(Received for publication, October 21, 1996, and in revised form, December 30, 1996)
From the Department of Medicine, Division of Cardiology, University
of Washington, Seattle, Washington 98195 and ISIS
Pharmaceuticals, Carlsbad, California 92008
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-,
-
1, -
, -
, and -
in relative abundance. All isoforms except
PKC-
were down-regulated by 1 µM PDBU for 24 h
suggesting that PKC-
was responsible for angII-mediated ERK1/2
activation. In response to angII, PKC-
associated with Ras as shown
by co-precipitation of PKC-
with anti-H-Ras antibody. To
characterize further the role of PKC-
, PKC-
protein was depleted specifically by transfection with antisense PKC-
oligonucleotides. Antisense PKC-
oligonucleotide treatment significantly decreased PKC-
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-
and Ras
in angII stimulation of ERK1/2.
Angiotensin II (angII)1 plays an important role in hypertrophic 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 µM 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. PKC-
has 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 PKC-
closely resembles c-Raf-1 (12). Within the PKC family, PKC-
represents an atypical PKC isoform in that it lacks the C2 domain
making its kinase activity Ca2+-independent, and it
possesses only one zinc finger region in its regulatory domain (12).
Consequently, PKC-
does not bind Ca2+ 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 PKC-
activity
(15).
In the present study, we investigated the role of PKC- in
agonist-mediated ERK1/2 activation. The results show that PKC-
associates with Ras in response to angII, and PDGF. Antisense PKC-
oligonucleotides 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
PKC-
association with Ras.
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 µM TES, 50 µg/ml ascorbic acid, 10 µg/ml insulin, 10 µg/ml transferrin, 10 ng/ml sodium selenite, 30 µg/ml endothelial cell growth supplement, and 10% fetal bovine serum. Cells were growth arrested by incubation in DMEM medium with 1% platelet-depleted serum for 48 h.
Western Blot AnalysisAfter 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
MgCl2, 1 mM EDTA, 25 mM NaF)
containing fresh 100 µM Na3VO4,
20 µg/ml leupeptin, 1 µg/ml pepstatin A, 4 µg/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 × g. Protein concentration in the supernatant was
determined by Bradford protein assay, and the samples were stored at
80 °C. For Western blot analysis, 20 µg of protein was subjected
to SDS-PAGE under reducing conditions, and proteins were then
transferred to nitrocellulose (HybondTM-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).
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.
Human vascular smooth muscle cells were plated
in 6-well tissue culture dishes with DMEM/F-12 containing 10 µM TES, 50 µg/ml ascorbic acid, 10 µg/ml insulin, 10 µg/ml transferrin, 10 ng/ml sodium selenite, 30 µg/ml endothelial
cell growth supplement, and 10% fetal bovine serum at 5 × 105 cells/well and grown overnight in a CO2
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 GACGCACGCGGCCTCACACC, and the scrambled oligonucleotide
sequence was AAGCGCGCACCAGCGCCTCC. A complex of LipofectAMINE and
oligonucleotides (2.5 µg/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).
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 AnalysisData are presented as mean ± S.E. for all experiments that were performed at least three times. Significant differences were determined by Student's t test (p < 0.05).
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 µM 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 ± 11% of control at 5 min,
n = 11, p > 0.1 versus
control). In contrast, there was >70% inhibition of PDGF- and
PMA-stimulated ERK1/2 activity (n = 5 and 8, respectively, p < 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.
To determine which PKC isoforms are expressed in VSMC, Western blot
analysis with PKC isoform-specific antibodies was performed on whole
cell lysates. VSMC express 5 different PKC isoforms in relative
abundance, PKC-, -
1, -
, -
, and -
(Fig. 1B).
No significant immunoreactive PKC-
2, -
, -
, -
, or -
isoforms were detected. The upper band in the PKC-
Western blot is actually PKC-
that cross-reacts with the PKC-
antibody (24). Treating cells with PDBU (1 µM for 24 h) caused PKC-
, -
1, -
, and -
to be down-regulated completely while PKC-
was unaffected (Fig. 1B). Note that
the PKC-
detected by the PKC-
antibody was completely
down-regulated. These results are consistent with the described phorbol
ester binding characteristics of the different PKC isoforms (25).
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 angII-stimulated 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 Raf-independent pathway
as observed by other investigators for different agonists and cells
(28-31). PKC- is a candidate protein to mediate the Raf-independent
pathway because PKC-
has 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 PKC-
antibody. 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 PKC-
to associate with Ras. There was a dramatic
increase in the association of PKC-
with 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 µM 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 immunoprecipitates. 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 µM PDBU did not
prevent the association of PKC- with 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
PKC-
may serve as the Ras-associated MEK kinase stimulated by angII
in VSMC.
