(Received for publication, December 31, 1996, and in revised form, March 25, 1997)
From the Department of Internal Medicine (1st Division), Kobe University School of Medicine, Kobe 650, Japan and the § Molecular Cardiology Unit, Research Institute of Angiocardiology and Cardiovascular Clinic, Kyushu University School of Medicine, Fukuoka 812-82, Japan
Angiotensin II (ANG II), a potent hypertrophic factor of vascular smooth muscle cells (VSMC), induces activation of the ras protooncogene product (Ras) and mitogen-activated protein (MAP) kinases and subsequent stimulation of protein synthesis in VSMC. In the present study, we examined whether Ras activation is required for ANG II-induced MAP kinase activation and stimulation of protein synthesis in cultured rat VSMC. Pretreatment with tyrosine kinase inhibitors, genistein and herbimycin A, or a putative phosphatidylinositol 3-kinase inhibitor, wortmannin, completely blocked ANG II-induced Ras activation, whereas neither of them had an effect on ANG II-induced MAP kinase activation. Adenovirus-mediated expression of a dominant negative mutant of Ha-Ras completely inhibited ANG II-induced Ras activation but failed to inhibit MAP kinase activation and stimulation of protein synthesis by this vasoconstrictor. These results indicate that ANG II stimulates MAP kinases and protein synthesis by a Ras-independent pathway in VSMC.
ANG II,1 the main peptide hormone of the renin-angiotensin system, has been known to play an important role in the growth of VSMC in addition to its key regulatory role in the regulation of blood pressure and circulating volume (1). Because this growth-promoting effect of ANG II is considered to contribute to the development of various cardiovascular diseases characterized by VSMC growth such as hypertension, atherosclerosis, and restenosis following balloon angioplasty (2-4), it is important to define the signaling pathways of ANG II that mediate the growth response of VSMC.
ANG II acts via a high affinity cell surface receptor called the AT1 receptor. Although this receptor is a seven-transmembrane, heterotrimeric G protein-coupled receptor, some of the intracellular signals mediated by the AT1 receptor are similar to the signaling pathways activated by receptor tyrosine kinases such as platelet-derived growth factor and epidermal growth factor receptors. For instance, ANG II induces tyrosine and threonine phosphorylation and activation of MAP kinases (5, 6) and stimulates expression of early growth response genes such as c-fos, c-jun, and c-myc (7-10). ANG II, however, subsequently induces cell hypertrophy as a result of increased protein synthesis rather than cell proliferation in cultured VSMC (11, 12).
Recently, we and others have shown that ANG II induces activation of the ras protooncogene product (Ras), a membrane-bound GTPase with a relative molecular mass of 21 kDa, via the AT1 receptor in cultured VSMC (13-15). Because Ras plays a pivotal role in signaling from the tyrosine kinase receptors of various growth factors to the MAP kinase pathway (16-18), this finding tempts us to speculate that Ras activation may be involved in the mechanism by which ANG II induces MAP kinase activation in VSMC. However, recent studies in various cells suggest that the relative contribution made by Ras to the activation mechanisms of the MAP kinase pathway by heterotrimeric G protein-coupled receptors varies for different agonists and in different cell types (18, 19). Moreover, it has not yet been clarified whether Ras activation is necessary for hypertrophic response of VSMC to ANG II.
In the present study, we assessed the requirement of Ras activation for the ANG II-induced MAP kinase activation and stimulation of protein synthesis in cultured VSMC. The results clearly indicate that Ras activation is dispensable not only for ANG II-induced MAP kinase activation but also for stimulation of protein synthesis in VSMC.
ANG II and MBP were obtained from Sigma. The
anti-Ras rat monoclonal antibody Y13-259 was purchased from Oncogene
Science (Cambridge, MA). L-[4,5-3H]leucine
(139 Ci/mmol) and [-32P]ATP (3000 Ci/mmol) were from
Amersham Life Science (Tokyo, Japan). [32P]orthophosphate
was from Du Pont. Protein G-Sepharose 4 Fast Flow and rabbit
antiserum to rat IgG were from Pharmacia (Uppsala, Sweden)
and Cappel (Durham, NC), respectively. Genistein and wortmannin were
from Kyowa (Tokyo, Japan). Herbimycin A was from Life Technologies, Inc. Other materials and chemicals were obtained from commercial sources.
