Protein Kinase C-ζ Mediates Angiotensin II Activation of ERK1/2 in Vascular Smooth Muscle Cells

Activation of 44 and 42 kDa extracellular signal-regulated kinases (ERK)1/2 by angiotensin II (angII) plays an important role in vascular smooth muscle cell (VSMC) function. The dual specificity mitogen-actived protein (MAP) kinase/ERK kinase (MEK) activates ERK1/2 in response to angII, but the MEK activating kinases remain undefined. Raf is a candidate MEK kinase. However, a kinase other than Raf appears responsible for angII-mediated signal transduction because we showed previously that treatment with 1 μM phorbol 12,13-dibutyrate (PDBU) for 24 h completely blocked Raf-Ras association in VSMC but did not inhibit activation of MEK and ERK1/2 by angII. We hypothesized that an atypical protein kinase C (PKC) isoform, which lacks a phorbol ester binding domain, mediated ERK1/2 activation by angII. Western blot analysis of rat aortic VSMC with PKC isoform-specific antibodies showed PKC-α, -β1, -δ, -ε, and -η in relative abundance. All isoforms except PKC-η were down-regulated by 1 μM PDBU for 24 h suggesting that PKC-η was responsible for angII-mediated ERK1/2 activation. In response to angII, PKC-η associated with Ras as shown by co-precipitation of PKC-η with anti-H-Ras antibody. To characterize further the role of PKC-η, PKC-η protein was depleted specifically by transfection with antisense PKC-η oligonucleotides. Antisense PKC-η oligonucleotide treatment significantly decreased PKC-η protein expression (without effect on other PKC isoforms) and angII-mediated ERK1/2 activation in a concentration-dependent manner. In contrast, ERK1/2 activation by platelet-derived growth factor and phorbol ester was not significantly inhibited. These results demonstrate an important difference in signal transduction by angII compared with PDGF and phorbol ester in VSMC, and suggest a critical role for PKC-η and Ras in angII stimulation of ERK1/2.

Angiotensin II (angII) 1 plays an important role in hyper-trophic and hyperplastic growth of vascular smooth muscle cells (VSMC) (1,2). It not only rapidly increases intracellular calcium and activates protein kinase C (PKC) but also stimulates many of the same signal transduction events as growth factors, including protein-tyrosine phosphorylation (3), stimulation of c-fos (4), and activation of mitogen-activated protein (MAP) kinases or extracellular-regulated signal kinases (ERK) (5). ERK1/2 are a family of serine/threonine protein kinases activated as an early response to a variety of stimuli involved in cellular growth, transformation, and differentiation. It appears likely that ERK1/2 activation is required for many of the effects of angII on gene expression, such as induction of c-fos and c-myc (6). Stimulation of ERK1/2 requires phosphorylation of a dual specific protein kinase, MAP kinase kinase or MEK, which is itself regulated by MEK kinase and/or Raf kinase. It has been suggested that Raf is phosphorylated in response to angII in mesangial cells (7), and Raf phosphorylation is potentially regulated by PKC (8). However, our previous experiments strongly indicated that Raf may not be the predominant MEK kinase responsible for angII stimulation of ERK1/2 (9). Specifically, we showed that angII-stimulated ERK1/2 activation was not inhibited by PKC down-regulation (1 M PDBU for 24 h) while both Raf association with Ras, and Raf activation by angII, were inhibited by PKC down-regulation. In addition, angII-stimulated MEK kinase activity was significantly greater in Ras immunoprecipitates than in Raf immunoprecipitates. These results imply that a kinase other than Raf may be required for angII-mediated signal transduction via ERK1/2.
