwin for the 3´NF B-Luc reporter and RelA/p65 expression plasmids, S Tapscott for 10T1/2-MyoD-ER cells, D Ballard for the IB SR expression plasmid and D Guttridge for helpful suggestions and for generously communicating unpublished results. This work was supported by Public Health Service Grant AR41115-07 (to SF Konieczny) and a grant awarded to EJ Taparowsky from the Indiana Elks. A Kudla was supported by an American Heart Association Postdoctoral Fellowship 9804743W and by Public Health Service Grant T32 CA09634.

Aktas H, Cai H, Cooper GM. (1997). Mol. Cell. Biol. 17, 3850-3857. MEDLINE

Albanese C, Johnson J, Watanabe G, Eklund N, Vu D, Arnold A, Pestell RG. (1995). J. Biol. Chem. 270, 23589-23597. Article MEDLINE

Anrather J, Csizmadia V, Soares MP, Winkler H. (1999). J. Biol. Chem. 274, 13594-13603. Article MEDLINE

Baldwin Jr AS. (1996). Annu. Rev. Immunol. 14, 649-683. MEDLINE

Bar-Sagi D, Feramisco JR. (1985). Cell 42, 841-848. MEDLINE

Bargou RC, Emmerich F, Krappmann D, Bommert K, Mapara MY, Arnold W, Royer HD, Grinstein E, Greiner A, Scheidereit C, Dorken B. (1997). J. Clin. Invest. 100, 2961-2969. MEDLINE

Beg AA, Ruben SM, Scheinman RI, Haskill S, Rosen CA, Baldwin Jr AS. (1992). Genes Dev. 6, 1899-1913. MEDLINE

Bennett AM, Tonks NK. (1997). Science 278, 1288-1291. Article MEDLINE

Brockman JA, Scherer DC, McKinsey TA, Hall SM, Qi X, Lee WY, Ballard DW. (1995). Mol. Cell. Biol. 15, 2809-2818. MEDLINE

Dobrowolski S, Harter M, Stacey DW. (1994). Mol. Cell. Biol. 14, 5441-5449. MEDLINE

Dorman CM, Johnson SE. (1999). Oncogene 18, 5167-5176. MEDLINE

Feig LA, Cooper GM. (1988). Mol. Cell. Biol. 8, 3235-3243. MEDLINE

Feramisco JR, Gross M, Kamata T, Rosenberg M, Sweet RW. (1984). Cell 38, 109-117. MEDLINE

Finco TS, Westwick JK, Norris JL, Beg AA, Der CJ, Baldwin Jr AS. (1997). J. Biol. Chem. 272, 24113-24116. MEDLINE

Gille H, Downward J. (1999). J. Biol. Chem. 274, 22033-22040. Article MEDLINE

Gossett LA, Zhang W, Olson EN. (1988). J. Cell. Biol. 106, 2127-2137. MEDLINE

Grumont RJ, Rourke IJ, O'Reilly LA, Strasser A, Miyake K, Sha W, Gerondakis S. (1998). J. Exp. Med. 187, 663-674. MEDLINE

Guttridge DC, Albanese C, Reuther JY, Pestell RG, Baldwin Jr AS. (1999). Mol. Cell. Biol. 19, 5785-5799. MEDLINE

Halevy O, Novitch BG, Spicer DB, Skapek SX, Rhee J, Hannon GJ, Beach D, Lassar AB. (1995). Science 267, 1018-1021. MEDLINE

Hinz M, Krappmann D, Eichten A, Heder A, Scheidereit C, Strauss M. (1999). Mol. Cell. Biol. 19, 2690-2698. MEDLINE

Hitomi M, Stacey DW. (1999). Mol. Cell. Biol. 19, 4623-4632. MEDLINE

Hollenberg SM, Cheng PF, Weintraub H. (1993). Proc. Natl. Acad. Sci. USA 90, 8028-8032. MEDLINE

Jo H, Zhang R, Zhang H, McKinsey TA, Shao J, Beauchamp RD, Ballard DW, Liang P. (2000). Oncogene 19, 841-849. MEDLINE

