The mitogen-activated protein kinase signaling cascade is used by many G protein-coupled receptors to initiate functional events. In this study, activation of the Gq/G11-coupled thromboxane A2 (TxA2) receptor (TP) by the TxA2 mimetic IBOP in ECV304 cells was found to induce extracellular regulated kinase (ERK) phosphorylation and tyrosine phosphorylation of the epidermal growth factor receptor (EGFR), which were inhibited by the TP antagonist SQ29548, the EGFR kinase inhibitor AG1478, the Src family kinase inhibitor PP1, the Gi/o protein inhibitor pertussis toxin (PTX), or the protein kinase C (PKC) inhibitor calphostin C. TP activation also increased Src kinase activity, which was blocked by PTX, PP1, and calphostin C, but not by AG1478, indicating that Src activation occurs before phosphorylation of EGFR. Blockade of Src activity by expression of dominant negative mutant of Src inhibits mitogen-activated protein kinase (MAPK) activation induced by TxA2. ERK activation induced by the PKC activator phorbol myristate acetate was inhibited by PTX, PP1, AG1478, and calphostin C. In contrast, activation of ERK by lysophosphatidic acid, a Gi-coupled receptor activator, was inhibited by PTX, PP1, and AG1478, but not by calphostin C. Thus, TP-stimulated ERK activation requires Gi, which in turn requires PKC activation. Immunoprecipitation of Gαi showed increased association of Gαi with TPα following PKC activation. In conclusion, TPα is directly coupled to the Gi protein by a PKC-regulated mechanism; Gi coupling causes Src-dependent transactivation of the EGFR, which is the dominant pathway in TP-mediated ERK activation.
The thromboxane A2 (TxA2) receptor (TP), a member of the G protein-coupled receptor (GPCR) superfamily (Coleman et al., 1994), mediates TxA2-induced platelet aggregation and vasoconstriction (Fitzgerald et al., 1987). Dysregulation of TxA2 synthesis and function has been implicated in the pathogenesis of a number of disease states including myocardial ischemia (Dorn et al., 1990), asthma (Devillier and Bessard, 1997), pregnancy-induced hypertension (Meagher and FitzGerald, 1993), and a variety of kidney diseases (Patrono et al., 1993). The TPs are expressed in a number of tissues including platelets, placenta, and endothelial cells (Hirata et al., 1991;Raychowdhury et al., 1994). Two isoforms of human TPs have been cloned from placenta (TPα) (Hirata et al., 1991) and endothelium (TPβ) (Raychowdhury et al., 1994) that differ in their mechanisms and kinetics of desensitization and internalization (Yukawa et al., 1997;Parent et al., 1999).
The TPs are linked via the Gq/G11 class of G proteins to phospholipase C (PLC), which hydrolyzes phosphoinositides to two potent second messengers: inositol 1,4,5-trisphosphate, which leads to an increase in cytoplasmic free calcium, and diacylglycerol (DAG), which activates protein kinase C (PKC) (Brass et al., 1987; Shenker et al., 1991). The TPs have also been shown to couple to G12/G13 (Offermanns et al., 1994;Allan et al., 1996), G16 (van der Vuurst et al., 1997), Gh (Vezza et al., 1999), and perhaps Gs and Gi, although reports conflict regarding the ability of TPs to couple to Gs or Gi (Shenker et al., 1991;Offermanns et al., 1994; Ushikubi et al., 1994; Allan et al., 1996).
