Signaling from G Protein-coupled Receptors to the c- jun Promoter Involves the MEF2 Transcription Factor EVIDENCE FOR A NOVEL c-Jun AMINO-TERMINAL KINASE-INDEPENDENT PATHWAY*

The c-Jun amino-terminal kinases (JNKs) are a sub- family of mitogen-activated protein kinases that phosphorylate c-Jun and ATF2, and it has been postulated that phosphorylated c-Jun enhances its own expression through AP-1 sites on the c- jun promoter. In this study, we asked whether signals activating JNK regulate the c- jun promoter. Using NIH 3T3 cells expressing G pro- tein-coupled m1 acetylcholine receptors as an experimental model, we have recently shown that the cholin- ergic agonist carbachol, but not platelet-derived growth factor, potently elevates JNK activity. Consistent with these findings, carbachol, but not platelet-derived growth factor, increased the activity of a c- jun pro- moter-driven reporter gene (for chloramphenicol acetyltransferase). However, coexpression of JNK ki- nase kinase (MEKK) effectively increased JNK activity, but resulted in surprisingly limited induction of the c- jun promoter. This raised the possibility that pathway(s) distinct from JNK control the c- jun promoter, and prompted us to explore which of its regulatory elements participate in transcriptional control. We ob- served that deletion of the 3 (cid:42) AP-1 site used are: TPA, 12- O -tetradecanoylphorbol-13- acetate; TRE, TPA-responsive element; MAPK, mitogen-activated protein kinase; JNK, c-Jun amino-terminal kinase; mAChR, muscarinic acetylcholine receptor; PDGF, platelet-derived growth factor; CAT, chloramphenicol acetyltransferase; DMEM, Dulbecco’s modified Eagle’s medium; MEKK, JNK kinase kinase; MEKEE, constitutively active mutant of MEK; MEK, MAP kinase kinase; MEF, myocyte family.

The jun and fos gene families are nuclear proto-oncogenes whose expression is induced in quiescent cells by the addition of serum or other growth promoting stimuli (1,2). The expression of these genes is rapid and does not require newly synthe-sized proteins, thus suggesting that their induction involves posttranslational modification of pre-existing transcription factors. For example, c-jun is stimulated in a rapid and transient fashion by a variety of growth factors (1,2), including those acting on G protein-coupled receptors (3). The protein product of the c-jun gene, c-Jun, then homodimerizes or heterodimerizes with various Jun and Fos family members to form the AP-1 transcription factor (1,2,4). In turn, AP-1 binds a palindromic DNA sequence, known as the TPA 1 -responsive element or TRE, that is present within the regulatory region of a variety of genes. Interestingly, c-jun itself has been described among the genes that display TRE sequences in their promoter regions, thus suggesting that the product of the c-jun gene regulates its own expression (5). The induction of c-jun expression appears to be critical for cell proliferation, as microinjection of neutralizing anti-c-Jun antibodies blocks cell cycle progression upon serum stimulation (6).
Molecules affecting the function of nuclear transcription factors have just begun to be identified. One such example is the family of extracellular signal-regulated kinases or MAPKs (7,8). The role of MAPKs is to phosphorylate and regulate the activity of key molecules which ultimately control the expression of genes essential for many cellular processes, including cell growth, differentiation, programmed cell death, and neoplastic transformation (9). Recently, it has been shown that a novel family of kinases structurally related to MAPK, termed stress-activated protein kinases or c-Jun NH 2 -terminal kinases (JNKs), phosphorylate in vivo Ser-63 and Ser-73 in the NH 2terminal transactivating domain of the c-Jun protein, thereby positively regulating its transcriptional activity (10).
In our laboratory, we have genetically engineered NIH 3T3 murine fibroblasts to express subtypes of the human muscarinic acetylcholine receptors (mAChRs), and have utilized these cells as a model system for the study of proliferative signaling through G protein-coupled receptors (11). In this biological system, the m1 class of mAChR can effectively transduce mitogenic signals and, if persistently activated, can behave as a potent agonist-dependent oncogene (12). Recently, we have shown that activation of m1 G protein-coupled receptors by its agonist, carbachol, induces the expression of a clearly distinct set of nuclear transcription factors when compared with those * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  expressed in response to platelet-derived growth factor (PDGF) (3), a mitogen acting on endogenous tyrosine kinase receptors. In particular, carbachol caused a much greater induction of c-jun mRNA expression and AP-1 activity (3). Interestingly, we have observed that the m1 agonist, carbachol, but not PDGF, can induce a remarkable increase in JNK activity, following a temporal pattern distinct from that of MAPK activation (3).
