Regulation of Vascular Smooth Muscle Cell Apoptosis
Modulation of Bad by a Phosphatidylinositol 3-Kinase–Dependent Pathway
Abstract—Our objective was to define the signaling mechanisms by which mitogens such as insulin-like growth factor-I (IGF-I) regulate vascular smooth muscle cell (VSMC) apoptosis. We confirmed that IGF-I inhibits serum withdrawal–induced apoptosis of cultured VSMCs in a dose-dependent and time-dependent fashion. To test the hypothesis that the phosphatidylinositol (PI) 3-kinase signaling pathway regulates VSMC survival, we examined the relationship between PI 3-kinase activity and cell fate. PI 3-kinase was elevated in viable VSMCs maintained in serum-containing medium, declined significantly in response to serum withdrawal, and increased in response to IGF-I–induced survival. Moreover, blockade of PI 3-kinase with 2 structurally dissimilar inhibitors (wortmannin or LY294002) abolished the capacity of IGF-I to maintain VSMC viability. Similarly, transient transfection of a dominant-negative Δp85 PI 3-kinase mutant construct abrogated the capacity of IGF-I to prevent VSMC death. Thus, PI 3-kinase is a critical antiapoptotic signal in VSMCs. To define the distal element of the antiapoptotic cascade, we tested the hypothesis that IGF-I inhibits the influence of the proapoptotic gene Bad. Indeed, IGF-I stimulates increased expression of the inactive, phosphorylated form of Bad by a PI 3-kinase–dependent pathway. Moreover, the proapoptotic effect of Bad was attenuated by the stimulation of IGF-I. Thus, growth factors appear to prevent VSMC death by activating signal transduction pathways linked to apoptotic regulatory genes.
- vascular smooth muscle cell
- insulin-like growth factor-I
- phosphatidylinositol 3-kinase
- programmed cell death
The pathogenesis of various forms of vascular disease involves an abnormal accumulation of cells within the intimal space. Although many studies have focused on alterations in the regulation of cell growth as a fundamental feature in the pathogenesis of vascular disease,1 it is becoming clear that perturbations in the regulation of cell death may be equally important.2 3 4 5 6 7
An emerging theme suggests that many of the regulatory mechanisms originally characterized as governing cell growth also modulate cell death.2 3 It is postulated that the increased expression of growth factors observed within vascular lesions may promote an increase in the vascular smooth muscle cell (VSMC) population by preventing apoptosis as well as enhancing cell cycle progression. Indeed, recent studies by Bennett et al8 have documented that growth factors such as insulin-like growth factor-I (IGF-I) and platelet-derived growth factor-BB are potent inhibitors of VSMC death. However, the molecular basis by which growth factors prevent VSMC death remains to be further defined.
A recent report has demonstrated that phosphatidylinositol (PI) 3-kinase is involved in mediating the antiapoptotic effects of nerve growth factor in rat pheochromocytoma cell lines.9 However, given the cell-type specificity of apoptosis regulation, it remains unclear whether PI 3-kinase plays a role in regulating VSMC fate. Moreover, the downstream elements of this antiapoptotic signaling pathway remain to be further defined. It is postulated that the signal transduction apparatus of membrane receptors must be coupled to the regulation of apoptosis regulatory genes in the Bcl-2 family. Cell fate is determined by the relative balance of proapoptotic members of the Bcl-2 family, such as Bad, versus antiapoptotic mediators, such as Bcl-XL.10 One current working model suggests that cell suicide is induced when the proapoptotic mediator Bad forms heterodimers with Bcl-XL and thereby inactivates the antiapoptotic functions of Bcl-XL. The signal transduction pathway that regulates the dimerization of apoptosis regulatory genes remains to be further elucidated. The present study tests the hypothesis that the regulation of VSMC survival is mediated by a PI 3-kinase–dependent signal transduction pathway that modulates the proapoptotic effects of Bad.
Materials and Methods
Human umbilical artery smooth muscle cells were purchased from Clonetics Corp and were grown according to the manufacturer’s instructions. Passages 5 to 7 were used for experimentation. The clonal embryonic rat thoracic VSMC line A7r5 was obtained from American Type Culture Collection. In addition, we have established several polyclonal lines of neomycin-selected A7r5 transfected with a retroviral construct to constitutively overexpress Bcl-2.11 These Bcl-2 stable transfectants were used in the experiments involving transient transfection of the Bad expression construct. VSMCs were maintained in 75-cm2 polystyrene flasks in DMEM/F-12 HAM (GIBCO-BRL) supplemented with 10% heat-inactivated FBS (GIBCO-BRL), penicillin (100 U/mL), streptomycin (100 mg/mL), and 25 mmol/L HEPES buffer.