To demonstrate further the role of PKC- in
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 PKC-
oligonucleotides 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-
, -
, -
, and -
were performed four
days later. Protein levels for PKC-
were 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 PKC-
oligonucleotides had no effect on PKC-
expression
at 1,000 nM. Expression of the PKC-
, -
, and -
isoforms was not affected by antisense or scrambled PKC-
oligonucleotides, indicating that the antisense PKC-
oligonucleotides 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).
Antisense PKC-
To determine the role of PKC- in
ERK1/2 activation by agonists, human VSMC treated with antisense or
scrambled PKC-
oligonucleotides were exposed to angII, PDGF, and PMA
4 days after transfection. All three agonists stimulated ERK1/2
activity in these rapidly growing cells (Fig.
4A, compare lanes 2-4 with
lane 1). The increase in ERK1/2 activity by the agonists
ranged from 8- to 10-fold (Fig. 4C; n = 4).
Scrambled PKC-
oligonucleotides had no apparent effect on ERK1/2
activity in either control unstimulated cells (Fig. 4A,
compare lanes 1 and 5) or ang II-stimulated cells
(Fig. 4A, compare lanes 3 and 6).
Scrambled PKC-
oligonucleotides also had no effect on PKC-
protein expression (Fig. 4B). Analysis of multiple
experiments showed that there was no significant effect of scrambled
PKC-
oligonucleotides on angII-stimulated ERK1/2 activity (Fig.
4C; 9.1 ± 2.3-fold versus 8.6 ± 1.8-fold, n = 3, p > 0.1). Antisense
PKC-
oligonucleotide treatment had no significant effect on ERK1/2
activity in control, unstimulated cells (Fig. 4A, compare
lanes 1 and 7, and Fig. 4C). It should
be noted that, at 1,000 nM antisense PKC-
oligonucleotides, there was significant but not complete inhibition of
PKC-
expression (Fig. 4B). Higher oligonucleotide
concentrations were not used because of cell toxicity. Antisense
PKC-
oligonucleotide treatment did not significantly inhibit ERK1/2
activity in cells stimulated by PDGF and PMA (Fig. 4A,
compare lanes 2 and 4 with lanes 8 and
10). However, ERK1/2 activity stimulated by angII was
clearly inhibited by antisense PKC-
oligonucleotides (Fig.
4A, compare lanes 3 and 9). Analysis of multiple experiments indicated that ERK1/2 activation by angII was
significantly inhibited by 48% (from 9.1 ± 2.3-fold to 4.5 ± 1.1-fold, p < 0.05, n = 3),
respectively, while PDGF and PMA were not significantly inhibited (Fig.
4C). The extent to which antisense PKC-
oligonucleotides
inhibited ang II-mediated ERK1/2 activation correlated well with the
extent to which treatment inhibited PKC-
protein expression:
reductions of 40%, 56%, and 66% in PKC-
expression resulted in
inhibition of angII-stimulated ERK1/2 activity of 37%, 46%, and 61%,
respectively. Additional controls for the specificity of the antisense
PKC-
oligonucleotides were the findings that the activity of a
90-kDa kinase (Fig. 4A) and the expression of PKC-
(Fig.
4B) remained unaffected. These findings indicate that
PKC-
is required for ERK1/2 activation by angII, but not by PDGF and
PMA, in VSMC.
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 PKC-
in angII-mediated signaling include the following. 1) PKC-
association 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 PKC-
protein with antisense PKC-
oligonucleotides inhibited angII-mediated activation of ERK1/2, while scrambled PKC-
oligonucleotides
showed no effect. 4) Work from the laboratory of Moscat has shown that PKC-
may 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 PKC-
is 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 PKC- as a
MEK kinase because experiments in which PKC-
was immunoprecipitated
after angII stimulation failed to show an increase in
activity.3 The inability to demonstrate
increased PKC-
activity 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 PKC- regulation are unclear (35).
Phosphatidylinositol 3-kinase (PI 3-K) may regulate PKC-
by
generation of activating molecules (e.g. PIP3)
and/or by acting as a "linker" protein to bring PKC-
in contact
with other activating molecules. It has been shown that
PIP3, a PI 3-K product, is a PKC-
activator. Nakanishi
et al. (13) showed that PIP3 potently and
selectively activated PKC-
in 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
activation. However, other investigators (40, 41) have found that
PKC-
is 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 PKC- in
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 PKC-
association with Ras. Previous studies have shown that PKC is required for angII stimulation of Na+/H+ 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 PKC-
oligonucleotides, there
was only minimal effect on PDGF-stimulated ERK1/2 activity. It is
possible that if PKC-
expression 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 PKC-
is involved in PDGF signal transduction pathway in several cell lines (32). The present findings
indicate that, while PDGF stimulates association of PKC-
with 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
PKC-
by these agonists may provide important insights into
regulation of VSMC growth.
We thank Arnold Baas and members of the Berk laboratory for helpful discussions.