A Dual Inhibitor of Platelet-Derived Growth Factor β-Receptor and Src Kinase Activity Potently Interferes With Motogenic and Mitogenic Responses to PDGF in Vascular Smooth Muscle Cells
A Novel Candidate for Prevention of Vascular Remodeling
Abstract—PP1 has previously been described as an inhibitor of the Src-family kinases p56Lck and FynT. We have therefore decided to use PP1 to determine the functional role of Src in platelet-derived growth factor (PDGF)–induced proliferation and migration of human coronary artery smooth muscle cells (HCASMCs). A synthetic protocol for PP1/AGL1872 has been developed, and the inhibitory activity of PP1/AGL1872 against Src was examined. PP1/AGL1872 potently inhibited recombinant p60c-src in vitro and Src-dependent tyrosine phosphorylation in p60c-srcF572-transformed NIH3T3 cells. PP1/AGL1872 also potently inhibited PDGF-stimulated migration of HCASMCs, as determined in the modified Boyden chamber, as well as PDGF-stimulated proliferation of HCASMCs. Surprisingly, in addition to inhibition of Src kinase, PP1/AGL1872 was found to inhibit PDGF receptor kinase in cell-free assays and in various types of intact cells, including HCASMCs. PP1/AGL1872 did not inhibit phosphorylation of the vascular endothelial growth factor receptor KDR (VEGF receptor-2; kinase-insert domain containing receptor) in cell-free assays as well as in intact human coronary artery endothelial cells. In line with the insensitivity of KDR, PP1/AGL1872 had only a weak effect on vascular endothelial growth factor–stimulated migration of human coronary artery endothelial cells. On treatment of cells expressing different receptor tyrosine kinases, the activities of the epidermal growth factor receptor, fibroblast growth factor receptor-1, and insulin-like growth factor-1 receptor were resistant to PP1/AGL1872, whereas PDGF α-receptor was susceptible, albeit to a lesser extent than PDGF β-receptor. These data suggest that the previously described tyrosine kinase inhibitor PP1/AGL1872 is not selective for the Src family of tyrosine kinases. It is also a potent inhibitor of the PDGF β-receptor kinase but is not a ubiquitous tyrosine kinase inhibitor. PP1/AGL1872 inhibits migration and proliferation of HCASMCs probably by interference with 2 distinct tyrosine phosphorylation events, creating a novel and potent inhibitory principle with possible relevance for the treatment of pathological HCASMC activity, such as vascular remodeling and restenosis.
- platelet-derived growth factor β-receptor
- coronary artery smooth muscle cell
Protein tyrosine phosphorylation has been established as a crucial step in the regulation of normal cell proliferation, migration, differentiation, and survival. Specific tyrosine phosphorylation events are triggered by extracellular stimuli such as growth factors and processed by intracellular signal transducers.1 2 Aberrant tyrosine phosphorylation has been associated with numerous pathological states. Therefore, the development of tyrosine phosphorylation inhibitors (tyrphostins) for specific protein tyrosine kinases appears to be promising in the framework of the general concept of “signal transduction therapy.”3 4 Indeed, in the last few years it has become apparent that tyrphostins with a high degree of selectivity and potency can be identified.4 5 6 For example, the tyrphostins AG1295 and AG1296, 2 closely related quinoxaline derivatives, selectively inhibit the platelet-derived growth factor (PDGF) β-receptor tyrosine kinase, PDGF-stimulated proliferation, and cell transformation driven by autocrine PDGF receptor activation.7 8 The kinase inhibition is direct and involves interaction of the compounds with the ATP binding domain.9 Interestingly, KDR, a functional receptor for vascular endothelial growth factor (VEGF), which is structurally related to the PDGF β-receptor, is not affected at all by AG1295 and AG1296, as are other tyrosine kinases such as the epidermal growth factor (EGF) receptor and Src.7
Inhibition of PDGF β-receptor signaling has important potential clinical applications. Besides treatment of malignancies, which involve PDGF β-receptor activation,10 11 12 the treatment of pathological vascular smooth muscle cell migration and proliferation in the context of atherosclerotic and restenotic processes appears as a promising application for PDGF β-receptor kinase blockers.13 14 15 16 17 18 19 Because AG1295 and AG1296 do not inhibit the VEGF receptor KDR, and because endothelial cells of large arteries do not express PDGF receptors, it was postulated that AG1295 and AG1296 would not inhibit endothelial cell activation but only smooth muscle cell activity.9 20 This prediction turned out to be correct indeed, as AG1295 was shown to inhibit neointima formation in pigs by blocking proliferation and migration of smooth muscle cells without blocking endothelial cells.21
PDGF β-receptor signaling involves interaction of the autophosphorylated receptor kinase with multiple SH2-domain–containing downstream effector molecules.22 Among other signaling events induced by PDGF, activation of Src-family kinases has been shown to be critical for the generation of the mitogenic signal.23 24 Activation occurs by the binding of Src to the autophosphorylated PDGF β-receptor at phosphotyrosine residues 579 and 581.25 Src has also been reported to phosphorylate the PDGF β-receptor at several sites, but its biological significance has not been fully established.26 To further evaluate the involvement of Src in PDGF signaling in vascular smooth muscle cells, we have used an Src-family kinase blocker, PP1, which was described recently.27 In the absence of a published synthesis protocol, we synthesized the compound, named it PP1/AGL1872, and tested its potential as a PDGF β-receptor signaling antagonist. PP1/AGL1872 inhibits recombinant Src kinase and Src-dependent tyrosine phosphorylation. It potently inhibits human coronary artery smooth muscle cell (HCASMC) migration and also proliferation, 2 critical steps in the restenotic process.13 18 28 Surprisingly, however, we find that in addition to Src kinase inhibition, PP1/AGL1872 directly inhibits the PDGF β-receptor tyrosine kinase, but it does not inhibit the kinase of the VEGF receptor KDR, the EGF receptor kinase, the fibroblast growth factor (FGF) receptor-1 kinase, or the insulin-like growth factor-1 (IGF-I) receptor kinase. In this article we describe in detail the dual Src and PDGF β-receptor kinase inhibitor PP1/AGL1872.
