In vascular smooth muscle cells (VSMCs), platelet-derived growth factor (PDGF) plays a major role in inducing phenotypic switching from contractile to proliferative state. Importantly, VSMC phenotypic switching is also determined by the phosphorylation state/expression levels of insulin receptor substrate (IRS), an intermediary signaling component that is shared by insulin and IGF-I. To date, the roles of PDGF-induced key proliferative signaling components including Akt, p70S6kinase, and ERK1/2 on the serine phosphorylation/expression of IRS-1 and IRS-2 isoforms remain unclear in VSMCs. We hypothesize that PDGF-induced VSMC proliferation is associated with dysregulation of insulin receptor substrates. Using human aortic VSMCs, we demonstrate that prolonged PDGF treatment led to sustained increases in the phosphorylation of protein kinases such as Akt, p70S6kinase, and ERK1/2, which mediate VSMC proliferation. In addition, PDGF enhanced IRS-1/IRS-2 serine phosphorylation and downregulated IRS-2 expression in a time- and concentration-dependent manner. Notably, phosphoinositide 3-kinase (PI 3-kinase) inhibitor (PI-103) and mammalian target of rapamycin inhibitor (rapamycin), which abolished PDGF-induced Akt and p70S6kinase phosphorylation, respectively, blocked PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. In contrast, MEK1/ERK inhibitor (U0126) failed to block PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. PDGF-induced IRS-2 downregulation was prevented by lactacystin, an inhibitor of proteasomal degradation. Functionally, PDGF-mediated IRS-1/IRS-2 dysregulation resulted in the attenuation of insulin-induced IRS-1/IRS-2-associated PI 3-kinase activity. Pharmacological inhibition of PDGF receptor tyrosine kinase with imatinib prevented IRS-1/IRS-2 dysregulation and restored insulin receptor signaling. In conclusion, strategies to inhibit PDGF receptors would not only inhibit neointimal growth but may provide new therapeutic options to prevent dysregulated insulin receptor signaling in VSMCs in nondiabetic and diabetic states.
- platelet-derived growth factor
- insulin resistance
- phosphoinositide 3-kinase
- extracellular signal-regulated kinase-1/2
platelet-derived growth factor (PDGF) is a potent mitogen for vascular smooth muscle cells (VSMCs), and it plays a major role in the pathogenesis of atherosclerosis and restenosis after angioplasty (6, 25, 41, 50). During early atherogenesis, the vessel wall exhibits synthetic and secretory activities for a number of cytokines and growth factors, which modulate the phenotypic characteristics of VSMCs through autocrine and paracrine mechanisms of action (15, 25, 50). In particular, acute vascular injury promotes increased synthesis and release of PDGF from several vascular cells, including endothelial cells, VSMCs, activated monocytes, and monocyte-derived macrophages (4, 25, 51). Increased expression of PDGF and PDGF receptors has also been observed in human coronary arteries after angioplasty (62, 63). Of importance, PDGF has the ability to induce phenotypic switching in VSMCs from the contractile state to the proliferative state, and this has been attributed to the unique property of phenotypic plasticity in VSMCs (41). Studies by several investigators suggest that VSMC phenotypic switching is also determined by the phosphorylation state or the expression levels of insulin receptor substrate (IRS) (23, 24, 36, 37, 46, 64), a key intermediary component that is shared by insulin and insulin-like growth factor I (IGF-I) receptor signaling. To date, the role of PDGF on insulin receptor signaling components has not been examined in VSMCs.
Under physiological conditions, an increase in IRS-1 tyrosine phosphorylation and the consequent downstream activation of IRS-1-associated phosphoinositide 3-kinase (PI 3-kinase) signaling in response to insulin or IGF-I maintain the expression of contractile proteins in VSMCs (23, 24, 36, 64). In addition, IRS-1-mediated recruitment or sequestration of protein tyrosine phosphatase (SHP-2) upon IGF-I stimulation results in the blockade of extracellular signal-regulated kinase (ERK)-mediated VSMC proliferation (24, 46). Thus, site-specific IRS-1 tyrosine phosphorylation and IRS-1/SHP-2 interaction promote the contractile phenotype while opposing the proliferative phenotype in VSMCs (24, 46). Furthermore, enhanced IRS-1 serine phosphorylation and/or degradation in VSMCs results in the downregulation of contractile proteins, including smooth muscle α-actin, smooth muscle myosin heavy chain, and calponin (36). In this regard, activation of mammalian target of rapamycin (mTOR)/p70S6kinase as evidenced in serum-incubated VSMCs of the synthetic phenotype (36, 37) promotes IRS-1 serine phosphorylation/degradation to diminish the expression of contractile proteins. Contrarily, pharmacological inhibition of p70S6kinase phosphorylation by rapamycin restores contractile protein expression in VSMCs by attenuating IRS-1 serine phosphorylation and enhancing IRS-1 tyrosine phosphorylation (36, 37). The role of IRS-1 in controlling VSMC phenotypic switching mechanism is further supported by IRS-1 overexpression and gene-silencing studies (46). While ectopic expression of IRS-1 cDNA inhibits ERK-mediated VSMC proliferation and maintains the contractile phenotype, IRS-1 silencing by short hairpin RNA enhances ERK phosphorylation to promote VSMC proliferation. In particular, IRS-1 downregulation allows IGF-I to promote ERK-mediated VSMC proliferation via SHPS-1/SHP-2/Src/Shc/Grb2 signaling complex formation (46). Since PDGF stimulation or IRS-1 serine phosphorylation/degradation diminishes contractile protein expression while enhancing VSMC proliferation, it is important to determine whether PDGF exerts regulatory effects on IRS isoforms (IRS-1 and IRS-2) in VSMCs.
