This article has a correction. Please see:

Abstract

Transforming growth factor-β (TGF-β) is upregulated at the time of arterial injury; however, the mechanism through which TGF-β enhances the development of intimal hyperplasia is not clear. Recent studies from our laboratory suggest that in the presence of elevated levels of Smad3, TGF-β stimulates smooth muscle cell (SMC) proliferation. This is a novel phenomenon in that TGF-β has traditionally been known as a potent inhibitor of cellular proliferation. In these studies we explore the signaling pathways through which TGF-β mediates its proliferative effect in vascular SMCs. We found that TGF-β phosphorylates and activates Akt in a time-dependent manner, and this effect is significantly enhanced by overexpression of Smad3. Furthermore, both chemical and molecular inhibition of Smad3 can reverse the effect of TGF-β on Akt. Although we found numerous signaling pathways that might function as intermediates between Smad3 and Akt, p38 appeared the most promising. Overexpression of Smad3 enhanced p38 phosphorylation and inhibition of p38 with a chemical inhibitor or a small interfering RNA blocked TGF-β-induced Akt phosphorylation. Moreover, TGF-β/Smad3 enhancement of SMC proliferation was blocked by inhibition of p38. Phosphorylation of Akt by TGF-β/Smad3 was not dependent on gene expression or protein synthesis, and immunoprecipitation studies revealed a physical association among p38, Akt, and Smad3 suggesting that activation requires a direct protein-protein interaction. Our findings were confirmed in vivo where overexpression of Smad3 in a rat carotid injury model led to enhancement of p-p38, p-Akt, as well as SMC proliferation. Furthermore, inhibition of p38 in vivo led to decreased Akt phosphorylation and SMC proliferation. In summary, our studies reveal a novel pathway whereby TGF-β/Smad3 stimulates SMC proliferation through p38 and Akt. These findings provide a potential mechanism for the substantial effect of TGF-β on intimal hyperplasia and suggest new targets for chemical or molecular prevention of vascular restenosis.

  • intimal hyperplasia
  • transforming growth factor-β
  • phosphoinositide-3 kinase/Akt
  • p38 mitogen-activated protein kinase

despite recent progress with coated coronary stents, restenosis following vascular intervention remains an unresolved problem. Intimal hyperplasia is a complex process occurring after arterial injury that leads to restenosis and failure of vascular reconstructions. The response to arterial injury is a multifaceted process initially characterized by inflammation and infiltration of monocytes, followed by vascular smooth muscle cell (SMC) migration and proliferation as well as recruitment of bone marrow-derived or tissue stem cells (56, 57). Although the precise mechanism that underlies the formation of intimal hyperplasia is not clear, the involvement of the cytokine transforming growth factor-β (TGF-β) is undisputed.

TGF-β is a multifunctional cytokine that mediates a number of complex responses that contribute to various phases of tissue repair following injury. TGF-β primarily signals through the Smad protein family. Specifically, TGF-β binds to its type I and II receptors followed by Smad2/3 phosphorylation. Phospho-Smad2/3 then forms a complex with Smad4 that translocates to the nucleus where it is involved in the transcriptional regulation of various target genes (2, 7, 15, 59). TGF-β also signals through Smad-independent pathways that have been found to modulate cellular process such as survival and differentiation (2, 15).

TGF-β has been implicated as a contributor to intimal hyperplasia for over two decades (32, 33, 54). In a number of in vivo models of intimal hyperplasia, increased expression of TGF-β has been observed (6). Elevated expression of TGF-β has also been observed in human restenotic lesions (42, 60). In animal models, blocking TGF-β at a receptor level has resulted in inhibition of intimal hyperplasia. However, the ubiquitous nature of TGF-β and its multiple downstream effects has made inhibitory strategies directed at TGF-β or its receptor unappealing (38, 48, 58). Deciphering the signaling mechanisms through which TGF-β promotes intimal hyperplasia may provide more effective and discriminatory therapeutic targets.

The mechanism through which TGF-β promotes intimal hyperplasia is unclear. Its effects were initially thought to be mediated through the enhancement of extracellular matrix; however, it has been recently shown that extracellular matrix is not a major component of hyperplastic plaque. Alternatively, intimal hyperplasia is a highly cellular process of which SMC proliferation is a central feature. Thus the fact that TGF-β profoundly inhibits SMC proliferation in vitro is perplexing (36). Emerging data suggest that TGF-β can have diverse effects on cells that depend on the cell type and the cellular environment. Our recent studies have shown that TGF-β is paradoxically a potent stimulant of SMC proliferation in vivo. At the time of arterial injury, there is significant expression of Smad3. Moreover, it appears in the presence of elevated levels of Smad3, TGF-β stimulates rather than inhibits vascular SMC proliferation (57). Overexpression of Smad3 in a rat carotid injury model led to enhancement of SMC proliferation as well as intimal hyperplasia (57). The mechanism through which TGF-β mediates its proliferative effect on vascular SMCs remains elusive and is the subject of this analysis.