The mitogen-activated protein kinase (MAPK) signaling cascade is a common cellular pathway used by many growth factors, hormones, and neurotransmitters. The MAPKs comprise a family of serine/threonine kinases Erk1 and Erk2, the Jun N-terminal kinase/stress-activated protein kinase, and p38 MAPK. Activation of extracellular regulated kinases (ERKs) by receptor tyrosine kinases (RTKs), such as the receptors for epidermal growth factor (EGF), fibroblast growth factor (FGF), platelet-derived growth factor (PDGF), and insulin and insulin-like growth factor 1 (IGF-1), is well recognized (Luttrell et al., 1999). Many GPCRs, including the lysophosphatidic acid (LPA) (Howe and Marshall, 1993), angiotensin II (Duff et al., 1992), α-thrombin (LaMorte et al., 1993), α2A adrenergic (van Biesen et al., 1995), M2 muscarinic acetylcholine, D2 dopamine, and A1 adenosine receptors (Faure et al., 1994), have also been reported to activate ERK. These GPCRs interact with distinct subsets of heterotrimeric G proteins, including the PTX-insensitive Gq/G11 and PTX-sensitive Gi/Go families. In the case of receptors coupled to PTX-sensitive Gi/Go proteins, such as the α2A adrenergic, M2 muscarinic acetylcholine, D2 dopamine, and A1 adenosine receptors, the MAPK pathway is initiated largely by the release of the βγ subunits from Gi proteins (Crespo et al., 1994). The EGFR tyrosine kinase has been identified as an essential link in the GPCR-mediated ERK activation in Rat-1 fibroblasts, HaCaT keratinocytes, primary mouse astrocytes, and COS-7 cells (Daub et al., 1996). In addition, tyrosine kinases of the Src family have been implicated in the mediation of both the tyrosine phosphorylation of EGFR and MAPK activation from both Gq- and Gi-coupled receptors (Della Rocca et al., 1997). Yet the mechanism of TP activating MAPK and how this signaling pathway is regulated are poorly characterized. In this study, we report that in ECV304 cells, TP-induced ERK activation is regulated by PKC via the coupling of TP and Gi proteins and that Src kinase activity and phosphorylation of EGFR are critical components in this signaling pathway.
Materials and Methods
Reagents and Cells.
The TxA2 mimetic [15-(1α,2β(5Z),3α-(1E,3S),4α)]-7-[3-hydroxy-4-(p-iodophenoxy)-1-butenyl-7-oxabicycloheptenoic acid (IBOP) and TP antagonist SQ29548 were purchased from Cayman Chemical (Ann Arbor, MI). Phorbol-12-myristate-13-acetate (PMA), calphostin C (Cal C), 4-amino-5-(4-methylphenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine (PP1), and tyrphostin AG1478 [4-(3-chloroanilino)-6,7-dimethoxyquinazoline] were obtained from Calbiochem (San Diego, CA). Pertussis toxin was purchased from Sigma Chemical Co. (St. Louis, MO). The ECV304 cells (American Type Culture Collection, Manassas, VA), a human bladder cancer cell line, were cultured in M199 medium supplemented with 10% fetal bovine serum and antibiotics (Life Technologies, Gaithersburg, MD) at 37°C in a humidified 5% CO2 atmosphere. Before stimulation, 60 to 80% confluent cells were cultured in serum-free M199 medium for 48 h. Dominant negative (DN) Src was a gift from Dr. Joan S. Brugge (Department of Cell Biology, Harvard Medical School, Boston, MA). The kinase-negative c-Src K295R was inserted in a mammalian expression vector pCMV-5. ECV cells were transfected either with pEGFP-C1 [green fluorescence protein (GFP)] alone or with DN Src plus GFP, using the Lipofectin method (Life Technologies). One day after transfection, cells were sorted by GFP fluorescence using a fluorescence-activated cell sorter. Cells with GFP expression were recultured for 1 to 2 days before the preparation of cell lysates.
ERK (MAPK) Assay.
Stimulation of cells was carried out at 37°C in serum-free medium. After stimulation, monolayers were washed once with ice-cold phosphate-buffered saline (PBS), lysed in RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 0.25% sodium deoxycholate, 1% Nonidet P-40, 1 mM EDTA, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, and 1 μg/ml pepstatin), sonicated briefly, clarified by centrifugation, and proteins (30 μg/lane) were resolved by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose membranes. Phosphorylation of Erk1/2 was detected by protein immunoblotting using a 1:1000 dilution of rabbit polyclonal phospho-Erk1/2 antibody (New England Biolabs, Beverly, MA) with horseradish peroxidase-conjugated goat anti-rabbit IgG as a secondary antibody. Immune complexes on nitrocellulose were visualized by enhanced chemoluminescence detection (Amersham Corp., Arlington Heights, IL) and quantified by densitometry using Bio-Rad (Hercules, CA) molecular analysis software. The membrane was stripped with stripping buffer (65 mM Tris, pH 6.8, 2% SDS, and 100 mM β-mercaptoethanol) for 30 min at 50°C and reblotted with monoclonal anti-α-tubulin antibody (Sigma Chemical Co.) for protein loading control.
