Role of the Raf/mitogen-activated protein kinase pathway in p21ras desensitization.

Desensitization of p21ras after stimulation of cells by growth factors and phorbol 12-myristate 13-acetate (PMA) correlates with hyperphosphorylation of the guanine nucleotide exchange factor Son-of-sevenless (Sos) and its dissociation from the adaptor protein Grb2 (Cherniack, A., Klarlund, J. K., Conway, B. R., and Czech, M. P. (1995) J. Biol. Chem. 270, 1485-1488). To test the role of the Raf/mitogen-activated protein (MAP) kinase pathway, we utilized cells expressing a chimera composed of the catalytic domain of p74Raf-1 and the hormone binding domain of the estradiol receptor (ΔRaf-1:ER). Estradiol markedly stimulated ΔRaf-1:ER and the downstream MEK and MAP kinases in these cells as well as Sos phosphorylation. However, the dissociation of Grb2 from Sos observed in response to PMA was not apparent upon ΔRaf-1:ER activation. Furthermore, stimulation of ΔRaf-1:ER did not impair GTP loading of p21ras in response to platelet-derived growth factor or epidermal growth factor. We conclude that activation of the Raf/MAP kinase pathway alone in these cells is insufficient to cause disassembly of Sos from Grb2 or to interrupt the ability of Sos to catalyze activation of p21ras.

The proto-oncogene product p21 ras serves as a pivotal intermediate in the transmission of signals from numerous growth factor receptors and the insulin receptor to downstream effectors (1)(2)(3)(4)(5)(6)(7)(8). p21 ras exists in an inactive GDP-bound state but is converted to the active GTP-bound state through the activities of guanine nucleotide exchange factors, such as Son-of-sevenless (Sos 1 ). Proline-rich regions in the COOH terminus of Sos bind src homology 3 (SH3) groups present in adaptor proteins such as Grb2, which also contains an src homology 2 (SH2) group. Upon stimulation of cells with growth factors, the Sos⅐Grb2 complex is recruited to the plasma membrane via the binding of the SH2 group of Grb2 to tyrosine-phosphorylated sites on activated receptors, to phosphorylated tyrosine 317 on SHC, or possibly to other tyrosine phosphates on membraneassociated proteins (9 -16).
GTP-loaded p21 ras binds to the amino-terminal part of the protein kinase Raf, which is thereby translocated to the plasma membrane where it undergoes an activation process (17)(18)(19)(20)(21).
These events initiate the activation of a cascade of protein kinases. Mitogen-activated protein kinase kinase (MEK) is apparently directly phosphorylated by Raf, resulting in its activation. MEK in turn stimulates mitogen-activated protein kinase (MAPK) by causing its phosphorylation on a tyrosine and a threonine residue. Whereas the substrate specificities of Raf and MEK are extremely limited, MAPK can phosphorylate a broad spectrum of proteins and has been shown to regulate numerous cellular enzymes such as phospholipase A 2 , nuclear transcription factors, and other protein kinases (22).
An important aspect of the function of p21 ras and other intracellular signaling molecules is that their activation is reversible. The activity of p21 ras is typically found to peak 2-5 min after growth factor stimulation, and it declines to near basal levels after 20 min. The deactivation mechanism, however, is not well understood. Grb2 in 3T3-L1 adipocytes and CHO-IR cells partially dissociates from Sos after stimulation with insulin or platelet-derived growth factor, which presumably results in disruption of p21 ras -activating complexes (23)(24)(25). After stimulation and desensitization of 3T3-L1 adipocytes with epidermal growth factor (EGF), no such dissociation is seen, but p21 ras deactivation in this case may be explained by the rapid down-regulation of EGF receptors observed in this system (24).
