Article

• The EMBO Journal (1997) 16, 6439 - 6451
• doi:10.1093/emboj/16.21.6439

Biphasic activation of p21^ras by endothelin-1 sequentially activates the ERK cascade and phosphatidylinositol 3-kinase

Marco Foschi^1,2, Sunita Chari³, Michael J. Dunn¹ and Andrey Sorokin¹

1. Department of Medicine and Cardiovascular Research Center, Medical College of Wisconsin, Milwaukee, WI 53226, USA
2. Istituto di Medicina Interna, Universita' di Firenze, Florence 50134, Italy
3. Department of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA

Correspondence to:

Andrey Sorokin, E-mail: sorokin@post.its.mcw.edu

Received 23 January 1997; Revised 1 July 1997

• Keywords:

  • adaptor proteins,
  • adenovirus-mediated gene transfer,
  • desensitization,
  • MKP-1,
  • Sos phosphorylation

Introduction

Ras proteins are members of the superfamily of small GTPases claimed to play a key role in signaling pathways leading to cell proliferation, differentiation and transformation (Satoh et al., 1992; Boguski and McCormick, 1993). p21^ras activity is regulated by guanosine nucleotides, the active form being bound to GTP and the inactive form being bound to GDP. Activation of p21^ras and the exchange of GTP for GDP, thereby increasing the GTP/GDP ratio, is promoted by guanosine nucleotide exchange factors (GEF) (Boguski and McCormick, 1993). The altered conformation which GTP-bound p21^ras adopts enables it to associate with other signaling proteins such as Raf-1 (Barnard et al., 1995) and phosphatidylinositol 3-kinase (PI-3-K) (Rodriguez-Viciana et al., 1994). The interaction of p21^ras with Raf-1 leads to phosphorylation and activation of downstream protein kinase MAP kinase/ERK kinase (MEK) and extracellular signal-regulated kinase (ERK) (Payne et al., 1991; Alessi et al., 1994), whereas PI-3-K activation leads to phosphorylation of phosphatidylinositides whose role, although not yet completely understood, seems to be important in regulation of cellular growth, differentiation, adhesion and migration (Carpenter and Cantley, 1996).

Most growth factors and insulin are shown to activate p21^ras by recruitment of the GEF Son of Sevenless (Sos) to a complex with adaptor protein Grb2 (Satoh et al., 1992; Buday and Downward, 1993). Grb2 is subsequently able to associate with Shc through its Src homology 2 (SH2) domain, recognizing phosphoTyr317 on Shc, while binding proline-rich domains on Sos through its SH3 domains (Egan et al., 1993). Recently it has been shown that cell stimulation with growth factors and insulin results in serine/threonine phosphorylation of Sos1 and its dissociation from Grb2, leading to p21^ras deactivation (Cherniack et al., 1995). The C-terminal domain of Sos1 contains multiple ERK consensus phosphorylation sites. Moreover, the ability of ERK to phosphorylate Sos1 in vitro and in vivo has been shown (Porfiri and McCormick, 1996). These data suggest Sos1 phosphorylation as a possible negative feedback mechanism in p21^ras/Raf/MEK/ERK signaling. An additional mechanism of regulation of ERK activity is achieved by dual specificity phosphatases which exhibit catalytic activity toward phosphotyrosine and phosphothreonine (Hunter, 1995). MAP kinase phosphatase-1 (MKP-1), also referred to as 3CH134 (Charles et al., 1992), the mouse homolog of human dual specificity phosphatase CL100 (Keyse and Emslie, 1992), shows high affinity for ERK and is responsible for its inactivation in vivo (Sun et al., 1993). Expression of MKP-1 in COS cells (Sun et al., 1993) or rat embryonic cells (Sun et al., 1994) prevents activation of ERK not only by different extracellular stimuli but also by oncogenic v-Ras and constitutively active Raf.

Recent findings suggest that p21^ras activation plays a role in mitogenic signal transduction by G-protein-coupled receptors (Alblas et al., 1993; Chen et al., 1996; Dikic et al., 1996; Sadoshima and Izumo, 1996). Endothelin-1 (ET-1) is a member of a family of 21 amino acid peptides possessing potent vasoconstrictor properties which are also known to modulate gene expression as well as cell growth and differentiation (Simonson and Dunn, 1993; Schramek and Dunn, 1997). ET-1 binds to a specific heterotrimeric G-protein-coupled receptor evoking several intracellular pathways that, in rat glomerular mesangial cells (GMC), include protein tyrosine phosphorylation, activation of different phospholipases, mobilization of Ca²⁺ and activation of protein kinase C (Schrameck and Dunn, 1997). In GMC ET-1 is able to activate transiently both isoforms of ERK, p42^ERK and p44^ERK, with a peak of activation at 2–5 min (Wang et al., 1992), and to induce nuclear signaling through p21^ras (Herman and Simonson, 1995), since expression of a dominant-negative form of Ras inhibits ET-1-stimulated gene induction (Herman and Simonson, 1995). Further, ET-1 has been shown to induce Shc phosphorylation in astrocytes and the association of tyrosine-phosphorylated Shc with the SH2 domain of a Grb2 fusion protein (Cazaubon et al., 1994).

To date, no detailed studies have appeared on the molecular mechanism involved in p21^ras activation and deactivation following ET-1 stimulation. In the present study we demonstrate that ET-1 is able to activate p21^ras in a biphasic manner with the first peak at 2–5 min and a second, longer lasting activation beginning at 30 min. ET-1 induces persistent tyrosine phosphorylation of the adaptor protein Shc, enabling it to associate with the adaptor protein Grb2 and thus the guanosine exchange factor Sos1. The first recruitment of Sos1 to a complex with Shc and Grb2 results in activation of the ERK cascade. Activation of the MEK/ERK pathway correlates with Sos1 phosphorylation, causing its dissociation from the Shc/Grb2 complex. Subsequent to p21^ras and ERK deactivation Sos1 returns to the dephosphorylated condition, enabling it to again associate in complex with Grb2 and Shc, in turn leading to the second activation of p21^ras. However, consequent to the second activation of p21^ras at 30 min a secondary ERK activation does not occur, possibly due to ET-1-induced expression of MKP-1. We also demonstrate that the second activation of p21^ras results in increased p21^ras-associated PI-3-K activity, providing evidence that biphasic activation of p21^ras induced by ET-1 in GMC activates multiple signaling pathways.

Endothelin-1 induces biphasic activation of p21^ras

The activity of p21^ras is regulated through its binding of guanine nucleotides, the GTP-bound active form of the protein being able to interact with effector proteins like Raf-1 and PI-3-K (Rodriguez-Viciana et al., 1994; Barnard et al., 1995). The kinetics of ET-1 stimulated p21^ras activation have, as yet, not been studied. In the present study p21^ras activation was assessed by two different methods: (i) measuring the relative amount of nucleotides bound to p21^ras; (ii) evaluating the quantity of Raf-1 protein able to associate with p21^ras. The relative amount of [³²P]GTP and [³²P]GDP bound to p21^ras was evaluated over a 90 min period following ET-1 (100 nM) stimulation of GMC. Cell lysates were subjected to immunoprecipitation with an anti-p21^ras monoclonal antibody (Y13-259) and guanine nucleotides bound to immunoprecipitated p21^ras were separated by thin layer chromatography (TLC). As shown in Figure 1A, the initial increase in p21^ras-bound GTP, which peaks at 2–5 min, is followed by a recovery to baseline values at 15 min and thereafter by a second and more stable reassociation of GTP with p21^ras, which is maximal at 30 min and then decreases again to the basal level at 90 min post-stimulation. When guanine nucleotides were eluted from immunoprecipitates performed without addition of the specific antibody or following substitution of normal rabbit serum for rabbit anti-rat antibody (refer to Materials and methods for the procedure) only background levels of ³²P activity were found (not shown), indicating that the GTP and GDP detected on TLC were originally complexed with p21^ras. Further, fetal bovine serum (17%) and platelet derived growth factor (50 ng/ml) were included as positive controls for p21^ras activation in several experiments (data not shown). Binding of p21^ras with the Ras binding domain (RBD) of Raf-1 was evaluated over a 2 h time course. Cell lysates of GMC stimulated with 100 nM ET-1 were subjected to affinity precipitation with a glutathione S-transferase–RBD fusion protein (see Materials and methods). The eluted proteins were then subjected to SDS–PAGE and immunoblotted with anti-p21^ras antibody. As is shown in Figure 1B, the GST–RBD of Raf-1 associates in a biphasic manner with p21^ras, with a first peak at 2 and 5 min and a second at 30 min following ET-1 stimulation. Thereafter, the baseline level is gradually attained and recovery is complete at 2 h. These data provide evidence that ET-1 activates p21^ras in a biphasic mode characterized by a first sudden and transient peak followed by a second more stable activation.

Figure 1.

