PPARα Inhibits TGF-β–Induced β5 Integrin Transcription in Vascular Smooth Muscle Cells by Interacting With Smad4
Integrins play an important role in vascular smooth muscle cell (VSMC) migration, a crucial event in the development of restenosis and atherosclerosis. Transforming growth factor-β (TGF-β) is highly expressed in restenotic and atherosclerotic lesions, and known to induce integrin expression. Peroxisome proliferator-activated receptor α (PPARα), a member of the nuclear receptor superfamily, regulates gene expression in a variety of vascular cells. We investigated the effects of PPARα ligands on TGF-β–induced β3 and β5 integrin expression and potential interaction between PPARα and TGF-β signaling. PPARα ligands WY-14643 (100 μmol/L) and 5,8,11,14-eicosatetranoic acid (ETYA, 50 μmol/L) inhibited TGF-β–induced β5 integrin protein expression by 72±6.8% and 73±7.1%, respectively (both P<0.05). TGF-β–stimulated β3 integrin expression was not affected by PPARα ligands. Both PPARα ligands also suppressed TGF-β–induced β5 integrin mRNA levels. PPARα ligands inhibited TGF-β–inducible transcription of β5 integrin by an interaction with a TGF-β response element between nucleotides −63 and −44, which contains a Sp1/Sp3 transcription factor binding site. Nuclear complexes binding to the TGF-β response region contained Sp1/Sp3 and TGF-β–regulated Smad 2, 3, and 4 transcription factors. TGF-β–stimulated Sp1/Smad4 nuclear complex formation was inhibited by WY-14643 and ETYA with a parallel induction of PPARα/Smad4 interactions. However, in vitro pull-down experiments failed to demonstrate direct binding between PPARα/Smad4. Both PPARα ligands blocked PDGF-directed migration of TGF-β–pretreated VSMCs, a process mediated, in part, by β5 integrins. The present study demonstrates that PPARα activators inhibit TGF-β–induced β5 integrin transcription in VSMCs through a novel indirect interaction between ligand-activated PPARα and the TGF-β–regulated Smad4 transcription factors. The full text of this article is available at http://www.circresaha.org.
- peroxisome proliferator-activated receptor α
- transforming growth factor-β
- vascular smooth muscle cell
Migration and proliferation of vascular smooth muscle cells (VSMCs) are crucial events in the development of restenosis after vascular interventions and atherosclerosis.1,2⇓ Both require the interaction of cell surface integrin receptors with the surrounding extracellular matrix.2,3⇓ Integrins are a family of heterodimeric transmembrane glycoproteins consisting of noncovalently associated α and β chains.4 The integrin complexes αvβ3 and αvβ5 have been shown to be expressed on VSMCs and to regulate their migration through interactions with the extracellular matrix proteins vitronectin and osteopontin.5,6⇓ Furthermore, αvβ3 and αvβ5 are upregulated in atherosclerotic lesions and during neointima formation after vascular injury underscoring their important role in these processes.7,8⇓ Expression of integrins can be regulated by cytokines and growth factors present in atherosclerotic and neointimal lesions.9
Members of the transforming growth factor-β (TGF-β) family play a critical role in the regulation of cellular growth, differentiation, migration, and death.10 TGF-β isoforms are highly expressed in human atherosclerotic lesions where they exert both pro- and antiatherosclerotic actions on vascular cells.11,12⇓ TGF-β is also an important mediator of restenosis after vascular interventions.13 TGF-β mediates its effects through receptor serine/threonine kinases at the cell surface and their substrates, the Smad proteins.10 On receptor activation, Smad2 and 3 are phosphorylated, resulting in complex formation with Smad4. These complexes translocate into the nucleus where they activate transcription of target genes.10 TGF-β induces integrin expression and extracellular matrix production in a variety of cells including VSMCs.14,15⇓ More specifically, TGF-β upregulates the VSMC expression of β3 integrin subunits and increases cell migration.16
Members of the peroxisome proliferator-activated receptor (PPAR) family are ligand-activated nuclear hormone receptors that function as transcriptional regulators of genes linked to lipid metabolism and glucose homeostasis.17 Three different isoforms have been identified: PPARα, PPARγ, and PPARδ.
PPARα is expressed in vascular cells including endothelial cells, monocyte/macrophages, and VSMCs.18 PPARα can be activated by hypolipidemic fibrates, eicosanoids, or polyunsaturated fatty acids.19 Activation of PPARα has been shown to inhibit proinflammatory processes in these cells, therefore, playing an important role in the development of atherosclerosis and restenosis after vascular injury.18 In VSMCs, PPARα ligands inhibited the expression of inflammatory genes such as interleukin 6,6-keto prostaglandin F1α and cyclooxygenase-2.20 Antiinflammatory actions of PPARα in VSMCs have been intensively studied.21 In contrast, the effects of PPARα ligands on other VSMC functions, including the regulation of genes involved in migration, are mainly unexplored.
