ATF-1 mediates protease-activated receptor-1 but not receptor tyrosine kinase-induced DNA synthesis in vascular smooth muscle cells.

Previously we have demonstrated that activation of p38 mitogen-activated protein kinase (MAPK) and induction of DNA synthesis in response to receptor tyrosine kinase (RTK) and G protein-coupled receptor (GPCR) agonists require NADH/NADPH-like oxidase activity in vascular smooth muscle cells (VSMC). Here we tested the role of p38 MAPK in RTK and GPCR agonist-induced DNA synthesis in VSMC. Platelet-derived growth factor (PDGF)-BB and thrombin (RTK and GPCR agonists, respectively) activated p38 MAPK in a time-dependent manner in VSMC. Inhibition of p38 MAPK led to a 50% decrease in the DNA synthesis induced by thrombin but not PDGF-BB. ATF-1 was found to be the predominant member of the cyclic AMP response element (CRE)-DNA complex formed in VSMC in response to PDGF-BB and thrombin, and both agonists induced its phosphorylation. Regardless of this, inhibition of p38 MAPK reduced only thrombin- but not PDGF-BB-induced ATF-1 phosphorylation. Similarly, inhibition of p38 MAPK caused a 50% decrease in thrombin- but not PDGF-BB-induced CRE promoter-dependent transcription. Ectopic expression of an inhibitory anti-ATF-1 single-chain antibody fragment, ScFv, significantly interfered with DNA synthesis induced by thrombin but not PDGF-BB. Together, these results suggest the following conclusions. 1) Both RTK and GPCR agonists activate p38 MAPK and induce CRE promoter-dependent transcription; 2) both RTK and GPCR agonists induce ATF-1 phosphorylation, and ATF-1 is a predominant member in the CRE-DNA complexes formed in response to these agents; and 3) p38 MAPK-dependent ATF-1 phosphorylation and CRE promoter-mediated transcription are associated with GPCR agonist-induced VSMC growth.

Increased vascular smooth muscle cell growth is a contributing factor in the pathogenesis of atherosclerosis and restenosis (1). Increased levels of mitogens such as platelet-derived growth factor (PDGF) 1 and fibroblast growth factor (FGF) were reported in atherosclerotic arteries compared with normal (1)(2)(3)(4). These mitogens modulate VSMC growth both in an autocrine and paracrine manner in VSMC (1). In addition, in vitro studies have shown that a variety of receptor tyrosine kinase (RTK) and G protein-coupled receptor (GPCR) agonists and of oxidants are potent mitogens to VSMC (5)(6)(7)(8)(9)(10), suggesting that many of the molecules that are produced at the site of arterial injury by various cell types can account for the increased VSMC growth during the formation of these lesions. Towards understanding the role of specific mitogens in the induction of VSMC growth during the formation of these lesions, several investigators have used neutralizing antibodies and antisense oligonucleotides (4,7,10,11). Use of neutralizing antibodies against PDGF or FGF resulted only in partial inhibition of VSMC growth and lesion progression (4,10). Similarly, the use of antisense oligonucleotides against a molecule that appears to be critical in the mitogenic signaling events, such as c-Myc, resulted only in partial inhibition of VSMC growth and lesion progression (11). Thus, inhibition of individual mitogens or proto-oncogenes has so far resulted in minimal reduction in VSMC growth and lesion progression in animal models of atherosclerosis and restenosis.