In view of these observations, we set out to investigate whether signals activating JNK regulate expression from the c-jun promoter, using NIH 3T3 cells expressing m1 G proteincoupled receptors (NIH-m1.2 cells) as an experimental model. In these cells, we observed that carbachol, but not PDGF, potently increased the expression from a reporter plasmid containing the chloramphenicol acetyltransferase (CAT) gene under the control of the murine c-jun promoter. However, this effect was not mimicked by expression of activated molecules acting downstream from m1 receptors in the JNK pathway, although they activated JNK to an extent much greater than that caused by the m1 agonist alone. This prompted us to explore the existence of JNK-independent signaling pathways regulating the expression from the c-jun promoter. Using deletion and point mutational analysis of the c-jun promoter, we found a critical role for a DNA sequence that binds the MEF2 family of transcription factors. Furthermore, cotransfection with plasmids carrying the MEF2C and MEF2D cDNAs potently enhanced expression from the c-jun promoter in response to carbachol. In addition, we found that signaling from m1 receptors, but not direct JNK activation, induced expression from a MEF2-responsive plasmid. Taken together, these findings strongly suggest that signaling from G protein-coupled receptors to the c-jun promoter involves a novel, JNK-independent pathway acting on the MEF2 family of transcription factors.

MATERIALS AND METHODS
Cell Lines and Transfection-NIH 3T3 fibroblasts expressing approximately 20,000 human m1 mAChRs/cell, designated NIH-m1.2 cells (11), were maintained in Dulbecco's modified Eagle's medium (DMEM, Life Technologies, Inc.) supplemented with 10% calf bovine serum. Transfections were performed by the calcium phosphate precipitation technique, adjusting the total amount of plasmid DNA to 5-10 g/plate with vector alone.
Reporter Gene Assays-NIH-m1.2 cells were transfected with different expression plasmids together with 1 g of pCMV-␤gal, a plasmid expressing the enzyme ␤-galactosidase, and 1 g of each of the reporter plasmids. After overnight incubation, the cells were washed with serum-free DMEM, and kept for 24 h in DMEM supplemented with 0.5% fetal bovine serum. Cells were then stimulated with agonists for an additional 6 h and lysed using reporter lysis buffer (Promega).
CAT activity was assayed in the cell extracts by incubation for 6 -16 h in the presence of 0.25 Ci of [ 14 C]chloramphenicol (100 mCi/mmol) and 200 g/ml butyryl-CoA in 0.25 M Tris-HCl, pH 7.4. Labeled butyrylated products were extracted using a mixture of xylenes (Aldrich) and counted as described (3). Luciferase activity present in cellular lysates was assayed using D-luciferin and ATP as substrates, and light emission was quantitated with the aid of a luminometer, Monolight 2010, as specified by the manufacturer (Analytical Luminescence Laboratory). ␤-Galactosidase activity present in each sample was assayed by a colorimetric method, and used to normalize for transfection efficiency.
Kinase Assays-Phosphorylating activity of endogenous MAPK, JNK, and p38, or of transfected, epitope-tagged MAPK, JNK, and p38 was determined as follows. Cells were seeded at 10% confluence, and the following day cells were left untreated for endogenous activity, or transfected with pcDNA3-HA-MAPK, pcDNA3-HA-JNK, and pcDNA3-HA-p38 and additional DNAs by the LipofectAMINE technique (Life Technologies, Inc.). Total amount of plasmid DNA was adjusted to 5-10 g/plate with vector DNA (pcDNA3, Invitrogen) when necessary. Two days later, NIH-m1.2 cells were incubated in serum-free medium overnight for MAPK or for 2 h for JNK and p38, respectively. Cells were then stimulated with agonists for different periods of time at 37°C, washed with cold phosphate-buffered saline, and lysed as described (3). Cleared lysates were immunoprecipitated on ice for 2 h with an antibody against ERK-2 (SC-154, Santa Cruz Biotechnology) for endogenous MAPK, with an antibody against JNK-1 (15701A, Phar-Mingen) for endogenous JNK, and with the anti-HA monoclonal antibody (HA.11, Berkeley Antibody Co.) for the epitope-tagged MAPK, JNK, and p38. The immunoprecipitates were washed, and in vitro kinase assays were performed using 1.5 g/l myelin basic protein (Sigma) as substrate for MAPK or 1 g of purified, bacterially expressed, GST-ATF2 protein as substrate for JNK and p38. Samples were analyzed by SDS-gel electrophoresis on 12% acrylamide gels, and autoradiography was performed with the aid of an intensifying screen.