Phosphorylation of EGFR.
IBOP stimulation was carried out at 37°C in serum-free medium as described in the figure legends. After stimulation, monolayers were washed once with ice-cold PBS, lysed in RIPA buffer, sonicated briefly, clarified by centrifugation, and diluted with RIPA buffer to a protein concentration of 1 mg/ml. EGFR-phosphorylated tyrosine was detected by immunoprecipitating the EGFR using anti-human EGFR monoclonal antibody (Upstate Biotechnology, Lake Placid, NY). The immunoprecipitates were resolved by SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with an anti-phosphotyrosine polyclonal antibody (Upstate Biotechnology). Immune complexes on nitrocellulose were visualized and analyzed as described above. The membrane was stripped with stripping buffer and reblotted with monoclonal anti-EGFR antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA).
Src Kinase Assay.
Src kinase activity was determined using the Src kinase assay kit (Upstate Biotechnology) following the manufacturer's protocol. Briefly, endogenous Src kinase was immunoprecipitated from 1 ml of cell lysate using 4 μg/sample monoclonal anti-Src antibody plus 100 μl of a 50% slurry of Protein G Plus/Protein A agarose (Santa Cruz Biotechnology Inc.) agitated for 2 h at 4°C. Immune complexes were washed with ice-cold buffer A (50 mM Tris-HCl, pH 7.5, 1% Triton X-100, 0.1% β-mercaptoethanol, 1 mM EDTA, 1 mM EGTA, 1 mM sodium orthovanadate, 50 mM sodium fluoride, 5 mM sodium pyrophosphate, 10 mM sodium β-glycerol phosphate, 1 μM microcystin-LA, 1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotonin, 1 μg/ml leupeptin, and 1 μg/ml pepstatin) and reaction buffer (100 mM Tris-HCl, pH 7.2, 125 mM MgCl2, 25 mM MnCl2, 2 mM EGTA, 0.25 mM sodium orthovanadate, and 2 mM dithiothreitol). Phosphorylation of a specific Src kinase substrate peptide (KVEKIGEGTYGVVYK) was performed at 30°C for 10 min in 30 μl of reaction mixture containing 200 μM substrate peptide, 125 μM ATP, and 10 μCi of [γ-32P]ATP. After reaction, 25-μl aliquots of supernatant were added to 20 μl of ice-cold 40% trichloroacetic acid, precipitated for 20 min, and centrifuged. Forty-microliter aliquots of each clarified supernatant were spotted onto P81 paper and washed three times in 0.75% phosphoric acid and once in acetone, then substrate peptide phosphorylation was quantified by scintillation counting.
Monolayers of cells were pretreated in serum-free medium with GDP (100 μM), IBOP (50 nM), PMA (500 nM), or calphostin C (500 nM) as indicated in the figure legends at 37°C for 30 min. After stimulation, cells were washed once with ice-cold PBS, lysed in modified RIPA buffer plus 50 mM sodium acetate, 0.2 mM EGTA, 1.0 mM benzamidine, 2 mM MgCl2, and 100 μM GDP, and kept in the same condition as in pretreatment to maintain the interactive conformation of Gi. Cell lysates were sonicated briefly and clarified by centrifugation. Anti-Gαi antibody (Santa Cruz Biotechnology Inc.) was added to immunoprecipitate TP-Gi complex for 2 h at 4°C, then protein A/G agarose (Santa Cruz Biotechnology Inc.) was added for 1 h at 4°C. Beads were washed four times with ice-cold RIPA buffer, denatured in Laemmli sample buffer, and resolved by SDS-PAGE. Proteins were transferred onto nitrocellulose membranes. Gαi and TPα were detected by immunoblotting using the rabbit polyclonal anti-Gαi antibody (Santa Cruz Biotechnology Inc.) and anti-TPα antibody (a gift from Dr. Garret A. FitzGerald, Center for Experimental Therapeutics, University of Pennsylvania, Philadelphia, PA) with horseradish peroxidase-conjugated goat anti-rabbit IgG as secondary antibody. Immune complexes on nitrocellulose were visualized and analyzed as described above.
