Gα12/13-mediated Up-regulation of TRPC6 Negatively Regulates Endothelin-1-induced Cardiac Myofibroblast Formation and Collagen Synthesis through Nuclear Factor of Activated T Cells Activation*

Sustained elevation of [Ca2+]i has been implicated in many cellular events. We previously reported that α subunits of G12 family G proteins (Gα12/13) participate in sustained Ca2+ influx required for the activation of nuclear factor of activated T cells (NFAT), a Ca2+-responsive transcriptional factor, in rat neonatal cardiac fibroblasts. Here, we demonstrate that Gα12/13-mediated up-regulation of canonical transient receptor potential 6 (TRPC6) channels participates in sustained Ca2+ influx and NFAT activation by endothelin (ET)-1 treatment. Expression of constitutively active Gα12 or Gα13 increased the expression of TRPC6 proteins and basal Ca2+ influx activity. The treatment with ET-1 increased TRPC6 protein levels through Gα12/13, reactive oxygen species, and c-Jun N-terminal kinase (JNK)-dependent pathways. NFAT is activated by sustained increase in [Ca2+]i through up-regulated TRPC6. A Gα12/13-inhibitory polypeptide derived from the regulator of the G-protein signaling domain of p115-Rho guanine nucleotide exchange factor and a JNK inhibitor, SP600125, suppressed the ET-1-induced increase in expression of marker proteins of myofibroblast formation through a Gα12/13-reactive oxygen species-JNK pathway. The ET-1-induced myofibroblast formation was suppressed by overexpression of TRPC6 and CA NFAT, whereas it was enhanced by TRPC6 small interfering RNAs and cyclosporine A. These results suggest two opposite roles of Gα12/13 in cardiac fibroblasts. First, Gα12/13 mediate ET-1-induced myofibroblast formation. Second, Gα12/13 mediate TRPC6 up-regulation and NFAT activation that negatively regulates ET-1-induced myofibroblast formation. Furthermore, TRPC6 mediates hypertrophic responses in cardiac myocytes but suppresses fibrotic responses in cardiac fibroblasts. Thus, TRPC6 mediates opposite responses in cardiac myocytes and fibroblasts.

In the heart, a Ca 2ϩ -responsive serine/threonine phosphatase calcineurin is activated by sustained [Ca 2ϩ ] i increase, and calcineurin activation has been implicated in hypertrophic growth of cardiomyocytes (11,12). Activated calcineurin dephosphorylates the transcriptional factor NFAT, facilitating translocation of NFAT from cytosol to the nucleus, where it acts synergistically with other partners to mediate the expression of prohypertrophic genes. However, the physiological and pathological roles of calcineurin-NFAT signaling in cardiac fibroblasts are still controversial when the roles of the signaling pathway are analyzed by CysA, a calcineurin inhibitor. It has been reported that treatment with CysA inhibits pressure overload-induced cardiac hypertrophy and fibrosis (13) and enhances cardiac dysfunction during postinfarction failure (14). Other reports have demonstrated that myocardial fibrosis is promoted in transplanted hearts (15) or chronically aortic banded hearts (16) treated with CysA. Cardiac fibrosis in vivo often follows the development of cardiac hypertrophy. To understand fibrotic pathways, it is essential to examine mechanisms of cardiac fibrosis and hypertrophy individually. Although the role of NFAT in cardiac hypertrophy is established, the role of NFAT in cardiac fibrosis is still unknown.
We previously reported that G␣ 12/13 mediate Ang II-induced NFAT activation in cardiac fibroblasts (17). We also demonstrated that activation of G␣ 13 increases basal Ca 2ϩ influx activity, which is completely suppressed by SK&F96365, an inhibitor of RACCs (18). Members of the TRPC family channels are thought to be molecular candidates for RACCs (19). Recent reports have implicated the involvement of TRPC up-regulation in diseases with abnormal Ca 2ϩ handling and resultant heart failure (20,21). We recently reported that TRPC3 and TRPC6 are involved in Ang II-induced hypertrophic responses of rat neonatal cardiomyocytes (12). However, it is still unknown whether TRPC channels participate in G␣ 12/13 -mediated NFAT activation and agonist-induced myofibroblast formation in cardiac fibroblasts.