DNA Constructs-A reporter plasmid containing the MEF2 DNA binding sequence (pGL3MEF2) was engineered by inserting the sequence 5Ј-TCGACGGGCTATTTTTAGGGCCG-3Ј in the SalI site of the pGL3 plasmid DNA (Promega), upstream of an SV40 minimal promoter and a luciferase gene. The mutant sequence 5Ј-TCGACGGGCGATTTT-TCGGGCCG-3Ј, that does not bind MEF2 proteins (13), was used to generate pGL3MEF2mut. pJC6, pJC8, pJC9, pJTX, pJSX, and pJSTX have been described previously (14). MEF2C and MEF2D cDNAs were kindly provided by Dr. Bruce Paterson (National Cancer Institute, National Institutes of Health, Bethesda, MD) and subcloned in the pcDNAIIIB expression vector (Invitrogen). Other cDNA expression vectors have already been described (15).

RESULTS
In previous studies, we have observed that stimulation of G protein-coupled receptors of the m1 class by the cholinergic agonist, carbachol, potently elevates JNK activity (3,16). In turn, JNK has been shown to phosphorylate the NH 2 -terminal transactivating domain of c-Jun, thereby stimulating AP-1mediated transcription (17). We have also reported that, consistent with these observations, stimulation of quiescent NIH-m1.2 cells with carbachol induces AP-1 activity and greatly enhances the expression of c-jun mRNA (3). As an approach to explore the relationship between the enzymatic activity of JNK and the regulation of c-jun expression, we transfected NIH-m1.2 cells with a reporter construct containing murine c-jun promoter sequences from Ϫ225 to ϩ150 fused to the CAT gene (pJC6) (14) (see below). In agreement with our previous observations (3), we show in Fig. 1 that, whereas carbachol, the tyrosine kinase-receptor ligand PDGF, and the phorbol ester TPA induced MAPK activity to similar extents, only carbachol elevated the phosphorylating activity of JNK in these cells. When reporter gene assays were performed in parallel, we observed that c-jun promoter activity was strongly enhanced by addition of carbachol, whereas treatment with PDGF and TPA did not result in demonstrable activation. Thus, these data suggest that the activity of the c-jun promoter correlated with JNK, but not MAPK activation.
To further explore the roles of JNK and MAPK in the regulation of expression from the c-jun promoter, we employed a constitutively active mutant of MEK, MEKEE, which results in a remarkable increase in MAPK activity (15,18), and a truncated JNK kinase kinase, MEKK, which potently activates JNK (19,20). Accordingly, we cotransfected expression plasmids for MEKEE and MEKK together with plasmids expressing epitope-tagged MAPK and epitope-tagged JNK, or the c-jun promoter containing reporter plasmid. As shown in Fig. 2, whereas MEKEE strongly induced MAPK activity, it had no effect on JNK or on the expression of the c-jun promoter-driven reporter plasmid. On the other hand, MEKK did not elevate MAPK activity but potently stimulated JNK-phosphorylating activity, to an extent much greater than even that elicited by m1 receptor stimulation. Surprisingly, however, activation of CAT expression in response to carbachol was severalfold higher than that induced by MEKK (Fig. 2). These observation raised the possibility that signaling pathways in addition to JNK, may participate in the activation of the c-jun promoter by m1 G protein-coupled receptors.