TP stimulation by the TxA2 mimetics has been shown to activate MAPK in guinea pig coronary artery, rat aortic smooth muscle cells, and rabbit platelets (Morinelli et al., 1994; Jones et al., 1995; Ohkubo et al., 1996). We have found that activation of the endogenous TP by IBOP induces Erk1/2 phosphorylation in a time- and concentration-dependent manner in ECV304 cells. We have determined that ECV304 cells constitutively express both subtypes of TPs with reverse transcription-polymerase chain reaction and protein immunoblotting (data not shown), which is consistent with a previous report (Miggin and Kinsella, 1998). IBOP-induced ERK activation occurred within 2 min, reached its peak in 30 min, and recovered to the basal level in 1 h (Fig. 1A). IBOP induced a maximum 4- to 5-fold increase in phosphorylated ERK level at the concentration of 100 nM (Fig. 1B), which is similar to the level of ERK activation induced by other G protein-coupled receptors (Della Rocca et al., 1997). To search for potential intracellular targets distal to the TP, we measured ERK phosphorylation by TP activation in the presence of several inhibitors. IBOP-induced ERK phosphorylation is diminished almost to the basal level by the EGFR kinase inhibitor AG1478, the Src kinase inhibitor PP1, the Gi protein inhibitor PTX, and the PKC inhibitor calphostin C (Fig. 1C). These results indicate that EGFR, members of the Src kinase family, Gα subunits Gi/o, and PKC are required elements in the signal transduction pathway that leads to ERK phosphorylation following TP stimulation.
Our data show that the EGFR tyrosine kinase inhibitor AG1478 blocked TP-mediated ERK activation, indicating both that the EGFR was required for ERK activity and that the EGFR was activated upon TP stimulation. To test this assumption, we assayed the effects of TP activation on the tyrosine phosphorylation of the EGF receptor. EGFR tyrosine phosphorylation occurred 2 min after IBOP addition to the culture medium of ECV cells. In 30 min, the phosphorylation reached a peak and lasted about 1 h. The effect was blocked by the TP antagonist SQ29548 (Fig. 2A). The IBOP-induced tyrosine phosphorylation of EGFR was blocked by AG1478 and was also blocked by PP1, PTX, and calphostin C (Fig. 2B). Thus, as is the case with a few other GPCRs (Daub et al., 1997), TP stimulation resulted in EGFR phosphorylation leading to MAPK activation.
The prevention of EGFR phosphorylation by these specific inhibitors indicated that TP-induced EGFR activation was mediated by PTX-sensitive Gα subunits, Src kinase, and PKC. To explore further the role of Src kinases in the TP-ERK signaling pathway, we tested whether dominant negative Src blocks TP-induced ERK activation. We transfected ECV cells with either plasmid pEGFP-C1, which expresses GFP as a marker, or DN Src containing plasmid pCMV-5 plus pEGFP-C1. Transfected ECV cells were sorted by GFP fluorescence using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA) and recultured for 1 to 2 days. Expression of DN Src was demonstrated by immunoblotting with anti-Src antibody. Figure 3A shows higher Src expression in lanes 4 and 5 than in lanes l to 3. After IBOP stimulation and treatment with various inhibitors, the Src kinase activity of control cells and DN Src cells was measured. In the control cells, the Src kinase activity was increased about 3- to 4-fold. Activation of Src kinases was inhibited by PP1, PTX, and calphostin C, but not by AG1478 (Fig. 3B). On the other hand, the Src kinase activity of DN Src cells was dramatically attenuated. These data indicate that Src kinases were activated before EGFR but following the PTX-sensitive Gα subunits and PKC after TP activation.
