Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The prostanoid thromboxane A₂ (TXA₂) mediates a number of cellular responses, including platelet aggregation and contraction of vascular smooth muscle (Narumiya et al., 1999). TXA₂ may stimulate mitogenic and/or hypertrophic growth of vascular smooth muscle (Morinelli et al., 1994), and a number of studies have shown that the TXA₂ mimetic U46619 elicits extracellular signal-regulated kinase (ERK) activation in porcine, rat, bovine, and human smooth muscle cells, respectively (Morinelli et al., 1994; Jones et al., 1995; Grosser et al., 1997; Miggin and Kinsella, 2001).

The TXA₂ receptor (TP), a member of the G protein-coupled receptor (GPCR) superfamily, is primarily coupled to Gq-dependent activation of phospholipase (PL) C (Narumiya et al., 1999). Previous studies have shown that down-regulation of diacyl glycerol (DAG)-regulated protein kinase (PK) C isozymes inhibits the ability of the G_q-coupled GPCRs to activate ERK signaling (Bogoyevitch et al., 1994). This effect is mediated through the ubiquitously expressed serine/threonine kinase c-Raf (Hagemann and Rapp, 1999). c-Raf, the most studied member of the Raf family, which also includes A-Raf and B-Raf, interacts directly with GTP-bound Ras, activating the ERK signaling cascade (Hagemann and Rapp, 1999). The Raf isoforms may be differentially regulated in response to diverse stimuli (Hagemann and Rapp, 1999); for example, although c-Raf is inhibited by cAMP-dependent PKA, A-Raf is insensitive (Hagemann and Rapp, 1999; Sutor et al., 1999). On the other hand, PKA not only blocks c-Raf activation but also may phosphorylate the low-molecular-mass guanine nucleotide binding protein Rap 1, which in turn recruits and causes sustained activation of B-Raf (Schmitt and Stork, 2000).

Phosphoinositide 3-kinase (PI3K) has been identified as one of the main mediators of growth factor-regulated cell proliferation and cell survival, transmitting antiapoptotic signals. The PI3K class I_A family members consist of a p110, , or  catalytic subunit and a p85 or  adaptor subunit. Growth factor stimulation of cells is mediated in part by interaction of the src-homology (SH)2 domains of the p85 adaptor subunits with tyrosine-phosphorylated proteins, such as receptor tyrosine kinases and growth factor receptor binding protein-2 (Wang et al., 1995; Sugden and Clerk, 1997). The PI3K class I_B family, consisting of a p110 catalytic subunit, is stimulated by G protein subunits and does not interact with SH2 domain-containing adaptors (Lopez-Ilasaca et al., 1998). Both class I_A and I_B PI3Ks interact with GTP-bound Ras, which can act as both an effector and regulator of PI3K (Sugden and Clerk, 1997). Several studies suggest that PKB/Akt activation, by complex formation with the phosphatidyl inositol-3,4,5-trisphosphate lipid products of PI3K, is a key event in the realization of the antiapoptotic effect of PI3K (Datta et al., 1999).

In humans, molecular cloning has identified two receptors for TXA₂, termed TP and TP, which arise by differential splicing and which differ exclusively in their carboxyl terminal (C) tails (Kinsella, 2001). Although both TP and TP exhibit identical G protein-dependent coupling to PLC, the main effector of TP receptor activation, they display differential regulation of their secondary effector adenylyl cyclase and the novel, high-molecular-weight G protein Gh (Walsh et al., 1998; Kinsella, 2001); they exhibit different patterns of expression (Miggin and Kinsella, 1998); and they are subject to differential homologous desensitization (Kinsella, 2001) and heterologous desensitization by a number of other prostanoids, including prostaglandin (PG)I₂, PGE₂, and PGD₂ (Walsh et al., 2000; Walsh and Kinsella, 2000; Kinsella, 2001). In view of these findings highlighting critical differences between the TP isoforms, the aim of the current study was to define the key regulatory elements involved in TXA₂-mediated activation of the ERK signaling cascade and to investigate whether the individual TP receptors may regulate this essential mitogenic cascade. These studies provide the first in-depth study defining the mechanisms of TXA₂-mediated ERK activation through the human TP receptors.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials

PD 98059, GF 109203X, Tyrphostin AG1478, PP2, and H-89 were purchased from Calbiochem-Novabiochem (Nottingham, UK). 5-Heptenoic acid, 7-[6-93-hydroxy-1-octenyl-2-oxabicyclo[2.2.1]hept-5-yl]-[1R-[1,4,5(z), 6(1E,3S*)]-9,11-dideoxy-9,11-methano epoxy prostaglandin F₂ (U46619) and [1S-[1,2-(5Z)-3,4-]]-7-[3-[[2-(phenylamino)carbonyl]hydrazine] methyl]-7-oxabicyclo[-2,2,1-] hept-2yl]-5-heptenoic acid (SQ29,548) were purchased from Cayman Chemical (Ann Arbor, MI). [³²P]ATP (6000 Ci/mmol; 10 mCi/ml) was purchased from PerkinElmer Life Sciences (Boston, MA). Anti-ACTIVE mitogen-activated protein kinase rabbit polyclonal antibody was purchased from Promega (Madison, WI). Chemiluminescence blotting substrate (peroxidase), rat monoclonal 3F10 anti-hemagglutinin (HA)-peroxidase, and -galactosidase staining kit (#1828-673) were purchased from Roche Applied Science (Sussex, UK). Affinity-purified rabbit polyclonal anti-ERK1 (691); anti-Ha-Ras 259 rat monoclonal antibody (Sc 35); anti-GRK2/-ARK1 (C15, Sc-562); anti-Rap1B (C-17, Sc 1481); anti-G1(C-16, Sc 379); anti-G2 (A-16, Sc 374); anti-PI3K (H-199; Sc 7177); and horseradish peroxidase-conjugated anti-rabbit, -mouse, or -rat IgG were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Aprotinin, benzamidine, leupeptin, myelin basic protein (MBP), phorbol-12-myristate-13-acetate (PMA), protein A-Sepharose CL-4B, protein G-agarose, phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002), epidermal growth factor, and wortmannin were purchased from Sigma Chemical (St. Louis, MO). Anti-HA 101r was purchased from BABCO (Berkeley, CA). I-Block was purchased from Tropix (Bedford, MA). Anti-phosphorylation-specific (pS⁴⁷³) PKB was purchased from BioSource International (Camarillo, CA). [³H]SQ29,548 (50.4 Ci/mmol) was obtained from PerkinElmer Life Sciences. All other chemicals and reagents were of AnalaR grade or molecular biology grade and were used without further purification.

