Report

Requirement of Ras-GTP-Raf Complexes for Activation of Raf-1 by Protein Kinase C

See allHide authors and affiliations

Science  03 Apr 1998:
Vol. 280, Issue 5360, pp. 109-112
DOI: 10.1126/science.280.5360.109

This article has a correction. Please see:

Abstract

Receptor tyrosine kinase–mediated activation of the Raf-1 protein kinase is coupled to the small guanosine triphosphate (GTP)–binding protein Ras. By contrast, protein kinase C (PKC)–mediated activation of Raf-1 is thought to be Ras independent. Nevertheless, stimulation of PKC in COS cells led to activation of Ras and formation of Ras–Raf-1 complexes containing active Raf-1. Raf-1 mutations that prevent its association with Ras blocked activation of Raf-1 by PKC. However, the activation of Raf-1 by PKC was not blocked by dominant negative Ras, indicating that PKC activates Ras by a mechanism distinct from that initiated by activation of receptor tyrosine kinases.

The PKC family of lipid-dependent Ser-Thr kinases has at least 11 members (1). The typical and atypical PKC isozymes are activated by diacylglycerol (DAG), which is produced by the metabolism of phosphatidyl inositols. In the case of 7-transmembrane (7-TM) receptors that are coupled to the heterotrimeric guanosine nucleotide–binding protein (G protein) Gαq, DAG is generated by the activation of phospholipase C-β (PLC-β) (2). Both the typical and atypical PKCs are activated by phorbol esters. Activation of PKC leads to short-term responses such as altered metabolic activity and to long-term responses such as differentiation or effects on proliferation and apoptosis (1). The extracellular signal–regulated kinases (ERKs) are mitogen-activated protein kinases (MAPKs), which are activated by PKC (3-7) and appear to mediate the effects of PKC on differentiation, secretion, proliferation, and hypertrophy (8). Signaling from receptor tyrosine kinases to ERKs is dependent on Ras proteins and the protein kinase Raf-1 (9). However, the role of Ras in transducing signals from PKC to the ERKs is unclear, because expression of a dominant negative Ras in which amino acid 17 is changed to Asn (N17Ras) does not block ERK activation by PKC in a number of cell types (4, 5). This Ras mutant, which is thought to function by inhibiting guanine nucleotide exchange factors (10), blocks activation of ERKs in response to stimulation of receptor tyrosine kinases in many cell types (4-6).

To investigate whether PKC activates the ERK MAPK pathway by a mechanism independent of Ras, we used a monkey kidney cell line (COS cells), because in these cells, N17Ras does not block PKC-mediated ERK activation (4). We blocked Ras signaling by microinjection of the Ras-neutralizing monoclonal antibody Y13-259 (11). To detect ERK activation, we used an antibody that recognizes only the dually phosphorylated, active form of ERK. Activation of ERK in COS cells treated with 12-O-tetradecanoyl-phorbol-13-acetate (TPA) was completely blocked by microinjection of Y13-259 (Fig.1). This demonstrates an essential role for Ras in activation of the ERK MAPK cascade by PKC. These data are consistent with studies on overexpression of Ras guanosine triphosphatase–activating protein (p120Ras-GAP), which blocks TPA-stimulated activation of ERKs, although it is unclear in those studies whether the p120Ras-GAP was acting on Ras or a related protein (7).

Figure 1

Requirement of Ras for activation of ERK by PKC. COS cells were microinjected with an expression vector for myc-tagged ERK2 together with either control rat immunoblobulin G (IgG) (top and middle panels) or with Y13-259 (bottom panels). The cells were deprived of serum (Con) or stimulated with TPA for 20 min as indicated. Staining with the 9E10 monoclonal antibody to myc (anti-Myc) and with an antibody to rat IgG (anti-Rat IgG) were used to reveal the injected cells. Activated ERK was detected with an antibody that recognizes dually-phosphorylated ERK (phospho-ERK; Promega, no. V6671). For each treatment, identical fields of cells are shown stained with the three antisera.

