Transforming growth factor-beta1-induced expression of smooth muscle marker genes involves activation of PKN and p38 MAPK.

Differentiated vascular smooth muscle cells (SMCs) exhibit a work phenotype characterized by expression of several well documented contractile apparatus-associated proteins. However, SMCs retain the ability to de-differentiate into a proliferative phenotype, which is involved in the progression of vascular diseases such as atherosclerosis and restenosis. Understanding the mechanisms involved in maintaining SMC differentiation is critical for preventing proliferation associated with vascular disease. In this study, the molecular mechanisms through which transforming growth factor-beta1 (TGF-beta1) induces differentiation of SMCs were examined. TGF-beta1 stimulated actin re-organization, inhibited cell proliferation, and up-regulated SMC marker gene expression in PAC-1 SMCs. These effects were blocked by pretreatment of cells with either HA1077 or Y-27632, which inhibit the kinases downstream of RhoA. Moreover, TGF-beta1 activated RhoA and its downstream target PKN. Overexpression of active PKN alone was sufficient to increase the transcriptional activity of the promoters that control expression of smooth muscle (SM) alpha-actin, SM-myosin heavy chain, and SM22alpha. In addition, PKN increased the activities of serum-response factor (SRF), GATA, and MEF2-dependent enhancer-reporters. RNA interference-mediated inhibition of PKN abolished TGF-beta1-induced activation of SMC marker gene promoters. Finally, examination of MAPK signaling demonstrated that TGF-beta1 increased the activity of p38 MAPK, which was required for activation of the SMC marker gene promoters. Co-expression of dominant negative p38 MAPK was sufficient to block PKN-mediated activation of the SMC marker gene promoters as well as the serum-response factor, GATA, and MEF2 enhancers. Taken together, these results identify components of an important intracellular signaling pathway through which TGF-beta1 activates PKN to promote differentiation of SMCs.

Quiescent vascular smooth muscle cells (SMCs) 1 exhibit a differentiated work phenotype characterized by the expression of several contractile apparatus-associated proteins such as smooth muscle myosin heavy chain (SM-MHC), smooth muscle ␣-actin (SM ␣-actin), calponin, and SM22␣ (1)(2)(3)(4). Expression of these SMC marker genes results in the increased myofibrillar organization required for contraction. Differentiated SMCs retain the ability to "de-differentiate" into a synthetic, proliferative phenotype, characterized by the loss of SMC marker gene expression. This phenotype reappears with certain vascular disease states, including atherosclerosis and restenosis (1,5). Consequently, understanding the mechanisms that promote and maintain the differentiated phenotype will provide insight into the aberrant growth seen in pathologic SMC proliferation.
Considerable strides have been made toward understanding the molecular mechanisms that regulate SMC differentiation. In particular, a role for TGF-␤1 in SMC differentiation has been demonstrated both in vitro and in vivo, although the signal transduction pathways required for these effects are not completely clear (6 -10). The classic signaling cascade downstream of TGF-␤1 involves activation of the Smads family of transcription factors (11). However, TGF-␤1 signaling has been linked recently to the activation of the small GTPase RhoA in other cell types (12)(13)(14)(15). RhoA signaling is known to induce many of the same phenotype changes in SMCs as TGF-␤1 (including myofibrillar re-organization and up-regulation of SMC marker gene expression) and thus may be a novel target involved in producing the downstream effects of TGF-␤1 in SMCs (1,3,16).
The signal transduction pathways through which RhoA elicits its downstream effects in SMCs have yet to be elucidated completely. Two key downstream targets of RhoA are Rho kinase and PKN. Although Rho kinase has been established to participate in the up-regulation of SMC marker gene expression, use of a dominant negative form of Rho kinase cannot completely block the effects of active RhoA (17). Moreover, constitutively active Rho kinase does not completely recapitulate the phenotype seen with active RhoA (18). This implies that there are additional signaling molecules (PKN or others) through which RhoA mediates its downstream effects.
The goal of the present study was to identify the signaling pathway through which TGF-␤1 regulates changes in SMC phenotype, with an emphasis on the control of transcriptional regulation of several SMC marker genes. For these studies, we examined the rat pulmonary arterial smooth muscle cell line, PAC-1, based upon their ability to de-differentiate in response to serum exposure and to differentiate into smooth muscle myocytes upon serum withdrawal (19). Here we demonstrate that treatment with TGF-␤1 induces PAC-1 cells to adopt a more differentiated phenotype, as shown by actin filament organization, reduced proliferation, and up-regulation of SMC marker gene expression. Furthermore, these effects of TGF-␤1 require the activity of the kinases downstream of RhoA. Specifically, our results demonstrate the importance of PKN, one of the downstream targets of RhoA, in mediating the effects of TGF-␤1 on SMC marker gene expression. In addition, we show that PKN stimulates SRF, GATA, and MEF2-dependent transcription. These transcription factors have been demonstrated to be required for SMC differentiation and thus may be important for PKN-mediated activation of the SMC marker gene promoters (1,20,21). Finally, we report that the effects of TGF-␤1 and PKN on SMC marker gene expression require the activity of p38 MAP kinase. Together, these results delineate a novel signaling pathway whereby TGF-␤1 activates PKN and p38 MAP kinase to induce phenotypic changes in PAC-1 SMCs indicative of differentiation.