Methods

Plasmids. The plasmid pCMV5:Ha-ras has been described previously (Kinsella et al., 1991). The plasmid pCMV5^:Ha-ras^N17 encoding the dominant negative (DN) Ha-Ras^N17, whereby Ser¹⁷ of Ha-Ras was mutated to Asn¹⁷, was constructed by site-directed mutagenesis with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). The DN [or kinase dead (KD)] plasmids pEF-BOS-c-raf^KD-HA² (K375W), pEF-BOS-A-raf^KD-HA² (K336W), pEF-BOSRI:p85-HA² (p85 DN), and the wild-type pEF-BOSRI:p85-HA², described previously (Sutor et al., 1999), were kindly donated by Dr. Larry M. Karnitz (Division of Radiation Oncology, Mayo Clinic, Rochester, MN). The plasmid pcDNA3:PI3KDN (K832R) was a generous gift from Dr. Reinhard Wetzker (Max-Planck Research Unit, Jena, Germany). The plasmid pRK5:ARK1(495-689), as described previously (Koch et al., 1994), was a generous gift from Prof. Robert Lefkowitz (Howard Hughes Medical Institute, Duke University Medical Center, Durham, NC). The plasmid pCMVrap1b^N17, as described previously (Schmitt and Stork, 2000), was a generous gift from Dr. Philip Stork (Vollum Institute, Oregon Health Sciences University, Portland, OR). The full-length rat G₁ and bovine G2 were amplified by reverse transcriptase-polymerase chain reaction followed by subcloning into pcDNA3 (Invitrogen, Carlsbad, CA); the cloned cDNAs were verified by nucleotide sequence analysis and were confirmed to be identical to the previously published sequences for the rat G1 (GenBank accession no. U34958) and bovine G2 (GenBank accession no. M37183) sequences, respectively.

Cell Culture and Transfection. Human embryonic kidney (HEK) 293 cells were obtained from the American Type Culture Collection (Manassas, VA) and were routinely grown in minimal essential medium (MEM), 10% fetal bovine serum (FBS), unless otherwise indicated. The previously described recombinant HEK.TP10 and HEK.TP3 cell lines, stably overexpressing TP and TP, respectively (Walsh et al., 1998, 2000), were routinely grown in MEM, 10% FBS unless otherwise indicated. For transient transfections, HEK.TP10 and HEK.TP3 cells were plated in 10-cm culture dishes 24 h before transfection at a density of 2 × 10⁶ cells/10-cm dish in MEM, 10% FBS. Cells were cotransfected with 25 µg of pCMV5-, pcDNA3-, or pEF-BOS-based vectors, or as negative controls, with 25 µg of the appropriate empty vector, either pCMV5 or pcDNA3, plus 10 µg of pADVA by using the calcium phosphate/DNA coprecipitation technique (Walsh et al., 1998). HEK 293 cells transiently transfected in this way were routinely harvested 48 h post-transfection. However, for the measurement of mitogenic responses, 48 h post-transfection the complete medium (MEM, 10% FBS) was replaced with serum-free medium (MEM, 0% FBS) and the cells were incubated for a further 48 h to induce quiescence. In all cases, transient expression of proteins was confirmed by Western blot analysis at 48 to 96 h post-transfection. Specifically, expression of Ha-Ras and Ha-Ras^N17 was confirmed by immunoblot analysis with anti-Ha-Ras (Sc-35); expression of HA-tagged c-Raf, A-Raf, and p85^DN were confirmed using the anti-HA 3F10 peroxidase conjugate; expression of Rap1b^N17 was confirmed using an anti-Rap1B (Sc 1481); expression of ARK1(495-689) was confirmed using anti-GRK 2 (Sc-562); expression of PI3K was established using anti-PI3K (Sc 7177) antibody; and expression G1 and G2 were confirmed using the anti-G1 (Sc 379) and anti-G2 (Sc 374) antisera, respectively.

Saturation Radioligand Binding. Radioligand binding was performed essentially as described previously (Miggin and Kinsella, 1998). Briefly, HEK 293 cells transiently transfected with pCMV5:TP were harvested by centrifugation at 500g at 4°C for 5 min and then washed three times in phosphate-buffered saline (PBS). Protein concentrations were determined using the Bradford assay (Miggin and Kinsella, 1998). Cells were then harvested and resuspended in modified Ca²⁺/Mg²⁺-free Hanks' buffered salt solution, containing 10 mM HEPES, pH 7.67, 0.1% bovine serum albumin at a final concentration of 1 mg/ml. Saturation radioligand binding experiments with the TP antagonist [³H]SQ29,548 (20 nM; 50.4 Ci/mmol) were carried out at 30°C for 30 min in 100-µl reactions by using approximately 100 µg of total cellular protein per assay. Nonspecific binding was determined in the presence of excess nonlabeled 10 µM SQ29,548. Reactions were terminated by the addition of ice-cold 10 mM Tris, pH 7.4, followed by filtration through Whatman GF/C glass filters (Whatman, Maidstone, UK). After washing of the filters three times with 4 ml of 10 mM Tris, pH 7.4, liquid scintillation counting of the filters in 5 ml of scintillation cocktail was performed. Results are expressed as picomoles of [³H]SQ29,548 incorporated per milligram of cell protein ± S.E.M., where n = 3.