Previously, we and several other investigators have shown that PDGF-induced VSMC proliferation is mediated by activation of PI 3-kinase/Akt and Ras/Raf/MEK1 signaling pathways (2, 7, 35). In brief, PDGF induces PDGF receptor tyrosine phosphorylation to recruit the p85 regulatory subunit followed by activation of the p110 catalytic subunit of PI 3-kinase. Activated PI 3-kinase induces the phosphorylation of downstream effectors such as Akt, mTOR, and p70S6kinase (2, 7). In addition, PDGF receptor tyrosine phosphorylation activates Ras/Raf/MEK1 signaling to phosphorylate ERK1/2 (35). Previous studies using different cell types have shown that protein kinases such as Akt, p70S6kinase, and ERK1/2 induce serine phosphorylation of IRS isoforms (IRS-1 and IRS-2) or downregulate IRS protein expression via a ubiquitin-proteasome-dependent pathway (22, 39, 48, 57, 59). However, the role of PDGF-induced key proliferative signaling components on IRS-1/IRS-2 serine phosphorylation or expression has not been examined in VSMCs.
In the present study, we tested the hypothesis that PDGF-induced VSMC proliferation is associated with dysregulated insulin receptor signaling. Using human aortic VSMCs, we determined the effects of prolonged PDGF treatment on 1) the phosphorylation of key proliferative signaling components such as Akt, p70S6kinase, and ERK1/2; and 2) the serine phosphorylation and expression levels of IRS-1 and IRS-2. Using specific chemical inhibitors, we examined the roles of PDGF-induced key proliferative signaling components on IRS-1 and IRS-2 serine phosphorylation and expression levels. To examine the functional consequences of IRS-1/IRS-2 dysregulation, we determined the effects of PDGF pretreatment on insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity. In addition, we determined the effects of imatinib, a PDGF receptor tyrosine kinase inhibitor, on PDGF-induced proliferative response and PDGF dysregulation of insulin receptor signaling in VSMCs.
MATERIALS AND METHODS
Recombinant human PDGF-BB was purchased from R&D Systems (Minneapolis, MN). Human insulin (Novolin R) and imatinib mesylate (Gleevec) were obtained from Hershey Medical Center Pharmacy. The primary antibodies for IRS-1, IRS-2, and phospho-ERK1/2 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The primary antibodies for phospho-IRS-1(Ser636/639), phospho-PDGF receptor-β(Tyr751), phospho-Akt(Ser473), and phospho-p70S6kinase(Thr421/Ser424) were purchased from Cell Signaling (Beverly, MA). The primary antibody for phospho-serine was purchased from BD Biosciences (San Jose, CA). The primary antibody for p85α (N-SH2, clone UB93-3) was purchased from Upstate Biotechnology/Millipore (Temecula, CA). The primary antibody for β-actin was purchased from Novus Biologicals (Littleton, CO). PI-103 and U0126 were purchased from Calbiochem (Gibbstown, NJ). Silica gel 60 TLC plates were purchased from EMD Biosciences (San Diego, CA). [γ-32P]ATP [specific (Sp) activity: 4,500 Ci/mmol] and [3H]thymidine (sp. activity: 6.7 Ci/mmol) were purchased from MP Biomedicals (Solon, OH). All other chemicals were from Fisher Scientific (Fair Lawn, NJ) or Sigma Chemical (St. Louis, MO).
Cell culture and treatments.
Human aortic VSMCs were purchased from Lonza (Allendale, NJ). Medium 231 and smooth muscle growth supplement (SMGS) were purchased from Invitrogen (Carlsbad, CA). VSMCs (passages 3 to 7) were maintained in medium 231, SMGS, and antibiotic-antimycotic solution in a humidified atmosphere of 95% air and 5% CO2. After the attainment of confluence (∼ 6–7 days), VSMCs were trypsinized, centrifuged, and seeded onto petri dishes or multiwell plates. The subconfluent VSMCs were subjected to SMGS deprivation, pretreatments with chemical inhibitors, or stimulation with PDGF [1 ng/ml (0.04 nM), 10 ng/ml (0.4 nM), 30 ng/ml (1.2 nM), or 100 ng/ml (4.0 nM)] or insulin (100 nM), as described in the legends to the respective figures.