Biphasic activation of p21^ras by ET-1. (A) Quiescent GMC labeled with [³²P]orthophosphate were stimulated with ET-1 (100 nM) for the times indicated. p21^ras was isolated by immunoprecipitation and p21^ras-bound guanine nucleotides were eluted and separated by TLC. Results were quantitated by phosphorimager analysis and expressed as GTP/GTP + GDP100. Data are the mean of five separate experiments (n = 3 for the extended 90 min time course). Bars represent standard deviations. The inset represents Western blot analysis of GMC lysates immunoprecipitated with anti-p21^ras and immunoblotted with the same antibody to confirm that equal amounts of p21^ras protein are immunoprecipitated throughout the time course of ET-1 stimulation. (B) GTP-bound active p21^ras was isolated from lysates of quiescent and ET-1-stimulated GMC by affinity precipitation with a GST–RBD fusion protein followed by immunoblot analysis with anti-p21^ras antibody. The fusion protein (42 kDa) was detected by Coomassie brilliant blue staining of the polyacrylamide gel. Shown is a representative result; the experiment was repeated three times.

View full figure (27 KB)

Endothelin-1 induces a time-dependent activation of ERK

p21^ras interacts with effector proteins through a stretch of amino acids, known as the effector region, that assume an active conformation when the protein is bound to GTP (Barnard et al., 1995). To date, the p21^ras effectors known to mediate predominantly growth stimulatory signals are the serine/threonine Raf kinases (Marshall, 1994). p21^ras is reported to interact, through the effector region, with the N-terminal domain of Raf, regulating its kinase activity (Vojtek et al., 1993; Marshall, 1994). Raf exhibits high substrate specificity toward MEK, phosphorylating its two regulatory serine residues and achieving MEK activation (Alessi et al., 1994). MEK1 and MEK2 are dual specificity kinases capable of phosphorylating regulatory serine/threonine and tyrosine residues on ERK (also referred to as p44 and p42 MAP kinases) (Payne et al., 1991). Thus activated, ERK is itself able to phosphorylate serines and threonines of a number of cytoplasmic and nuclear proteins, such as transcription factors, thereby transducing proliferative and differentiation signals to the nucleus (Seger and Krebs, 1995). In this study we investigated the kinetics of ERK activation following ET-1 stimulation. ERK activation was assessed both by delayed mobility of the phosphorylated form on SDS–PAGE and the ability of immunoprecipitated p42^ERK to phosphorylate myelin basic protein (MBP). As shown in Figure 2, data obtained by both these methods are consistent with an ET-1-induced peak of ERK activation occurring at 5 min. Quantification of ³²P incorporation into MBP revealed a 1.7-fold increase in ET-1-stimulated cells in comparison with their unstimulated counterparts. These experiments confirm the ability of ET-1 to induce ERK activation under our experimental conditions. The monophasic activation of ERK, over a 60 min period, suggests the presence of mechanisms downstream of p21^ras involved in regulation of ERK activity.

Figure 2.

Monophasic activation of ERK by ET-1. Quiescent GMC were stimulated with ET-1 (100 nM) for the times indicated. (A) Immunoblot detection of the two isoforms of ERK. Activation is identifiable by the appearance of bands with delayed mobility related to the phosphorylated form of the protein. (B) ERK activity assayed by the ability of immunoprecipitated p42^ERK to phosphorylate MBP as substrate.

View full figure (35 KB)

Endothelin-1 stimulates tyrosine phosphorylation of Shc and its association with Grb2 and Sos1

It has been shown that Shc adaptor proteins are involved in signal transduction of mitogenic stimuli arising through G-protein-coupled receptors via p21^ras (Cook et al., 1993; Chen et al., 1996; Sadoshima and Izumo, 1996). It has also been shown that ET-1 is able to induce Shc tyrosine phosphorylation in cultured astrocytes (Cazaubon et al., 1994). In the present study we evaluated whether ET-1 could bring about Shc phosphorylation in GMC. Shc immunoprecipitates were immunoblotted with an anti-phosphotyrosine monoclonal antibody. The results of these experiments demonstrate (Figure 3A) that ET-1 is indeed able to phosphorylate Shc, principally the p46 and p52 kDa isoforms, within 1 min and that this phosphorylation is persistent for at least 60 min, with a peak at 15 min. The anti-Shc antibody was able to immunoprecipitate all three isoforms of Shc, as determined by immunoprecipitation and immunoblotting with the same antibody (Figure 3B).

Figure 3.

Shc phosphorylation is persistent during the time period of ET-1-stimulation of rat glomerular mesangial cells. Lysates from quiescent cells stimulated with ET-1 (100 nM) for the times indicated were immunoprecipitated with anti-Shc polyclonal antibody (Transduction Laboratories). Immunoprecipitates were analyzed by SDS–PAGE and immunoblotted with anti-phosphotyrosine (anti-PY) antibody (A) and anti-Shc antibody (B) as indicated. Shown is a representative result; the experiment was repeated three times.

View full figure (59 KB)

Shc phosphorylation creates a binding site for the SH2 domain of the SH2/SH3 domain-containing adaptor protein Grb2 (Salcini et al., 1994). SH3 domains of Grb2 have been shown to interact with the proline-rich motifs located in the C-terminus of the guanine exchange factor Sos1 (Buday and Downward, 1993; Egan et al., 1993; Li et al., 1993), leading to recruitment of Sos1 to the membrane and its ability to activate p21^ras. To examine whether, and with what kinetics, Grb2 associates with Shc in GMC stimulated with ET-1 we performed immunoprecipitations with anti-Shc antibody at different time points post ET-1 stimulation, followed by immunoblotting with an anti-Grb2 antibody. As shown in Figure 4A, there is no evident association between Shc and Grb2 in unstimulated cells and a time-dependent increase in Grb2/Shc association is apparent following addition of ET-1. To confirm further association of Shc with Grb2, Grb2 immunoprecipitates were subjected to SDS–PAGE and blotted with anti-Shc antibodies. As shown in Figure 4B, Grb2 binds rapidly to Shc following ET-1 stimulation of GMC and the association lasts throughout the time course evaluated. Western blot analysis shows that the p52^Shc isoform, the most prevalent in GMC, plays a major role in association of Shc with Grb2. The identity of the band as p52^Shc was confirmed by comparison with Western blots of cell lysates and anti-Shc immunoprecipitates probed with anti-Shc antibody (not shown). Moreover, to further support the ability of Grb2 to associate with the phosphorylated form of p52^Shc, Grb2 immunoprecipitates were subjected to SDS–PAGE and immunoblotted with an anti-phosphotyrosine antibody. Figure 4C shows that after ET-1 stimulation Grb2 associates with the phosphorylated p52^Shc isoform. The onset of association is rapid and lasts throughout the 60 min time course. These data indicate that Grb2 forms a complex with Shc after ET-1 stimulation in GMC and that this association is persistent for at least 60 min.

Figure 4.

ET-1 induces association of Shc with Grb2. Lysates from quiescent cells stimulated with ET-1 (100 nM) for the times indicated were immunoprecipitated with anti-Shc (A) or anti-Grb2 antibodies (B and C). The resultant immunoprecipitates were resolved by SDS–PAGE and then immunoblotted with anti-Grb2 (A), anti-Shc (B) and anti-PY antibodies (C). (B) The anti-Shc antibody specifically recognizes the three Shc isoforms in the lysate. The band between the p52 and p66 isoforms is characterized as non-specific, since it is still recognized when immunoprecipitation is carried out in the absence of antibody (No abs). (C) GMC lysate was subjected to SDS–PAGE analysis and immunoblotted with anti-Shc antibody (far right lane) in order to define the positions of the Shc isoforms in the corresponding anti-phosphotyrosine immunoblot (left lanes).

View full figure (34 KB)

Although Grb2 associates with the phosphorylated form of Shc, in ET-1-stimulated GMC the amount of Grb2 detected in Shc immunoprecipitates does not completely correlate with the amount of phosphorylated Shc. This apparent discrepancy may be explained in a number of ways. Several agonists have been shown to induce Shc phosphorylation not only on Tyr317 but also on Tyr239 and Tyr240 (Gotoh et al., 1996, 1997; van der Geer et al., 1996). Phosphorylation on positions Tyr239 and Tyr240 has been reported not to contribute to p21^ras/ERK activation and to have little involvement in Grb2 association (Gotoh et al., 1997). Therefore, it may be assumed that Shc phosphorylation does not exclusively lead to Grb2/Shc association. Moreover, the SH2 domain and phosphotyrosine binding (PTB) domain of Shc have been shown to associate with a vast array of tyrosine-phosphorylated proteins, some not yet characterized (Kavanaugh et al., 1996). In GMC ET-1 induces rapid and transient tyrosine phosphorylation of a large number of proteins (Force et al., 1991) that may compete with Grb2 for Shc interaction in the early phase but not at later time points of ET-1 stimulation.

In order to examine association of Grb2 with Sos1, Sos1 immunoprecipitates were immunoblotted with anti-Grb2 antibody. Figure 5A shows that Sos1 is able to associate with Grb2 in quiescent cells and this association strongly and rapidly increases following ET-1 stimulation, in a biphasic manner. The first peak of association, observed at 1 and 2 min, is followed by a second and longer lasting association at 30 and 60 min. The identity of the band as Grb2 was confirmed by comparison of cell lysates probed with anti-Grb2 antibody (not shown). To further confirm the bimodal association of Grb2 with Sos1, Grb2 immunoprecipitates were blotted with anti-Sos1 antibody, showing a rapid Grb2/Sos1 association at 1 and 2 min, followed by a new association predominant at 30 min (Figure 5B). Quantitative immunoprecipitation experiments indicate that only a small fraction of cellular Grb2 is in complex with Sos1 (Waters et al., 1996). The relative excess of cytosolic Grb2 could account for the small amount of Sos1 immunoprecipitated with Grb2 under our experimental conditions. These results indicate that ET-1 increases the Grb2/Sos1 complex in a biphasic manner that correlates temporally with p21^ras activation.