The lack of a strict correlation between JNK activation and the expression from the c-jun promoter-containing reporter plasmid prompted us to search for those regulatory elements in the c-jun promoter that participate in the m1-induced response. In this regard, binding sites for Sp1, AP-1, CTF, and MEF2 transcription factors have been identified in the c-jun promoter at nucleotides Ϫ117, Ϫ92, Ϫ72, and Ϫ59, respectively, and two GATAA elements are found at Ϫ33 and Ϫ11 (Fig. 3A) (14). In addition, a putative site for AP-1, referred to here as AP-1-like, is located at Ϫ180 (4). To assess the contribution of the AP-1-like and Sp1 sites, and the CAAT box in the response to m1 receptors, we used two deletion mutants, pJC8 and pJC9, containing fragments Ϫ115 to ϩ150 and Ϫ80 to ϩ150 of the c-jun promoter, respectively (Fig. 3A). As shown in Fig. 3B, carbachol addition to cells transfected with pJC8 or pJC9 caused an even higher level of CAT expression than that observed for pJC6. These data suggest that the AP-1-like and the Sp1 sites that are absent in both constructs, as well as the CAAT box that is deleted in pJC9, are dispensable for signaling from m1 receptors to the c-jun promoter.
To explore the role of the AP-1 and MEF2 sites, we used a plasmid, pJTX, that harbors four point mutations in the TRE element located at Ϫ72 thereby abolishing AP-1 binding (5, 14); a plasmid, pJSX, containing two point mutations, at nucleotides Ϫ51 and Ϫ58, that have been shown to abolish MEF2 binding (13,14); and a double mutant, pJSTX, that lacks both AP-1 and MEF2 binding sites (14) (Fig. 3A). As expected, in control experiments we observed that mutation of the AP-1 site, but not of the MEF2 site, nearly abolishes the expression from the reporter plasmid elicited by cotransfection of the FIG. 1. The cholinergic agonist carbachol, but not PDGF or TPA, induces the c-jun promoter in NIH 3T3 cells expressing m1 G protein-coupled receptors. NIH-m1.2 cells were cotransfected with pJC6 and pCMV-␤-gal reporter plasmid DNAs (1 g/plate each) by the calcium phosphate technique. 40 h later, cells were left untreated or exposed for 4 h to 100 M carbachol, 10 ng/ml PDGF, or 50 ng/ml TPA, as indicated. Cells were collected, and the lysates were assayed for CAT and ␤-galactosidase activity. The data represent CAT activity normalized by the ␤-galactosidase activity present in each sample, and are the average Ϯ S.E. of triplicate samples from a typical experiment. Similar results were obtained in five independent experiments. The autoradiogram corresponds to a representative experiment resolved on a thin layer chromatography (TLC) plate. Confluent plates of NIH-m1.2 cells were serum-starved and treated with 100 M carbachol, 10 ng/ml PDGF, or 50 ng/ml TPA as indicated, for 5 or 15 min for MAPK or JNK activity, respectively. In vitro kinase reactions were performed in immunoprecipitates from cellular lysates, and the reaction mixtures were fractionated by 12% SDS-polyacrylamide gel electrophoresis. Positions of the labeled products are indicated by arrows. Similar results were obtained in three independent experiments. MBP, myelin basic protein; n-but-14 C-chl., 14 C-labeled chloramphenicol.