Human recombinant IGF-I (R&D Systems) was supplemented to the serum-free medium at various concentrations (0, 0.1, 1, 10, and 100 ng/mL). In the experiments to determine the time window of the stimulus, IGF-I or a vehicle control was added at different time points (0, 4, and 8 hours) after serum withdrawal. Two structurally dissimilar inhibitors of PI 3-kinase were used to selectively inhibit the PI 3-kinase activity, wortmannin (Sigma Chemical Co) and LY294002 (Biomol). Previous studies have established the specificity of PI 3-kinase blockade with these agents in the dose range used in this study.9 12 13 In pilot studies, we have also confirmed blockade of VSMC PI 3-kinase in our model system in this dose range. On the basis of these studies, we observed that the optimal inhibitory concentrations were 100 nmol/L wortmannin and 50 μmol/L LY294002 administered 30 minutes before the addition of IGF-I in VSMCs. Each drug was dissolved in DMSO, and the final concentrations of DMSO did not exceed 0.1%. This concentration of DMSO had no direct effect on cell viability. In some experiments with wortmannin, it was added every 4 hours to the medium to achieve sustained blockade given the short half-life of its biological efficacy.12 In each experiment, cells were treated with a DMSO vehicle control. As an additional control in the experiments with LY294002, comparisons were also made with cells treated with the structurally related but inactive analogue LY298619 (kindly provided by Lilly Research Laboratory).13
Determination of Apoptosis
Nuclear chromatin morphology analysis and caspase-3 activity assay were used as quantitative indices of apoptosis. Nuclear chromatin morphology analysis has been extensively validated and cross-correlated with other complementary assays of apoptotic cell death in our laboratory, including time-lapse video microscopy, DNA electrophoresis, terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) staining, caspase-3 activity assays, and fluorescence-activated cell sorting analysis.14 Cells to be analyzed for apoptosis by nuclear chromatin morphology were stained with H33342 and PI (Molecular Probes, Inc) and viewed under fluorescence microscopy at 12 hours after serum withdrawal. Using both membrane-permeable (H33342) and -impermeable (PI) dyes in the assay allows for the determination of cell viability and plasma membrane integrity. The proportion of cells rescued by IGF-I treatment (% rescue) was calculated as the percentage of apoptotic cells under serum-free conditions minus the percentage of the apoptotic cells after IGF-I treatment divided by the percentage of apoptotic cells under serum-free conditions.
Caspase-3 activity assay was performed according to the manufacturer’s protocol (Clontech Laboratory, Inc). The caspase-3 activity was determined by assay of each sample normalized to the basal activity in the viable cells maintained in serum-containing medium.
PI 3-Kinase Assay
The measurement of PI 3-kinase activity was performed as previously described.15 VSMCs were incubated with cell lysis buffer (in mmol/L, HEPES [pH 7.5] 25, NaCl 50, NaF 50, EDTA 5, okadaic acid 10, sodium orthovanadate 1, and phenylmethylsulfonyl fluoride 1; 1% NP-40; and 1 μg/mL each of antipain, aprotinin, and leupeptin) for 10 to 15 minutes on ice. Insoluble material was removed by centrifugation at 12 000g for 20 minutes. The amount of cell lysate was normalized by protein content in each experiment (Bio-Rad Laboratories). The cell lysate (0.5 mg of protein) was incubated with an antibody against the p85α subunit of PI 3-kinase (2 μg/mg of protein) (Upstate Biotechnology) with constant agitation for 3 hours and then further incubated with 25 μL of protein A/G plus agarose (Santa Cruz Biotechnology) with agitation for 1 hour. The beads containing the immunoprecipitates were washed 4 times with lysis buffer and washed once with buffer (0.1 mol/L NaCl, 1 mmol/L EDTA, and 20 mmol/L Tris-HCl, pH 7.5). The washed pellets were resuspended in 50 μL of kinase reaction buffer (in mmol/L, Tris-HCl [pH 7.5] 20, NaCl 100, and EGTA 0.5) and incubated at 25°C for 10 minutes with 0.5 mL of 20 mg/mL PI dissolved in chloroform to make micelles of PI. Assays were initiated with the addition of 5 μL of ATP solution (0.4 μmol/L ATP, 0.1 mol/L MgCl2, and 1 μCi/mL [γ-32P]ATP) and incubated at room temperature for 30 minutes. The reaction was stopped after the addition of 100 μL of chloroform, methanol, and 11.6N HCl (100:200:2). After centrifugation, the lower organic phase was taken for thin-layer chromatography on silica gel plates (Aldrich Chemical Co) and developed in chloroform, methanol, 25% ammonium hydroxide, and water (43:38:5:7). The plate was exposed to Kodak X-Omat film at –70°C with an intensifying screen and quantified by computer-enhanced video densitometry using the NIH image analysis program.
Plasmid Constructs and Transfection
To determine whether PI 3-kinase is an essential signaling molecule mediating the inhibition of apoptosis, a dominant-negative mutant bovine 85-kDa subunit of PI 3-kinase (Δp85) that lacks the binding site for the 110-kDa subunit of PI 3-kinase was overexpressed in VSMCs by transient transfection. An expression vector containing a promoter, which is composed of the simian virus 40 (SV40) early promoter and the R segment and part of the U5 sequence of the long terminal repeat of human T-cell leukemia virus type I (SRα), and the Δp85 PI 3-kinase mutant construct was kindly provided by Dr W. Ogawa (Kobe University, Japan). Several previous studies have documented that overexpression of the dominant-negative Δp85 PI 3-kinase mutant selectively interferes with endogenous PI 3-kinase function.16 17 To investigate the proapoptotic effect of Bad, 6-amino acid epitope of the bovine papilloma virus capsid protein (AU1)-tagged wild-type Bad construct subcloned into a pCDNA3 vector (a gift from Dr G. Nunez, University of Michigan, Ann Arbor)18 was transiently transfected into VSMCs stably expressing Bcl-2. The A7r5 cells stably transfected with a retroviral construct containing Bcl-2 exhibit normal viability for >24 hours despite serum withdrawal and therefore provide a means to assess the effects of Bad on cell fate under serum-free conditions in the presence or absence of IGF-I stimulation.