Several findings suggest that PKC-may act as a MEK kinase. PKChas been shown to activate MEK kinase and ERK1/2 in vitro (10) and in vivo (11). In addition, PKC isoforms are serine/threonine kinases like Raf, and the protein structure of PKCclosely resembles c-Raf-1 (12). Within the PKC family, PKCrepresents an atypical PKC isoform in that it lacks the C2 domain making its kinase activity Ca 2ϩ -independent, and it possesses only one zinc finger region in its regulatory domain (12). Consequently, PKCdoes not bind Ca 2ϩ and cannot be activated by diacylglycerol or phorbol esters (13). In addition, prolonged treatment with phorbol esters does not down-regulate PKC- (12,14), and most PKC inhibitors do not decrease PKCactivity (15).
In the present study, we investigated the role of PKCin agonist-mediated ERK1/2 activation. The results show that PKCassociates with Ras in response to angII, and PDGF. Antisense PKColigonucleotides decreased ERK1/2 activation by angII but had no significant effect on ERK1/2 activation by PDGF and PMA. These findings demonstrate for the first time a novel pathway for angII stimulation of ERK1/2 that is separate from PDGF and PMA, defined by a requirement for PKCassociation with Ras. * This work was supported by grants from the National Institutes of Health. 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

MATERIALS AND METHODS
Cell Culture-VSMC were isolated from 200 -250 g male Harlan Sprague Dawley rats and maintained in 10% calf serum/Dulbecco's modified Eagle's medium (DMEM) as described previously (16). Passage 5 to 13 VSMC at 70 -80% confluence in 100-mm dishes were growth arrested by incubation in 0.1% calf serum/DMEM for 48 h prior to use. Neonatal human VSMC, kindly provided by Drs. E. Raines and R. Ross at the University of Washington, were maintained in DMEM/ F-12 supplemented with 10 M TES, 50 g/ml ascorbic acid, 10 g/ml insulin, 10 g/ml transferrin, 10 ng/ml sodium selenite, 30 g/ml endothelial cell growth supplement, and 10% fetal bovine serum. Cells were growth arrested by incubation in DMEM medium with 1% plateletdepleted serum for 48 h.
Western Blot Analysis-After treatment, the cells were washed with phosphate-buffered saline (PBS), and 0.5 ml of TME lysis buffer (10 mM Tris, pH 7.5, 5 mM MgCl 2 , 1 mM EDTA, 25 mM NaF) containing fresh 100 M Na 3 VO 4 , 20 g/ml leupeptin, 1 g/ml pepstatin A, 4 g/ml aprotinin, and 1 mM DTT was added (17). Cell lysates were prepared by freezing, thawing on ice, scraping, Dounce homogenization (30 strokes), sonication for 1 s, and centrifugation for 30 min at 15,000 ϫ g. Protein concentration in the supernatant was determined by Bradford protein assay, and the samples were stored at Ϫ80°C. For Western blot analysis, 20 g of protein was subjected to SDS-PAGE under reducing conditions, and proteins were then transferred to nitrocellulose (Hybond TM -ECL, Amersham) as described previously (18). The membrane was blocked for 2 h at room temperature with a commercial blocking buffer from Life Technologies, Inc. The blots were incubated for 1 h at room temperature with the primary antibodies (PKC isoform-specific and c-Raf-1 antibodies from Santa Cruz Biotechnology, and H-Ras antibody from Boehringer Mannheim) followed by incubation for 1 h with secondary antibody (horseradish peroxidase-conjugated). Immunoreactive bands were visualized using chemiluminescence (ECL, Amersham International plc., United Kingdom).
Analysis of PKC-Association with Ras-VSMC were lysed with TME lysis buffer, and the lysates were subjected to immunoprecipitation with anti-H-Ras antibody. Immune complexes were recovered by the addition of protein A-agarose (Life Technologies, Inc.), incubation overnight at 4°C, and centrifugation. The beads were washed once with TME buffer, twice with TTBS buffer (20 mM Tris, pH 7.5, 500 mM NaCl, 1% Triton X-100, and 0.1% ␤-mercaptoethanol), and once with TME buffer. Immunoprecipitated proteins were then electrophoresed on a 9% SDS-polyacrylamide gel, transferred to nitrocellulose, and proteins identified by ECL.