PPARs are capable of both positive and negative regulation of gene expression in response to ligand binding.17 PPARs positively regulate gene expression by binding to PPAR response elements in target genes as a heterodimer with retinoid X receptors (RXRs) followed by the recruitment of coactivators and consequent transcription of the target genes.17 Negative regulation of gene expression by PPARs can be mediated by transrepression, a mechanism involving either competition for limiting amounts of essential coactivators utilized by many other transcription factors, or through direct physical interactions between PPARs and specific transcription factors.22,23⇓ Direct interactions with NF-κB or AP-1 have been identified as a mechanism for inhibition of gene transcription by PPARα.22
Given the importance of integrins in VSMC functions involved in atherosclerosis and restenosis, we investigated the effects of PPARα ligands on TGF-β–induced β3 and β5 integrin expression in VSMCs. Furthermore, we identified potential novel interactions of ligand-activated PPARα with the TGF-β–regulated transcription factor, Smad4.
Materials and Methods
Platelet-derived growth factor (PDGF), dimethylsulfoxide (DMSO), vitronectin, glutamine, and antibiotics were purchased from Sigma. Dulbeccos’s modified Eagle’s medium (DMEM), Opti-MEM I medium, Trizol Reagent, and LipofectAMINE 2000 were from Life Technologies. Fetal bovine serum (FBS) was purchased from Irvine Scientific. Hybond ECL nitrocellulose membrane, horseradish peroxidase-linked anti-rabbit and anti-mouse antibody, ECL Western blotting detection reagents, random prime labeling system (Redi prime II), rapid hybridization buffer, and nylon Hybond-N+ membranes were from Amersham Life Sciences. Dual-luciferase reporter assay and gel shift assay systems were from Promega. Cell fixation and staining was performed using the Quik-Diff stain set from DADE. TGF-β1 was from R&D Systems. WY-14643 and 5,8,11,14-Eicosatetraynoic acid (ETYA) were from BIOMOL. Antibodies were purchased from the following providers: β5 integrin (sc-5401), Sp1 (sc-420X, sc-59G), Sp3 (sc-644X), Smad2/3 (sc-6032X), Smad4 (sc-7154X), and PPARα (sc-9000X) were from Santa Cruz Biotechnology; β3 integrin (22440D) was from BD Pharmingen; β5 integrin (AB 1926) was from Chemicon International Inc; and normal rabbit IgG, normal goat IgG, and normal mouse IgG were from Zymed.
Rat aortic smooth muscle cells (RASMCs) were prepared from thoracic aortae of 2 to 3 month old Harlan Sprague-Dawley rats by using the explant technique.24 RASMCs were cultured as previously described.25 WY-14643, ETYA, or vehicle DMSO were added to cells 30 minutes before the treatment with TGF-β. For all data shown, each individual experiment represented in the n value was performed using independent preparations of RASMCs.
Western Blot Analysis and Immunoprecipitation
RASMC were harvested after stimulation with TGF-β in the absence or presence of PPARα ligands. Western immunoblotting and immunoprecipitation experiments were performed as previously described.26 Nuclear Extracts were prepared as described.27
RNA Isolation and Northern Blot Analysis
After stimulation with TGF-β±PPARα ligands, total RNA was isolated using Trizol reagent. RNA (20 μg) was electrophoresed through 1% agarose gels containing formaldehyde, transferred to charged nylon membranes, and hybridized with cDNA labeled with [α-32P] dCTP by random labeling. The hybridization signals of the specific mRNA of β5 integrin were normalized to those of CHOB, a constitutively expressed gene initially isolated as Chinese hamster ovary clone B that encodes a ribosomal protein, to correct for differences in loading or transfer. The β5 integrin cDNA was kindly provided by R.S. Ross (University of California, Los Angeles, Calif). All experiments were repeated at least 3 times with a different cell preparation. Band intensity was analyzed by densitometry.
β5 Promoter Luciferase Reporter Constructs
Generation of progressive 5′- to 3′-deletion constructs of murine β5 integrin promoters has been described previously, and were kindly provided by S.L. Cheng and F.P. Ross (Washington University, St Louis, Mo).28 A 1-kb fragment was isolated from the AccI-digested products of the β5 genomic 9-kb fragment. Deletion constructs of this 1-kb fragment were obtained by using the Exonuclease III/Mung Bean Nuclease kit from Stratagene. All the fragments were subcloned into pGL3-basic vector containing the luciferase reporter gene. The promoter constructs used in this study spanned from −875, −63, and −43 to +110 from the transcription start site.
Transient Transfection and Luciferase Assay
RASMCs were transfected using LipofectAMINE 2000 for promoter activity analysis. Cells were grown to 70% to 80% confluence in 6-well plates and placed in Opti-MEM I medium. After 16 hours incubation, cells were transfected with 0.1 μg promoter construct to be tested and 5 ng pRL-CMV, a renilla luciferase control reporter vector. After 24 hours starvation in Opti-MEM I medium, cells were treated with WY-14643 and ETYA before stimulation with TGF-β. After 24 hours, cells were lysed in reporter lysis buffer followed by measurement of luciferase activity using a Dual-Luciferase Reporter Assay System. Firefly luciferase activities from the β5 reporter constructs were normalized with renilla luciferase activities from pRL-CMV. All experiments were repeated at least 3 times with a different cell preparation.