Identification of molecules that are critical in the signaling pathways of several mitogens may provide more successful targets. Mitogen-activated protein kinases (MAPK) are a group of serine/threonine kinases that are ubiquitously expressed (12)(13)(14). These are grouped primarily into three major categories, namely 1) extracellular signal-regulated kinases, 2) Jun N-terminal kinases, and 3) p38 MAPKs (13,14). Furthermore, differences were observed in the responsiveness of different groups of MAPKs to various external stimuli. Specifically, extracellular signal-regulated kinases have been reported to respond preferentially to agents that induce cell growth and differentiation (15)(16)(17)(18)(19)(20), whereas Jun N-terminal kinases and p38 MAPK have been reported to be potently activated by cellular stressors and cytokines (21)(22)(23)(24). Despite these differences in responsiveness, cross-talk between different groups of MAPK has been observed in mediating cellular responses to certain agonists (25). In addition, activation of one or all three groups of MAPKs are involved in the regulation of activities of various transcriptional factors, including activator protein 1 (AP-1) and nuclear factor B (NFB) (12). Previously we have demonstrated that activation of p38 MAPK and induction of DNA synthesis in response to RTK and GPCR agonists require NADH/NADPH-like oxidase activity in VSMC (26). The purpose of the present investigation was to study the role of p38 MAPK in the induction of growth by both RTK and GPCR agonists in VSMC. We show that 1) both RTK and GPCR agonists activate p38 MAPK in VSMC; 2) although activation of p38 MAPK is not required for PDGF-BB-induced DNA synthesis, it is involved in thrombin-induced DNA synthesis; 3) thrombin-but not PDGF-BB-induced p38 MAPK activation is required for ATF-1 phosphorylation and CRE promoter-dependent transcription; 4) thrombin-stimulated p38 MAPK-dependent ATF-1 phosphorylation and CRE promoter-mediated transcription are associated with growth in VSMC; and 5) disruption of ATF-1 activity by intracellular expression of an inhibitory anti-ATF-1 single-chain antibody fragment, ScFv, interfered with only thrombin but not PDGF-BB-induced DNA synthesis in VSMC. Together, these findings reveal that ATF-1 plays a role in thrombin-but not PDGF-BB-induced VSMC growth.

MATERIALS AND METHODS
Reagents-Acetyl-coenzyme A, aprotinin, dithiothreitol, HEPES, leupeptin, phenylmethylsulfonyl fluoride, poly(dI-dC), sodium orthovanadate, sodium deoxycholate, and thrombin were purchased from Sigma. Recombinant human PDGF-BB was bought from R&D Systems Inc. Cell Culture-VSMC were isolated from the thoracic aortas of 200 -300-g male Sprague-Dawley rats by enzymatic dissociation as described previously (5). Cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS), 100 units/ml penicillin, and 100 g/ml streptomycin. Cultures were maintained at 37°C in a humidified 95% air and 5% CO 2 atmosphere. Cells were growth-arrested by incubating in DMEM containing 0.1% (v/v) FBS for 72 h and used to perform the experiments unless otherwise stated.
DNA Synthesis-VSMC with and without appropriate treatments were pulse-labeled with 1 Ci/ml [ 3 H]thymidine for the indicated times. After labeling, cells were washed with cold PBS, trypsinized, and collected by centrifugation. The cell pellet was suspended in cold 10% (w/v) trichloroacetic acid and vortexed vigorously to lyse cells. After standing on ice for 20 min, the cell lysis mixture was passed through a glass fiber filter (GF/C, Whatman). The filter was washed once with cold 5% trichloroacetic acid and once with cold 70% (v/v) ethanol. The filter was dried and placed in a liquid scintillation vial containing the scintillant fluid, and the radioactivity was measured in a liquid scintillation counter (LS 5000TA, Beckman).
Electrophoretic Mobility Shift Assay-Nuclear extracts were prepared from VSMC that were subjected to appropriate treatments or left untreated as described previously (26). Protein-DNA complexes were formed by incubating 5 g of nuclear protein in a total volume of 20 l consisting of 15 mM HEPES, pH 7.9, 3 mM Tris-HCl, pH 7.9, 60 mM KCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 4.5 g of bovine serum albumin, 2 g of poly(dI-dC), 15% glycerol, and 100,000 cpm of 32 P-labeled oligonucleotide probe for 30 min on ice. The protein-DNA complexes were then resolved by electrophoresis on a 4% polyacrylamide gel using 1ϫ Tris-glycine-EDTA buffer (25 mM Tris-HCl, pH 8.5, 200 mM glycine, 0.1 mM EDTA). Double-stranded oligonucleotides were labeled with [␥-32 P]ATP using T4 polynucleotide kinase kit following the supplier's protocol (Invitrogen).
