Ras-dependent Mitogen-activated Protein Kinase Activation by G Protein-coupled Receptors

Many receptors that couple to heterotrimeric guanine-nucleotide binding proteins (G proteins) have been shown to mediate rapid activation of the mitogen-activated protein kinases Erk1 and Erk2. In different cell types, the signaling pathways employed appear to be a function of the available repertoire of receptors, G proteins, and effectors. In HEK-293 cells, stimulation of either α1B- or α2A-adrenergic receptors (ARs) leads to rapid 5–10-fold increases in Erk1/2 phosphorylation. Phosphorylation of Erk1/2 in response to stimulation of the α2A-AR is effectively attenuated by pretreatment with pertussis toxin or by coexpression of a Gβγ subunit complex sequestrant peptide (βARK1ct) and dominant-negative mutants of Ras (N17-Ras), mSOS1 (SOS-Pro), and Raf (ΔN-Raf). Erk1/2 phosphorylation in response to α1B-AR stimulation is also attenuated by coexpression of N17-Ras, SOS-Pro, or ΔN-Raf, but not by coexpression of βARK1ct or by pretreatment with pertussis toxin. The α1B- and α2A-AR signals are both blocked by phospholipase C inhibition, intracellular Ca2+chelation, and inhibitors of protein-tyrosine kinases. Overexpression of a dominant-negative mutant of c-Src or of the negative regulator of c-Src function, Csk, results in attenuation of the α1B-AR- and α2A-AR-mediated Erk1/2 signals. Chemical inhibitors of calmodulin, but not of PKC, and overexpression of a dominant-negative mutant of the protein-tyrosine kinase Pyk2 also attenuate mitogen-activated protein kinase phosphorylation after both α1B- and α2A-AR stimulation. Erk1/2 activation, then, proceeds via a common Ras-, calcium-, and tyrosine kinase-dependent pathway for both Gi- and Gq/11-coupled receptors. These results indicate that in HEK-293 cells, the Gβγ subunit-mediated α2A-AR- and the Gαq/11-mediated α1B-AR-coupled Erk1/2 activation pathways converge at the level of phospholipase C. These data suggest that calcium-calmodulin plays a central role in the calcium-dependent regulation of tyrosine phosphorylation by G protein-coupled receptors in some systems.

GTP-binding protein (G protein) 1 -coupled receptors (GPCRs) comprise a family of heptahelical membrane-bound receptors that mediate responses to a vast array of ligands (1). While the effects of these receptors on intermediary metabolism have been extensively studied, recent data have suggested that they play important roles in the regulation of cell growth and differentiation. Constitutively activating mutations of the thyrotropin and luteinizing hormone receptors are associated with hyperfunctioning thyroid adenomas and idiopathic male precocious puberty (1,2). Expression of a constitutively active mutant of the ␣1B-adrenergic receptor (AR) in myocardial cells induces myocardial hypertrophy in transgenic animals (3), and ␣1-adrenergic agonists stimulate hypertrophy in cultured neonatal rat ventricular myocytes (4).
Mitogen-activated protein (MAP) kinases represent a point of convergence for cell surface signals regulating cell growth and division. The MAP kinases comprise a family of serine/ threonine kinases, which include the extracellular signal-regulated kinases Erk1 and Erk2, the Jun N-terminal kinase/ stress-activated protein kinase, and p38 mapk (5). MAP kinases are regulated via protein phosphorylation cascades whose basic pattern has been highly conserved throughout evolution. In the mammalian Erk1/2 pathway, the proximal kinases Raf-1 and B-Raf phosphorylate and activate the dual function threonine/ tyrosine kinases MAP/Erk kinases 1 and 2, which in turn phosphorylate Erk1/2. Once phosphorylated, activated Erk1/2 translocate to the cell nucleus, where they phosphorylate and activate nuclear transcription factors (6). Many signals received at the cell surface, including those mediated by growth factor receptor tyrosine kinases (7) and integrins, which mediate cell adhesion (8), initiate the MAP kinase cascade via activation of the low molecular weight GTP-binding protein, p21 ras (9). Association with GTP-bound p21 ras localizes Raf to the plasma membrane, which is sufficient to induce its activation (10).