To further confirm the position of Src in this signaling pathway, we stimulated the cells transfected either with GFP alone or with DN Src plus GFP with the TP activator IBOP, the PKC activator PMA, the Gi activator LPA, or the EGFR activator EGF, and tested MAPK activation. The results show that in cells transfected with GFP alone, IBOP, PMA, LPA, and EGF all induced ERK phosphorylation; in cells transfected with DN Src plus GFP, only EGF enhanced ERK phosphorylation (Fig.4). This result further demonstrates that in the pathway by which TP activated MAPK, Src was downstream of TP, PKC, and Gi protein, but was required for activation of EGFR.
Our data indicate that one or more PTX-sensitive G proteins were involved in an early step in the TP-MAPK signaling pathway. Thus, we asked whether a direct stimulation of Gi leads to MAPK activation. When cells were stimulated with LPA, an activator of a Gi-coupled receptor, ERK phosphorylation was dramatically increased, an effect that was inhibited by PTX, PP1, and AG1478, but not by calphostin C (Fig.5). These results suggest that in the TP-MAPK signaling pathway, the position of Gi followed that of PKC.
Since the PKC inhibitor calphostin C attenuated ERK activity following TP stimulation, we tested whether a direct PKC activator would increase ERK activity. When cells were stimulated with IBOP, addition of PMA increased ERK activity by 5- to 6-fold; this PKC-mediated increase in ERK activity was attenuated by calphostin C, PP1, AG1478, and surprisingly, also by the Gi inhibitor PTX, by approximately 70%, as demonstrated by densitometric analysis (Fig.6). It is well known that Gq activation initiates PLC/DAG and PKC activation, which places PKC downstream of Gq. Our data suggest that PKC mediation of ERK activation occurred upstream of Gi, that is, between Gq and Gi. Therefore, we considered the possibility that PKC could regulate the TP-ERK pathway by altering the interaction of TP with Gi.
To test this hypothesis, we examined the association of TP with Gi proteins following ligand stimulation and PKC activation. IBOP and GDP were applied to the immunoprecipitation system to maintain the interactive conformations of TP and Gi, respectively. Our results demonstrate that Gαi was directly coupled with TPα (Fig.7). In the presence of IBOP and GDP, the amount of TPα-Gαi complex was significantly increased. When PMA was added, the interaction of TP and Gi was enhanced by 2- to 3-fold. In contrast, addition of calphostin C decreased the interaction by about 50%. These results suggest that PKC was an important modulator in TP-Gi coupling and thus in the MAPK signaling pathway.
MAPKs play a central role in regulating cell growth and differentiation. The activation of MAPK may result from stimulation of either RTKs, which possess intrinsic tyrosine kinase activity, or GPCRs, which can transactivate RTKs. In most cases, signaling from GPCR to MAPK involves βγ subunits of heterotrimeric G proteins acting on a Ras-dependent pathway; the GPCR signaling pathway converges at the level of Ras with that emerging from RTKs. This process includes the rapid tyrosine phosphorylation of the adapter protein Shc by Src family kinases and the consequent formation of Shc-Grb2 complexes, which recruit Sos, a guanine-nucleotide exchange factor, to the membrane, thereby inducing the exchange of GDP for GTP on Ras, which sequentially activates a classical kinase cascade, from Raf to MEK to MAPK (Gutkind, 1998; Luttrell et al., 1999).
In this study, we demonstrated that a Gq-PKC-Gi-Src-EGFR proximal signaling chain is the major pathway by which TP-induced ERK phosphorylation is mediated. PKC has been linked to activation of the MAPK by plasma membrane receptor stimulation, such as the M1 muscarinic receptor that couples to Gq and Go and activates ERK in a PKC-dependent manner (van Biesen et al., 1996). A proposed model for this sequence is shown in Fig. 8. In this model, stimulation of TP leads to receptor-Gq/Gi coupling. By activating PKC, Gq signaling enhances Gi coupling, which is a mechanism similar to that of the “switching” mechanism proposed for β-adrenergic receptor-Gs/Gi coupling (Daaka et al., 1997). PKC regulates the receptor, most probably by altering the TP conformation or the TP-Gi interaction rather than by directly affecting Gi function. We do not know yet if TP itself or another intermediate phosphorylated by PKC is responsible for this regulation.