Transient Transfection and Reporter Gene Assay-VSMC were plated evenly onto 60-mm dishes the day before transfection and grown in DMEM containing 10% (v/v) heat-inactivated FBS, 100 units/ml penicillin, and 100 g/ml streptomycin. Cells were transfected with appropriate plasmid DNA (20 g/60-mm dish) using calcium phosphate precipitation as described previously (26). Cells were washed with PBS 16 h after transfection and incubated in DMEM containing 0.1% (v/v) FBS for 36 h at 37°C. Cells were then treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for the indicated times, and cell lysates were prepared. Cell lysates normalized for protein were assayed for either CAT activity using [ 14 C]chloramphenicol and acetylcoenzyme A as substrates or luciferase activity using Luciferase assay system (Promega) and a Turner luminometer (TD-20/20).
Western Blot Analysis-After appropriate treatments, VSMC were rinsed with cold phosphate-buffered saline (PBS) and frozen immediately in liquid nitrogen. Cells were lysed by thawing in 250 l of lysis buffer (PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 100 g/ml phenylmethylsulfonyl fluoride, 100 g/ml aprotinin, 1 g/ml leupeptin, and 1 mM sodium orthovanadate) and scraped into 1.5-ml Eppendorf tubes. After standing on ice for 20 min the cell lysates were cleared by centrifugation at 12,000 rpm for 15 min at 4°C. The protein content of the supernatants was determined using Micro BCA protein assay reagent kit (Pierce). Cell lysates containing equal amounts of protein were resolved by electrophoresis on 0.1% SDS and 10% polyacrylamide gels. The proteins were transferred electrophoretically to a nitrocellulose membrane (Hybond, Amersham Biosciences). After blocking in 10 mM Tris-HCl buffer, pH 8.0, containing 150 mM sodium chloride, 0.1% Tween 20, and 5% (w/v) nonfat dry milk, the membrane was treated with appropriate primary antibodies followed by incubation

FIG. 1. PDGF-BB and thrombin activate p38 MAPK in VSMC.
Panel A, equal amounts of protein from VSMC that were treated for the indicated times with PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) or left untreated were analyzed by Western blotting for p38 MAPK using its phosphospecific antibodies. Panel B, equal amounts of protein from VSMC that were treated with PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for 30 min or left untreated were immunoprecipitated with anti-p38 MAPK antibodies, and the immunocomplexes were divided into two halves and subjected to immunocomplex kinase assay in the presence and absence of SB203580 (10 M) using [␥-32 P]ATP and recombinant ATF-2 protein as substrates. After termination of the reaction, the assay mixture was separated by electrophoresis on 0.1% SDS, 12% acrylamide gel and subjected to autoradiography.
with horseradish peroxidase-conjugated secondary antibodies. The antigen-antibody complexes were detected using a chemiluminescence reagent kit (Amersham Biosciences).
Statistics-All the experiments were repeated at least three times with similar results. Data on luciferase activity and [ 3 H]thymidine incorporation were presented as mean Ϯ S.D., and the treatment effects were analyzed by Student's t test. p values Ͻ 0.05 were considered to be statistically significant. In the case of CAT assay, EMSA, and Western blot analysis, one representative set of data is shown.

RESULTS
Previously we have demonstrated that activation of p38 MAPK and induction of DNA synthesis in response to RTK and GPCR agonists require NADH/NADPH-like oxidase activity in VSMC (26). Here we have tested the role of p38 MAPK in RTK and GPCR agonist-induced VSMC growth. Lysates of VSMC that were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) were analyzed by Western blotting for phosphorylated p38 MAPK using its phosphospecific antibodies. Both PDGF-BB and thrombin activated p38 MAPK in VSMC in a time-dependent manner with a maximum effect of 10-fold at 30 min and declining thereafter (Fig. 1A). To confirm these observations, lysates of VSMC that were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for 30 min were immunoprecipitated with anti-p38 MAPK antibodies and the immunocomplexes were assayed for p38 MAPK activity using recombinant ATF-2 and [␥-32 P]ATP as substrates in the presence and absence of SB203580 (10 M), a potent inhibitor of p38 MAPK. As expected, both PDGF-BB and thrombin increased p38 MAPK activity, and it was completely inhibited by SB203580 (Fig. 1B). To learn whether p38 MAPK plays a role in PDGF-BB-and thrombin-induced VSMC growth, we tested the effect of SB203580 on DNA synthesis induced by these agonists. Quiescent VSMC were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M), and DNA synthesis was measured by [ 3 H]thymidine incorporation into trichloroacetic acid-precipitable material. PDGF-BB and thrombin stimulated VSMC DNA synthesis 3-5-fold, and SB203580 inhibited 50% of the thrombin-induced, but not PDGF-BB-induced, DNA synthesis (Fig. 2).