Transient transfection was performed with 2 μg of total DNA per 35-mm dish and LipofectAMINE (GIBCO-BRL) according to the manufacturer’s instructions (DNA:LipofectAMINE ratio, 1:3). Pilot studies have documented a transfection efficiency of ≈15% to 25% in A7r5 VSMCs using this approach. Serum withdrawal was performed to induce apoptosis in the presence or absence of IGF-I 48 hours after transfection. Cells transfected with an expression vector containing the same promoter but without the subcloned transgene cassette (empty expression vector) served as a control. The effect of the transgene on cell viability was determined by assaying the activity of a cotransfected reporter gene19 20 21 or nuclear morphology with H33342 in the cells coexpressing the reporter gene, as previously described.22 23 Briefly, VSMCs were cotransfected with an expression vector containing the reporter gene luciferase driven by an SV40 promoter (Promega), along with transfection of either the Δp85 PI 3-kinase mutant transgene or the wild-type Bad expression construct, versus the control vector at a 3:1 transgene:reporter ratio. Cells that were successfully transfected and remained viable were assayed by measurement of luciferase activity using standard methods with a luminometer. In pilot studies conducted in VSMCs stably transfected with Bcl-2, we confirmed that serum withdrawal failed to have a direct inhibitory effect on reporter gene activity in viable cells using this constitutively active viral promoter construct. This assay of luciferase activity provides an objective, sensitive, and quantitative index of cell viability among the subpopulation of transfected cells under these conditions. The proportion of viable transfected cells after serum withdrawal was determined by assay of luciferase activity in this group in relation to the basal level of luciferase activity expressed in transfected cells maintained in serum-containing medium. We also used the expression vector containing the reporter gene green fluorescent protein (GFP) driven by the cytomegalovirus promoter (GIBCO-BRL) to confirm that the loss of reporter gene expression in response to serum withdrawal in transfected VSMCs is due to apoptotic cell death, as assessed by nuclear chromatin morphology with H33342.22 23 The characterization of nuclear morphology in GFP-positive cells provides an assessment of the effect of the transgene under study on cell fate using a single-cell analysis approach.
Cells were harvested 6 hours after serum withdrawal in the presence or absence of IGF-I (100 ng/mL) and lysed in a modified radioimmunoprecipitation assay buffer (in mmol/L, NaCl 150, Tris [pH 7.4] 10, EDTA 1, and PMSF 1, as well as 1% Triton X-100 and 1% deoxycholic acid, with the addition of a complete protease inhibitor mixture [Boehringer Mannheim]). To assess the role of PI 3-kinase, cells were pretreated with wortmannin (100 nmol/L), LY294002 (50 μmol/L), or a DMSO vehicle at the time of serum withdrawal. Protein concentrations were determined and normalized as above. Cell lysates (40 μg protein/sample) were run in a 15% SDS-PAGE under DTT-reducing conditions. Proteins were then transferred to a nitrocellulose filter by semidry blotting in (in mmol/L) Tris 25, glycine 200, and SDS 1.3, and 20% methanol (pH 8.3) at 160 mA for 3 hours. The filters were stained with 10% Ponceau S solution (Sigma) for 10 minutes to verify equal loading and transfer efficiency. The filters were then blocked in 10% dry milk for 1 hour at room temperature and probed with polyclonal antibodies against Bad (catalog No. sc-941, Santa Cruz Biotechnology) (1 μg/mL) overnight at 4°C. After washing with TBS-T (20 mmol/L Tris, 137 mmol/L NaCl, and 0.1% Tween 20), signals for Bad expression were detected by horseradish peroxidase–conjugated secondary antibody (Transduction Laboratories) for 1 hour at room temperature. The specificity of the signal was confirmed by the absence of signal when membranes were incubated with nonimmune IgG. After extensive washing in TBS-T, electrochemical luminescence was developed (Amersham), and membranes were exposed to Hyperfilm-ECL (Amersham) and quantified by computer-enhanced video densitometry using the NIH image analysis program.
Previous studies have established that Bad is phosphorylated on serine residues and that this phosphorylated form of Bad migrates as a retarded band on gel electrophoresis relative to the nonphosphorylated form.20 21 22 We observed both the phosphorylated and nonphosphorylated forms of Bad endogenously expressed in VSMCs maintained in serum as determined by a gel mobility shift assay. To confirm that the retarded band on gel electrophoresis reflected serine phosphorylation, we assessed the effect of in vitro treatment of cellular extracts with a protein phosphatase (Lambda Protein Phosphatase, New England Biolabs Inc) at 37°C for 30 minutes, in accord with previous reports in nonvascular cells.20 21 22 Furthermore, as an additional verification, we also analyzed VSMCs overexpressing Bad after transient transfection of AU1-tagged Bad constructs. Modulation of Bad phosphorylation in VSMCs was determined by sequential immunoprecipitation with an anti-AU1 antibody (BAbCO Co) and immunoblotting with a phosphospecific (serine 136) anti-Bad antibody (New England Biolabs, Inc). The anti-Bad monoclonal antibody (Transduction Laboratories) was used to determine the total Bad protein expression.