Materials and Methods
“Workup” refers to the process of adding water to the reaction mixture and extracting it with CH2Cl2 or ethyl acetate (EtAc); washing the organic phase; drying, filtering, and evaporating to dryness. All compounds were characterized spectroscopically and had the correct elemental analysis. Nuclear magnetic resonance (NMR) was recorded in CDCl3, with shifts in ppm downfield to tetramethyl silane. Compounds 1a, 1b, 2a, and 2b were prepared according to Reference 2929 . Compounds 4a (PP1/AGL1872) and 4b (PP2/AGL1879) were reported previously, but no synthesis or chemical data were described.27 Figure 1⇓ presents the synthesis of PP1/AGL1872 and PP2/AGL1879. Compounds 1a through 4b are described below; their numbers correspond to those appearing in Figure 1⇓.
1. Acyl Malononitriles
Compound 1a is a light yellow solid, yield 60%, with melting point 180°C (from CCl4). NMR (acetone d6) measures 7.69, 7.40 (4H, ABq, JAB=8.0 Hz), and 2.43 (3H, s, CH3).
Compound 1b is a light yellow solid, yield 55%, with melting point 187°C. NMR 7.50, 4H, narrow ABq.
2. Benzylidene Malononitriles
Compound 2a is a white solid, 40% yield, with melting point 95°C (from chromatography and trituration with methanol). NMR 7.40 (4H, AB, JAB=8.8 Hz), 3.95 (3H, s, methoxy), and 2.46 (3H, s, methyl).
Compound 2b is a white solid (chromatography and trituration with hexane), yield 51%, with melting point 120°C. NMR 7.50 (4H, JAB=8.5 Hz) and 3.95 (3H, s, methoxy).
3. 1–Tertiary Butyl (t-Bu)–3 Aryl-4 Cyano-5 Amino Pyrazoles
Compounds 3a (0.3 g, 1.5 mmol/L) and 2a (0.22 g, 1.8 mmol/L), t-Bu hydrazine HCl, and 0.09 g KOH in 20 mL ethanol were refluxed for 1.5 hours. Workup (EtAc) and trituration with benzene-hexane gave a 0.134 g white solid, 35% yield, with melting point 157°C. NMR 7.82, 7.23 (4H, ABq, JAB=8.6Hz), 4.37 (2H, br.s, NH2), 2.39 (3H, s, methyl), and 1.69 (9H, s, t-Bu).
Compounds 3b (1.5 g, 6.8 mmol/L) and 2b (1 g, 8 mmol/L), t-Bu hydrazine HCl, and 0.5 g KOH in 40 mL ethanol were refluxed for 3 hours. Workup (EtAc) and recrystallization from cyclohexane gave a 0.58-g white solid, 31% yield, with melting point 153°C. NMR 7.86, 7.38 (4H, ABq, JAB=8.8Hz), 4.36 (2H, br.s, NH2), and 1.68 (9H, s, t-Bu).
4. 1-t-Bu-3-Aryl-4 Amino Pyrazolo[3.4-d] Pyrimidines
Compound 4a (PP1/AGL1872 [0.13 g]). Compound 3a and 3 mL formamide were heated at 180°C for 3 hours. To the cooled reaction water was added the precipitated solid that was filtered and recrystallized from benzene-hexane to give a 0.12 g white solid, 83% yield, with melting point 175°C. NMR 8.35 (1H, s, H2), 7.58, 7.32 (4H, ABq, JAB=8.2 Hz), 5.36 (2H, br.s, NH2), 2.43 (3H, s, methyl), and 1.84 (9H, s).
Compound 4b (PP2/AGL1879 [0.25 g]). Compound 3b and 4 mL formamide were heated at 180°C for 3 hours. Workup (CH2Cl2) and chromatography gave a 0.16 g white solid, 58% yield, with melting point 203°C. NMR 8.35 (1H, s, H2), 7.65, 7.50 (4H, ABq, JAB=8.2 Hz), 5.54 (2H, br.s, NH2), and 1.82 (9H, s).
Cells and Reagents
Swiss 3T3 cells (American Type Culture Collection No. 92), NIH 3T3 cells stably transfected with a constitutively active mutant (F527) of chicken c-src gene,7 and TRMP cells overexpressing human PDGF β-receptor9 were grown in DMEM, supplemented with 4.5 g/L glucose, glutamine, antibiotics, and 10% FCS (Life Technologies). PAE cells, stably transfected with PDGF β-receptor (PAE/PDGFβ-R cells), PDGF α-receptor (PAE/PDGFα-R cells), or FGF receptor-1 (PAE/FGFR-1 cells), were kindly provided by Dr Lena Claesson-Welsh (Department of Medical Chemistry, Uppsala University, Uppsala, Sweden) and were grown in DMEM/Ham’s F12 1:1 medium, supplemented as above. VEGF receptor KDR-expressing cells (PAE/KDR cells) were previously described.30 31 Human coronary artery endothelial cells (HCAECs) and HCASMCs were purchased from Clonetics and grown in EBM medium (supplemented with EGM-MV SingleQuots) or in SmBM medium (supplemented with SmGM-2 SingleQuots), respectively (Clonetics).
