p160 Bcr Mediates Platelet-Derived Growth Factor Activation of Extracellular Signal-Regulated Kinase in Vascular Smooth Muscle Cells
Background— The human Bcr gene was originally identified by its presence in the chimeric Bcr/Abl oncogene, which is causative for chronic myeloblastic leukemia. Because Bcr encodes a serine/threonine protein kinase, we studied its kinase activity and determined the role of Bcr in the PDGF signaling pathway to ERK1/2 activation and DNA synthesis in rat aortic smooth muscle cells (RASMCs).
Methods and Results— In RASMCs, platelet-derived growth factor-BB (PDGF) stimulated Bcr kinase activity, with a maximum at 1 minute. Because phosphatidylinositol 3′-kinase (PI3-K) is essential for Bcr/Abl leukemogenesis, we evaluated the role of mouse PDGF-β-receptor binding sites for PI3-K (Y708, Y719) and for phospholipase C-γ (Y977, Y989) in PDGF-mediated Bcr kinase activation. The mutant PDGF receptor Y708F/Y719F but not Y977F/Y989F showed significantly reduced Bcr kinase activity. To determine the role of Bcr in PDGF-mediated signal transduction events leading to ERK1/2 and its downstream Elk1 transcription activation, wild-type (WT) and kinase-negative (KN) Bcr were transiently expressed in RASMCs. Bcr WT enhanced, whereas Bcr KN inhibited, PDGF-stimulated ERK1/2 and Elk1 transcriptional activity. Overexpression of Bcr also enhanced PDGF-induced Ras/Raf-1 activity and DNA synthesis, but this regulation is independent of the kinase activity of Bcr. Finally, we found that Bcr expression was increased in the neointimal layer after balloon injury of rat carotid artery.
Conclusions— These results demonstrated the importance of Bcr in PDGF-mediated events, such as activation of Ras, Raf-1, ERK1/2, and Elk1, and stimulation of DNA synthesis.
Received January 25, 2001; revision received June 4, 2001; accepted June 7, 2001.
The Bcr gene was originally defined as the breakpoint of the Philadelphia (Ph) chromosome translocation associated with chronic myelogenous leukemia and acute lymphocytic leukemia. This translocation fuses the c-abl proto-oncogene on chromosome 9 with the 5′ region of the Bcr gene on chromosome 22, resulting in the formation of a chimeric oncogene.1 The Bcr/Abl gene is expressed as a constitutively activated protein-tyrosine kinase (p210 Bcr/Abl) that is believed to contribute directly to the pathogenesis of the disease.2 The function of the endogenous Bcr protein in vascular smooth muscle cells remains undefined. The cDNA sequence of Bcr predicts several functional domains.3 The amino-terminus domain of Bcr contains an oligomerization domain,4 a serine/threonine kinase activity,5 and a region that binds Src-homology 2 (SH2) domains.6 The middle of the protein has a region of sequence similarity to guanine nucleotide exchange factors for the Rho family of GTP binding proteins7 and a pleckstrin homology domain.8 The carboxyl terminus encodes a GTPase-activating function for the small GTP-binding protein Rac (Rac-GAP).9
The platelet-derived growth factor (PDGF) receptor appears to use Bcr as a signal mediator, because Ridley et al10 reported that expression of the Rac-GAP domain of Bcr inhibited PDGF-induced membrane ruffling. In this study, we showed that PDGF stimulated Bcr kinase activity. We found that Bcr regulates extracellular signal-regulated kinase (ERK)1/2 and Elk1 activation in RASMCs. Overexpression of Bcr increased Ras/Raf-1 activity and DNA synthesis by PDGF, and Bcr expression was increased in neointima after balloon injury in rat carotid artery. These results suggest a role for Bcr in PDGF signaling as a modulator on Ras/Raf-1 activity and cell proliferation and its possible role in the process of restenosis.