FIG. 2. Effects of the activation of the MAPK and JNK cascades
on the c-jun promoter activity. NIH-m1.2 cells were cotransfected with pJC6 (1 g/plate) and pCMV-␤-gal (1 g/plate) plasmid DNAs by the calcium phosphate technique. Expression vectors for MEKK or the constitutively activated mutant of MEK (MEKEE) were included in the transfection mixtures, as indicated, as an approach to activate the JNK or MAPK pathway, respectively. 48 h later, cells were left untreated or exposed for 4 h to 100 M carbachol, as indicated. The cells were collected, and the lysates were assayed for CAT and ␤-galactosidase activity. The data represent CAT activity normalized by the ␤-galactosidase activity present in each sample, and are the average Ϯ S.E. of triplicate samples from a typical experiment. Similar results were obtained in three independent experiments. In parallel, NIH-m1.2 cells were transfected with pcDNA3-HA-JNK1 (1 g/plate) or pcDNA3-HA-MAPK (1 g/plate) for MAPK and JNK assays, respectively, together with pcDNA3 vector (control) or expression vectors carrying cDNAs for MEKK or MEKEE, as indicated (2 g/plate in each case). Treatment of cells with 100 M carbachol for 5 or 15 min was done for MAPK or JNK, respectively. Kinase reactions were performed in anti-HA immunoprecipitates from the corresponding cellular lysates as described under "Materials and Methods." 32 P-Labeled products are indicated. Autoradiograms correspond to representative experiments for MAPK and JNK assays, which were repeated three times with similar results. activated JNK kinase kinase, MEKK (data not shown). However, the TRE-deficient mutant, pJTX, showed only a relatively small reduction in the response to m1 stimulation (Fig. 3B). In contrast, the MEF2-binding mutant, which contains an intact TRE site (pJSX), showed a drastic reduction in the m1-induced response (70 -80%). Furthermore, as shown in Fig. 3B, mutation of both MEF2 and TRE sites in the c-jun promoter (pJSTX) nearly abolished the expression of CAT elicited by carbachol in m1-expressing cells. These results suggest that both the MEF2 and AP-1 regulatory sites are critical for the regulation of expression from the c-jun promoter in response to intracellular signals transmitted by G protein-coupled receptors. In addition, these data suggest that, in these cells, the majority of the stimulatory response on the c-jun promoter-containing reporter plasmid is exerted through the MEF2 binding site and not through the TRE, as we had initially hypothesized.
MEF2 proteins bind the consensus sequence motif CTA(A/ T) 4 TAG (13,14), which is present in the c-jun promoter between positions Ϫ50 and Ϫ59. As an attempt to further investigate if that sequence participates in the transcriptional response triggered by stimulation of m1 receptors, we engineered a reporter construct by introducing oligonucleotides that match exactly the sequence CTATTTTTAG found in the c-jun promoter, upstream of a minimal promoter in the FIG. 3. Regulatory elements controlling the activity of the cjun promoter. A, pJC6 is a pBLCAT3-based reporter construct that contains bases Ϫ225 to ϩ150 of the murine c-jun promoter. pJC8 and pJC9 contain bases Ϫ115 to ϩ150 and Ϫ80 to ϩ150, respectively. The sites of point mutations in the AP-1 binding site that abolish AP-1 binding (pJTX) and point mutations at the MEF2 binding site that abolish MEF2 binding (pJSX) are indicated by double vertical lines. The plasmid pJSTX contains mutations for both the AP-1 and MEF2 sites. B, NIH-m1.2 cells were cotransfected with the reporter plasmids pJC6, pJC8, pJC9, pJTX, pJSX, or pJSTX (1 g/plate) as indicated. In each case, 1 g of pCMV-␤-gal reporter plasmid DNA was included in the transfection mixtures. 40 h later, cells were left untreated or exposed for 4 h to 100 M carbachol. Cells were collected, and the lysates were assayed for CAT and ␤-galactosidase activity. The data represent CAT activity normalized by the ␤-galactosidase activity present in each sample, and are the average Ϯ S.E. of triplicate samples from a typical pGL3promoter plasmid (Promega). The resulting construct expresses a luciferase reporter gene and was designated pGL3MEF2. As a control, we generated a similar reporter plasmid containing the mutant sequence CGATTTTTCG, which does not bind MEF2 (13,14) and therefore is not able to respond to MEF2 (pGL3MEF2mut). Fig. 4 shows that upon transfection in NIH-m1.2 cells with pGL3MEF2, carbachol induced a remarkable increase in the luciferase activity when compared with the unstimulated control cells. Furthermore, coexpression of MEF2C (Fig. 4) or MEF2D (data not shown) also elevated the reporter gene activity. However, no demonstrable response was observed upon transfection with MEKK, thus suggesting that the MEF2 binding site is insensitive to the activation of the JNK pathway. On the other hand, cells transfected with the construct pGL3MEF2mut exhibited the same level of luciferase activity in each of these conditions, supporting that an intact MEF2 binding site is required to elicit a transcriptional response. Taken together, these findings strongly suggest that the MEF2 binding site could be the target for signaling pathways linking G protein-coupled receptors to the c-jun promoter.