The phosphoinositide-3 kinase (PI3K)/Akt signal transduction pathway, a known mediator of cell-cycle progression, has previously been shown to control vascular SMC proliferation both in vivo and in vitro (53). Thus we hypothesize that Smad3 may mediate its proliferative effect through a pathway involving Akt. Our findings demonstrate that TGF-β can indeed stimulate SMC proliferation both in vitro and in vivo through a novel pathway that involves Smad3, p38, and Akt.

MATERIALS AND METHODS

Reagents.

Chemical inhibitors for Smad3 (SIS3), extracellular signal-regulated kinase (ERK) 1 and 2 mitogen-activated protein (MAP) kinase (PD98059), p38 MAP kinase (SB203580), Jun NH2-terminal kinase (JNK) MAP kinase (SP600125), and PI3K (wortmannin) were obtained from Calbiochem (San Diego, CA). TGF-β receptor kinase inhibitor (SB431542) was purchased from Selleck Chemicals (Houston, TX). Recombinant TGF-β1 was purchased from R&D Systems (Minneapolis, MN). Dulbecco's modified Eagle's medium (DMEM) and cell culture reagents were from Invitrogen (Carlsbad, CA). Other reagents, if not specified, were purchased from Sigma (St. Louis, MO).

Smooth muscle cell culture.

Rat aortic vascular SMCs were isolated from the thoracoabdominal aorta of male Sprague-Dawley rats based on a protocol described by Clowes et al. (8) and maintained in DMEM containing 10% FBS at 37°C with 5% CO2. Cell viability was assayed by the trypan blue exclusion method, which indicated that <5% of the cells took up the dye both before and after the infection of adenoviral vectors or treatment with recombinant TGF-β1 or chemical inhibitors. For experiments using adenoviral overexpression, vascular SMCs were infected with adenovirus (3 × 104 particles/cell) in DMEM containing 2% FBS for 4 h at 37°C followed by recovery in 10% FBS overnight. Cells were then starved in DMEM containing 0.5% FBS for 24 h. The cells were then treated with recombinant TGF-β1 (5 ng/ml) or solvent (4 mm HCl with 2% bovine serum albumin) for 12 h.

Construction of adenoviral vectors and infection.

Adenoviral (Ad) vectors expressing Smad3 (AdSmad3) and green fluorescent protein (GFP) (AdGFP) were constructed as previously described (62). AdGFP was used as a control.

siRNA knockdown of Smad3 and p38 MAPK.

Vascular SMCs were plated at 50–60% confluence in DMEM culture medium in six-well plates and incubated for 24 h. Cells were then transfected in Opti-MEM I medium with 20 nM of small interfering RNA (siRNA) for Smad3 (Invitrogen), 20 nM of siRNA for p38 MAP kinase (MAPK, Santa Cruz Biotechnology, Santa Cruz, CA), or control siRNA using RNAiMax transfection reagent (Invitrogen), as described by the manufacturer's protocol. After 6 h, the Opti-MEM I medium was replaced by DMEM containing 0.5% FBS for 24 h. For experiments combining knockdown and overexpression, cells were infected with adenovirus as described above and allowed to recover overnight before being transfected with siRNA.

Western blot analysis.

Cells were lysed in RIPA buffer (50 mm Tris, 150 mm NaCl, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate, and 10 μg/ml aprotinin). Thirty micrograms of protein from each sample were separated on 10% SDS-PAGE. The protein samples were then transferred to nitrocellulose membranes. Protein expression was assessed by immunoblotting with the following antibodies: rabbit anti-pAkt-Ser473 or total Akt, rabbit anti-phosphorylated p38 (p-p38) or total p38 (Cell Signaling, Boston, MA), rabbit anti-Smad3 (Invitrogen), and mouse anti-β-actin (Sigma). After incubation with the appropriate primary and horseradish peroxidase-conjugated secondary antibodies, the membranes were developed with enhanced chemiluminescence reagent (Pierce, Davenport, IL).

Immunoprecipitation.

Immunoprecipitation was performed as previously described (18). Briefly, cells were lysed in Nonidet P-40 buffer. The protein concentration was determined by a Bio-Rad kit and about 250 μg total protein of each sample was used.Protein A-Sepharose beads (Santa Cruz Biotechnology) and 5 μg of rabbit anti-Smad3 primary antibody (Invitrogen) or 5 μg of rabbit IgG (Cell Signaling) were added to lysates and incubated at 4°C with constant rotation for 12 h. After centrifugation was completed, pellets were washed five times with Nonidet P-40 buffer and once with 50 mM Tris. The final pellet was resuspended in 20 μl of sample buffer and heated to 100°C for 5 min. Samples were then separated on 10% SDS-PAGE gels. The protein samples were then transferred to nitrocellulose membranes. Protein expression was confirmed by immunoblotting with the following antibodies: rabbit anti-p-Akt-Ser473, anti-p-p38, and anti-p-Smad3 (Cell Signaling).

Cell viability and proliferation assay.