Rat aortic smooth muscle cells (RASMCs) and human umbilical vein endothelial cells (HUVECs) were isolated and maintained as described previously.11,12 Chinese hamster ovary (CHO) cells were grown in F-12 medium supplemented with 10% FBS. Rat fibroblast cell (Rat1) was cultured in DMEM with 10% FBS. Human aortic smooth muscle cells (HASMCs) and human K562 leukemia cells (K562) were a kind gift from Dr R. Ross and from Dr John M. Harlan, respectively (both from University of Washington, Seattle). The HASMCs and K562 cells were incubated in DMEM/10% FBS and RPMI 1640/10% FBS, respectively.
Plasmids and Transfection
Bcr wild-type, the kinase-negative form of Bcr (Y328F/Y360F double mutant), and dominant negative Ras [Ras(N17)] were generated as described previously.13,14 For transient expression experiments, cells were transfected with the lipofectamine plus method (Gibco BRL) as described previously.15 Using the lipofectamine technique, we have determined that transfection efficiency is 80% to 90% in CHO cells and 5% to 10% in RASMCs with the LacZ expression plasmid pcDNA3.1Lacz (Invitrogen). Cells at 70% to 80% confluence in 60- to 100-mm dishes were growth-arrested by incubation in DMEM (for RASMCs) or F-12 (for CHO cells) for 24 hours before use. The cells were treated with PDGF-BB (Boehringer Mannheim) and were harvested.
PDGF Receptor Mutants
CHO cells transfected with PDGF receptors were a kind gift from Dr Harlan E. Ives, University of California, San Francisco. In brief, plasmids containing the mutated PDGF-β-receptor cDNA were used to stably transfect the receptor in CHO cells, which normally lack PDGF receptors. Mutation of Tyr-708 to Phe (Y708F) and Tyr-719 to Phe (Y719F) prevents the association of p85 phosphatidylinositol 3′-kinase (PI3-K) (Y708F, Y719F), and mutation of Tyr-977 to Phe (Y977F) and Tyr-989 to Phe (Y989) prevents the association of phospholipase C (PLC)-γ.16 Y708/Y719 and Y977/Y989 in mouse PDGF-β-receptor are equivalent to Y740/Y751 and Y1009/Y1021 in humans.16,17
Immunoprecipitation and Western Blot
After treatment with reagents, the cells were washed with PBS (−) and harvested in 0.5 mL of lysis buffer as described previously.18 Immunoprecipitation was performed as described previously with mouse anti-Bcr or anti-hemagglutinating (HA) (F-7) antibody (Santa Cruz) or PDGF-β receptor (U.B.I.).18 Western blot analysis was performed as previously described.18 In brief, the blots were incubated for 4 hours at room temperature with the anti-Bcr (Santa Cruz), SH3 p85 (PI3-K), PLC-γ, or PDGF-β-receptor (U.B.I.) antibody, followed by incubation with horseradish peroxidase-conjugated secondary antibody (Amersham Life Science). For ERK1/2 activation in HA (−ERK2) immunoprecipitates, the blot were incubated 12 hours with anti-phospho ERK1/2 (New England Biolabs) or ERK2 or anti-HA antibodies (Santa Cruz). For Raf-1, MEK1/2, and ERK1/2 activation in total cell lysates, the blots were incubated for 12 hours with anti-phospho Raf-1, MEK1/2, or ERK1/2 (New England Biolabs) or non-phospho Raf-1 or ERK1/2 (Santa Cruz) or MEK1/2 (New England Biolabs) antibodies.
Bcr Kinase Assay
PathDetect trans-Reporting System
A PathDetect trans-reporting system (Stratagene) was used for detection of Elk1 transcription activity as described previously.21
PAK Kinase Assay
p21-activated protein kinase (PAK) kinase activity was assayed by immunocomplex myelin basic protein (MBP from UPI) in-gel kinase assay, in which the PAK immunocomplex was precipitated with anti-α-PAK antibody (Santa Cruz) as described previously.22
Activated Ras Affinity Precipitation Assay
Affinity precipitation of activated p21 Ras was performed with Ras antibody (Santa Cruz) as described previously.23
Preparation of Balloon Injury Model and Immunohistochemistry
Tissue preparation of balloon-injured rat carotid artery was performed as described previously.24 Immunohistochemistry was performed as previously described with monoclonal anti-Bcr antibody (1:1000 dilution) as primary antibody.25 For a negative control, primary antibody was substituted with normal mouse IgG at a corresponding dilution. The cross sections were counterstained with hematoxylin.