Each experimental condition was tested in triplicate, and each experiment was repeated a minimum of 3 times. Statistical analyses were performed by ANOVA or unpaired 2-tailed Student test. Data are presented as mean±SD.
Antiapoptotic Effect of IGF-I on VSMCs
First, we confirmed a previous report that IGF-I is a potent antiapoptotic factor for human VSMCs.8 In addition to the morphological features of apoptotic cell death, we observed that human VSMCs maintained in serum-free medium showed a marked increase of caspase-3 activity within 24 hours (5-fold increase of caspase-3 activity in serum-free versus serum control; P<0.001). However, the addition of IGF-I (100 ng/mL) in the serum-free medium significantly inhibited caspase-3 activation (1.8±0.6, with IGF-I, versus 5.1±0.3, serum free; P<0.001). To further elucidate the molecular mediators regulating VSMC fate, we initiated our studies by establishing an in vitro model system using a phenotypically stable, clonal cell line that was amenable to the use of recombinant DNA approaches. We observed that the rat A7r5 clonal VSMCs grown to near confluence in 10% FBS undergo apoptosis when switched to serum-free medium. Using a quantitative nuclear chromatin morphology assay, we determined that 20% of cells underwent cell suicide within 12 hours under these conditions (Figure 1⇓). However, the addition of IGF-I at the time of serum withdrawal inhibited apoptosis in a dose-dependent manner, with significant effects noted at nanomolar concentrations (Figure 1⇓).
We postulated that a critical time window exists in which the activation of a countervailing antiapoptotic signaling cascade is sufficient to prevent cellular suicide. We therefore examined whether the survival-promoting effects of IGF-I are a time-dependent phenomenon. Accordingly, cell viability was assessed in response to adding IGF-I or a vehicle control at different time points after serum withdrawal as shown in Figure 2⇓. The addition of IGF-I (100 ng/mL) at 4 hours after serum withdrawal had survival-promoting effects similar to those of IGF-I administration at the time of serum withdrawal (time 0). However, the addition of IGF-I at 8 hours after serum withdrawal failed to rescue a significant portion of VSMCs committed to cell suicide compared with IGF-I administration at 4 hours after serum withdrawal. These findings suggested that VSMC survival may depend on the activity level of an antiapoptotic signaling pathway during a critical time window between 4 and 8 hours after serum withdrawal.
IGF-I Induces PI 3-Kinase Activation in VSMCs
To further define the molecular mediators of VSMC survival, we tested the hypothesis that the activity of the PI 3-kinase pathway determines VSMC viability. According to our analysis of the time course kinetics of VSMC death, we postulated that an antiapoptotic signal active in the presence of serum may decay in activity over the course of 4 to 8 hours after serum withdrawal. Indeed, as shown in Figure 3⇓, analysis of PI 3-kinase activity by thin-layer chromatography documented high levels in serum-containing medium and a substantial decline that is evident within 4 hours after serum withdrawal. Moreover, the addition of IGF-I stimulated a 5.6-fold increase in PI 3-kinase activity within 5 minutes of administration compared with control (n= 5, P<0.01). The IGF-I–induced activation of PI 3-kinase activity was maintained for at least 40 minutes. Similar results were obtained in both human and rat VSMCs. These results indicate a striking association between VSMC survival and the activity of the PI 3-kinase signaling pathway.
Antiapoptotic Signaling Role of PI 3-Kinase: Pharmacological Blockade
On the basis of this apparent association between VSMC survival and PI 3-kinase activity, we directly tested the hypothesis that the survival-promoting effects by IGF-I are mediated via the activation of a PI 3-kinase–dependent signal transduction pathway. We used 2 structurally dissimilar pharmacological probes (wortmannin and LY294002)12 13 to selectively inhibit PI 3-kinase activity and define the role of this cellular mediator as a determinant of IGF-I–stimulated VSMC survival.
As compared with the induction of apoptosis in control rat VSMCs exposed to serum withdrawal, IGF-I treatment rescued 78±14% stimulated to undergo cell suicide. However, IGF-I treatment failed to rescue a substantial portion of VSMCs from apoptotic death in the context of PI 3-kinase pathway blockade with wortmannin (78±14% rescue in DMSO vehicle-treated cells versus 28±7% rescue with wortmannin treatment; P<0.001). To further confirm that these findings reflect a specific response to blockade of PI 3-kinase, we also used the structurally dissimilar PI 3-kinase inhibitor LY294002. In accord with the results obtained with wortmannin, blockade of PI 3-kinase activation with LY294002 significantly inhibited the capacity of IGF-I to promote VSMC survival (67±16% rescue in vehicle-treated cells versus 37±16% rescue with LY294002 treatment, P<0.01). As an additional experimental control to confirm the specificity of the response, we observed that an inactive analogue of LY294002, LY298619, did not have any effect on IGF-I–induced VSMC survival.