In this study, we investigate how the expression of TRPC channels is regulated and whether TRPC channels are involved in the G␣ 12/13 -mediated increase in basal Ca 2ϩ influx required for NFAT activation. We also examine the roles of TRPC channels and NFAT in transformation of cardiac fibroblasts to myofibroblasts.

EXPERIMENTAL PROCEDURES
Materials and Plasmid Construction-JNK inhibitor II (SP600125), cyclosporine A, and PP2 were purchased from Calbiochem. Ang II, ET-1, DPI, U73122, Igepal CA-630, BQ123, BQ788, and anti-␣-SMA antibody (clone 1A4) were purchased from Sigma. Fura-2/AM was from Dojindo (Kumamoto, Japan). Collagenase and Fugene 6 were from Roche Applied Science. Dual luciferase reagents was from Promega. pNFAT-Luc and pRL-SV40 vectors were from Stratagene. Anti-TRPC3 and anti-TRPC6 antibodies were from Alomone Laboratories. Anti-TRPC7 antibody was prepared by Y. Mori. Anti-phospho-ERK, anti-ERK, anti-phospho-Src (Tyr 416 ), and anti-Src antibodies were from Cell Signaling. Anti-glyceraldehyde-3phosphate dehydrogenase, anti-JNK1, horseradish peroxidaseconjugated anti-rabbit IgG, and anti-mouse IgG antibodies were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-Rac antibody was from Transduction Laboratories. CA NFAT was first inserted into the pEGFP-C1 vector (Clontech) and transferred into pShuttle-cytomegalovirus vector to produce recombinant adenovirus. A 913-bp fragment containing the upstream region and the 5Ј-untranslated region of the rat TRPC6 gene 3 was isolated by PCR using rat genomic DNA as a template, and the fragment was inserted upstream of a luciferase gene in the pGL3-Basic vector using KpnI/BglII sites.
Cell Culture-Cardiac fibroblasts and myocytes were prepared from ventricles of 1-2-day-old Sprague-Dawley rats (17,22). Briefly, after digestion of ventricles with 0.1% collagenase, isolated fibroblasts were plated on a noncoated dish in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 50 units/ml penicillin/streptomycin. Subconfluent cells were serum-starved for 48 h and used for the experiments. Considering the possibility that cardiac fibroblasts may lose the original characteristics after prolonged culture, cells were used within two passages in all experiments.
Measurement of Intracellular Ca 2ϩ Measurement and NFAT Activity-The intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) of cardiac fibroblasts was determined by a method described previously (17). Fluorescence images of GFP-positive cells were recorded and analyzed with a video image analysis system (Aquacosmos, Hamamatsu Photonics). Measurement of NFAT activity was performed as described previously (17). For measuring the translocation of GFP-NFAT4, cells (1.5 ϫ 10 5 ) plated on glass bottom 35-mm dishes were infected for 48 h with adenovirus coding GFP-NFAT4 at a multiplicity of infection of 30. After ET-1 stimulation (100 nM) for 48 h, the localization of GFP-NFAT4 was determined with a laser-scanning confocal imaging system (LSM510; Carl Zeiss).
Measurement of ROS Production-Measurement of intracellular ROS concentration was performed in 2 mM Ca 2ϩ -containing Hepes-buffered saline solution (107 mM NaCl, 6 mM KCl, 11.5 mM glucose, 20 mM HEPES, pH 7.4, 1.2 mM MgSO 4 , 2 mM CaCl 2 ) with a fluorescent dye, 2Ј,7Ј-dichlorofluorescein diacetate as described previously (17). Fluorescence images were recorded and analyzed with a video image analysis system (Aquacosmos, Hamamatsu Photonics). The peak changes (⌬F/ F 0 ) of DCF fluorescence intensity were identified as values obtained by subtracting the basal fluorescence intensity (F 0 ) from the maximal intensity during 5-min ET-1 treatment.