To address directly whether proteins of the MEF2 family could regulate the activity of the c-jun promoter, we next cotransfected pJC6 with expression plasmids harboring the cDNAs for MEF2C or c-Jun in NIH-m1.2 cells. As shown in Fig.  5, CAT expression driven by the c-jun promoter was elevated by cotransfection with the c-Jun and MEF2C expression plasmids. A similar result was obtained by overexpressing MEF2D (data not shown). When cells were stimulated with carbachol, we observed that coexpression of c-Jun potentiated the response by only 30 -40%, while both MEF2C (Fig. 5) and MEF2D (data not shown) nearly tripled CAT activity in response to the cho-linergic agonist. Taken together, these findings support a critical role for MEF2 proteins in signaling from G protein-coupled receptors to the c-jun promoter.
A recent study (29) showed that one of the members of the MEF2 family (MEF2C) is phosphorylated by a novel member of the MAPK superfamily, p38, in myeloid lineage cells. Therefore we explored whether p38, by regulating the activity of MEF2, might mediate in the induction of the c-jun promoter by G protein-coupled receptors. To that end, we transfected NIH-m1.2 cells with pJC6 and measured CAT activity upon carbachol stimulation in the presence or absence of the specific p38 inhibitor, SB203580 (30). As shown in Fig. 6, incubation with SB203580 nearly abolished p38 activation by anisomycin or carbachol but did not have any demonstrable effect on transcriptional activation by the c-jun promoter. Similar results were obtained using pJTX, the mutant of the c-jun promoter lacking the AP-1 site but containing a MEF2-responsive element (Fig. 6). These results suggest that p38 does not regulate MEF2 activation of the c-jun promoter in NIH-m1.2 cells. DISCUSSION Positive autoregulation has been considered the major regulatory event in the response of the c-jun promoter to proliferative stimuli. Two steps have been described as participating in that process: 1) post-translational modification of preexisting c-Jun proteins and 2) transcriptional activation of c-jun that leads to an increase in the pool of c-Jun protein present in the cell (1,4,5,21). c-Jun is a constitutive part of the AP-1 transcription factor, and binding of such a factor to the TRE element in the c-jun promoter has been shown to result in the stimulation of transcription (5). Using NIH-3T3 cells expressing m1 G protein-coupled receptors, we had previously ob- served that addition of the cholinergic agonist, carbachol, results in a remarkable increase in JNK activity and in the level of c-jun expression. Thus, we expected a major participation of the JNK pathway on the regulation of c-jun promoter activity. Indeed, we observed a positive correlation between the c-jun promoter activity and JNK, but not with MAPK. Surprisingly, direct activation of the JNK pathway by overexpression of MEKK resulted in activation of the c-jun promoter but to a much more limited extent than that induced upon m1 stimulation. This observation prompted us to ask whether other biochemical routes, in addition to the JNK pathway, participate in the transduction of signals from G protein-coupled receptors to the promoter of c-jun.
Mapping of the regulatory elements in the c-jun promoter has yielded the existence of several consensus binding sites for different transcription factors such as Sp1, CTF, AP-1, and MEF2 (5,14). Using a series of c-jun promoter mutants fused to a reporter gene, we observe that deletion of the binding sites for Sp1 or CTF did not affect reporter gene activity. Furthermore, we found that mutations that abolish AP-1 binding prevent expression from the reporter plasmid when induced by MEKK, but display only a partial reduction in the response to carbachol. In contrast, a more dramatic decrease was observed when mutations that inhibit binding of MEF2 factors were introduced, thus suggesting a role for the MEF2 factors in regulating c-jun promoter activity, perhaps even more critical than that played by AP-1 itself. Furthermore, the fact that the double mutant that does not bind AP-1 or MEF2 fails to respond to carbachol, indicates that, for signals initiated at the level of m1 G protein-coupled receptors, the AP-1 and MEF2 sites are sufficient to account for the transcriptional response of the c-jun promoter.