Joneson T, White MA, Wigler MH, Bar-Sagi D. (1996). Science 271, 810-812. MEDLINE

Kaliman P, Vinals F, Testar X, Palacin M, Zorzano A. (1996). J. Biol. Chem. 271, 19146-19151. MEDLINE

Karin M. (1999). Oncogene 18, 6867-6874. MEDLINE

Kong Y, Johnson SE, Taparowsky EJ, Konieczny SF. (1995). Mol. Cell. Biol. 15, 5205-5213. MEDLINE

Konieczny SF, Drobes BL, Menke SL, Taparowsky EJ. (1989). Oncogene 4, 473-481. MEDLINE

Lassar AB, Thayer MJ, Overell RW, Weintraub H. (1989). Cell 58, 659-667. MEDLINE

May MJ, Ghosh S. (1997). Semin. Cancer Biol. 8, 63-73. MEDLINE

Mayo MW, Wang CY, Cogswell PC, Rogers-Graham KS, Lowe SW, Der CJ, Baldwin Jr AS. (1997). Science 278, 1812-1815. Article MEDLINE

Meriane M, Roux P, Primig M, Fort P, Gauthier-Rouviere C. (2000). Mol. Biol. Cell 11, 2513-2528. MEDLINE

Miller MJ, Prigent S, Kupperman E, Rioux L, Park SH, Feramisco JR, White MA, Rutkowski JL, Meinkoth JL. (1997). J. Biol. Chem. 272, 5600-5605. MEDLINE

Mulcahy LS, Smith MR, Stacey DW. (1985). Nature 313, 241-243. MEDLINE

Norris JL, Baldwin Jr AS. (1999). J. Biol. Chem. 274, 13841-13846. MEDLINE

Olson EN, Spizz G, Tainsky MA. (1987). Mol. Cell. Biol. 7, 2104-2411. MEDLINE

Pahl HL. (1999). Oncogene 18, 6853-6866. MEDLINE

Pinset C, Garcia A, Rousse S, Dubois C, Montarras D. (1997). CR Acad. Sci. III 320, 367-374.

Prendergast GC, Khosravi-Far R, Solski PA, Kurzawa H, Lebowitz PF, Der CJ. (1995). Oncogene 10, 2289-2296. MEDLINE

Ramocki MB, Johnson SE, White MA, Ashendel CL, Konieczny SF, Taparowsky EJ. (1997). Mol. Cell. Biol. 17, 3547-3555. MEDLINE

Ramocki MB, White MA, Konieczny SF, Taparowsky EJ. (1998). J. Biol. Chem. 273, 17696-176701. MEDLINE

Rao SS, Chu C, Kohtz DS. (1994). Mol. Cell. Biol. 14, 5259-5267. MEDLINE

Robbins DJ, Cheng M, Zhen E, Vanderbilt CA, Feig LA, Cobb MH. (1992). Proc. Natl. Acad. Sci. USA 89, 6924-6928. MEDLINE

Rodriguez-Viciana P, Warne PH, Dhand R, Vanhaesebroeck B, Gout I, Fry MJ, Waterfield MD, Downward J. (1994). Nature 370, 527-532. MEDLINE

Rodriguez-Viciana P, Warne PH, Khwaja A, Marte BM, Pappin D, Das P, Waterfield MD, Ridley A, Downward J. (1997). Cell 89, 457-467. MEDLINE

Skapek SX, Rhee J, Spicer DB, Lassar AB. (1995). Science 267, 1022-1024. MEDLINE

Takano H, Komuro I, Oka T, Shiojima I, Hiroi Y, Mizuno T, Yazaki Y. (1998). Mol. Cell. Biol. 18, 1580-1589. MEDLINE

Thomas SM, DeMarco M, D'Arcangelo G, Halegoua S, Brugge JS. (1992). Cell 68, 1031-1040. MEDLINE