Significant heterogeneity may also exist between cell types. Activation of Erk1/2 by ␣1-adrenergic receptors in neonatal rat ventricular myocytes (4) and by prostaglandin F 2␣ receptors in NIH-3T3 cells (15) is G␣ q/11 -mediated and p21 ras -dependent, suggesting that G␣ q/11 subunits also activate p21 ras in some cell types. This pathway differs markedly from the G q/11 -coupled receptor-mediated p21 ras -independent MAP kinase activation that has been described in COS-7 cells (13). In this paper, we characterize the mechanisms of Erk1/2 activation employed by the G i -coupled ␣2Aand by the G q/11 -coupled ␣1B-adrenergic receptors, heterologously expressed in HEK-293 cells. We find that both receptors mediate p21 ras -dependent Erk1/2 activation via phospholipase C and calcium-dependent activation of Src family kinases. These data suggest that, in some cell types, the G␤␥ subunit complex-dependent ␣2A-AR and G␣ q/11 subunit-dependent ␣1B-AR signals converge at the level of PLC and proceed via a common, p21 ras -dependent, signaling pathway.
Cell Culture and Transfection-HEK-293, Rat-1, and PC12 cells were from the American Type Culture Collection. HEK-293 cells were maintained in minimum essential medium with Earle's salts (Life Technologies, Inc.) supplemented with 10% fetal bovine serum (Life Technologies) and 100 g/ml gentamicin (Life Technologies), at 37°C in a humidified 5% CO 2 atmosphere. Rat-1 cells were maintained in Dulbecco's modified Eagle's medium (Life Technologies) supplemented with 10% fetal bovine serum and 100 g/ml gentamicin under similar con-ditions. PC12 cells were maintained in RPMI medium 1640 (Life Technologies) supplemented with 10% heat-inactivated horse serum (Life Technologies), 5% fetal bovine serum, 100 g/ml gentamicin, and 20 g/ml L-glutamic acid (Life Technologies) under similar conditions. Transfections of HEK-293 cells were performed on 80 -90% confluent monolayers in six-well dishes. Cells were transfected using the calcium phosphate coprecipitation method as described previously (32). Empty pRK5 vector was added to transfections as needed to keep the total mass of DNA added per well constant within an experiment.
Pyk2 Immunoblotting-Unstimulated PC12 and HEK-293 cell monolayers were lysed directly with 100 l/well Laemmli sample buffer. Cell lysates were sonicated briefly, and approximately 30 g of protein/lane were loaded for resolution via SDS-polyacrylamide gel electrophoresis. Pyk2 was detected by protein immunoblotting using a 1:1000 dilution of rabbit polyclonal anti-Pyk2 IgG with horseradish peroxidase-conjugated goat anti-rabbit IgG (Santa Cruz Biotechnology) as secondary antibody. Chemiluminescent detection of Pyk2 was performed after development of membranes with ECL reagent (Amersham Corp.), according to the manufacturer's instructions and exposure to Biomax XR MAP Kinase Assay and Immunoblotting-Stimulations were carried out at 37°C in serum-starving medium as described in the figure legends. After stimulation, monolayers were lysed directly with 100 l/well Laemmli sample buffer. Cell lysates were sonicated briefly to disrupt DNA, and proteins (30 g/lane) were resolved by SDS-polyacrylamide gel electrophoresis. Phosphorylation of Erk1/2 was detected by protein immunoblotting using a 1:1000 dilution of rabbit polyclonal phospho-specific MAP kinase IgG (New England Biolabs) with alkaline phosphatase-conjugated goat anti-rabbit IgG (Amersham) as secondary antibody. Quantitation of Erk1/2 phosphorylation was performed after development of membranes with Vistra ECF reagent (Amersham) by scanning on a Storm PhosphorImager (Molecular Dynamics). After scanning, membranes were treated for 30 min with 40% methanol to remove the Vistra ECF reagent, stripped by treatment with stripping buffer (62.5 mM Tris-Cl, pH 6.8, 2% SDS, 100 mM ␤-mercaptoethanol) for 30 min at 50°C, and reprobed with rabbit polyclonal anti-Erk2 IgG (Santa Cruz Biotechnology) to quantitate total p42 mapk .