Subconfluent VSMCs were maintained in medium devoid of SMGS (serum deprivation) for 6 days in the absence or presence of PDGF (30 ng/ml) or insulin (100 nM). To ensure stability of PDGF and its constant stimulus, we replaced the medium with fresh SMGS-free medium containing PDGF every 2 days, as previously described (55). In parallel, we replaced the medium with fresh SMGS-free medium containing insulin every 2 days. For treatments that also included the use of chemical inhibitors and PDGF, we changed the medium containing the respective agents every 2 days. After the 6-day incubation period, VSMCs were trypsinized and the changes in cell number were determined using hemocytometer.
[3H]thymidine incorporation studies.
SMGS-deprived (48 h) VSMCs were pretreated with the indicated concentrations of imatinib for 30 min and then exposed to 30 ng/ml PDGF for 24 h. During the last 4-h incubation period, the cells were labeled with [3H]thymidine (1 μCi/ml). After labeling, the cells were washed three times with ice-cold PBS and then exposed to 10% trichloroacetic acid (TCA; 10 min × 3). After complete removal of the TCA, the acid-insoluble material was extracted with 0.1 N NaOH, and the incorporation of [3H]thymidine into DNA was determined using liquid scintillation counter, as previously described (7).
VSMC lysates (10 μg protein each) were electrophoresed using precast 4–12% NuPage mini-gels (Invitrogen), and the resolved proteins were transferred to nitrocellulose membranes (Hybond C, GE Healthcare). The membranes were blocked, and probed with the indicated primary antibodies. After extensive washes, the immunoreactivity was detected using specific horseradish peroxidase-conjugated secondary antibodies followed by enhanced chemiluminescence (Amersham Biosciences), as previously described (5, 7). The protein bands were quantified using a Bio-Rad GS-800 calibrated densitometer.
VSMC lysates were obtained using buffer A that consisted of 50 mM Tris·HCl (pH 7.5), 0.1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 50 mM sodium fluoride, 10 mM sodium β-glycerophosphate, 5 mM sodium pyrophosphate, 1 mM sodium orthovanadate, protease inhibitor cocktail (Sigma), 0.1% β-mercaptoethanol, and 1 μM LR-microcystin. The lysates were sonicated (15 s × 4) and centrifuged at 14,000 rpm (4°C) for 10 min. The respective supernatants were then used for protein assays (Coomassie protein reagent, Pierce, Rockford, IL). The aliquots of supernatants (60 μg protein) were subjected to immunoprecipitation (4°C, overnight) with 2 μg each of anti-IRS-1 or anti-IRS-2 primary antibody that was preconjugated (2 h at 4°C) to Gammabind G Sepharose. The immunocomplexes were then washed with buffer A and TNE buffer [consisting of 10 mM Tris·HCl (pH 7.4) 150 mM NaCl, 5 mM EGTA, and 0.1 mM sodium orthovanadate] before PI 3-kinase assays, as previously described (5, 7).
In vitro PI 3-kinase assays.
PI 3-kinase assays were performed as described previously with slight modifications (5, 7). After immunoprecipitation of proteins using specific primary antibodies, the respective immunocomplexes were subjected to PI 3-kinase assays by incubation at 35°C for 10 min in the presence of 50 μl TNE buffer (pH 7.4), phosphatidylinositol substrate (20 μg/assay), and [γ-32P]ATP (10 μCi/assay). The reactions were stopped by addition of 20 μl of 6 N HCl and 160 μl of CHCl3/CH3OH (1:1). Subsequently, the assay tubes were vortexed for 20 s and centrifuged at 14,000 rpm (room temperature) for 5 min. The phospholipid-containing lower organic phase from the respective reaction tubes was recovered and spotted on to silica gel thin layer chromatography plates (that were preheated to 100°C for ∼1 h). The thin layer chromatography plates were then subjected to ascending chromatography using the freshly prepared solvent mixture (CHCl3/CH3OH/H2O/NH4OH: 60:47:11.3:2). Phosphatidylinositol 3-phosphate (PI3P) spots thus resolved were visualized and quantified by autoradiography and phosphorimager analyses (Molecular Dynamics, Sunnyvale, CA), respectively. As negative controls, mock immunoprecipitations were performed using lysis buffer, which revealed negligible formation of 32P-labeled PI3P.
Results are expressed as means ± SE values. Statistical analyses of the data were performed by one-way repeated-measures ANOVA followed by Bonferroni t-test. Values of P < 0.05 were considered statistically significant.
PDGF-induced increase in DNA synthesis and proliferation in VSMCs is dependent on PDGF receptor tyrosine kinase activation and PI 3-kinase/Akt, mTOR/p70S6kinase, and MEK1/ERK signaling.
As shown in Fig. 1A, treatment of VSMCs with 30 ng/ml PDGF for 24 h led to ∼ 4.2-fold increase in DNA synthesis, an index of VSMC proliferation. Pretreatment of VSMCs with 0.3 μM to 5 μM imatinib (11), a PDGF receptor tyrosine kinase inhibitor, produced significant diminutions in PDGF-induced DNA synthesis. Imatinib at 0.3 μM concentration produced ∼34% inhibition, whereas at 1 μM and 5 μM concentrations completely abolished PDGF-induced DNA synthesis.