Figure 5.

A biphasic increase in Grb2/ Sos1 association induced by ET-1. Serum-starved cells were stimulated with ET-1 for the indicated times. Cell lysates were immunoprecipitated with anti-Sos1 antibody (A) or anti-Grb2 antibody (B). Immunoprecipitates were analyzed by SDS–PAGE and then immunoblotted with anti-Sos1 antibody and anti-Grb2 antibody. Shown is a representative result; the experiment was repeated three times.

View full figure (22 KB)

We next examined whether Sos1 is complexed with Shc, as a result of association of Grb2 with both Sos1 and Shc, after ET-1 stimulation and whether association of Sos1/Shc follows the kinetics of association of Shc with Grb2. Sos1 was immunoprecipitated from quiescent and ET-1-stimulated GMC and immunoblotted with anti-Shc antibody. During ET-1 stimulation a biphasic association of Sos1 with Shc occurs; the first transient association is evident within the first minute and the second phase of association starts at 30 min and lasts at least until 60 min (Figure 6A). Association of Sos1 is mainly with the p52^Shc isoform. In our experiments p46^Shc and p52^Shc were the predominant isoforms phosphorylated on tyrosine, while only the p52^Shc isoform associated with Grb2 during ET-1 stimulation. Differential recruitment of the three isoforms of Shc in the signaling leading to p21^ras activation is consistent with data obtained by other authors using different cell lines (Langlois et al., 1995; Chen et al., 1996; Sadoshima and Izumo, 1996). It is interesting to note that in our experiments Sos1 is able to associate with the p66^Shc isoform, although this protein appears to be poorly phosphorylated and not associated with Grb2 in ET-1-stimulated GMC. The p66^Shc/Sos1 association could, however, be mediated by other SH2/SH3-containing proteins involved in signaling by ET-1 through p21^ras. Shc immunoprecipitates blotted with anti-Sos1 antibody (Figure 6B) confirm the biphasic association of Sos1 with Shc with the same time course displayed by Sos1 immunoprecipitation. These data demonstrate the ability of ET-1 to induce a biphasic Shc/Sos1 association. When comparing the time course of Shc/Grb2/Sos1 association with that of p21^ras activation it is evident that there is a close correlation between these events, suggesting that Sos1 dislocation and association with Shc and Grb2 are responsible for both phases of p21^ras activation.

Figure 6.

Grb2 mediates the association of Sos1 and Shc. Lysates from quiescent cells stimulated with ET-1 (100 nM) for the times indicated were immunoprecipitated with anti-Sos1 antibody (A) or anti-Shc antibody (B). The resultant immunoprecipitates were resolved by SDS–PAGE and then immunoblotted with anti-Sos1 antibody and anti-Shc antibody. Shown is a representative result; the experiment was repeated three times.

View full figure (26 KB)

The kinetics of association of Sos1 with Grb2 and p21^ras desensitization are determined by Sos1 phosphorylation

It has been shown that exposure of cells to growth factors and cytokines results in a reduction in the electrophoretic mobility of Sos1 on SDS–PAGE as a result of protein phosphorylation (Cherniack et al., 1994, 1995; Waters et al., 1995). Phosphorylation of Sos1 occurs on serine/threonine residues located in the C-terminal portion of the protein. Phosphorylation of Sos1 acts as a feedback mechanism, leading to dissociation of Sos1 from the complex with Grb2 and Shc (Cherniack et al., 1994, 1995; Waters et al., 1995). Although ERK has been shown to be able to phosphorylate Sos1 (Porfiri and McCormick, 1996), it was recently reported that an as yet undefined MEK-dependent, ERK-independent kinase might phosphorylate Sos1 in CHO cells stimulated by insulin (Holt et al., 1996). Here we show that during ET-1 stimulation of GMC Sos1 is phosphorylated in a time-dependent manner. Phosphorylation of Sos1, resulting in delayed mobility on SDS–PAGE, is evident at 5 min and reaches a peak at 15 min (Figure 7A). Comparing the time course of Sos1 phosphorylation with that of its association with Grb2/Shc, shown above, strongly suggests that phosphorylation is responsible for the dissociation of Sos1 from the complex with Grb2 and the desensitization of p21^ras we observe in our experiments.

Figure 7.

MEK-dependent mechanism of Sos1 phosphorylation in ET-1-stimulated rat glomerular mesangial cells. Quiescent GMC were left in serum-free medium or exposed to the MEK-inhibitor PD 98059 (100 M) for 1 h prior to being stimulated with ET-1 (100 nM) for the indicated times. Lysates were resolved by SDS–PAGE and then immunoblotted with anti-Sos1 antibody. (A) Kinetics of ET-1-induced phosphorylation of Sos1 seen as delayed mobility of the phosphorylated form on SDS–PAGE. (B) ET-1-induced phosphorylation of Sos1 is compared in GMC pretreated or not with PD 98059. (C) Phosphorylation of Sos1, as delayed motility, in serum-starved NIH 3T3 cells transfected with wild-type (WT) or constitutively active MEK (DN3-S222D). These experiments were repeated three times.

View full figure (31 KB)

To determine whether the MEK/ERK cascade plays a role in Sos1 phosphorylation we used the specific MEK inhibitor PD 98059 in order to block phosphorylation and hence activation of ERK. As shown in Figure 7B, PD 98059 almost completely abolished the Sos1 phosphorylation induced by ET-1. These experiments indicate that the MEK/ERK cascade plays a major role in Sos1 phosphorylation in GMC. Moreover, in order to better elucidate the mechanism of Sos1 phosphorylation, we used NIH 3T3 cells stably transfected with the wild-type or a constitutively active form of MEK (Mansour et al., 1994). In serum-starved constitutively active MEK-expressing cells we detected a delayed mobility of ERK, indicating ERK phosphorylation and activation. In contrast, no shift was detected in serum-starved cells expressing wild-type MEK (data not shown). Lysates from these cell lines were analyzed by SDS–PAGE and blotted with anti-Sos1 antibody. As shown in Figure 7C, we found a delayed mobility of Sos1 in constitutively active MEK-expressing cells, while with wild-type MEK no phosphorylation of Sos1 was observed. This experiment confirms the role of the MEK/ERK cascade in Sos1 phosphorylation. Moreover, it confirms that the reduced phosphorylation of Sos1 observed in GMC preincubated with PD 98059 is wholly due to MEK inhibition and not to non-specific mechanisms. In order to define the role of Sos1 phosphorylation in the mechanism of p21^ras desensitization, we used a recombinant adenovirus to express a constitutively active form of MEK (AdMEK_CA) in GMC. A transgenic adenovirus encoding bacterial -galactosidase activity (AdLacZ) was used to define the multiplicity of infection (MOI) transducing expression in practically 100% of the cells (Figure 8A). Lysates from serum-starved cells transduced with AdMEK_CA and with non-transgenic viruses (Ad-dl327) as a control were resolved by SDS–PAGE and immunoblotted with anti-ERK, anti-MEK and anti-Sos1 antibodies. As a result of strong expression of the catalytically active form of MEK in AdMEK_CA-transduced cells, ERK presented a delay in gel mobility, indicating phosphorylation and activation (Figure 8B). Also, cells expressing the active form of MEK show a pronounced phosphorylation of Sos1, assessed as a delay in gel mobility (Figure 8B). Further, to confirm the role of phosphorylated Sos1 in p21^ras desensitization and the requisite recovery of the non-phosphorylated condition of Sos1 in the second activation of p21^ras, we stimulated AdMEK_CA-transduced and control cells for 30 and 60 min with ET-1 and determined p21^ras activation by affinity binding to the RBD of Raf-1. As shown in Figure 8B and C, in Ad-dl327-infected cells after ET-1 stimulation for 30 and 60 min, as well as in uninfected cells, Sos1 is almost completely dephosphorylated and, consequently, the second peak of p21^ras activation can be detected. In contrast, p21^ras activation is completely abolished and persistent phosphorylation of Sos1 is induced in AdMEK_CA-infected cells. These experiments confirm the role of the MEK/ERK cascade in Sos1 phosphorylation and the requirement for Sos1 phosphorylation in p21^ras desensitization. Further, we obtained evidence that in GMC recovery of the non-phosphorylated status of Sos1 is required for onset of the second phase of p21^ras activation.

Figure 8.

p21^ras desensitization is determined by Sos1 phosphorylation. (A) GMC were infected with adenoviral constructs at a MOI of 60 for 24 h. As assessed by X-gal staining, AdLacZ infects 100% of cells. (B) Lysates from AdMEK_CA- or Ad-dl327-infected cells, stimulated or not for the indicated times with ET-1, were resolved by SDS–PAGE and immunoblotted with anti-Sos1, anti-MEK and anti-ERK antibodies. An ERK shift in AdMEK_CA-infected cells is evidence of ERK activation. (C) Lysates, equalized for protein, from AdMEK_CA- and Ad-dl327-infected cells were subjected to affinity precipitation with GST–RBD fusion protein. p21^ras proteins were then detected by immunoblotting with anti-p21^ras antibody. The fusion protein (42 kDa) was detected in polyacrylamide gels by Comassie brilliant blue staining. This experiment was repeated twice with identical results.