HEK-293 and Rat-1 Cells Exhibit Distinct Patterns of Erk1/2 Activation by Endogenous
GPCRs-LPA receptor-mediated Erk1/2 activation in Rat-1 fibroblasts is mediated by G␤␥ subunits derived from PTX-sensitive G proteins (33) and is independent of changes in intracellular cAMP, calcium, or PKC (23). As shown in Fig. 1A, stimulation of endogenous LPA or thrombin receptors in these cells resulted in a 3-6-fold increase in Erk1/2 phosphorylation, which was completely inhibited by treatment with PTX. Acute activation of PKC by treatment with phorbol ester resulted in a less than 2-fold increase in Erk1/2 phosphorylation. Exposure to the calcium ionophore A23187 resulted in a less than 2-fold increase in Erk1/2 phosphorylation.
HEK-293 cells exhibit a distinct pattern of Erk1/2 activation. As shown in Fig. 1B, Erk1/2 phosphorylation via endogenous LPA and thrombin receptors is mediated via both PTX-sensitive and -insensitive G proteins in these cells. LPA and thrombin induce a similar 8 -10-fold increase in Erk1/2 phosphorylation. Like Rat-1 cells, the LPA signal is PTX-sensitive. In contrast, the thrombin receptor mediates PTX-insensitive Erk1/2 phosphorylation, indicating that a distinct Erk1/2 activation pathway, mediated by PTX-insensitive G proteins, exists in these cells. Also, acute stimulation of HEK-293 cells with phorbol ester or with A23187 resulted in 24-and 15-fold stimulations of Erk1/2 phosphorylation, respectively, in stark contrast to the above results obtained in Rat-1 cells.

Erk1/2 Phosphorylation in HEK-293 Cells Is Mediated by G␣ Subunits for G q/11 -coupled Receptors and by the G␤␥ Subunit
Complex for G i -coupled Receptors-To characterize the mechanisms of Erk1/2 activation via PTX-sensitive and insensitive G proteins in HEK-293 cells, we employed a transiently transfected model system in which G i -coupled ␣2A-AR and G q/11coupled ␣1B-AR were heterologously expressed. As shown in Fig. 2A, stimulation of the ␣2A-AR resulted in PTX-sensitive Erk1/2 phosphorylation, while ␣1B-AR-mediated Erk1/2 phosphorylation in response to stimulation was insensitive to pretreatment with PTX. Since G␤␥ subunits mediate Erk1/2 activation by several GPCRs, we determined whether cellular expression of a G␤␥-sequestrant polypeptide derived from ␤ARK1ct would inhibit ␣1B-ARand ␣2A-AR-mediated MAP kinase activation. As shown in Fig. 2B, expression of the ␤ARK1ct peptide attenuated Erk1/2 phosphorylation in response to ␣2A-AR, but not ␣1B-AR, stimulation. EGF-stimu- with the addition of either 1 g/well plasmid DNA coding for ␤ARK1ct or 1 g/well empty vector, and stimulated as described. C, HEK-293 cells were transiently transfected with plasmid DNA encoding G␣ q -Q209L (1 g/well), G␣ i2 -Q204L (1 g/well), empty vector (1 g/well), or both G␤ 1 (0.5 g/well) and G␥ 2 (0.5 g/well) as described. Erk1/2 phosphorylation in these cells was determined after 24 h of serum starvation. Data are expressed as -fold Erk1/2 phosphorylation, in which the Erk1/2 phosphorylation produced in unstimulated, empty vector-transfected cells was defined as 1.0. In the absence of transfected ␣1B-AR and ␣2A-AR, phenylephrine and UK-14304 resulted in 2.3-and 0.9-fold stimulation of Erk1/2 phosphorylation, respectively. Values shown represent means Ϯ S.E. from three separate experiments each performed in duplicate. *, greater than not stimulated (NS) (p Ͻ 0.05, two-tailed T test). lated Erk1/2 phosphorylation was not sensitive to pretreatment of cells with PTX or to overexpression of cDNA coding for the ␤ARK1ct peptide. This suggests that the ␣2A-AR signals primarily via the G␤␥ subunit complex from PTX-sensitive G proteins, whereas the ␣1B-AR signal is mediated by the G␣ subunit from PTX-insensitive G proteins. As shown in Fig. 2C, overexpression of a constitutively active mutant of G␣ q (G␣ q -Q209L), but not of G␣ i2 (G␣ i2 -Q204L), was sufficient to induce Erk1/2 phosphorylation. Overexpression of G␤ 1 ␥ 2 resulted in a consistent 2-fold stimulation of Erk1/2 phosphorylation, unlike the 6 -8-fold stimulations of MAP kinase activity observed previously in COS-7 cells (13,34).