-Galactosidase Staining. Briefly, HEK 293 cells were transiently transfected with pHM6:lacZ (Roche Applied Science) by using the calcium phosphate/DNA coprecipitation technique (Walsh et al., 1998). After 48 and 96 h, -galactosidase activity was assessed using the -galactosidase staining kit, essentially as described by the manufacturer (Roche Applied Science). Briefly, cells were washed once with PBS. After removal of the PBS, the cells were fixed with PBS containing 2% formaldehyde (3 ml/10-cm dish) for 15 min at room temperature. Cells were washed three times with PBS followed by analysis of -galactosidase activity with a staining solution (1 part X-gal:19 parts iron buffer). After incubation of the cells for 60 min at 37°C, 5% CO₂, stained and nonstained cells were counted in a number of random fields of the dish. Transfection efficiency was determined as mean percentage (%) of stained cells per total cell population ± S.E.M., where n = 3.

Determination of ERK Activity by Using an in Vitro Kinase Assay. HEK.TP10 and HEK.TP3 cells were seeded at 4 × 10⁵ cells/60-mm dish in MEM, 10% FBS. After 24 h, cells were exposed to MEM, 0.5% FBS to induce quiescence. After a further 48 h, the cells were exposed to test agents for the appropriate times, as indicated in the respective figure legends. ERK activity was determined by monitoring ERK-mediated phosphorylation of its substrate myelin basic protein as described previously (Coso et al., 1995; Miggin and Kinsella, 2001).

Determination of ERK/PKB Activation by Immunoblot Analysis. HEK.TP10 and HEK.TP3 cells were seeded at 1.5 × 10⁶ cells/10-cm dish in MEM, 10% FBS. After 24 h, cells were exposed to MEM, 0% FBS to induce quiescence. After 48 h, the cells were exposed to test compounds, as indicated in the respective figure legends. Cellular lysates were prepared as described previously (Boulton et al., 1991). Aliquots (30 µg) of the resultant supernatant were subjected to 12.5% SDS-polyacrylamide gel electrophoresis followed by electroblotting onto PVDF membranes. Thereafter, membranes were screened by immunoblot analysis with a rabbit anti-ACTIVE ERK antibody (Promega) or a rabbit anti-phosphorylated PKB/Akt antibody (BioSource) as appropriate, to detect dual phosphorylated ERK (ppERK) and phosphorylated S⁴⁷³PKB (ppPKB), respectively, essentially as recommended by the supplier. Where appropriate, membranes were rescreened with an affinity-purified rabbit anti-ERK antibody (Santa Cruz Biotechnology) to detect total ERK protein. Immunocomplexes were visualized using the chemiluminescence detection system, as described by the supplier (Roche Applied Science); signals were quantified by scanning densitometry with a UVP gel documentation system. Results presented are representative of at least three independent experiments. Alternatively, signals from ppERK1 and ppERK2 (combined signals) or ppPKB were quantified by densitometry with a UVP gel documentation system and the combined ppERK1/2 or ppPKB signals are presented as mean fold increase of basal ERK/PKB phosphorylation ± S.E.M., where the levels of basal ERK/PKB phosphorylation in vehicle-treated cells are assigned a value of 1.0. During densitometric analyses, the signals from ppERK1 and ppERK2 were combined because the ratio of ppERK1/ppERK2 was constant. Statistical analyses were performed using the unpaired Students' t test with the Statworks analysis package. p values  0.05 indicate a statistically significant difference.

Coimmunoprecipitation of p85 with TP and TP. HEK.TP10 and HEK.TP3 cells were transiently transfected with pEF-BOSRI:p85-HA² and pADVA by using the Effectene transfection reagent essentially as described by the manufacturer (QIAGEN, Valencia, CA). After 48 h, cells were stimulated with U46619 (100 nM; 10 min) with nonstimulated cells serving as a reference. After harvesting, aliquots (80 µg) of whole cell protein were resolved by SDS-PAGE and electroblotted onto PVDF membranes. For immunoprecipitations, whole cell protein (500 µg) from each transfection was immunoprecipitated using the anti-TP (1/100) and anti-TP (1/100) antisera directed to peptide sequences unique to TP (amino acid residues SLSLQPQLTQRSGLQ;  peptide) and TP (amino acid residues LPFEPPTGKALSRKD;  peptide) C-tail sequences, as described previously (Miggin and Kinsella, 2001). Alternatively, as negative controls, whole cell protein (500 µg) from each transfection was subjected to immunoprecipitation by using the control preimmune TP (1/100) and TP (1/100) sera. Immunoprecipitations were carried out essentially as described previously (Walsh et al., 2000; Miggin and Kinsella, 2001). Briefly, cells were washed once with ice-cold phosphate-buffered saline (3 ml/dish) and were then lysed with 0.6 ml of radioimmune precipitation buffer [50 mM Tris-Cl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40 (v/v), 0.5% sodium deoxycholate (w/v), 0.1% SDS (w/v) containing 10 mM sodium fluoride, 25 mM sodium pyrophosphate, 1 µg/ml leupeptin, 0.5 mM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml antipain, and 1 mM sodium orthovanadate]. After incubation on ice for 15 min, cells were harvested and disrupted by sequentially passing through hypodermic needles of decreasing bore size (gauge 20, 23, and 26), and soluble lysates were harvested by centrifugation at 13,000g at room temperature for 5 min. Coimmunoprecipitation of TP and TP receptors from the cell lysates was performed using either anti-TP (1/100) or anti-TP (1/100) antisera as appropriate. Immunoprecipitates were resolved by SDS-PAGE followed by electroblotting onto PVDF membranes. Thereafter, membranes were screened by immunoblot analysis by using either the anti-HA 3F10 peroxidase conjugate (to detect coimmunoprecipitation of HA-tagged p85 with TP and TP) or by using the anti-HA-101r (BABCO) followed by peroxidase-conjugated anti-mouse IgG (to detect HA-tagged p85 cellular protein). To confirm the specificity of the anti-TP and anti-TP antisera to immunoprecipitate TP and TP, the previously described cell lines HEK.HATP and HEK.HATP were used (Walsh et al., 2000). Both HA-tagged TP and HA-tagged TP were subjected to immunoprecipitation by using the anti-TP antisera (1/100), anti-TP (1/100), or anti-HA 101r (1/300; BABCO) antisera as described previously (Walsh et al., 2000) with HEK 293 cells serving as a control. Immunoprecipitates were resolved by SDS-PAGE followed by electroblotting onto PVDF membranes. Thereafter, membranes were screened by immunoblot analysis by using the anti-HA 3F10 peroxidase conjugate (to detect immunoprecipitation of HA-tagged TP and TP).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