As has been noted by several groups, Sos is highly phosphorylated after growth factor stimulation (12, 23, 26 -28). The time course of this phosphorylation suggests a potential role in the deactivation rather than the activation phase of p21 ras regulation. A negative feedback role of MAPK in deactivating p21 ras is attractive for several reasons. The existence of negative feedback by a downstream kinase is likely in Saccharomyces cerevisiae since the p21 ras exchange factor CDC25 is phosphorylated by protein kinase A, and the exchange factor is concomitantly released from the membrane (29). A large number of consensus sites for MAPK phosphorylation are found in the COOH-terminal region of Sos protein, and we have previously reported that at least two sites are phosphorylated in vitro and in intact cells heterologously expressing Drosophila Sos (30). Inhibition of the Raf/MAPK pathway at the level of MEK either by a cell-permeable inhibitor or by a dominant negative mutant of MEK blocked Grb2/Sos dissociation in response to insulin and the deactivation of p21 ras (28,31). Taken together, these observations strongly imply that MAPK or other protein kinases downstream of MEK are necessary for both Grb2/Sos dissociation and for the deactivation of p21 ras seen after the initial activation by insulin. The aim of the work presented here was to assess whether the protein kinases in the Raf/MAPK pathway or other downstream kinases are indeed sufficient to desensitize p21 ras to growth factor stimulation or whether other events are also necessary. We report here that activation of the Raf/MAPK pathway, independent of growth factor stimulation, does not cause dissociation of Grb2 * This work was supported by National Institutes of Health Grant DK 30648. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 The abbreviations used are: Sos, Son of sevenless; EGF, epidermal growth factor; PAGE, polyacrylamide gel electrophoresis; PMA, phorbol 12-myristate 13-acetate; PDGF, platelet-derived growth factor; MEK, mitogen-activated protein kinase kinase; MAPK, mitogen-activated protein kinase.
from Sos or p21 ras desensitization. One or more other regulatory mechanisms must also play a role in mediating disassembly of complexes of Grb2 and Sos in response to growth factors.

EXPERIMENTAL PROCEDURES
Materials-The anti-phosphotyrosine antibody PY20 and monoclonal anti-Grb2 antibody was from Transduction Laboratories. For immunoprecipitation of Sos, an antibody raised against a peptide corresponding to amino acids 100 -120 was used (23). For immunoblotting of Sos, an antibody to a COOH-terminal part of the Sos-1 was kindly provided by Dr. H. Meisner (32). PMA was from LC laboratories. Estrogen was from Sigma. Platelet-derived growth factor-BB (PDGF-BB) and EGF were from Upstate Biotechnology.
Cell Lysis, Immunoprecipitation, and Immunoblotting--After stimulation, cells were washed in 10 ml of ice-cold phosphate-buffered saline (1.8 mM KH 2 PO 4 , 171 mM NaCl, 1.0 mM Na 2 HPO 4 , and 3.4 mM KCl) and lysed in 1 ml of cold lysis buffer (30 mM Hepes, pH 7.5, 100 mM NaCl, 1 mM EGTA, 1% Triton X-100, 2 mM p-nitrophenyl phosphate, 50 mM NaF, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 5 g/ml each of leupeptin, aprotinin and pepstatin, and 1 mM benzamidine). Lysates were spun in a microfuge at 15,000 ϫ g for 10 min at 4°C. The supernatants were removed and assayed for total protein content using the Bradford method (34). Equal amounts of protein from each lysate were then precleared by the addition of 10 l of protein A-Sepharose (Pharmacia Biotech Inc.) and incubated on an end-over-end mixer at 4°C for 1 h. Samples were then centrifuged at 15,000 ϫ g for 2 min at 4°C, and supernatants were incubated on an end-over-end mixer with 10 l of anti-Sos and 25 l of protein A-Sepharose for 16 h. The Sepharose was pelleted by centrifugation at 15,000 ϫ g for 2 min at 4°C. Pellets were washed five times with cold wash buffer (phosphatebuffered saline with 0.1% Triton X-100, 2 mM P-nitrophenyl phosphate, 50 mM NaF, 1 mM Na 3 VO 4 ), and the protein was dissolved in SDS-PAGE sample buffer. Samples were loaded on SDS-PAGE gels and transferred to nitrocellulose filters. Filters were probed with antibodies, and antibody-bound proteins were visualized using ECL (Amersham) according to the manufacturer's specifications.
Ras Activation Assay-The cells were transferred to 4.5 ml of phosphate-free Dulbecco's modified Eagle's medium with 0.5% fetal bovine serum when they had just reached confluence and were used 48 h later. 1 mCi of 32 P i (DuPont NEN) was added to each plate for the last 24 h of incubation. Cells were then rapidly washed with ice-cold phosphatebuffered saline and lysed by addition of 800 l of lysis buffer (20 mM Tris, pH 8.0, 100 mM NaCl, 1 mM MgCl 2 , 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl chloride, 1 mM benzamidine, 1% Triton X-100, and aprotinin, pepstatin, and leupeptin at 0.5 g/ml each) with 10% tissue culture supernatant from the hybridoma cell line Y13-259. After clarification by centrifugation, lysates were incubated for 1 h with 5 l of Sepharose 4B coupled to goat anti-rat antibodies (Organon Teknika, Cappel). Antibody-bound beads were collected by centrifugation and washed extensively in 50 mM Hepes, pH 7.4, 0.5 M NaCl, 5 mM MgCl 2 , 0.1% Triton X-100, and 0.005% SDS. Bound GDP and GTP was eluted and subjected to chromatography on polyethylenimine cellulose plates (Merck) as described with GDP and GTP as unlabeled markers (35). The locations of the GDP and GTP were visualized by UV light, and relevant areas were cut out and analyzed for radioactivity in a beta counter.