Expression Analysis of TRPC mRNAs-The RT-PCR protocol used for the expression analysis and the PCR primers used in this study were as described previously (26). Real time RT-PCR for quantitative measurement was performed as described pre- 3 The sequence was obtained from Ref. 37 (GenBank TM accession number  NW_047798).
Measurement of Rac Activity-Activation of small G proteins was determined by the method of Nishida et al. (25). Activated Rac was pulled down with 10 g of glutathione S-transferase-fused Racinteracting domain of p21-activated kinase (PAK-CRIB). Pulled down small G proteins were detected with anti-Rac1 antibody.
Measurement of Cardiac Myofibroblast Formation-Transformation of cardiac fibroblasts to myofibroblasts was assessed by the expression of ␣-SMA and production of ECM components (6). Briefly, 24 h after infection, cardiomyocytes were stimulated with ET-1 (100 nM) for 48 h. The cells were washed, fixed, incubated with anti-␣-SMA antibody, and then stained with Alexa Fluor 546 goat antimouse IgG. Morphological changes and ␣-SMA expression of cardiac fibroblasts were visualized with confocal microscopy. For the measurement of collagen synthesis, cells were harvested with lysis buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Igepal CA-630, and protease inhibitor mixture (Nacalai, Japan). Production of ECM components was assessed by [ 3 H]proline incorporation. After cells were stimulated with ET-1 (100 nM) for 2 h, [ 3 H]proline (1 Ci/ml) was added to the culture medium and further incubated for 6 h. The incorporated [ 3 H]proline was extracted by 1 N NaOH overnight and measured using a liquid scintillation counter.
Measurement of Hypertrophic Responses of Cardiomyocytes-Preparation of rat neonatal cardiomyocytes and measurement of hypertrophic responses were performed as described (12).   13 Activation-We previously demonstrated that the expression of GTPase-defective mutants of G␣ 12 (Q229L; CA G␣ 12 ) and G␣ 13 (Q226L; CA G␣ 13 ) increases [Ca 2ϩ ] i and NFAT activity through an SK&F96365-sensitve Ca 2ϩ influx pathway (17,18). TRPC chan-nels are thought to be molecular candidates for RACCs. Thus, we examined which TRPC channels are expressed in rat cardiac fibroblasts. Among TRPC1 to -7 channels, TRPC1, TRPC3, TRPC6, and TRPC7 mRNAs were detected (Fig.  1A). Interestingly, only TRPC6 mRNA expression was increased 2.5-fold by the expression of CA G␣ 13 . The expression of TRPC6 proteins was also increased by CA G␣ 12 (Fig. 1B). Expression of CA G␣ q (R183C) and WT G␣ i2 did not increase TRPC6 protein levels, although both mutants significantly increased ERK activity (supplemental Fig. 1). Treatment with forskolin (50 M) rather decreased basal expression of TRPC6 proteins. These results indicate that G␣ 12/13 specifically up-regulate TRPC6 proteins in cardiac fibroblasts. In order to examine whether up-regulation of TRPC6 participates in CA G␣ 12/13induced enhancement of Ca 2ϩ responses, knockdown experiments of TRPC6 proteins were performed by using TRPC6 siRNAs (12). Since the extent of NFAT activation induced by CA G␣ 13 was larger than that by CA G␣ 12 , CA G␣ 13 -expressing cells were used to examine the involvement of TRPC6 in G␣ 12/13mediated Ca 2ϩ responses. Treatment with TRPC6 siRNAs completely suppressed CA G␣ 13 -induced up-regulation of TRPC6 protein expression (Fig. 1C). We confirmed that TRPC6 siRNAs did not affect the expression levels of TRPC3 and TRPC7 proteins (data not shown). The basal Ca 2ϩ influx activity was determined by the increase in [Ca 2ϩ ] i induced by the addition of 2 mM extracellular Ca 2ϩ to cells in Ca 2ϩ -free Tyrode's solution (18). This basal Ca 2ϩ influx activity was enhanced by the expression of CA G␣ 13 and CA G␣ 12 (Fig. 1D). The CA G␣ 13 -induced increase in [Ca 2ϩ ] i was completely suppressed by the treatment with TRPC6 siRNAs (Fig. 1, C-E). The sustained increase in [Ca 2ϩ ] i activated NFAT activity, and the treatment with TRPC6 siRNAs inhibited CA G␣ 13 -induced NFAT activation (Fig. 1F). These results suggest that up-regulation of TRPC6 by G␣ 13 activation enhances Ca 2ϩ influx and NFAT activation in cardiac fibroblasts.