U46619-Mediated ERK Activation in HEK.TP10 and HEK.TP3 Cells. Maximal concentration-dependent activation of ERK1 and ERK2 occurred in HEK.TP10 cells after their exposure to 100 nM U46619 (Fig. 1, A and C). In HEK.TP3 cells, maximal ERK1/2 activation occurred after their exposure to 100 to 300 nM U46619 (Fig. 1, B and C).

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Concentration-dependent effect of U46619 on ERK1 and ERK2 activation in HEK.TP10 and HEK.TP3 cells. HEK.TP10 cells (A) or HEK.TP3 cells (B) were stimulated for 10 min with 0 to 1000 nM U46619. A and B, blots (top) were screened with anti-ACTIVE-ERK to detect the phosphorylated, active forms of ERK (ppERK1/2), whereas blots (bottom) were screened with anti-ERK antibodies to detect ERK1/2 immunoreactive protein. Results are representative of three independent experiments. C, fold increases in ERK (ppERK1/2) phosphorylation in A (HEK.TP10 cells) and B (HEK.TP3 cells), respectively, are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 3), where the levels of basal ERK phosphorylation in vehicle-treated cells are assigned a value of 1.0. *, p < 0.05 and **, p < 0.01 indicate that the levels of U46619-mediated ERK activation (ppERK) were significantly greater compared with basal levels.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Time-dependent effect of U46619 on ERK1 and ERK2 activation in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (C) cells were stimulated for 0 to 60 min with 100 nM U46619. Additionally, HEK. TP10 (A) and HEK.TP3 (C) cells were preincubated with SQ29,548 (1 µM; 1 min) or PD 98059 (10 µM; 30 min) before stimulation for 10 min with 100 nM U46619. A and C, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and C, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and C, bottom). Results are representative of three independent experiments. B and D, HEK.TP10 cells and HEK.TP3 cells, respectively, were stimulated with 100 nM U46619 for 0 to 15 min and U46619-mediated ERK activation was determined using an in vitro kinase assay. Positions of the 32P-labeled MBP are indicated by the arrow. E, fold increases in ERK (ppERK1/2) phosphorylation in A (HEK.TP10 cells) and C (HEK.TP3 cells), respectively, are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 3), where the levels of basal ERK phosphorylation in vehicle-treated cells are assigned a value of 1.0. *, p < 0.05 and **, p < 0.01 indicate that the levels of U46619-mediated ERK activation (ppERK) were significantly greater compared with basal levels. F, fold increases in MBP phosphorylation in B (HEK.TP10 cells) and D (HEK.TP3 cells), respectively, are presented as mean fold increases of basal MBP phosphorylation ± S.E.M. (n = 3), where the levels of basal MBP phosphorylation in vehicle-treated cells are assigned a value of 1.0.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Effect of coexpression of Ha-Ras and Ha-RasN17 on U46619-induced activation of ERK in HEK. TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) were transiently transfected with either pCMV5:Ha-Ras, pCMV5:Ha-RasN17, or as a control, with the vector pCMV5. Cells were stimulated with U46619 for 10 min (+), with cells exposed to vehicle alone () serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. *, p < 0.05 and **, p < 0.01 indicate that the levels of U46619-mediated ppERK activation were significantly enhanced/reduced in the presence of Ha-ras and Ha-rasN17, respectively. D, immunoblot analysis of total cellular protein (100 µg) by using anti-Ha-Ras antibody (259; Santa Cruz Biotechnology) to confirm expression of Ha-Ras (lane 2) and Ha-RasN17 (lane 3) with HEK 293 cells serving as a reference (lane 1).