View full figure (63 KB)

Endothelin-1 induces expression of the MAP kinase-specific phosphatase MKP-1

ERK requires phosphorylation on both threonine and tyrosine regulatory residues for activation. Protein phosphorylation is a dynamic process that can be reversed by phosphatases that catalyze dephosphorylation. Recently a new dual specificity phosphatase has been cloned that is able to dephosphorylate tyrosine- and threonine-phosphorylated ERK 15- to 200-fold more rapidly than the other tyrosine-phosphorylated substrate examined (Charles et al., 1993; Sun et al., 1993) and for this reason this phosphatase has been named MKP-1. In order to assess if MKP-1 plays a role in preventing second phase activation of ERK, we evaluated MKP-1 expression during the time course of ET-1 stimulation. MKP-1 was immunoprecipitated from quiescent and ET-1-stimulated GMC and immunoblotted with the same polyclonal antibody. As shown in Figure 9, ET-1 induces expression of MKP-1 30 min post-stimulation, with expression persisting for at least 90 min. These results confirm our previous finding that ET-1 induces the MKP-1 mRNA transcript with a peak at 30 min in GMC (our unpublished data). These experiments suggest that MKP-1 may play an important role in dephosphorylation of ERK and deactivation of the ERK pathway during ET-1 stimulation in GMC. The increased and persistent expression of MKP-1 after 30 min exposure to ET-1 could be one of the factors explaining why late phase activation of p21^ras does not result in a secondary activation of ERK.

Figure 9.

ET-1 induces expression of the dual specificity phosphatase MKP-1. Lysates from quiescent cells stimulated with ET-1 (100 nM) for the times indicated were analyzed by immunoprecipitation and successive immunoblotting with anti-MKP-1 antibody. Shown is a representative result; the experiment was repeated twice.

View full figure (31 KB)

Endothelin-1 increases p21^ras-associated PI-3-K activity

PI-3-K, which is responsible for phosphorylation of the 3'-OH of the inositol ring of phosphatidylinositides, is present in the cell as a heterodimer composed of a 110 kDa subunit, accounting for catalytic activity, and a 85 kDa subunit, with a regulatory role. Although the biological function of PI-3-K has not yet been completely elucidated, it has been implicated in transduction of signaling leading to cell proliferation, differentiation, adhesion and migration by several growth factors (Carpenter and Cantley, 1996). Several regulatory mechanisms involving the p85 subunit of PI-3-K have been proposed to be involved in controlling the activity of this lipid kinase in response to extracellular signals. In H-ras-transformed rat liver epithelial cells PI-3-K activity has been detected in p21^ras immunoprecipitates (Sjolander et al., 1991). Recently p21^ras has been demonstrated to directly interact with the catalytic subunit of PI-3-K. The interaction between p21^ras and PI-3-K is regulated in a GTP-dependent manner, occurring through the effector region of p21^ras (Rodriguez-Viciana et al., 1994). In order to verify if ET-1, secondary to its ability to activate p21^ras, can regulate PI-3-K activity we evaluated PI-3-K activity in p21^ras immunoprecipitates following ET-1 stimulation. In quiescent and ET-1-stimulated GMC p21^ras was immunoprecipitated with monoclonal antibody Y13-259 and subsequently incubated with phosphatidylinositol. Lipids were extracted from the reaction product and resolved by TLC. Following ET-1 stimulation p21^ras-associated PI-3-K activity shows a major increase at 30 min (Figure 10A and B). When immunoprecipitates were performed without addition of specific antibody no activity was found (Figure 10, lane 9), indicating that the PI-3-K activity measured is specifically associated with p21^ras. The kinetics of activation of PI-3-K do not directly follow the kinetics of p21^ras activation following ET-1 stimulation. Even if an increase in PI-3-K activity is present at 5 min, it appears that PI-3-K is predominantly recruited by p21^ras during its second phase of activation (Figure 10A). These experiments demonstrate the ability of ET-1 to induce a p21^ras-associated PI-3-K activity in GMC with a peak at 30 min. The kinetics of PI-3-K activity suggest that mechanisms other than the GTP binding state of p21^ras contribute to regulation of p21^ras-associated PI-3-K activity.

Figure 10.

ET-1 induces PI-3-K activity in anti-p21^ras immunoprecipitates. Lysates from quiescent cells stimulated with ET-1 (100 nM) for the times indicated were immunoprecipitated with either anti-p21^ras antibody (lanes 1–6), anti-p85 antibody (lane 7 and 8) or with neither (lane 9). PI-3-K activity was assayed by its ability to phosphorylate phosphatidylinositol. In lane 7 phosphatidylinositol 4–phosphate was used as substrate (refer to Material and methods). Ori, origin. (A) A quantitative presentation (bars represent standard deviation). (B) A representative result (n = 4).

View full figure (25 KB)

Our data represent the first demonstration of biphasic p21^ras activation; further, we provide evidence that this biphasic model of activation results in sequential activation of two downstream pathways, ERK and PI-3-K.

The proto-oncogene product p21^ras has been shown to act as a pivotal point in transmission of signals from tyrosine kinase receptors to downstream kinase cascades, ultimately leading to cell proliferation, differentiation and gene expression (Satoh et al., 1992; McCormick, 1994). More recently it has been shown that engagement of G-protein-coupled receptors also leads to activation of p21^ras (Alblas et al., 1993; Herman and Simonson, 1995; Sadoshima and Izumo, 1996). Indeed, in yeast which lacked tyrosine kinase receptors, activation of a GTPase analog of p21^ras was promoted by binding of a mating pheromone to a G-protein-coupled receptor. These data indicate that transduction of signals from G-protein-coupled receptors via p21^ras phylogenetically predates even the link between this GTPase and tyrosine kinase receptors (Herskowitz, 1995).

ET-1-induced biphasic activation of p21^ras represents a new model of p21^ras activation that is facilitated by persistent tyrosine phosphorylation of Shc over a 60 min period. Prolonged phosphorylation of Shc induced by several agonists has been described (Chen et al., 1996; Holt et al., 1996) and the crucial role of Shc phosphorylation and its association with Grb2 in ERK pathway activation, secondary to G-protein-coupled receptor engagement, has been extensively documented (Chen et al., 1996). However, the role of prolonged phosphorylation of Shc has not been addressed. In the present study we show that persistent phosphorylation of Shc, stabilizing it in complex with Grb2, results in a permanent docking site for sequential recruitment of Sos1 and hence biphasic activation of p21^ras. Rapid and transient initial activation of p21^ras is followed by an equally rapid and transient deactivation, as a direct consequence of Sos1 phosphorylation and its release from the Shc/Grb2/Sos1 complex (Figure 11). We have also demonstrated ET-1-induced MEK/ERK-dependent phosphorylation of Sos1, leading to its release from the Grb2/Shc complex and the return of p21^ras to the inactive GDP-bound form. Dephosphorylation and deactivation of ERK is followed by recovery of the non-phosphorylated form of Sos1, with its ability to again complex with Shc/Grb2, leading to secondary activation of p21^ras (Figure 11). In fact, transgenic adenovirus-mediated persistent activation of ERK, responsible for prolonged phosphorylation of Sos1, blocks agonist-induced p21^ras activation. Serine/threonine phosphorylation of Sos1 has been shown to play a role in feedback mechanisms leading to p21^ras desensitization in insulin- and epidermal growth factor (EGF)-stimulated cells. Rapid activation of p21^ras with a peak at 1.5–2 min is followed by recovery of the resting condition within 10 min (Klarlund et al., 1995, 1996; Waters et al., 1996). Return of p21^ras to the inactive form correlates with Sos1 phosphorylation and dissociation from the Grb2/Shc complex (Langlois et al., 1995; Klarlund et al., 1996). Recruitment of Sos1 to the membrane seems to be the main requisite for p21^ras activation, since expression in cells of Sos mutants containing membrane anchoring myristoylation or farnesylation signals results in persistent stimulation of p21^ras signaling and consequent oncogenic transformation (Aronheim et al., 1994). Therefore, Sos1 phosphorylation appears to desensitize the p21^ras pathway, resulting in dissociation of Sos1 from Grb2 and its release from the membrane. However, C-terminal deletion generates a considerably more active form of Sos (Aronheim et al., 1994; Wang et al., 1995) and in vitro phosphorylation of Sos1 does not appear to affect its binding to Grb2, although the ability of Sos1 to associate with Shc and with the epidermal growth factor receptor is strongly reduced (Porfiri and McCormick, 1996). These data suggest that Grb2 interacting with the C-terminal region of Sos1 plays a more complex role than merely recruiting it to the membrane. In accordance with these findings, in the present study we also demonstrated an ET-1-dependent increase in Grb2/Sos1 association. Initial results obtained in mammalian cells indicated that Grb2 exists constitutively in complex with Sos1 and that the degree of association does not change upon ligand stimulation (Buday and Downward, 1993; Egan et al., 1993; Li et al., 1993). Recently, however, the ability of ligands to increase association of Grb2 with Sos1 has been reported in several studies using a wide range of agonists (Cherniack et al., 1995; Waters et al., 1995, 1996; Hu and Bowtell, 1996).