Electrophoretic Mobility Shift Assay
For electrophoretic mobility shift assays (EMSA), radioactive double-stranded oligonucleotide −66/−42, labeled with T4 polynucleotide kinase and [γ-32P] ATP, was incubated with nuclear extracts (2 μg) for 20 minutes in gel shift binding buffer (20% glycerol, 5 mmol/L MgCl2, 2.5 mmol/L EDTA, 2.5 mmol/L DTT, 250 mmol/L NaCl, 50 mmol/L Tris-HCl [pH 7.5], and 0.25 mg/mL poly [dI-dC]). Assays were terminated by addition of 1 μL 10× gel loading buffer (250 mmol/L Tris-HCl [pH 7.5], 0.2% bromophenol blue, and 40% glycerol) and analyzed by electrophoresis using a 4% nondenaturing acrylamide gel (40:1 acrylamide: bisacrylamide) in 0.5× TBE-buffer. Gels were dried, and autoradiography was performed. For competitive oligonucleotide or immunodepletion assays, 100-fold unlabeled double-stranded oligonucleotides and 2 μg antibodies were incubated with nuclear extracts for 30 minutes before the addition of radiolabeled probe. For supershift-experiments, 2 μg of antibodies were added after incubation of the radiolabeled probe with nuclear proteins.
GST (Glutathione S-Transferase) Pull-Down Assay
The pSG5-hPPARα plasmid was described previously.20 The pSG5-hPPARγ1 plasmid was obtained from A. Elbrecht (Merck Research Laboratories). Expression vectors were in vitro transcribed and their transcripts in vitro translated in the presence of [35S] methionine to label their recombinant protein products. GST-Smad3 and GST-Smad4 expression vectors were as previously published, and were produced in E. coli and purified using glutathione-Sepharose beads according to the manufacturer’s protocol (Amersham Life Sciences).29 Approximately 10 μg of purified GST-fusion protein were mixed with 15 μL of [35S] methionine-labeled PPARα or PPARγ at 4°C in 500 μL of binding buffer (20 mmol/L Tris-HCl [pH 7.4], 100 mmol/L KCl, 0.7 mmol/L EDTA, 0.05% NP-40, 0.5 mmol/L PMSF, 10 μg/mL leupeptin and pepstatin, 2 μg/mL aprotinin) for 2 hours in the absence or presence of WY-14643 or rosiglitazone. Samples were centrifuged and washed 4 times to remove unbound protein. The washed beads were resuspended in 50 μL of SDS-PAGE loading dye and boiled to elute bound protein. Bound protein samples were size-fractionated by SDS-PAGE, stained with Coomassie Brilliant Blue R250, and analyzed by autoradiography.
RASMC migration was studied as previously described.30 Transwell chambers were coated with vitronectin (10 μg/mL). Cells were pretreated with TGF-β for 12 hours. PPARα ligands were added 30 minutes before stimulation with TGF-β. For antibody experiments, RASMCs were incubated for 30 minutes with the indicated anti-integrin antibodies before plating in migration chamber.
Analysis of variance was performed for statistical analysis, and values of P<0.05 were considered to be statistically significant. Data are expressed as mean±SEM.
PPARα Ligands Inhibit TGF-β–Induced β5 Integrin Expression
Treatment of quiescent RASMCs with TGF-β induced β3 and β5 integrin protein expression by 2.7±0.7-fold and 3.3±0.5-fold, respectively, reaching a maximum at 5 ng/mL after 8 hours (P<0.01 versus untreated RASMCs) (Figures 1A through 1C). Neither higher concentrations of TGF-β or longer periods of growth factor treatment further increased β3 and β5 integrin levels.
Pretreatment with the PPARα ligands WY-14643 (50 to 250 μmol/L) and ETYA (10 to 100 μmol/L) suppressed TGF-β–mediated induction of β5 integrin expression reaching maximal inhibition at 100 μmol/L for WY-14643 (72±6.8% inhibition versus TGF-β+DMSO; P<0.05) and at 50 μmol/L for ETYA (73±7.1% inhibition versus TGF-β+DMSO; P<0.05) (Figures 1A and 1B). Similar concentrations of these PPARα-ligands have previously been used, without toxicity, in other studies, demonstrating inhibitory biological effects of PPARα-ligands.22,31,32⇓⇓ The PPARγ-ligands rosiglitazone (10 μmol/L) and troglitazone (10 μmol/L) had no effect on TGF-β–induced β5 integrin protein expression (data not shown). TGF-β–induced β3 integrin expression was not affected by either PPARα ligand (Figure 1C). The binding partner of β3/β5 integrins, namely αv integrin, was neither stimulated by TGF-β nor regulated by the PPARα-ligand WY-14643 (100 μmol/L) (data not shown).