To further investigate whether Ras has a role in activation of Raf-1 by PKC, we used a mutant Raf-1 protein in which Arg89is replaced with Leu (R89LRaf-1). This mutant does not bind the GTP-bound form of Ras (Ras-GTP) (12). Unlike transiently expressed wild-type Raf-1 (mRaf-1), R89LRaf-1 was not activated in cells treated with TPA (Fig. 2A). Exogenous mRaf-1 was activated with similar kinetics to those of the endogenous protein (13), and activation was blocked in cells treated with the PKC inhibitor Ro 31-8220 (Fig. 2B) (14). Ro 31-8220 had only a small effect on activation of mRaf-1 in cells treated with epidermal growth factor (EGF) (Fig. 2B). Expression of N17Ras, which has no effect on ERK activation in COS cells treated with TPA (4), did not block activation of mRaf-1 following TPA treatment, but did block mRaf-1 activation in response to EGF (Fig. 2B). To show that activation of Raf-1 by a physiological activator of PKC also depends on Ras, we examined the activation of mRaf-1 in cells expressing human muscarinic type 1 receptor (Hm1), which is coupled to Gαq (15). There was no activation of R89LRaf-1 when the Hm1 receptor was activated in cells treated with carbachol (Fig. 2A), whereas wild-type mRaf-1 was activated 8- to 10-fold in an Ro 31-8220–sensitive but N17Ras-insensitive manner (Fig. 2B). The inability of R89LRaf-1 to be activated by tyrosine kinase signaling can be overcome by targeting it to the plasma membrane with a CAAX motif (16, 17). R89LRafCAAX was activated in cells treated with TPA (Fig. 2C), which shows that activation of Raf-1 by PKC requires Ras-GTP–mediated translocation to the plasma membrane, but that additional events at the membrane are required for full activation.

Figure 2

Activation of mRaf-1 by PKC. (A) Activation of Raf-1 by PKC requires Raf-1 interaction with Ras. mRaf-1 or R89LRaf-1 were expressed in COS cells (17) alone or together with Hm1 as indicated. The cells were deprived of serum for 24 hours and stimulated for 20 min with EGF (10 ng/ml), TPA (40 nM), or carbachol (Carb, 100 μM). The activity of mRaf-1 was measured from COS cell extracts by immunoprecipitation with the 9E10 monoclonal antibody and kinase cascade assay, using GSTMek-1, GSTERK2, and myelin basic protein as substrates (24). In the absence of added GSTMek-1, GSTERK2 was not activated (13). The results shown are for a representative experiment assayed in triplicate and are corrected for background counts (immunoprecipitations done with control antibodies). Error bars show standard deviations. One unit of activity represents mRaf-1 activity from cells deprived of serum. Similar results were obtained in three independent experiments. (B) Requirement of PKC for activation of Raf-1 by TPA and Hm1. mRaf-1 was expressed in COS cells alone or together with N17Ras and Hm1 as indicated. Where indicated, the cells were pretreated with the PKC inhibitor Ro 31-8220 (10 μM for 10 min). Similar results were obtained in four independent experiments. (C) Activation of membrane-targeted Raf-1 by TPA. R89LRaf and R89LRafCAAX (17) were expressed in COS cells that were deprived of serum (Cont) or treated with TPA (TPA). Similar results were obtained in two independent experiments.

The requirement of Ras in the activation of Raf-1 is consistent with observation that mutations in the Ras-binding domain (RBD) and Cys-rich domain of Raf-1 affect its activation by EGF and TPA (18). However, our results differ from those showing that R89LRaf-1 is activated by EGF and TPA (19). To further examine the role of Ras in Raf-1 activation by PKC, we investigated whether Ras–Raf-1 complexes form in cells treated with TPA. Endogenous Ras was immunoprecipitated from extracts of TPA-treated cells, using the rat monoclonal antibody Y13-238, which can immunoprecipitate Ras-GTP–Raf complexes from Ras-transformed or EGF-stimulated cells (20). The immunoprecipitates were analyzed for association of endogenous Raf-1 by immunoblotting. We also performed a reciprocal analysis by immunoprecipitating endogenous Raf-1. In cells treated with TPA, complexes were formed between Ras and Raf-1 (Fig.3A).