Figure 11.

Schematic representation of the proposed signaling through p21^ras induced by ET-1. ET-1 induces a first phase of p21^ras GTP binding which drives ERK pathway activation. ERK phosphorylation of Sos1 leads to p21^ras desensitization. Dephosphorylation and deactivation of ERK, corresponding to increased expression of MKP-1, correlates with the renewed ability of Sos1 to complex with Shc/Grb2, leading to the secondary activation of p21^ras and hence of PI-3-K. The PI-3-K products could act by regulating ribosomal function through the involvement of Akt and p70^S6 kinases.

View full figure (58 KB)

Other mechanisms of desensitization are known to operate at the receptor level in G-protein-mediated signaling determining agonist-induced down-regulation of receptor function and/or expression. Desensitization at the receptor level, which occurs within a few minutes, is mediated by the sequential action of the G-protein-coupled receptor kinases, which phosphorylate agonist-occupied receptors, facilitating binding of a second family of proteins, the arrestins, which block receptor signaling and facilitate receptor internalization (Lefkowitz, 1993). Receptor desensitization has been shown to take place following ET A receptor engagement and has been hypothesized to modulate ET responsiveness (Cyr et al., 1993). Although ET receptor desensitization could down-regulate ET-induced ion transport, protein kinase A and C and phospholipase activation, we show here that ET-1-stimulated p21^ras signaling undergoes an independent mechanism of desensitization.

To date, no other studies on the kinetics of p21^ras activation subsequent to G-protein-coupled receptor engagement have been reported, although persistent phosphorylation of Shc in response to thrombin has been documented (Chen et al., 1996) and transient activation of ERK consequent on G-protein-coupled receptor activation is a common finding. It is possible, therefore, that the biphasic p21^ras activation that we observe in cells stimulated with ET-1 may represent a general event in cellular responses to G-protein-coupled receptor agonists. It is noteworthy that a single activation of p21^ras is apparent in the available kinetic data on tyrosine kinase receptor-mediated signaling. This could be due to the fact that following tyrosine kinase receptor engagement the status of the signaling molecules does not favor the biphasic kinetics of p21^ras activation or, alternatively, that biphasic activation of p21^ras in response to growth factors escaped the attention of investigators because the time course considered was not long enough. While phosphorylation of the receptor is a transient phenomenon, Shc phosphorylation has been shown to be persistent under stimulation with EGF and to reach a maximum at 60 min during insulin stimulation (Waters et al., 1996). Thus, prolonged phosphorylation of Shc persisting after Sos1 dephosphorylation could, in theory, lead to its association with Grb2/Sos1, driving a second activation of p21^ras.

Activation of ERK follows promptly on the first p21^ras stimulation but not after the second phase of p21^ras activity. This lack of late phase ERK activation suggests that other factors involved in regulation of its activity come into play. MKP-1, which selectively dephosphorylates ERK, has been shown to be regulated mainly at the transcriptional level by different MAP kinase pathways (Bokemeyer et al., 1997). We demonstrate that, following 30 min ET-1 stimulation, there is strong and persistent expression of MKP-1. The onset of expression of MKP-1 correlates with deactivation of ERK following the second phase activation of p21^ras activity, suggesting that MKP-1 may play a part in blunting ERK activity (Figure 11). However, further studies will be needed to verify if other mechanisms are involved in this process.

In the present study we show increments in PI-3-K activity in p21^ras immunoprecipitates with a peak at 30 min, corresponding to the second peak of p21^ras activation. Our data support the involvement of p21^ras in regulating PI-3-K activity in ET-1-stimulated cells. The biological significance of this activation remains to be defined and, considering that the functions of the phosphorylated inositide products of PI-3-K activity are largely unknown, the issue is clearly a difficult one. Some lines of evidence indicate that PI-3-K regulates the translation process via other effector molecules, such as Akt kinase and/or p70^S6 kinase, known to regulate ribosomal function (Burgering and Coffer, 1995). In accordance with these data and our findings one could formulate a working hypothesis in which the first activation of p21^ras drives the ERK cascade, inducing gene transcription, whereas the second activation of p21^ras, through the effector PI-3-K, could regulate gene translation (Figure 11).

While in this study we have determined the proteins involved in transduction of the signal from phosphorylation of Shc to p21^ras activation, some steps in ET-1-induced activation of p21^ras remain to be elucidated. For example, the signaling pathways that lead to activation of intracellular tyrosine kinase by ET-1 are still under investigation, as are which kinase(s) phosphorylates Shc on tyrosines. It has been suggested that the subunit of Gi and Gq proteins (G) mediates activation of p21^ras by the G-protein-coupled receptor through recruitment of the Src family kinases (Crespo et al., 1994; van Biesen et al., 1995). Moreover, evidence that a dominant interfering Src kinase mutant blocks ET-1-stimulated c-fos transcription in GMC supports the involvement of Src in ET-1 signaling (Simonson et al., 1996), although the involvement of other tyrosine kinases cannot be ruled out. Further, as shown in stimulated rat fibroblasts overexpressing the human EGF receptor, ET-1 appears to be able to induce EGF receptor phosphorylation, which in turn could be responsible for adaptor protein recruitment (Doub et al., 1996). We were not able to detect ET-1-induced tyrosine phosphorylation of the EGF receptor in primary cultures of GMC (data not shown), possibly due to low-level expression of the EGF receptor itself.

In conclusion, as schematically represented in Figure 11, the ET receptor, like tyrosine kinase receptors, is capable of recruiting the signaling proteins Shc, Grb2 and Sos1, leading to an initial first activation of p21^ras, which in turn results in activation of the downstream ERK pathway. As negative feedback mechanisms, Sos1 is phosphorylated and MKP-1 is expressed, playing determining roles in p21^ras and ERK deactivation respectively. Inactivation of ERK correlates with dephosphorylation of Sos1 and its return to the Shc/Grb2 complex. The consequence is a second delayed activation of p21^ras, resulting in PI-3-K activation. Therefore, ET-1 in GMC results in sequential activation of the ERK and PI-3-K pathways.

Materials

Tissue culture media and reagents were from Life Technologies Inc. (Grand Island, NY). Purified human ET-1 was from Calbiochem-Novabiochem Corp. (La Jolla, CA). ECL reagent was supplied by Amersham (Little Chalfont, UK). PD 98059 was kindly provided by Dr A.Saltiel (Parke-Davis, Ann Arbor, MI). Free carrier ³²P, [-³²P]ATP and [³H]GDP were from Dupont-NEN Research Products (Boston, MA). Polyethyleneimine (PEI)–cellulose and silica TLC sheets were from J.T.Baker Inc. (Phillipsburg, NJ) and Whatman (Clifton, NJ) respectively. The BCA protein assay kit was from Pierce (Rockford, IL). All other reagents were from Sigma Chemical Co. (St Louis, MO).

Cell culture

Primary GMC from male Sprague–Dawley rats were isolated and characterized as previously reported (Simonson and Dunn, 1990). Cells were cultured in RPMI 1640 medium supplemented with 17% fetal bovine serum, 100 U/ml penicillin, 100 g/ml streptomycin, 5 g/ml each insulin and transferrin and 5 ng/ml selenite at 37°C in a 5% CO₂ incubator. NIH 3T3 cells expressing wild-type and catalytically active (DN3-S222N) MEK were grown in Dulbecco's modified Eagle's medium supplemented with 10% calf serum, 100 U/ml penicillin and 100 g/ml streptomycin at 37°C in a 5% CO₂ incubator. As previously described, the catalytically active mutant of MEK was obtained by substitution of Asp for the regulatory Ser222 and by deletion of a predicted -helix encompassing residues 32–51 (Mansour et al., 1994). All experiments were performed with cells in 100 mm Petri dishes made quiescent for 24 h in serum-free medium before stimulation with 100 nM ET-1. Rat hybridoma Y13-259 cells and human embryonic kidney (HEK) 293 cells were purchased from the American Type Culture Collection (Rockville, MD) and cultured according to the supplier's protocols.

Antibodies

Rat monoclonal anti-Ras antibodies which react with both H- and K-Ras were obtained by culturing rat hybridoma Y13-259 cells. Polyclonal anti-Sos1 and polyclonal anti-Grb2 antibodies, both raised against C-terminal peptides of the respective proteins, and rabbit normal IgG, used as non-specific antibody, were from Santa Cruz Biotecnology (Santa Cruz, CA). Rabbit polyclonal anti-Shc antibodies, generated with a GST–Shc fusion protein, and rabbit polyclonal anti-p85 antibodies directed against a GST fusion protein containing amino acids 265–523 of human p85 were kindly provided by Dr J.Schlessinger (New York University Medical Center, New York, NY) (Hu et al., 1992; Pelicci et al., 1992). Monoclonal anti-phosphotyrosine antibodies and a polyclonal anti-Shc antibody (used where indicated) raised against a C-terminal peptide of Shc protein were from Transduction Laboratories (Lexington, KY). Polyclonal anti-p42^ERK/p44^ERK antibodies were raised by immunizing rabbits with synthetic peptides (Wang et al., 1992). Anti-MKP-1 antibody was produced by immunizing rabbits with a synthetic peptide derived from the C-terminus of the protein (Bokemeyer et al., 1997).