TGF-β (5 ng/mL) also induced β5 integrin mRNA by 2.9±0.9-fold reaching a maximum after 6 hours (P<0.01 versus untreated RASMC) (Figure 1D). Both PPARα ligands blocked TGF-β–induced β5 integrin mRNA expression with a maximal inhibition of 78.5±6.7% for WY-14643 (100 μmol/L; P<0.05 versus TGF-β+DMSO) and 84.6±7.3% for ETYA (50 μmol/L; P<0.05 versus TGF-β+DMSO) (Figure 1D).
PPARα Ligands Inhibit TGF-β–Induced β5 Integrin Transcription
To determine whether the regulation of TGF-β–induced β5 integrin mRNA by PPARα ligands resulted from an inhibition of gene transcription, RASMCs were transfected with a β5 integrin promoter (−875 to +110) region/luciferase reporter gene construct. TGF-β (5 ng/mL) activated transcription from this β5-promoter region by 2.8±0.1-fold (P<0.05 versus untreated RASMCs) (Figure 2A). TGF-β induction of this β5 integrin promoter activity was potently reduced by both PPARα ligands (WY 100 μmol/L, 1.2±0.4-fold activation versus untreated RASMCs; ETYA 50 μmol/L, 1±0.3-fold activation versus untreated RASMC; both P<0.05 versus TGF-β+DMSO) (Figure 2A). These data suggest that PPARα ligands exert their inhibitory effects on β5 integrin mRNA expression through a transcriptional mechanism.
Recently, a TGF-β response element within the β5 integrin promoter (β5 TβRE) was identified between nucleotides −63 and −44.33 In order to locate the region essential for the inhibitory actions of PPARα ligands on TGF-β–induced β5 integrin promoter activity, RASMCs were transfected with β5 integrin promoter deletion constructs spanning from −63 to +110 and from −43 to +110. TGF-β (5 ng/mL) induced β5 integrin promoter activity by 2.3±0.14-fold when the promoter construct was deleted to −63 (P<0.05 versus untreated RASMCs), whereas further deletion of the construct to −43 resulted in a complete loss of the inductive effects of TGF-β, corroborating a major β5 TβRE between nucleotides −63 and −44 (Figure 2A). These data suggest that PPARα ligands inhibit TGF-β–inducible transcription of β5 integrin by a potential interaction with the β5 TβRE between nucleotides −63 to −44.
Transient transfection of a pGL3-basic vector in RASMC revealed no regulation of luciferase activity compared with the β5 integrin promoter (−875 to +110) construct (Figure 2B).
PPARα Ligands Inhibit Binding of Nuclear Factors to the TGF-β Response Element of the β5 Integrin Promoter
To confirm that the β5 TβRE between nucleotides −63 and −44 was indeed the target of PPARα ligands, EMSAs were performed with double-stranded radiolabeled (β5 TβRE) oligonucleotides corresponding to the −66/−42 region of the β5 integrin promoter. Treatment with TGF-β (5 ng/mL) increased complex formation between nuclear proteins and the −66/−42 oligonucleotide by 1.8±0.1-fold reaching a maximum after 4-hour treatment (P<0.05 versus untreated RASMCs) (Figure 3A). Pretreatment with WY-14643 (100 μmol/L) and ETYA (50 μmol/L) in TGF-β–treated cells decreased complex formation at the β5 TβRE region to basal levels (both P<0.05 versus TGF-β+DMSO) (Figure 3A). These data demonstrate that PPARα ligands prevent the binding of transcription factors to the β5 TβRE.
TGF-β Response Element of the β5 Integrin Promoter Binds Sp1/Sp3/Smad Proteins
Previous studies revealed the presence of a Sp1/Sp3 transcription factor binding site between nucleotides −53 and −48 in the β5 TβRE.33 To identify transcription factors binding to the β5 TβRE region and to confirm the presence of a Sp1/Sp3 site within this region, a series of additional EMSA experiments was performed with labeled −66/−42 oligonucleotides. Consistent with the presence of a Sp1/Sp3 site, pretreatment of nuclear extracts with 100-fold excess unlabeled consensus Sp1 oligonucleotide decreased complex formation of nuclear extracts with the β5 TβRE probe (Figure 3B). Incubation with 100-fold consensus AP-2 oligonucleotide did not affect binding activity (Figure 3B). Treatment of nuclear extracts with anti-Sp1/Sp3 antibodies after addition of the radioactive probe induced a new, supershifted band, indicating the presence of these transcription factors in the complex binding to β5 TβRE (Figure 3C). Because Smad proteins are important mediators of TGF-β signaling, nuclear extracts were incubated with anti-Smad 2, 3, and 4 antibodies after addition of labeled β5 TβRE oligonucleotide. Incubation with Smad 2, 3, and 4 antibodies also induced an additional supershifted band, demonstrating the presence of Smad transcription factors in nuclear complexes formed at the −66/−42 region (Figure 3C). Addition of anti-PPARα antibody showed no effect on the binding reaction, which indicates that the complexes does not contain PPARα (Figure 3C).
Ligand-Activated PPARα Inhibits Sp1/Smad4 Complex Formation by Interacting With Smad4
Immunoprecipitation experiments with nuclear extracts were performed to examine further potential mechanisms of PPARα’s inhibitory actions on Sp1/Sp3: Smad 2, 3, and 4 interactions at the β5 TβRE.