Recombinant human Src kinase (p60c-src) was purified from overexpressing Sf9 cells as described earlier.32 The monoclonal anti-Src antibody (OPO 7L) and the agarose-coupled anti-Src antibody (OPO 7A) were obtained from Dianova. Antibodies against phosphotyrosine (RC20H, horseradish peroxidase conjugated) were from Transduction Laboratories. For detection of PDGF β-receptors by immunoblotting, antibody DIG-17 or antibody 06-498 from Upstate Biotechnology was used. The same antibody, 06-498, was used for immunoprecipitation of PDGF β-receptor33 and NEF antiserum for immunoprecipitation of the VEGF receptor KDR,30 respectively. Wheat germ agglutinin coupled to agarose was obtained from Pharmacia, and [γ-32P]ATP was purchased from NEN-DuPont. The “optimal Src substrate” peptide AEEEIYGEFEAKKKK34 was kindly provided by U. Engström (Ludwig Institute for Cancer Research, Uppsala, Sweden). Human recombinant PDGF-BB, PDGF-AA, basic FGF (bFGF), acidic FGF (aFGF), and IGF-I were obtained from TEBU, and human recombinant EGF was from Biomol. Human recombinant VEGF165 was a kind gift from D. Gospodarowicz (Chiron, Emeryville, Calif).
In Vitro Phosphorylation Assay of Src
For in vitro assays of Src kinase activity, 0.2 to 0.4 U of p60c-src were incubated in the absence or presence of the inhibitors or vehicle (DMSO, final concentration 1%) and with the optimal Src substrate peptide (30 μmol/L) in assay buffer containing (in mmol/L) HEPES (pH 7.4) 50, MnCl2 10, and DTT 2 (final concentrations) for 15 minutes on ice. Then, [γ-32P]ATP was added (0.25 μCi, 10 μmol/L final concentration), and the mixture (25 μL final volume) was transferred to 30°C and incubated for 30 minutes. Thereafter, the reaction was quenched by adding 5 μL of 120 mmol/L EDTA, and 10 μL of the reaction mixture was spotted on Whatman P81 phosphocellulose paper, dried for 30 minutes, washed 4 times for 10 minutes with 75 mmol/L phosphoric acid, and dried. Radioactivity in the spots was quantified with a phosphor imager (Molecular Imager GS250, Bio-Rad). Background correction was made using reactions without peptide substrate. This assay was also applied to PDGF β-receptor preparations (see below) to reveal possible Src-like kinase activity in the preparation. Preparations (5 μL) were analyzed as above. IC50 values were obtained by fitting the data points using a nonlinear regression program (Graph III).
In Vitro Phosphorylation Assays of the PDGF β-Receptor, KDR, IGF-I Receptor, and FGF Receptor-1
In vitro autophosphorylation of PDGF β-receptor with purified Swiss 3T3 cell membranes was carried out as described.7 To test the effect of inhibitors on partially purified PDGF β-receptor from TRMP cell membranes, aliquots of a receptor preparation obtained as in Reference 99 were preincubated with the compounds or vehicle (DMSO, final concentration 1%) for 15 minutes on ice in the presence of 50 mmol/L HEPES (pH 7.4), 5 mmol/L MnCl2, and 100 μmol/L sodium orthovanadate (final concentrations). Then, PDGF-BB (or the respective solvent containing 0.5 mol/L NaCl, 20 mmol/L HEPES [pH 7.4], and 5 mg/mL BSA) was added for 20 minutes on ice, and subsequently, autophosphorylation was allowed in the presence of [γ-32P]ATP (3.5 μCi, 10 μmol/L final concentration, final volume 25 μL) for 10 minutes. The reaction was quenched by addition of SDS-PAGE sample buffer, and receptor phosphorylation was evaluated by SDS-PAGE followed by phosphor imager analysis of the fixed, stained, and dried gels.
For in vitro phosphorylation of PDGF β-receptor, FGF receptor-1 and KDR expressed in PAE cells, subconfluent PAE/PDGFβ-R cells, PAE/FGFR1 cells, and PAE/KDR cells were used as previously described.30 In addition, primary HCASMCs and HCAECs were used in the same way to analyze the PDGF β-receptor, FGF receptor-1, and IGF-I receptor. After a 10-minute preincubation with different concentrations of PP1/AGL1872, 100 μmol/L Na3VO4 was added 5 minutes before stimulation of cells at 37°C with 50 ng/mL PDGF-BB (5 minutes), 50 ng/mL bFGF (10 minutes), 50 ng/mL IGF-I (2 minutes), or 50 ng/mL VEGF for 5 minutes, respectively. Immunoprecipitation was performed with a PDGF β-receptor–specific antiserum (06–498), a monoclonal chicken anti-FGF receptor antiserum (kindly provided by Sabine Werner, Max-Planck-Institute for Biochemistry, Martinsried/Munich, Germany), an anti–IGF-I receptor-specific antiserum (Santa Cruz Biotechnology, catalog No. sc-713), or an antiserum recognizing KDR (NEF). Immunoprecipitates immobilized on protein A–Sepharose CL 4B (Pharmacia) were used for an immune complex–kinase assay, which was carried out for 7 minutes at room temperature in 25 μL using the previously described buffer. After SDS-PAGE, gels were incubated for 30 minutes in 2.5% glutaraldehyde, washed twice for 15 minutes in 10% acetic acid/40% methanol, treated for 1 hour at 55°C in 1 mol/L KOH, washed 3 times for 20 minutes in 10% acetic acid/40% methanol, dried, and exposed to Hyperfilm MP (Amersham).