Earlier studies from other laboratories have reported that p38 MAPK plays a role in c-Fos gene induction in response to cytokines and ultraviolet B light in some cell types (28 -30). Therefore, to understand the mechanism by which p38 MAPK participates in thrombin-induced DNA synthesis, we studied its role in PDGF-BB-and thrombin-induced expression of the Fos and Jun family proteins. Lysates of VSMC that were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M) for 2 h were analyzed for Fos and Jun family proteins using their specific antibodies. PDGF-BB and thrombin induced expression of c-Fos, Fra-1, c-Jun, and Jun-B by 10-to more than 100-fold (Fig. 3). Interestingly, Fos-B expression was induced more potently (Ͼ100-fold) by thrombin than by PDGF-BB. Other RTK agonists, such as epidermal growth factor (50 ng/ ml) and basic FGF (50 ng/ml), also failed to induce the expression of Fos-B (data not shown), indicating that GPCR agonists specifically induce the expression of this proto-oncogene product. Jun-D is expressed constitutively in VSMC, and its levels were increased 3-fold in response to PDGF-BB and thrombin. SB203580, while having no effect on c-Fos, Fos-B, and c-Jun induction, inhibited Fra-1 expression induced by both PDGF-BB and thrombin and Jun-B and Jun-D expression induced by thrombin only. The Fos and Jun family proteins form the transcriptional factor, AP-1, which has been shown to play an important role in cell proliferation, differentiation, and apoptosis (31). Because inhibition of p38 MAPK depleted the levels of Jun-B and Jun-D in thrombin-treated cells, we suspected a role for AP-1 in p38 MAPK-dependent thrombin-induced growth in these cells. To address this we first determined the composition of AP-1 in PDGF-BB and thrombin-treated VSMC. Quiescent VSMC were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for 2 h, and nuclear proteins were isolated. Equal amounts of nuclear protein from control and agonist-treated VSMC were analyzed for AP-1-DNA binding activity by EMSA, using a 32 P-labeled AP-1 consensus oligonucleotide probe. AP-1-DNA binding activity was increased by both PDGF-BB and thrombin compared with control (Fig. 4A). Supershift EMSA for each of the Fos and Jun family proteins showed the presence of c-Fos, Fos-B, Fra-1, Jun-B, and Jun-D in the AP-1 complexes formed in response to both PDGF-BB and thrombin, although their levels differed between the two treatments (Fig. 4, B-D, F, and G). Fra-1, Jun-B, and Jun-D were present in the AP-1 complexes formed in response to both PDGF-BB and thrombin. Fos-B was present more abundantly in thrombin-induced AP-1 complexes. Consistent with the Western blot analyses, inhibition of p38 MAPK by SB203580 resulted in reduced amounts of Fra-1 and Jun-B in the AP-1 complexes formed in response to PDGF-BB and thrombin. Previously, we have reported that Jun-B forms the majority of the AP-1 complex in VSMC in response to PDGF-BB and thrombin and in concert with c-Fos drives AP-1-responsive reporter gene expression (26). Because SB203580 significantly reduced the expression of Jun-B and Fra-1 and thereby their levels in AP-1 complexes, we wanted to examine whether these decreases have any impact on PDGF-BB-and thrombin-induced AP-1-responsive reporter gene expression. VSMC were transiently transfected with an AP-1-responsive reporter plasmid, pCOLL-CAT, growth-arrested, treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 for 4 h and cell lysates prepared. Cell lysates containing equal amounts of protein were assayed for CAT activity using coenzyme A and [ 14 C]chloramphenicol as substrates. As shown in Fig. 5, both PDGF-BB and thrombin induced AP-1-responsive CAT expression in VSMC by 1.8 -2.4-fold. Inhibition of p38 MAPK, while causing a 50% reduction in thrombin-induced DNA synthesis, attenuated only the basal but not the PDGF-BB-or thrombininduced AP-1-responsive reporter gene expression. This result suggests that p38 MAPK mediates thrombin-induced VSMC growth independent of its effects on AP-1.