Given a central role of Smad4 in TGF-β–mediated β5 integrin regulation, we focused our studies on this transcription factor.33 Treatment with TGF-β (5 ng/mL) for 4 hours increased binding of nuclear Smad4 to nuclear Sp1, whereas binding of Smad4 to Sp3 was not affected (Figure 4A and 4B). Immunoprecipitation with nonspecific anti-goat and anti-rabbit IgGs did not yield any Smad4 bands. Sp1/Sp3 partner proteins for Smad4 are efficiently immunoprecipitated by their corresponding antibodies (Figures 4A and 4B). Pretreatment with WY-14643 (100 μmol/L) and ETYA (50 μmol/L) inhibited TGF-β–induced Smad4/Sp1 binding in nuclear extracts, suggesting a potential interaction between ligand-activated PPARα and Smad4/Sp1 complexes (Figure 4A). In contrast, Smad4/Sp3 complexes were not affected by either of the PPARα-ligands (Figure 4B). Cytosolic and nuclear protein levels of Sp1/Sp3 or Smad4 were not affected by PPARα-ligands (data not shown).
Because direct interactions between PPARα and transcription factors have been described recently, we hypothesized an interaction between ligand-activated PPARα and either Smad4 or Sp1. Treatment with TGF-β (5 ng/mL) alone did not induce binding of PPARα to Smad4 (Figure 4C). Pretreatment of TGF-β–stimulated cells with WY-14643 (100 μmol/L) or ETYA (50 μmol/L) led to enhanced formation of complexes containing PPARα and Smad4 (Figure 4C), whereas PPARα/Sp1 binding was not affected (data not shown). To further clarify whether PPARα directly interacts with Smad4, GST-pull-down experiments with in vitro translated PPARα protein and GST-Smad4 fusion proteins were performed. A direct interaction between PPARα and Smad4 could not be detected with or without the PPARα ligand WY-14643 (250 μmol/L) (Figure 5A). As a positive control we used in vitro translated PPARγ, which is known to directly interact with Smad3 (Figure 5B).34 This interaction is enhanced by treatment with the PPARγ ligand rosiglitazone (10 μmol/L) (Figure 5B).
In combination, these data demonstrate that ligand-activated PPARα blocks TGF-β–induced Smad4/Sp1 interactions. Although a direct PPARα/Smad4 association could not be detected, these two proteins may be a part of a multiprotein complex that is the target for the observed inhibitory effects of PPARα ligands on TGF-β–induced binding of Sp1/Sp3/Smad2,3,4 complexes to the β5 TβRE.
PPARα Ligands Inhibit PDGF-Directed Migration of TGF-β–Pretreated RASMCs
To elucidate whether the inhibitory effects of PPARα ligands on TGF-β–induced β5 integrin expression translated into an inhibition of RASMC migration, we studied the migratory response of TGF-β–treated and nontreated cells toward PDGF in the absence or presence of PPARα ligands.
Pretreatment of RASMCs with TGF-β for 12 hours increased PDGF-directed migration on vitronectin (10 μg/mL) by 2.7±1.2-fold at 5 ng/mL TGF-β compared with untreated RASMCs (P<0.05 versus PDGF alone) (Figure 6A). Incubation of RASMCs with WY-14643 (100 μmol/L) and ETYA (50 μmol/L) attenuated PDGF-directed migration without reaching statistical significance (WY-14643, 24.3±5.9% inhibition; ETYA, 25±13% inhibition versus PDGF alone) (Figure 6B). The inhibitory effects of PPARα ligands WY-14643 and ETYA became more pronounced when RASMCs were pretreated with TGF-β (5 ng/mL) for 12 hours. Pretreatment with PPARα ligands blocked migration of these cells, reaching maximal inhibition at 100 μmol/L WY-14643 and at 50 μmol/L ETYA (WY-14643 100 μmol/L, 65.5±1.1% inhibition; ETYA 50 μmol/L, 71±1.8% inhibition; P<0.05 versus TGF-β–treated RASMC+PDGF/DMSO) (Figure 6B). To study the involvement of β3 and β5 integrins in the antimigratory actions of PPARα ligands, RASMCs were treated with the PPARα ligands followed by an incubation with either anti–β3 integrin antibody or anti–β5 integrin antibody, and migration experiments were performed. Both PPARα ligands potently blocked PDGF-directed migration under conditions of β5 integrin–dependent migration (RASMCs incubated with anti-β3 antibody) (WY-14643 100 μmol/L, 71.9±1% inhibition; ETYA 50 μmol/L, 77.5±1.7% inhibition; both P<0.01) (Figure 6C). However, PPARα ligands had no significant inhibitory effect under conditions of β3 integrin–dependent migration (RASMCs incubated with anti-β5 antibody), implicating that PPARα-ligands antimigratory actions are predominantly mediated through β5 integrins (Figure 6C).