[3H]Thymidine Incorporation Assay
Measurement of [3H]thymidine incorporation into DNA was performed as described previously.26
Data are reported as mean±SD. Statistical analysis was performed with the StatView 4.0 package (ABACUS Concepts). Differences were analyzed with 1-way or 2-way repeated-measures ANOVA as appropriate, followed by Scheffé’s correction.
Bcr Is Expressed and Activated Rapidly by PDGF in Vascular Smooth Muscle Cells
To determine whether Bcr was expressed in vascular cells, we analyzed cell lysates by Western blot for Bcr in several different cell types. An immunoreactive band of 160 kDa was present in all cell types examined, with relatively high expression in HASMCs but much less in HUVECs (Figure 1A). Both p160 Bcr and p210 Bcr/Abl were identified in K562 cells, which are a Ph1-positive chromosome cell line.
We next investigated whether PDGF stimulated Bcr serine/threonine kinase activity. Immunoprecipitation followed by Western blot analysis of lysates revealed a single 160-kDa protein band in addition to IgG in RASMCs and CHO cells (Figure 1B). First, Bcr kinase activity was determined by in vitro kinase assay with histone H1 as a substrate. PDGF (10 ng/mL) rapidly activated Bcr, with a maximal increase in activity of 3.4±1.3-fold (P<0.01) within 1 minute (Figure 2A). To confirm this Bcr kinase activation by PDGF, we also performed an autokinase assay of Bcr as we described previously.19,20 Similarly, Bcr autokinase activity was increased within 1 minute after PDGF stimulation (Figure 2B). Bcr kinase was activated by 2.5 ng/mL PDGF and was maximal (P<0.01) at 10 to 30 ng/mL PDGF (Figure 2C). Bcr autokinase activation was also stimulated by PDGF dose-dependently in RASMCs (Figure 2D).
PDGF β-Receptor Y708F/Y719F but Not Y977F/Y989F Mutations Inhibit Bcr Kinase Activation by PDGF
It has been reported that PI3-K has an essential role in Bcr/Abl-mediated leukemia. To determine whether the association of p85 (PI3-K) and PDGF-β receptor was required for Bcr kinase activation, we studied the effects of mutations in the PDGF-β receptor (Y708F/Y719F) that prevent binding of PI3-K to the activated receptor16 in transfected CHO cells. We also used another PDGF-β-receptor mutant (Y977F/Y989F) that prevents PLC-γ binding to PDGF-β receptor. We confirmed wild-type PDGF receptor interaction with p85 (PI3-K) by PDGF stimulation (Figure 3A and 3B). We further confirmed that the mutant PDGF receptor (Y977F/Y989F) associated weakly with p85 (PI3-K) but not with PLC-γ, whereas the mutant PDGF receptor (Y708F/Y719F) also bound PLC-γ weakly but not p85 (PI3-K)17 (Figure 3B).
In CHO cells expressing wild-type PDGF receptor, PDGF rapidly stimulated Bcr kinase activity (Figure 3C, top, and 3D), with a maximal increase (P<0.05) at 1 minute. In CHO cells expressing mutant PDGF receptor (Y708F/Y719F), PDGF was unable to increased Bcr kinase activity (Figure 3C, middle, and 3D). The Bcr kinase activity, however, was similar (P<0.01) to wild-type in Y977F/Y989F mutant-expressing cells (Figure 3C, bottom, and 3D).