In addition, we confirmed that the antiapoptotic effect of IGF-I and the reversal by pharmacological blockade of PI 3-kinase were also observed in human VSMCs. As shown in Figure 4⇓, IGF-I significantly inhibited the activation of caspase-3 activity in response to serum withdrawal (5.1±0.3-fold increase of caspase-3 activity in vehicle-treated serum-free conditions versus 1.8±0.6-fold with IGF-I; P<0.01). However, the blockade of PI 3-kinase completely abolished the antiapoptotic effect of IGF-I on caspase-3 activity (1.8±0.6-fold with IGF-I versus 5.7±0.8 with IGF-I+wortmannin; P<0.01). Although there was a slight but significant potentiating effect of wortmannin on the induction of caspase activity in response to serum withdrawal (5.1±0.3-fold in vehicle-treated serum-free conditions versus 6.4±0.7 with wortmannin; P<0.05), the level of caspase-3 activation was not significantly different in wortmannin-treated cells exposed to IGF-I compared with wortmannin-treated vehicle controls (5.7±0.8-fold with IGF-I+wortmannin versus 6.4±0.7-fold in vehicle-treated serum-free conditions with wortmannin; P>0.05). These data indicate that IGF-I–induced survival-promoting effects are mediated by PI 3-kinase in both human and rat VSMCs.
Dominant-Negative PI 3-Kinase Mutant Expression Inhibits IGF-I–Induced Antiapoptotic Effects
To further confirm the mediator role of PI 3-kinase as a determinant of VSMC survival induced by IGF-I, we also used a gene transfer strategy to selectively block the activity of the PI 3-kinase pathway. Previous studies have established that overexpression of the dominant-negative Δp85 mutant construct of PI 3-kinase results in the suppression of endogenous PI 3-kinase function.15 16 To assess the effect of this transgene on VSMC fate, we used a cotransfection protocol with an expression vector containing the reporter gene luciferase to focus on the assessment of cell viability among the subpopulation of successfully transfected cells. The comparison of reporter gene expression between VSMCs cotransfected with the PI 3-kinase Δp85 mutant construct versus a control expression vector provides a means of assessing the role of this signaling pathway on cell viability in response to serum withdrawal.
Rat VSMCs cotransfected with the reporter gene luciferase and a control vector exhibited preserved cell viability, as indicated by high luciferase activity in serum-containing medium. However, the withdrawal of serum induced cell death, as reflected by a significant decrease in luciferase activity of 58% (Figure 5⇓). The induction of cell death in these control vector-transfected cells was attenuated by the administration of IGF-I as indicated by the significantly higher level of luciferase activity in IGF-I–treated cells compared with those in the serum-free condition (58±11% in serum-free conditions versus 79±8% with IGF-I; n=9, P<0.01). In contrast to the antiapoptotic effect of IGF-I stimulation observed in cells transfected with the control vector, cotransfection with the dominant-negative Δp85 mutant PI 3-kinase abolished the capacity of IGF-I to preserve VSMC viability under serum-free conditions.
Taken together, these findings using a gene transfer experimental strategy complement and confirm the pharmacological studies and provide corroborating evidence that IGF-I promotes VSMC survival via a PI 3-kinase–dependent signaling pathway. However, the link between this signal transduction pathway and the downstream elements of the cell death program governing VSMC fate remained to be defined.
IGF-I Phosphorylates Bad via PI 3-Kinase–Dependent Pathway
It is postulated that extracellular factors may promote cell survival by stimulating the phosphorylation of Bad, reduce Bad–Bcl-2 heterodimer formation by inducing this post-translational modulation, and thereby potentiate the effects of antiapoptotic mediators such as Bcl-2.
Given that the expression pattern of members of the Bcl-2 family is cell specific, we initially explored whether VSMCs express Bad. As shown in Figure 6A⇓, immunoblots of Bad protein expression confirmed that VSMCs maintained in serum-containing growth medium exhibited a doublet of ≈24 to 25 kDa, which is consistent with the retarded mobility of the phosphorylated form as well as a faster-migrating nonphosphorylated form of Bad, as previously described in nonvascular cells.21 We confirmed that the upper band reflected retarded mobility because of protein phosphorylation in the samples stably expressing Bad phosphorylation by in vitro treatment of cellular extracts with a protein phosphatase. As shown in Figure 6B⇓, in vitro treatment with the phosphatase eliminated the retarded band signal.
In contrast to the cells in serum, VSMCs maintained in serum-free medium for 6 hours exhibited substantially less expression of the inactive, phosphorylated form of Bad (Figure 6A⇑). Moreover, IGF-I treatment of VSMCs in serum-free medium increased the inactive, phosphorylated form of Bad to the levels comparable with the serum-stimulated baseline. On the basis of the evidence that PI 3-kinase mediates the IGF-I–induced antiapoptotic signals in VSMCs, we tested the hypothesis that this signaling pathway is coupled to the phosphorylation of Bad. Indeed, blockade of PI 3-kinase activity with either wortmannin or LY294002 attenuated the increase in expression of the phosphorylated form of Bad in response to IGF-I stimulation (Figure 6A⇑). The densitometric analysis revealed that IGF-I stimulation increased the expression of the phosphorylated form of Bad by 160% (n=4, P<0.05) but had no significant effect on the overall protein expression of Bad.