Together these data suggest that the inhibition of TGF-β–treated RASMC migration by PPARα ligands, may result, at least in part, from their effect to inhibit TGF-β–induced β5 integrin expression.
The present study demonstrates that PPARα activators inhibit TGF-β–induced transcription of β5 integrin in RASMCs. We further show that a PPARα/Smad4 association, which is likely mediated by additional nuclear proteins, may prevent formation of Sp1/Sp3/Smad2,3,4 complexes on a TGF-β response element and constitutes a potential mechanism for PPARα ligand-mediated repression of β5 integrin promoter activity.
PPARα is expressed at substantial levels in VSMCs where it has been shown to function as a negative regulator of proinflammatory processes.20 Its role in regulating VSMC migration, however, is largely unexplored. Integrins are important modulators of VSMC migration and are known to be regulated by TGF-β.2,3⇓ We show that two different PPARα ligands inhibit TGF-β–induced β5 integrin expression in RASMCs. To our knowledge, this is the first report describing the regulation of integrin expression by ligands of the PPAR class. Other nuclear hormone receptors, including the glucocorticoid receptor and the vitamin D receptor, have also been shown to regulate integrin expression in a variety of cells.35,36⇓
Recently, a TGF-β response element has been identified between nucleotides −63 and −44 in the β5 integrin promoter in a murine osteoblastic cell line.33 Transfection experiments in RASMCs revealed that the inhibitory effects of PPARα ligands on TGF-β–induced β5 integrin expression are mediated by blocking transcriptional activity of the TGF-β response element (−63/−44) in the β5 integrin promoter. Several molecular mechanisms can be invoked to explain the inhibitory actions of PPARα on gene transcription. Inhibition of transcription by PPARα might occur by competing for and sequestering of limiting transcriptional coactivators, such as CBP or p300.17 Delerive and colleagues22 demonstrated that direct physical interactions between PPARα and the p65 NF-κB subunit or c-Jun account, at least in part, for PPARα-mediated repression of NF-κB/AP-1–regulated gene transcription. Our data suggest a more specific effect of ligand-activated PPARα on the TGF-β–regulated transcription factor Smad4 as the mechanism for repressing TGF-β–stimulated β5 integrin transcription.
Previous reports have shown that Smad4 is required for TGF-β–induced β5 integrin transcription.33 On TGF-β stimulation Smad4 binds to Smad 2 and 3, and translocates to the nucleus where it binds to Sp1 transcription factors resulting in increased binding of Sp1/Sp3/Smad2,3,4 complexes to a Sp1/Sp3 binding site within the TGF-β response region of the β5 integrin promoter.33 By gel shift experiments, we demonstrated that nuclear protein complexes binding to the TGF-β response region in the β5 integrin promoter contain Sp1/Sp3 and Smad 2, 3, and 4 transcription factors. More importantly, we observed that PPARα ligands potently inhibited formation of Sp1/Smad4 complexes. Recently, Fu and colleagues34 demonstrated that PPARγ inhibits TGF-β–induced gene expression by directly interacting with Smad3, suggesting that physical interplay between PPAR family members and Smad transcription factors may constitute a mechanism for PPAR-mediated inhibition of TGF-β–regulated gene expression. It appears that Smad3 does not play a major role in regulating β5 integrin promoter activity, because its binding partner, ligand-activated PPARγ, did not affect β5 integrin protein expression. In contrast, based on the central role for Smad4 in β5 integrin promoter regulation, sequestering of Smad4 by PPARα may prevent the formation of Sp1/Sp3/Smad2,3,4 complexes at −66/−42 of the β5 integrin promoter. Although we could not detect a direct protein-protein interaction between PPARα and Smad4 in pull-down experiments, PPARα ligands still induced nuclear PPARα/Smad4 association in coimmunoprecipitation experiments. This data are consistent with a study by Pouponnot and colleagues,37 demonstrating an interaction of Smad4 with the coactivator p300 in coimmunoprecipitation experiments, whereas a direct protein-protein interaction was not detected in experiments with a GST-Smad4 fusion protein. There are two potential explanations for these data. Smad4 may have a weak affinity to its binding partners, which mitigates the detection of an interaction in in vitro assays. Alternatively, Smad4 interactions with its binding partners may be mediated by association with additional nuclear proteins present within a multiprotein complex. Additional studies will be required to identify other proteins present in a PPARα/Smad4 multisubunit complex.
Interestingly, treatment of RASMCs with PPARα ligands did not affect TGF-β–induced β3 integrin expression. Little is known about β3 integrin regulation by TGF-β. β3 integrin mRNA expression is upregulated by TGF-β.16 Although the β3 integrin promoter contains binding sites for a number of transcription factors (Sp1, AP-1, STAT, and NF-κB) known to be regulated by PPARα, the role of those elements in TGF-β–regulated transcription from the β3 integrin promoter is still unknown.38 Our findings predict, that Smad4 proteins are not involved in TGF-β–stimulated β3 integrin transcription. Future studies are required to more fully elucidate transcriptional mechanisms involved in TGF-β–mediated β3 integrin expression.