Up-regulation of TRPC6 Proteins by G␣
Involvement of Tyrosine Kinase in TRPC6-mediated Ca 2ϩ Influx by G␣ 12/13 Activation-We next examined whether the channel activity of TRPC6 proteins is actually enhanced by G␣ 13 activation. WT TRPC6 or two DN mutants of TRPC6 (C6-3A and C6-⌬N) were overexpressed in cardiac fibroblasts ( Fig. 2A). The expression of DN TRPC6 significantly sup- pressed the sustained [Ca 2ϩ ] i increase of CA G␣ 13 -expressing cells induced by the addition of 2 mM extracellular Ca 2ϩ (Fig. 2, B and C), indicating that the increase in [Ca 2ϩ ] i of CA G␣ 13expressing cells is enhanced by Ca 2ϩ influx through TRPC6 channels. It has been reported that TRPC6 activity is regulated by DAG (28) and tyrosine phosphorylation (29), and activated G␣ 13 induces activation of tyrosine kinase (30,31) and PLC (32). We examined the involvement of TRPC6 in CA G␣ 13induced [Ca 2ϩ ] i increase by determining the requirement of tyrosine phosphorylation or PLC for TRPC6 activation. Treatment with PP2, a Src tyrosine kinase inhibitor, and SP600125, a JNK inhibitor, but not U73122, a PLC inhibitor, significantly suppressed the increase in [Ca 2ϩ ] i of CA G␣ 13 -expressing cells induced by the addition of 2 mM Ca 2ϩ (Fig. 2, D and E). Consistent with the involvement of Src, the expression of CA G␣ 12 and CA G␣ 13 increased Src activity about 2.5-fold (Fig. 2F). CA G␣ 13 -induced Src activation was suppressed by treatment with AG1478, an EGFR kinase inhibitor (Fig. 2F), suggesting that EGFR kinase mediates the CA G␣ 13 -induced Src activation. Treatment with SP600125, but not PP2 and AG1478, suppressed CA G␣ 13induced up-regulation of TRPC6 protein, whereas treatment with PP2 and AG1478, but not SP600125, suppressed CA G␣ 13 -induced Src activation (Fig. 2F). These results suggest that EGFR-and PP2sensitive tyrosine kinase(s) participate in G␣ 13 -mediated enhancement of Ca 2ϩ influx independently of TRPC6 up-regulation. These findings are consistent with the report that EGFR participates in G␣ 13 -mediated responses (33).