MATERIALS AND METHODS
Cell Culture and Regents-Rat pulmonary arterial smooth muscle cells (PAC-1) have been described previously (19). PAC-1 cells were cultured in media 199 (Invitrogen) supplemented with 10% fetal bovine serum (Atlanta Biologicals) and gentamicin (Fisher) unless otherwise specified. All experiments were performed on cells between pass 3 and pass 15. TGF-␤1 (used at 2.5 ng/ml) was purchased from Sigma. HA1077 (used at 20 M), Y-27632 (used at 10 M), and SB203580 (used at 10 M) were purchased from Calbiochem.
Immunocytochemistry-PAC-1 cells were cultured on 1.5% gelatincoated coverslips in 12-well tissue culture plates at a starting density of 1 ϫ 10 4 cells/well. 24 h post-plating, cells were pretreated with HA1077, Y-27632, or vehicle control (water) for 3 h prior to the addition of TGF-␤1 or vehicle control (4 mM HCl ϩ 0.1% BSA) for 48 h. As a positive control, cells were stimulated to differentiation by serum withdrawal (media 199 supplemented with 0.5% FBS) for 48 h. Following treatment, cells were fixed in 1:1 acetone:methanol solution. The coverslips were blocked with 5% BSA in HBSS and then incubated in primary antibody (monoclonal anti-␣ smooth muscle actin, Sigma) overnight at 4°C in HBSS ϩ 1% BSA followed by the addition of secondary antibody (AlexaFluor® 594 goat anti-mouse IgG, Molecular Probes), in 1% BSA in HBSS at 37°C for 1 h. Fluorescence was visualized using an Olympus AX70 fluorescent microscope. The fields shown are representative data collected from three independent experiments.
Cell Proliferation Assay-PAC-1 cells were cultured in 6-well tissue culture plates at a starting density of 4 ϫ 10 4 cells per well. 6 h post-plating, cells were pretreated with HA1077, Y-27632, or vehicle control for 3 h. Following pretreatment, TGF-␤1 or vehicle control was added. Total cell number was then determined following 0, 12, 24, 36, and 48 h of TGF-␤1 stimulation using a Coulter cell counter (Beckman Coulter). As a control, cells were stimulated to differentiate by serum withdrawal (media 199 supplemented with 0.5% FBS). Each group was tested in triplicate for each time point per experiment. Data shown are representative of three independent experiments and are expressed as the mean Ϯ S.E.
Transient Transfection and Luciferase Assay-PAC-1 cells were cultured in 12-well tissue culture plates at a starting density of 2.4 ϫ 10 4 cells/well. For TGF-␤1 experiments, PAC-1 cells were transfected 40 h post-plating by using Lipofectamine TM supplemented with Plus TM reagent (Invitrogen) per the manufacturer's instructions. After transfection, cells were pretreated with HA1077 and Y-27632 and treated with TGF-␤1 as described for immunocytochemistry. For all other transfections, PAC-1 cells were transfected 48 h post-plating by using Lipofectamine TM per the manufacturer's instructions (Invitrogen), and luciferase activity was measured 24 h post-transfection. Plasmid concentrations (per well) were as follows: 30 ng of promoter-reporter, 100 ng of enhancer-reporter, 1 or 10 ng of PKN-AF3 and PKN-AF3(K644E) (1 ng for 6xGATA-luc and 4xSRFluc and 10 ng for all other reporters), and 100 ng of p38␣ AF . Total plasmid concentration per well was normalized with empty vector DNA. Luciferase activity was determined by luminometry (Turner Designs Luminometer model 20) using a commercially available substrate kit (Promega). Each data set was independently replicated a minimum of three times with each experimental group tested in triplicate. Luciferase activity is expressed as the mean Ϯ S.E. Initial experiments included co-transfection of an SV40-driven ␤-galactosidase construct as a control for transfection efficiency; however, minimal variability (Ͻ10%) among each independent experiment was seen. Moreover, Shimizu et al. (33) reported that the inclusion of viral driven control vectors alter SMC marker gene promoter activity by potentially sequestering common transcription factors. For these reasons, this vector was excluded from further experiments.
Western Blot for SMC Marker Proteins-PAC-1 cells were cultured in 100-mm tissue culture dishes at a starting density of 2.8 ϫ 10 5 cells per dish. 40 h post-plating, cells were treated with vehicle control or TGF-␤1 for 48 h. Following treatment, cells were lysed, and the relative abundance of SM ␣-actin, SM-MHC, and SM22␣ was determined with commercially available antibodies (Sigma, Biomedical Technologies, Inc., and Abcam, respectively).