Cell viability was determined by modified 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay (10). Cells were plated at 50–60% confluence on 24-well plates and incubated for 24 h. Cells were then infected with adenovirus (3 × 104 particles/cell) in DMEM containing 2% FBS for 4 h at 37°C followed by recovery in DMEM containing 10% FBS for 12 h. After recovery, Cells were then transfected in Opti-MEM I medium with 20 nM of siRNA for p38 MAPK (Santa Cruz Biotechnology) or control siRNA using RNAiMax transfection reagent (Invitrogen), as described by the manufacturer's protocol. After 6 h, the Opti-MEM I medium was replaced by DMEM containing 0.5% FBS for 24 h. Recombinant TGF-β1 (5 ng/ml) or solvent as a control (4 mm HCl with 2% bovine serum albumin) was then added for 96 h. MTT solution (25 μl; 0.5 mg/ml) in phenol red-free culture medium (0.5% FBS) was added to each well and incubated for 2 h at 37°C. The medium was aspirated, and 500 μl acidic isopranolol (0.04 M HCl) were added for 10 min to solubilize the intracellular formazan product. Triplicates (100 μl) of each well were transferred to a 96-well plate, and absorbance was measured at 570 nm using absorbance at 690 nm to correct for background.

Rat balloon injury model and in vivo gene delivery.

Male Sprague-Dawley rats (300–350 g) underwent balloon injury of the left common carotid artery as described elsewhere in accordance with institutional guidelines and approval by the Animal Care and Use Committee at the University of Wisconsin-Madison (18). Briefly, after induction of anesthesia with isoflurane, a 2-French balloon catheter was inserted through the left external carotid artery into the common carotid and insufflated with 2 atm of pressure three times. After injury, animals received intraluminal administration of adenoviral vectors (2.5 × 109 plaque-forming units in 200 μl of PBS over 20 min). The external carotid artery was then ligated, and flow was reestablished through the common carotid and internal carotid arteries. Rats were euthanized 3 days after injury and fixed in 4% paraformaldahyde overnight for paraffin embedding.

For in vivo experiments involving inhibition of p38, rats were treated with SB203580 (10 mg/kg ip) or vehicle (1% DMSO in saline ip) 30 min before injury. SB203580 (10 mg/kg ip) or vehicle (1% DMSO in saline ip) was administered once daily for 3 days, and rats were then euthanized. One-third of the artery was fixed in 4% paraformaldahyde overnight for paraffin-embedding, and two-thirds was snap frozen in liquid nitrogen for homogenization and Western blot analysis.

Immunohistochemistry.

Paraffin-embedded arteries were cut into 6-μm sections for analysis. Immunostaining for Smad3, p-Smad3, phosphorylated Akt (p-Akt), p-p38, and Ki67 were performed as described previously (17). Antibody controls included detection of species-matched IgG. For quantification, five sections from each animal were chosen. Six different fields were then imaged from each section at ×400, and two independent investigators then manually counted the number of immunopositive as well as the total number of cells. This was a binary decision. Cells were either positive or negative. All ratios were then averaged to generate the mean and standard deviation for that animal. The means were then averaged, and the standard error of the mean was calculated for each group of animals.

Statistical analysis.

Values were expressed as means ± SE derived from at least three independent experiments, unless stated otherwise. Differences between two groups were analyzed by two-tailed Student's t-test. For comparisons between more than two sets of experimental conditions, we used a one-way analysis of variance (ANOVA). If significant, the ANOVA was followed by Tukey's multiple comparison test. P values <0.05 were considered as statistically significant.

RESULTS

TGF-β induces Akt activation in vascular SMCs.

We have previously shown that TGF-β/Smad3 promotes vascular SMC proliferation through cytoplasmic sequestration of p27. Cytoplasmic sequestration of p27 has been associated with serine-10 phosphorylation, which is known to be regulated by PI3K/Akt (3, 4, 22, 26). Consequently, we explored whether TGF-β might activate Akt in vascular SMCs. Treatment with all doses of TGF-β tested (1 to 10 ng/ml) significantly increased Akt phosphorylation (Fig. 1A). Consequently, 5 ng/ml was chosen as the dose for all future experiments. Akt phosphorylation peaked between 8 and 12 h when stimulated with 5 ng/ml of TGF-β (Fig. 1B). To verify that Akt phosphorylation requires activation of the TGF-β receptor, we preincubated SMCs for 30 min with the TGF-β receptor kinase inhibitor (SB431542) and then stimulated with TGF-β for 12 h. This inhibitor completely abolished TGF-β-induced Akt phosphorylation (Fig. 1C). These data suggest that Akt may be the signaling link between TGF-β and p27.

Fig. 1.