Determination of GTP/GDP ratio

Determination of the GTP/GDP ratio from nucleotides immunoprecipitated with p21^ras was essentially as described elsewhere (Downward et al., 1990). Briefly, cells were rendered quiescent for 24 h in serum-deprived medium and subsequently labeled for 4 h with 400 Ci [³²P]orthophosphate/dish in phosphate-free/serum-free medium. Cells were stimulated with ET-1 (100 nM) for the times indicated. Following ET-1 stimulation cells were placed on ice and rapidly washed with ice-cold Tris-buffered saline and lysed in 50 mM HEPES buffer, pH 7.4, 1% Triton X-100, 100 mM NaCl, 5 mM MgCl₂, 1 mg/ml bovine serum albumin (BSA), 10 g/ml leupeptin, 10 g/ml aprotinin and 1 mM phenylmethylsulfonyl fluoride. Nuclei were removed by centrifugation at 15 000 g for 3 min and 0.5 M NaCl, 0.5% deoxycolate and 0.05% SDS added to the lysates. Immunoprecipitation with Y13-259 rat anti-p21^ras monoclonal antibody (300 l hybridoma supernatant) was carried out for 1 h. Rabbit anti-rat or normal rabbit serum, as a control, coupled to protein A–Sepharose, was added to the lysate and incubation carried out for an additional 1 h at 4°C with rotation. Immunoprecipitates were collected and washed eight times with 1 ml 50 mM HEPES buffer, pH 7.4, 0.1% Triton X-100, 500 mM NaCl, 5 mM MgCl₂ and 0.005% SDS. Nucleotides were eluted in 2 mM EDTA, 2 mM dithiothreitol, 0.2% SDS, 0.5 mM GDP and 0.5 mM GTP at 65°C for 20 min and separated on PEI–cellulose resin plates developed in 1 M KH₂PO₄, pH 3.4. The positions of GTP and GDP were monitored by chromatography of standards. The results are expressed as percentage GTP, which is the percentage of the amount of GTP relative to total GTP plus GDP determined by scintillation counting and phosphorimager analysis.

Activated Ras affinity precipitation assay

The expression vector encoding the fusion protein GST–RBD was obtained by ligation of the portion of the raf-1 gene encoding the first 149 amino acids of the protein into the SmaI site of the pGEX 2T vector (Pharmacia Biotech, Piscataway, NJ) (Taylor and Shalloway, 1997). GST–RBD expression was induced in transformed bacteria with 1 mM isopropyl -D-thiogalactoside (IPTG) for 3–4 h, after which time bacteria were harvested and lysed by sonication. The GST–RBD fusion protein was then purified on glutathione–Sepharose beads. Affinity precipitation of activated p21^ras was performed as described elsewhere (Taylor and Shalloway, 1997). Briefly, lysates, equalized for protein, were incubated on a rocker plate at 4°C for 30 min with 50–60 g GST–RBD bound to glutathione–Sepharose beads. The beads were then extensively washed with 20 mM HEPES, pH 7.5, 120 mM NaCl, 10% glycerol, 0.5% Triton X-100, 2 mM EDTA, 10 g/ml leupeptin and 10 g/ml aprotinin. The eluted proteins were resolved on an 11% polyacrylamide gel. Comassie brilliant blue was used to stain the fusion protein in the gel (molecular weight 42 kDa). The polyacrylamide gel was processed as described in the Western blot analysis section and probed with anti-p21^ras antibody.

Immunoprecipitation and Western blot analysis

Stimulation of cells was terminated by washing twice with ice-cold phosphate-buffered saline (PBS). Cell were than lysed with 50 mM HEPES buffer, pH 7.5, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl₂, 1 mM EGTA, 10 g/ml leupeptin, 10 g/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride and 0.2 mM sodium orthovanadate for 20 min at 4°C. Cells were scraped from dishes and centrifuged at 15 000 g for 20 min at 4°C. Supernatants were either loaded for SDS–PAGE or subjected to further analysis.

Immunoprecipitation of samples standardized for proteins was performed for 1 h at 4°C with constant rotation. After this time protein A–Sepharose was added and incubation was carried out for an additional 1 h. Immunoprecipitates were washed five times with 1 ml ice-cold 50 mM Tris–HCl, pH 7.5, 500 mM NaCl, 0.1% Triton X-100, 0.5 mM MgCl₂ and 0.5 mM CaCl₂ and resuspended in 50 l 2 sample buffer. The sample were boiled for 5 min, subjected to SDS–PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore, Bedford, MA).

For Western blotting the membrane was first blocked with 20 mM Tris–HCl, pH 7.8, 150 mM NaCl and 2% BSA for 1 h at 42°C for all antibodies or with 10 mM Tris–HCl, pH 7.5, 100 mM NaCl, 1% BSA and 0.1% Tween 20 for anti-phosphotyrosine antibody and then probed in the same solution with primary antibody for 2 h at room temperature. After three washes membranes were incubated for 45 min at room temperature with horseradish peroxidase-conjugated protein A in 20 mM Tris–HCl, pH 7.8, 150 mM NaCl and 5% non-fat dry milk. The signal from immunoreactive bands was detected by ECL.

ERK activity assay

Cells were stimulated and harvested as above. Lysates, containing 400 g protein, were immunoprecipitated with 2 l anti-p42^ERK antibody. Immunoprecipitation was carried out at 4°C for 1 h before adding protein A–Sepharose and incubating for an additional 1 h. Immunoprecipitates were washed twice in lysis buffer and twice in 20 mM HEPES, pH 7.4, 10 mM MgCl₂ and 0.2 mM sodium orthovanadate and incubated in 0.25 mg/ml MBP, 50 M ATP and 5 Ci [-³²P]ATP for 15 min at 30°C. The reaction was terminated by addition of 4 sample buffer and the proteins subjected to electrophoresis in a 12.5% SDS–polyacrylamide gel and visualized by autoradiography. The bands corresponding to phosphorylated MBP were cut out and radioactivity measured using liquid scintillation counting.

Recombinant adenoviral vectors

The recombinant adenovirus vector AdMEK_CA, expressing the catalytically active form of MEK, was constructed from replication-deficient adenovirus type 5 (Ad5) with deletions in the E1 and E3 genes, Ad-dl327 and a plasmid containing Ad5 sequences from bp 22 to 5790 with a deletion of the E1 region from bp 342 to 3523, a polycloning site under control of the CMV promoter, the mutated human MEK gene and the SV40 polyadenylation signal. As previously described, the catalytically active mutant of MEK was achieved by substitution with Glu and Asp respectively of the regulatory residues Ser218 and Ser222 and by deletion of a predicted -helix encompassing residues 32–51 (Mansour et al., 1994). Aliquots of 10 g linearized plasmid were co-transfected with 10 g of the large fragment of ClaI-digested Ad-dl327 DNA into HEK 293 cells, to allow homologous recombination to occur, followed by replication and encapsidation of recombinant adenoviral DNA into infection virions and formation of plaques as a result of lysis of the infected cells. Individual plaques were isolated, amplified in HEK 293 cells and viral DNA isolated (Hirt, 1967) and analyzed for identification of recombinant viruses by Hind III DNA restriction analysis and DNA sequencing. Recombinant viruses were propagated in 50 T175 flasks of HEK 293 cells infected at a MOI of five. Cells were recovered 36–48 h after infection and viruses released by five cycles of freeze–thawing. All viral preparations were purified by CsCl density gradient centrifugation (Graham and Van Der Eb, 1973), dialyzed and stored at -70°C in 10 mM Tris–HCl, pH 7.4, 1 mM MgCl₂ and 10% glycerol until used. Titers of the viral stocks were determined by plaque assay using HEK 293 cells (Graham and Van Der Eb, 1973).

Adenoviral infection

GMC were infected with varying titers of AdLacZ in 1 ml RPMI 1640 medium containing 2% fetal bovine serum. After 1 h infection culture medium was added to the plate. Twenty-four hours after infection cells were washed twice with PBS and fixed in 1% glutaraldehyde, 1 mM MgCl₂ in PBS for 15 min. Cells were then washed three times with PBS and stained with 5 mM K₄Fe(CN)₆.3H₂O, 5 mM K₄Fe(CN)₆, 2 mM MgCl₂, 0.2% X-gal in PBS for 1–2 h at 37°C.

GMC were infected with AdMEK_CA at a MOI of 60. After 1 h infection serum-free medium was added to the plate and the cells maintained for 24 h before agonist stimulation.