In addition, we determined the effects of PI-103, rapamycin, and U1026 pretreatment on PDGF-induced DNA synthesis in VSMCs. PI-103 is an inhibitor of PI 3-kinase and diminishes downstream phosphorylation of Akt (10). Rapamycin is an inhibitor of mTOR and diminishes downstream phosphorylation of p70S6kinase (58). U0126 is an inhibitor of MEK1 and abrogates ERK1/2 phosphorylation (18). As shown in Fig. 1B, PDGF-induced DNA synthesis was significantly inhibited by 1 μM PI-103, 20 nM rapamycin, and 10 μM U0126.
To confirm whether PDGF-induced DNA synthesis is reflected by increases in VSMC proliferation, we incubated the VSMCs with PDGF for 6 days as described in materials and methods. As shown in Fig. 1C, exposure to 30 ng/ml PDGF for 6 days resulted in ∼ 3.3-fold increase in VSMC proliferation compared with basal conditions. Pretreatments with increasing concentrations of imatinib, PI-103, rapamycin, or U0126 led to concentration-dependent diminutions in PDGF-induced VSMC proliferation. Imatinib at 0.3 μM concentration produced ∼48% inhibition, whereas at 1 μM and 5 μM concentrations completely abolished PDGF-induced VSMC proliferation. PI-103 at 0.1 μM concentration produced ∼63% inhibition, whereas at 0.3 μM and 1 μM concentrations completely abolished PDGF-induced VSMC proliferation. In addition, rapamycin at 2 nM, 6 nM, and 20 nM concentrations diminished PDGF-induced VSMC proliferation by ∼26%, ∼74%, and 100%, respectively. Furthermore, U0126 at 0.3 μM produced ∼77% inhibition, whereas at 1 μM and 10 μM concentrations completely abolished PDGF-induced VSMC proliferation.
To determine whether insulin has the ability to induce VSMC proliferation, we incubated the VSMCs with 100 nM insulin for 6 days (in parallel to PDGF treatments for 6 days). Unlike PDGF, there was no significant increase in VSMC proliferation after exposure to insulin.
Together, these data indicate that PDGF receptor tyrosine kinase activation promotes VSMC proliferation, and it is mediated by PI 3-kinase/Akt, mTOR/p70S6kinase, and MEK1/ERK signaling.
Prolonged PDGF treatment produces a sustained increase in the phosphorylation of Akt, p70S6kinase, and ERK1/2 in VSMCs.
To determine whether PDGF-induced VSMC proliferation is associated with temporal changes in the activation of Akt, p70S6kinase, and ERK1/2, we determined the phosphorylation states of these key protein kinases. Previous studies using rat VSMCs have shown that PDGF produces sustained increases in Akt phosphorylation for up to 24 h (29). In the present study using human aortic VSMCs, prolonged PDGF treatment led to sustained increases in Akt phosphorylation for up to 48 h (Fig. 2). In parallel, PDGF produced a sustained increase in the phosphorylation of p70S6kinase and ERK1/2 in human aortic VSMCs. Together, the data shown in Figs. 1 and 2 indicate that PDGF-induced VSMC proliferation is mediated by intermediary activation of protein kinases such as Akt, p70S6kinase, and ERK1/2.
PDGF induces IRS-1 and IRS-2 serine phosphorylation and downregulates IRS-2 expression in VSMCs in a time- and concentration-dependent manner.
Previous studies examining the role of PDGF on insulin-induced glucose transport in adipocytes have shown that acute PDGF treatment promotes IRS-1 serine phosphorylation with or without IRS-1 degradation (22, 48, 57). The primary sequence of IRS-1 and IRS-2 contains over 30 potential serine/threonine phosphorylation sites (39). In the present study using human aortic VSMCs, we performed time dependency experiments to determine the effects of prolonged PDGF treatment on IRS-1 and IRS-2 serine phosphorylation and expression levels. To assess IRS-1 serine phosphorylation, we used VSMC lysates for immunoblot analysis using phospho-IRS-1(Ser636/639) primary antibody (42). To assess IRS-2 serine phosphorylation, we used VSMC lysates for immunoprecipitation using IRS-2 primary antibody followed by immunoblotting using phospho-serine primary antibody. Figure 3A shows that PDGF induced significant increases in the serine phosphorylation of IRS-1 for up to 48 h with maximal effects occurring at 20 min. In parallel, PDGF induced significant increases in the serine phosphorylation of IRS-2 for up to 6 h with maximal effects occurring between 20 min and 6 h. To determine whether IRS-1 and IRS-2 serine phosphorylation is associated with the changes in IRS-1 and IRS-2 expression levels, we performed immunoblot analysis using the aliquots of PDGF-treated VSMC lysates. Figure 3B shows that PDGF treatment did not produce significant changes in the expression of IRS-1 for up to 48 h. In parallel, PDGF induced significant decreases in the expression of IRS-2 by 48 ± 13%, 66 ± 14%, 79 ± 7%, and 82 ± 5% at 60 min, 6 h, 24 h, and 48 h, respectively.