Up-regulation of TRPC6 Proteins by ET-1 and Ang II Stimulation-We next examined whether upregulation of TRPC6 is actually induced by G 13 -coupled receptor stimulation. We have previously demonstrated that G␣ 12/13 mediate Ang II-and ET-1-induced activation of JNK and hypertrophic responses in rat neonatal cardiomyocytes (24,25). The TRPC6 protein expression levels were significantly increased by Ang II treatment and more potently by ET-1 treatment, in a concentrationdependent manner (Fig. 3, A and B). Treatment with ET-1 actually increased TRPC6 mRNA levels (supplemental Fig. 2A). The up-regulation of TRPC6 protein was detected at 6 h after ET-1 stimulation and the expression of TRPC6 proteins reached a peak at 48 h (data not shown). In contrast to TRPC6, Ang II and ET-1 treatment did not affect the expression levels of TRPC3 and TRPC7 proteins (Fig. 3A). These results indicate that ET-1 treatment selectively up-regulates TRPC6 proteins in cardiac fibroblasts. Since the ET-1-induced increase in TRPC6 protein expression levels was completely suppressed by BQ123, a selective ET A receptor antagonist, but not by BQ788, a selective ET B receptor antagonist (Fig. 3C), the ET A receptor predominantly mediates ET-1induced TRPC6 up-regulation. It has been reported that the activation of JNK and STAT3 is required for up-regulation of TRPC6 proteins induced by ET-1 and PDGF stimulation in pulmonary vascular smooth muscle cells (34,35). We previously demonstrated that G␣ 12/13 mediate Ang II-induced Rac activation, ROS production, and JNK activation in cardiac myocytes and fibroblasts (17,25). Thus, we examined whether ROS and JNK are involved in ET-1-induced TRPC6 up-regulation. Expression of p115-RGS, a selective inhibitory polypeptide of G␣ 12/13 , completely suppressed ET-1-induced increase in  AUGUST 10, 2007 • VOLUME 282 • NUMBER 32 JOURNAL OF BIOLOGICAL CHEMISTRY 23121 TRPC6 proteins, as expected (Fig. 3D). The ET-1-induced TRPC6 up-regulation was also suppressed by DPI and SP600125 but not by a PLC inhibitor, U73122, and CysA. Since DPI is a selective inhibitor of ROS production by NADPH oxidase complex and SP600125 selectively inhibits JNK activation, these results suggest that the G␣ 12/13 -ROS-JNK pathway participates in ET-1-induced TRPC6 up-regulation. NFAT is activated through its dephosphorylation by calcineurin that is inhibited by CysA. Since the treatment with CysA increased TRPC6 mRNA levels and the expression of CA NFAT decreased TRPC6 mRNA levels (supplemental Fig. 2B), NFAT may negatively regulate ET-1-induced TRPC6 expression in cardiac fibroblasts. Furthermore, the ET-1-induced increase in TRPC6 promoter-dependent luciferase activity was completely suppressed by DPI and SP600125 but not by CysA (Fig. 3E). These results suggest that the G␣ 12/13 -ROS-JNK pathway participates in the transcriptional activation of the TRPC6 gene. ET-1-stimulated up-regulation of TRPC6 protein was completely suppressed by siRNAs against TRPC6 (Fig. 3F), suggesting the utility of these siRNAs for future analysis.

Inhibition of Myofibroblast Formation by TRPC6-NFAT Signaling
Up-regulation of TRPC6 Enhances Receptor-activated Ca 2ϩ Responses but Not Store-operated Ca 2ϩ Responses-TRPC proteins have been emerged as a candidate subunit of RACCs and SOCs (19). To examine whether up-regulation of TRPC6 enhances receptor-activated or store-operated Ca 2ϩ entry, we have individually determined store-operated and receptor-activated Ca 2ϩ entry. The increase in [Ca 2ϩ ] i through receptoractivated Ca 2ϩ entry is determined by the addition of Ca 2ϩ after receptor stimulation in the absence of extracellular Ca 2ϩ . The increase in [Ca 2ϩ ] i through store-operated Ca 2ϩ entry is determined by the addition of Ca 2ϩ after the treatment of cells with Ca 2ϩ ionophore ionomycin in the absence of extracellular Ca 2ϩ . Since the cells were treated with ET-1 for long period of time, we used Ang II as a receptor stimulant to avoid the underestimation of ET-1-stimulated increase in [Ca 2ϩ ] i through receptor-activated Ca 2ϩ entry due to desensitization. The expression of TRPC6 did not affect the Ang II-induced increase in [Ca 2ϩ ] i in the absence of extracellular Ca 2ϩ but enhanced the increase in [Ca 2ϩ ] i by