RESULTS AND DISCUSSION
A chimeric protein containing the catalytic domain of oncogenic Raf-1 fused to the hormone binding region of the estradiol receptor (⌬Raf-1:ER) exhibits protein kinase activity that can be rapidly activated by addition of estradiol, resulting in activation of endogenous MEK and MAPK (33,36). Cells stably expressing ⌬Raf-1:ER therefore allow examination of the effects of activation of Raf-1/MAPK kinase cascade independently of other signaling pathways. We first examined the time course of p21 ras activation in response to EGF and PDGF in C7 3T3 cells expressing ⌬Raf-1:ER (3T3⌬Raf-1:ER cells) to verify that deactivation does occur in these cells after initial activation by the growth factors. As is seen in Fig. 1, both hormones rapidly induce a 3-4-fold increase in GTP loading of p21 ras , and this is followed by a deactivation phase resulting in the return of GTP⅐p21 ras concentrations to basal levels after 20 -30 min. As seen in other systems, the deactivation of p21 ras occurs in parallel with hyperphosphorylation of mSos as evidenced by its decreased mobility on SDS-PAGE (data not shown). Sos contains numerous consensus sites for MAPK phosphorylation in its COOH-terminal portion and can be phosphorylated by MAP kinase both in vitro and in intact cells (30). We thus tested whether MEK inhibition abolishes Sos hyperphosphorylation and Grb2 dissociation from Sos, as reported recently for Chinese hamster ovary cells or rat-1 cells overexpressing the insulin receptor (28,31). 3T3⌬Raf-1:ER cells were treated with or without the potent MEK inhibitor, PD98059 (37), and incubated in the presence or absence of PMA to initiate the phosphorylation and disassembly of Sos from Grb2. Fig. 2 shows that these two previously reported effects of PMA in 3T3-L1 adipocytes are also observed in 3T3⌬Raf-1:ER cells. Importantly, the MEK inhibitor blocked both the retarded electrophoretic mobility of Sos (panel A) as well as the 50% loss of Grb2 from immunoprecipitated Sos (panels B and C) caused by PMA. These data confirm that MEK and presumably MAPK are required for these processes to proceed normally.
To determine whether selective activation of the Raf/MAPK pathway downstream of receptor or protein kinase C elements causes phosphorylation of Sos and its dissociation from Grb2, 3T3⌬Raf-1:ER cells were treated with or without estradiol or PMA. Estradiol treatment of 3T3⌬Raf-1:ER cells caused a marked electrophoretic mobility shift of the Sos protein, which was even more extensive than that caused by PMA (Fig. 3A). It is noteworthy that all of the Sos from the untreated cells appears to be converted to the slower migrating species, indicating that virtually all Sos molecules are phosphorylated in response to ⌬Raf-1:ER activation. This system therefore allows rigorous testing of the hypothesis that phosphorylation of Sos by protein kinases in the Raf/MAPK pathway constitutes a feedback mechanism that results in Grb2/Sos dissociation and desensitization of p21 ras to growth factor stimulation. Surprisingly, the apparent hyperphosphorylation of Sos by ⌬Raf-1:ER activation was not accompanied by any detectable dimunition of Grb2 from anti-Sos immune complexes (Fig. 3, B and C). Under these same experimental conditions, PMA treatment of 3T3⌬Raf-1:ER cells elicited release of over half of the Grb2 band to cellular Sos protein. These data lead to the unexpected conclusion that selective stimulation of MAPK via ⌬Raf-1:ER activation can cause Sos hyperphosphorylation but is insufficient to signal Grb2/Sos disassembly in 3T3⌬Raf-1:ER cells.
Although Sos disassembly from Grb2 does not occur in response to MAP kinase activation (Fig. 3), it is possible that GTP⅐p21 ras levels are regulated by MAP kinase through other mechanisms. We tested this hypothesis by treating 32 P-labeled 3T3⌬Raf-1:ER cells with or without estradiol for 30 min prior to a 5-min exposure to either EGF or PDGF and then analyzing the labeled guanosine nucleotide content of p21 ras . Fig. 4 demonstrates the failure of estradiol to modulate the extent of p21 ras activation by EGF or PDGF under these experimental conditions. Parallel plates of cells treated with estradiol for 30 min were shown to exhibit marked hyperphosphorylation of endogenous Sos, verifying the effectiveness of estradiol to stimulate the Raf/MAPK pathway (data not shown, see Figs. 2 and 3). Thus, activation of MAPK via the ⌬Raf-1:ER is not sufficient to cause desensitization of p21 ras under these conditions.