Using both pharmacological and molecular approaches we have demonstrated that Src kinase is a critical mediator in the TP-ERK signaling pathway. Previous studies have revealed that Src is activated by many GPCRs in a PTX-sensitive manner (Gutkind, 1998; Luttrell et al., 1999; Ptasznik and Gewirtz, 2000). Src can also directly interact with PKC, either regulating PKC or conversely being affected by PKC activation (Miranti et al., 1999). Our data indicate that by regulating the interaction of TP with G proteins, PKC modulates signal transduction from a PTX-sensitive G protein to Src, thus regulating Src activity. This appears to be the major interaction of Src and PKC within the pathway by which TP activates ERK.
Substantial evidence indicates the important role of Src family nonreceptor tyrosine kinases in GPCR stimulation of MAPK. Some data suggest that Src kinase is required for “downstream” signaling of the transactivated RTKs, as inhibiting Src activity dramatically reduces LPA- and EGF-induced tyrosine phosphorylation of Shc and Gab1, and ERK activation (Luttrell et al., 1996; Daub et al., 1997). Additional data suggest that Src kinase activity may also play an “upstream” role in GPCR-induced RTK transactivation. Inhibition of Src activity by expression of either the Src inhibitor kinase Csk or a catalytically inactive mutant of c-Src attenuates LPA and α2A adrenergic receptor-mediated EGFR phosphorylation in COS-7 cells (Luttrell et al., 1997). Our data indicate that Src-dependent EGFR phosphorylation is critical for TP-induced ERK activation, suggesting that Src activation precedes EGFR transactivation and that TxA2-mediated transactivation of the EGFR is an essential step in the pathway. It cannot be ruled out that an alternative pathway for PKC to activate ERK exists in these cells [for example, one that bypasses EGFR, links to Raf, and activates MAPK (Kolch et al., 1993) or one that enhances Gi interaction with other GPCRs], but our evidence suggests that any such pathway is of less magnitude. We also found that PKC activation increased TP-Gi interaction without TP stimulation. One possible mechanism is that PKC activation could cause transactivation or phosphorylation of TP, which, in either case, could lead to an increase in TP-Gαi association and subsequent ERK activation. Taken together, the evidence favors a model in which Gq-initiated PKC activation, followed by PKC-regulated TP-Gi coupling, mediates the signal from TP to Src and to EGFR. This may represent a unique signaling pathway to MAPK activation following TP stimulation by TxA2.
We thank Dr. Garret A. FitzGerald for TP antibodies, Dr. Joan S. Brugge for Src constructs, and Drs. Ryoji Yokota and Anthony Ashton for helpful discussion.
- Received July 20, 2000.
- Accepted September 29, 2000.
Send reprint requests to: Dr. J. Anthony Ware, Cardiovascular Division, Department of Medicine, Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461. E-mail:
This work was supported by National Institutes of Health Grants HL47032 and HL51043.
- thromboxane A2
- thromboxane A2 receptor
- [15-(1α,2β(5Z),3α-(1E,3S),4α)]-7-[3-hydroxy-4-(p-iodophenoxy)-1-butenyl-7-oxabicycloheptenoic acid
- G protein coupled receptor
- phospholipase C
- protein kinase C
- Cal C
- calphostin C
- lysophosphatidic acid
- mitogen-activated protein kinase
- extracellular regulated kinase
- receptor tyrosine kinase
- epidermal growth factor
- epidermal growth factor receptor
- phosphate-buffered saline
- polyacrylamide gel electrophoresis
- dominant negative
- green fluorescence protein
- The American Society for Pharmacology and Experimental Therapeutics