To determine whether kinase activity of PI3-K is required for Bcr activation by PDGF, effects of PI3-K inhibitors on PDGF-increased Bcr kinase activity were tested in CHO cells expressing PDGF-β receptor (CHO-PDGFR). After pretreatment with 100 nmol/L Wortmannin and 100 μmol/L LY294002 (Calbiochem) for 10 minutes before the addition of 20 ng/mL PDGF for 1 minute, the cell lysates were applied for Bcr kinase assay. Both PI3-K inhibitors inhibited PDGF-mediated Akt activation in CHO-PDGFR (data not shown) but did not suppress PDGF-induced Bcr kinase activity (Figure 3E). We also determined the similar lack of inhibitory effect of PI3-K inhibitors on Bcr kinase activation in RASMCs (data not shown).
Bcr Mediates PDGF-Induced ERK1/2 and Elk1 Activation in RASMCs
To identify downstream signaling molecules activated by Bcr, we determined the effect of Bcr wild-type (WT) and Bcr kinase-negative (KN) expression on PDGF-β receptor-dependent ERK1/2 activation. After cotransfection of RASMCs with Bcr WT or Bcr KN as well as HA epitope-tagged ERK2, the activity of immunoprecipitated HA-ERK2 was then determined as described in Methods. Compared with control cells, Bcr WT-transfected RASMCs exhibited increased HA-ERK2 activation (P<0.01) by PDGF, but this was significantly inhibited (P<0.05) in Bcr KN-transfected cells (Figure 4A and 4B). We observed the same inhibition by Bcr KN in CHO-PDGFR (data not shown).
To corroborate this observation, we further investigated whether Bcr regulates PDGF-mediated Elk1 transcription activation, as a downstream effector of ERK1/2, in RASMCs by use of the PathDetect trans-reporting system. Compared with vector-transfected cells, Bcr WT-transfected cells exhibited increased Elk1 transcription activity (P<0.05; Figure 4C), but Bcr KN significantly inhibited PDGF-induced Elk1 activity dose-dependently (P<0.05 and P<0.01; Figure 4D).
An important downstream component of Rac in the signaling pathway is PAK.22 Because the carboxyl terminus of Bcr encodes a GTPase-activating function for Rac, we determined the PDGF-induced activation of PAK in Bcr WT-transfected CHO-PDGFR. As shown in Figure 4E, we could not detect any significantly different PAK activity in vector- and Bcr WT-transfected cells. These data suggest that Bcr WT modulated ERK1/2 activation but not Rac/PAK activation by PDGF stimulation.
Ras Regulates PDGF-Induced Raf-1, MEK1/2, and ERK1/2 Activation but Not Bcr Kinase Activation
To evaluate the role of Ras in PDGF-induced Bcr and ERK1/2 activation, we transfected Ras(N17) and evaluated Bcr, Raf-1, MEK1/2, and ERK1/2 activity in CHO-PDGFR. We found that Ras(N17) did not inhibit PDGF-induced Bcr kinase activity (Figure 5A) but completely inhibited Raf-1 activation (P<0.05 and P<0.01) by PDGF (Figure 5B). Interestingly, Ras(N17) significantly but partially inhibited PDGF-induced MEK1/2 by ≈55% (P<0.05 and P<0.01; Figure 5C) and ERK1/2 activity (P<0.05 and P<0.01) by ≈60% at 5 minutes (Figure 5D).
Bcr Stimulates Ras and Raf-1 Activity and Enhances PDGF-Induced DNA Synthesis
Numerous studies have shown that Ras/Raf signaling is involved in the regulation of cell proliferation. To determine the role of Bcr in cell proliferation, we evaluated the effect of Bcr on PDGF-mediated Ras and Raf-1 activation. As shown in Figure 6A and 6B, not only Bcr WT but also Bcr KN significantly enhanced PDGF-induced Ras and Raf-1 activity (P<0.05 and P<0.01). These data suggest that Bcr is an upstream regulator of Ras and Raf-1, but this regulation is independent of Bcr kinase activity.
