Angiotensin II controls p21ras activity via pp60c-src.

Angiotensin II is the major effector peptide of the renin-angiotensin system, and it exerts its physiologic functions via a G protein-coupled cell surface receptor called AT1. We found that in rat aortic smooth muscle cells, angiotensin II stimulated the formation of Ras-GTP, Ras-Raf-1 complex formation, and the tyrosine phosphorylation of two important Ras GTPase-activating proteins (GAPs), p120 Ras-GAP and p190 Rho-GAP. Electroporation of anti-pp60c-src antibody into cultured, adherent smooth muscle cells blocked the angiotensin II stimulation of Ras-GAP and Rho-GAP tyrosine phosphorylation. In contrast electroporation of antibodies against c-Yes or c-Fyn had no effect. Anti-pp60c-src antibody also blocked angiotensin II-stimulated Ras activation and Ras-Raf-1 complex formation. These data strongly suggest that a G protein-coupled receptor such as the AT1 receptor can activate the Ras protein cascade via the tyrosine kinase pp60c-src.

Angiotensin II is the major effector molecule of the reninangiotensin system. This octapeptide stimulates vascular smooth muscle contraction, elevates vascular resistance, and increases intravascular volume. These effects combine to raise systemic blood pressure (1,2). There is also substantial experimental evidence that angiotensin II acts as a growth factor (3). For instance, cultured vascular smooth muscle cells respond to angiotensin II by expressing early growth response genes such as c-fos, c-jun, and c-myc and by increasing thymidine incorporation (4,5). This correlates with in vivo data that infusion of angiotensin II into animals injured with a vascular balloon catheter markedly exacerbates the resulting myoproliferative lesion (6). A role for angiotensin II in cell growth and tissue remodeling has also been shown in animal models of hypertension, heart failure, and atherosclerosis (7)(8)(9).
Angiotensin II exerts its physiologic functions via a high affinity cell surface receptor now called the AT 1 receptor. This receptor was first cloned in 1991, and it contains the structural features of a seven transmembrane, heterotrimeric G proteinassociated receptor (10). In vascular smooth muscle cells, ligand activation of the AT 1 receptor leads to the rapid activation of phospholipase C and the production of 1,4,5-inositol trisphosphate (11).
Recently it has become clear that many of the intracellular signals mediated by the AT 1 receptor are similar to the signaling pathways activated by receptor tyrosine kinases. For instance, ligand activation of the AT 1 receptor leads to the rapid tyrosine phosphorylation and activation of phospholipase C-␥1 in vascular smooth muscle cells (12). This is a critical event for downstream signaling, because inhibition of tyrosine phosphorylation markedly reduces angiotensin II stimulation of 1,4,5inositol trisphosphate production. Separate studies have also shown that angiotensin II, acting through the AT 1 receptor, stimulates JAK2 tyrosine phosphorylation and activation (13,14). Finally, several studies have indicated that angiotensin II leads to the phosphorylation and activation of mitogen-activated protein kinases (15,16). The known link between p21 ras activity and mitogen-activated protein kinase stimulation, as well as the central role of p21 ras in cell growth, prompted us to ask if this molecule is activated by angiotensin II. p21 ras and other small G proteins are membrane bound guanine nucleotide-binding proteins that are active when complexed with GTP and inactive when bound to GDP (17,18). The ratio of bound GTP to GDP is regulated by the rate of nucleotide exchange and by the rate of GTP hydrolysis. Proteins such as mammalian son-of-sevenless (mSOS) stimulate the exchange of GTP for bound GDP on p21 ras . The deactivation of p21 ras is regulated by GTPase-activating proteins, called GAPs, 1 which markedly accelerate intrinsic Ras GTPase activity and lead to Ras-GDP. The best defined GAP is Ras-GAP, a 120-kDa protein that binds p21 ras via a Src homology II domain (19,20). A second protein implicated in Ras deactivation is Rho-GAP, a 190-kDa protein (21). Although the exact role of Rho-GAP is not known, it is thought to have a regulatory role in stress fiber induction and focal adhesion (22). In cells stimulated with growth factors the GAPs are phosphorylated on tyrosine and form a complex that enhances the GTPase activity of p21 ras (23).