Both ␣1B-ARand ␣2A-AR-mediated Erk1/2 Phosphorylation in HEK-293 Cells Is Dependent upon p21 ras Activity-In COS-7 cells, ␣2A-AR stimulation results in Erk1/2 activation by a p21 ras -dependent mechanism, whereas ␣1B-AR-mediated Erk1/2 activation is insensitive to overexpression of a p21 ras dominant-negative mutant and is inhibited by down-regulation of PKC (34). To determine the role of p21 ras in adrenergic receptor-mediated Erk1/2 phosphorylation in HEK-293 cells, cDNA coding for either the ␣1B-AR or ␣2A-AR was coexpressed with cDNA coding for dominant-negative mutant forms of p21 ras (N17-Ras), mSOS1 (SOS-Pro), or p74 raf-1 (⌬N-Raf). As shown in Fig. 3, phosphorylation of Erk1/2 in response to stimulation of both the ␣1B-AR and the ␣2A-AR was attenuated in cells coexpressing N17-Ras, SOS-Pro, or ⌬N-Raf. Acute stimulation with phorbol esters was attenuated by overexpression of ⌬N-Raf, but not by overexpression of N17-Ras or SOS-Pro, indicating that PKC-mediated Erk1/2 activation is Ras-independent. As expected, EGF-stimulated Erk1/2 phosphorylation was sensitive to the effects of overexpressed N17-Ras, SOS-Pro, and ⌬N-Raf. Phosphorylation of Erk1/2 as a result of G␣ q -Q209L expression was similarly attenuated in cells coexpressing N17-Ras (data not shown). These data suggest that in HEK-293 cells, stimulation of both G q/11 -and G i -coupled receptors leads to Erk1/2 phosphorylation in a manner that is dependent upon mSOS, p21 ras , and p74 raf-1 activation.
Erk1/2 Phosphorylation Mediated by ␣1B-AR and ␣2A-AR in HEK-293 Cells Is Phospholipase C-and Calcium-dependent-Pertussis toxin-sensitive G␤␥ subunit-mediated activation of p21 ras in COS-7 cells is sensitive to inhibitors of tyrosine kinases and requires recruitment of the Ras guanine-nucleotide exchange factor, mSOS (19). The G␤␥ subunit effectors responsible for these signals are unknown. Since PLC␤ isoforms are regulated by both G␣ q/11 and G␤␥ subunits and PLC␤ overexpression results in Erk1/2 activation in COS-7 cells (34), we tested whether PLC activation was required for ␣1B-ARand ␣2A-AR-mediated Erk1/2 phosphorylation in HEK-293 cells. As shown in Fig. 4, pretreatment of HEK-293 cells with the PLC inhibitor, U73122, markedly attenuated both ␣1B-ARand ␣2A-AR-mediated Erk1/2 phosphorylation. Phorbol estermediated Erk1/2 phosphorylation was insensitive to the effects of U73122. The results suggest that one or more isoforms of PLC are required for both G q/11 -and G i -coupled receptor-mediated MAP kinase activation in HEK-293 cells. Recent reports have suggested that Ras-dependent Erk1/2 activation in vascular smooth muscle cells (35) and in neuronal cells (36) may be calcium-dependent. As shown in Fig. 5, treatment of HEK-293 cells with the calcium ionophore A23187 resulted in 5-10-fold increases in Erk1/2 phosphorylation, similar to that observed after 5-min stimulation of cells expressing transfected ␣1Band ␣2A-AR. Pretreatment of HEK-293 cells with the cell membrane-permeable Ca 2ϩ chelating agent BAPTA abrogated ␣1B-ARand ␣2A-AR-mediated, as well as A23187-induced, Erk1/2 phosphorylation. Erk1/2 phosphorylation after stimulation of BAPTA-pretreated cells with EGF was unaffected. These data suggest that increased intracellular Ca 2ϩ concentration, resulting from G␤␥-or G␣-mediated PLC activation, is required for Erk1/2 activation in HEK-293 cells.