In this same series of experiments (Fig. 3), the cells were also incubated with or without okadaic acid, a potent phosphatase inhibitor known to induce or potentiate the activation of several cellular protein kinases including MAPK (38,39). When added alone to 3T3⌬Raf-1:ER cells, this agent only slightly decreased the electrophoretic mobility of Sos (Fig. 3A) and did not detectably dissociate Grb2 from Sos (Fig. 3, B and C). However, incubation of okadaic acid with estradiol-treated 3T3⌬Raf-1:ER cells potentiated Sos hyperphosphorylation and caused a marked, 60% decrease in Grb2 associated with Sos protein. Thus, the combined signaling pathways elicited by ⌬Raf-1:ER and okadaic acid are able to cause disassembly of complexes containing Sos and Grb2, whereas each of these pathways acting alone is insufficient to mediate this response.
Previous results showed that prolonged activation of ⌬Raf-1:ER in 3T3 cells induces a partial block in activation of endogenous Raf/MAPK pathway by PDGF, which does not appear to be due to impaired function of the PDGF receptor (36). However, in these cells Sos is heavily phosphorylated in response to ⌬Raf-1:ER activation, and it was therefore hypothesized that the block was at the level of p21 ras activation. We therefore examined whether prolonged activation of ⌬Raf-1:ER results in diminished responsiveness of p21 ras to activation by PDGF or EGF (Fig. 4). After activation of ⌬Raf-1:ER for 15 h, the cells appeared morphologically transformed as described previously (33). However, GTP⅐p21 ras levels were stimulated to the same extent in estradiol-treated cells as in control cells in response to PDGF under these experimental conditions (Fig. 4). The block of endogenous Raf-1 kinase activation by PDGF seen in ⌬Raf-1:ER cells treated for prolonged periods with estradiol is therefore not at the level of GTP loading of p21 ras . Rather, the underlying inhibitory mechanism is likely due to modifications of Raf-1 itself or of other components that modulate its protein kinase activity.
Interestingly, prolonged treatment of 3T3⌬Raf-1:ER cells with estradiol did inhibit the ability of EGF to activate p21 ras (Fig. 4). To examine whether EGF receptors function normally in these cells, extracts were subjected to Western blotting with anti-phosphotyrosine antibodies. As shown in Fig. 4C, the tyrosine phosphorylation of the PDGF receptor upon treatment with PDGF was unaffected by the estradiol treatment of the ⌬Raf-1:ER cells. In contrast, activation of ⌬Raf-1:ER for 15 h, but not for 30 min, resulted in a dramatic reduction of the signal from the EGF receptor (Fig. 4). 2 Such a reduction was not seen in untransfected C7 3T3 cells (data not shown). Thus, the reduction in p21 ras activation seen after prolonged activation of ⌬Raf-1:ER is likely due to down-regulation of the EGF receptor as a result of activation of the Raf/MAP kinase pathway.
Taken together, the data presented here (Figs. 3 and 4) are inconsistent with the simple model that MAP kinase-mediated phosphorylation of Sos proteins causes their disassembly from Grb2 as well as p21 ras desensitization. This conclusion is reinforced by the observation that EGF does not cause Grb2/Sos dissociation in 3T3-L1 adipocytes whereas insulin and PMA do, even though EGF activates MAP kinase and induces Sos hyperphosphorylation in these cells (24, 40 -43). On the other hand, MAPK seems necessary for dissociation of Sos from Grb2 to occur (Refs. 28 and 31; Fig. 2). One reasonable hypothesis is that insulin and PMA activate another protein kinase or kinases that are necessary for Grb2/Sos dissociation but that this kinase is not activated by other growth factors such as EGF. Thus, activation of both MAP kinase and another unidentified protein kinase might be required to phosphorylate the putative multiple sites on Sos necessary for Grb2 dissociation. This dissociation presumably contributes to the desensitization of p21 ras (for detailed discussion, see Ref. 24). The deactivation that occurs after initial p21 ras activation in response to EGF could be caused by a different mechanism: down-regulation of EGF receptors (24,44). Clearly, a more precise definition of the protein kinases that phosphorylate Sos will be of importance in understanding the molecular basis of Grb2/Sos dissociation in response to insulin and PMA.