Transforming growth factor-β (TGF-β)/Smad3 induces Akt phosphorylation in vascular smooth muscle cells (SMCs). A: vascular SMCs were treated with either 1, 2, 5, or 10 ng/ml of TGF-β for 12 h, and phosphorylated Akt (p-Akt) was measured by Western blot analysis (*P < 0.05, compared with solvent). B: vascular SMCs were treated with TGF-β (5 ng/ml) for the specified times, and phosphorylated Akt (p-Akt) was measured by Western blot analysis (*P < 0.05). C: vascular SMCs were pretreated with the TGF-β receptor inhibitor SB431542 (10 μM) for 30 min and then stimulated with TGF-β (5 ng/ml) for 12 h, and p-Akt was measured by Western blot analysis (*P < 0.05). Data shown are representative of three independent experiments *Significant ANOVA and Tukey comparisons.

Akt activation in response to TGF-β is mediated through a Smad3-dependent pathway.

We next explored whether the effect of TGF-β on Akt is mediated by a Smad-dependent or independent pathway. Since TGF-β enhances SMC proliferation only in the presence of elevated levels of Smad3, we hypothesized that the effect of TGF-β on Akt is Smad dependent. To prove our hypothesis, we infected cultured vascular SMCs with adenovirus-expressing Smad3 (AdSmad3) or control (AdGFP) followed by stimulation with or without TGF-β for 12 h. TGF-β once again led to an increase in p-Akt; however, this effect was substantially enhanced in cells overexpressing Smad3 (Fig. 2A). To further confirm the role of Smad3 in phosphorylation of Akt, we employed two distinct methods of Smad3 inhibition. First, we preincubated SMCs with a Smad3 inhibitor (SIS3) for 15 min before stimulating cells with TGF-β. We found that inhibition of Smad3 with SIS3 reduced TGF-β-induced Akt phosphorylation to basal levels (Fig. 2B). To further prove that TGF-β activates Akt through a Smad3-mediated pathway, we knocked down Smad3 using a siRNA. Smad3 siRNA, compared with scramble, markedly diminished TGF-β-induced Akt phosphorylation (Fig. 2C). In another set of experiments, we overexpressed Smad3, then incubated these cells with Smad3 siRNA, and then stimulated with TGF-β for 12 h. Overexpression of Smad3 greatly enhanced p-Akt, whereas inhibition with Smad3 siRNA eliminated this effect (Fig. 2D). Through these experiments, we conclusively prove that TGF-β induces Akt phosphorylation in vascular SMCs in a Smad3-dependent fashion.

Fig. 2.

Blockade of TGF-β receptor or Smad3 reduces activation of Akt. A: vascular SMCs were infected with control virus [adenoviral (Ad) green fluorescent protein (AdGFP)] or AdSmad3 followed by stimulation with TGF-β (5 ng/ml) for 12 h, and p-Akt was measured by Western blot analysis (*P < 0.05, compared with AdGFP alone). B: vascular SMCs were pretreated with the Smad3 inhibitor SIS3 (2.5 μM) for 15 min followed by treatment with TGF-β (5 ng/ml) for 12 h (*P < 0.05). C: vascular SMCs were then infected with either scramble or Smad3 siRNA for 6 h, starved for 24 h in 0.5% media, and then stimulated with TGF-β (5 ng/ml) for 12 h (*P < 0.05). D: vascular SMCs were transfected with AdSmad3 for 4 h, recovered, and then incubated with 20 nM small interfering RNA (siRNA) Smad3 for 6 h. Cells were then starved for 24 h in 0.5% media followed by stimulation with TGF-β (5 ng/ml) for 12 h. p-Akt and Smad3 were measured by Western blot analysis (*P < 0.05). Data shown are representative of three independent experiments. *Significant ANOVA and Tukey comparisons.

TGF-β/Smad3-induces activation of Akt through a pathway that involves p38 MAPK.

Having verified that TGF-β-induced activation of Akt is Smad3 dependent, we searched for a signaling intermediate that would connect Smad3 and Akt. As a first step, we infected vascular SMCs with AdSmad3, pretreated cells with inhibitors to ERK (PD98059), p38 MAPK (SB203580), JNK (SP 600125), PKA (H89), PKC-δ (rottlerin), and PI3K (wortmannin), followed stimulation with TGF-β (5 ng/ml) for 12 h. Using Akt phosphorylation as the end point, we found that the p38 MAPK, JNK, and PI3K inhibitors completely suppressed TGF-β/Smad3-induced Akt phosphorylation. Alternatively, the ERK and PKC-δ inhibitors only partially blocked TGF-β/Smad3-induced Akt activation, whereas the PKA inhibitor had no effect (Fig. 3A). These results suggest p38 MAPK and JNK and PI3K are important intermediates for TGF-β/Smad3-induced Akt phosphorylation. We continued our exploration, focusing on p38 MAPK. We first examined p38 activation following TGF-β/Smad3 stimulation by treating cultured vascular SMCs with TGF-β (5 ng/ml) for 12 h and assessing p38 activation by Western blot analysis. We found that TGF-β produced a rapid increase in p-p38 without affecting total p38 MAPK levels. Overexpression of Smad3 in vascular SMCs further enhanced TGF-β-induced p38 activation, paralleling our findings with Akt (Fig. 3B). To further establish p38 as an intermediate between TGF-β/Smad3 and Akt we inhibited p38 using a siRNA. Vascular SMCs overexpressing Smad3 were pretreated with siRNA to p38 and stimulated with TGF-β for 12 h followed by Western blot analysis for p-Akt. Elimination of p38 completely inhibited TGF-β/Smad3-induced phosphorylation of Akt (Fig. 3C). These results suggest that p38 acts as a signaling intermediate between Smad3 and Akt.

Fig. 3.

TGF-β/Smad3-induced activation of Akt requires p38 MAPK. A: vascular SMCs infected with AdSmad3 were preincubated for 30 min with inhibitors to extracellular signal-regulated kinase-1/2 (ERK1/2) (PD98059, 10 μM), p38 mitogen-activated protein kinase (MAPK) (SB203580, 10 μM), Jun NH2-terminal kinase (JNK) (SP600125, 10 μM), PKA (H89, 10 μM), PKC-δ (rotterlin, 2.5 μM), and phosphoinositide-3 kinase (PI3K) (wortmannin,1 μM). These cells were then stimulated with TGF-β (5 ng/ml) for 12 h followed by Western blot analysis for p-Akt (*P < 0.05 compared with solvent). B: vascular SMCs infected with AdGFP or AdSmad3 were stimulated with TGF-β (5 ng/ml) for 12 h followed by Western blot analysis for p-p38 (*P < 0.05 compared with AdGFP alone). C: vascular SMCs infected with AdSmad3 were transfected with 20 nM of control or p38 siRNA for 6 h followed by starvation in 0.5% serum for 24 h. Cells were then stimulated with TGF-β (5 ng/ml) for 12 h followed by Western blot analysis for p-Akt (*P < 0.05). Data shown are representative of three independent experiments. *Signficant ANOVA and Tukey comparisons.

Blockade of p38 MAPK decreases TGF-β/Smad3-induced cell proliferation.

Next, we explored the physiological relevance of these findings by investigating the role of the p38 pathway in TGF-β/Smad3-induced vascular SMC proliferation. TGF-β inhibits proliferation in many cell types. However, we have previously found that TGF-β promotes proliferation in vascular SMCs in the presence of elevated levels of Smad3 (14, 39, 57). In our first set of experiments we evaluated the role TGF-β alone on SMC proliferation (Fig. 4A). SMCs were starved for 24 h and then incubated with TGF-β (5 ng/ml) for 96 h followed by evaluation of proliferation through a MTT assay. Consistent with our previous findings, TGF-β without excess Smad3 produces an ∼15% reduction in SMC proliferation (38). Next, we overexpressed GFP or Smad3 and then pretreated these cells with SB203580, an inhibitor of p38 MAPK for 30 min. Cells were then incubated with TGF-β for 96 h (Fig. 4B). In another set of experiements, cells overexpressing GFP or Smad3 were incubated with either control siRNA or siRNA p38 for 6 h and then stimulated with TGF-β for 96 h (Fig. 4C). In the forgoing experiments, proliferation was assessed using MTT assay. We show that TGF-β, in cells overexpressing Smad3, stimulates SMC proliferation. Interestingly, cells overexpressing Smad3 alone also showed a substantial increase in proliferation. Chemical inhibition or siRNA knockdown of p38 produced a marked diminution in cell numbers (Fig. 4, B and C). Thus p38 appears to be essential for TGF-β/Smad3 stimulation of vascular SMC proliferation.

Fig. 4.

Knockdown of p38 MAPK attenuates TGF-β/Smad3 induced SMC proliferation. A: vascular SMCs were starved in 0.5% serum for 24 h and then treated with TGF-β (5 ng/ml) or solvent for 96 h. Cell proliferation was evaluated by a 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay (*P < 0.05, t-test). B: vascular SMCs were infected with AdSmad3 or AdGFP. Cells were then pretreated for 30 min with the p38 MAPK inhibitor SB203580 (10 μM) followed by TGF-β (5 ng/ml) stimulation for 96 h. Proliferation was assessed using MTT (*P < 0.05, ANOVA and Tukey comparisons) C: vascular SMCs infected with AdGFP or AdSmad3 were transfected with 20 nM of control or p38 siRNA for 6 h followed by starvation in 0.5% serum for 24 h. Cells were then stimulated with TGF-β (5 ng/ml) for 96 h, and proliferation was assessed using MTT (*P < 0.05, #P < 0.01, ANOVA and Tukey comparisons). Data shown are representative of three independent experiments.

Phosphorylation of Akt by Smad3 does not require gene expression or protein synthesis but rather a direct protein-protein interaction.