RhoA Activity Assay-PAC-1 cells were cultured in 100-mm tissue culture dishes at a starting density of 2.8 ϫ 10 5 cells per dish. 48 h post-plating, cells were treated with vehicle control or TGF-␤1 for 2-60 min. Following treatment, RhoA activity was determined using the EZ-Detect TM Rho activity kit (Pierce) per the manufacturer's instructions.
p38 MAP Kinase Activity Assay-PAC-1 cells were cultured in 100-mm tissue culture dishes seeded at a starting density of 2.8 ϫ 10 5 cells per dish. 48 h post-plating, cells were treated with vehicle control or TGF-␤1 for 2-60 min prior to lysis. The kinase activity of p38 MAPK was examined using a nonradioactive p38 MAPK assay kit (Cell Signaling Technology) per the manufacturer's instructions. In addition, SB203580 (Calbiochem) was used to test the specificity of the kinase activity. Data shown are representative of three independent experiments.
RNAi-The pSUPER-PKN RNAi vector was constructed by ligation of the following oligonucleotide pair to the pSUPER.neoϩgfp vector, which was obtained from Oligoengine, according to manufacturer's instructions: 5Ј-gatccccGATTGACATCATCCGCATGttcaagagaCATGC-GGATGATGTCAATCttttta-3Ј (sense) and 5Ј-agcttaaaaaGATTGACAT-CATCCGCATGtctcttgaaCATGCGGATGATGTCAATCggg-3Ј (antisense). The scramble siRNA negative control was constructed by ligation of the following oligonucleotide pair to the pSUPER.neoϩgfp vector: 5Ј-gatccccACGCACACGTGCATATGTTttcaagagaAACATATGC-ACGTGTCGCtttttta-3Ј (sense) and 5Ј-agcttaaaaaACGCACACGTGCA-TATGTTtctcttgaaAACATATGCACGTGTGCGTggg-3Ј (antisense). For analysis of PKN knock-down, PAC-1 cells were plated in 6-well tissue culture plates at a starting density of 5.0 ϫ 10 4 cells per well. Cells were transfected with 250, 500, or 1000 ng of either PKN siRNA or scramble siRNA using Lipofectamine TM supplemented with Plus reagent (Invitrogen) per the manufacturer's instructions. 48 h following transfection, cells were lysed and subjected to Western blotting to detect PKN (mouse anti-PKN; Pharmingen). For TGF-␤1 experiments, PAC-1 cells were plated in 12-well tissue culture plates at a starting density of 2.4 ϫ 10 4 cells/ml. 40 h post-plating, cells were transfected with 30 ng of SM ␣-actin-luc, SM-MHC-luc, or SM22␣-luc along with 250 ng of either PKN siRNA or scramble siRNA using Lipofectamine TM supplemented with Plus reagent (Invitrogen) per the manufacturer's instructions. Total DNA per well was normalized with empty vector. 48 h posttransfection, cells were treated with either vehicle control or TGF-␤1 for another 48 h prior to luciferase assay.
Statistical Analysis-All results are expressed as the mean Ϯ S.E. Data were analyzed with GraphPad Prism software (version 4.0, GraphPad Software Inc.) using one-way analysis of variance and Bonferroni's post-hoc test for inter-group comparisons. p values Ͻ 0.05 were considered statistically significant.

TGF-␤1 Stimulates Actin Re-organization in PAC-1 Cells-A
role for TGF-␤1 in the differentiation of SMCs has been demonstrated both in vitro and in vivo (6 -10). Although the effects of TGF-␤1 have been tested on several SMC lines and models, the response of PAC-1 cells to TGF-␤1 stimulation has not been characterized previously. To test whether TGF-␤1 induces phenotypic changes in PAC-1 cells toward a more differentiated phenotype, we first examined morphological changes in the actin organization of proliferating cells cultured in the absence or presence of TGF-␤1. In addition, the actin organization of PAC-1 cells stimulated to differentiate by serum withdrawal was examined as a control. PAC-1 cells treated with TGF-␤ 1 for 48 h displayed increased actin organization compared with untreated proliferating cells (Fig. 1A, compare middle panel to left panel). Moreover, PAC-1 cells that had been cultured in low serum for 48 h displayed increased actin organization, similar to TGF-␤1-treated cells (Fig. 1A, compare right panel to middle panel). These data demonstrate that TGF-␤1 stimulates actin re-organization of PAC-1.
TGF-␤1 Slows Proliferation of PAC-1 Cells-Differentiation of SMC is associated with cell cycle arrest and a subsequent reduction in proliferation (34). Therefore, we sought to examine whether TGF-␤1 affects the proliferative capacity of PAC-1 cells. To do this, PAC-1 cells were plated at an equal starting density and allowed to proliferate for 12, 24, 36, or 48 h in the absence or presence of TGF-␤1. Following treatment, the total cell number per group was counted. Cells cultured in the presence of TGF-␤1 displayed reduced proliferation compared with untreated cells, which proliferated in a relatively linear fashion (Fig. 1B). Cells that were stimulated to differentiate by serum withdrawal showed very little proliferation (Fig. 1B). These data show that TGF-␤1 attenuates serum-induced proliferation of PAC-1 cells.
TGF-␤1 Increases Expression of SMC Marker Genes in PAC-1 Cells-To demonstrate further that TGF-␤1 promotes a more differentiated phenotype in PAC-1 cells, we examined changes in the expression of mRNA encoding SMC marker genes in actively proliferating cells cultured in the absence or presence of TGF-␤1 or in PAC-1 cells stimulated to differentiate by serum withdrawal. Untreated proliferating PAC-1 cells showed only low level expression of the three definitive SMC marker genes, SM ␣-actin, SM-MHC, and SM22␣, as measured by RT-PCR (Fig. 1C). In contrast, cells treated with TGF-␤1 for 48 h displayed increased expression of all three SMC marker genes, in a manner similar to cells stimulated to differentiate by serum withdrawal (Fig. 1C). In addition to measuring the relative expression of the SMC marker genes by RT-PCR, we tested whether TGF-␤1 alters the activity of the promoters that control the expression of these genes. Indeed, the activity of all three SMC marker gene promoters was increased in PAC-1 cells treated with TGF-␤1 (Fig. 1D), arguing that TGF-␤1 increases the expression of these genes at the level of transcription. Finally, protein levels of SM ␣-actin, SM-MHC, and SM22␣ were increased in TGF-␤1 treated PAC-1 cells compared with vehicle-treated cells (Fig. 1E). When combined, the results from Fig. 1 strongly argue that TGF-␤1 stimulates PAC-1 cells to adopt a more differentiated phenotype.
TGF-␤1 Effects Require RhoA Signaling-To determine whether RhoA signaling is required for the effects of TGF-␤1 on PAC-1 cells, we utilized two chemical inhibitors, HA1077 and Y-27632. These inhibitors have been reported recently to block specifically the function of the RhoA-associated kinases Rho kinase and PKN/PRK2 (35)(36)(37). PAC-1 cells were pretreated with HA1077 or Y-27632 prior to the addition of TGF-␤1, and the relative actin re-organization, proliferation rate, and expression of SMC marker genes of these cells were compared with proliferating PAC-1 cells or to cells stimulated to differentiate by the addition of TGF-␤1 alone. Pretreatment of cells with either HA1077 or Y-27632 attenuated TGF-␤1-induced actin re-organization ( Fig. 2A). In addition, pretreatment with either inhibitor prevented TGF-␤1-induced reduction of cell proliferation, restoring total cell number at each time point to near control values (Fig. 2B). Finally, pretreatment with HA1077 or Y-27632 blocked TGF-␤1-mediated up-regulation of SM ␣-actin, SM-MHC, and SM22␣ as assessed by RT-PCR ( TGF-␤1 Activates RhoA and PKN-Data from Fig. 2 argue that the downstream targets of RhoA are required for TGF-␤1induced changes in the phenotype of PAC-1 cells. To determine directly whether TGF-␤1 activates RhoA, PAC-1 cells were treated with TGF-␤1 for various times up to 60 min, and GTP-bound (active) RhoA was pulled down from the whole cell lysate by using a glutathione S-transferase-rhotekin Rho binding domain fusion protein. Two minutes of TGF-␤1 treatment was sufficient to increase the relative amount of GTP-bound RhoA compared with untreated cells (Fig. 3A). RhoA activity remained elevated through 5 min of TGF-␤1 stimulation and then quickly declined. Thus, TGF-␤1 activates RhoA in a rapid and transient manner in PAC-1 cells.
Next, we examined whether TGF-␤1 activates PKN, one of the downstream targets of RhoA. Again, PAC-1 cells were treated with TGF-␤1 for various times up to 60 min. Following treatment, the phosphorylation status of PKN was measured by Western blot analysis by using an antibody that recognizes the active form of PKN (phospho-Thr-778) and its closely related family member, PRK2 (phospho-Thr-816). Both proteins were phosphorylated following 5 min of treatment with TGF-␤1 and remained elevated through 60 min (Fig. 3B). The total quantity of PKN was not altered by TGF-␤1 treatment (Fig.  3B). Taken together, results from Fig. 3 demonstrate that TGF-␤1 activates both RhoA and its downstream target, PKN, in PAC-1 cells.
PKN Regulates Transcriptional Activation of SMC Marker Genes-To determine whether active PKN alone (independently of other signals activated by TGF-␤1) is sufficient to stimulate transcription of SMC marker genes, PAC-1 cells were co-transfected with one of three SMC marker gene promoterreporter constructs (SM ␣-actin, SM-MHC, or SM22␣), and an expression vector encoding constitutively active PKN (PKN-AF3, which lacks the regulatory domain). Expression of active PKN increased the transcriptional activity of these promoters in a dose-dependent manner (data not shown). In addition, the kinase activity of PKN was required for this effect, as expression of active, but not a kinase-dead form of PKN (K644E, which contains a point mutation in the ATP-binding domain), stimulated the transcriptional activity of the SM ␣-actin (Fig.  4A), SM-MHC (Fig. 4B), and SM22␣ (Fig. 4C) promoters ϳ3fold over empty vector control. These data argue that activation of PKN is sufficient to activate transcription of SMC marker genes.
PKN Increases the Transcriptional Activity of SRF, GATA, and MEF2-SRF is by far the single most important transcription factor characterized to date to control the expression of SMC marker gene expression. However, proper expression of SMC marker genes requires the activation of multiple transcription factors that cooperate and interact in concert to drive tissue-specific expression of these genes (3,38). SRF, GATA, and MEF2 are three transcription factors that have each been characterized to play a role in the proper expression of the SMC marker genes (20, 24, 39 -41). Thus, we sought to examine whether PKN might regulate the overall activity of these transcription factors, in order to increase the expression of the SMC marker genes. For these studies, PAC-1 cells were co-transfected with active PKN along with an SRF-dependent enhancer-reporter (containing four tandem CArG box elements), a GATA-dependent enhancer-reporter (containing six tandem GATA consensus binding sites), or a MEF2-dependent enhancer-reporter (containing three tandem MEF2 consensus binding sites). These enhancer-reporters are sensitive to changes in the relative abundance, DNA binding, and post-translational mod-ifications that alter the transcriptional activity of SRF, GATA, and MEF2, respectively. Thus, although these enhancer-reporters cannot provide insight to the mechanism of activation of the respective transcription factor, they do provide a sensitive readout of the overall transcriptional activity of each protein. Expression of active PKN stimulated all three enhancerreporters in a dose-dependent manner (data not shown). Moreover, this effect was dependent upon the kinase activity of PKN as expression of active PKN, but not the kinase-dead mutant, increased the activity of the SRF (Fig. 5A), GATA (Fig.  5B), and MEF2 (Fig. 5C) enhancer-reporters ϳ3-fold over empty vector control. These data show that PKN increases the overall activity of the transcription factors SRF, GATA, and MEF2, which may be important for PKN-mediated activation of SMC marker gene expression.
TGF-␤1 Activation of the SMC Marker Gene Promoters Requires PKN-As discussed previously, TGF-␤1 can activate multiple downstream signaling pathways including Smads and RhoA. Moreover, there are multiple targets downstream of RhoA that may mediate the downstream effects of TGF-␤1 on SMC marker gene expression. To determine the relative contribution of PKN to TGF-␤1-mediated activation of the SMC marker gene promoters, we used RNAi to specifically knock-out PKN expression. As shown in Fig. 6A, the PKN siRNA reduced the expression of PKN, whereas the scrambled control siRNA had no effect. Moreover, knock-out of PKN using the PKN siRNA completely blocked the ability of TGF-␤1 to activate each of the three SMC marker gene promoters tested (Fig. 6B). These data strongly argue that PKN is required for TGF-␤1mediated up-regulation of SMC marker gene expression.
TGF-␤1 Alters MAP Kinase Phosphorylation-MAP kinases play an integral role in mediating intracellular signaling. They are involved in numerous cellular processes such as growth, proliferation, differentiation, and apoptosis. There are three distinct families of MAP kinases (ERKs, p38 MAP kinases, and JNKs) (42). PKN has been shown previously to bind and activate p38 MAP kinase in other cell types (43,44). Therefore, we sought to examine the activity of endogenous MAP kinase activity following TGF-␤1 treatment, when PKN is active. PAC-1 cells were stimulated with TGF-␤1 for various times up to 60 min, and the phosphorylation status of ERK, JNK, and p38 MAP kinase was measured by Western blot analysis using antibodies that detect the active (phosphorylated) forms of these kinases. Phosphorylation of ERK1/2 (p44/p42) and JNK1/2 (p46/p54) decreased in a time-dependent manner following TGF-␤1 treatment (Fig. 7). There was no significant change in the total cellular quantity of these proteins, demonstrating a true decrease in their phosphorylation status, indicative of decreased activity. In contrast, phosphorylation of p38 MAP kinase increased within 5-10 min of TGF-␤1 treatment without change in the total expression of this protein (Fig. 7), indicative of increased activity. We also directly tested the activity of p38 MAP kinase following TGF-␤1 treatment using an in vitro kinase assay with ATF-2 as the substrate. In agreement with the data shown in Fig. 6, the activity of the p38 MAP kinase was increased within 5-10 min of TGF-␤1 treatment (Fig. 8). Based on these results, p38 MAP kinase may play an important role in mediating the downstream effects of TGF-␤1.
TGF-␤1 Activation of the SMC Marker Gene Promoters Requires p38 MAP Kinase-To test whether p38 MAP kinase is required for TGF-␤1-induced activation of the SMC marker gene promoters, we used a specific p38 MAP kinase inhibitor, SB203580, to block p38 MAP kinase activity. PAC-1 cells were transfected with the SM ␣-actin, SM-MHC, or SM22␣ promoter-reporter plasmids and were treated with 10 M SB203580 for 1 h prior to TGF-␤1 treatment. Pretreatment of cells with SB203580 completely blocked the ability of TGF-␤1 to activate the SMC marker gene promoters, demonstrating the necessity of this enzyme for TGF-␤1-mediated effects on SMC marker gene expression (Fig. 9).