␣1B-AR-, ␣2A-AR-, and Calcium Ionophore-stimulated Erk1/2 Phosphorylation Requires Tyrosine Kinase Activity-Tyrosine phosphorylation of the Shc adaptor protein, which supports the SH2 domain-mediated recruitment of Grb2-Sos to the plasma membrane, has been implicated in both receptortyrosine kinase-and GPCR-mediated Erk1/2 activation in some cell types (37). To test whether the calcium-dependent ␣1Band ␣2A-AR signals in HEK-293 cells are also dependent upon tyrosine protein phosphorylation, we determined the effects of two tyrosine kinase inhibitors, genistein and herbimycin A, on ␣1Band ␣2A-mediated Erk1/2 phosphorylation in HEK-293 cells. As shown in Fig. 6A, pretreatment of HEK-293 cells with the tyrosine kinase inhibitors markedly attenuated ␣1B-AR-, ␣2A-AR-, and EGF-R-mediated Erk1/2 phosphorylation. Erk1/2 phosphorylation induced by the calcium ionophore A23187 was also tyrosine kinase inhibitor-sensitive, suggesting that elevation of intracellular Ca 2ϩ levels is sufficient to induce tyrosine phosphorylation in these cells.
Both pertussis toxin-sensitive (38) and -insensitive (39) activation of Src family nonreceptor tyrosine kinases have been described by several laboratories, and Src activity appears to be required for G␤␥ subunit-mediated Erk1/2 activation in COS-7 cells (40). To determine whether Src family tyrosine kinases are involved in GPCR-mediated Erk1/2 activation in HEK-293 cells, we measured ␣1B-ARand ␣2A-AR-mediated Erk1/2 phosphorylation in cells coexpressing either a catalytically inactive mutant of p60 c-src , Src-K298M, or a negative regulatory protein of p60 c-src , p50 csk , which phosphorylates and inactivates p60 c-src (41). As shown in Fig. 6B, overexpression of either a constitutively activated mutant form of p60 c-src (Src-Y530F) or of wild-type p60 c-src in HEK-293 cells was sufficient to induce Erk1/2 phosphorylation. As shown in Fig. 6C, overexpression of either Src-K298M or p50 csk significantly attenuated Erk1/2 phosphorylation induced by either ␣1B-AR and ␣2A-AR stimulation or treatment with calcium ionophore. Erk1/2 phosphorylation induced by acute stimulation with phorbol ester or by overexpression of wild-type c-Src was unaffected. The inability of overexpressed p50 csk to significantly inhibit Erk1/2 phosphorylation mediated by overexpressed wild-type c-Src probably reflects ineffective competition between the two overexpressed proteins. These data suggest that both ␣1Band ␣2A-AR signals in HEK-293 cells are calcium-dependent and mediated by Src family tyrosine kinase activity.
Calcium-dependent Erk1/2 Phosphorylation in HEK-293 Cells Is Sensitive to Inhibitors of both Calcium-calmodulin and the Calcium-regulated Tyrosine Kinase, Pyk2, but Not Inhibitors of PKC-Acute stimulation of PKC with phorbol ester is sufficient to induce Erk1/2 phosphorylation in HEK-293 cells. Unlike the ␣1B-ARand ␣2A-AR-mediated signals, the acute PMA signal is insensitive to the effects of N17-Ras, SOS-Pro, overexpressed p50 csk , Src-K298M, and tyrosine kinase inhibitors. As shown in Fig. 7, the PKC inhibitor GFX, which abolished acute PMA-stimulated Erk1/2 phosphorylation, had no effect on Erk1/2 phosphorylation after stimulation of cells with adrenergic agonists, EGF, or calcium ionophore. Similar results were obtained through down-regulation of endogenous PKC expression after chronic treatment of cells with phorbol ester. These data suggest that PKC activation mediates Erk1/2 phosphorylation via a pathway that is distinct from the calcium-and tyrosine kinase-dependent pathway employed by ␣1Band ␣2A-ARs.