Transfection Protocol for Antisense PKC-Oligonucleotides-Human vascular smooth muscle cells were plated in 6-well tissue culture dishes with DMEM/F-12 containing 10 M TES, 50 g/ml ascorbic acid, 10 g/ml insulin, 10 g/ml transferrin, 10 ng/ml sodium selenite, 30 g/ml endothelial cell growth supplement, and 10% fetal bovine serum at 5 ϫ 10 5 cells/well and grown overnight in a CO 2 incubator at 37°C to 70% confluence. Cells were washed once with pre-warmed PBS solution and once with Opti-MEM medium (Life Technologies, Inc.). A series of antisense oligonucleotides directed against the PKC-were screened, and the most active sequences were identified as described (19). The PKC-antisense oligonucleotide sequence was GACGCACGCGGCCT-CACACC, and the scrambled oligonucleotide sequence was AAGCGCG-CACCAGCGCCTCC. A complex of LipofectAMINE and oligonucleotides (2.5 g/100 nM) in Opti-MEM was added directly to cells at a final concentration of 1000 nM oligonucleotide and incubated for 6 h at 37°C (18). The transfection media was removed, and cells were washed once with PBS and refed with complete media. The cells were growth arrested by incubation in DMEM with 1% platelet-depleted serum for 48 -96 h prior to agonist stimulation and prepared as described above for measuring protein expression and ERK1/2 activity. Preliminary experiments demonstrated that maximal depletion of PKC isoforms occurred at 96 h, consistent with the half-life of PKC (about 24 h) (20) and previous studies with antisense PKC oligonucleotides (21). ERK 1 ⁄2 Kinase Assays-A myelin basic protein in-gel-kinase assay to measure ERK1/2 phosphotransferase activity was performed exactly as described previously (18). ERK1/2 activity was measured by densitometry of autoradiograms (in the linear range of film exposure) using NIH Image Version 1.59.
Statistical Analysis-Data are presented as mean Ϯ S.E. for all experiments that were performed at least three times. Significant differences were determined by Student's t test (p Ͻ 0.05).

Effects of PDBU Treatment on ERK1/2 Activation and PKC Isoform Expression in VSMC-
The goal of this study was to determine the role of specific PKC isoforms in activation of ERK1/2 by angII compared with PMA and PDGF. PKC has been suggested to be both "upstream" and "downstream" of ERK1/2 in signal transduction cascades (22, 23). To investigate which PKC isoforms were required for agonist-mediated activation of ERK1/2, phorbol ester-responsive PKC isoforms were down-regulated by PDBU (1 M for 24 h), cells were stimulated for 5 min with 100 nM angII, 10 ng/ml PDGF, and 200 nM PMA, and ERK1/2 activity was determined by in-gel-kinase assay. All three agonists increased ERK1/2 activity (Fig. 1A, left). PDBU treatment caused no significant decrease in angII-stimulated ERK1/2 activity (Fig. 1A, right; 89 Ϯ 11% of control at 5 min, n ϭ 11, p Ͼ 0.1 versus control). In contrast, there was Ͼ70% inhibition of PDGF-and PMA-stimulated ERK1/2 activity (n ϭ 5 and 8, respectively, p Ͻ 0.01). These results suggest that the classical and novel PKC isoforms, which are phorbol ester-responsive, are required for PDGF-and PMA-mediated ERK1/2 activation in VSMC. In contrast, if a PKC isoform is required for angII-mediated ERK1/2 activation, it must be an atypical isoform which is phorbol ester-unresponsive.