We further explored the capacity of extracellular stimuli to regulate VSMC fate by modulating the proapoptotic activity of Bad using a gene transfer experimental approach. As a complementary strategy to assess the effect of IGF-I on Bad phosphorylation, we transiently transfected rat A7r5 VSMCs stably overexpressing Bcl-2 with an epitope-tagged Bad construct that enabled us to perform sequential immunoprecipitation using the anti-AU1 antibody followed by immunoblot with the phosphospecific anti-Bad antibody that selectively recognizes the phosphorylated epitope at serine 136. In addition, total Bad expression was assessed by immunoblot using the same anti-Bad antibody used in Figure 6A⇑ and 6B⇑. As shown in Figure 6C⇑, IGF-I treatment of transfected cells under serum-free conditions induced significant phosphorylation of Bad as demonstrated by the increased retarded mobility of the doublet. Furthermore, blockade of the PI 3-kinase signaling pathway attenuated the IGF-I–induced phosphorylation of Bad. This observation was confirmed using the phosphospecific anti-Bad antibody. As shown in Figure 6D⇑, IGF-I treatment induced serine phosphorylation of Bad, and blockade of the PI 3-kinase signaling pathway attenuated the IGF-I–induced serine phosphorylation of Bad. We confirmed that neither stimulation with IGF-I nor blockade of PI 3-kinase activity modulated the total expression levels of Bad protein. Overall, these data provide further evidence that serine phosphorylation of Bad is regulated in a PI 3-kinase–dependent manner in VSMCs.
IGF-I Modulates the Proapoptotic Effect of Bad
The working hypothesis poses that VSMC survival depends on the interplay between the proapoptotic mediator Bad and antiapoptotic factors such as Bcl-2. As noted above, treatment of VSMCs with IGF-I is associated with the serine phosphorylation of Bad by a PI 3-kinase–dependent pathway. However, it remained to be determined whether the IGF-I–induced phosphorylation of Bad influences the proapoptotic activity of Bad on the fate of VSMCs. To examine whether the proapoptotic activity of Bad is modulated by extracellular stimuli such as IGF-I, we assessed the effect of transient transfection of Bad in rat A7r5 cells stably overexpressing Bcl-2. This experimental strategy takes advantage of the fact that these Bcl-2 stable transfectants do not undergo apoptosis under serum-free conditions for 24 hours,11 and therefore the modulatory influence of IGF-I on Bad-induced cell death can be determined. These experiments used a cotransfection protocol with the GFP reporter gene to quantify apoptosis among the subpopulation of successfully transfected cells. Figure 7A⇓ illustrates GFP staining in a microscopic field and concomitant analysis of nuclear chromatin morphology. In VSMCs stably overexpressing Bcl-2, we observed preserved cell viability after 24 hours of serum withdrawal to levels comparable with those of cells maintained in serum. However, transfection with the Bad expression vector construct abrogated the survival-promoting influence of Bcl-2 and induced apoptosis in 31% of cells under serum-free conditions (31±4% Bad transfected in serum-free conditions versus 5±1% Bad transfected in serum; P<0.01). Furthermore, the proapoptotic effect of Bad transfection was significantly attenuated by IGF-I treatment (31±4% Bad transfected in serum-free conditions versus 15±2% Bad transfected with IGF-I; n=6, P<0.01). Thus, the IGF-I-induced serine phosphorylation of Bad is associated with an attenuation of its proapoptotic effects on VSMC fate.
In the present study we have characterized a signal transduction pathway involved in regulating VSMC apoptosis. Our findings indicate that activation of PI 3-kinase plays a critical mediator role in preventing VSMC death. Moreover, we document for the first time that VSMCs express the proapoptotic mediator Bad and that the inactivation of Bad by serine phosphorylation is mediated by a PI 3-kinase–dependent pathway in these cells. Overall, these data provide evidence for an antiapoptotic signal transduction pathway in VSMCs in which protein kinases involved in generating second messengers are coupled to apoptosis regulatory genes that govern cell fate.
A rapidly emerging body of evidence suggests that vascular remodeling and lesion formation are determined in large part by the balance between cell proliferation and apoptotic cell death.2 4 5 6 We have previously shown that vasoactive substances such as angiotensin II may be important determinants of VSMC fate by modulating programmed cell death.14 Similarly, Bennett et al8 have documented that classic peptide mitogens such as IGF-I and platelet-derived growth factor-BB are also potent inhibitors of VSMC death. The present study has confirmed this observation in both human and rat VSMCs at physiological concentrations in a dose-dependent and time-dependent manner. However, the intracellular signaling pathways by which growth factors or vasoactive substances promote VSMC survival remain to be elucidated.
Several recent studies in neuronal and hematopoietic cell lines have suggested that a signal transduction pathway that includes the lipid kinase PI 3-kinase plays an important role in regulating cell death.9 24 However, it has become clear that the regulation of apoptosis is cell specific and contextual, and the role of these pathways as determinants of VSMC fate remained unclear. PI 3-kinase is a heterodimeric protein composed of 85- and 110-kDa subunits that catalyze the phosphorylation of the D3 portion of PI.25 The regulatory p85 subunit of PI 3-kinase has no intrinsic catalytic activity but forms complexes with activated growth factor receptors as well as adapter proteins such as IRS-1 and Shc via SH2 domains. The resultant association leads to a translocation of the catalytic p110 subunit from the soluble fraction to a vesicular fraction enriched in Golgi membranes. This maneuver enhances the catalytic activity of p110 and leads to the production of PI3-phosphate, PI(3,4)-biphosphate, and PI(3,4,5)-triphosphate. These lipids subsequently serve as second messengers for specific downstream signal transduction events such as receptor endocytosis, intracellular vesicular trafficking, and the regulation of cell survival.25
The present study has examined the IGF-I–induced survival-signaling cascade in VSMCs and documented that it is mediated by a PI 3-kinase–dependent pathway according to the following lines of evidence: (1) IGF-I induces activation of PI 3-kinase, (2) 2 structurally dissimilar PI 3-kinase inhibitors abolished IGF-I–induced survival effects, and (3) overexpression of a dominant-negative Δp85 mutant construct of PI 3-kinase abolished the capacity of IGF-I to preserve VSMC viability under serum-free conditions. Taken together, these findings provide substantial evidence that PI 3-kinase is an important antiapoptotic signaling pathway in VSMCs. However, the downstream mediators of this survival signal remained unclear.