Effect of GF 109203X and H-89 on TP- and TP-Mediated ERK Activation. Preincubation of HEK.TP10 cells with the PKC inhibitor GF 109203X, at a previously established PKC-selective inhibitory concentration (Martiny-Baron et al., 1993), led to near complete inhibition of U46619-induced ERK1/2 activation (Fig. 4A). In HEK.TP3 cells, GF 109203X only partial inhibited U46619-mediated ERK1/2 activation (Fig. 4B). Specifically, GF 109203X pretreatment reduced U46619-mediated ERK activation in HEK.TP10 cells from a 7.57 ± 0.45- to 2.15 ± 0.42-fold increase over basal levels (Fig. 4C; n = 8). However, in HEK.TP3 cells, U46619-mediated ERK activation was reduced from a 5.25 ± 0.36- to 3.11 ± 0.22-fold increase over basal levels (Fig. 4C; n = 8). Moreover, the effects of GF 109203X on U46619-induced ERK activation in HEK.TP10 and HEK.TP3 cells were concentration-dependent (data not shown). Specifically, IC₅₀ values of 206 and 840 nM GF 109203X were determined in HEK.TP10 and HEK.TP3 cells, respectively. As a control, GF 109203X inhibited PMA-induced ERK activation, indicating that it was used at an effective, inhibitory concentration (Fig. 4, A and B).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Effect of GF 109203X on U46619-induced activation of ERK1 and ERK2 in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) cells were preincubated with GF 109203X (GF; 500 nM; 30 min). Subsequently, U46619 (U4; 100 nM) or PMA (250 nM; 20 min) was added for 10 min, with cells exposed to U46619 alone (100 nM; 10 min), PMA alone (250 nM; 20 min), or vehicle alone (Control) serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of six to eight independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 8), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. **, p  0.01 indicates that the levels of U46619-induced ppERK activation were significantly reduced in the presence of GF 109203X.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Effect of H-89 on U46619-induced activation of ERK1 and ERK2 in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) cells were preincubated for 5 min in the absence or presence of 10 µM H-89 before stimulation for 10 min with 100 nM U46619, 10 ng/ml EGF with cells exposed to vehicle alone serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. **, indicates that the levels of U46619-mediated ppERK activation were significantly reduced in the presence (p  0.01) of H-89.

Role of A-Raf, c-Raf, and Rap1b in TP- and TP-Mediated ERK Activation. Coexpression of the DN forms of either A-Raf (A-Raf^DN) or c-Raf (c-Raf^DN) decreased U46619-mediated ERK activation in both HEK.TP10 and HEK.TP3 cells but did not affect the level of basal ERK activation in either cell type (Fig. 6, A and B). Overexpression of both A-Raf^DN and c-Raf^DN were confirmed by western blot analyses (Fig. 6D). Additionally, HEK.TP10 and HEK.TP3 cells transiently transfected with either A-Raf^DN or c-Raf^DN neither elicited ERK activation in the absence of U46619 (Fig. 6, A and B) nor demonstrated altered levels of TP expression compared with control nontransfected cells (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Effect of coexpression of DN A-Raf and DN c-Raf on U46619-induced ERK activation in HEK. TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) were transiently transfected with either pEF-BOS:A-Raf.HA2 DN (DN A-Raf) or pEF-BOS:c-Raf.HA2 DN (DN c-Raf) or the empty vector (Control). Cells were stimulated with U46619 for 10 min (+), with cells exposed to vehicle alone () serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. *, p  0.05 and **, p  0.01 indicate that the levels of U46619-mediated ppERK activation were significantly reduced in the presence of DN A-Raf and c-Raf. D, immunoblot analysis of total cellular lysate (30 µg) to confirm expression of c-Raf (lane 2) and A-Raf (lane 3) with nontransfected HEK 293 cells serving as a reference (lane 1).

View larger version (37K): [in this window] [in a new window]   Fig. 7.   Effect of coexpression of Rap1bN17 on U46619-induced ERK activation in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) were transiently transfected with pCMV:Rap1bN17 (RapN17) or with pCMV (Control). Cells were stimulated with U46619 for 10 min (+), with cells exposed to vehicle alone () serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. **, p  0.01 indicates that the levels of U46619-mediated ppERK activation were significantly reduced in the presence of Rap1bN17.

Role of Class I_B PI3K in TP- and TP-Mediated ERK Activation. Pretreatment of quiescent HEK.TP10 and HEK.TP3 cells with the PI3K inhibitor wortmannin, at an established PI3K-selective inhibitory concentration (Leopoldt et al., 1998), led to near complete inhibition of U46619-mediated ERK activation (Fig. 8, A and C). Moreover, pretreatment of HEK.TP10 and HEK.TP3 cells with the more selective PI3K inhibitor LY294002, at an established inhibitory concentration (Vlahos et al., 1994), also significantly inhibited U46619-mediated ERK activation compared with cells exposed exclusively to U46619 under similar conditions (Fig. 8, A-C).

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Effect of wortmannin and LY294002 on U46619-induced activation of ERK1 and ERK2 in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) cells were preincubated with either wortmannin (wort; 400 nM; 30 min) or LY294002 (LY; 50 µM; 30 min). Subsequently, the medium was supplemented with U46619 (U4; 100 nM; 10 min), with cells exposed exclusively to U46619 or vehicle alone (Control) serving as references. A and B, blots were screened with anti ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A and B are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. *, p  0.05 and **, p  0.01 indicate that the levels of U46619-mediated ppERK activation were significantly reduced in the presence of wortmannin and LY294002.

View larger version (32K): [in this window] [in a new window]   Fig. 9.   Role of class IB PI3Ks on U46619-induced activation of ERK in HEK.TP10 and HEK.TP3 cells. HEK. TP10 (A) and HEK.TP3 (B) cells were transiently transfected with either pcDNA3:G1 plus pcDNA3:G2 (G), pRK5:ARK1(495-689) minigene (-ARK), pcDNA3:PI3KDN (PI3KDN), or with the vector pcDNA3 (Control). In addition, cells were preincubated with PTx (50 ng/ml; 16 h). Subsequently, cells were stimulated for 10 min with 100 nM U46619 (+), with cells exposed exclusively to U46619 or vehicle alone () serving as a reference. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C and D, fold increases in ERK phosphorylation (ppERK1/2) in A and B, respectively, are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0. The levels of U46619-mediated ppERK activation were not significantly altered in the presence of PTX, G, -ARK, or PI3KDN.