Trent JC 2nd, McConkey DJ, Loughlin SM, Harbison MT, Fernandez A, Ananthaswamy HN. (1996). EMBO J. 15, 4497-4505. MEDLINE

Vaidya TB, Weyman CM, Teegarden D, Ashendel CL, Taparowsky EJ. (1991). J. Cell. Biol. 114, 809-820. MEDLINE

Verma IM, Stevenson JK, Schwarz EM, Van Antwerp D, Miyamoto S. (1995). Genes Dev. 9, 2723-2735. MEDLINE

Wang J, Walsh K. (1996). Science 273, 359-361. MEDLINE

Weyman CM, Ramocki MB, Taparowsky EJ, Wolfman A. (1997). Oncogene 14, 697-704. MEDLINE

White MA, Nicolette C, Minden A, Polverino A, Van Aelst L, Karin M, Wigler MH. (1995). Cell 80, 533-541. MEDLINE

White MA, Vale T, Camonis JH, Schaefer E, Wigler MH. (1996). J. Biol. Chem. 271, 16439-16442. MEDLINE

Wood KW, Sarnecki C, Roberts TM, Blenis J. (1992). Cell 68, 1041-1050. MEDLINE

Wu Z, Woodring PJ, Bhakta KS, Tamura K, Wen F, Feramisco JR, Karin M, Wang JYJ, Puri PL. (2000). Mol. Cell. Biol. 20, 3951-3964. MEDLINE

Zabludoff SD, Csete M, Wagner R, Yu X, Wold BJ. (1998). Cell Growth Differ. 9, 1-11. MEDLINE

Zetser A, Gredinger E, Bengal E. (2000). J. Biol. Chem. 274, 5193-5200.

Zhang JM, Wei Q, Zhao X, Paterson BM. (1999). EMBO J. 18, 926-933. MEDLINE

Figure 1 H-Ras G12V proteins induce NFB transcriptional activity. 10T1/2 cells were transfected with 1 g of 3´B-Luc, 0.25 g of pEM-MyoD and 0.5 g of pDCR vector (control), Ras G12V, Ras G12V,T35S, Ras G12V,E37G, Ras G12V,Y40C, RelA/p65 or 1 g of IB SR as described in Materials and methods. Twenty-four hours after transfection, the culture medium was switched to DM for 24 h. Cell extracts were prepared, normalized for protein content and assayed for luciferase activity as described previously (Ramocki et al., 1997). Luciferase activity is expressed relative to the control group which is set at 100. Error bars indicate the standard error of the mean

Figure 2 H-Ras G12V proteins inhibit myogenesis. 10T1/2-MyoD-ER cells were transiently transfected with 2 g of pDCR vector (control), Ras G12V, Ras G12V,T35S, Ras G12V,E37G or Ras G12V,Y40C. Twenty-four hours following transfection, the culture medium was switched to DM plus -estradiol. Where indicated, 20 ng/ml TNF was added at the time of addition of DM and again 12 h afterwards. Forty-eight hours following induction, the cells were fixed and immunostained for troponin T (TnT) and H-Ras expression as described in Materials and methods. (a) The number of H-Ras-positive cells that also express TnT was determined for each experimental group. The experiment was performed multiple times and the percentages averaged from 10 independent microscope fields per experiment. Error bars represent the standard error of the mean. (b) Representative microscope fields from the indicated experimental groups are shown. Arrows indicate the position of the H-Ras positive cells in both panels

Figure 3 Ras effector-loop variants do not compromise cell cycle arrest in differentiation-induced cultures. 10T1/2-MyoD-ER cells were transiently transfected with 2 g of pDCR vector (control); Ras G12V, Ras G12V,T35S, Ras G12V,E37G or Ras G12V,Y40C. Twenty-four hours after transfection the culture medium was switched to DM plus -estradiol. Where indicated, 20 ng/ml TNF was added with the DM and every 12 h afterwards. Thirty hours following induction, 10 M 5-bromo-2'-deoxy-uridine (BrdU) was added to the differentiation medium and 18 h later, the cells were fixed and immunostained for BrdU incorporation into DNA as described in Materials and methods. The number of H-Ras-positive cells that were also BrDU positive was determined for each experimental group. The experiment was performed multiple times and the percentages averaged from 10 independent microscope fields per experiment. Error bars represent the standard error of the mean