Smad3 often affects cell function through gene expression and protein synthesis. To explore the mechanism through which Smad3 leads to enhancement of Akt phosphorylation, we pretreated vascular SMCs infected with Smad3 with either actinomycin D (Act D, 5 μg/ml) or cycloheximide (CHX, 10 μg/ml) before incubating these cells with TGF-β or solvent for 12 h. TGF-β robustly stimulated phosphorylation of Akt even in the presence of Act D or CHX (Fig. 5A), suggesting that TGF-β/Smad3 regulates Akt via a mechanism independent of transcription or translation. We then used studies involving immunoprecipitation to evaluate whether there might be a physical association between Smad3, p38, and Akt. Vascular SMCs were treated with AdSmad3 followed by stimulation with TGF-β (5 ng/ml) for 12 h. Cells were then lysed and immunoprecipitated with an antibody to Smad3 or IgG control. The immunocomplexes were then immunoblotted with antibodies to p-Akt, p-p38, and p-Smad3. At baseline, there appears to be a physical association between Smad3 and p-p38 and p-Akt. This association is significantly enhanced in cells overexpressing Smad3 and stimulated by TGF-β. This enhancement was 2.1 ± 0.23-fold for p-Akt, 2.9 ± 0.39 for p-p38, and 2.7 ± 0.57 for p-Smad3. These data suggest that Smad3 activation of Akt may require a direct physical association between these three proteins.

Fig. 5.

Phosphorylation of Akt by Smad3 does not require gene expression or protein synthesis but rather a direct protein-protein interaction. A: vascular SMCs were infected with AdSmad3 and then preincubated for 30 min with inhibitors to transcription (Act D, 5 μg/ml) or translation (cycloheximide, CHX, 10 μg/ml) followed by stimulation with TGF-β (5 ng/ml) or solvent for 12 h. Western blot analysis was performed for p-Akt. Data shown are representative of three independent experiments. B: vascular SMCs were treated with AdSmad3 or AdGFP followed by stimulation with TGF-β (5 ng/ml) for 12 h. Cell lysates were immunoprecipitated with Smad3 antibody. The immunocomplexes were then immunoblotted by with antibodies to p-Akt, p-p38, and p-Smad3. Data shown are representative of three independent experiments.

Smad3 overexpression enhances activation of p-p38, p-Akt, and SMC proliferation in an in vivo carotid injury model, and p38 inhibition reduces Akt phosphorylation and SMC proliferation.

Finally after establishing in vitro a signaling sequence that involves TGF-β, Smad3, p38, and Akt leading to SMC proliferation, we sought to confirm these findings in vivo. Male Sprague-Dawley rats underwent either sham surgery, standard carotid balloon angioplasty, or carotid balloon injury followed by intraluminal infusion of an adenoviral vector expressing either Smad3 or GFP. Animals were euthanized 3 days following injury, and immunostaining was performed for IgG, Smad3, p-p38, p-Akt and for the marker of proliferation Ki67 (Fig. 6A). Animals receiving standard balloon angioplasty alone had an increase in Smad3, p-Akt, p-p38, and Ki67 medial positivity compared with uninjured animals. Those vessels transfected with Smad3 displayed a markedly higher percentage of medial SMCs that were positive for Smad3. As well, there were an increase number of cells positive for p-p38 (AdGFP = 5.0 ± 1.3 vs. AdSmad3 = 24.5 ± 6.5) and p-Akt (AdGFP = 6.6 ± 1.6 vs. AdSmad3 = 22.6 ± 2.5) (Fig. 6B). Similar changes were seen on Western blotting of carotid artery homogenates, where p38 phosphorylation increased 1.8 ± 0.14-fold and Akt phosphorylation increased 1.6 ± 0.09-fold (Fig. 6C). Further analysis revealed that the activation of p38 and Akt was accompanied by a significant elevation in vascular SMC proliferation as measured by Ki67 (AdGFP = 18.2 ± 1.4 vs. AdSmad3 = 28.4 ± 4.8). Thus increased levels of Smad3 in vivo resulted in enhancement of both p-p38 and p-Akt, as well as SMC proliferation. These data demonstrate by association that the signaling pathway that we identified in vitro may exist in vivo.

Fig. 6.

Upregulation of Smad3 in vivo increases phosphorylation of both p38 and Akt as well as SMC proliferation. Inhibition of p38 in vivo decreases phosphorylation of Akt and SMC proliferation. Male Sprague-Dawley rats underwent sham surgery (n = 3), standard carotid balloon injury alone (n = 3), or balloon angioplasty followed by infusion of 200 μl of AdSmad3 (n = 6) or AdGFP (n = 3) (2.5 × 109 pfu/ml) over 20 min. Rats were treated intraperitoneally with SB203580 (SB, 10 mg/kg) or vehicle (1% DMSO in saline) 30 min before surgery. SB203580 or solvent was administered daily postoperatively. Animals were euthanized at 3 days, and one-third of each carotid artery was used for paraffin embedding and two-thirds were homogenized to extract protein for Western blot analysis. A: paraffin-embedded sections (6 μm) of carotid artery were stained with IgG, Smad3, p-Akt, p-p38, or Ki67 (magnification, ×400). B: graphical representation of the percentage of positive p-Akt, p-p38, and Ki67 cells per high-powered field (HPF) (*P < 0.05). C: Western blot analysis of homogenized carotid arteries. Results are expressed as the ratio of p-p38/p38 or p-Akt/Akt (*P < 0.05, represents signficant ANOVA and Tukey comparisons).