VSMC were isolated from rat thoracic aorta by enzymatic dissociation as described previously (20). Cells were grown and passaged as described previously (21) and used at passage levels 7-18.
Analysis of Ras-bound GDP and GTPDetection and quantification of Ras-bound GDP and GTP was performed as described previously (13). Briefly, the quiescent VSMC on 60-mm dishes were incubated with phosphate-free DMEM supplemented with 0.2 mCi/ml [32P]orthophosphate for 12 h. Fifty µM sodium orthovanadate was added to the cells during the last 30 min. After stimulation with ANG II, the cells were lysed. Ras was immunoprecipitated from the cell lysates with anti-Ras monoclonal antibody Y13-259 and analyzed by thin layer chromatography. The radioactivity was analyzed using a Fujix bioimaging analyzer BAS2000.
MAP Kinase AssayMAP kinase activity was measured by a MAP
kinase renaturation assay in MBP-containing polyacrylamide gels as
described previously (13). Briefly, the cell lysates from VSMC were
electrophoresed on 10% SDS-polyacrylamide gel containing 0.5 mg/ml
MBP. After washing the gel, the enzymes were denatured in 6 M guanidine HCl and then renatured in 50 mM
Tris-Cl, pH 8.0, containing 5 mM 2-mercaptoethanol and
0.04% Tween 40. Kinase reaction was carried out by incubating the gel
with [-32P]ATP. After incubation, the gel was
extensively washed and dried. The radioactivity was analyzed using a
Fujix bioimaging analyzer BAS2000.
Replication-defective E1 and
E3
adenoviral vectors containing CA promoter comprising a
cytomegalovirus enhancer and chicken
-actin promoter were prepared
as described previously (22-24). The adenoviruses expressing either a
dominant negative mutant of Ha-Ras (AdRasY57), in which tyrosine
replaces aspartic acid at residue 57 (25), or the bacterial
-galactosidase (AdLacZ) were constructed as described previously
(24). Subconfluent VSMC grown on 60-mm dishes were incubated with DMEM
containing either AdRasY57 or AdLacZ (20 plaque-forming units/cell) for
2 h at room temperature and then washed once with 2 ml of DMEM and incubated with DMEM supplemented with 10% fetal bovine serum for more
2 days. Then cells were growth-arrested for 48 h in DMEM prior to
use.
Protein synthesis was measured by [3H]leucine incorporation as described previously (26). The quiescent VSMC were stimulated with ANG II in serum-free DMEM containing 0.5 µCi/ml [3H]leucine. After 24 h, the radioactivity incorporated into trichloroacetic acid-precipitable material was measured by liquid scintillation spectrometry after solubilization in 0.1 N NaOH.
Protein DeterminationCell protein was determined by the method of Bradford (27) with bovine serum albumin as a standard.
Since a number of investigators have
shown that tyrosine kinase activity is essential for Ras and/or MAP
kinase activation by various stimuli in their respective target cells
(16, 19), we examined whether tyrosine kinase activity was required for ANG II-induced Ras and MAP kinase activation, using two distinct tyrosine kinase inhibitors, genistein and herbimycin A. As has been
previously reported (13), the unstimulated VSMC displayed a low level
of Ras-GTP (Fig. 1). ANG II caused an approximately 3-fold increase in the accumulation of Ras-GTP in VSMC. Ras activation by ANG II was prevented completely by pretreatment with genistein (100 µM) or herbimycin A (3 µM). Using a MAP
kinase renaturation assay in MBP-containing polyacrylamide gels, we
investigated whether genistein and herbimycin A also inhibited ANG
II-induced MAP kinase activation (Fig. 2). These two
tyrosine kinase inhibitors, however, had no effect on ANG II-induced
MAP kinase activation.
Effects of Wortmannin on ANG II-induced Ras and MAP Kinase Activation in VSMC
Recent studies have revealed that
phosphatidylinositol 3-kinase is involved in receptor-mediated
activation of Ras/MAP kinase pathway in some cell types (28-32). We
next tested the effect of wortmannin, which is a fungal metabolite that
has been characterized as an inhibitor of phosphatidylinositol 3-kinase
at less than 100 nM (33, 34) on ANG II-induced Ras and MAP
kinase activation. Wortmannin (50 nM) completely blocked
the stimulatory effect of ANG II on Ras activation (Fig.