Protein Phosphorylation in Intact Cells
To analyze Src-dependent tyrosine phosphorylation in intact cells, NIH3T3 fibroblasts transfected with a cDNA for chicken p60srcF527 were treated in 24-well plates for 2 hours under serum-free conditions with the inhibitors or corresponding vehicle (DMSO, final concentration 0.3% to 1%). Thereafter, the cells were washed 3 times with PBS and extracted with lysis buffer as described.7 Lysate protein (10 μg) was subjected to SDS-PAGE, immunoblotting using RC20H anti-phosphotyrosine antibodies (1:2500, Transduction Laboratories), and detection with the enhanced chemiluminescence reagent (Amersham). To quantify inhibitory effects, major phosphotyrosine bands at ≈65 or 120 kDa were analyzed by scanning the films with a flatbed office scanner and evaluating intensity using the program NIH Image 1.57. To detect the Src phosphorylation level, lysates of respectively treated Src-transformed cells (1 nearly confluent well of a 6-well plate per point) were subjected to immunoprecipitation with agarose-coupled anti-Src OPO 7A antibodies (Dianova). The immunoprecipitates were washed 3 times with lysis buffer and subjected to SDS-PAGE and immunoblotting with anti-phosphotyrosine antibodies or anti-Src antibodies.
For demonstration of tyrosine phosphorylation of PDGF β-receptor or KDR in transfected PAE cells and in HCASMCs or HCAECs, respectively, cells were starved overnight in serum-free medium and incubated for 10 minutes at 37°C with different concentrations of PP1/AGL1872 and for another 5 minutes with 100 μmol/L Na3VO4 before stimulation of cells with 50 ng/mL PDGF-BB or 50 ng/mL VEGF for 5 minutes at 37°C. After a wash with ice-cold PBS containing 100 μmol/L Na3VO4, cells were solubilized in lysis buffer as previously described.35 The cell lysates were centrifuged at 10 000g for 15 minutes, and the supernatants were used for immunoprecipitation with antisera recognizing the PDGF β-receptor (R3) or KDR (NEF). Immunoprecipitates were bound on protein A–Sepharose CL 4B, subjected to SDS-PAGE, and blotted onto a nitrocellulose membrane (Hybond C Extra, Amersham). Phosphorylated proteins were detected by immunoblotting using the horseradish peroxidase–conjugated anti-phosphotyrosine antibody RC20H (Transduction Laboratories) followed by a chemoluminescence-based detection system (enhanced chemiluminescence, Amersham).
PDGF β-receptor and EGF receptor autophosphorylation in intact Swiss 3T3 cells was estimated as described.7 Ligand-dependent receptor phosphorylation in PAE cells overexpressing PDGF β-receptor, PDGF α-receptor, or FGF receptor-1 was estimated similarly, by starving confluent cultures for 6 to 16 hours in serum-free medium, treating for 2 hours with inhibitors or vehicle, and stimulating for 10 minutes with 50 ng/mL PDGF-BB, 10 minutes with 50 ng/mL PDGF-AA, and 5 minutes with 100 ng/mL aFGF, respectively. Cell lysates (usually the lysates of 1 well of a 6-well plate per point) were incubated with 30 μL of lysis buffer–equilibrated wheat germ agarose beads for 1 hour with end-over-end rotation. The beads were washed 3 times with lysis buffer, and bound proteins were extracted by boiling with SDS-PAGE sample buffer and analyzed by immunoblotting.
Quantitative evaluation of the immunoblots was carried out by scanning appropriately exposed films with a flatbed scanner and analyzing the data with the program NIH Image 1.57.
Assessment of Cell Migration (Chemotaxis Assay)
The migratory response of HCASMCs and PAE/KDR cells was assessed using the modified Boyden chamber (Neuro Probe, Inc) and collagen-coated polycarbonate filters (Nuclepore) with pore diameters of 8 μm, as previously described.36 In brief, HCASMCs were starved for 24 hours in DMEM (Biochrom) containing 0.1% BSA before the assay. To detach cells from the culture flasks, an ice-cold 0.02% EDTA solution was added to the cells for ≈5 minutes. Then, cells were collected by centrifugation and resuspended in DMEM containing 0.1% BSA. PAE/KDR cells were cultivated and assayed in Hamm F12 medium containing 10% FCS. To the medium in the lower part of the Boyden chamber, PDGF-BB, VEGF, bFGF, or IGF-I (each 10 ng/mL respectively), or 20% FCS, was added. PP1/AGL1872 was added to both the upper and the lower chambers. Suspended cells were given 4 hours for migration after a preincubation period of 30 minutes in PP1/AGL1872. The number of cells that migrated without specific stimulation was referred to as 100% baseline migration (chemokinesis). The assay was performed in triplicate, and 5 medium-power fields per well were counted using a light microscope (Jenalab). Data were reproduced in at least 2 independent experiments.
Assessment of Cell Proliferation
Cell proliferation was assessed using a DNA synthesis assay based on measurement of [3H]thymidine incorporation. Assays with Swiss 3T3 cells were performed exactly as previously described, and HCAEC and HCASMC proliferation was assessed similarly.7 In brief, HCAECs or HCASMCs were seeded sparsely in 12-well culture dishes. After 24 hours, cells were washed 2 times with endothelial basal medium containing 1% FCS (HCAECs) or smooth muscle basal medium containing 1% FCS (HCASMCs), respectively, and incubated for an additional 48 hours with 1 renewal of medium. Cells were incubated for 15 minutes with different concentrations of PP1/AGL1872 (0.1, 1, and 10 μmol/L) or with the solvent DMSO alone and were stimulated with 3 ng/mL VEGF or with 15 ng/mL PDGF-BB for 20 hours, followed by addition of 0.25 μCi of [3H]thymidine per milliliter (Amersham) for 2 hours. Finally, precipitated DNA was quantitated by liquid scintillation counting.