Furthermore, we determined the role of Bcr in PDGF-induced DNA synthesis. Because the half-maximal effects of DNA synthesis by PDGF-B were obtained at 2.5 to 5.0 ng/mL in CHO-PDGFR, we used 2.5 ng/mL of PDGF to determine the enhancement effect of Bcr. As shown in Figure 6C, both BCR WT and BCR KN overexpression also significantly enhanced PDGF-induced DNA synthesis (P<0.01). These results show that Bcr expression, but not its kinase activity, has a significant role in PDGF-induced DNA synthesis.
Bcr is Expressed Predominantly in the Neointima in Balloon-Injured Rat Carotid Artery
To determine the cellular distribution of Bcr protein expression, immunohistochemical analysis was performed on cross sections of sham, 7-day, and 14-day balloon-injured rat carotid arteries (Figure 7). Immunohistochemical analysis shows increased Bcr expression in the neointima, present at both 7 and 14 days. At 14 days, a gradient of Bcr protein expression was observed from the lumen to the deeper layers of the neointima. In contrast, carotids from sham-operated rats and the media of injured vessels exhibited very little Bcr expression.
In the present study, we have shown that Bcr is a downstream component of PDGF receptor signaling and that Bcr acts as an adapter molecule in transmitting signals to ERK1/2 and Elk1 transcription activity for PDGF-β receptor. Many investigators, including us, have reported the importance of PDGF in the process of restenosis and atherogenesis in vascular disease. For example, we reported previously that adenovirus-mediated gene transfer of the extracellular region of the PDGF-β receptor, which acts as its antagonist, into injured arteries resulted in a >50% reduction in the neointimal area of injured arteries.27 These results provide direct evidence that PDGF-β-receptor activation plays an essential role in neointima formation. In the present study, we found that Bcr transmits signals from the PDGF-β receptor to Ras and ERK1/2-Elk1. The activation of Ras initiates a variety of protein kinase cascades that include PI3-K, Raf-1 kinase, mitogen-activated protein kinases, and subsequent cell transformation and proliferation.28 Therefore, if the function of Ras could be abolished, cellular proliferation might be stopped. In fact, Ueno et al29 reported that application of an adenoviral vector expressing a potent dominant negative mutated form of Ras into balloon-injured rat carotid arteries significantly reduced neointima formation. These data suggest the critical role of Ras in proliferative arterial disease. Interestingly, in the present study, we found that overexpression of Bcr enhanced PDGF-induced Ras/Raf-1 and ERK1/2-Elk1 activation, but this enhancement effect of Bcr to Ras/Raf-1 activity is independent of the kinase activity of Bcr. In fact, our laboratory and others have shown that there is a Ras(N17)-independent pathway that regulates ERK1/2 activation in multiple cell types, including VSMCs.15 Marais et al30 demonstrated that PKC-mediated ERK1/2 activation required Ras protein but was not inhibited by Ras(N17). We showed that dominant negative Ras could not completely inhibit ERK1/2 activation (Figure 5), supporting the complexity of PDGF-induced ERK1/2-Elk1 activation. Elk1, a c-fos proto-oncogene regulator that belongs to the ETS-domain family of transcriptional factors, plays an important role in the induction of immediate early gene expression and subsequent protein synthesis in response to a variety of extracellular signals.31 Therefore, although the kinase activity of Bcr does not regulate DNA synthesis, it is possible that Bcr kinase activity is crucial to control mitogen-induced protein synthesis, which leads to hypertrophy in vascular smooth muscle cells.32 Future studies will be necessary to define the precise role of Bcr kinase activation in arterial diseases.
PI3-Ks comprised 2 subunits of 85- and 110-kDa molecular mass. We found that PDGF stimulates Bcr serine/threonine kinase activity and that the sites of mouse PDGF-β-receptor tyrosine 708 and 719, which are important for binding PI3-K, are critical for Bcr kinase activation. We used wortmannin and LY294002 to inhibit p110 PI3-K activity, but Bcr kinase activation by PDGF was not inhibited by these PI3-K kinase inhibitors (Figure 3E). It is important to note that Nck and Shc33 share the same binding site as PI3-K (Y751) in the human PDGF-β receptor. Future studies will be necessary to define the precise mechanisms for Bcr kinase activation in PDGF-mediated signaling events.