Dominant Negative p38 MAP Kinase Blocks PKN-mediated Effects-To determine whether p38 MAP kinase is required for PKN-mediated activation of SMC marker gene expression, we examined whether a dominant negative p38 MAP kinase mutant could block the ability of PKN to activate transcription. PAC-1 cells were transfected with active PKN in the absence or presence of dominant negative p38␣ MAP kinase, and the activity of the SRF, GATA, or MEF2 enhancers or the SMC marker gene promoters was measured. PKN enhanced the activity of each enhancer or promoter, as expected. Co-expression of the dominant negative p38 MAP kinase abolished the ability of PKN to enhance transcription. This demonstrates the necessity of p38 MAP kinase for PKN-mediated activation of both the SMC marker gene promoters (Fig. 10A) and the SRF, GATA, and MEF2 enhancers (Fig. 10B). DISCUSSION A role for TGF-␤1 in the differentiation of SMCs has been demonstrated both in vitro and in vivo (6 -10). However, despite the importance of TGF-␤1 on SMC phenotype, very little is known regarding the molecular mechanisms through which TGF-␤1 elicits its downstream effects. Of interest, it has been demonstrated recently that TGF-␤1 can activate the small GTPase RhoA in epithelial cells undergoing epithelial-to-mesenchymal transformation, which results in a myofibroblast-like phenotype because of the up-regulation of SM ␣-actin expression in these cells (12,14). Moreover, TGF-␤1 and RhoA are known to produce similar phenotypic changes in SMCs, causing them to adopt a more differentiated phenotype (16,45). When taken together, these data support the hypothesis that TGF-␤1 may mediate its downstream effects on SMC differentiation through the activation of a RhoA-driven signaling cascade. The goals of the present study were as follows: (i) demonstrate that the effects of TGF-␤1 on SMC phenotype requires RhoA signaling and (ii) evaluate the specific roles of PKN and p38 MAP kinase on SMC marker gene expression.
Data from the present study demonstrate that TGF-␤1 promotes PAC-1 smooth muscle cells to adopt a more differentiated phenotype. TGF-␤1 stimulated SM ␣-actin re-organization and inhibited serum-induced proliferation of PAC-1 cells. Despite the fact that TGF-␤1 could reduce the rate of proliferation of PAC-1 cells, TGF-␤1 was not sufficient to completely block all of the mitogenic signals present in serum, as evidenced by the higher rate of proliferation of TGF-␤1-treated PAC-1 cells compared with cells simulated to differentiate by serum withdrawal. Finally, TGF-␤1 up-regulated the expression of SMC marker genes in PAC-1 cells, presumably by increasing the activity of the promoters that regulate their expression resulting in an overall increase in the relative abundance of these proteins. However, it is unclear whether the level of SMC marker gene expression achieved in this cultured SMC model is comparable with that found in SMCs in vivo within intact tissue.
Through the use of two chemical inhibitors, HA1077 and Y-27632, TGF-␤1-induced changes in PAC-1 SMC phenotype were shown to require the activity of the main downstream targets of RhoA. These inhibitors, although marketed as Rho kinase-specific inhibitors, have been shown to inhibit the kinase activity of Rho kinase and PKN/PRK2 (35)(36)(37). Thus, although these inhibitors cannot distinguish between these proteins, their use can provide direct evidence that one or more of these kinases are important for mediating the downstream functions of TGF-␤1. Because both HA1077 and Y-27632 blocked the effects of TGF-␤1 on PAC-1 SMC phenotype, RhoA and its downstream signaling targets must play an important role mediating the downstream responses of TGF-␤1 on SMC phenotypic changes.
TGF-␤1 has been shown recently to stimulate RhoA activity in multiple cell types, although to date a link between TGF-␤1 and RhoA in SMCs has not been reported (12)(13)(14)(15). Supporting the data collected from the use of HA1077 and Y-27632, TGF-␤1 stimulation of PAC-1 cells increased the amount of active, GTP-bound RhoA. This occurred in a time-dependent manner and was transient in nature. The mechanism of how TGF-␤1 increased RhoA activity is unclear. However, Shen et al. (15) reported recently that treatment of Swiss 3T3 cells with TGF-␤1 led to an increase in expression of the Rho-GEF, NET1, resulting in increased RhoA activity. The authors also reported that this mechanism required Smad-dependent up-regulation of NET1 gene expression. Because of the rapid activation of RhoA in response to TGF-␤1 in our studies, it is not likely that this same mechanism occurs in PAC-1 cells in response to TGF-␤1. However, it remains possible that TGF-␤1 can increase the activity of a Rho-GEF, such as NET1, which would be consistent with the rapid activation of RhoA presented here. Further studies will be needed to assess directly the mechanism of TGF-␤1 activation of RhoA in SMCs.
In addition to RhoA activity, phosphorylation of the activation loop of PKN and PRK2 was also increased in a time-dependent manner by TGF-␤1. Because PKN/PRK2 requires the binding of GTP-bound RhoA for their activation, it correlates well that the phosphorylation of PKN/PRK2 appears immediately after the increase in GTP-bound RhoA occurs. Although the remainder of this study focused on the downstream effects of PKN, it is likely that PKN and PRK2 may have overlapping roles in this system because both are activated in a similar manner in response to TGF-␤1. Further studies will be needed FIG . 5. PKN increases the activity of the transcription factors SRF, GATA, and MEF2 in a kinase-dependent manner. PAC-1 cells were transfected with the 4xSRF-(A), 6xGATA-(B), or 3xEF2-luciferase (C) enhancer-reporter plasmids along with either empty vector control (Control), active PKN (PKN-AF3), or kinase-dead PKN (K644E) using Lipofectamine TM reagent. 12 h post-transfection, cells were re-fed with media supplemented with 10% FBS. Luciferase activity was measured 24 h after re-feeding (*, p Ͻ 0.001 compared with control; †, p Ͻ 0.001 compared with PKN-AF3).