Figure 3

Formation of Ras–Raf-1 complexes. Cells deprived of serum (–) or treated with TPA (+) for 60 min were extracted in medium-salt buffers, and endogenous Ras or Raf-1 was immunoprecipitated (25). (A) Immunoblot analysis of Ras-Raf complexes. Upper panels show immunoblots for Raf-1, and lower panels show immunoblots for Ras. Lanes 1 and 2: 0.5% of total cell extract; lanes 3 and 4: immunoprecipitation of Ras; lanes 5 and 6: immunoprecipitation of Raf-1. Molecular size markers (in kilodaltons) are shown to the right. (B) Association of Raf-1 activity with Ras. Y13-238 immunoprecipitates from cells deprived of serum (–) or treated with TPA (+) were analyzed for Mek activators in the kinase cascade assay (17). Lanes 1 and 2: Ras-associated GSTMek activator. Lanes 3 and 4: Ras immunocomplexes treated with high salt concentrations to elute the Mek activator (25); the amount of activator remaining associated with the immunoprecipitated Ras after elution was measured by returning the beads to a kinase assay. Lanes 5 through 8: reprecipitation of the Mek activator from parallel samples with a monoclonal antibody to Raf-1 (lanes 5 and 6) or with the 9E10 monoclonal antibody as a control (lanes 7 and 8). The assays were done in triplicate, and similar results were observed in two independent experiments. CPM, counts per minute.

To determine whether the Raf-1 associated with Ras from TPA-stimulated cells was active, we tested whether the Y13-238 immunoprecipitates could activate a glutathione-S transferase (GST)–tagged version of the kinase activator of ERK, Mek-1 (GSTMek-1). Ras immunoprecipitates from TPA-stimulated cells activated GSTMek-1 (Fig. 3B). Approximately 65% of this activity could be eluted by treating the immunoprecipitates with buffers containing high concentrations of salt; of this, 50% could be reprecipitated with a monoclonal antibody to Raf-1 (Fig. 3B). Thus, Raf-1 from TPA-stimulated cells forms a complex with Ras and is activated.

Formation of complexes between Ras and Raf-1 requires that Ras be in the GTP-bound form (20, 21). Therefore, we examined whether activation of PKC leads to increased amounts of Ras-GTP by monitoring association between Ras-GTP and the Ras-binding domain of Raf-1 (22, 23). Small amounts of Ras-GTP were detected in cells deprived of serum (Fig. 4A). Treatment of cells with TPA or EGF activated Ras to similar extents, but the response was slower with TPA, continuing to rise for 40 min after stimulation (Fig. 4A). The activation of Ras in response to TPA was blocked by the PKC inhibitor Ro 31-8220, whereas the activation in response to EGF was not (Fig. 4B). Stimulation of the Hm1 receptor also activated Ras in a PKC-dependent manner (Fig. 4B). H-Ras, N-Ras, and K-Ras were all activated in cells treated with TPA (13). The kinetics of the increases in Ras-GTP formation following TPA treatment were similar to those of increases in Raf-1 activity (Fig. 4C).

Figure 4

Activation of Ras in cells treated with TPA or EGF. (A) Time course of Ras-GTP formation after exposure to TPA or EGF. Cells deprived of serum were treated with TPA or EGF for the indicated times, and the amount of Ras-GTP was measured (23). Upper panels show Ras-GTP; lower panels show total amount of Ras in 10% of the extract. (B) Inhibition of Ras activation by the PKC inhibitor Ro 31-8220. Cells deprived of serum (Cont) were stimulated with EGF or TPA, or cells expressing Hm1 were treated with carbachol (Carb); the amount of Ras-GTP was then determined. Where indicated, the cells were treated with Ro 31-8220 before stimulation (10 μM for 10 min). The levels of Ras-GTP are shown in the upper panels, and the total Ras levels in 10% of the extracts are shown in the lower panels. Molecular size markers (in kilodaltons) are shown to the right. (C) Activation of endogenous Raf-1 by TPA. Cells deprived of serum were treated with TPA (40 nM) for the times indicated, and the stimulation of the endogenous Raf-1 activity was measured. Similar results were seen in two independent experiments. Endogenous Raf-1 was immunoprecipitated with a monoclonal antibody (Transduction Laboratories).

Our data show that Raf-1 activation by signals that activate PKC is a consequence of Ras activation and the formation of Ras-GTP–Raf-1 complexes. Although Raf-1 activation by PKC is mediated through Ras activation, it differs from activation by receptor tyrosine kinases in that it is not blocked by N17Ras. This observation emphasizes that the absence of an effect of N17Ras expression cannot be used to conclude that Ras is not part of a signaling pathway. Rather, our results indicate that Ras activation is a component of PKC signaling.

  • * To whom correspondence should be addressed. E-mail: chrism{at}icr.ac.uk

REFERENCES AND NOTES

View Abstract

Navigate This Article