This study was supported by grants from the National Institutes of Health to Dr Abe (HL-61319) and to Dr Berk (HL-44721 and HL-49192). We thank Dr Harlan E. Ives, Cardiovascular Research Institute, University of California, San Francisco, for providing the PDGF receptor wild-type and mutant-overexpressing CHO cells. We also thank members of the Berk laboratory for helpful discussions, especially Drs Jane Sottile, Joseph Miano, and Wang Min.
The first 2 authors contributed equally to this work.
Duff JL, Marrero MB, Paxton WG, et al. Angiotensin II induces 3CH134, a protein-tyrosine phosphatase, in vascular smooth muscle cells. J Biol Chem. . 1993; 268: 26037–26040.
Haendeler J, Ishida M, Hunyady L, et al. The third cytoplasmic loop of the angiotensin type I receptor exerts differential effects on ERK1/2 and apoptosis via Ras- and Rap1-dependent pathways. Circ Res. . 2000; 86: 729–736.
Liao DF, Monia B, Dean N, et al. Protein kinase C-ζ mediates angiotensin II activation of ERK1/2 in vascular smooth muscle cells. J Biol Chem. . 1997; 272: 6146–6150.
Ma YH, Reusch HP, Wilson E, et al. Activation of Na+/H+exchange by platelet-derived growth factor involves phosphatidylinositol 3′-kinase and phospholipase C gamma. J Biol Chem. . 1994; 269: 30734–30739.
Abe J, Kusuhara M, Ulevitch RJ, et al. Big mitogen-activated protein kinase 1 (BMK1) is a redox-sensitive kinase. J Biol Chem. . 1996; 271: 16586–16590.
Yan C, Luo H, Lee JD, et al. Molecular cloning of mouse ERK5/BMK1 splice variants and characterization of erk5 functional domains. J Biol Chem. . 2001; 276: 21686–21691.
Schmitz U, Ishida T, Ishida M, et al. Angiotensin II stimulates p21-activated kinase in vascular smooth muscle cells: role in activation of JNK. Circ Res. . 1998; 82: 1272–1278.
Abe J, Okuda M, Huang Q, et al. Reactive oxygen species activate p90 ribosomal S6 kinase via fyn and ras. J Biol Chem. . 2000; 275: 1739–1748.
Melaragno MG, Wuthrich DA, Poppa V, et al. Increased expression of Axl tyrosine kinase after vascular injury and regulation by G protein-coupled receptor agonists in rats. Circ Res. . 1998; 83: 697–704.
Abe J, Zhou W, Takuwa N, et al. A fumagillin derivative angiogenesis inhibitor, AGM-1470, inhibits activation of cyclin-dependent kinases and phosphorylation of retinoblastoma gene product but not protein tyrosyl phosphorylation or protooncogene expression in vascular endothelial cells. Cancer Res. . 1994; 54: 3407–3412.
Vojtek AB, Der CJ. Increasing complexity of the Ras signaling pathway. J Biol Chem. . 1998; 273: 19925–19928.
Ueno H, Yamamoto H, Ito S, et al. Adenovirus-mediated transfer of a dominant-negative H-ras suppresses neointimal formation in balloon-injured arteries in vivo. Arterioscler Thromb Vasc Biol. . 1997; 17: 898–904.
Marais R, Light Y, Mason C, et al. Requirement of Ras-GTP-Raf complexes for activation of Raf-1 by protein kinase C [published erratum appears in Science. 1998;280:987]. Science. . 1998; 280: 109–112.
Berk BC, Vekshtein V, Gordon HM, et al. Angiotensin II-stimulated protein synthesis in cultured vascular smooth muscle cells. Hypertension. . 1989; 13: 305–314.
Yokote K, Mori S, Hansen K, et al. Direct interaction between Shc and the platelet-derived growth factor beta-receptor. J Biol Chem. . 1994; 269: 15337–15343.