Earlier studies have reported that MAPKs via phosphorylating CREB/ATF1 play a role in CRE-dependent gene expression (32,33). To test the role of p38 MAPK in CRE-dependent gene expression and its role in VSMC growth, we next studied the effect of PDGF-BB and thrombin on CRE activity. Quiescent VSMC were treated with and without PDGF-BB (20 ng/ ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M) for 2 h, and nuclear proteins were prepared and analyzed by EMSA and supershift EMSA for CRE-DNA binding activity and CRE constituents, respectively. Both

FIG. 5. Effect of SB203580 on PDGF-BB-and thrombin-induced AP-1-dependent reporter gene expression in VSMC.
VSMC that were transfected with an AP-1-dependent reporter gene plasmid were growth-arrested and treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M) for 4 h, and cell lysates were prepared. Cell lysates were assayed for CAT activity using acetyl-coenzyme A and [ 14 C]chloramphenicol as substrates, and the substrate-products were separated by TLC and subjected to autoradiography.
PDGF-BB and thrombin induced CRE-DNA binding activity, and SB203580 inhibited ϳ60% of the thrombin-induced, but not PDGF-BB-induced, CRE-DNA binding activity (Fig. 6). Supershift EMSA using anti-ATF-1 antibodies revealed that ATF-1 constitutes most of the PDGF-BB-and thrombin-induced CRE-DNA complexes. Use of anti-CREB-1 or anti-ATF-2 antibodies neither abolished nor caused a supershift of the CRE-DNA binding activity (data not shown). However, the lack of effect of anti-CREB-1 or anti-ATF-2 antibodies to cause a supershift or abolish the CRE-DNA binding activity does not rule out the presence of these transcriptional factors in the complexes. To confirm CRE-dependent transactivation, VSMC were transfected with a CRE-LUC reporter plasmid, growtharrested, treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M) for 4 h, and cell lysates prepared. Cell lysates from control and each treatment were assayed for luciferase activity. Both PDGF-BB and thrombin induced CREdependent luciferase activity (Fig. 7). SB203580 blocked only the thrombin-increased luciferase activity by ϳ50%. To delineate the mechanism underlying this, quiescent VSMC were treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) in the presence and absence of SB203580 (10 M) for 30 min and cell lysates were prepared. Equal amounts of protein from control and each treatment were analyzed for the phosphorylation state of CREB-1 and ATF-1 using a phosphospecific antibody that recognizes both the phosphoproteins. Both PDGF-BB and thrombin induced phosphorylation of CREB-1 and ATF-1 (Fig. 8). PDGF-BB and thrombin had no significant effect on the levels of CREB-1 or ATF-1, suggesting that the increase in the phosphorylation of these transcriptional factors was not because of an increase in their levels (data not shown). SB203580, while inhibiting both PDGF-BBand thrombin-induced CREB-1 phosphorylation, blocked 50% of thrombin-induced ATF-1 phosphorylation only. To obtain additional evidence for the role of ATF-1 in thrombin-induced DNA synthesis, ATF-1 and CREB-1 activities were down-regulated by intracellular expression of an inhibitory anti-ATF-1 single-chain antibody fragment (ScFv) or a dominant negative mutant of CREB-1, respectively, growth-arrested, treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for 24 h, and DNA synthesis measured by [ 3 H]thymidine incorporation. Ectopic expression of ScFv but not the dominant negative mutant of CREB-1 attenuated the DNA synthesis induced by thrombin (Fig. 9). DNA synthesis induced by PDGF-BB was not inhibited. DISCUSSION The important findings of the present study are as follows. 1) p38 MAPK plays a selective role in GPCR agonist-induced DNA synthesis in VSMC, and 2) phosphorylation of ATF-1 appears to be the likely mechanism for the selectivity of p38 MAPK. A role for p38 MAPK in cell growth has been reported previously (34). The present study, however, shows that p38 MAPK plays a differential role in the induction of VSMC growth by RTK and GPCR agonists. The most intriguing find- ing of the present study is the phosphorylation of ATF-1, a transcriptional factor that belongs to the leucine zipper family of proteins (35,36). ATF-1, like CREB, binds to a consensus motif 5Ј-TGACGTCA-3Ј in the promoter region of genes and activates transcription. These transcriptional factors contain a kinase-inducible transactivation domain that possesses consensus phosphorylation sites for several kinases including protein kinase A (37). In fact, several kinases, including calcium calmodulin kinase IV (38) and p90 ribosomal S6 kinase 2 (39), have been reported to phosphorylate and activate CREB. Although CREB/ATF-1 family proteins have been extensively studied for their role in cAMP and Ca 2ϩ -responsive gene regulation, their role in cell growth is less clear. The absence of ATF-1 in normal melanocytes and its detection in metastatic melanoma cells, however, suggest that this transcriptional factor may be involved in the regulation of cell growth (40). The most convincing evidence for the role of ATF-1 and its related protein CREB in cell growth comes from depletion studies (27,41,42). Ectopic expression of single-chain antibody fragment of ATF-1 or dominant negative mutant of CREB-1 inhibited the tumor growth and metastasis of human melanoma cells and the survival of these cells, respectively (27,41).