Role of Class I_A PI3K in TP- and TP-Mediated ERK Activation. Because the G-regulated class 1_B subfamily of PI3K does not seem to be involved in TP-mediated ERK activation, the role of PI3K class 1_A was investigated. Coexpression of a dominant negative form of the class 1_A adaptor subunit (p85^DN; Sutor et al., 1999) resulted in a marginal decrease in U46619-mediated ERK activation in both HEK.TP10 and HEK.TP3 cells (Fig. 10, A-C). In addition, overexpression of the wild-type p85 in both HEK.TP10 and HEK.TP3 cells augmented U46619-mediated ERK activation (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 10.   Effect of dominant negative p85 adaptor subunit of PI3K on TP- and TP-mediated ERK activation. HEK. TP10 (A) and HEK.TP3 (B) cells were transiently transfected with pEF-BOSRI:p85. HA2 (p85DN) or with the empty vector (Control). Cells were stimulated with 100 nM U46619 for 10 min (+), with cells exposed to vehicle alone () serving as references. A and B, blots were screened with anti-ACTIVE-ERK to detect ppERK1/2 (A and B, top) or with anti-ERK antibodies to detect ERK1/2 immunoreactive protein (A and B, bottom). Results are representative of four independent experiments. C, fold increases in ERK phosphorylation (ppERK1/2) in A (HEK.TP10) and B (HEK.TP3), respectively, are presented as mean fold increases of basal ERK phosphorylation ± S.E.M. (n = 4), where the levels of basal ERK phosphorylation in vehicle-treated cells () are assigned a value of 1.0.

View larger version (50K): [in this window] [in a new window]   Fig. 11.   Association of the p85 adaptor subunit of PI3K with TP and TP.HEK. TP10 (A) and HEK.TP3 (C) cells were transiently transfected with pEF-BOSRI:p85.HA2 (p85DN). Cells were stimulated with 100 nM U46619 for 10 min before immunoprecipitation of TP and TP with anti-TP (A, lane 1) and anti-TP (C, lane 1) antisera, respectively, or as negative controls, with the preimmune TP (A, lane 2) and TP (C, lane 2) sera, respectively. Thereafter, immunoprecipitates were resolved by SDS-PAGE and immunoblots were screened with anti-HA 3F10 antibody to detect coimmunoprecipitation of HA-tagged p85DN. B and D, HEK:HATP cells (B, lanes 2, 3, and 5) and HEK:HATP cells (D, lanes 2, 3, and 5) or HEK 293 cells (B and D, lanes 1 and 4) were immunoprecipitated with anti-TP (B, lanes 1 and 2), anti-TP (B, lane 3; D, lanes 1 and 3), or anti-HA 101R (B and D, lanes 4 and 5) antisera. Immunoprecipitates were resolved by SDS-PAGE and immunoblots were screened with anti-HA 3F10 antibody. E, immunoblot analysis of total cellular lysate (100 µg) by using anti-HA 3F10 antibody to confirm expression of HA-tagged p85DN (lane 2) with nontransfected cells serving as a reference (lane 1). DN, dominant negative.

View larger version (33K): [in this window] [in a new window]   Fig. 12.   Time-dependent effect of U46619 on PKB activation in HEK.TP10 and HEK.TP3 cells. HEK.TP10 (A) and HEK.TP3 (B) cells were stimulated for 0 to 60 min with 100 nM U46619. Additionally, in A (HEK.TP10 cells) and B (HEK.TP3 cells), cells were preincubated with wortmannin (wort; 400 nM; 30 min) before stimulation of cells with U46619 (100 nM; 5 min). A and B, blots were screened with anti-phosphorylation-specific PKB to detect the phosphorylated, active forms of PKB (ppPKB). Results are representative of three independent experiments. C, fold increases in PKB (ppPKB) phosphorylation in A (HEK.TP10 cells) and B (HEK.TP3 cells), respectively, are presented as mean fold increases of basal PKB phosphorylation ± S.E.M. (n = 3), where the levels of basal PKB phosphorylation in vehicle-treated cells are assigned a value of 1.0. *, p  0.05 and **, p  0.01) indicate that the levels of U46619-mediated PKB activation (ppPKB) were significantly greater compared with basal levels.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

TXA₂ has been implicated as a positive mediator of mitogenic/hypertrophic responses in vascular smooth muscle (Morinelli et al., 1994). Both the TXA₂ mimetics [1S-[1,2(Z), 3(1E,3S*),4]]-7-[3-[3-hydroxy-4-(4-iodophenoxy)-1-butenyl]-7-oxabicyclo[2.2.1]hept-2-yl]-5-heptenoic acid and U46619 activate ERK1/2 in porcine, rat, and bovine aortic SMCs, respectively (Morinelli et al., 1994; Jones et al., 1995; Grosser et al., 1997), although the mechanisms of ERK activation remain poorly defined. Recently, Gao et al. (2001) reported that TP-mediated ERK activation in human endothelial ECV304 cells involved transactivation of the EGF receptor. They proposed that TP-mediated mitogenesis occurred in a PTx-sensitive, G_i/o-src-EGF receptor-dependent mechanism (Gao et al., 2001). In another study in cultured human uterine SMCs, we have established that TP-mediated ERK activation in response to U46619 and the F2 isoprostane 8-epi PGF₂ also involves transactivation of the EGF receptor; however, ERK activation was insensitive to PTx but occurred in a PKA-, PKC-dependent manner in a pathway that also required the participation of PI3K (Miggin and Kinsella, 2001). Thus, it seems that there are differences in TP-mediated ERK activation in human endothelial and in human uterine SMCs, readily distinguishable on the basis of their apparent sensitivity to PTx and the requirement for other signaling elements in human SMCs. Thus, in the current study, we first sought to fully define the mechanisms leading to TXA₂-mediated ERK activation. Second, we sought to establish whether the individual TP and TP receptor isoforms regulate ERK signaling and to define the key elements of those cascade(s).