PPARα-mediated interference with Smad4 activity may be functionally significant. PPARα-ligands inhibited migration of TGF-β–treated RASMCs, at least in part, by their inhibitory effects on TGF-β–induced β5 integrin expression. These data demonstrate that the inhibitory actions of PPARα on β5 integrin expression translates into the blockade of an important integrin-mediated VSMC behavior. PPARα ligands did not affect PDGF-directed migration of untreated RASMCs. This finding suggests that PPARα ligands do not inhibit VSMC migration through a generalized effect on cell movement or chemotaxis. Instead, integrin upregulation by TGF-β was shown to be the specific pathway targeted by PPARα. Antimigratory activity of PPARα ligands, therefore, may be limited to pathophysiological states characterized by elevated levels of TGF-β. Marx and colleagues31 and our group30 showed that activation of PPARγ blocks VSMC migration, which was associated with a blockade of matrix metalloproteinase (MMP)-9 activity. Interestingly, Marx and colleagues did not observe any effects of PPARα ligands on VSMC MMP-activity.31 These data suggest that PPARγ inhibits VSMC migration by targeting a matrix-degrading, invasive component of the migratory process, whereas PPARα ligands are affecting integrin-mediated cell movement.
Beside their important role in cell migration, integrins are also involved in the regulation of cell proliferation.39 The migratory and proliferative response of VSMCs after vascular injury are major contributors to the development of intimal hyperplasia followed by restenosis of the injured vessel.1,2⇓ Blocking of integrin function with anti-integrin antibodies has been shown to reduce postinjury restenosis in animals and decreased ischemic long-term complications after vascular interventions in humans.40,41⇓ In addition, blocking TGF-β expression or signaling by ribozyme oligonucleotides against TGF-β, soluble TGF-β type II receptor, or anti–TGF-β antibodies has been shown to attenuate neointima formation after vascular injury.13,42,43⇓⇓ Decreased intimal lesion formation in these studies resulted from an inhibition of TGF-β–mediated VSMC migration and/or proliferation as well as matrix accumulation. Inhibition of TGF-β–mediated β5 integrin expression by activation of PPARα may, therefore, have beneficial effects on the development of postinjury intimal hyperplasia.
This study was supported by an NIH Grant to W.A.H. (HL-58328-03). U.K. and D.B. are supported by a research fellowship by the Gonda (Goldschmied) Diabetes Center, University of California, Los Angeles. S.W. is supported by a fellowship by the Mary K. Iacocca Foundation. D.B. is supported by a grant from MSD Sharp&Dohme.
Original received July 12, 2001; resubmission received October 16, 2002; accepted October 29, 2002.
- ↵Liaw L, Skinner MP, Raines EW, Ross R, Cheresh DA, Schwartz SM, Giachelli CM. The adhesive and migratory effects of osteopontin are mediated via distinct cell surface integrins: role of αvβ3 in smooth muscle cell migration to osteopontin in vitro. J Clin Invest. 1995; 95: 713–724.
- ↵Smith JW, Vestal DJ, Irwin SV, Burke TA, Cheresh DA. Purification and functional characterization of integrin αvβ5: an adhesion receptor for vitronectin. J Biol Chem. 1990; 265: 11008–11013.
- ↵Dufourcq P, Louis H, Moreau C, Daret D, Boisseau MR, Lamaziere JM, Bonnet J. Vitronectin expression and interaction with receptors in smooth muscle cells from human atheromatous plaque. Arterioscler Thromb Vasc Biol. 1998; 18: 168–176.
- ↵Kim LT, Yamada KM. The regulation of expression of integrin receptors. Proc Soc Exp Biol Med. 1997; 214: 123–131.
- ↵Bobik A, Agrotis A, Kanellakis P, Dilley R, Krushinsky A, Smirnov V, Tararak E, Condron M, Kostolias G. Distinct patterns of transforming growth factor-β isoform and receptor expression in human atherosclerotic lesions: colocalization implicates TGF-β in fibrofatty lesion development. Circulation. 1999; 99: 2883–2891.
- ↵Topper JN. Transforming growth factor-β (TGF-β) and vascular disease: CARP as a putative TGF-β target gene in the vessel wall. Circ Res. 2001; 88: 5–6.
- ↵Yamamoto K, Morishita R, Tomita N, Shimozato T, Nakagami H, Kikuchi A, Aoki M, Higaki J, Kaneda Y, Ogihara T. Ribozyme oligonucleotides against transforming growth factor-β inhibited neointimal formation after vascular injury in rat model: potential application of ribozyme strategy to treat cardiovascular disease. Circulation. 2000; 102: 1308–1314.
- ↵Ignotz RA, Heino J, Massague J. Regulation of cell adhesion receptors by transforming growth factor-β: regulation of vitronectin receptor and LFA-1. J Biol Chem. 1989; 264: 389–392.
- ↵Brown SL, Lundgren CH, Nordt T, Fujii S. Stimulation of migration of human aortic smooth muscle cells by vitronectin: implications for atherosclerosis. Cardiovasc Res. 1994; 28: 1815–1820.