The biochemical pathway by which a G protein-coupled receptor such as the AT 1 receptor regulates p21 ras activity is not known. These cell surface receptors lack intrinsic kinase activity and must activate Ras in a fashion different from that of tyrosine kinase receptors. In this study we show that angiotensin II stimulates p21 ras , leading to p21 ras -GTP and Ras-Raf-1 complex formation. Angiotensin II also stimulates the tyrosine phosphorylation of Ras-GAP and Rho-GAP. The tyrosine kinase pp60 c-src appears to play a critical early role in angiotensin II signaling; neutralization of pp60 c-src activity by the electroporation of anti-pp60 c-src antibodies into smooth muscle cells blocked both Ras-GTP accumulation and Ras-Raf-1 complex * This work was supported by National Institutes of Health Grants DK39777, DK44280, and DK45215 as well as grants-in-aid from the American Heart Association and the Georgia affiliate of the American Heart Association. 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  formation in response to angiotensin II. This is the first observation that a G protein-coupled receptor such as the AT 1 receptor can activate the Ras protein cascade via pp60 c-src .

EXPERIMENTAL PROCEDURES
Materials-Human recombinant epidermal growth factor and all cell culture media were purchased from Life Technologies, Inc. Monoclonal antibodies against Rho-GAP (p190), Ras-GAP (p120), p21 ras , and phosphotyrosine (clone PY20) were obtained from Transduction Laboratories (Kensington, KY). Polyclonal anti-pp60 c-src , pp60c-src peptide, antic-Yes, c-Fyn, and monoclonal rat anti-p21 ras (Y13-259) antibodyagarose conjugate were obtained from Santa Cruz Biotech. (Santa Cruz, CA). [ 32 P]Orthophosphate was obtained from DuPont NEN. The enhanced chemiluminescence kit was obtained from Amersham Corp. Angiotensin II, goat anti-mouse IgG, and all other chemicals were purchased from Sigma.
Cell Culture-Rat vascular smooth muscle (RASM) cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine serum, 10 g/ml streptomycin, and 100 units/ml penicillin. Cells to be used for experiments were grown to 75-85% confluence, washed once with serum-free Dulbecco's modified Eagle's medium, and growth arrested in 5 ml of serum-free Dulbecco's modified Eagle's medium for 24 h prior to use. The cell number at this state of growth was approximately 10 7 cells/100-mm Petri dish (12,13).
Immunoprecipitation and Western Blotting-RASM cells were stimulated with 10 Ϫ7 M angiotensin II. The reaction was terminated by washing twice with phosphate-buffered saline and cells were lysed in 25 mM Tris, pH 7.4, 1% Nonidet P-40, 10% glycerol, 50 mM NaF, 10 mM NaP 2 O 4 , 137 mM NaCl, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, and 10 ng/ml aprotinin. The supernatant was precleared with Immunoprecipitin at 4°C. Protein concentration was determined in the cleared supernatant (24). To immunoprecipitate proteins from the cleared lysate, antibodies were added for 4 -8 h at 4°C. We used the following monoclonal antibodies: anti-Raf-1 (2 g/ml), anti-p21 ras (4 g/ml), anti-Ras-GAP (2 g/ml), anti-Rho-GAP (2 g/ml), or anti-phosphotyrosine (PY20 clone, 5 g/ml lysate). The immunoprecipitates were then recovered, separated on SDS-polyacrylamide gel electrophoresis, and visualized as described previously (12). The tyrosine phosphorylation of Ras-GAP and Rho-GAP was scanned by using a Silverscanner (La Cie, Inc.) interfaced with a personal computer. Each band was scanned in two dimensions, and the density was corrected for the background present in the lane.
Antibody Preabsorption-The anti-pp60 c-src antibody used for electroporation was raised against residues 509 -533 of the N-terminal site of human Src. (26). It was preabsorbed with a 10-fold excess of peptide for 2 h at 37°C in Hanks' balanced salt solution. The preabsorbed antiserum was immediately used for electroporation experiments. An aliquot of antibody was also sham absorbed without peptide.
Ras-GTP/Ras-GDP Analysis-Rat vascular smooth muscle cells were quiesced in serum and phosphate-free Dulbecco's modified Eagle's medium supplemented with 1 mM sodium pyruvate, 100 units/ml penicillin and 100 g/ml streptomycin. After 12 h, [ 32 P]orthophosphate was added (0.2 mCi/plate), and the cells were cultured for 6 h. Cells were stimulated with 100 ng/ml recombinant human epidermal growth factor (EGF) or 10 Ϫ7 M angiotensin II. Cells were then washed twice with ice-cold phosphate buffered saline, and p21 ras -associated nucleotides were isolated as described by Vaillancourt et al. (27). Immunoprecipitated nucleotides were eluted with 2 mM EDTA, 2 mM dithiothreitol, 0.2% SDS, 0.5 mM GTP, 0.5 mM GDP at 68°C for 20 min and separated by thin-layer chromatography on polyethyleneimine cellulose plates with fluorescence indicators run in 1.2 M ammonium formate and 0.2 M HCl. Nonradioactive GTP and GDP were used as standards. Radiolabeled nucleotides were quantitated by densitometry using a Silverscanner (LaCie, Inc.). The ratio of GTP/(GTPϩGDP) guanine nucleotides bound to p21 ras was calculated.