To confirm our in vitro findings that p38 had an essential role in the phosphorylation of Akt and SMC proliferation, we moved to our carotid balloon injury model. Animals were treated intraperitoneally with SB203580 (10 mg/kg) or vehicle (1% DMSO in saline) 30 min before injury. After balloon injury, animals received an intraluminal infusion of an adenoviral vector expressing Smad3 and administration of SB203580 (10 mg/kg) or vehicle (1% DMSO in saline) once daily for 3 days. Animals were euthanized 3 days following injury, and immunostaining was performed for Smad3, p-p38, p-Akt, and Ki67 (Fig. 6A). Consistent Smad3 overexpression was confirmed in all animals. In the group receiving SB203580, there was a significant reduction in p-p38-positive cells in the medial layer (AdSmad3 only group = 39.4 ± 3.69 vs. AdSmad3+SB203580 group = 13.5 ± 3.47). Parallel to our in vitro findings, inhibition of p38 reduced Akt phosphorylation, as evidenced by a decrease in p-Akt-positive cells (AdSmad3 only group = 28.7 ± 4.82 vs. AdSmad3+SB203580 group = 11.8 ± 3.63). The decrease in p38 and Akt phosphorylation was confirmed on Western blot analysis of carotid artery homogenates. Furthermore, expression of Ki67, the marker for proliferation, was also decreased with inhibition of p38 in this carotid balloon injury model (AdSmad3 only group = 44.1 ± 3.56 vs. AdSmad3+SB203580 group = 18.7 ± 2.47).

DISCUSSION

Herewithin we report a novel signaling pathway involving vascular SMCs through which TGF-β induces proliferation in the presence of elevated levels of Smad3 via a mechanism involving p38 MAPK and Akt. We found that TGF-β leads to a significant increase in p-Akt that is markedly enhanced in the presence of elevated levels of Smad3. We then demonstrated that p38 MAPK acts as an intermediate between TGF-β/Smad3 and PI3K/Akt. Finally, in vivo studies reveal that balloon injury of the carotid artery results in an enhancement of Smad3 expression, which coincides with an increase in p-Akt as well as cellular proliferation, which is dependent on p-p38. Taken together, these data suggest that TGF-β may, at least in part, produce its profound effect on intimal hyperplasia by stimulating vascular SMC proliferation through a pathway involving the forgoing sequence of signaling proteins.

A vast number of studies have confirmed the critical role of TGF-β in the development of vascular restenosis. The addition of TGF-β at the time of arterial injury results in an enhancement of intimal hyperplasia, and inhibition of TGF-β through neutralizing antibodies, antisense DNA, or receptor inhibitors diminishes the vascular response to injury (38, 48). In other disease processes, excess levels of TGF-β produce fibrosis related to robust production of matrix proteins, most often collagen. Accordingly, the influence of TGF-β on restenosis was long thought to be related to the deposition of extracellular matrix (ECM) (57). In fact, our own laboratory has demonstrated that TGF-β, through Smad3, increases synthesis of the ECM protein fibronectin in vascular SMCs (49). However, multiple investigations have now shown that hyperplastic plaque is highly cellular and is the consequence of migration and proliferation of SMCs, myofibroblasts, and bone marrow progenitor cells, rather than fibrosis.Thus there is the need for an alternative explanation for the effect of TGF-β on the arterial wall other than fibrosis.

Interestingly, TGF-β in the majority of cell types including epithelial cells and lymphocytes has been found to be a suppressant of cell growth and proliferation and a promoter of apoptosis (14, 36, 48). We and others (14, 39, 50, 57) have demonstrated that TGF-β inhibits both proliferation as well as migration of vascular SMCs in vitro. Moreover, we have demonstrated that this growth inhibitory effect in vitro is mediated through a TGF-β-induced complex that includes cAMP and cAMP response element binding protein, which in turn inhibits cyclin A (23). Alternatively, our most recent findings have reinforced that although in vitro, stimulation of vascular SMCs with TGF-β inhibits proliferation, in the presence of elevated levels of Smad3, and TGF-β stimulates cell proliferation (56). We have found in the rat carotid injury model that levels of Smad3 increase substantially within 3 days following arterial injury and peak around 14 days (56). Moreover, our studies of human plaque derived from atherectomy specimens likewise reveal elevated levels of Smad 3 in restenotic lesions (9). In the carotid balloon injury model, elevated levels of Smad3 are associated with an increase in vascular SMC proliferation and consequently intimal hyperplasia. Thus circumstances present at the time of arterial injury, specifically the increased expression of Smad-3, lead to TGF-β changing from an inhibitor to an agonist of SMC proliferation.