PDBU Treatment Inhibits angII-stimulated Ras-Raf Associ- 1. PDBU treatment effect on ERK1/2 activity and expression of PKC isoforms. A, growth-arrested VSMC were stimulated with 10 ng/ml PDGF, 100 nM angII, or 200 nM PMA for 5 min, cells were harvested, and lysates were analyzed for ERK1/2 activity by in-gelkinase assay. Arrows indicate the molecular masses of ERK1/2 (44 and 42 kDa, respectively). To down-regulate PKC, cells were treated with 1 M PDBU for 24 h prior to stimulation (right lanes). B, growth-arrested VSMC were exposed to 1 M PDBU for 24 h (or vehicle) to downregulate PKC. Cells were harvested with TME buffer, and Western blot analysis was performed on whole cell lysates using PKC isoform-specific antibodies. Care was taken to ensure equal loading of cell protein, antibody dilutions, and ECL exposure. Asterisks indicate the position of the correct PKC isoform band based on molecular mass. ation but Not Ras-PKC-Association-Activation of ERK1/2 by growth factors has been shown to require Raf interaction with membrane-bound Ras (26). We and other investigators have previously demonstrated that angII activates Ras (27), stimulates Raf association with Ras (27), and activates Raf in VSMC (9). In previous work (9), however, we showed that angIIstimulated association of Raf with Ras was blocked by PDBU treatment while ERK1/2 activation was not blocked. These results suggested that ERK1/2 activation occurred via a Rafindependent pathway as observed by other investigators for different agonists and cells (28 -31). PKCis a candidate protein to mediate the Raf-independent pathway because PKChas recently been shown to stimulate MEK in vitro (10), to associate with Ras (32), and it is not down-regulated by PDBU treatment (Fig. 1B). To study the association of Ras with PKC-, growth-arrested VSMC were stimulated with 100 nM angII, 200 nM PMA, and 10 ng/ml PDGF, H-Ras was immunoprecipitated, and Western blot analysis was performed with PKCantibody. Minimal amounts of PKC-associated with H-Ras in unstimulated cells ( Fig. 2A and B, Control). Treatment for 5 min with angII and PDGF caused PKCto associate with Ras. There was a dramatic increase in the association of PKCwith Ras in response to angII as shown by Western blot analysis ( Fig. 2A, compare lanes 1 and 2). PMA alone occasionally (1 of 3 experiments) caused a small increase in association. Treatment with 1 M PDBU for 24 h completely depleted PKC-␣ but had no effect on PKC- (Fig. 2A, compare lanes 5  and 6). In addition, it is clear that PKC-, but not PKC-␣, associated with Ras in response to agonists as only the lower immunoreactive band was observed in H-ras immunoprecipi-tates. This finding was confirmed by Western blot analysis of H-Ras immunoprecipitates with antibodies against the other four PKC isoforms present in VSMC (not shown).
PKC down-regulation by treatment with 1 M PDBU did not prevent the association of PKCwith H-Ras in cells stimulated by angII and PDGF (Fig. 2B, compare lanes 2 and 3 with lanes  5 and 6). In contrast, the association of Raf with Ras stimulated by angII and PDGF was significantly inhibited after treatment with PDBU (Fig. 2C), as previously reported (9). AngII-stimulated phosphorylation of Raf, which is reflected as the retardation of Raf electrophoretic mobility ("band shift") (3), was also blocked by PDBU treatment for 24 h (Fig. 2D, compare lanes 2  and 3 with lanes 5 and 6). These findings suggest that PKCmay serve as the Ras-associated MEK kinase stimulated by angII in VSMC.
Antisense PKC-Oligonucleotides Specifically Reduce PKC-Protein Expression-To demonstrate further the role of PKCin ERK1/2 activation by angII, antisense PKC-oligonucleotides and their corresponding scrambled controls were employed. Human VSMC were chosen for these experiments because the efficacy of antisense PKColigonucleotides was defined based on human PKC-mRNA and protein expression. 2 Human VSMC were exposed to antisense PKC-oligonucleotides for 6 h, and Western blot analyses for PKC-␣, -␦, -⑀, andwere performed four days later. Protein levels for PKCwere reduced in a concentration-dependent manner with reductions of 20%, 55%, and 70% at 100, 300, and 1,000 nM antisense PKC-oligonucleotide, respectively (Fig. 3A). Scrambled PKColigonucleotides had no effect on PKC-expression at 1,000 nM. Expression of the PKC-␣, -␦, and -⑀ isoforms was not affected by antisense or scrambled PKColigonucleotides, indicating that the antisense PKColigonucleotides were specific for PKC- (Fig. 3B-D). Treatment with angII for 5 min had no effect on protein levels of PKC-␣, -␦, -⑀, and -(compare lanes 1 and 2 in Fig. 3A-D).