Figure 4 H-Ras G12V, but not the Ras effector-loop variants, cause increased cyclin D1 protein expression in DM. 10T1/2-MyoD-ER cells were transiently transfected with the indicated plasmid DNA as in Figure 3. Twenty-four hours following transfection, the culture medium was switched to DM plus -estradiol. Forty-eight hours following induction of differentiation, the cells were fixed and immunostained for cyclin D1 and H-Ras expression as described in Materials and methods. (a) The per cent of cyclin D1 positive cells was calculated as described in previous figures. Error bars represent the standard error of the mean. (b) Representative microscope fields from the indicated experimental groups are shown. Arrows indicate the cells expressing H-Ras

Figure 5 TNF-induced activation of NFB in 10T1/2 cells is Ras-independent and Ras inhibition of myogenesis does not rely on NFB. (a) 10T1/2 cells were transiently transfected with 1 g of (Gal4)₅-Luc, 0.5 g of Gal4- Elk1 and 0.5 g of Ras N17 in growth medium. Twenty-four hours after transfection, the culture medium was switched to DMEM-H. Twenty-four hours later, EGF was added (to a final concentration of 100 ng/ml) and the cells stimulated for 3 h. Cell extracts were prepared, normalized for protein content and assayed for luciferase activity as described previously (Ramocki et al., 1997). (b) 10T1/2 cells were transiently transfected with 1 g of 3´B-Luc, 0.25 g of MyoD and 0.5 g of pDCR vector or pSVX Ras N17. Twenty-four hours following transfection the culture medium was switched to DM and TNF (20 ng/ml) was added to the indicated groups. TNF was added again and at 12 and 24 h following induction, cell extracts were prepared, normalized for protein content and assayed for luciferase activity. (c) 10T1/2 cells were transiently transfected with 1 g of TnI-Luc, 0.25 g of MyoD and 0.5 g of pDCR vector or pSVX Ras N17. Cells were treated the same way as described in (b) and luciferase activity for each experimental group was expressed relative to the control which was set at 100. Error bars indicate the standard error of the mean

Figure 6 IB SR expression does not reverse the inhibitory effect of Ras on myogenesis. (a) 10T1/2 cells were transfected with 1 g of TnI-Luc, 0.25 g of MyoD and 0.5 g of pDCR vector (control), Ras G12V, Ras G12V,T35S; Ras G12V,E37G, Ras G12V,Y40C or p65/RelA. Where indicated (black bars), cells also received 1 g of IB SR. Twenty-four hours following transfection the culture medium was switched to DM for 24 h. Cell extracts were prepared, normalized for protein content and assayed for luciferase activity as described previously (Ramocki et al., 1997). Luciferase activity is expressed relative to the control which was set at 100. Error bars indicate the standard error of the mean. (b) 10T1/2 cells were transiently transfected with the indicated plasmids and treated with DM as in (a). Cell extracts were prepared and analysed by Western blot hybridization as described in Materials and methods. Anti-HA antiserum was used to detect H-Ras and anti-Flag antiserum was used to detect the IB SR and RelA/p65 proteins. (c) 10T1/2 cells were transiently transfected with 2 g of MyoD and 3 g of pDCR vector (control), Ras G12V or p65/RelA DNA. Where indicated, 6 g of IB SR were added to inhibit NFB activity. After 48 h in DM, the cells were fixed, immunostained for myosin heavy chain protein and scored for myofiber formation as described in Materials and methods. The values presented are the average myofiber formation of each group expressed relative to the MyoD control group, which is set to 100. Error bars indicate the standard error of the mean of the five counts used to obtain the average for each experimental group. (d) Representative photographs of the immunostained cultures scored for myofiber formation in (c)