CREB/ATF-1 transcription factors form homodimers or heterodimers with members of the Jun family proteins, and they preferentially bind to CRE (31,43). Despite the presence of both CREB-1 and ATF-1 in VSMC, only ATF-1 was found in the CRE-DNA complexes formed in response to PDGF-BB and thrombin in these cells. This result implies that ATF-1 exists as either homodimers or heterodimers with other transcriptional factors, such as Jun-B or Jun-D. Although both PDGF-BB and thrombin stimulated ATF-1 phosphorylation, only thrombininduced phosphorylation was sensitive to p38 MAPK. This finding also suggests that ATF-1 phosphorylation induced by PDGF-BB is independent of p38 MAPK. Regardless of the mechanisms of its phosphorylation, ATF-1 is present both in PDGF-BB-and thrombin-induced CRE-DNA complexes. If it was involved in growth induced by both agonists, then inhibition of its activity via ectopic expression of its single-chain antibody should have interfered with the DNA synthesis induced by both PDGF-BB and thrombin and not the latter only. It is possible that, although it is the predominant component in PDGF-BB-and thrombin-induced CRE-DNA complexes, it may exist as heterodimers with different Jun family proteins in response to PDGF-BB and thrombin and that the ATF-1 complex in GPCR agonist-treated cells is involved in growth induction.
PDGF-BB and thrombin also stimulated the phosphorylation of CREB-1 in a p38 MAPK-dependent manner. CREB-1 phosphorylation by PDGF-BB and thrombin, although mediated by p38 MAPK, appears not to be on the path to growth stimulation. The lack of a role for p38 MAPK-dependent CREB-1 phosphorylation in the induction of growth was further confirmed by the inability of its dominant negative mutant to suppress either PDGF-BB-or thrombin-induced DNA synthesis in VSMC. It was recently reported that CREB plays an important role in VSMC differentiation-specific gene regulation (44). This study further showed that CREB is a negative regulator of VSMC growth. The present findings are consistent with these observations. CREB/ATF-1 act as survival factors in metastatic melanoma cells (27,41,42). In addition, the expression of the thrombin receptor (protease-activated receptor-1) is directly correlated with the metastatic ability of human melanoma cells (46,47). Our observation that thrombin-induced p38 MAPK-dependent FIG. 9. Endogenous expression of an inhibitory anti-ATF-1 ScFv reduces thrombin-but not PDGF-BB-induced DNA synthesis in VSMC. VSMC were transfected with mock or ScFv anti-ATF-1 or KCREB plasmids, growth-arrested, and treated with and without PDGF-BB (20 ng/ml) or thrombin (0.1 unit/ml) for 24 h, and DNA synthesis was measured by pulse-labeling cells with 1 Ci/ml [ 3 H]thymidine for the last 2 h of the 24-h incubation period and counting the trichloroacetic acid-precipitable radioactivity. *, p Ͻ 0.01 versus control; **, p Ͻ 0.01 versus thrombin treatment.