Furthermore, we performed concentration dependency experiments to determine the effects of PDGF on IRS-1 and IRS-2 serine phosphorylation and expression levels. Figure 3C shows that PDGF at 10–100 ng/ml concentrations induced significant increases in IRS-1 serine phosphorylation for 24 h. In addition, PDGF at 1–100 ng/ml concentrations induced significant increases in IRS-2 serine phosphorylation for 1 h. Figure 3D shows that PDGF at 1–100 ng/ml concentrations did not affect IRS-1 expression for 24 h. In contrast, PDGF at 10–100 ng/ml concentrations produced significant downregulation of IRS-2 at the 24-h time point.
Together, time and concentration dependency experiments reveal that PDGF induces IRS-1 and IRS-2 serine phosphorylation and IRS-2 downregulation in VSMCs.
PDGF induces proteasomal degradation of IRS-2 in VSMCs.
To determine whether IRS-2 downregulation occurs due to proteasomal degradation, we employed the use of lactacystin, an inhibitor of 26S proteasome (59). Figure 4 shows that pretreatment of VSMCs with 10 μM lactacystin prevented PDGF-induced downregulation of IRS-2 expression. In parallel, lactacystin did not affect IRS-1 expression levels. Together, the data shown in Figs. 3 and 4 indicate that PDGF induces IRS-1 and IRS-2 serine phosphorylation and IRS-2 degradation in VSMCs.
PDGF receptor-mediated increases in Akt and p70S6kinase phosphorylation but not ERK1/2 phosphorylation mediate IRS-1 serine phosphorylation and IRS-2 downregulation in VSMCs.
To determine the causal link between PDGF receptor stimulation, PDGF-induced key protein kinases (Akt, p70S6kinase, and ERK1/2), and IRS-1/IRS-2 dysregulation in VSMCs, we used specific inhibitors of PDGF receptor tyrosine kinase and the respective upstream kinases for Akt, p70S6kinase, and ERK1/2. We chose 5 μM imatinib, 1 μM PI-103, 20 nM rapamycin, and 10 μM U0126 to inhibit PDGF receptor tyrosine kinase, PI 3-kinase, mTOR, and MEK1, respectively, as described in Fig. 1, A and B.
Figure 5A shows that pretreatment of VSMCs with imatinib abolished PDGF-induced PDGF receptor tyrosine phosphorylation. In addition, imatinib abolished PDGF-induced phosphorylation of Akt, p70S6kinase, and ERK1/2 (data not shown). Importantly, imatinib pretreatment significantly prevented PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. These data indicate that increases in PDGF receptor tyrosine phosphorylation and activation of PDGF-induced downstream signaling events contribute to IRS-1/IRS-2 dysregulation in VSMCs.
Figure 5B shows that pretreatment of VSMCs with PI-103 abolished PDGF-induced Akt phosphorylation. In addition, PI-103 pretreatment led to significant abrogation of PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. These data indicate that PDGF-induced Akt phosphorylation mediates IRS-1/IRS-2 dysregulation in VSMCs.
Figure 5C shows that pretreatment of VSMCs with rapamycin abolished PDGF-induced p70S6kinase phosphorylation. In addition, rapamycin pretreatment significantly blocked PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. These data indicate that PDGF-induced p70S6kinase phosphorylation mediates IRS-1/IRS-2 dysregulation in VSMCs.
Figure 5D shows that pretreatment of VSMCs with U0126 abolished PDGF-induced ERK1/2 phosphorylation. In contrast to the effects observed with all three inhibitors described in Fig. 5, A–C, pretreatment of VSMCs with U0126 did not significantly affect PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. These data indicate that PDGF-induced ERK1/2 phosphorylation does not mediate IRS-1/IRS-2 dysregulation in VSMCs. Furthermore, IRS-1 expression remained unaffected in the presence of all four inhibitors as shown in Fig. 5, A–D.
Together, the data shown in Figs. 2 through 5 reveal that PDGF receptor-mediated increases in Akt and p70S6kinase phosphorylation but not ERK1/2 phosphorylation mediate IRS-1 serine phosphorylation and IRS-2 degradation in VSMCs.
Prolonged PDGF treatment diminishes insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity in VSMCs.
Next, we determined the effects of PDGF-mediated IRS-1/IRS-2 dysregulation on insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity in VSMCs.
Figure 6, A and B, shows that incubation of VSMCs under serum-deprived conditions (48 h) followed by acute stimulation with insulin enhanced PI 3-kinase activity associated with IRS-1 and IRS-2 immunocomplexes. The immunoblots for the respective immunocomplexes showed that insulin stimulation increased the association of PI 3-kinase p85α regulatory subunit with IRS-1 and IRS-2.
Inclusion of PDGF in serum-free medium for 48 h led to significant decreases in IRS-1-associated and IRS-2-associated PI 3-kinase activity (P < 0.05) compared with the respective basal PI 3-kinase activity (Fig. 6, A and B). The immunoblots for the respective immunocomplexes showed that PDGF treatment inhibited the association of PI 3-kinase p85α regulatory subunit with IRS-1 and IRS-2. Imatinib pretreatment prevented PDGF-mediated decreases in the basal PI 3-kinase activities.