In the present study, the previously described HEK 293 cell lines stably overexpressing either TP (HEK.TP10) or TP (HEK.TP3) (Walsh et al., 1998, 2000) were used to investigate the mechanisms of TP- and TP-mediated ERK signaling. U46619 elicited concentration- and time-dependent activations of ERK1/2 through both TP and TP. Maximal U46619-induced ERK activations in HEK.TP10 and in HEK.TP3 cells were observed at 10 and 5 min, respectively. The relevance of differential rates of ERK activation through the TP isoforms remains to be determined, but may be related to a requirement for rapid ERK activation through TP in certain tissues that express higher levels of this isoform, for example, fetal vascular smooth muscle (Miggin and Kinsella, 1998).

The role of Ras in GPCR-mediated activation of the ERK signaling cascade is well documented. Indeed, U46619 induced rapid Ras activation platelets (Shock et al., 1997). In this study, overexpression of Ha-Ras augmented U46619-induced ERK activation in both HEK.TP10 and HEK.TP3 cells, whereas overexpression of dominant negative Ha-Ras^N17 partially inhibited U46619-induced ERK activation in both cell lines. Concurring with findings in uterine SMCs and ECV304 cells (Gao et al., 2001; Miggin and Kinsella, 2001), we have also established that both TP- and TP-mediated ERK activation involves transactivation of the EGF receptor, and is src-dependent.

Previous studies have demonstrated that PKC activation was necessary for TXA₂-stimulated hypertrophic growth in vascular smooth muscle (Craven et al., 1996) and that U46619-induced activation of Ras in platelets was PKC-dependent (Shock et al., 1997). Furthermore, inhibition of GF 109203X-sensitive PKCs led to a decrease in U46619-induced ERK activation in cultured human SMCs (Miggin and Kinsella, 2001). Thus, the role of PKC in U46619-mediated activation of ERK through both TP and TP was investigated. In contrast to the strong PKC-dependent activation of ERK in HEK.TP10 cells, U46619-induced ERK activation in HEK.TP3 cells was only partially dependent on GF 109203X-sensitive PKCs. The precise mechanism of the differential PKC dependence of TXA₂-induced ERK activation through either TP or TP remains to be fully investigated; however, it is possible that the TP isoforms may represent differential targets of PKC isoform activation. In keeping with this, we have previously established that the TP and TP isoforms are subject to differential PGE₂-induced desensitization in a mechanism most probably involving direct PKC phosphorylation of the TP isoforms within their unique C-tail domains (Walsh and Kinsella, 2000).

Typically, increased levels of intracellular cAMP were thought to attenuate the ERK signaling cascade (Burgering et al., 1993). Recent evidence now suggests that, in certain situations, elevations of cAMP may actually activate the ERK signaling cascade (Frödin et al., 1994), through both the G_S subunit and the G subunits of the activated G_S (Faure et al., 1994). Interestingly, correlating with findings in human uterine SMCs and in COS-7 cells (Faure et al., 1994; Miggin and Kinsella, 2001), we found that U46619-induced ERK activation was partially dependent on PKA, because H-89 significantly decreased ERK activation through both TP and TP.

Taken together, our studies have shown that both PKC and PKA are involved in ERK activation through the TP isoforms. It is postulated that the positive effects of PKC may be mediated through any one of the PKC-sensitive Raf isozymes, such as c-Raf (Anton and Wennogle, 1998). However, because c-Raf is inhibited by PKA, it is also postulated that other Raf isoforms may be involved in TXA₂-mediated ERK activation. To address this question, the direct role of A-Raf and c-Raf in TP-mediated ERK activation was examined. Both the PKA-insensitive A-Raf and the PKA-sensitive c-Raf were demonstrated to regulate U46619-induced ERK activation through TP and TP. Additionally, overexpression of Rap1b^N17 significantly reduced ERK activation through TP and TP, suggesting that the somewhat novel PKA-modulated Rap1/B-Raf mechanism of ERK regulation (Schmitt and Stork, 2000) may also be involved in TP-mediated ERK activation. The availability of dominant negative forms of B-Raf would add further clarification to this point.

GPCRs have been shown to regulate the class I family of PI3Ks (Vanhaesebroeck et al., 1997), although a mechanism has not yet been clearly defined. In the current study, using the PI3K inhibitors wortmannin and LY294002, we have shown that ERK activation via TP and TP is mediated through PI3K, correlating with findings in human uterine SMCs (Miggin and Kinsella, 2001). Additionally, the dominant negative PI3K^K832P (Lopez-Ilasaca et al., 1998) did not affect U46619-mediated ERK activation in either cell type. Moreover, overexpression of G12 or the G antagonist -ARK1(495-689) minigene (Koch et al., 1994) did not affect ERK activation through TP and TP. Additionally, PTx did not affect ERK activation through TP and TP. These data suggest that PI3K class 1_B is not involved in TP-mediated ERK activation.