FIG. 2. Effect of PDBU on association of PKC-with Ras stimulated by angII, PMA, and PDGF.
Growth-arrested VSMC were exposed to 100 nM angII, 200 nM PMA, or 10 ng/ml PDGF for 5 min and harvested with TME buffer. A, Ras was immunoprecipitated with anti-H-Ras antibody, the immunoprecipitates were size-fractionated by SDS-PAGE, and Western blot analysis was performed with anti-PKCantibody (left 4 lanes). Total cell lysates (TCL) were analyzed similarly but without immunoprecipitation. The upper band is cross-reacting PKC-␣, which is down-regulated by PDBU treatment. B, to downregulate PKC, cells were treated with 1 M PDBU for 24 h. Ras was immunoprecipitated with anti-H-Ras antibody, the immunoprecipitates were size-fractionated by SDS-PAGE, and Western blot analysis was performed with anti-PKCantibody. C, growth-arrested cells were treated with PDBU or vehicle to down-regulate PKC and then exposed for 5 min to angII or PDGF. Cell lysates were prepared, Ras was immunoprecipitated with anti-H-Ras antibody, the immunoprecipitates were size-fractionated by SDS-PAGE, and Western blot analysis was performed with anti-c-Raf-1 antibody. D, to down-regulate PKC, cells were treated with 1 M PDBU for 24 h. Cells were exposed to angII or PDGF, lysates were prepared, proteins were size-fractioned by SDS-PAGE, and Western blot analysis was performed with anti-c-Raf-1 antibody.

DISCUSSION
The major finding of this study is that PKC-associates with Ras in an agonist-dependent manner and is required for activation of ERK1/2 by angII in VSMC. Data that support an essential role for PKCin angII-mediated signaling include the following. 1) PKCassociation with Ras was unaffected by PKC down-regulation (PDBU treatment for 24 h) as was ERK1/2 activation. 2) In contrast, translocation and association of Raf with Ras was inhibited by PKC down-regulation (9).
3) Specific depletion of PKCprotein with antisense PKColigonucleotides inhibited angII-mediated activation of ERK1/2, while scrambled PKColigonucleotides showed no effect. 4) Work from the laboratory of Moscat has shown that PKCmay function as a MEK kinase in vitro. Our results are the first to show that PKC-, which is structurally related to Raf (12), may substitute functionally for Raf in vivo and suggest that PKCis a Ras-associated MEK kinase in VSMC.
Several investigators have showed that Raf phosphorylation is rapidly stimulated by angII, suggesting an important role for Raf in angII signal transduction (3, 7) (33). However, our previous investigations (9) indicate that angII stimulation of ERK1/2 occurs via a c-Raf-1 independent pathway as discussed above. In addition, we previously found that the magnitude of MEK kinase activity was significantly greater in Ras immunoprecipitates than in Raf immunoprecipitates. These findings suggested that MEK kinases other than Raf were stimulated by angII. We cannot be more definitive regarding the role of PKCas a MEK kinase because experiments in which PKCwas immunoprecipitated after angII stimulation failed to show an increase in activity. 3 The inability to demonstrate increased PKCactivity is not unexpected given that its activation requires interactions with various phosphoinositides (13) and possible protein mediators (34) that may be removed during immunoprecipitation.