3A). In contrast, this concentration of
wortmannin failed to inhibit ANG II-induced MAP kinase activation (Fig.
3B).
Effects of a Dominant Negative Mutant of Ras on ANG II-induced Ras and MAP Kinase Activation and Stimulation of Protein Synthesis in VSMC
The results of experiments using the tyrosine kinase
inhibitors and wortmannin strongly suggest that ANG II stimulates MAP kinases by a Ras-independent pathway in VSMC. Because these
pharmacological drugs may affect other reactions besides Ras
activation, we used a more specific molecular tool, the adenovirus
expressing a dominant negative mutant of Ha-Ras (AdRasY57) in the next
set of experiments. The adenovirus expressing the bacterial
-galactosidase (AdLacZ) was used as a control. As has been described
(24), AdRasY57-infected VSMC expressed markedly elevated levels of Ras
protein as compared with both uninfected and AdLacZ-infected control
cells (data not shown). As shown in Fig. 4, expression
of RasY57 decreased the basal levels of Ras-GTP and completely
prevented the stimulatory effect of ANG II on Ras activation.
Expression of
-galactosidase had no effect on basal levels of
Ras-GTP and ANG II-increased Ras-GTP accumulation. In consistent with
the results of the pharmacological experiments described above,
expression of RasY57 had no effect on ANG II-induced MAP kinase
activation (Fig. 5A). Moreover, even in
AdRasY57-infected VSMC, ANG II stimulated protein synthesis to extents
similar to those observed in uninfected and AdLacZ-infected control
cells (Fig. 5B). In all experiments so far described, we
used a high concentration (100 nM) of ANG II to induce
maximal responses of VSMC. It is possible that Ras activation may
contribute to MAP kinase activation and stimulation of protein
synthesis by lower concentrations of ANG II and that an alternative
Ras-independent pathway(s) may be utilized at higher concentrations of
ANG II. Therefore, in the last set of experiments, we used a lower
concentration (1 nM) of ANG II and examined the effects of
expression of RasY57 on Ras and MAP kinase activation and stimulation
of protein synthesis. Expression of RasY57 also inhibited Ras
activation by 1 nM ANG II (Fig.
6A) but failed to prevent MAP kinase
activation and stimulation of protein synthesis even by this low
concentration of ANG II (Fig. 6, B and C).
In 1992, we showed that ANG II induces tyrosine and threonine phosphorylation and activation of MAP kinases in cultured VSMC (5). ANG II also induces activation of the raf protooncogene product (Raf) and MAP kinase kinase in this cell type (35, 36). Recently, several investigators including us have clearly demonstrated that ANG II induces activation of Ras via AT1 receptors in VSMC (13-15). In our study (13), however, we found that pertussis toxin treatment, which markedly inhibited ANG II-induced Ras activation, had no inhibitory effect on ANG II-induced MAP kinase activation. This differential sensitivity to pertussis toxin of these reactions suggests that Ras activation may be dispensable for ANG II-induced MAP kinase activation in VSMC. However, since the pertussis toxin treatment did not completely inhibit ANG II-induced Ras activation, we could not rule out the possibility that ANG II was activating MAP kinases via residual Ras activity. The results reported herein extended these observations and showed that treatment with two tyrosine kinase inhibitors (genistein and herbimycin A) or wortmannin, which completely inhibited ANG II-induced Ras activation, had no effect on ANG II-induced MAP kinase activation. Further, we showed that expression of a dominant negative mutant of Ha-Ras, which completely blocked ANG II-induced activation of endogenous Ras, presumably by interfering with exchange of GDP for GTP (25), had no effect on ANG II-induced MAP kinase activation. These results indicate that ANG II induces MAP kinase activation by a Ras-independent pathway in VSMC.