Phosphatidylinositol 3-kinase activity assay

The determination of PI-3-K activity was performed essentially as described elsewhere (Fukui and Hanafusa, 1989). Briefly, cells were made quiescent and then stimulated with ET-1 (100 nM) as described above. Stimulation of cells was terminated by washing twice with ice-cold PBS. Cells were lysed with 20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA, 1% NP-40, 1 mM sodium orthovanadate and 1 mM phenylmethylsulfonyl fluoride for 20 min at 4°C. Lysates, containing an equal amount of protein, were immunoprecipitated with Y13-259 rat anti-p21^ras (800 l hybridoma supernatant) previously coupled with rabbit anti-rat antibody and protein A–Sepharose. The Sepharose beads containing the immunoprecipitates were washed twice with lysis buffer, once with PBS, once with 0.5 M LiCl, 100 mM Tris–HCl, pH 7.5, 1 mM sodium orthovanadate, once with distilled water and once with 20 mM Tris–HCl, pH 7.5, 0.5 mM EGTA, 100 mM NaCl, and resuspended in 50 l final wash solution to which was added 0.5 l 20 mg/ml phosphatidylinositol (Avanti Polar Lipids Inc., Alabaster, AL). Anti-p85 immunoprecipitates were finally resuspended either in phosphatidylinositol or in phosphatidylinositol 4-monophosphate in order to identify phosphatidylinositol monophosphate and biphosphate respectively on TLC. Samples were incubated at 25°C for 10 min, at which time [-³²P]ATP (10 Ci/sample) and MgCl₂ (20 mM final concentration) were added and incubation allowed to continue at 25°C for an additional 15 min. The reaction was stopped by addition of 150 l chloroform/methanol/11.6 N HCl (100:200:2) and subsequently 100 l chloroform was added to separate the organic phase. Reaction products were separated by TLC and resolved in chloroform/methanol/ammonium hydroxide/water (129:114:15:21). Spots were visualized by autoradiography of TLC plates and quantified by phosphorimager analysis.

The authors wish to thank Dr Nathalie Ahn for providing the MEK gene and the NIH 3T3 cells expressing the MEK construct, Dr Stephen Taylor for providing the GST–RBD construct, Dr Joseph Schlessinger for providing the anti-Shc and anti-p85 antisera and Dr Alan Saltiel for providing the MEK inhibitor PD 98059. We are also indebted to Dr Vladimir Poltoratsky for advice on PI-3-K assay, Dr Kirkwood Pritchard for his help with transgenic adenovirus construction and Dr Ann McGinty for her excellent editorial assistance. This work was supported by NIH grants HL 22563 and DK 41684 to M.J.D. and in part by a grant from the Milheim Foundation for Cancer Research (95-54) to A.S. Sunita Chari was supported by NIH training grant ST32 DK07470-12.

Alblas J, van Corven EJ, Hordijk PL, Milligan G and Moolenaar WH (1993) Gi-mediated activation of the p21^ras-mitogenic-activated protein kinase pathway by alfa2-adrenergic receptor expressed in fibroblasts. J Biol Chem, 268, 22235–22238. | PubMed | ISI | ChemPort |

Alessi DR, Saito Y, Campbell DG, Cohen P, Sithanandam G, Rapp U, Ashworth A, Marshall CJ and Cowley S (1994) Identification of the sites in MAP kinase kinase-1 phosphorylated by p74raf-1. EMBO J, 13, 1619–1619.

Aronheim A, Engelberg D, Li N, Al-Alawi N, Schlessinger J and Karin M (1994) Membrane targeting of the nucleic exchange factor Sos is sufficient for activating the Ras signaling pathway. Cell, 78, 949–961. | Article | PubMed | ISI | ChemPort |

Barnard D, Diaz B, Hettich L, Chuang E, Zhang XF, Avruch J and Marshall M (1995) Identification of the sites of interaction between c-Raf-1 and Ras–GTP. Oncogene, 10, 1283–1290. | PubMed | ISI | ChemPort |

Boguski MS and McCormick F (1993) Protein regulating Ras and its relatives. Nature, 366, 643–654. | Article | PubMed | ISI | ChemPort |

Bokemeyer D, Sorokin A and Dunn MJ (1997) The dual specificity protein tyrosine phosphatase CL100, B23 and PAC1: differential regulation and expression. J Am Soc Nephrol, 8, 40–50. | PubMed | ChemPort |

Buday L and Downward J (1993) Epidermal growth factor regulates p21^ras through the formation of a complex of receptor, Grb2 adaptor protein, and Sos nucleotide exchange factor. Cell, 73, 611–620. | Article | PubMed | ISI | ChemPort |

Burgering BM and Coffer PJ (1995) Protein kinase B (c-akt) in phosphatidylinositol-3-OH kinase signal transduction. Nature, 376, 599–602. | Article | PubMed | ISI | ChemPort |

Carpenter CL and Cantley LC (1996) Phosphoinositide kinases. Curr Opin Cell Biol, 8, 153–158. | Article | PubMed | ISI | ChemPort |

Cazaubon SM, Ramos-Morales F, Fischer S, Schweighoffer F, Stromberg AD and Couraud PO (1994) Endothelin induces tyrosine phosphorylation and Grb2 association of Shc in astrocytes. J Biol Chem, 269, 24805–24809. | PubMed | ISI | ChemPort |

Charles CH, Abler AS and Lau LF (1992) cDNA sequence of growth factor-inducible immediate early gene and characterization of its encoded protein. Oncogene, 7, 187–190. | PubMed | ISI | ChemPort |

Charles CH, Sun H, Lau LF and Tonks NK (1993) The growth factor-inducible immediate early gene 3CH134 encodes a protein-tyrosine-phosphatase. Proc Natl Acad Sci USA, 90, 5292–5296. | Article | PubMed | ChemPort |

Chen Y, Grall D, Salcini EA, Pelicci PG, Pouyssegur J and van Obberghen-Shilling E (1996) Shc adaptor proteins are key transducers of mitogen signaling mediated by the G protein-coupled thrombin receptor. EMBO J, 15, 1037–1044. | PubMed | ISI | ChemPort |

Cherniack AD, Klarlund JK and Czech MP (1994) Phosphorylation of the Ras nucleotide exchange factor son of sevenless by mitogen-activated protein kinase. J Biol Chem, 269, 4717–4720. | PubMed | ISI | ChemPort |

Cherniack AD, Klarlund JK, Conway BR and Czech MP (1995) Disassembly of Son-of-Sevenless protein from Grb2 during p21^ras desensitization by insulin. J Biol Chem, 270, 1485–1488. | Article | PubMed | ISI | ChemPort |

Cook SJ, Rubinfeld B, Albert I and McCormick F (1993) RapV12 antagonizes Ras-dependent activation of ERK1 and ERK2 by LPA and EGF in Rat-1 fibroblasts. EMBO J, 12, 3475–3485. | PubMed | ISI | ChemPort |

Crespo P, Xu N, Simonds WF and Gutkind JD (1994) Ras-dependent activation of MAP kinase pathway mediated by G-protein subunits. Nature, 369, 418–420. | Article | PubMed | ISI | ChemPort |

Cyr CR, Rudy B and Kris RM (1993) Prolonged desensitization of the human endothelin A receptor in Xenopus oocytes. Comparative studies with human neurokin A receptor. J Biol Chem, 268, 26071–26074. | PubMed | ChemPort |

Dikic I, Tokiwa G, Lev S, Courtneiidge SA and Schlessinger J (1996) A role for Pyk2 and Src in linking G protein-coupled receptors with MAP kinase activation. Nature, 383, 547–550. | Article | PubMed | ISI | ChemPort |

Doub H, Weiss FU, Wallasch C and Ullrich A (1996) Role of transactivation of the EGF receptor in signaling by G-protein-coupled receptor. Nature, 379, 557–560. | Article | PubMed | ISI | ChemPort |

Downward J, Graves JD, Warne PH, Rayter S and Cantrell DA (1990) Stimulation of p21^ras upon T-cell activation. Nature, 346, 719–723. | Article | PubMed | ISI | ChemPort |

Egan SE, Giddings BW, Brooks MW, Buday L, Sizeland AM and Weinberg RA (1993) Association of Sos Ras exchange protein with Grb2 is implicated in tyrosine kinase signal transduction and transformation. Nature, 363, 45–51. | Article | PubMed | ISI | ChemPort |

Force T, Kyriakis JM, Avruch J and Bonventre JV (1991) Endothelin, vasopressin, and angiotensin II enhance tyrosine phosphorylation by protein kinase C-dependent and -independent pathways in glomerular mesangial cells. J Biol Chem, 266, 6650–6656. | PubMed | ISI | ChemPort |

Fukui Y and Hanafusa H (1989) Phosphatidylinositol kinase activity associates with viral p60^src protein. Mol Cell Biol, 9, 1651–1658. | PubMed | ChemPort |

Gotoh N, Tojo A and Shibuya M (1996) A novel pathway from phosphorylation of tyrosine residues 239/240 of Shc, contributing to suppress apoptosis by IL-3. EMBO J, 15, 6197–6204. | PubMed | ISI | ChemPort |

Gotoh N, Toyoda M and Shibuya M (1997) Tyrosine phosphorylation sites at amino acids 239 and 240 of Shc are involved in epidermal growth factor-induced mitogenic signaling that is distinct from Ras/mitogen-activated protein kinase activation. Mol Cell Biol, 17, 1824–1831. | PubMed | ISI | ChemPort |

Graham FL and van der Eb AJ (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology, 52, 456–467. | Article | PubMed | ISI | ChemPort |

Herman WH and Simonson MS (1995) Nuclear signaling by endothelin-1. J Biol Chem, 270, 11654–11661. | Article | PubMed | ISI | ChemPort |