the addition of extracellular Ca 2ϩ (Fig. 4,  A and B). The treatment with TRPC6 siRNA (C6-1609) did not affect the initial increase in [Ca 2ϩ ] i of ET-1-treated cells (Fig. 4,  A and B). However, TRPC6 siRNA (C6-1609) completely inhibited the increase in [Ca 2ϩ ] i induced by the addition of Ca 2ϩ . These results are consistent with the findings that the prolonged treatment of cells with ET-1 increases TRPC6 expression. It has been known that TRPC6-mediated Ca 2ϩ influx is enhanced by DAG derivative OAG (19). We examined whether OAG-stimulated increases in [Ca 2ϩ ] i are suppressed by TRPC6 siRNA. Stimulation with OAG enhanced the increase in [Ca 2ϩ ] i induced by the addition of extracellular Ca 2ϩ of WT TRPC6-expressing or ET-1-treated cells (Fig. 4, C  and D). TRPC6 siRNA (C6-1609) completely suppressed the OAG-induced increase in [Ca 2ϩ ] i of WT TRPC6-expressing or ET-1-treated cells. Although compensatory up-regulation of TRPC3 mRNA expression is reported in TRPC6-deficient mice (36), TRPC6 siRNAs did not affect TRPC3 protein expression (Fig. 3F). These results suggest that up-regulation of TRPC6 enhances receptor-activated Ca 2ϩ entry through a DAG-de-pendent mechanism. In contrast to the involvement of TRPC6 in receptor-activated Ca 2ϩ entry, overexpression of TRPC6 did not affect Ca 2ϩ entry-dependent increase in [Ca 2ϩ ] i induced by ionomycin treatment that completely depletes intracellular Ca 2ϩ stores (27) (Fig. 4, E and F). Up-regulation of TRPC6 by ET-1 treatment did not affect the increase in [Ca 2ϩ ] i induced by Ca 2ϩ entry after ionomycin treatment. These results suggest that the increased expression of TRPC6 does not contribute to the increase in [Ca 2ϩ ] i through SOC-mediated Ca 2ϩ entry in cardiac fibroblasts. Since the Ang II-induced Ca 2ϩ release and ionomycin-induced Ca 2ϩ release were not affected by ET-1 treatment or TRPC6 siRNA treatment, the changes in TRPC6 proteins may not influence Ca 2ϩ content of intracellular Ca 2ϩ stores or IP 3 -mediated functions.
Requirement of TRPC6 for ET-1-induced NFAT Activation-Since CA G␣ 13 -induced TRPC6 up-regulation enhances basal Ca 2ϩ influx activity, we examined whether ET-1 treatment enhances basal Ca 2ϩ influx activity through G␣ 12/13 -mediated TRPC6 up-regulation. Treatment with ET-1 for 48 h enhanced the increase in [Ca 2ϩ ] i induced by the addition of 2 mM Ca 2ϩ to ET-1-containing Ca 2ϩ -free solution, which was completely suppressed by TRPC6 siRNA (C6-1609) and DN TRPC6 (C6-⌬N) (Fig.  5, A and B). The ET-1-induced enhancement of Ca 2ϩ influx was suppressed by p115-RGS and PP2. Treatment with ET-1 increased Src activity, which was completely suppressed by p115-RGS and PP2 but not by SP600125 (Fig. 5C). We confirmed that ET-1-induced TRPC6 up-regulation was not affected by PP2 (supplemental Fig. 2C). Src activation by ET-1 stimulation was mediated by ET A receptor, since ET A receptor-selective (BQ123) but not ET B receptor-selective (BQ788) blockers inhibited ET-1-stimulated Src activation (supplemental Fig.  2D). These results suggest that G␣ 12/13 -mediated activation of Src tyrosine kinase participate in ET-1induced increase in basal Ca 2ϩ influx activity apart from TRPC6 up-regulation. On the other hand, treatment with ET-1 for 48 h induced nuclear localization of GFP-NFAT4 proteins, which were inhibited by p115-RGS and SP600125 (Fig. 5, D and F). Treatment with SP600125 (1 M) significantly inhibited the ET-1 treatment-induced increase in basal Ca 2ϩ influx activity (⌬Ratio ϭ 0.70 Ϯ 0.05, n ϭ 67 cells) and TRPC6 protein levels (Fig. 3D), suggesting that G␣ 12/13 -mediated activation of JNK participates in ET-1-induced NFAT translocation through TRPC6 up-regulation. The ET-1 treatment-induced NFAT translocation was also suppressed by TRPC6 siRNAs (Fig. 5, E and F). The expression of WT TRPC6 increased basal NFAT activity 2.5fold, and it also enhanced NFAT activity induced by ET-1 treatment (Fig. 5G). In contrast, the expression of DN TRPC6 mutants completely suppressed the ET-1 treatment-induced increase in NFAT activity. These results suggest that G␣ 12/13 -mediated up-regulation and activation of TRPC6 channels participate in sustained activation of NFAT induced by ET-1 treatment in cardiac fibroblasts.
Inhibition of Cardiac Myofibroblast Formation by TRPC6 Activation-Transformation of cardiac fibroblasts to myofibroblasts is characterized by expression of ␣-SMA and production of ECM components that are a key event in connective tissue remodeling (6). Since NFAT is activated by the sustained increase in [Ca 2ϩ ] i and the calcineurin-NFAT pathway plays a critical role in the pathogenesis of heart failure, we examined whether TRPC6-mediated NFAT activation participates in ET-1 treatment-induced transformation of cardiac fibroblasts. Treatment with ET-1 increased ␣-SMA expression in a concentration-dependent manner (Fig.  6, A and B). Treatment with ET-1 (1 nM) caused maximal ␣-SMA expression, whereas ET-1 treatment (Ͼ10 nM) slightly decreased ␣-SMA expression. Knockdown of TRPC6 did not affect basal ␣-SMA protein expression levels. However, it weakly but significantly enhanced ET-1-induced ␣-SMA expression (Fig. 6, A and C). Furthermore, TRPC6 siRNAs significantly enhanced the basal incorporation of [ 3 H]proline, an index of synthesis of ECM proteins (Fig. 6D). TRPC6 siRNAs also enhanced ET-1-induced [ 3 H]proline incorporation. Collagen expression is another index of transformation of cardiac fibroblasts to myofibroblasts. TRPC6 siRNA increased mRNAs of collagen type I and III, whereas overexpression of WT TRPC6 significantly suppressed the basal expression of collagen mRNAs (Fig. 6E). These results suggest that TRPC6 negatively regulates ET-1-induced transformation of cardiac fibroblasts to myofibroblasts and synthesis of ECM components.
Inhibition of Cardiac Myofibroblast Formation by NFAT Activation-The expression of TRPC6 WT inhibited the collagen mRNA expression and enhanced NFAT activation by ET-1 treatment (Figs. 5G and 6E). In addition to cardiac fibroblasts, we have previously demonstrated in cardiac myocytes that TRPC6 regulates NFAT activation (12). Then we examined whether TRPC6-mediated NFAT activation negatively regulates ET-1-induced transformation of cardiac fibroblasts to myofibroblasts. The expression of CA NFAT increased basal NFAT activity (8.3 Ϯ 0.9-fold) and suppressed the increase in ␣-SMA expression by ET-1 treatment (Fig. 7, A and B). Expression of ␣-SMA by ET-1 treatment was also suppressed by expression of TRPC6 WT. In contrast, treatment with CysA increased the basal ␣-SMA expression levels and [ 3 H]proline incorporation, which was further enhanced by ET-1 treatment (Fig. 7, B and D). These results suggest that TRPC6-mediated NFAT activation inhibits ET1-induced ␣-SMA expression in cardiac fibroblasts. Furthermore, the ET-1-induced increase in ␣-SMA expression and [ 3 H]proline incorporation was suppressed by the expression of p115-RGS and DN Rac and treatment with SP600125 (Fig. 7, C and D). These results suggest that the G␣ 12/13 -mediated signaling pathway (G␣ 12/13 -Rac-JNK) mediates ET-1-induced myofibroblast formation in cardiac fibroblasts. However, G␣ 12/13 -mediated NFAT activation negatively regulates ET-1-induced cardiac myofibroblast formation.