PP1/AGL1872 Is a Potent Inhibitor of p60c-src and Src Activity in Intact Cells
Because the synthetic protocols for PP1 and PP2 are not available,27 both compounds were synthesized by us and named PP1/AGL1872 and PP2/AGL1879, respectively. We assayed PP1/AGL1872 and PP2/AGL1879 for their activity on recombinant p60c-src using the optimal Src peptide substrate, AEEEIYGEFEAKKKK.34 PP1/AGL1872 inhibits p60c-src in vitro with an IC50 of 0.2 μmol/L (Table 1⇓); ie, we found an IC50 ≈50 times higher than that reported for inhibition of p56Lck in vitro.27 PP2/AGL1879 had a somewhat lower activity in this assay with an IC50 of 0.3 μmol/L (Table 1⇓). Therefore, further analysis was carried out primarily with PP1/AGL1872; however, in all types of assays described below, PP2/AGL1879 yielded similar results (data not shown). In NIH3T3 cells transformed by expression of the constitutively active Src variant p60c-srcF527, highly elevated tyrosine phosphorylation of multiple proteins is detectable as compared with untransfected NIH3T3 cells (see also Reference 3737 . When Src-transformed NIH3T3 cells were exposed to PP1/AGL1872, tyrosine phosphorylation was inhibited in a dose-dependent manner (Figure 2⇓). Also, p60c-srcF527 autophosphorylation detected in anti-Src immunoprecipitates from these cells was inhibited in a dose-dependent manner (Figure 2⇓). Quantification revealed IC50 values of 0.7 and 0.9 μmol/L for both effects, respectively (Table 1⇓), ie, close to the value reported for inhibition of p56Lck by PP1 in intact T cells.27 Taken together, PP1/AGL1872 inhibits p60c-src in vitro and in intact cells.
PP1/AGL1872 Potently Inhibits PDGF-BB–Induced Migration of Human Vascular Smooth Muscle Cells but Not VEGF-Induced Migration of Human Endothelial Cells
To study the role of Src-family kinases in growth factor–induced cell migration, we used a modified Boyden chamber assay to measure PDGF-stimulated migration of HCASMCs. PDGF-BB (10 ng/mL) stimulates HCASMC migration to about twice the baseline value (Figure 3A⇓). PP1/AGL1872 exerts a strong and dose-dependent inhibitory effect on this process. At a concentration of 2 μmol/L, the PDGF-BB–stimulated value is reduced to 69% of the stimulated control value (IC50=3 μmol/L), and PP1/AGL1872 at 10 μmol/L abolishes the majority of the stimulatory effect of PDGF-BB. PP1/AGL1872 does not affect the baseline value of HCASMC migration in the absence of PDGF-BB. As compared with the selective PDGF β-receptor blocker AG1295 (IC50=7 μmol/L; J. Waltenberger et al, unpublished data), PP1/AGL1872 is more potent in this assay. In contrast, PP1/AGL1872 has only a weak effect on VEGF-induced migration of endothelial cells (HCAECs). Even at a concentration of 10 μmol/L, the VEGF-induced migration of these cells was inhibited by ≈30% (Figure 3B⇓).
In addition, we have compared the inhibitory effect of PP1/AGL1872 on PDGF-BB–stimulated migration of HCASMCs with its effect on migration stimulated by IGF-I, bFGF, and FCS. The latter 2 agents had little effect on HCASMC migration in our assays (not shown), whereas IGF-I treatment resulted in a small but reproducible migratory response (Figure 3C⇑). Given the small stimulation by IGF-I, sensitivity to PP1/AGL1872 is difficult to evaluate. Within these limits, the available data suggest that PP1/AGL1872 exerts only very weak if any inhibitory action on IGF-I–stimulated migration of HCASMCs (IC50≫10 μmol/L, Figure 3C⇑).
PP1/AGL1872 Inhibits PDGF β-Receptor Kinase
We tested the effect of PP1/AGL1872 on PDGF β-receptor kinase autophosphorylation. Interestingly, we found PP1/AGL1872 to potently inhibit PDGF β-receptor autokinase activity in Swiss 3T3 cell membranes (IC50=0.1 μmol/L, Figure 4A⇓, upper panel, Table 2⇓), in preparations of partially purified human PDGF β-receptor (IC50=0.1 μmol/L, Figure 4⇓, lower panel, Table 2⇓) and in intact Swiss 3T3 cells (IC50=0.5 μmol/L, Figure 4B⇓). Because heterologous phosphorylation of PDGF β-receptor by Src-family kinases has been previously described,26 we wondered whether this observed inhibition could be secondary to Src-family kinase inhibition. Therefore, we tested the possible presence of Src-family kinases in the semipurified PDGF β-receptor fraction from TRMP cells using the Src kinase in vitro assay with the optimal Src peptide substrate peptide (see Materials and Methods). Only a very small activity toward the Src substrate peptide could be detected in these preparations, and the activity was completely suppressed by 10 μmol/L AG1296, the selective PDGF receptor blocker (Figure 5⇓). In contrast, as described earlier,7 recombinant p60c-src is not affected by AG1296 (Figure 5⇓). Thus, the small activity toward the Src substrate peptide in the PDGF β-receptor preparation represents some minor activity of PDGF receptor toward this substrate. We conclude that the PDGF β-receptor preparations used in our assays are devoid of Src-family kinase activity and, therefore, the inhibition of PDGF β-receptor autokinase activity by PP1/AGL1872 is not mediated by Src kinase inhibition, but is a direct effect.
We further investigated whether inhibition of the PDGF β-receptor kinase by PP1/AGL1872 would occur with some specificity or whether other receptor tyrosine kinases would be affected similarly. EGF receptor autophosphorylation in intact Swiss 3T3 cells was completely refractory to inhibition by PP1/AGL1872 (Figure 4B⇑). Also, EGF receptor autophosphorylation in Swiss 3T3 cell membranes was inhibited with an IC50 of >3 μmol/L (ie, at least 30 times higher than that for inhibition of PDGF β-receptor autophosphorylation in the same type of assay [data not shown]).
Using an in vitro kinase assay, PP1/AGL1872 inhibits PDGF-BB–stimulated phosphorylation of the PDGF β-receptor in receptor immunoprecipitates from PAE/PDGFβ-R cells in a dose-dependent manner (IC50<1 μmol/L, Figure 6A⇓, Table 2⇑). In contrast, VEGF-induced phosphorylation of KDR in PAE/KDR cells remained essentially unaffected by PP1/AGL1872 (Figure 6B⇓). Importantly, VEGF-stimulated autophosphorylation of KDR in intact HCAECs is also resistant to PP1/AGL1872, whereas PDGF-BB–stimulated PDGF receptor phosphorylation in HCASMCs is potently inhibited by PP1/AGL1872 (Figure 6C⇓ and 6D⇓). Subjecting PAE cells overexpressing human PDGF β-receptor, PDGF α-receptor, or FGF receptor-1 to PP1/AGL1872 treatment revealed resistance of FGF receptor-1 to the compound (Figure 7A⇓); rather than inhibition we reproducibly observed enhancement of FGF receptor-1 phosphorylation in the presence of PP1/AGL1872. The PDGF β-receptor was similarly susceptible to inhibition in this background (Table 2⇑), as in Swiss 3T3 cells or HCASMCs, and PDGF α-receptor was inhibited at high concentrations of PP1/AGL1872 but was clearly less susceptible than the PDGF β-receptor (Figure 7A⇓, Table 2⇑).
Unlike the PDGF β-receptor, the FGF receptor-1 showed a high baseline level of tyrosine phosphorylation in HCASMCs. Stimulation with bFGF (50 ng/mL) had only a very weak effect on FGF receptor-1 phosphorylation. In the presence of PP1/AGL1872, the phosphorylation pattern of the FGF receptor-1 was not affected, and there was definitely no reduction below the unstimulated baseline level (Figure 7B⇑). In contrast, IGF-I (50 ng/mL) strongly stimulated the autophosphorylation of the IGF-I receptor, and the presence of PP1/AGL1872 (>1 μmol/L) led not to inhibition, but to a further enhancement of tyrosine phosphorylation (Figure 7C⇑).
In summary, PP1/AGL1872 inhibits PDGF β-receptor kinase with efficacy similar to that of Src-family kinase activity in vitro and in intact cells, whereas PDGF α-receptor is less sensitive and the agent has little effect on EGF receptor kinase, VEGF receptor kinase KDR, FGF receptor-1, and IGF-I receptor kinase.
PP1/AGL1872 Differentially Affects Cell Proliferation Driven by Different Growth Factors
Given the putative role of Src-family kinases for downstream signaling of PDGF β-receptor and the susceptibility of both Src-family kinases and PDGF β-receptor to PP1/AGL1872, we expected potent inhibition of PDGF-stimulated proliferation by the compound. PDGF-BB (15 ng/mL) stimulated [3H]thymidine incorporation of HCASMCs strongly (Figure 8A⇓) and PP1/AGL1872 inhibited DNA synthesis potently, with only a minor effect also on basal DNA synthesis (Figure 8A⇓) (mean IC50=1 μmol/L in multiple independent experiments). VEGF-induced stimulation of DNA synthesis in HCAECs was comparable with that of PDGF-BB in HCASMCs, and PP1/AGL1872 inhibited this stimulatory effect with an IC50 of ≈3 μmol/L (Figure 8B⇓). PP1/AGL1872 did not affect cell viability within the studied concentration range of up to 30 μmol/L. Also, adhesion of endothelial cells and smooth muscle cells to culture flasks was not affected by PP1/AGL1872. Cells neither detached nor changed in shape. When DNA synthesis of HCASMCs was stimulated with IGF-I (10 ng/mL), stimulation was only weak and the inhibitory effect of PP1/AGL1872 was difficult to judge but apparently less pronounced than that toward PDGF stimulation (IC50 of ≈4 μmol/L). Likewise, FCS (20%)–stimulated proliferation of HCASMCs was inhibited by PP1/AGL1872, with a somewhat higher IC50 of ≈3 μmol/L (Table 3⇓).
PP1/AGL1872 Is a Dual Inhibitor of Src Kinase and PDGF β-Receptor Kinase Activity
Because activation of Src-family kinases is an important downstream mediator of PDGF β-receptor signaling, we wondered whether selective Src inhibitors might be used to interfere with PDGF-mediated mitogenesis and migration of HCASMCs in the context of atherosclerosis and restenosis. Also, the utilization of selective Src kinase inhibitors is expected to yield important information as to the extent to which Src is involved in the signaling of different receptor tyrosine kinases implicated in the proliferation of vascular cells as PDGF β-receptor and the VEGF receptor KDR. The compound PP1 has been previously described as a selective inhibitor for the Src-family kinases p56Lck and FynT27 and was likely also to inhibit p60c-src. We therefore examined its inhibitory activity on this kinase. As expected, we found PP1/AGL1872 to potently inhibit p60c-src in vitro and Src-dependent tyrosine phosphorylation in intact cells (Figure 2⇑). Because Src has been demonstrated to be essential for PDGF β-receptor signaling,26 we examined the ability of PP1/AGL1872 to inhibit PDGF-dependent proliferation and migration of HCASMCs. We found PP1/AGL1872 to be a potent inhibitor of both processes. On detailed examination we found, to our surprise, that PP1/AGL1872 potently inhibits the PDGF β-receptor kinase itself but has no inhibitory activity toward KDR, FGF receptor-1, IGF-I receptor, or, at least in intact cells, EGF receptor. Also, PP1/AGL1872 is only moderately effective on the PDGF α-receptor. These observations reveal that PP1/AGL1872 is not a selective Src kinase inhibitor but interacts with Src kinase and PDGFβ-R kinase domain with similar efficacy. The specificity profile of the compound, which is expected to reflect differential affinity to the ATP-binding pockets of the various kinases,9 gives rise to the interesting speculation that the PDGF β-receptor kinase ATP-binding pocket may be more related to the Src kinase ATP-binding pocket than to that of the other receptors, including the PDGF α-receptor. This point requires further attention, including future experiments to obtain detailed kinetic data on inhibitor kinase interaction.
Dual Inhibition of PDGF β-Receptor Kinase and Src-Family Kinases as a Novel Concept to Prevent HCASMC Activation in Atherogenesis and Restenosis
In line with the insensitivity of KDR to PP1/AGL1872, the compound had little effect on VEGF-induced migration of endothelial cells. Also, we observed little effect of PP1/AGL1872 on the small increase of HCASMC migration in the presence of IGF-I, whereas PDGF-stimulated migration was effectively blocked. Thus, the inhibitory efficacy of PP1/AGL1872 on migration induced by various motogens seems to be related to the susceptibility of the receptor kinases to inhibition by the compound. Interestingly, we even observed an enhanced phosphorylation of the FGF receptor-1 and IGF-I receptor in the presence of PP1/AGL1872. One could speculate that negative regulators of receptor phosphorylation as protein-tyrosine phosphatases are positively regulated by Src-family kinases, leading to elevated receptor phosphorylation secondary to Src-family kinase inhibition. Activation of protein-tyrosine phosphatases by tyrosine-specific phosphorylation has been reported.38 39 40
VEGF-stimulated DNA synthesis of HCAECs was susceptible to inhibition by PP1/AGL1872 (IC50=3 μmol/L). Likewise, DNA synthesis induced by FCS or IGF-I in HCASMCs was inhibited by PP1/AGL1872, albeit to a somewhat lesser extent than PDGF-stimulated DNA synthesis in HCASMCs. Similar data have been obtained for DNA synthesis stimulated by various mitogens in Swiss 3T3 fibroblasts (not shown). It seems tempting to speculate that the PP1/AGL1872 susceptibility of DNA synthesis induced by growth factors the receptor kinases of which are resistant to the agent reflects the involvement of Src-family kinases in the downstream mitogenic signaling of these receptors. In line with this assumption, PDGF-stimulated DNA synthesis of HCASMCs is rather sensitive to PP1/AGL1872 (IC50=1 μmol/L), possibly as the result of dual inhibition of the PDGF β-receptor kinase itself and of downstream Src-family kinases. This interpretation of the data would imply that Src-family kinases are to some extent involved in mitogenic signaling of various growth factor receptors, including KDR, but not, or to a lesser extent, in the signaling events leading to cell migration/chemotaxis. Indeed, for the PDGF β-receptor, abrogation of receptor interaction with Src by mutating a Src phosphorylation site of the receptor leads to a shift of receptor signaling activity in that the mitogenic response is attenuated and the motogenic response is enhanced.26 These data are also in line with the hypothesis that Src is primarily involved in mitogenic signaling of the PDGF β-receptor and that the chemotactic response is not mediated by Src. Thus, the inhibitory effect of PP1/AGL1872 on migration observed by us is more likely to result from the direct effect of the compound on the receptor and not as a result of its anti-Src activity.
PDGF signaling inhibitors acting by different mechanisms have been shown to combat successfully PDGF-dependent pathological processes in restenosis.13 14 17 19 41 42 According to our data, PP1/AGL1872 potently interferes with PDGF-driven migration and proliferation of HCASMCs, including basal DNA synthesis, and differentially affects KDR signaling. It has little effect on VEGF-stimulated migration of PAE/KDR cells and no effect on basal DNA synthesis of HCAECs. The inhibitory effect on VEGF-stimulated DNA synthesis of HCAECs occurs at 10-fold higher concentrations than that on PDGF-stimulated DNA synthesis of HCASMCs. VSMC migration and proliferation, shown here to be both inhibited by PP1/AGL1872, are crucial events underlying neointima formation.43 It would, therefore, be worthwhile testing PP1/AGL1872 in different in vivo models to investigate the functional significance of a combined Src/PDGF β-receptor inhibition on neointima formation (ie, in the rat and the pig). It will be interesting to examine whether the dual inhibition is more potent than just PDGF β-receptor inhibition as achieved by using AG129516 to prevent vascular remodeling and restenosis.
This work was supported in part by a joint German-Israeli research project (DISMED87) grant (to A.L., F.-D.B., and J.W.), a grant from The Israel Science Foundation (The Israel Academy of Sciences and Humanities) (to A.L.), a grant from the Max-Planck-Society (to F.-D.B.), and a grant from the Land Baden-Württemberg (Landesforschungsschwerpunkt “Wachstumsfaktor-Modulation als Therapieprinzip”) (to J.W.). We are grateful to Ulla Engström for synthesis of the optimal Src peptide and to Dr L. Claesson-Welsh for kind provision of various cell lines.
- Received June 12, 1998.
- Accepted April 15, 1999.
- © 1999 American Heart Association, Inc.
Levitzki A, Gazit A. Tyrosine kinase inhibition: an approach to drug development. Science. 1995;267:1782–1788.
Fry DW. Recent advances in tyrosine kinase inhibitors. Annu Rep Med Chem. 1996;31:151–160.
Kovalenko M, Gazit A, Bohmer A, Rorsman C, Ronnstrand L, Heldin CH, Waltenberger J, Bohmer FD, Levitzki A. Selective platelet-derived growth factor receptor kinase blockers reverse sis-transformation. Cancer Res. 1994;54:6106–6114.