In previous studies, Morinelli et al. (1997) demonstrated that exposure of A7r5 cells overexpressing TP to I-BOP led to tyrosine phosphorylation of both the TP itself and the p85 adaptor subunit of class I_A PI3Ks. In view of these findings and our studies indicating a specific role for wortmannin- and LY294002-sensitive PI3Ks, distinct from the class 1_B family, we sought to investigate a possible role of class I_A PI3Ks in TP- and TP-mediated ERK activation. Overexpression of a dominant negative form of PI3K class 1_A adaptor subunit, p85^DN (Sutor et al., 1999), partially inhibited U46619-induced ERK activation, indicating that p85 is involved in TP-mediated ERK activation. The finding that p85^DN only partially inhibited TP-mediated ERK activation can most probably be explained by the high endogenous levels of p85 in HEK 293 cells (Dr. Len Stephens, Babraham Institute, Cambridge, UK; personal communication). Moreover, TP isoform-specific antisera, but not the preimmune sera, permitted immunoprecipitation of both the wild-type p85 (data not shown) and the p85^DN, confirming association of both TP and TP with PI3K class 1_A p85^DN adaptor subunit. Additionally, initial immunoprecipitation of HA-tagged p85 resulted in coimmunoprecipitation of both TP and TP (data not shown). Although the precise mechanism of TP:p85 association remains to be defined, the fact that the p85^DN, devoid of a portion of the intracellular SH2 domain (Sutor et al., 1999), could associate with both TP and TP indicates that other domains, such as the amino (N) or carboxyl (C) terminal SH2 domains of p85, are involved (Wymann and Pirola, 1998). Our studies do not, however, exclude the possibility that other (adaptor) protein(s) may mediate association of TP/TP with p85.

Thus, through these studies, we provide novel evidence suggesting that class 1_A, not class 1_B, PI3K mediates ERK activation through TP and TP. Several studies have indicated that PKB/Akt activation is a key event in the realization of the antiapoptotic effect of PI3K (Datta et al., 1999). Through follow-up studies, we established that TP and TP induced activation of PKB/Akt through wortmannin- and LY294002-sensitive phosphorylation of PKB (S⁴⁷³). Thus, in keeping with its mitogenic actions, it seems that TXA₂ and its mimetics may not only promote mitogenesis through activation of the ERK cascade but also may stimulate antiapoptotic signaling through activation of PKB and its downstream signaling. These findings contrast those of Gao et al. (2000) who demonstrated that I-BOP partially inhibited PKB activation in cultured ECV304 cells. An explanation of these conflicting data will require further investigation but may possibly be attributed to differences in the cell type under study.

In summary, based on the current general understanding of ERK signal transduction cascades, we now propose a model of TXA₂-mediated ERK activation (Fig. 13). The TP agonist U46619 induces ERK activation through both TP and TP in a concentration- and time-dependent manner. TP-dependent ERK activation involves transactivation of the EGF receptor, in a mechanism possibly involving src, but is also dependent on PKA-, PKC-, and PI3K-mediated signaling. In view of the strong dependence of PKA on TP-mediated ERK activation coupled to the fact that EGF signaling may be inhibited by PKA through direct phosphorylation of the EGF receptor itself (Barbier et al., 1999), it is unlikely that all of the downstream effects of U46619-mediated ERK activation are EGF-dependent.

View larger version (30K): [in this window] [in a new window]   Fig. 13.   Proposed model for TP-mediated mitogen-activated protein kinase activation. A, ligand activation of the TXA2 receptor TP and TP isoforms leads to Gq-mediated activation of PLC, leading to increases in intracellular concentrations of IP3 and DAG, and concomitant activation of Ca2+-sensitive, DAG-regulated PKC. A key target of PKC is its positive regulation of c-Raf and, to a lesser extent, A-Raf leading to mitogen-activated protein kinase kinase 1 and 2 phosphorylation and activation of ERK1 and -2. B, in addition, ligand activation of TP and TP may lead to Gs-dependent activation of adenylyl cyclase (AC) resulting in increases in intracellular concentrations of cAMP and, in turn, activation of H-89-sensitive PKA. PKA may, in turn, phosphorylate c-Raf to inhibit its actions but may also phosphorylate Rap1b, which, in turn, sequesters B-Raf leading to sustained activation of B-Raf, leading to ERK activation. C, ligand activation of TP and TP may also signal through activation of wortmannin and LY294002-sensitive PI3K class 1A members, through unknown mechanisms but which may involve direct interaction of TP and TP with the p85 adaptor subunit, leading to Ras signaling and ERK activation (a) and leading to PKB activation (b). D, ligand activation of TP and TP may also signal through transactivation of AG1478-sensitive epidermal growth factor receptor (EGF-R) through an unknown mechanisms involving PP2-sensitive src, leading to activation of ERK1 and -2. Although the TP antagonist SQ29,548 completely inhibits TP-mediated mitogenesis, PTx, G subunits, and ARK1(495-689) minigene are without an effect on this pathway. Solid arrows and/or (+) indicate positive effects; broken lines and/or () indicate inhibitory effects. ? indicates whether PI3K acts as upstream or downstream of ras.

Although TP-mediated ERK activation is dependent on DAG-activated PKCs, TP-mediated ERK activation is only partially dependent on DAG-activated PKCs. Ras, PI3K class 1_A, but not class 1_B, and the Raf isoforms c-Raf, A-Raf, and PKA-regulated Rap1/B-Raf transduce U46619-mediated ERK activation through both TP and TP. Additionally, PKB/Akt was shown to be activated in a PI3K-dependent manner. In our proposed model (Fig. 13), our data are in partial agreement with those of Gao et al. (2001) in that TP-mediated ERK activation involves transactivation of the EGF receptor; however, our study expands this mechanism and indicates the clear involvement of additional PKA-, PKC-, and PI3K-dependent, PTx-independent mechanisms not identified by those studies in ECV304 cells (Gao et al., 2001). Thus, it is clear that the regulation of the ERK signaling cascades through TP and TP is multifaceted with multiple intermediates participating in this highly regulated signaling network, and studies presented herein have identified the key components involved in the regulation of TXA₂-mediated ERK activation.