FIG. 6. PKN is required for TGF-␤1-mediated activation of the SMC marker gene promoters. A, PAC-1 cells were transfected with control vector (pSUPER.neoϩgfp) or increasing amounts of either PKN siRNA or scramble siRNA using Lipofectamine TM and Plus TM reagents. 48 h post-transfection, cells were lysed, and total PKN was measured by Western blot. Following PKN detection, membranes were stripped, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was measured to ensure equal protein loading. B, PAC-1 cells were transfected with 250 ng/well of control vector (pSUPER.neoϩgfp), PKN siRNA, or scramble siRNA using Lipofectamine TM and Plus TM reagents. Following 3 h of transfection, cells were re-fed with media supplemented with 10% FBS and incubated for 48 h to ensure PKN knock-down. Cells were then treated with TGF-␤1 or vehicle control for 48 h prior to luciferase assay (*, p Ͻ 0.001 compared with control; †, p Ͻ 0.001 compared with TGF-␤1 alone).
to determine the relative similarities and differences between PKN and PRK2 with respect to their effects on SMC phenotype.
Corresponding with its increased activity following TGF-␤1 stimulation, the specific activation of PKN alone was capable of regulating SMC marker gene expression. Constitutively active PKN trans-activated the SM ␣-actin, SM-MHC, and SM22␣ promoters in a dose-and kinase-dependent manner. Because the kinase activity of PKN was required for its ability to increase promoter activity, PKN most likely elicits its effects through the activation of other downstream targets (such as other signaling molecules or transcription factors). Accordingly, PKN was also capable of increasing the overall activity of the transcription factors SRF, GATA, and MEF2 in a dose-and kinase-dependent manner. Although there are no SMC-specific transcription factors, most SMC marker genes contain one or more consensus SRF-binding sites (CArG box) within the promoter region, and it is well accepted that these sites are required for proper expression of these genes in SMCs (24,39,46,47). In addition, expression and activity of GATA-6, the prevalent isoform of GATA in SMCs, is regulated by SMC phenotype. Suzuki et al. (48) reported that expression of GATA-6 is rapidly down-regulated when SMCs are stimulated to proliferate. Mano et al. (20) further expanded these results by demonstrating that restoration of GATA-6 expression following balloon injury of rat carotid artery lessened neointimal formation and restored proper SMC marker gene expression. Finally, Lin et al. (21) demonstrated that MEF2 is required for normal SMC differentiation during development, and Katoh et al. (41) reported that MEF2 is important for SM-MHC expression, implying that MEF2 may also participate in SMC differentiation. It is also interesting to note that RhoA signaling has been reported to regulate the activity and/or expression of each of these transcription factors (16,49,50). Mack et al. (16) reported that the effect of RhoA on SRF activity could be blocked by the use of Y-27632, which they conclude to be due to the inhibition of Rho kinase. However, based on the data presented here, the effects of RhoA could also have been caused by activation of PKN. When taken together, the ability of PKN to trans-activate the SMC marker gene promoters as well as increase the overall activity of these key transcription factors strongly argues that this signaling pathway is important for promoting expression of SMC marker genes in SMCs.
Because both TGF-␤1 and RhoA can activate multiple downstream targets, we wanted to assess the specific contribution of PKN to TGF-␤1-mediated activation of SMC marker gene expression. This was accomplished through the use of RNAi to specifically inhibit PKN expression. The PKN siRNA reduced PKN expression in PAC-1 cells, which completely blocked the effect of TGF-␤1 on the SMC marker gene promoters. These results imply that PKN is the critical signaling pathway through which TGF-␤1 activates SMC marker gene expression. However, it is important to note that whereas the promoters used for these studies do contain the most critical transcription factor binding sites (such as the main CArG boxes), they are not the full-length promoters required for in vivo expression of these genes and thus may not completely recapitulate the expression of the endogenous SMC marker genes found in SMCs in vivo (39,46). From the studies presented here, we can conclude that PKN obviously plays a role in mediating the effects of TGF-␤1; however, further studies are needed to assess the relative contribution of PKN as well as other potential downstream targets of TGF-␤1 (including PRK2, Rho kinase, and Smads) to TGF-␤1-mediated activation of SMC marker gene expression in vivo.