- ↵Forman BM, Chen J, Evans RM. Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are ligands for peroxisome proliferator-activated receptors α and δ. Proc Natl Acad Sci U S A. 1997; 94: 4312–4317.
- ↵Delerive P, Fruchart JC, Staels B. Eurosterone Meeting: peroxisome proliferator-activated receptors in inflammation control. J Endocrinol. 2001; 169: 453–459.
- ↵Delerive P, De Bosscher K, Besnard S, Vanden Berghe W, Peters JM, Gonzalez FJ, Fruchart JC, Tedgui A, Haegeman G, Staels B. Peroxisome proliferator-activated receptor α negatively regulates the vascular inflammatory gene response by negative cross-talk with transcription factors NF-κB and AP-1. J Biol Chem. 1999; 274: 32048–32054.
- ↵Li M, Pascual G, Glass CK. Peroxisome proliferator-activated receptor γ-dependent repression of the inducible nitric oxide synthase gene. Mol Cell Biol. 2000; 20: 4699–4707.
- ↵Chamley-Campbell J, Campbell GR, Ross R. The smooth muscle cell in culture. Physiol Rev. 1979; 59: 1–61.
- ↵Wakino S, Kintscher U, Kim S, Yin F, Hsueh WA, Law RE. Peroxisome proliferator-activated receptor γ ligands inhibit retinoblastoma phosphorylation and G1→S transition in vascular smooth muscle cells. J Biol Chem. 2000; 275: 22435–22441.
- ↵Goetze S, Kintscher U, Kawano H, Kawano Y, Wakino S, Fleck E, Hsueh WA, Law RE. Tumor necrosis factor α inhibits insulin-induced mitogenic signaling in vascular smooth muscle cells. J Biol Chem. 2000; 275: 18279–18283.
- ↵Dignam JD, Lebovitz RM, Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 1983; 11: 1475–1489.
- ↵Feng X, Teitelbaum SL, Quiroz ME, Towler DA, Ross FP. Cloning of the murine β5 integrin subunit promoter: identification of a novel sequence mediating granulocyte-macrophage colony-stimulating factor–dependent repression of β5 integrin gene transcription. J Biol Chem. 1999; 274: 1366–1374.
- ↵Pardali K, Kurisaki A, Moren A, ten Dijke P, Kardassis D, Moustakas A. Role of Smad proteins and transcription factor Sp1 in p21(Waf1/Cip1) regulation by transforming growth factor-β. J Biol Chem. 2000; 275: 29244–29256.
- ↵Marx N, Schonbeck U, Lazar MA, Libby P, Plutzky J. Peroxisome proliferator-activated receptor gamma activators inhibit gene expression and migration in human vascular smooth muscle cells. Circ Res. 1998; 83: 1097–1103.
- ↵Marx N, Mackman N, Schonbeck U, Yilmaz N, Hombach VV, Libby P, Plutzky J. PPARα activators inhibit tissue factor expression, and activity in human monocytes. Circulation. 2001; 103: 213–219.
- ↵Lai CF, Feng X, Nishimura R, Teitelbaum SL, Avioli LV, Ross FP, Cheng SL. Transforming growth factor-β up-regulates the β5 integrin subunit expression via Sp1 and Smad signaling. J Biol Chem. 2000; 275: 36400–36406.
- ↵Fu M, Zhang J, Zhu X, Myles DE, Willson TM, Liu X, Chen YE. Peroxisome proliferator-activated receptor gamma inhibits transforming growth factor β–induced connective tissue growth factor expression in human aortic smooth muscle cells by interfering with Smad3. J Biol Chem. 2001; 276: 45888–45894.
- ↵Cao X, Ross FP, Zhang L, MacDonald PN, Chappel J, Teitelbaum SL. Cloning of the promoter for the avian integrin β3 subunit gene and its regulation by 1,25-dihydroxyvitamin D3. J Biol Chem. 1993; 268: 27371–27380.
- ↵Filep JG, Delalandre A, Payette Y, Foldes-Filep E. Glucocorticoid receptor regulates expression of L-selectin and CD11/CD18 on human neutrophils. Circulation. 1997; 96: 295–301.
- ↵Pouponnot C, Jayaraman L, Massague J. Physical and functional interaction of SMADs and p300/CBP. J Biol Chem. 1998; 273: 22865–22868.
- ↵Villa-Garcia M, Li L, Riely G, Bray PF. Isolation and characterization of a TATA-less promoter for the human β3 integrin gene. Blood. 1994; 83: 668–676.
- ↵Slepian MJ, Massia SP, Dehdashtii B, Fritz A, Whitesell L. β3-Integrins rather than β1-integrins dominate integrin-matrix interactions involved in postinjury smooth muscle cell migration. Circulation. 1998; 97: 1818–1827.
- ↵Smith JD, Bryant SR, Couper LL, Vary CP, Gotwals PJ, Koteliansky VE, Lindner V. Soluble transforming growth factor-β type II receptor inhibits negative remodeling, fibroblast transdifferentiation, and intimal lesion formation but not endothelial growth. Circ Res. 1999; 84: 1212–1222.