Recent studies provide a rapidly growing body of evidence that indicates that Akt/PKB (protein kinase B) is a critical downstream effector of PI 3-kinase that promotes cell survival in nonvascular cells.19 26 Akt/PKB, the cellular homologue of the viral oncogene v-akt, encodes a serine/threonine protein kinase that is ubiquitously expressed and of which the catalytic domain is closely related to the catalytic domain of all the members of protein kinase C family. Akt/PKB differs in that it contains an N-terminal domain related in part to the pleckstrin homology domain.27 28 We speculate that Akt may be the critical distal element of the PI 3-kinase–dependent VSMC antiapoptotic signaling pathway. Studies are underway to determine the potential role of Akt/PKB as a determinant of VSMC fate.
It is now well established that members of the Bcl-2 family are critical regulators of apoptosis in a variety of cell types. The expression pattern and role of different members of the family appears to be cell specific.29 Our laboratory has recently demonstrated that vascular lesion formation is associated with an upregulation of the antiapoptotic gene Bcl-XL within intimal VSMCs in animal models and human specimens of vascular disease. Moreover, we have demonstrated that downregulation of Bcl-XL expression within intimal cells using antisense oligonucleotides induces VSMC apoptosis and regression of vascular lesions.30 These findings indicate that Bcl-XL is a critical determinant of intimal VSMC viability and lesion formation.
Bcl-XL and Bcl-2 appear to inhibit cell suicide by several mechanisms that include the following: binding to the caspase CED-4 (Apaf-1) to prevent the activation of the proteolytic caspase-mediated cell execution cascade, altering mitochondrial membrane potential, and inhibiting cytochrome c.31 32 33 The antiapoptotic activity of mediators such as Bcl-XL is modulated in part by the formation of heterodimers with proapoptotic members of the Bcl-2 family such as Bad, Bak, and Bax.10 It is postulated that these proapoptotic mediators may promote cell death in part by sequestering the antiapoptotic mediators such as Bcl-XL or Bcl-2. Recent studies have demonstrated that formation of Bad-Bcl-XL or Bad-Bcl-2 heterodimers is regulated by post-translational modifications of Bad. The phosphorylation of Bad at 2 sites, serine 112 and serine 136, substantially reduces its capacity to bind to Bcl-XL or Bcl-2. In addition, this serine phosphorylated form of Bad becomes associated with the docking protein 14-3-3.20 21 Thus, the phosphorylation of Bad promotes its sequestration from the apoptosis regulatory machinery and thereby plays a permissive role in promoting the antiapoptotic effect of Bcl-XL or Bcl-2. In hematopoietic cells, the cytokine interleukin-3 stimulates cell survival in association with phosphorylation of Bad at serine 112 and serine 136.21 Mutagenesis studies indicate that phosphorylation of these serine residues is a necessary condition for the antiapoptotic effect of interleukin-3 and that Akt/PKB is one of the serine kinases that may induce Bad phosphorylation.18 34 The present study has documented that VSMC survival is associated with phosphorylation of Bad and that the inactivation of Bad by this post-translational modification is mediated by a PI 3-kinase–dependent pathway in VSMCs. During the preparation of this manuscript, this observation was confirmed by reports using nonvascular cells.18 34 Furthermore, we observed that Bad inhibits the antiapoptotic influence of Bcl-2 in VSMCs and that this proapoptotic influence of Bad can be modulated by extracellular stimuli such as IGF-I.
In summary, the present study has defined an antiapoptotic signal transduction pathway that couples extracellular stimuli to the activity of apoptosis regulatory genes in VSMCs. Our data suggest a working model in which VSMC fate is critically dependent on a cell-survival signaling cascade mediated by PI 3-kinase, the generation of phosphoinositols, and activation of the serine kinase Akt. This receptor-activated, antiapoptotic signaling pathway is linked to the cell death program by its capacity to phosphorylate Bad and thereby reduce its capacity to inhibit antiapoptotic mediators such as Bcl-XL or Bcl-2. The findings of these in vitro experiments combined with our in vivo studies30 suggest that factors that modulate the expression level and/or activity of antiapoptotic genes such as Bcl-XL may play a fundamental pathogenic role in vascular disease by regulating VSMC apoptosis.
The authors gratefully acknowledge grant support from the NIH, the American Heart Association, and the Pew Charitable Trusts.
- Received January 15, 1999.
- Accepted May 19, 1999.
- © 1999 American Heart Association, Inc.
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature. 1993;62:801–809.