Inclusion of PDGF in serum-free medium for 48 h followed by acute challenge with insulin revealed that prolonged PDGF treatment attenuated insulin-induced IRS-1-associated PI 3-kinase activity by 72 ± 3% (Fig. 6A). This was due to PDGF-induced IRS-1 serine phosphorylation (Fig. 3, A and C) and impairment in the association of p85α regulatory subunit with serine phosphorylated IRS-1 (Fig. 3, A and C, and Fig. 6A). In addition, prolonged PDGF treatment resulted in complete inhibition of insulin-induced IRS-2-associated PI 3-kinase activity (Fig. 6B). This was due to PDGF-induced IRS-2 degradation (Fig. 3, B and D, Fig. 4, and Fig. 6B) and therefore the lack of recruitment of p85α regulatory subunit (Fig. 6B).
Together, these data reveal that prolonged PDGF treatment promotes IRS-1 serine phosphorylation and IRS-2 downregulation, which impair p85α recruitment to IRS-1 and IRS-2, respectively, and therefore diminish basal and insulin-induced increases in IRS-1 and IRS-2-associated PI 3-kinase activity in VSMCs.
Imatinib abolishes PDGF-induced proliferative signaling and prevents PDGF dysregulation of insulin receptor signaling in VSMCs.
From the results of the present study (Fig. 1, A and C, and Fig. 5A), the effects of imatinib are summarized as follows. 1) Imatinib abolished PDGF receptor tyrosine phosphorylation (Fig. 5A), PDGF-induced phosphorylation of Akt, p70S6kinase, and ERK1/2 (data not shown), and PDGF-induced VSMC proliferation (Fig. 1, A and C). 2) In addition, imatinib prevented PDGF-induced IRS-1 serine phosphorylation and PDGF-induced IRS-2 downregulation (Fig. 5A). As a sequel to these observations, Fig. 6, A and B, shows that imatinib pretreatment before PDGF exposure abrogated PDGF-induced diminutions in insulin-induced IRS-1 and IRS-2-associated PI 3-kinase activity. The ability of imatinib to inhibit PDGF receptor tyrosine kinase and improve insulin-induced PI 3-kinase activity was reflected by increased association of p85α regulatory subunit with IRS-1 and IRS-2.
Together, these data reveal that imatinib abolishes PDGF-induced proliferative signaling and PDGF-mediated dysregulation of insulin receptor signaling in VSMCs.
The present study provides the first direct evidence that PDGF-induced VSMC proliferation is associated with dysregulation of insulin receptor signaling components such as IRS-1 and IRS-2 (Fig. 7). The use of specific chemical inhibitors of PI 3-kinase, mTOR, and MEK1 reveals that the activation of protein kinases such as Akt, p70S6kinase, and ERK, respectively, mediates PDGF-induced VSMC proliferation. Nevertheless, the activation of Akt and p70S6kinase but not ERK promotes IRS-1 and IRS-2 serine phosphorylation and IRS-2 downregulation in VSMCs, as evidenced by the use of specific chemical inhibitors. Recent studies demonstrate that IRS-1 downregulation seen under hyperglycemic conditions allows IGF-I to promote VSMC proliferation via ERK activation (46). In particular, IRS-1 downregulation relieves SHP-2 tyrosine phosphatase that is sequestered to IRS-1, thereby allowing the formation of SHPS-1/SHP-2/Src/Shc/Grb2 signaling complex to promote IGF-I-induced ERK-mediated VSMC proliferation (46). In a similar manner, PDGF-induced IRS-1/IRS-2 dysregulation may invoke the participation of SHP-2 to enhance ERK activation and VSMC proliferation. In this regard, it is pertinent to note that in PDGF receptor overexpressing aortic endothelial cells (49), PDGF receptor tyrosine phosphorylation recruits SHP-2 to promote Ras/Raf/MEK1-mediated ERK phosphorylation (12, 49). Further studies are clearly warranted to determine the relationship between IRS-1/IRS-2 downregulation/overexpression and SHP-2-mediated ERK phosphorylation in the context of PDGF-induced VSMC proliferation.
The primary sequence of IRS-1 and IRS-2 contains over 30 potential serine/threonine phosphorylation sites (39). Several lines of evidence suggest that IRS-1 and IRS-2 undergo serine phosphorylation with or without proteasomal degradation as a function of agonists, cellular phenotype, and species (22, 34, 39, 43, 48, 52, 57, 59). A number of studies demonstrate that in mouse embryonal 3T3-L1 adipocytes, insulin induces IRS-1 and IRS-2 serine phosphorylation but only IRS-1 undergoes degradation (43, 59). However, in 3T3-L1 preadipocytes and mouse embryo fibroblasts, insulin/IGF-I induces IRS-2 degradation but not IRS-1 (52). In human neuroblastoma cells, IGF-I induces preferential degradation of IRS-2 but not IRS-1 (30). Thus, in several nonvascular cells, insulin/IGF-I has been shown to promote the degradation of either IRS-1 or IRS-2. Interestingly, PDGF-induced IRS-1 serine phosphorylation does not result in IRS-1 degradation as evidenced in earlier studies using 3T3-L1 adipocytes (34, 48, 57) and in the present study using human aortic VSMCs. As suggested by Pederson et al. (43), PDGF-induced mTOR/p70S6kinase signaling may not activate a downstream kinase required for site-specific phosphorylation of IRS-1 to promote its degradation. Furthermore, intrinsic differences in the compartment-specific activation of PI 3-kinase/Akt signaling pathway and subcellular localization pattern of IRS-1 and IRS-2 (30, 40) may be attributable to the present observations of IRS-2 degradation but not IRS-1 in response to PDGF in human aortic VSMCs.