TGF-β has been found recently in a small number of cell types including myofibroblast and fibroblasts to enhance proliferation (1). Interestingly, the dichotomous role that we have observed of TGF-β enhancing and inhibiting proliferation in a single cell type (in this case SMCs) has not been previously described. In an attempt to better understand the mechanism through which TGF-β and Smad3 stimulate SMC proliferation, we previously studied the cyclin-dependent kinase inhibitor p27. We found that nuclear exportation of phosphorylated p27 may in part be responsible for TGF-β/Smad3-induced increase in vascular SMC proliferation (57). The purpose of the current investigation was to identify the upstream factors that mediate proliferative effect of TGF-β.

It is well established that the PI3K/Akt pathway plays an important role in promoting cell survival and proliferation in a variety of cell types (11, 16, 17, 27, 29). In endothelial cells, for example, PI3K/Akt regulates angiogenesis (40). Moreover, in vascular SMCs, the PI3K/Akt pathway has been implicated in both survival and proliferation (5, 12, 13, 34, 40, 46). Clinically, the PI3K/Akt pathway has been exploited in the development of coronary stents eluting rapamycin, an inhibitor of mammalian target of rapamycin (mTOR), the downstream target of PI3K/Akt (45, 52). The novelty of our study is the establishment of a relationship between TGF-β and Akt; the latter having been shown to be of substantial importance in the development of restenosis.

There is some prior evidence of a relationship between Smad proteins and p38. Smad2-dependent p38 activation leads to Rho activation in some cell types (30). Moreover, TGF-β leads to Smad4-dependent activation of p38 in pancreatic acinar cells (51). In a variant on this theme, it has been reported that Smad can activate p38 through an autocrine signaling mechanism facilitated by the secreted soluble factor GADD5b (55). In our model, an intermediate autocrine signaling pathway is highly unlikely in that neither inhibitors to transcription nor translation were able to block TGF-β-induced p38 phosphorylation.

A relationship between p38 and Akt has also been previously demonstrated. p38 has been shown to activate PI3K/Akt in fibroblasts, epithelial cells, and cardiomyocytes (19, 20, 41, 47, 61). Interestingly, there have been conflicting findings regarding the relationship between p38 and Akt. Li and colleagues (28) have found that in myofibroblasts, myostatin activates p38 MAPK and subsequently Akt to stimulate cell proliferation. Alternatively, Kulasekaran et al. (25) also found that p38 MAPK activates Akt in myofibroblasts but with the functional effect of apoptosis rather than cell proliferation. In addition to a relationship between p38 and Akt, a relationship between p38 and SMC proliferation has been previously demonstrated. In vascular SMCs, Zhan et al. (61) found that p38 activation leads to proliferation. Ohashi et al. (44) demonstrated that p-p38 is upregulated in rat carotid arteries following balloon injury. Moreover, similar to our in vivo findings with SB203580, Ohashi et al. found blockade of p38 with a chemical inhibitor FR167653 reduced proliferation and thus intimal hyperplasia. In contrast, Seay et al. (50) found that p38 activation inhibited proliferation of mouse vascular SMCs. These discrepant findings may reflect a species-specific difference in p38 signaling. Interestingly, the C57BL/6 strain of mice used by Seay et al. is known to be resistant to carotid intimal hyperplasia (our observations and Hui et al.) (21). In summary, our finding that p38 is upstream from Akt and proliferation is in line with the observations of others.

The findings of the current study fit nicely with our previous observations regarding TGF-β and the cyclin-dependent kinase inhibitor p27. In a recent study, we reported that nuclear exportation of the phosphorylated cyclin-dependent kinase inhibitor p27 is at least in part responsible for TGF-β/Smad3-induced vascular SMC proliferation (57). One of the known mechanisms through which Akt stimulates cellular proliferation is through phosphorylation and sequestration of p27 in the cytoplasm, preventing p27 from inhibiting nuclear cyclins and thus facilitating cell cycle progression (24, 31, 35, 37, 43). By combining our previous and current findings we postulate that a complete signaling pathway for the effect of TGF-β on proliferation might include TGF-β, Smad3, p38, Akt, and finally p27.

This is the first study to systematically delineate the signaling cascade through which TGF-β promotes vascular SMC proliferation. Although relationships between several of these signaling proteins have been previously delineated, this sequence of signaling events and the role of these proteins in TGF-β induced SMC proliferation is novel. We have also found evidence for this signaling pathway in an in vivo model of arterial injury. In conclusion, we demonstrate that TGF-β in the presence of elevated Smad3 activates p38 MAPK, then Akt resulting in SMC proliferation. The ultimate goal is to develop innovative therapies or combinations of therapies that can interrupt this pathway and block the negative effect of TGF-β on intimal hyperplasia.

GRANTS

This work was supported by National Heart, Lung, Blood Institute Grants R01-HL-068673 (to K. C. Kent and B. Liu) and T32-HL-07899 (to P. A. Suwanabol), a Society of University Surgeons-Ethicon Surgical Research fellowship award (to P. A. Suwanabol), and the Howard Hughes Medical Institute-Medical Research training fellowship (to S. M. Seedial).

ACKNOWLEDGMENTS

We thank Stephanie Morgan for intellectual input and Jun Ren for assistance with adenovirus preparation.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
View Abstract