Gibbons GH, Dzau VJ. Molecular therapies for vascular diseases. Science. 1996;272:689–693.
Isner JM, Kearney M, Bortman S, Passeri J. Apoptosis in human atherosclerosis and restenosis. Circulation. 1995;91:2703–2711.
Mizutani M, Kern TS, Lorenzi M. Accelerated death of retinal microvascular cells in human and experimental diabetic retinopathy. J Clin Invest. 1996;7:2883–2890.
Bennett MR, Evans GI, Schwartz SM. Apoptosis of human vascular smooth muscle cells derived from normal vessels and coronary atherosclerotic plaques. J Clin Invest. 1995;95:2266–2274.
Yao R, Cooper GM. Requirement for phosphatidylinositol-3 kinase in the prevention of apoptosis by nerve growth factor. Science. 1995;267:2003–2006.
Pollman MJ, Sherwood SW, Gibbons GH. Cell cycle arrest induces vascular smooth muscle cell programmed cell death: implications for antiproliferative therapies for restenosis. Circulation. 1995;92 (suppl I):I-101. Abstract.
Kimura K, Hattori S, Kabuyama Y, Shizawa Y, Takayanagi J, Nakamura S, Toki S, Matsuda Y, Onodera K, Fukui Y. Neuric outgrowth of PC12 cells is suppressed by wortmannin, a specific inhibitor of phosphatidylinositol 3-kinase. J Biol Chem. 1994;269:18961–18967.
Vlahos CJ, Matter WF, Hui KY, Brown RF. A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002). J Biol Chem. 1994;269:5241–5248.
Pollman MJ, Yamada T, Horiuchi M, Gibbons GH. Vasoactive substances regulate vascular smooth muscle cell apoptosis: countervailing influences of nitric oxide and angiotensin II. Circ Res. 1996;79:748–756.
Hu Z, Shi X, Lin RZ, Hoffman BB. α1 adrenergic receptors activate phosphatidyl inositol 3-kinase in human vascular smooth muscle cells. J Biol Chem. 1996;271:8977–8982.
Hara K, Yonezawa K, Sakue H, Ando A, Kotani K, Kitamura T, Kitamura Y, Ueda H, Stephens L, Jackson TR, Hawkins PT, Dhand R, Clark AE, Holman GD, Waterfield MD, Kasuga M. 1-Phosphatidylinositol 3-kinase activity is required for insulin-stimulated glucose transport but not for RAS activation in CHO cells. Proc Natl Acad Sci U S A. 1994;91:7415–7419.
Parrizas M, Saltiel AR, LeRoith D. Insulin-like growth factor 1 inhibits apoptosis using the phosphatidylinositol 3-kinase and mitogen-activated protein kinase pathways. J Biol Chem. 1997;272:154–161.
Peso LD, Gonzalez-Garcia M, Page C, Herrera R, Nunez G. Interleukin-3-induced phosphorylation of Bad through the protein kinase Akt. Science. 1997;278:687–689.
Dudek H, Dotta SR, Franke TF, Birnbaum MJ, Yao R, Cooper GM, Segal RA, Kaplan DR, Greenberg ME. Regulation of neuronal survival by the serine-threonine protein kinase Akt. Science. 1997;275:661–665.
Zha J, Harada H, Osipov K, Jockel J, Waksman G, Korsmeyer SJ. BH3 domain of BAD is required for heterodimerization with Bcl-XL and pro-apoptotic activity. J Biol Chem. 1997;272:24101–24104.
Stokoe D, Stephens LR, Copeland T, Gaffney PRJ, Reese CB, Painter GF, Holmes AB, McCormick F, Hawkins PT. Dual role of phosphatidylinositol-3,4,5-triphosphate in the activation of protein kinase B. Science. 1997;277:567–570.
Marianna P, Cooper GM. Role of glycogen synthase kinase-3 in the phosphatidylinositol 3-kinase/Akt cell survival pathway. J Biol Chem. 1998;273:19929–19932.
Minshall C, Arkins S, Freund CG, Kelley KW. Requirement for phosphatidylinositol 3-kinase to protect hemopoietic progenitors against apoptosis depends upon the extracellular survival factor. J Immunol. 1996;156:936–947.
Coffer PJ, Jin J, Woodgett JR. Protein kinase B (c-Akt): a multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem J. 1998;335:1–13.
Evan G, Littlewood T. A matter of life and cell death. Science. 1998;281:1317–1322.
Chinnaiyan AM, O’Rourke K, Lane BR, Dixit YM. Interaction of ced-3 and ced-9: a molecular framework for cell death. Science. 1997;275:1122–1126.
Decaudin D, Geley S, Hirsch T, Castedo M, Marchetti P, Macho A, Kofler R, Kroemer G. Bcl-2 and Bcl-XL antagonize the mitochondrial dysfunction preceding nuclear apoptosis induced by chemotherapeutic agents. Cancer Res. 1997;57:62–67.
Kharbanda S, Pandey P, Schofield L, Israels S, Roncinske R, Yoshida K, Bharti A, Yuan Z, Saxena S, Weichselbaum R, Nalin C, Kufe D. Role of Bcl-XL as an inhibitor of cytosolic cytochrome C accumulation in DNA damage-induced apoptosis. Proc Natl Acad Sci U S A. 1997;94:6939–6942.