Herskowitz I (1995) MAP kinase pathways in yeast: for mating and more. Cell, 80, 187–197. | Article | PubMed | ISI | ChemPort |

Hirt B (1967) Selective extraction of polyoma DNA from infected mouse cell culture. J Mol Biol, 26, 365–369. | Article | PubMed | ISI | ChemPort |

Holt KH, Kasson BG and Pessin JE (1996) Insulin stimulation of a MEK-dependent but ERK-independent Sos protein kinase. Mol Cell Biol, 16, 577–583. | PubMed | ISI | ChemPort |

Hu P, Margolis B, Skolnik EY, Lammers A, Ullrich A and Schlessinger J (1992) Interaction of phosphatidylinositol 3-kinase-associated p85 with epidermal growth factor and platelet-derived growth factor receptors. Mol Cell Biol, 12, 981–990. | PubMed | ISI | ChemPort |

Hu Y and Bowtell DDL (1996) Sos1 rapidly associate with Grb2 and is hypophosphorylated when complexed with the EGF receptor after EGF stimulation. Oncogene, 12, 1865–1872. | PubMed | ISI | ChemPort |

Hunter T (1995) Protein kinase and phosphatases: the Yin and Yang of protein phosphorylation and signaling. Cell, 80, 225–236. | Article | PubMed | ISI | ChemPort |

Kavanaugh WM et al. (1996) Multiple forms of an inositol polyphosphate 5-phosphatase form signaling complexes with Shc and Grb2. Curr Biol, 6, 438–445. | Article | PubMed | ISI | ChemPort |

Keyse SM and Emslie EA (1992) Oxidative stress and heat shock induce a gene encoding a protein-tyrosine phosphatase. Nature, 359, 644–647. | Article | PubMed | ISI | ChemPort |

Klarlund JK, Cherniack AD and Czech MP (1995) Divergent mechanism for homologous desensitization of p21^ras by insulin and growth factors. J Biol Chem, 270, 23421–23428. | Article | PubMed | ChemPort |

Klarlund JK, Cherniack AD, McMahon M and Czech MP (1996) Role of the Raf/mitogen-activated protein kinase pathway in p21^ras desensitization. J Biol Chem, 271, 16674–16677. | Article | PubMed | ChemPort |

Langlois WJ, Sasaoka T, Saltiel AR and Olefsky JM (1995) Negative feedback regulation and desensitization of insulin-and epidermal growth factor-stimulated p21^ras activation. J Biol Chem, 270, 25320–25323. | Article | PubMed | ISI | ChemPort |

Lefkowitz RJ (1993) G protein-coupled receptor kinase. Cell, 74, 409–412. | Article | PubMed | ISI | ChemPort |

Li N, Batzer A, Daly R, Yajnik R, Skolnik E, Bar-Sagi D, Margolis B and Schlessinger J (1993) Guanine-nucleotide-releasing factor hSos1 binds to Grb2 and links receptor tyrosine kinase to Ras signaling. Nature, 363, 85–88. | Article | PubMed | ISI | ChemPort |

Mansour SJ, Wayne TM, Hermann AS, Candia JM, Rong S, Fukasawa K, Vande Woude GF and Ahn NG (1994) Transformation of mammalian cells by constitutively active MAP kinase kinase. Science, 265, 966–970. | Article | PubMed | ISI | ChemPort |

Marshall CJ (1994) MAP kinase kinase kinase, MAP kinase kinase, MAP kinase. Curr Opin Genet Dev, 4, 82–89. | Article | PubMed | ChemPort |

McCormick F (1994) Activators and effectors of p21 proteins. Curr Opin Genet Dev, 4, 71–76. | Article | PubMed | ChemPort |

Payne DM, Rossamando AJ, Martino P, Erickson AK, Her J-H, Shabanowitz J, Hunt DF, Weber MJ and Sturgill TW (1991) Identification of a regulatory phosphorylation sites in p42/mitogen-activated protein kinase (MAP kinase). EMBO J, 10, 885–892. | PubMed | ISI | ChemPort |

Pelicci G, Lanfrancone L, Grignani F, McGlade J, Cavallo F, Forni G, Nicoletti I, Pawson T and Pelicci PG (1992) A novel transforming protein (SHC) with SH2 domain is implicated in mitogenic signal transduction. Cell, 70, 93–104. | Article | PubMed | ISI | ChemPort |

Porfiri E and McCormick F (1996) Regulation of epidermal growth factor receptor signaling by phosphorylation of the ras exchange factor hSos1. J Biol Chem, 271, 5871–5877. | Article | PubMed | ISI | ChemPort |

Rodriguez-Viciana P, Warne PH, Dhand R, Vanhaesebroeck B, Gourt I, Fry MJ, Waterfield MD and Downward J (1994) Phosphatidylinositol-3-OH kinase as a direct target of Ras. Nature, 370, 527–532. | Article | PubMed | ISI | ChemPort |

Sadoshima J and Izumo S (1996) The heterotrimeric Gq protein-coupled angiotensin II receptor activates p21^ras via the tyrosine kinase–Shc–Grb2–Sos pathway in cardiac myocytes. EMBO J, 15, 775–787. | PubMed | ISI | ChemPort |

Salcini AE, McGlade J, Pelicci G, Nicoletti I, Pawson T and Pelicci PG (1994) Formation of Shc–Grb2 complexes is necessary to induce neoplastic transformation by overexpression of Shc proteins. Oncogene, 9, 2827–2836. | PubMed | ISI | ChemPort |

Satoh T, Nakafuku M and Kaziro Y (1992) Function of Ras as a molecular switch in signal transduction. J Biol Chem, 267, 24149–24152. | PubMed | ISI | ChemPort |

Schramek H and Dunn MJ (1997) Endothelin-induced intracellular signaling pathways. In Huggins,J.P. and Pelton,J.T. (eds), Endothelins in Biology and Medicine. CRC Press, Boca Raton, FL, pp. 81–100.

Seger R and Krebs EG (1995) The MAPK signaling cascade. FASEB J, 9, 726–735. | PubMed | ISI | ChemPort |

Simonson MS and Dunn MJ (1990) Eicosanoid biochemistry in cultured glomerular mesangial cells. Methods Enzymol, 187, 544–553. | PubMed | ChemPort |

Simonson MS and Dunn MJ (1993) Endothelin peptides and the kidney. Annu Rev Physiol, 55, 249–265. | Article | PubMed | ChemPort |

Simonson MS, Wang Y and Herman WH (1996) Nuclear signaling by endothelin-1 requires Src protein-tyrosine kinases. J Biol Chem, 271, 77–82. | Article | PubMed | ISI | ChemPort |

Sjolander A, Yamamoto K, Huber BE and Lapetina EG (1991) Association of p21^ras with phosphatidylinositol 3-kinase. Proc Natl Acad Sci USA, 88, 7908–7912. | Article | PubMed | ChemPort |

Sun H, Charles CH, Lau LF and Tonks NK (1993) MKP-1 (3CH134) an immediate early product, is a dual specificity phosphatase that dephosphorylates MAP kinase in vivo. Cell, 75, 487–493. | Article | PubMed | ISI | ChemPort |

Sun H, Tonks NK and Bar-Sagi D (1994) Inhibition of Ras-induced DNA synthesis by expression of the phosphatase MKP-1. Science, 266, 258–288.

Taylor SJ and Shalloway D (1996) Cell cycle-dependent activation of Ras. Curr Biol, 6, 1621–1627. | Article | PubMed | ISI | ChemPort |

van Biesen T, Hawes BE, Luttrell DK, Krueger KM, Touhara K, Porfiri E, Sakaue M, Luttrell LM and Lefkowitz RJ (1995) Receptor-tyrosine-kinase and G-mediated MAP kinase activation by a common signaling pathway. Nature, 376, 781–784. | Article | PubMed | ChemPort |

Van der Geer P, Wiley S, Gish GD and Pawson T (1996) The Shc adaptor protein is highly phosphorylated at conserved, twin tyrosine residues (Y239/240) that mediate protein–protein interaction. Curr Biol, 6, 1435–1444. | Article | PubMed | ChemPort |

Vojtek AB, Hollenberg SM and Cooper JA (1993) Mammalian Ras interacts directly with the serine/threonine kinase Raf. Cell, 74, 205–214. | Article | PubMed | ISI | ChemPort |

Wang Y, Simonson MS, Pouysseguer J and Dunn MJ (1992) Endothelin rapidly stimulates mitogen-activated protein kinase activity in rat mesangial cells. Biochem J, 287, 589–594. | PubMed | ISI | ChemPort |

Wang W, Fisher ECM, Jia Q, Dunn JM, Porfiri E, Downward J and Egan SE (1995) The Grb2 binding domain of mSos1 is not required for downstream signal transduction. Nature Genet, 10, 294–300. | Article

Waters SB, Yamauchi K and Pessin JE (1995) Insulin-stimulated dissociation of the Sos–Grb2 complex. Mol Cell Biol, 15, 2791–2799. | PubMed | ISI | ChemPort |

Waters SB, Chen D, Kao AW, Okada S, Holt KH and Pessin JE (1996) Insulin and epidermal growth factor receptor regulate distinct pools of Grb2/Sos in the control of Ras activation. J Biol Chem, 271, 18224–18230. | Article | PubMed | ISI | ChemPort |

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising