Basic Fibroblast Growth Factor Antagonizes Transforming Growth Factor-?1–Induced Smooth Muscle Gene Expression Through Extracellular Signal–
http://www.100md.com
《动脉硬化血栓血管生物学》
From the Department of Medicine and Biological Science, Gunma University Graduate School of Medicine, Gunma, Japan.
ABSTRACT
Objective— Transforming growth factor-?1 (TGF?1) and fibroblast growth factor (FGF) families play a pivotal role during vascular development and in the pathogenesis of vascular disease. However, the interaction of intracellular signaling evoked by each of these growth factors is not well understood. The present study was undertaken to examine the molecular mechanisms that mediate the effects of TGF?1 and basic FGF (bFGF) on smooth muscle cell (SMC) gene expression.
Methods and Results— TGF?1 induction of SMC gene expression, including smooth muscle protein 22- (SM22) and smooth muscle -actin, was examined in the pluripotent 10T1/2 cells. Marked increase in these mRNA levels by TGF?1 was inhibited by c-Src-tyrosine kinase inhibitors and protein synthesis inhibitor cycloheximide. Functional studies with deletion and site-directed mutation analysis of the SM22 promoter demonstrated that TGF?1 activated the SM22 promoter through a CC(A/T-rich)6GG (CArG) box, which serves as a serum response factor (SRF)–binding site. TGF?1 increased SRF expression through an increase in transcription of the SRF gene. In the presence of bFGF, TGF?1 induction of SMC marker gene expression was significantly attenuated. Transient transfection assays showed that bFGF significantly suppressed induction of the SM22 promoter–driven luciferase activity by TGF?1, whereas bFGF had no effects on the TGF?1-mediated increase in SRF expression and SRF:DNA binding activity. Mitogen-activated protein kinase kinase-1 (MEK1) inhibitor PD98059 abrogated the bFGF-mediated suppression of TGF?1-induced SMC gene expression.
Conclusion— Our data suggest that bFGF-induced MEK/extracellular signal-regulated kinase signaling plays an antagonistic role in TGF?1-induced SMC gene expression through suppression of the SRF function. These data indicate that opposing effects of bFGF and TGF?1 on SMC gene expression control the phenotypic plasticity of SMCs.
TGF?1 induces SMC gene expression through an increase in SRF gene expression, which activates CArG-dependent transcription, and Src-tyrosine kinase is required for such an induction in 10T1/2 cells and vascular SMC. bFGF antagonizes TGF?1-induced SMC gene expression, although MEK1 activation without interfering with SRF:DNA binding activity and SRF gene expression.
Key Words: basic fibroblast growth factor ? transforming growth factor-?1 ? serum response factor ? SM22 ? smooth muscle cells
Introduction
Phenotypic modulation of smooth muscle cells (SMCs) contributes to development of atherosclerotic and restenotic lesions. There is considerable interest in identifying the various extracellular signals that regulate SMC phenotype and the molecular mechanisms underlying such SMC plasticity. Transforming growth factor-?1 (TGF?1) is one of the primary differentiation factors for SMCs.1 TGF?1 upregulates several SMC differentiation markers, such as smooth muscle -actin (SM-actin), smooth muscle myosin heavy chain, SM22, and h1 calponin in vitro. Moreover, TGF?1 induces expression of these SMC differentiation maker genes in a variety of nonsmooth muscle precursor cell types in culture, including multipotent embryonic fibroblast (10T1/2 cells) and neural crest cells.2,3 These results suggest that TGF?1 evokes an important signal that induces SMC differentiation.
In contrast, basic fibroblast growth factor (bFGF) is one of the most important mitogenic growth factors for SMCs1 and plays an important role in the onset and development of vascular disease. Several studies indicated that experimental reduction in bFGF expression inhibits SMC proliferation after intimal injury. Lindner et al suggested that in injured arteries, bFGF and FGF receptor-type 1 may be involved in the continued proliferative response of SMCs leading to neointima formation.4 SMCs respond to bFGF stimulation with proliferation, migration, and cytokine secretion occurring after vascular lesion.5
The promoter of almost all examined SMC-specific genes contain CArG (CC6GG) box,6,7 which serves as a binding site for serum response factor (SRF)8. An increasing number of studies provide direct evidence that the CArG boxes are required for promoter activation of SMC genes in vitro and in vivo.9–11 The smooth muscle calponin gene contains CArG box within the first intron, which mediates SMC-specific enhancer activity.12 However, the signaling pathways that regulate SRF expression and function have been poorly understood.
In the present study, we investigated the effects of TGF?1 and bFGF on SMC gene expression in the pluripotent 10T1/2 cells and in vascular SMCs. The results demonstrated that TGF?1 induces CArG-dependent SM22 gene expression via an increase in SRF expression, and Src family of tyrosine kinase is important for this effect. Furthermore, bFGF inhibits TGF?1-induced SM22 gene expression via mitogen-activated protein kinase kinase-1 (MEK1) by repressing SRF function without affecting SRF expression. Thus, we suggest that TGF?1 and bFGF act as antagonistic growth factors that regulate the CArG-dependent SMC marker gene expression by modulating SRF expression and function, respectively. These findings shed light on the role of TGF?1 and bFGF for modulating SMC gene expression during development and in vascular disease, in which SMC phenotypic change plays a crucial role.
Materials and Methods
The Materials and Methods section can be found in an online supplement available at http://atvb.ahajournals.org.
Results
TGF?1 Induces Expression of SMC Marker Genes in 10T1/2 Cells
Northern blot analyses revealed that the levels of mRNAs for SM22 and SM-actin were significantly induced by TGF?1 in 10T1/2 cells. Embryonic type of smooth muscle myosine heavy chain (SMemb) and Krüppel-like zinc-finger transcription factor 5/basic transcription regulatory element binding protein 2 (KLF5/BTEB2), a transcription factor implicated in the regulation of SMemb and others, were also induced. Calponin mRNA was scarcely detected by Northern blot, but RT-PCR analysis revealed its significant induction by TGF?1 in 10T1/2 cells. Analysis of temporal expression showed that SM-actin and SM22 mRNA levels were induced by TGF?1x2 hours and remained elevated for 12 hours after TGF?1 stimulation. SRF mRNA levels were peaked at 6 hours after TGF?1 stimulation (Figure IA through IC, available online at http://atvb.ahajournals.org).
As shown in Figure ID, cycloheximide (4 μg/mL, a nonspecific inhibitor for protein synthesis) alone modestly increased the expression of SM22 and SM-actin mRNAs. TGF?1-induced expression of SM22 and SM-actin mRNAs was rather attenuated by cycloheximide. These results suggest that de novo protein synthesis is partly required for the TGF?1-induced expression of SM22 and SM-actin genes.
Src Family of Tyrosine Kinase Mediates TGF?1-Induced Expression of SM22 and SM-Actin Genes
As shown in Figure 1A, TGF?1-induced SM22 mRNA expression in 10T1/2 cells was blocked by genistein (10 μmol/L; a tyrosine kinase inhibitor) but not by other protein kinase inhibitors, such as PD98059 (50 μmol/L; a specific inhibitor for MEK1, a mitogen-activated protein kinase kinase for extracellular signal–regulated kinase ), SB203580 (10 μmol/L; a specific inhibitor for p38MAP kinase), calphostin C (1 μmol/L; protein kinase C inhibitor), and wortmannin (1 μmol/L; a PI3 kinase inhibitor). We then examined the effect of daidzein (10 μmol/L; an inactive analog of genistein) and other tyrosine kinase inhibitors, such as herbimycin A (1 μmol/L; a relatively specific inhibitor for Src tyrosine kinases), tyrphostin 23 (100 μmol/L; a broad inhibitor for tyrosine kinase), and protein phosphatase 1 (PP1; 4 μmol/L; a more specific inhibitor for Src tyrosine kinases). As shown in Figure 1B, tyrphostin 23 as well as genistein but not daidzein dramatically inhibited induction of SM22 by TGF?1. Furthermore, herbimycin A and PP1 potently attenuated the TGF?1-induced SM22 expression. To examine the effect of Src family of tyrosine kinase in cultured rat aortic SMCs (RASMCs), levels of SM22 and SM-actin mRNA were measured by real-time RT-PCR in TGF?1-treated RASMCs (Figure II, available online at http://atvb.ahajournals.org). TGF?1-induced SM22 and SM-actin mRNA expression in RASMCs was significantly blocked by herbimycin A. These results suggest that Src family of tyrosine kinase pathways mediate TGF?1-induced SM22 gene expression.
Figure 1. Effect of Src-tyrosine kinase inhibitors on induction of SM22 and SM-actin expression by TGF?1. A, 10T1/2 cells were pretreated with PD98059, SB203580, calphostin C, genistein, and wortmannin for 1 hour and were exposed to TGF?1 (1 ng/mL) for 12 hours. Total cellular RNA (10 μg) was analyzed by Northern blotting for SM22 and SM -actin mRNA. B, 10T1/2 cells were pretreated with genistein, daidzein, herbimycin A, tyrphostin 23, and PP1 for 1 hour and were exposed to TGF?1 (1 ng/mL) for 12 hours. Methylene blue–stained 28S ribosomal RNA indicates that comparable amounts of total RNA actually blotted onto a membrane. PD indicates PD98059; SB, SB203580; Calph, calphostin C; Genist, genistein; Wort, wortmannin; Daidz, daidzein; Herb, herbimycin A; Tyr23, tyrphostin 23.
CArG Box Is Required for TGF?1-Mediated Increase in SM22 Promoter Activity
To determine the effects of TGF?1 on SM22 promoter activity, a series of 5'-deletion constructs of SM22 promoter was transfected into 10T1/2 cells and cultured RASMCs (Figure IIIA and IIIB, available online at http://atvb.ahajournals.org). TGF?1 induced promoter activity of –305Luc, –215Luc, and –158Luc, whereas –115Luc was unresponsive. Because the sequence between –158 and –115 contains 1 CArG box, (5'-CCAAATATGG-3', located at –150), we then tested the role of CArG box in the induction of promoter activity in response to TGF?1. We introduced nucleotide substitutions of 5 successive bases into the flanking or core sequence within the CArG box in the context of the –158Luc construct (Figure 2A and 2B). Luciferase activity in unstimulated cells transfected with –158m1Luc, –158m2Luc, and –158m3Luc was significantly lower than that of wild-type promoter construct –158Luc. More importantly, neither –158m2Luc nor –158m3Luc were responsive to TGF?1. These results suggest that integrity of CArG box is essential for activation of the SM22 promoter by TGF?1.
Figure 2. Effects of TGF?1on SM22 promoter activity. A, Mutated bases within the CArG box. Mutation 1 (m1), 2 (m2), and (m3) were introduced into the –158Luc context, which yielded –158m1Luc, –158m2Luc, and –158m3Luc, respectively. Mutations of wild-type sequence appear in boldface. B, Luciferase activity of the each construct shown in A. Cells transfected with indicated reporter genes were incubated with vehicle or TGF?1 (1 ng/mL) for 24 hours as described in Materials and Methods. *P<0.05 vs control.
TGF?1 Stimulates SRF Gene Expression at the Transcriptional Level Through Src Family of Tyrosine Kinase Pathway
To determine whether TGF?1 increases SRF binding activity, we performed electrophoretic mobility shift assay (EMSA) by using the nuclear protein extracts from unstimulated or TGF?1-stimulated 10T1/2 cells and the oligonucleotide-containing sequence between –158 and –133 of the SM22 promoter as probe. Shifted complex was increased in amounts in TGF?1-treated 10T1/2 cells, and anti-SRF antibody completely supershifted the complex (Figure 3A and Figure IVA, available online at http://atvb.ahajournals.org). An increase in SRF binding by TGF?1 was observed in the presence of PP1 because PP1 by itself increased SRF binding (Figure 3A). We next performed Northern blot analysis to test the effect of Src inhibition on TGF?1-induced SRF mRNA expression in 10T1/2 cells. TGF?1 induction of SRF was completely inhibited by PP1 or herbimycin, both of which are the Src family kinase inhibitors (Figure 3B). Western blot analysis showed the increase in SRF protein levels in response to TGF?1 (Figure IVB). TGF?1 effects on SRF promoter activity were evaluated by transfecting –2052SRF/Luc, a luciferase reporter gene containing SRF promoter region between –2052 and 114 into 10T1/2 cells and cultured RASMCs. As shown in Figure IVC, TGF?1 stimulated luciferase activity of –2052SRF/Luc by 2.5-fold in 10T1/2 cells and by 1.7-fold in RASMCs. These results indicate that TGF?1 stimulates SRF gene expression at least partly at the transcriptional level through Src family of tyrosine kinase pathway.
Figure 3. Effect of TGF?1 on SRF gene expression and binding to CArG box. A, EMSA. Nuclear extracts from 10T1/2 cells grown in absence or presence of TGF?1 were incubated with wild-type probe, which corresponds to SM22 (from –158 bp to –133 bp) sequence, and were subjected to electrophoresis. Nuclear extract from 10T1/2 cells in absence or presence of TGF?1 or PP1 were incubated with the probe using the antibody against SRF. Positions of the sequence-specific DNA protein complexes (C) and supershifted complexes (S) are indicated. B, Northern blot analysis. 10T1/2 cells were pretreated with either PP1 or herbimycin A for 1 hour, exposed to TGF?1 for 12 hours, and analyzed by Northern blotting for SRF.
bFGF Inhibits TGF?1-Induced SM22 and SM-Actin Gene Via Activation of MEK1
To investigate the effects of mitogenic stimulation on TGF?1-induced SM22 and SM-actin gene expression, we used various growth factors, including platelet-derived growth factor (PDGF)-BB, vascular endothelial growth factor (VEGF), interleukin (IL)-6, epidermal growth factor (EGF) and bFGF in 10T1/2 cells (Figure VA, available online at http://atvb.ahajournals.org). Among these, bFGF markedly inhibited TGF?1 induction of mRNA levels for SM22, SM-actin, and calponin. In contrast, dedifferentiated markers, such as SMemb and KLF5/BTEB2, were not affected (Figure 4A).
Figure 4. bFGF effects on TGF?1-stimulated SMC marker gene expression. A, Top, Total RNA was extracted from 10T1/2 cells treated with TGF?1 (1 ng/mL) or bFGF (10 ng/mL) and analyzed by Northern blotting for SM22, SM-actin, SMemb, and KLF5/BTEB2 mRNAs. Methylene blue–stained 28S ribosomal RNA indicates that comparable amounts of total RNA actually blotted onto a membrane. Bottom, Southern blot analysis of RT-PCR products for calponin and GAPDH. B, Top, 10T1/2 cells were pretreated with PD98059, SB203580, genistein, and wortmannin for 1 hour and were exposed to TGF?1 or bFGF for 12 hours. Total cellular RNA was analyzed by Northern blot for SM22. PD indicates PD98059; SB, SB203580; Genist, genistein; Wort, wortmannin. Bottom, Statistical analysis of the effect of protein kinase inhibitors on reduction of TGF?1-induced SM22 expression by bFGF. *P<0.05 vs bFGF(–). C, Phosphorylation of ERK1/2 by TGF?1 or bFGF for 5 minutes in 10T1/2 cells. 10T1/2 cells were treated with or without TGF?1 or bFGF, and total cellular lysates were prepared for Western blotting using either anti-phospho-ERK1/2 or anti-ERK1/2.
To investigate the signaling pathways involved in this repression, we tested the effects of inhibitors for protein kinases on TGF?1-induced SM22 gene expression. As shown in Figure 4B, bFGF inhibited TGF?1-induced SM22 expression by 55% and 17% in the absence and presence of PD98059 (MEK1 inhibitor), respectively, in 10T1/2 cells. In contrast, SB203580, genistein, and wortmannin had no effect on bFGF inhibition. To examine whether PD98059 inhibits bFGF effect in cultured RASMCs, levels of SM22 mRNA were measured by real-time RT-PCR in TGF?1-treated or bFGF-treated RASMCs. As shown in supplemental Figure VB, bFGF did not attenuate TGF?1-induced SM22 expression in the presence of PD98059 in RASMCs. These results indicated that bFGF represses TGF?1-induced expression of the SM22 gene via activation of MEK1.
To examine whether bFGF affects TGF?1-induced SM22 promoter activity, we performed luciferase assays. bFGF significantly attenuated TGF?1 effects on the SM22 promoter, and PD98059 inhibited such an effect of bFGF (Figure VC).
To evaluate TGF?1 or bFGF pathway for ERK1/2 activation, 10T1/2 cells were treated with TGF?1 or bFGF. Figure 4C and Figure VD show ERK phosphorylation was detected at 5 minutes and remained 60 minutes after bFGF stimulation. Although ERK phosphorylation was detected 5 minutes after TGF?1 stimulation, this phosphorylation significantly attenuated thereafter.
Next, we performed Western blot analyses to test whether activation of MEK1 pathways by bFGF inhibits TGF?1-induced c-Src activation. As shown in Figure VI (available online at http://atvb.ahajournals.org), bFGF as well as TGF?1 induced c-Src activation, and simultaneous stimulation with TGF?1 and bFGF exerts the additive effect on c-Src phosphorylation. PD98059 by itself induced c-Src phosphorylation, and thus an induction of c-Src phosphorylation by bFGF was not blocked by PD98059. However, these data imply that the inhibitory effects of bFGF on TGF?1-induced SM22 expression are not mediated through inhibition of c-Src phosphorylation.
MEK1 Inhibits SRF Function Independent of Expression and DNA Binding Activity of SRF
To determine whether MEK1 activation inhibits SM22 expression, 10T1/2 cells and cultured RASMCs were transfected with expression vector for MEK1 or empty vector pcDNA3 along with the SM22-luciferase reporter gene. Overexpression of MEK1 reduced TGF?1-stimulated luciferase activity of –305Luc (Figure 5 and Figure VIIA, available online at http://atvb.ahajournals.org). Furthermore, induction of the SM22 promoter activity by SRF was prevented by MEK1 overexpression in 10T1/2 cells (Figure 5).
Figure 5. MEK1 effects on TGF?1-stimulated SRF function. 10T1/2 cells were transfected with the SM22-305Luc along with pcDNA3 or MEK1/pcDNA3. Left, TGF?1 effect on SM22-305Luc promoter activity with or without MEK1. Right, Effect of SRF overexpression on SM22-305Luc promoter activity with or without MEK1. *P<0.05 vs control. Bottom, Activation of ERK by overexpressing MEK1. 10T1/2 cells were transfected with MEK1/pcDNA3 plasmid, and total cellular lysates were prepared for Western blotting using either anti-phospho-ERK1/2 or anti-ERK1/2.
The observation that bFGF inhibited the TGF?1-induced SM22 promoter activity led us to test whether bFGF inhibits SRF expression. Interestingly, bFGF had no measurable effects on SRF expression as assessed by Northern blot and Western blot analyses. Furthermore, EMSA showed that bFGF did not decrease and rather increased the binding of SRF to the CArG box (Figure VIIB through VIID). Collectively, these results suggest that bFGF inhibits SM22 expression not through inhibition of the SRF protein synthesis or inhibition of DNA binding but possibly through repression of SRF function (Figure 6).
Figure 6. A model for intracellular signaling that controls SMC gene expression in response to TGF?1 and bFGF. Signaling pathways mediating TGF?1-induced expression of the SMC gene and bFGF-mediated repression of TGF?1 signaling. Activation of MEK1 and subsequent activation of ERK1/2 play key roles in bFGF-mediated repression of SRF function.
Discussion
The current study shows TGF?1 actions on SMC gene expression in 10T1/2 cells and RASMCs. TGF?1-directed SMC gene expression is mediated through Src-tyrosine kinase activation and is associated with the increases in SRF gene expression at the transcriptional level. We also demonstrated that TGF?1 induction on SMC gene expression is significantly inhibited by bFGF. ERK1/2 activation mediates this effect of bFGF on TGF?1 induction of SMC gene expression; in fact, inducible expression of SM22 was attenuated in the presence of PD98059. In addition, phosphorylation of ERK1/2 by transfecting MEK1 expression plasmid inhibited TGF?1 action on SM22 promoter. Furthermore, bFGF did not inhibit TGF?1-induced SRF expression and DNA binding activity. Finally, when several growth factors were tested for their capacity to inhibit TGF?1 action on SMC gene expression, we observed that PDGF-BB, VEGF, PGI2, IL-6, and EGF did not possess the potent capacity of inhibiting TGF?1-driven SM22 expression in 10T1/2 cells. Thus, our data highlight the critical role of bFGF–MEK1 pathway in inhibition of TGF?1-driven SMC differentiation.
It is worth stressing that TGF?1 signaling was inhibited by Src family of tyrosine kinase inhibitor. This finding is somewhat surprising because it has been described that TGF?1 decreased Src kinase activity or induced degradation of activated Src kinase.13 Very few precedent reports showed the activation of Src tyrosine kinases by TGF?1. Su et al have shown that in bovine articular chondrocyte, TGF?1-mediated induction of tissue inhibitor of metalloproteinases-3 expression was inhibited by Src tyrosine kinase inhibitor as well as by serine/threonine protein kinase inhibitor.14 We confirmed the activation of Src family of tyrosine kinase by Western blot analysis using the anti-c-Src antibody. Although the precise mechanisms remain to be determined, it is possible that TGF?1 signaling varies depending on cell types.
Despite the fact that bFGF did not decrease but rather increased SRF mRNA and protein levels, bFGF significantly attenuated induction of TGF?1-mediated SMC-specific genes. Then how can bFGF inhibit the effect of TGF?1? We examined whether bFGF inhibits the binding of SRF to CArG box and found that bFGF had essentially no effect on the binding activity of SRF to the CArG box. These data led us to speculate that bFGF represses SRF function through: (1) post-transcriptional SRF modification; (2) an induction of repressors of SRF including SRF5,15 Id2,16 and SMART;17 (3) a repression of SRF coactivators such as ternary complex factor, p300/CBP, SRC-1, and myocardin;18,19 and (4) an alteration of chromatin structure regulated by histone acetylase or histone deacetylase activity.20,21 Studies to examine these possibilities are currently in progress.
Our finding that MEK1 inhibitor PD98059 attenuated bFGF-mediated repression of TGF?1-induced SM-actin and SM22 expression deserves further attention. Previous studies have provided ample evidence indicating that MEK1 activation leads to the increase in SRF-dependent transcription through activation of accessory factors that bind to the SRF–CArG box complex, including members of the Ets family of transcription factors, Elk-1, SAP-1a, b, SAP-2, and ERP-1.22 Thus, a critical question is how SRF function could selectively be upregulated or downregulated by MEK1. This issue appears to be analogous to the question regarding the differential effects of mitogenic stimulation on the expression of SMC-specific genes and c-fos gene, both of which contain functional CArG box.23 The most plausible explanation is that modulation of SRF function by MEK1 signaling is dependent on sequence flanking CArG box. However, our data indicate that PDGF and EGF, both of which are known to activate MEK1 signaling, are less potent than bFGF in inhibiting TGF?1-induced SM22 expression in 10T1/2 cells. Thus, it is tempting to speculate that bFGF-specific events other than MEK1 activation are necessary to inhibit TGF?1-induced SMC gene expression.
It has been shown that the nuclear factors involved in the TGF?1 control element–mediated transcription belong to the zinc finger family of KLFs, such as GKLF and KLF5/BTEB2.24 MacLellan et al have described previously that TGF?1-induced activation of skeletal -actin promoter required cooperation of SRF, YY1, and transcriptional enhancer factor-1.25 However, our data suggest that GKLF, BTEB2/IKLF, or YY1 do not play a major role in repressing the SM22 promoter by bFGF. The reasons for such an assumption are: (1) mRNA for BTEB2/IKLF, which positively regulates SM22 promoter, was induced by bFGF; (2) mRNA for GKLF, which functions as a negative regulator of SM22 expression, was not changed by bFGF; and (3) anti-YY1 antibody did not affect the DNA:protein complex as assessed by EMSA (data not shown).
What is the in vivo relevance of our observation? It has been shown previously that in early, simple, and advanced atherosclerotic lesions, both TGF?1 and bFGF were expressed in intimal SMCs. Although TGF?1 acts as a bifunctional regulator for SMC differentiation depending on growth status and the presence of other growth factors, an increased expression of bFGF was associated consistently with SMC proliferation of the atheromatous lesions.26 However, the role of bFGF in SMC differentiation has been described poorly. Our data indicate that SMC or embryonic fibroblasts exposed simultaneously to TGF?1 and bFGF express SMC marker genes significantly less than in response to TGF?1 alone. This suggests that SMCs in atherosclerotic lesions that contain abundant bFGF are less differentiated than those expressing predominant TGF?1. Given that undifferentiated SMCs that highly express the genes for proteases that degrade matrix proteins and bFGF may potentially be involved in plaque neovascularization, bFGF-driven events may contribute to formation of unstable plaques and to life-threatening complications of atheroma and provide new options for therapeutic intervention.
In summary, we demonstrated that 2 growth factors (TGF?1 and bFGF), which have been shown independently to play critical roles in regulation of smooth muscle development, antagonistically affect SRF-dependent SMC gene expression. In addition, we demonstrate that bFGF-induced MEK1 signaling attenuates SRF function but not its DNA binding activity and expression. These findings provide novel insight into SRF function regulation during SMC differentiation, which is influenced profoundly by growth factors.
Acknowledgments
This work was supported in part by a grant-in-aid for scientific research from the Ministry of Education, Science, Sport, and Culture of Japan and a grant from the Japan Cardiovascular Foundation. We thank Miki Yamazaki and Akemi Yoguchi for excellent technical help.
References
Owens GK. Regulation of differentiation of vascular smooth muscle cells. Physiol Rev. 1995; 75: 487–517.
Shah NM, Groves AK, Anderson DJ. Alternative neural crest cell fates are instructively promoted by TGF? superfamily members. Cell. 1996; 85: 331–343.
Hirschi KK, Rohovsky SA, D’Amore PA. PDGF, TGF?, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate. J Cell Biol. 1998; 141: 805–814.
Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc Natl Acad Sci U S A. 1991; 88: 3739–3743.
Nugent MA, Karnovsky MJ, Edelman ER. Vascular cell-derived heparan sulfate shows coupled inhibition of basic fibroblast growth factor binding and mitogenesis in vascular smooth muscle cells. Circ Res. 1993; 73: 1051–1060.
Madsen CS, Regan CP, Owens GK. Interaction of CArG elements and a GC-rich repressor element in transcriptional regulation of the smooth muscle myosin heavy chain gene in vascular smooth muscle cells. J Biol Chem. 1997; 272: 29842–29851.
Solway J, Seltzer J, Samaha FF, Kim S, Alger LE, Niu Q, Morrisey EE, Ip HS, Parmacek MS. Structure and expression of a smooth muscle cell-specific gene, SM22. J Biol Chem. 1995; 270: 13460–13469.
Miwa T, Kedes L. Duplicated CArG box domains have positive and mutually dependent regulatory roles in expression of the human -cardiac actin gene. Mol Cell Biol. 1987; 7: 2803–2813.
Mack CP, Thompson MM, Lawrenz SS, Owens GK. Smooth muscle -actin CArG elements coordinate formation of a smooth muscle cell-selective, serum response factor-containing activation complex. Circ Res. 1999; 86: 221–232.
Li L, Liu Z, Mercer B, Overbeek P, Olson EN. Evidence for serum response factor-mediated regulatory networks governing SM22 transcription in smooth, skeletal, and cardiac muscle cells. Dev Biol. 1997; 187: 311–321.
Kim S, Ip HS, Lu MM, Clendenin C, Parmacek MS. A serum response factor-dependent transcriptional regulatory program identifies distinct smooth muscle cell sublineages. Mol Cell Biol. 1997; 17: 2266–2278.
Miano JM, Carlson MJ, Spencer JA, Misra RP. Serum response factor-dependent regulation of the smooth muscle calponin gene. J Biol Chem. 2000; 275: 9814–9822.
Fukuda K, Kawata S, Tamura S, Matsuda Y, Inui Y, Igura T, Inoue S, Kudara T, Matsuzawa Y. Transforming growth factor-?1-induced degradation of activated Src tyrosine kinase in rat fibroblasts. Oncogene. 1998; 16: 3349–3356.
Su S, DiBattista JA, Sun Y, Li WQ, Zafarullah M. Up-regulation of tissue inhibitor of metalloproteinases-3 gene expression by TGF-? in articular chondrocytes is mediated by serine/threonine and tyrosine kinases. J Cell Biochem. 1998; 70: 517–527.
Belaguli NS, Zhou W, Trinh TH, Majesky MW, Schwartz RJ. Dominant negative murine serum response factor: alternative splicing within the activation domain inhibits transactivation of serum response factor binding targets. Mol Cell Biol. 1999; 19: 4582–4591.
Yates PR, Atherton GT, Deed RW, Norton JD, Sharrocks AD. Id helix-loop-helix proteins inhibit nucleoprotein complex formation by the TCF ETS-domain transcription factors. EMBO J. 1999; 18: 968–976.
Lee SK, Kim JH, Lee YC, Cheong J, Lee JW. Silencing mediator of retinoic acid and thyroid hormone receptors, as a novel transcriptional corepressor molecule of activating protein-1, nuclear factor-kappaB, and serum response factor. J Biol Chem. 2000; 275: 12470–12474.
Wang D, Chang PS, Wang Z, Sutherland L, Richardson JA, Small E, Krieg PA, Olson EN. Activation of cardiac gene expression by myocardin, a transcriptional cofactor for serum response factor. Cell. 2001; 105: 851–862.
Yoshida T, Shinha S, Dandre F, Wamhoff BR, Hoofnagle MH, Kremer BE, Wang D, Olson EN, Owens GK. Myocardin is a key regulator of CArG-dependent transcription of multiple smooth muscle marker gene. Circ Res. 2003; 92: 856–864.
Manabe I, Owens GK. Recruitment of serum response factor and hyperacetylation of histones at smooth muscle-specific regulatory regions during differentiation of a novel P19-derived in vitro smooth muscle differentiation system. Circ Res. 2002; 88: 1127–1134.
Qiu P, Li L. Histone acetylation and recruitment of serum responsive factor and CREB-binding protein onto SM22 promoter during SM22 gene expression. Circ Res. 2002; 90: 858–865.
Masutani H, Magnaghi JL, Ait SAS, Groisman R, Robin P, Harel BA, Ait SAS. Activation of the c-fos SRE through SAP-1a. Oncogene. 1997; 15: 1661–1669.
Chang PS, Li L, McAnally J, Olson EN. Muscle specificity encoded by specific serum response factor-binding sites. J Biol Chem. 2001; 276: 17206–17212.
Adam PJ, Regan CP, Hautmann MB, Owens GK. Positive- and negative-acting Kruppel-like transcription factors bind a transforming growth factor ? control element required for expression of the smooth muscle cell differentiation marker SM22 in vivo. J Biol Chem. 2000; 275: 37798–37806.
MacLellan WR, Lee TC, Schwartz RJ, Schneider MD. Transforming growth factor-? response elements of the skeletal alpha-actin gene. Combinatorial action of serum response factor, YY1, and the SV40 enhancer-binding protein, TEF-1. J Biol Chem. 1994; 269: 16754–16760.
Hughes SE, Crossman D, Hall PA. Expression of basic and acidic fibrobast growth factors and their receptor in normal and atherosclerotic human arteries. Cardiovasc Res. 1993; 27: 1214–1219.(Keiko Kawai-Kowase; Hirok)
ABSTRACT
Objective— Transforming growth factor-?1 (TGF?1) and fibroblast growth factor (FGF) families play a pivotal role during vascular development and in the pathogenesis of vascular disease. However, the interaction of intracellular signaling evoked by each of these growth factors is not well understood. The present study was undertaken to examine the molecular mechanisms that mediate the effects of TGF?1 and basic FGF (bFGF) on smooth muscle cell (SMC) gene expression.
Methods and Results— TGF?1 induction of SMC gene expression, including smooth muscle protein 22- (SM22) and smooth muscle -actin, was examined in the pluripotent 10T1/2 cells. Marked increase in these mRNA levels by TGF?1 was inhibited by c-Src-tyrosine kinase inhibitors and protein synthesis inhibitor cycloheximide. Functional studies with deletion and site-directed mutation analysis of the SM22 promoter demonstrated that TGF?1 activated the SM22 promoter through a CC(A/T-rich)6GG (CArG) box, which serves as a serum response factor (SRF)–binding site. TGF?1 increased SRF expression through an increase in transcription of the SRF gene. In the presence of bFGF, TGF?1 induction of SMC marker gene expression was significantly attenuated. Transient transfection assays showed that bFGF significantly suppressed induction of the SM22 promoter–driven luciferase activity by TGF?1, whereas bFGF had no effects on the TGF?1-mediated increase in SRF expression and SRF:DNA binding activity. Mitogen-activated protein kinase kinase-1 (MEK1) inhibitor PD98059 abrogated the bFGF-mediated suppression of TGF?1-induced SMC gene expression.
Conclusion— Our data suggest that bFGF-induced MEK/extracellular signal-regulated kinase signaling plays an antagonistic role in TGF?1-induced SMC gene expression through suppression of the SRF function. These data indicate that opposing effects of bFGF and TGF?1 on SMC gene expression control the phenotypic plasticity of SMCs.
TGF?1 induces SMC gene expression through an increase in SRF gene expression, which activates CArG-dependent transcription, and Src-tyrosine kinase is required for such an induction in 10T1/2 cells and vascular SMC. bFGF antagonizes TGF?1-induced SMC gene expression, although MEK1 activation without interfering with SRF:DNA binding activity and SRF gene expression.
Key Words: basic fibroblast growth factor ? transforming growth factor-?1 ? serum response factor ? SM22 ? smooth muscle cells
Introduction
Phenotypic modulation of smooth muscle cells (SMCs) contributes to development of atherosclerotic and restenotic lesions. There is considerable interest in identifying the various extracellular signals that regulate SMC phenotype and the molecular mechanisms underlying such SMC plasticity. Transforming growth factor-?1 (TGF?1) is one of the primary differentiation factors for SMCs.1 TGF?1 upregulates several SMC differentiation markers, such as smooth muscle -actin (SM-actin), smooth muscle myosin heavy chain, SM22, and h1 calponin in vitro. Moreover, TGF?1 induces expression of these SMC differentiation maker genes in a variety of nonsmooth muscle precursor cell types in culture, including multipotent embryonic fibroblast (10T1/2 cells) and neural crest cells.2,3 These results suggest that TGF?1 evokes an important signal that induces SMC differentiation.
In contrast, basic fibroblast growth factor (bFGF) is one of the most important mitogenic growth factors for SMCs1 and plays an important role in the onset and development of vascular disease. Several studies indicated that experimental reduction in bFGF expression inhibits SMC proliferation after intimal injury. Lindner et al suggested that in injured arteries, bFGF and FGF receptor-type 1 may be involved in the continued proliferative response of SMCs leading to neointima formation.4 SMCs respond to bFGF stimulation with proliferation, migration, and cytokine secretion occurring after vascular lesion.5
The promoter of almost all examined SMC-specific genes contain CArG (CC6GG) box,6,7 which serves as a binding site for serum response factor (SRF)8. An increasing number of studies provide direct evidence that the CArG boxes are required for promoter activation of SMC genes in vitro and in vivo.9–11 The smooth muscle calponin gene contains CArG box within the first intron, which mediates SMC-specific enhancer activity.12 However, the signaling pathways that regulate SRF expression and function have been poorly understood.
In the present study, we investigated the effects of TGF?1 and bFGF on SMC gene expression in the pluripotent 10T1/2 cells and in vascular SMCs. The results demonstrated that TGF?1 induces CArG-dependent SM22 gene expression via an increase in SRF expression, and Src family of tyrosine kinase is important for this effect. Furthermore, bFGF inhibits TGF?1-induced SM22 gene expression via mitogen-activated protein kinase kinase-1 (MEK1) by repressing SRF function without affecting SRF expression. Thus, we suggest that TGF?1 and bFGF act as antagonistic growth factors that regulate the CArG-dependent SMC marker gene expression by modulating SRF expression and function, respectively. These findings shed light on the role of TGF?1 and bFGF for modulating SMC gene expression during development and in vascular disease, in which SMC phenotypic change plays a crucial role.
Materials and Methods
The Materials and Methods section can be found in an online supplement available at http://atvb.ahajournals.org.
Results
TGF?1 Induces Expression of SMC Marker Genes in 10T1/2 Cells
Northern blot analyses revealed that the levels of mRNAs for SM22 and SM-actin were significantly induced by TGF?1 in 10T1/2 cells. Embryonic type of smooth muscle myosine heavy chain (SMemb) and Krüppel-like zinc-finger transcription factor 5/basic transcription regulatory element binding protein 2 (KLF5/BTEB2), a transcription factor implicated in the regulation of SMemb and others, were also induced. Calponin mRNA was scarcely detected by Northern blot, but RT-PCR analysis revealed its significant induction by TGF?1 in 10T1/2 cells. Analysis of temporal expression showed that SM-actin and SM22 mRNA levels were induced by TGF?1x2 hours and remained elevated for 12 hours after TGF?1 stimulation. SRF mRNA levels were peaked at 6 hours after TGF?1 stimulation (Figure IA through IC, available online at http://atvb.ahajournals.org).
As shown in Figure ID, cycloheximide (4 μg/mL, a nonspecific inhibitor for protein synthesis) alone modestly increased the expression of SM22 and SM-actin mRNAs. TGF?1-induced expression of SM22 and SM-actin mRNAs was rather attenuated by cycloheximide. These results suggest that de novo protein synthesis is partly required for the TGF?1-induced expression of SM22 and SM-actin genes.
Src Family of Tyrosine Kinase Mediates TGF?1-Induced Expression of SM22 and SM-Actin Genes
As shown in Figure 1A, TGF?1-induced SM22 mRNA expression in 10T1/2 cells was blocked by genistein (10 μmol/L; a tyrosine kinase inhibitor) but not by other protein kinase inhibitors, such as PD98059 (50 μmol/L; a specific inhibitor for MEK1, a mitogen-activated protein kinase kinase for extracellular signal–regulated kinase ), SB203580 (10 μmol/L; a specific inhibitor for p38MAP kinase), calphostin C (1 μmol/L; protein kinase C inhibitor), and wortmannin (1 μmol/L; a PI3 kinase inhibitor). We then examined the effect of daidzein (10 μmol/L; an inactive analog of genistein) and other tyrosine kinase inhibitors, such as herbimycin A (1 μmol/L; a relatively specific inhibitor for Src tyrosine kinases), tyrphostin 23 (100 μmol/L; a broad inhibitor for tyrosine kinase), and protein phosphatase 1 (PP1; 4 μmol/L; a more specific inhibitor for Src tyrosine kinases). As shown in Figure 1B, tyrphostin 23 as well as genistein but not daidzein dramatically inhibited induction of SM22 by TGF?1. Furthermore, herbimycin A and PP1 potently attenuated the TGF?1-induced SM22 expression. To examine the effect of Src family of tyrosine kinase in cultured rat aortic SMCs (RASMCs), levels of SM22 and SM-actin mRNA were measured by real-time RT-PCR in TGF?1-treated RASMCs (Figure II, available online at http://atvb.ahajournals.org). TGF?1-induced SM22 and SM-actin mRNA expression in RASMCs was significantly blocked by herbimycin A. These results suggest that Src family of tyrosine kinase pathways mediate TGF?1-induced SM22 gene expression.
Figure 1. Effect of Src-tyrosine kinase inhibitors on induction of SM22 and SM-actin expression by TGF?1. A, 10T1/2 cells were pretreated with PD98059, SB203580, calphostin C, genistein, and wortmannin for 1 hour and were exposed to TGF?1 (1 ng/mL) for 12 hours. Total cellular RNA (10 μg) was analyzed by Northern blotting for SM22 and SM -actin mRNA. B, 10T1/2 cells were pretreated with genistein, daidzein, herbimycin A, tyrphostin 23, and PP1 for 1 hour and were exposed to TGF?1 (1 ng/mL) for 12 hours. Methylene blue–stained 28S ribosomal RNA indicates that comparable amounts of total RNA actually blotted onto a membrane. PD indicates PD98059; SB, SB203580; Calph, calphostin C; Genist, genistein; Wort, wortmannin; Daidz, daidzein; Herb, herbimycin A; Tyr23, tyrphostin 23.
CArG Box Is Required for TGF?1-Mediated Increase in SM22 Promoter Activity
To determine the effects of TGF?1 on SM22 promoter activity, a series of 5'-deletion constructs of SM22 promoter was transfected into 10T1/2 cells and cultured RASMCs (Figure IIIA and IIIB, available online at http://atvb.ahajournals.org). TGF?1 induced promoter activity of –305Luc, –215Luc, and –158Luc, whereas –115Luc was unresponsive. Because the sequence between –158 and –115 contains 1 CArG box, (5'-CCAAATATGG-3', located at –150), we then tested the role of CArG box in the induction of promoter activity in response to TGF?1. We introduced nucleotide substitutions of 5 successive bases into the flanking or core sequence within the CArG box in the context of the –158Luc construct (Figure 2A and 2B). Luciferase activity in unstimulated cells transfected with –158m1Luc, –158m2Luc, and –158m3Luc was significantly lower than that of wild-type promoter construct –158Luc. More importantly, neither –158m2Luc nor –158m3Luc were responsive to TGF?1. These results suggest that integrity of CArG box is essential for activation of the SM22 promoter by TGF?1.
Figure 2. Effects of TGF?1on SM22 promoter activity. A, Mutated bases within the CArG box. Mutation 1 (m1), 2 (m2), and (m3) were introduced into the –158Luc context, which yielded –158m1Luc, –158m2Luc, and –158m3Luc, respectively. Mutations of wild-type sequence appear in boldface. B, Luciferase activity of the each construct shown in A. Cells transfected with indicated reporter genes were incubated with vehicle or TGF?1 (1 ng/mL) for 24 hours as described in Materials and Methods. *P<0.05 vs control.
TGF?1 Stimulates SRF Gene Expression at the Transcriptional Level Through Src Family of Tyrosine Kinase Pathway
To determine whether TGF?1 increases SRF binding activity, we performed electrophoretic mobility shift assay (EMSA) by using the nuclear protein extracts from unstimulated or TGF?1-stimulated 10T1/2 cells and the oligonucleotide-containing sequence between –158 and –133 of the SM22 promoter as probe. Shifted complex was increased in amounts in TGF?1-treated 10T1/2 cells, and anti-SRF antibody completely supershifted the complex (Figure 3A and Figure IVA, available online at http://atvb.ahajournals.org). An increase in SRF binding by TGF?1 was observed in the presence of PP1 because PP1 by itself increased SRF binding (Figure 3A). We next performed Northern blot analysis to test the effect of Src inhibition on TGF?1-induced SRF mRNA expression in 10T1/2 cells. TGF?1 induction of SRF was completely inhibited by PP1 or herbimycin, both of which are the Src family kinase inhibitors (Figure 3B). Western blot analysis showed the increase in SRF protein levels in response to TGF?1 (Figure IVB). TGF?1 effects on SRF promoter activity were evaluated by transfecting –2052SRF/Luc, a luciferase reporter gene containing SRF promoter region between –2052 and 114 into 10T1/2 cells and cultured RASMCs. As shown in Figure IVC, TGF?1 stimulated luciferase activity of –2052SRF/Luc by 2.5-fold in 10T1/2 cells and by 1.7-fold in RASMCs. These results indicate that TGF?1 stimulates SRF gene expression at least partly at the transcriptional level through Src family of tyrosine kinase pathway.
Figure 3. Effect of TGF?1 on SRF gene expression and binding to CArG box. A, EMSA. Nuclear extracts from 10T1/2 cells grown in absence or presence of TGF?1 were incubated with wild-type probe, which corresponds to SM22 (from –158 bp to –133 bp) sequence, and were subjected to electrophoresis. Nuclear extract from 10T1/2 cells in absence or presence of TGF?1 or PP1 were incubated with the probe using the antibody against SRF. Positions of the sequence-specific DNA protein complexes (C) and supershifted complexes (S) are indicated. B, Northern blot analysis. 10T1/2 cells were pretreated with either PP1 or herbimycin A for 1 hour, exposed to TGF?1 for 12 hours, and analyzed by Northern blotting for SRF.
bFGF Inhibits TGF?1-Induced SM22 and SM-Actin Gene Via Activation of MEK1
To investigate the effects of mitogenic stimulation on TGF?1-induced SM22 and SM-actin gene expression, we used various growth factors, including platelet-derived growth factor (PDGF)-BB, vascular endothelial growth factor (VEGF), interleukin (IL)-6, epidermal growth factor (EGF) and bFGF in 10T1/2 cells (Figure VA, available online at http://atvb.ahajournals.org). Among these, bFGF markedly inhibited TGF?1 induction of mRNA levels for SM22, SM-actin, and calponin. In contrast, dedifferentiated markers, such as SMemb and KLF5/BTEB2, were not affected (Figure 4A).
Figure 4. bFGF effects on TGF?1-stimulated SMC marker gene expression. A, Top, Total RNA was extracted from 10T1/2 cells treated with TGF?1 (1 ng/mL) or bFGF (10 ng/mL) and analyzed by Northern blotting for SM22, SM-actin, SMemb, and KLF5/BTEB2 mRNAs. Methylene blue–stained 28S ribosomal RNA indicates that comparable amounts of total RNA actually blotted onto a membrane. Bottom, Southern blot analysis of RT-PCR products for calponin and GAPDH. B, Top, 10T1/2 cells were pretreated with PD98059, SB203580, genistein, and wortmannin for 1 hour and were exposed to TGF?1 or bFGF for 12 hours. Total cellular RNA was analyzed by Northern blot for SM22. PD indicates PD98059; SB, SB203580; Genist, genistein; Wort, wortmannin. Bottom, Statistical analysis of the effect of protein kinase inhibitors on reduction of TGF?1-induced SM22 expression by bFGF. *P<0.05 vs bFGF(–). C, Phosphorylation of ERK1/2 by TGF?1 or bFGF for 5 minutes in 10T1/2 cells. 10T1/2 cells were treated with or without TGF?1 or bFGF, and total cellular lysates were prepared for Western blotting using either anti-phospho-ERK1/2 or anti-ERK1/2.
To investigate the signaling pathways involved in this repression, we tested the effects of inhibitors for protein kinases on TGF?1-induced SM22 gene expression. As shown in Figure 4B, bFGF inhibited TGF?1-induced SM22 expression by 55% and 17% in the absence and presence of PD98059 (MEK1 inhibitor), respectively, in 10T1/2 cells. In contrast, SB203580, genistein, and wortmannin had no effect on bFGF inhibition. To examine whether PD98059 inhibits bFGF effect in cultured RASMCs, levels of SM22 mRNA were measured by real-time RT-PCR in TGF?1-treated or bFGF-treated RASMCs. As shown in supplemental Figure VB, bFGF did not attenuate TGF?1-induced SM22 expression in the presence of PD98059 in RASMCs. These results indicated that bFGF represses TGF?1-induced expression of the SM22 gene via activation of MEK1.
To examine whether bFGF affects TGF?1-induced SM22 promoter activity, we performed luciferase assays. bFGF significantly attenuated TGF?1 effects on the SM22 promoter, and PD98059 inhibited such an effect of bFGF (Figure VC).
To evaluate TGF?1 or bFGF pathway for ERK1/2 activation, 10T1/2 cells were treated with TGF?1 or bFGF. Figure 4C and Figure VD show ERK phosphorylation was detected at 5 minutes and remained 60 minutes after bFGF stimulation. Although ERK phosphorylation was detected 5 minutes after TGF?1 stimulation, this phosphorylation significantly attenuated thereafter.
Next, we performed Western blot analyses to test whether activation of MEK1 pathways by bFGF inhibits TGF?1-induced c-Src activation. As shown in Figure VI (available online at http://atvb.ahajournals.org), bFGF as well as TGF?1 induced c-Src activation, and simultaneous stimulation with TGF?1 and bFGF exerts the additive effect on c-Src phosphorylation. PD98059 by itself induced c-Src phosphorylation, and thus an induction of c-Src phosphorylation by bFGF was not blocked by PD98059. However, these data imply that the inhibitory effects of bFGF on TGF?1-induced SM22 expression are not mediated through inhibition of c-Src phosphorylation.
MEK1 Inhibits SRF Function Independent of Expression and DNA Binding Activity of SRF
To determine whether MEK1 activation inhibits SM22 expression, 10T1/2 cells and cultured RASMCs were transfected with expression vector for MEK1 or empty vector pcDNA3 along with the SM22-luciferase reporter gene. Overexpression of MEK1 reduced TGF?1-stimulated luciferase activity of –305Luc (Figure 5 and Figure VIIA, available online at http://atvb.ahajournals.org). Furthermore, induction of the SM22 promoter activity by SRF was prevented by MEK1 overexpression in 10T1/2 cells (Figure 5).
Figure 5. MEK1 effects on TGF?1-stimulated SRF function. 10T1/2 cells were transfected with the SM22-305Luc along with pcDNA3 or MEK1/pcDNA3. Left, TGF?1 effect on SM22-305Luc promoter activity with or without MEK1. Right, Effect of SRF overexpression on SM22-305Luc promoter activity with or without MEK1. *P<0.05 vs control. Bottom, Activation of ERK by overexpressing MEK1. 10T1/2 cells were transfected with MEK1/pcDNA3 plasmid, and total cellular lysates were prepared for Western blotting using either anti-phospho-ERK1/2 or anti-ERK1/2.
The observation that bFGF inhibited the TGF?1-induced SM22 promoter activity led us to test whether bFGF inhibits SRF expression. Interestingly, bFGF had no measurable effects on SRF expression as assessed by Northern blot and Western blot analyses. Furthermore, EMSA showed that bFGF did not decrease and rather increased the binding of SRF to the CArG box (Figure VIIB through VIID). Collectively, these results suggest that bFGF inhibits SM22 expression not through inhibition of the SRF protein synthesis or inhibition of DNA binding but possibly through repression of SRF function (Figure 6).
Figure 6. A model for intracellular signaling that controls SMC gene expression in response to TGF?1 and bFGF. Signaling pathways mediating TGF?1-induced expression of the SMC gene and bFGF-mediated repression of TGF?1 signaling. Activation of MEK1 and subsequent activation of ERK1/2 play key roles in bFGF-mediated repression of SRF function.
Discussion
The current study shows TGF?1 actions on SMC gene expression in 10T1/2 cells and RASMCs. TGF?1-directed SMC gene expression is mediated through Src-tyrosine kinase activation and is associated with the increases in SRF gene expression at the transcriptional level. We also demonstrated that TGF?1 induction on SMC gene expression is significantly inhibited by bFGF. ERK1/2 activation mediates this effect of bFGF on TGF?1 induction of SMC gene expression; in fact, inducible expression of SM22 was attenuated in the presence of PD98059. In addition, phosphorylation of ERK1/2 by transfecting MEK1 expression plasmid inhibited TGF?1 action on SM22 promoter. Furthermore, bFGF did not inhibit TGF?1-induced SRF expression and DNA binding activity. Finally, when several growth factors were tested for their capacity to inhibit TGF?1 action on SMC gene expression, we observed that PDGF-BB, VEGF, PGI2, IL-6, and EGF did not possess the potent capacity of inhibiting TGF?1-driven SM22 expression in 10T1/2 cells. Thus, our data highlight the critical role of bFGF–MEK1 pathway in inhibition of TGF?1-driven SMC differentiation.
It is worth stressing that TGF?1 signaling was inhibited by Src family of tyrosine kinase inhibitor. This finding is somewhat surprising because it has been described that TGF?1 decreased Src kinase activity or induced degradation of activated Src kinase.13 Very few precedent reports showed the activation of Src tyrosine kinases by TGF?1. Su et al have shown that in bovine articular chondrocyte, TGF?1-mediated induction of tissue inhibitor of metalloproteinases-3 expression was inhibited by Src tyrosine kinase inhibitor as well as by serine/threonine protein kinase inhibitor.14 We confirmed the activation of Src family of tyrosine kinase by Western blot analysis using the anti-c-Src antibody. Although the precise mechanisms remain to be determined, it is possible that TGF?1 signaling varies depending on cell types.
Despite the fact that bFGF did not decrease but rather increased SRF mRNA and protein levels, bFGF significantly attenuated induction of TGF?1-mediated SMC-specific genes. Then how can bFGF inhibit the effect of TGF?1? We examined whether bFGF inhibits the binding of SRF to CArG box and found that bFGF had essentially no effect on the binding activity of SRF to the CArG box. These data led us to speculate that bFGF represses SRF function through: (1) post-transcriptional SRF modification; (2) an induction of repressors of SRF including SRF5,15 Id2,16 and SMART;17 (3) a repression of SRF coactivators such as ternary complex factor, p300/CBP, SRC-1, and myocardin;18,19 and (4) an alteration of chromatin structure regulated by histone acetylase or histone deacetylase activity.20,21 Studies to examine these possibilities are currently in progress.
Our finding that MEK1 inhibitor PD98059 attenuated bFGF-mediated repression of TGF?1-induced SM-actin and SM22 expression deserves further attention. Previous studies have provided ample evidence indicating that MEK1 activation leads to the increase in SRF-dependent transcription through activation of accessory factors that bind to the SRF–CArG box complex, including members of the Ets family of transcription factors, Elk-1, SAP-1a, b, SAP-2, and ERP-1.22 Thus, a critical question is how SRF function could selectively be upregulated or downregulated by MEK1. This issue appears to be analogous to the question regarding the differential effects of mitogenic stimulation on the expression of SMC-specific genes and c-fos gene, both of which contain functional CArG box.23 The most plausible explanation is that modulation of SRF function by MEK1 signaling is dependent on sequence flanking CArG box. However, our data indicate that PDGF and EGF, both of which are known to activate MEK1 signaling, are less potent than bFGF in inhibiting TGF?1-induced SM22 expression in 10T1/2 cells. Thus, it is tempting to speculate that bFGF-specific events other than MEK1 activation are necessary to inhibit TGF?1-induced SMC gene expression.
It has been shown that the nuclear factors involved in the TGF?1 control element–mediated transcription belong to the zinc finger family of KLFs, such as GKLF and KLF5/BTEB2.24 MacLellan et al have described previously that TGF?1-induced activation of skeletal -actin promoter required cooperation of SRF, YY1, and transcriptional enhancer factor-1.25 However, our data suggest that GKLF, BTEB2/IKLF, or YY1 do not play a major role in repressing the SM22 promoter by bFGF. The reasons for such an assumption are: (1) mRNA for BTEB2/IKLF, which positively regulates SM22 promoter, was induced by bFGF; (2) mRNA for GKLF, which functions as a negative regulator of SM22 expression, was not changed by bFGF; and (3) anti-YY1 antibody did not affect the DNA:protein complex as assessed by EMSA (data not shown).