RESULTS
Angiotensin II Activation of p21 ras -Activation of p21 ras is dependent upon the formation of a Ras-GTP complex. To measure the effects of angiotensin II on the formation of this complex, cultured RASM cells were quiesced for 24 h and then loaded with [ 32 P]orthophosphate. After exposure to angiotensin II for 0, 1, 5, or 10 min, cells were lysed and immunoprecipitated with anti-p21 ras antiserum. The ratio of GTP to GDP nucleotides bound to p21 ras was then determined by thin-layer chromatography. Without angiotensin II, p21 ras is found associated with GDP (Fig. 1A, lane C 0 ). However, 1 min after the addition of 10 Ϫ7 M angiotensin II, there is a significant increase in the amount of Ras-GTP complex formation (Fig. 1A, lane A 1 ). Levels of Ras-GTP then slowly decline. An identical experiment was performed using 100 ng/ml EGF in place of angiotensin II (Fig. 1, lanes E 1 , E 5 , and E 10 ). As previously demonstrated, this growth factor stimulated Ras-GTP complex formation (27). In order to quantitate the shift of Ras from the GDP to the GTP bound forms, the ratio of GTP to the total of GTP plus GDP was determined by densitometry. The averages of five separate experiments are shown in Fig. 1B. These data show that although the angiotensin II-mediated stimulation of Ras-GTP (3-fold) was less than that observed with EGF (4.5-fold), both FIG. 1. Ras activation. RASM cells were labeled with [ 32 P]orthophosphate, and p21 ras was collected by immunoprecipitation. Associated GTP and GDP were analyzed by thin-layer chromatography. A, the binding of 32 P-labeled guanine nucleotides to p21 ras under unstimulated conditions (C 0 ) and after 1, 5, and 10 min of exposure to 10 Ϫ7 M angiotensin II (A 1 , A 5 , and A 10 ) or 100 ng/ml EGF (E 1 , E 5 , and E 10 ). B, data from five separate experiments were quantitated by densitometry. The results are plotted as the ratio of GTP to the total of GTP plus GDP. After 1 min, angiotensin II (f) induced a 3-fold increase of p21 ras -GTP, whereas EGF (q) induced an 4.5-fold increase.
agents induced maximal Ras-GTP complex formation by 1 min. At this time point, the angiotensin II-stimulated ratio of Ras-GTP to the total of Ras-GTP plus Ras-GDP ranged from 58 to 74% of that stimulated by EGF (mean response 66% Ϯ 11).
Activated p21 ras is known to recruit Raf-1 into a protein complex (28). If angiotensin II stimulated Ras-GTP formation, we then hypothesized that angiotensin II would also induce Ras-Raf-1 complex formation. To measure this, RASM cells were exposed to angiotensin II, lysed, and immunoprecipitated with anti-p21 ras antiserum. The immunoprecipitated proteins were probed by Western blot analysis using anti-Raf-1 antibody. This showed that after angiotensin II addition, there is a marked increase in Ras-Raf-1 complex formation (Fig. 2). As with Ras-GTP levels, the greatest Ras-Raf-1 association was present at 1 min, after which levels declined. We have also measured Ras-Raf-1 complex formation by stimulating cells with angiotensin II, immunoprecipitating with anti-Raf-1 antibody, and probing by Western blot analysis using anti-Ras antibody. This protocol gave identical results to those shown in Fig. 2.