While extensive studies have established the role of AP-1 complexes in transcriptional control, very limited information is available about the functions of MEF2 proteins. The myocyte enhancer family (MEF) of transcription factors comprises a group of transcriptional activators MEF2A, -B, -C, and -D (22)(23)(24)(25). MEF2 binding sites have been found in the promoters and enhancers of many skeletal and cardiac muscle structural genes (26). Structurally, MEF2 proteins contain a common motif that was named the MEF2 domain, this feature being unique to all MEF2 family members described so far (26). Despite the importance of MEF2 factors in the regulation of muscle gene expression, very little is known about the mechanism by which these proteins promote transcription or the signal transduction pathways controlling their activity. Recently, MEF2C, but not the other members of the family, has been shown to be a substrate for the MAPK-related p38 (29). While the expression of MEF2A, -B, and -D proteins is ubiquitous, the expression of MEF2C seems to be restricted to cortical neurons, monomyelocytic and muscle cells (23,24,27). This wide distribution suggests that their function is not limited to the regulation of muscle specific gene expression (27). Interestingly, they share with other transcription factors, such as serum response factor, the presence of a common structural domain termed MADS box, and due to that characteristic they were originally named RSRFs (related to serum response factors) (13).
In this study, we provide evidence for a critical role for MEF2 in the regulation of c-jun promoter activity by stimulation of G protein-coupled receptors. In addition, we show that overexpression of the MEF2 protein potentiates carbachol-induced expression of the c-jun promoter, strengthening the concept of MEF2 proteins as the targets of a yet uncharacterized signal transduction pathway. We also report here that an isolated MEF2 binding site stimulates transcription of a reporter gene when m1-expressing cells are stimulated by carbachol, but not by cotransfection with the JNK-inducing construct MEKK. Taken together, our data support the existence of two distinct signaling pathways that initiating on the same receptor at the level of the plasma membrane converge on the c-jun promoter: a JNK-dependent pathway, acting on the AP-1 responsive element; and a JNK-independent pathway, acting on the MEF2 regulatory sequence. In this regard, a recent report showed that the transactivating activity of MEF2C can be regulated by p38 (29). However, we observed that SB205380, a potent p38 and p38␤ blocker, fails to inhibit c-jun promoter activity, thus ruling out these p38 proteins as mediators of the carbacholinduced signal to MEF2. Nevertheless, NIH 3T3 cells express MEF2A, -B, and -D but not MEF2C (23,24,27), and it has been recently shown that p38␥ and p38␦ are not inhibited by SB205380 (31). Thus, it is likely that these or other members of the MAPK superfamily might mediate MEF2 activity in these FIG. 6. The p38 inhibitor, SB203580, fails to block the induction of c-jun promoter activity by carbachol. NIH-m1.2 cells were transfected with 1 g of pJC6 or pJTX as indicated. In each case, 1 g of pCMV-␤-gal reporter plasmid DNA was included in the transfection mixture. 48 h later, cells were left untreated or exposed for 2 h to 20 M SB203580 or left untreated, followed by a subsequent incubation for 4 h in the presence or absence of 100 M carbachol, as shown. Cells were collected, and the lysates were assayed for CAT and ␤-galactosidase activity. The data represent CAT activity normalized by the ␤-galactosidase activity present in each sample, and are the average Ϯ S.E. of triplicate samples from a representative experiment. Similar results were obtained in two independent experiments. In parallel, NIH-m1.2 cells were transfected with pcDNA3-HA-p38 (1 g/plate) for p38 kinase assays and treated with 20 M SB203580 for 2 h or left untreated. Cells were then incubated with 100 M carbachol or 10 M anisomycin for 15 min. as indicated. Kinase reactions were performed in anti-HA immunoprecipitates from the corresponding cellular lysates as described under "Materials and Methods." 32 P-Labeled products are indicated. cells. The identity of such regulatory molecule acting on other members of the MEF2 family of transcription factors is under current investigation.
Our results assign a critical role for MEF2 in the regulation of c-jun expression by the G protein-linked class of cell surface receptors. Interestingly, Han et al. provided evidence that the MEF2 binding site in the c-jun promoter also participates in the activation of c-jun by epidermal growth factor receptors and serum in HeLa cells (14,28). Thus, the elucidation of the molecular entities that participate in the transduction of signals from the membrane to MEF2 proteins, as well as of the mechanisms whereby those signaling molecules control the transcriptional activity of MEF2, is expected to be central for understanding the regulation of c-jun expression by growth promoting factors. Furthermore, considering the ubiquitous expression of MEF2 proteins and their resemblance to other transcription factors such as serum response factor, we can envision a growing interest for dissecting the signaling pathways that converge on MEF2.