What is the in vivo relevance of our observation? It has been shown previously that in early, simple, and advanced atherosclerotic lesions, both TGF?1 and bFGF were expressed in intimal SMCs. Although TGF?1 acts as a bifunctional regulator for SMC differentiation depending on growth status and the presence of other growth factors, an increased expression of bFGF was associated consistently with SMC proliferation of the atheromatous lesions.26 However, the role of bFGF in SMC differentiation has been described poorly. Our data indicate that SMC or embryonic fibroblasts exposed simultaneously to TGF?1 and bFGF express SMC marker genes significantly less than in response to TGF?1 alone. This suggests that SMCs in atherosclerotic lesions that contain abundant bFGF are less differentiated than those expressing predominant TGF?1. Given that undifferentiated SMCs that highly express the genes for proteases that degrade matrix proteins and bFGF may potentially be involved in plaque neovascularization, bFGF-driven events may contribute to formation of unstable plaques and to life-threatening complications of atheroma and provide new options for therapeutic intervention.
In summary, we demonstrated that 2 growth factors (TGF?1 and bFGF), which have been shown independently to play critical roles in regulation of smooth muscle development, antagonistically affect SRF-dependent SMC gene expression. In addition, we demonstrate that bFGF-induced MEK1 signaling attenuates SRF function but not its DNA binding activity and expression. These findings provide novel insight into SRF function regulation during SMC differentiation, which is influenced profoundly by growth factors.
Acknowledgments
This work was supported in part by a grant-in-aid for scientific research from the Ministry of Education, Science, Sport, and Culture of Japan and a grant from the Japan Cardiovascular Foundation. We thank Miki Yamazaki and Akemi Yoguchi for excellent technical help.
References
Owens GK. Regulation of differentiation of vascular smooth muscle cells. Physiol Rev. 1995; 75: 487–517.
Shah NM, Groves AK, Anderson DJ. Alternative neural crest cell fates are instructively promoted by TGF? superfamily members. Cell. 1996; 85: 331–343.
Hirschi KK, Rohovsky SA, D’Amore PA. PDGF, TGF?, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate. J Cell Biol. 1998; 141: 805–814.
Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc Natl Acad Sci U S A. 1991; 88: 3739–3743.
Nugent MA, Karnovsky MJ, Edelman ER. Vascular cell-derived heparan sulfate shows coupled inhibition of basic fibroblast growth factor binding and mitogenesis in vascular smooth muscle cells. Circ Res. 1993; 73: 1051–1060.
Madsen CS, Regan CP, Owens GK. Interaction of CArG elements and a GC-rich repressor element in transcriptional regulation of the smooth muscle myosin heavy chain gene in vascular smooth muscle cells. J Biol Chem. 1997; 272: 29842–29851.
Solway J, Seltzer J, Samaha FF, Kim S, Alger LE, Niu Q, Morrisey EE, Ip HS, Parmacek MS. Structure and expression of a smooth muscle cell-specific gene, SM22. J Biol Chem. 1995; 270: 13460–13469.
Miwa T, Kedes L. Duplicated CArG box domains have positive and mutually dependent regulatory roles in expression of the human -cardiac actin gene. Mol Cell Biol. 1987; 7: 2803–2813.
Mack CP, Thompson MM, Lawrenz SS, Owens GK. Smooth muscle -actin CArG elements coordinate formation of a smooth muscle cell-selective, serum response factor-containing activation complex. Circ Res. 1999; 86: 221–232.
Li L, Liu Z, Mercer B, Overbeek P, Olson EN. Evidence for serum response factor-mediated regulatory networks governing SM22 transcription in smooth, skeletal, and cardiac muscle cells. Dev Biol. 1997; 187: 311–321.
Kim S, Ip HS, Lu MM, Clendenin C, Parmacek MS. A serum response factor-dependent transcriptional regulatory program identifies distinct smooth muscle cell sublineages. Mol Cell Biol. 1997; 17: 2266–2278.
Miano JM, Carlson MJ, Spencer JA, Misra RP. Serum response factor-dependent regulation of the smooth muscle calponin gene. J Biol Chem. 2000; 275: 9814–9822.
Fukuda K, Kawata S, Tamura S, Matsuda Y, Inui Y, Igura T, Inoue S, Kudara T, Matsuzawa Y. Transforming growth factor-?1-induced degradation of activated Src tyrosine kinase in rat fibroblasts. Oncogene. 1998; 16: 3349–3356.
Su S, DiBattista JA, Sun Y, Li WQ, Zafarullah M. Up-regulation of tissue inhibitor of metalloproteinases-3 gene expression by TGF-? in articular chondrocytes is mediated by serine/threonine and tyrosine kinases. J Cell Biochem. 1998; 70: 517–527.
Belaguli NS, Zhou W, Trinh TH, Majesky MW, Schwartz RJ. Dominant negative murine serum response factor: alternative splicing within the activation domain inhibits transactivation of serum response factor binding targets. Mol Cell Biol. 1999; 19: 4582–4591.
Yates PR, Atherton GT, Deed RW, Norton JD, Sharrocks AD. Id helix-loop-helix proteins inhibit nucleoprotein complex formation by the TCF ETS-domain transcription factors. EMBO J. 1999; 18: 968–976.
Lee SK, Kim JH, Lee YC, Cheong J, Lee JW. Silencing mediator of retinoic acid and thyroid hormone receptors, as a novel transcriptional corepressor molecule of activating protein-1, nuclear factor-kappaB, and serum response factor. J Biol Chem. 2000; 275: 12470–12474.
Wang D, Chang PS, Wang Z, Sutherland L, Richardson JA, Small E, Krieg PA, Olson EN. Activation of cardiac gene expression by myocardin, a transcriptional cofactor for serum response factor. Cell. 2001; 105: 851–862.
Yoshida T, Shinha S, Dandre F, Wamhoff BR, Hoofnagle MH, Kremer BE, Wang D, Olson EN, Owens GK. Myocardin is a key regulator of CArG-dependent transcription of multiple smooth muscle marker gene. Circ Res. 2003; 92: 856–864.
Manabe I, Owens GK. Recruitment of serum response factor and hyperacetylation of histones at smooth muscle-specific regulatory regions during differentiation of a novel P19-derived in vitro smooth muscle differentiation system. Circ Res. 2002; 88: 1127–1134.
Qiu P, Li L. Histone acetylation and recruitment of serum responsive factor and CREB-binding protein onto SM22 promoter during SM22 gene expression. Circ Res. 2002; 90: 858–865.
Masutani H, Magnaghi JL, Ait SAS, Groisman R, Robin P, Harel BA, Ait SAS. Activation of the c-fos SRE through SAP-1a. Oncogene. 1997; 15: 1661–1669.
Chang PS, Li L, McAnally J, Olson EN. Muscle specificity encoded by specific serum response factor-binding sites. J Biol Chem. 2001; 276: 17206–17212.
Adam PJ, Regan CP, Hautmann MB, Owens GK. Positive- and negative-acting Kruppel-like transcription factors bind a transforming growth factor ? control element required for expression of the smooth muscle cell differentiation marker SM22 in vivo. J Biol Chem. 2000; 275: 37798–37806.
MacLellan WR, Lee TC, Schwartz RJ, Schneider MD. Transforming growth factor-? response elements of the skeletal alpha-actin gene. Combinatorial action of serum response factor, YY1, and the SV40 enhancer-binding protein, TEF-1. J Biol Chem. 1994; 269: 16754–16760.
Hughes SE, Crossman D, Hall PA. Expression of basic and acidic fibrobast growth factors and their receptor in normal and atherosclerotic human arteries. Cardiovasc Res. 1993; 27: 1214–1219.(Keiko Kawai-Kowase; Hirok)