The mechanisms of PKCregulation are unclear (35). Phosphatidylinositol 3-kinase (PI 3-K) may regulate PKCby generation of activating molecules (e.g. PIP 3 ) and/or by acting as a "linker" protein to bring PKCin contact with other activating molecules. It has been shown that PIP 3 , a PI 3-K product, is a PKCactivator. Nakanishi et al. (13) showed that PIP 3 potently and selectively activated PKCin the absence of phosphatidylserine and/or phosphatidylethanolamine but was much less effective in activating conventional PKC. PI 3-K consists of an 85-kDa regulatory subunit and a 110-kDa catalytic subunit (36) and interacts directly with Ras through its catalytic subunit and the effector site of Ras in a GTP-dependent manner (37) . Wortmannin has been shown to inhibit PI 3-K activity and block PDGF-mediated activation of ERK1/2 (38, 39), suggesting that PI 3-K activity is required for ERK1/2 3 4. Antisense PKC-oligonucleotides inhibit ERK1/2 activation by angII but not by PMA and PDGF. A, human VSMC were grown in DMEM/F-12 for 24 h to 70% confluence and then transfected with 1,000 nM antisense or scrambled PKC-oligonucleotides as described under "Materials and Methods." After transfection, the cells were growth-arrested in DMEM supplemented with 1% platelet-depleted serum for 48 h and then treated with 10 ng/ml PDGF, 100 nM angII, or 200 nM PMA for 5 min. Cells were harvested, and cell lysates were analyzed for ERK1/2 activity by an in-gel-kinase assay. Arrows indicate the position of 44 and 42 kDa bands identified as ERK1/2. As a control for protein loading and nonspecific effects of oligonucleotides, an unidentified myelin basic protein kinase of ϳ90 kDa is also indicated. B, the same cell lysates used for in-gel-kinase assay were analyzed by Western blot with PKCantibody. C, the effect of antisense PKColigonucleotides on ERK1/2 activation by AngII, PDGF, and PMA was determined on a relative basis. ERK1/2 activity was measured by densitometry of autoradiograms in the linear range of film development. The densities of 42 and 44 kDa ERK were measured together. The results for each experiment were normalized to the density of the control (1% serum) sample, which was arbitrarily adjusted to 1.0. Results are the mean Ϯ S.E. of three to five determinations. *, p Ͻ 0.05 versus scrambled or without antisense. activation. However, other investigators (40,41) have found that PKCis activated by diacylglycerol and phosphatidylserine, suggesting multiple mechanisms for activation. In future work, we plan to identify the mechanism by which angII activates PKC-, focusing on interactions with PI 3-K.
The results of the present study strongly support a role for PKCin angII-stimulated signal transduction. VSMC have been reported to express PKC-␣, -␤, -␦, -⑀, and - (42), as confirmed in the present study. Our results are consistent with the reported characteristics of PKC- (43) in that PKC-was not down-regulated by PDBU, and PMA did not stimulate PKCassociation with Ras. Previous studies have shown that PKC is required for angII stimulation of Na ϩ /H ϩ exchange (44), c-fos expression (4), and mitogen-activated protein kinase phosphatase-1 expression (18). Because these experiments used PDBU treatment to down-regulate PKC, it is likely that they were mediated by phorbol ester responsive PKC isoforms (PKC-␣, -␤, -␦, or -⑀). The present study is thus the first to show a role for an atypical PKC isoform in angII signal transduction. Of interest, while angII signal transduction was significantly inhibited by antisense PKColigonucleotides, there was only minimal effect on PDGF-stimulated ERK1/2 activity. It is possible that if PKCexpression could have been inhibited by 100% that there may have been a larger effect on PDGF. In fact, Moscat and other laboratories have shown that PKCis involved in PDGF signal transduction pathway in several cell lines (32). The present findings indicate that, while PDGF stimulates association of PKCwith Ras, PKC-is not required for ERK1/2 activation by PDGF to the same extent as for angII. These results suggest a fundamental difference between the early events stimulated by angII and PDGF. Since angII causes primarily VSMC hypertrophy while PDGF causes primarily hyperplasia (45,46), understanding differences in activation of PKCby these agonists may provide important insights into regulation of VSMC growth.