Calcium-dependent activation of a novel focal adhesion kinase family protein-tyrosine kinase, Pyk2, has been shown to mediate calcium ionophore-, phorbol ester-, and G q/11 -coupled receptor-stimulated Erk1/2 activation in neuronal cells (36) via a direct interaction with c-Src (42). Although Pyk2 is expressed at high levels only in cells of neuronal origin (36), it is possible that this or a related kinase might link calcium flux to tyrosine kinase signaling pathways in other cell types. However, as shown in Fig. 8A, protein immunoblots of HEK-293 cell lysates using anti-Pyk2 antisera detect only low levels of Pyk2 expression. To determine whether Pyk2 is involved in the calciumdependent activation of Erk1/2 in these cells, ␣1B-ARand ␣2A-AR-mediated Erk1/2 phosphorylation was assayed in cells expressing a dominant negative mutant of Pyk2 (PKM; Ref. 42). As shown in Fig. 8B, Erk1/2 phosphorylation in response to adrenergic receptor stimulation or treatment with calcium ionophore was significantly attenuated, with no effect on EGFor phorbol ester-induced signals.
The mechanism whereby calcium influx regulates the activity of the focal adhesion kinase family member Pyk2 is unknown. Neither Ca 2ϩ nor PKC directly activate Pyk2 in vitro (36). Recently, calmodulin inhibitors have been shown to inhibit p21 ras -dependent Erk1/2 activation in cultured rat vascular smooth muscle cells (35). To determine whether calmodulin might play a role in AR-mediated Erk1/2 activation in HEK-293 cells, we determined the effect of three different calmodulin inhibitors on ␣1Band ␣2A-AR-stimulated Erk1/2 phosphoryl-ation. As shown in Fig. 9, pretreatment of HEK-293 cells with fluphenazine, calmidazolium, or ophiobolin resulted in marked attenuation of the phospho-MAP kinase signal, compared with Me 2 SO-pretreated controls. Erk1/2 phosphorylation resulting from stimulation of endogenous EGF receptors was unaffected. These data suggest that calcium-calmodulin may directly or indirectly contribute to the regulation of Pyk2 kinases and the Src-dependent activation of Erk1/2 by GPCRs. DISCUSSION These data suggest a model for ␣1B-ARand ␣2A-AR-mediated Erk1/2 activation that is mediated by calcium-dependent regulation of protein-tyrosine kinases. In HEK-293 cells, as in COS-7 cells (34), the ␣1B-AR-mediated signal is dependent upon the ␣ subunit of a pertussis toxin-insensitive G protein, while the ␣2A-AR-mediated signal is sensitive to PTX treatment and is dependent upon the release of free G␤␥ subunit complexes. Fig. 10 depicts a model of GPCR-mediated Erk1/2 activation in HEK-293 cells that is consistent with our data. G␣ q/11 -and G␤␥-dependent activation of PLC increases cytoplasmic levels of inositol 1,4,5-trisphosphate, resulting in an increase in cytoplasmic calcium concentration. High intracellular concentrations of calcium, perhaps through calmodulin, lead to activation of Pyk2 or a closely related tyrosine kinase, which regulates the activity of p60 c-src . Src-dependent tyrosine phosphorylation of adaptor proteins, such as Shc, results in recruitment of the Grb2-SOS complex to the plasma membrane, where it catalyzes p21 ras guanine nucleotide exchange. Ras-dependent recruitment of p74 raf-1 kinase to the membrane initiates the phosphorylation cascade leading to activation of Erk1/2. In this system the G␤␥ subunit-and G␣ q/11 subunitmediated pathways each require the PLC␤-dependent stimulation of calcium influx. This early convergence is distinct from findings in COS-7 and CHO cells (34) and more closely resembles the calcium-and Ras-dependent activation of Erk1/2, which has been reported in primary cultures of vascular smooth muscle cells and ventricular myocytes (35,43,44). The observation that dominant interfering mutants of p21 ras , p74 raf-1 , and SOS do not fully attenuate ␣1B-ARand ␣2A-ARmediated Erk1/2 phosphorylation may reflect incomplete inhibition of receptor-mediated p74 raf-1 activation. Alternatively, these data may be indicative of another, p21 ras -independent, mechanism of Erk1/2 phosphorylation, as has been described for G q -and G o -coupled MAP kinase activation in Chinese hamster ovary cells (34,45).
Activation of p60 c-src is required for G i -coupled receptor-mediated, G␤␥ subunit-dependent activation of Erk1/2 in COS-7 cells (37), and G i -and G q/11 -coupled receptor-stimulated Erk1/2 activation in PC12 cells (42). Src family kinase recruitment into Shc-containing protein complexes has been demonstrated following stimulation of formyl-methionyl peptide receptors in human neutrophils (46) and following stimulation of LPA and ␣2-adrenergic receptors in COS-7 cells (37). Our data indicate that Src kinases function as key intermediates in calcium-dependent regulation of Erk1/2, mediated by both G␤␥ and G␣ q/11 subunits in some cell types. Collectively, these findings indicate that regulation of Src family protein-tyrosine kinase activity, potentially via multiple mechanisms, is a common requirement for GPCR-mediated Erk1/2 activation.