In VSMCs, activation of select protein kinases and generation of reactive oxygen species promote IRS-1/IRS-2 dysregulation to affect insulin-induced IRS-1-/IRS-2-associated PI 3-kinase activity and Akt phosphorylation (17, 26, 28, 61). In the present study, PDGF (a tyrosine kinase-linked receptor agonist) increases Akt and p70S6kinase phosphorylation to dysregulate IRS-1 and IRS-2 and inhibit insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity in VSMCs. In addition, endothelin-1 and angiotensin II (G protein-coupled receptor agonists) and aldosterone (a steroid hormone) utilize one or more of the intermediary signaling components to dysregulate either IRS-1 or IRS-2 signaling in VSMCs (17, 26, 28, 61). For instance, endothelin-1 induces activation of protein kinase C to promote IRS-2 serine phosphorylation and diminish insulin-induced IRS-2-associated PI 3-kinase activity in VSMCs (28). Furthermore, angiotensin II and aldosterone induce the activation of protein kinases such as Src and phosphoinositide-dependent kinase-1 and increase the levels of reactive oxygen species to promote IRS-1 serine phosphorylation or degradation, thereby diminishing insulin-induced increases in Akt phosphorylation and glucose uptake in VSMCs (17, 26, 61). Together, these findings strongly suggest that tyrosine kinase-linked receptor agonists and G protein-coupled receptor agonists have the potential to dysregulate IRS-1- and/or IRS-2-dependent PI 3-kinase pathway in VSMCs.
In addition to the known mitogenic effects of PDGF (6, 41, 50), previous studies demonstrate that the vasoconstrictor peptide (e.g., endothelin-1) and the key components of the renin-angiotensin-aldosterone system (e.g., angiotensin II and aldosterone) exhibit pleiotropic effects including the ability to enhance VSMC proliferation (8, 16, 53). In this regard, the expression levels of ligands and receptors for all these four mitogens have been shown to be upregulated in the atherosclerotic lesion (8, 16, 50, 53). Notably, PDGF (present study), endothelin-1, angiotensin II, and aldosterone (previous studies) also have the potential to dysregulate IRS-1- and/or IRS-2-dependent insulin receptor signaling in VSMCs (17, 26, 28, 61). Together, these observations suggest that vascular pathologies associated with increases in the levels of these mitogens are prone to VSMC-specific defects in IRS-1 and/or IRS-2 signaling.
At this juncture, it is pertinent to note that hyperinsulinemic/insulin-resistant conditions in obese Zucker rats diminish IRS-1 and IRS-2 expression levels and attenuate insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity in the aorta while enhancing basal ERK phosphorylation (27). In addition, high-glucose conditions downregulate IRS-1 expression in VSMCs to promote IGF-I-induced ERK-mediated VSMC proliferation via SHPS-1/SHP-2/Src/Shc/Grb2 signaling complex formation (46). With regard to neointimal growth after vessel injury, increases in VSMC proliferation are much greater in IRS-2-deficient mice compared with IRS-1-deficient mice (31). Notably, the frequency of mutation of IRS-1 gene is significantly higher in patients with coronary artery disease (3, 13). Furthermore, metastatic mammary tumors are specifically associated with inactivation of IRS-1 but not IRS-2 (20). Together, these findings support the notion that diminutions in smooth muscle-specific IRS-1/IRS-2 expression or IRS-1/IRS-2-associated PI 3-kinase activity, if observed in diseased vessels such as those from subjects undergoing coronary artery bypass grafting surgery, may serve as a surrogate marker for enhanced neointimal growth in the grafted vessels.
Previous studies have shown that mTOR/p70S6kinase activation in synthetic VSMCs promotes IRS-1 serine phosphorylation/degradation to diminish the expression of contractile proteins (36, 37). Although PDGF dysregulates IRS-1 and IRS-2 via mTOR/p70S6kinase pathway (present study) and downregulates the expression of contractile proteins in VSMCs (41), a direct correlation between IRS-1/IRS-2 dysregulation and a global downregulation of contractile proteins in VSMCs should be viewed with caution. This is due to the fact that endothelin-1, which dysregulates IRS-2/PI 3-kinase signaling in VSMCs (28), enhances calponin phosphorylation to induce arterial contractility (65). In addition, angiotensin II that promotes IRS-1 serine phosphorylation/degradation (17, 61) induces the expression of calponin protein (without affecting smooth muscle α-actin) by increasing calponin promoter activity (14). Further studies using target-specific small interfering RNAs for IRS-1 and IRS-2 are clearly warranted to examine the relationship between IRS-1/IRS-2 downregulation and contractile phenotype in VSMCs in the absence or presence of tyrosine kinase-linked receptor agonists (e.g., PDGF and insulin) vs. G protein-coupled receptor agonists (endothelin-1 and angiotensin II).