Very little is known regarding the signal transduction pathway(s) through which PKN elicits its downstream effects on transcription factor activation and gene expression. However, Marinissen et al. (43) demonstrated that PKN alters gene expression though activation of p38 MAP kinase in NIH-3T3 and HEK 293 cells. These results were extended by Takahashi et al. (44) by demonstrating that PKN directly binds to p38 MAP kinase, resulting in increased p38 MAP kinase activity. Here we show that the activity of p38 MAP kinase is increased following TGF-␤1 treatment, which correlates with the increase in PKN phosphorylation seen under the same conditions. Moreover, inhibition of p38 MAP kinase with SB203580 blocked the effect of TGF-␤1 on the SMC marker gene promoters, showing the necessity of p38 MAP kinase for these events. Finally, inhibition of p38 MAP kinase (through the use of dominant negative p38 MAP kinase) blocked the ability of PKN FIG. 7. TGF-␤1 increases phosphorylation of p38 MAP kinase, while decreasing the phosphorylation of ERK and JNK. PAC-1 cells were cultured in media supplemented with 10% FBS. Cells were untreated (control) or treated with TGF-␤1 for 2, 5, 10, 30, or 60 min. Following treatment, the cells were lysed, and the phosphorylation status of ERK1/2, JNK1/2, and p38 MAP kinase was analyzed by Western blot. In addition, total ERK1/2, JNK1/2, and p38 MAP kinase was also analyzed by Western blot to detect any changes in the expression of these proteins following TGF-␤1 treatment.
FIG. 8. TGF-␤1 increases the kinase activity of p38 MAP kinase. PAC-1 cells were cultured in media supplemented with 10% FBS. Cells were untreated (control) or treated with TGF-␤1 for 2, 5, 10, 30, or 60 min. Following treatment, active, phosphorylated p38 MAP kinase was immunoprecipitated from whole cell lysate using a phospho-p38specific antibody conjugated to agarose beads. The relative activity of p38 MAP kinase following TGF-␤1 stimulation was then determined by in vitro kinase assay using an ATF-2 peptide as the substrate. Phosphorylated ATF-2 was analyzed by Western blot. In addition, SB203580 was used to inhibit p38 MAP kinase activity during the in vitro kinase assay to ensure the specificity.
to activate the SMC marker gene promoters as well as the transcription factor enhancers. Although p38 MAP kinase has not been characterized previously to play a role in SMC differentiation, it is known to be critical for skeletal muscle differentiation through activation of MEF2-dependent gene transcription (51). Moreover, p38 MAP kinase has been shown to phosphorylate directly and activate GATA in cardiomyocytes and is associated with an increase in SRF-dependent transcription in SMCs corresponding to up-regulation of SM ␣-actin gene expression (49,52). Based on these reports and the data presented here, it is easy to speculate that p38 MAP kinase may play an important role in the regulation of muscle-specific genes in all muscle types.
The p38 MAPK family of enzymes represents only one branch of the MAP kinase signaling cascade. We also examined whether TGF-␤1 affects the activity of the other MAP kinase pathways, ERK and JNK. Activation of the ERK and JNK branches of MAP kinase signaling has been shown to be associated with growth and proliferation in multiple cell types, including SMCs. Multiple researchers have reported that ERK and JNK proteins are rapidly activated following balloon injury (53,54). Moreover, Izumi et al. (55) found that inhibition of ERK and JNK blocked neointimal formation resulting from balloon injury by reducing proliferation of the underlying SMCs. In addition, they found that inhibiting ERK and JNK also attenuated serum-induced SMC proliferation in vitro. In agreement with these previous studies, we found that the activity of ERK1/2 and JNK1/2 was highest in actively proliferating PAC-1 cells. Furthermore, TGF-␤1 reduced the phosphorylation of ERK1/2 and JNK1/2, indicating a decrease in their activities. These results are consistent with our model of TGF-␤1-induced differentiation of SMCs. It is not clear whether TGF-␤1-induced activation of p38 MAP kinase plays a direct role in inhibiting ERK and JNK activity in SMCs or whether other signaling pathways downstream of TGF-␤1 mediate this effect. However, it is interesting to note that p38 MAP kinase has been shown to negatively regulate angiotensin II-and platelet-derived growth factor-induced activation of ERK as well as increases in expression of cyclin D1 and DNA synthesis in SMCs (56,57).
In summary, our results provide strong evidence that the small GTPase RhoA and its downstream target PKN play an important role in TGF-␤1-mediated changes in the SMC phenotype. Although PKN may have many other functions in SMCs that are not yet well established, it is clear that PKN is involved in TGF-␤1-mediated up-regulation of the SMC marker genes, SM ␣-actin, SM-MHC, and SM22␣, presumably through activation of the SRF, GATA, and MEF2 families of transcription factors. Finally, p38 MAP kinase is required for mediating the downstream effects of TGF-␤1 and PKN on the up-regulation of these genes. These data demonstrate, for the first time, the existence of a novel signaling cascade through which TGF-␤1 modulates changes in SMC phenotype. It is easy to envision a linear signaling pathway involving RhoA/PKN-mediated activation of p38 MAPK, which in turn activates SRF, GATA, and MEF2 leading to up-regulation of SMC marker gene expression. However, other downstream targets of TGF-␤1 and RhoA, such as Rho kinase and Smads, may also play a role in mediating these effects on SMCs (Fig. 11). We have identified PKN as an important component of this complex signaling pathway. Understanding how PKN regulates SMC phenotype in cooperation with other downstream targets of RhoA will provide insight into potential therapeutic targets designed to prevent the aberrant proliferation of SMCs that contributes to the pathophysiology of vascular disease.