In neuronal cells, association of p60 c-src with the calciumregulated focal adhesion kinase family member Pyk2 mediates both Shc phosphorylation and Erk1/2 activation (42). Pyk2 was previously thought to be active only in neuronal cells. The detection of Pyk2 in HEK-293 cell lysates as well as the sensitivity of ␣1B-ARand ␣2A-AR-mediated Erk1/2 activation in HEK-293 cells to both the dominant negative mutant of Pyk2 and specific inhibitors of p60 c-src suggest that a Pyk2-mediated Src-dependent mechanism of p21 ras activation may represent a paradigm for mitogenic signaling in a variety of non-neuronal cell types. The mechanism of Ca 2ϩ -mediated Pyk2 activation remains unclear, however, since calcium does not directly mod-ulate Pyk2 activity (36). Perhaps significantly, both adrenergic receptor-and calcium ionophore-mediated Erk1/2 phosphorylation in HEK-293 cells is sensitive to chemical inhibitors of calmodulin. Eguchi et al. (35) have suggested that calmodulin regulates Erk1/2 activation in cultured rat vascular smooth muscle cells. In NG108 cells, depolarization induces calciumdependent Erk1/2 activation, which is mediated by calmodulindependent kinase IV (47). Our data suggest that the calciummediated regulation of Src family tyrosine kinases proceeds through a calmodulin-dependent mechanism. These data also suggest that, if Pyk2 directly activates p60 c-src in HEK-293 cells, then perhaps calcium/calmodulin is involved in activation of Pyk2, either directly or through a calcium/calmodulin effector protein.
The elucidation of GPCR-mediated mitogenic signaling pathways has revealed significant degrees of heterogeneity between cell types. In Rat-1 fibroblasts, G i -coupled, but not G q/11 -coupled, receptors mediate tyrosine kinase-dependent Erk1/2 activation via a calcium-and PLC-independent mechanism (23). Stimulation of the G i -coupled ␣2A-AR and the G q/11 -coupled ␣1B-AR leads to phospholipase C activation via liberation of G␤␥ and G␣ q/11 -GTP subunits, respectively. Increases in intracellular calcium as a result of phosphoinositide hydrolysis by PLC cause activation of calmodulin (CaM) and the focal adhesion kinase family protein-tyrosine kinase Pyk2. Pyk2 activates c-Src, which results in phosphorylation of the Shc adaptor protein, recruitment of the Grb2-Sos complex to the membrane, and activation of Ras through guanine nucleotide exchange. Subsequent activation of Raf initiates the cascade of phosphorylation events leading to MAP kinase (Erk1/2) activation.
In Chinese hamster ovary cells, G q/11 -coupled receptor stimulation leads to G␣ q/11 -mediated activation of PKC, p74 raf-1 , and Erk1/2 in a tyrosine kinase-and p21 ras -independent manner (34). In PC12 neuroblastoma cells, both G i -and G q/11 -coupled receptors have been shown to activate Erk1/2 via calcium-dependent regulation of p112 pyk2 , p60 c-src , and p21 ras (42). Our data suggest that calcium-dependent regulation of Ras by both G i -and G q/11 -coupled receptors may represent a common mechanism of GPCR-mediated Erk1/2 activation in many non-neuronal cell types. Indeed, G q/11 -coupled receptors mediate calmodulin inhibitor-sensitive, Ras-dependent Erk1/2 activation in cultured vascular smooth muscle cells (35), a calcium-sensitive tyrosine kinase has been cloned from calf uterus (48), and G q/11 -coupled receptor-mediated hypertrophy of cultured rat ventricular myocytes is reportedly Ras-dependent (43). Characterization of receptor-and kinase-specific differences in the mechanisms of GPCR-mediated mitogenic signal transduction may permit the development of strategies for selective antagonism of distinct G protein-coupled receptor-mediated mitogenic signaling pathways, which ultimately may permit selective modulation of cell proliferation in a variety of pathophysiologic states.