GAP proteins play an intimate role in the Ras activationdeactivation cycle. Analysis of growth factor signaling has indicated that these proteins serve to limit the time Ras remains in the active Ras-GTP form. To investigate if angiotensin II stimulates the tyrosine phosphorylation of Rho-GAP and Ras-GAP, RASM cells were exposed to angiotensin II, lysed, and immunoprecipitated with an anti-phosphotyrosine antibody. A Western blot of the precipitated proteins was then probed with monoclonal antibodies against either Rho-GAP or Ras-GAP. In unstimulated cells, very little Ras-GAP was phosphorylated on tyrosine (Fig. 3A). However, 1 min after the addition of angiotensin II, Ras-GAP tyrosine phosphorylation increased to about 16-fold greater than at time 0. Even at 10 min, levels of Ras-GAP tyrosine phosphorylation remained elevated. By comparison, Rho-GAP showed less change in tyrosine phosphorylation levels after angiotensin II exposure (Fig. 3B). Densitometry of five experiments showed that on average the tyrosine phosphorylation of Rho-GAP increased 4-fold 1 min after angiotensin II addition. The tyrosine phosphorylation of Rho-GAP was also studied by immunoprecipitation with monoclonal anti-Rho-GAP followed by Western blot analysis using anti-phosphotyrosine. This experiment gave data virtually identical to that of Fig. 3B.
The Role of Src in Angiotensin II p21 ras Regulation-A central question in understanding the interaction of the AT 1 receptor with the p21 ras pathway is how a seven transmembrane receptor initiates phosphorylation of downstream signaling molecules. We have focused on the role of the soluble tyrosine kinase pp60 c-src , because previous work has demonstrated that this enzyme plays an important role in angiotensin II-induced signaling (25). To inactivate Src activity, we inserted polyclonal anti-Src antibody into cultured, adherent RASM cells using an electroporation technique previously shown effective in introducing immunoglobulins into these cells (25). We first asked if the neutralization of intracellular Src activity affected angiotensin II-stimulated phosphorylation of Ras-GAP and Rho-GAP. As control experiments, cells were electroporated in the presence of vehicle (Hanks' balanced salt solution), bovine serum albumin (BSA), or pooled rabbit IgG (Fig. 4A: E, EI, and  EB). None of these reagents affected the tyrosine phosphorylation of Ras-GAP 1 min after the addition of angiotensin II. Electroporation itself does appear to augment the angiotensin II-mediated tyrosine phosphorylation of Rho-GAP (Fig. 4A, E). In contrast to the results with BSA or pooled rabbit IgG, electroporation of rabbit anti-pp60 c-src antibody inhibited the angiotensin II-induced tyrosine phosphorylation of both Rho-GAP and Ras-GAP (Fig. 4A, ES). The specificity of the pp60 c-src antibody was controlled using two separate protocols. First, electroporation experiments were performed using antiserum absorbed with a 10-fold excess of Src peptide. The absorbed antiserum no longer prevents the angiotensin II-stimulated tyrosine phosphorylation of Ras-GAP or Rho-GAP (Fig. 4B,  EA). We have also used electroporation to test the effects of rabbit polyclonal antibodies directed against the Src family proteins c-Yes (pp62) and c-Fyn (pp59). As shown in Fig. 4C, neither anti-c-Yes (EY) nor anti-c-Fyn (EF) antibody blocked the angiotensin II-stimulated tyrosine phosphorylation of Ras-GAP or Rho-GAP. Western analysis had previously demonstrated that RASM cells lack c-Yes but contain c-Fyn (data not shown). These experiments suggest a major role for pp60 c-src in angiotensin II-mediated phosphorylation of Rho-GAP and Ras-GAP.
We also studied the role of pp60 c-src in Ras-GTP formation. When cells were electroporated in the presence of vehicle, they responded to angiotensin II or to EGF with the rapid conversion of Ras-GDP to Ras-GTP, very similar to nonelectroporated cells (Fig. 5A, E). When cells were electroporated with either rabbit IgG or BSA, these cells responded to angiotensin II or EGF in a fashion indistinguishable from cells electroporated with vehicle alone (Fig. 5A, EI and EB). In contrast, when cells were electroporated with rabbit anti-pp60 c-src antiserum, Ras activation in response to angiotensin II was abolished (Fig. 5A,  ES). This was not a toxic effect because these cells remained fully capable of activating Ras in response to EGF.
We also used the electroporation technique to investigate the role of pp60 c-src on angiotensin II-mediated formation of a Ras-Raf-1 complex. Consistent with the above data, anti-pp60 c-src antibody completely blocks Ras-Raf-1 complex formation (Fig.  5B, ES). No effect on angiotensin II-mediated complex formation was observed with either rabbit IgG or BSA (Fig. 5B, EI  and EB). These data strongly suggest that pp60 c-src or a highly related enzyme plays a critical role in angiotensin II-mediated activation of the Ras signaling pathway.