Insulin is known to exert both vasculoprotective (antiatherogenic) and growth-promoting (atherogenic) effects. With regard to its vasculoprotective effects, it not only promotes VSMC differentiation but also diminishes atherosclerosis and neointimal growth (9, 54, 64). For instance, orally administered insulin attenuates atherosclerosis in apoE-deficient mice (54), whereas hyperinsulinemic-euglycemic conditions diminish neointimal growth after arterial injury and also enhance reendothelialization (9). In endothelial cells, insulin-induced activation of IRS-associated PI 3-kinase signaling enhances the production of nitric oxide, which in turn inhibits VSMC growth (38). In conformity with these observations, recent studies using apoE-deficient mice demonstrate that a conditional knockout of insulin receptor gene in vascular endothelial cells accelerates atherosclerosis without altering systemic insulin sensitivity or lipoprotein profile (47). With regard to its growth-promoting effects, insulin has been shown to induce proliferation of human arterial VSMCs in culture (44). In addition, chronic insulin treatment enhances PDGF receptor tyrosine phosphorylation and neointima formation after balloon injury in the rat carotid artery model (45).
In the present study using human aortic VSMCs, unlike PDGF, insulin does not produce a significant increase in cell number in parallel experiments, thus supporting a lack of insulin effects on VSMC proliferation. Our data are in conformity with earlier findings that insulin does not stimulate the proliferation of human coronary artery VSMCs (56). In these studies, insulin induces Akt phosphorylation but fails to enhance the expression of Ki-67, a nuclear marker of actively cycling cells (56). In a different study using human coronary artery VSMCs, IGF-I, a weak mitogen, induces Akt phosphorylation but fails to downregulate the mRNA expression of p27kip1, a cell cycle inhibitor (1). Contrarily, PDGF, a potent mitogen, induces Akt phosphorylation and VSMC proliferation, which is mediated by enhanced Ki-67 expression and downregulation of p27kip1 expression (1, 32). Thus, PDGF- but not insulin/IGF-I-induced Akt phosphorylation promotes VSMC proliferation, which is characterized by an increase in cell cycle progression into S phase (1, 56). In addition, it is pertinent to note that PDGF- and IGF-I-induced Akt phosphorylation in human coronary artery VSMCs results in the downstream phosphorylation and nuclear exclusion of FoxO forkhead transcription factors (1). Nevertheless, only PDGF-induced nuclear exclusion of FoxO promotes the downregulation of p27kip1 mRNA expression, a critical step in cell cycle progression (1). Thus, the role of PDGF to enhance VSMC proliferation is attributed, in part, to the likely increases in the intensity and duration of FoxO inactivation by PDGF in VSMCs.
Importantly, imatinib inhibition of PDGF receptor tyrosine kinase diminishes PDGF receptor tyrosine phosphorylation and its downstream signaling to attenuate PDGF-induced VSMC proliferation, while at the same time relieves PDGF-induced dysregulation of insulin receptor signaling. Specifically, imatinib inhibition of PDGF receptor tyrosine kinase blocks PDGF-induced IRS-1 serine phosphorylation and IRS-2 downregulation. Furthermore, imatinib treatment abrogates PDGF-mediated diminutions in insulin-induced IRS-1- and IRS-2-associated PI 3-kinase activity. Thus, restoration of functional IRS/PI 3-kinase signaling in conjunction with inhibition of VSMC proliferation may promote the differentiation/contractile phenotype in VSMCs, as described in previous studies (23, 24, 36, 46, 64). Together, these findings increase the likelihood for the vasculoprotective effects of intact IRS/PI 3-kinase axis in VSMCs in the absence of detrimental factors such as enhanced PDGF secretion/action.
In vitro and in vivo strategies to inhibit PDGF receptor tyrosine kinase by imatinib have been shown to diminish the proliferation of VSMCs in culture (11) and attenuate atherosclerosis in diabetic apoE-deficient mice, respectively (33). While recent studies demonstrate a marked reduction of systemic insulin resistance by imatinib in diabetic mice (21), the present findings point toward the potential use of imatinib to improve insulin receptor signaling in VSMCs. At this juncture, it is pertinent to note that thiazolidinediones, angiotensin II type 1 receptor antagonists, and rapamycin (sirolimus) inhibit neointimal hyperplasia and restore dysregulated IRS/PI 3-kinase signaling in VSMCs (19, 36, 60). Targeting PDGF receptor tyrosine kinase may provide yet another pharmacological tool to diminish neointimal growth and improve VSMC-specific IRS/PI 3-kinase signaling in nondiabetic and diabetic subjects.
This work was supported by National Institutes of Health Grants DK-071893 and HL-097090.
No conflicts of interest, financial or otherwise, are declared by the author(s).
The authors thank Dr. Lawrence I. Sinoway for critical reading of this manuscript.
Present address of P. H. McNulty: Bassett Heart Care Institute, Bassett Healthcare, Cooperstown, NY 13326-1301.
- Copyright © 2011 the American Physiological Society