Previous reports showed that depletion of protein kinase C by pretreatment with a protein kinase C-activating phorbol ester, phorbol 12-myristate 13-acetate, markedly blocked ANG II-induced MAP kinase activation, suggesting that MAP kinase activation by ANG II is mediated mainly by protein kinase C activation (5, 37). Protein kinase C is known to activate Raf by direct phosphorylation (18). Indeed, protein kinase C-activating phorbol 12-myristate 13-acetate can stimulate MAP kinases without activating Ras in VSMC (5, 13). Based on the present results and these previous findings, it is likely that ANG II induces MAP kinase activation predominantly by a Ras-independent and protein kinase C-dependent pathway in VSMC. Recently, Eguchi et al. (15) also reported that ANG II induced Ras and MAP kinase activation in cultured rat VSMC. In their system, however, ANG II-induced MAP kinase activation was only partially impaired by pretreatment with phorbol 12-myristate 13-acetate but was abolished by treatment with an intracellular Ca2+ chelator, TMB-8, a calmodulin inhibitor, calmidazolium, and genistein. Furthermore, both ANG II-induced MAP kinase activation and Ras activation were insensitive to pertussis toxin treatment in their study. Based on these findings, they proposed that ANG II induces MAP kinase activation via Ras activation, which is mediated by an unidentified Ca2+/calmodulin-dependent tyrosine kinase in VSMC. The reason for the discrepancies between our results and theirs is unclear at present, but the discrepancies may be due to differences in phenotypes of our cells and theirs, because it is well known that various degrees of dedifferentiation occur when VSMC are placed in culture (38).
The mechanism by which ANG II activates Ras via the AT1 receptor and
Gi protein remains unclear. The inhibitory effect of tyrosine kinase inhibitors on Ras activation by ANG II suggests that
tyrosine phosphorylation may be involved in this process. It has been
shown that activation of the ANG II receptor as well as other
Gi-coupled-receptors such as 2-adrenergic,
lysophosphatidic acid, and endothelin receptors induces tyrosine
phosphorylation of Shc, resulting in its association with a Grb2-Sos
complex (39). These observations suggest that Shc-Grb2-Sos complexes
may propagate signals not only from growth factor receptors but also
from Gi protein-coupled receptors including the ANG II
receptor as well. The pathway leading to Shc phosphorylation may
involve
subunits of Gi protein because
lysophosphatidic acid- and
2-adrenergic receptor-induced
Shc phosphorylation is blocked by coexpression of a
binding
peptide derived from
ARK1 (40). Recently, Schieffer et
al. (14) have suggested that ANG II activates Ras by inhibiting the GAP activity through tyrosine phosphorylation of GAP by c-Src in
VSMC. Therefore, it is possible that, to fully activate Ras, ANG II may
utilize two mechanisms simultaneously, i.e. stimulating the
exchange of GDP for GTP by Sos through tyrosine phosphorylation of Shc
and inhibiting the hydrolysis of bound GTP by GAP through tyrosine
phosphorylation of GAP. In the present study, we also showed that
wortmannin inhibited ANG II-induced Ras activation in VSMC. Similar
inhibitory effect of wortmannin on Ras activation was observed in
adipocytes stimulated with insulin (32). In this cell type, inhibition
of phosphatidylinositol 3-kinase with wortmannin resulted in
significant activation of GAP and reduction in Ras-GTP. Therefore, it
is possible that phosphatidylinositol 3-kinase may be involved in the
mechanism by which ANG II induces Ras activation via inhibiting GAP
activity.
Using a dominant negative strategy, Ras has been shown to be involved in stimulation of VSMC proliferation by serum, platelet-derived growth factor, acidic fibroblast growth factor, epidermal growth factor, and thrombin (24, 41). By the same strategy, we also explored the role of Ras in ANG II-induced hypertrophic response of VSMC. We found that expression of a dominant negative mutant of Ras did not inhibit ANG II-induced stimulation of protein synthesis, indicating that Ras activation is not required for ANG II-induced hypertrophic response of VSMC. It is becoming apparent that Ras may have multiple effectors besides Raf and play diverse roles in cell responses other than cell growth (16, 17). Therefore, it is possible that Ras activated by ANG II may play roles unrelated to MAP kinase activation and cell growth in VSMC. Further studies are needed to clarify the roles of Ras in ANG II actions as well as the mechanism by which ANG II activates Ras in VSMC.