Strawn LM, Mann E, Elliger SS, Chu LM, Germain LL, Niederfellner G, Ullrich A, Shawver LK. Inhibition of glioma cell growth by a truncated platelet-derived growth factor-beta receptor. J Biol Chem. 1994;269:21215–21222.
Buchdunger E, Zimmermann J, Mett H, Meyer T, Muller M, Regenass U, Lydon NB. Selective inhibition of the platelet-derived growth factor signal transduction pathway by a protein-tyrosine kinase inhibitor of the 2-phenylaminopyrimidine class. Proc Natl Acad Sci U S A. 1995;92:2558–2562.
Waltenberger J. Modulation of growth factor action: implications for the treatment of cardiovascular diseases. Circulation. 1997;96:4083–4095.
Hart CE, Clowes AW. Platelet-derived growth factor and arterial response to injury. Circulation. 1997;95:555–556.
Fukumoto Y, Shimokawa H, Kozai T, Kadokami T, Kuwata K, Owada MK, Shiraishi T, Kuga T, Egashira K, Takeshita A. Tyrosine kinase inhibitor suppresses the (re)stenotic changes of the coronary-artery after balloon injury in pigs. Cardiovasc Res. 1996;32:1131–1140.
Waltenberger J, Kovalenko M, Böhmer F, Gazit A, Hombach V, Levitzki A. Characterization of receptor-tyrosine-kinase-inhibitors: selectivity as a molecular basis for the treatment of restenosis. Eur Heart J. 1994;15(suppl):427. Abstract.
Banai S, Wolf Y, Golomb G, Pearle A, Waltenberger J, Fishbein I, Schneider A, Golomb G, Gazit A, Perez LS, Huber R, Lazarovichi G, Levitzki A, Gertz SD. PDGF-receptor tyrosine kinase blocker AG1295 selectively attenuates smooth muscle cell growth in vitro and reduces neointima formation after balloon angioplasty in swine. Circulation. 1998;97:1960–1969.
Claesson-Welsh L. Platelet-derived growth factor receptor signals. J Biol Chem. 1994;269:32023–32026.
Twamley SG, Pepperkok R, Ansorge W, Courtneidge SA. The Src family tyrosine kinases are required for platelet-derived growth factor-mediated signal transduction in NIH 3T3 cells. Proc Natl Acad Sci U S A. 1993;90:7696–7700.
Hanke JH, Gardner JP, Dow RL, Changelian PS, Brissette WH, Weringer EJ, Pollok K, Connelly PA. Discovery of a novel, potent, and src family-selective tyrosine kinase inhibitor: study of lck-dependent and fynt-dependent T-cell activation. J Biol Chem. 1996;271:695–701.
Southwick PL, Dhawan B. Preparation of 4,6-diaminopyrazolo[3,4-d]pyrimidines with variation in substitution at the 1 and 3 positions. J Heterocyclic Chem. 1975;12:1199–1203.
Waltenberger J, Claesson-Welsh L, Siegbahn A, Shibuya M, Heldin C-H. Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem. 1994;269:26988–26995.
Westermark B, Siegbahn A, Heldin C-H, Claesson-Welsh L. B-type receptor for platelet-derived growth factor mediates a chemotactic response by means of ligand-induced activation of the receptor protein-tyrosine kinase. Proc Natl Acad Sci U S A. 1990;87:128–132.
Bjorge JD, Bellagamba C, Cheng HC, Tanaka A, Wang JH, Fujita DJ. J Biol Chem. 1995;270:24222–24228.
Claesson-Welsh L, Hammacher A, Westermark B, Heldin C-H, Nistér M. Identification and structural analysis of the A type receptor for platelet-derived growth factor: similarities with the B type receptor. J Biol Chem. 1989;264:1742–1747.
Waltenberger J, Mayr U, Pentz S, Hombach V. Functional upregulation of the vascular endothelial growth factor receptor KDR by hypoxia. Circulation. 1996;94:1647–1654.
Agbotounou WK, Levitzki A, Jacquemin SA, Pierre J. Effects of tyrphostins on the activated c-src protein in NIH/3T3 cells. Mol Pharmacol. 1994;45:922–931.
Uchida T, Matozaki T, Noguchi T, Yamao T, Horita K, Suzuki T, Fujioka Y, Sakamoto C, Kasuga M. Insulin stimulates the phosphorylation of Tyr(538) and the catalytic activity of Ptp1c, a protein tyrosine phosphatase with SRC homology-2 domains. J Biol Chem. 1994;269:12220–12228.
Vambutas V, Kaplan DR, Sells MA, Chernoff J. Nerve growth factor stimulates tyrosine phosphorylation and activation of Src homology-containing protein-tyrosine phosphatase 1 in PC12 cells. J Biol Chem. 1995;270:25629–25633.
Vogel W, Lammers R, Huang J, Ullrich A. Activation of a phosphotyrosine phosphatase by tyrosine phosphorylation. Science. 1993;259:1611–1614.
Banai S, Schneider A, Pearle A, Gertz SD, Golomb G, Levitzki A, Wolff Y. Effect of selective PDGF-receptor versus non-selective protein tyrosine kinase blockers on aortic SMCs and endothelial cells proliferation. J Am Coll Cardiol. 1996;27:255A. Abstract.
Sirois MG, Simons M, Edelman ER. Antisense oligonucleotide inhibition of PDGFR-β receptor subunit expression directs suppression of intimal thickening. Circulation. 1997;95:669–676.
Waltenberger J, Akyürek ML, Aurivillius M, Wanders A, Larsson E, Fellström B, Funa K. Ischemia-induced transplant arteriosclerosis in the rat: induction of peptide growth factor expression. Arterioscler Thromb Vasc Biol. 1996;16:1516–1523.