DISCUSSION
The AT 1 receptor is a seven transmembrane receptor responsible for virtually all of the physiologic actions of angiotensin II (1,2). Whereas the process of angiotensin II-mediated smooth muscle contraction has been intensely studied, the intracellular signals associated with angiotensin II-mediated cell growth are less understood. Previous investigators have established that angiotensin II activates the mitogen-activated protein kinase cascade (15,16). Here we show that angiotensin II stimulates the activation of p21 ras (Ras). This was first established by directly verifying the conversion of Ras-GDP to Ras-GTP and by demonstrating Ras-Raf-1 association. Thus, angiotensin II appears to use the Ras pathway to activate mitogen-activated protein kinase in a fashion analogous to growth factors . Cells were either harvested unstimulated (C 0 ) or treated for 1 min with either EGF (E 1 ) or angiotensin II (A 1 ). Ras-associated nucleotides were isolated and analyzed as described in Fig. 1. The bar graph plots the ratio of GTP compared with the total of GTP plus GDP. The data are given as mean Ϯ S.D. (n ϭ 3) for each time point. Anti-pp60 c-src antibody inhibits the angiotensin II-stimulated increase of p21 ras -GTP but has no effect on the EGF stimulation of p21 ras -GTP. B, RASM cells were electroporated as described for A. Angiotensin II was added for 0 or 1 min, and Ras-Raf-1 complex formation was measured as described in Fig. 2. Anti-pp60 c-src antibody inhibits the angiotensin II-stimulated formation of this complex. such as EGF. That said, there must be differences in the signaling initiated by angiotensin II and by classic growth factors because angiotensin II is a less potent growth factor than EGF or similar molecules. In analyzing the Ras activation in response to angiotensin II, we noted two differences from the Ras activation in response to EGF. The magnitude of angiotensin II-induced Ras-GTP formation was less, and Ras was more rapidly inactivated. A major intracellular mechanism used to regulate Ras activity is the action of Ras-GTPase proteins or GAPs. In response to angiotensin II, there is a rapid and marked tyrosine phosphorylation of Ras-GAP. In association with Rho-GAP, this molecule converts the active Ras molecule back to the Ras-GDP form (17). Thus, we hypothesize that the rapid activation of the Ras-GAP system by tyrosine phosphorylation acts to modulate the stimulatory potential of angiotensin II.
Another obvious difference between angiotensin II signaling and that of classic growth factors is that the AT 1 receptor lacks intrinsic kinase domains (10). Thus, the AT 1 receptor must recruit an intracellular kinase to initiate any type of kinase cascade. Recent experimental evidence indicates that pp60 c-src is activated by angiotensin II (29). In vascular smooth muscle cells it appears to play a major role in angiotensin II-mediated phosphorylation of phospholipase C-␥1 and 1,4,5-inositol trisphosphate generation (25). In response to growth factors, Chang and colleagues (30) showed that pp60 c-src exerts a signaling effect upstream of Ras. To investigate the role of pp60 c-src in stimulation of Ras, we established an electroporation protocol to insert anti-pp60c-src antibodies into cells to examine if this blocks angiotensin II-mediated Ras activation. This technique uses an electrode that allows electroporation of cultured cells while still attached to a tissue culture dish; it is efficient and gentle in that cells need not be trypsinized either before or after electroporation (25).
Anti-pp60 c-src antibody blocked the angiotensin II-mediated conversion of Ras-GDP to the active Ras-GTP form. This appeared to be a specific effect for angiotensin II because EGF activated Ras in the presence of the antibody. These data clearly support a role for pp60 c-src upstream of angiotensin II-mediated Ras activation. Whether pp60 c-src directly activates a nucleotide exchange protein complex or whether it activates Ras in some indirect fashion remains unknown. Potential indirect mechanisms include the release of intracellular Ca 2ϩ and the tyrosine phosphorylation of linkers such as SHC and IRS-1 (31,32). Experiments with the anti-pp60 c-src antibody also show that Src plays a role in Ras-GAP and Rho-GAP tyrosine phosphorylation. As with the activation of Ras, the precise mechanism by which Src leads to the tyrosine phosphorylation of GAPs is unknown. Both Ras-GAP and Rho-GAP have been shown to be substrates of Src family kinases (19,33).
Thus, in conclusion, angiotensin II activates the Ras pathway in vascular smooth muscle cells. This is consistent with its known role as a growth factor for these cells. In response to angiotensin II, there is the rapid tyrosine phosphorylation of GAP proteins which serve to regulate the active Ras complex.
Critical to angiotensin II-mediated activation of the Ras pathway is the intracellular kinase pp60 c-src . This enzyme plays a role upstream of p21 ras activation. It also leads to the tyrosine phosphorylation of GAPs perhaps through direct pp60 c-src kinase activity.