Inhibition of ET-1-induced JNK Activation by TRPC6-mediated NFAT Activation-In order to examine how NFAT attenuates myofibroblast formation by ET-1 treatment, we next examined the effects of NFAT on ET-1-induced mitogen-activated protein kinase activation. Treatment with ET-1 increased JNK activity, which was suppressed by the expression of CA NFAT and WT TRPC6 but was potentiated by CysA (Fig. 8A). However, the ET-1-induced ERK activation was not affected by CA NFAT, TRPC6 WT, and CysA (Fig. 8B). This is consistent with the result that ERK activation is not involved in the increased expression of TRPC6 by ET-1 treatment (supplemental Fig. 2C). These results suggest that NFAT activation through TRPC6 up-regulation specifically inhibits ET-1-induced JNK activation in cardiac fibroblasts. The ET-1-induced ROS production and Rac activation, both of which work upstream of JNK, were suppressed by TRPC6 overexpression and NFAT activation (Fig. 8, C and D). These results suggest that NFAT inhibits JNK activation through inhibition of Rac activation and ROS production.
Requirement of TRPC6 Up-regulation in ET-1-induced Hypertrophic Responses of Cardiomyocytes-It has been reported that up-regulation of TRPC6 amplifies calcineurin signaling, leading to pathologic cardiac remodeling in mice (37). We have previously reported that G␣ 12/13 -mediated JNK activation participates in ET-1-induced hypertrophic responses of cardiomyocytes (24). We examined whether TRPC6 up-regulation is involved in ET-1-induced hypertrophic growth of rat neonatal cardiomyocytes. Treatment with ET-1 increased TRPC6 protein levels in a concentrationdependent manner, and the EC 50 value was about 3 nM (Fig.  9A). The ET-1-induced increase in TRPC6 protein expression was completely suppressed by p115-RGS (Fig. 9B), suggesting that G␣ 12/13 mediate ET-1-induced TRPC6 up-regulation. Treatment with ET-1 increased BNP-luciferase activity, a marker for cardiac hypertrophy, and the EC 50 value was about 5 nM (Fig. 9C). ET-1-stimulated BNP-luciferase activity was blocked by expression of p115-RGS, an inhibitory polypeptide of G␣ 12/13 . This result suggests that G␣ 12/13 mediate ET-1-stimulated hypertrophic response. Although we could not reliably determine the effects of TRPC6 siRNAs on ET-1-stimulated BNP-luciferase activity, the ET-1-induced NFAT activation and hypertrophic responses (actin reorganization and protein synthesis) were significantly suppressed by TRPC6 siRNAs (Fig. 9, D-F). These results suggest that G␣ 12/13 also play a role in ET-1induced increase in TRPC6 protein expression in car-diomyocytes. However, in the case of cardiomyocytes, the upregulated TRPC6 may positively regulate hypertrophic responses induced by ET-1.

DISCUSSION
In this study, we demonstrated that TRPC6 proteins are up-regulated by ET-1 treatment in cardiac fibroblasts. Furthermore, we demonstrated that the ET-1-induced increase in TRPC6 expression is critical for NFAT activation, which negatively regulates myofibroblast formation. Previous reports have shown that TRPC6 proteins are upregulated by ET-1 and platelet-derived growth factor in pulmonary vascular smooth muscle cells (34,35). These reports are consistent with the present results. We first demonstrated that G␣ 12/13 mediate ET-1-induced up-regulation of TRPC6 mRNAs and proteins in cardiac fibroblasts. We previously reported that G␣ 12/13 mediate Ang II-induced JNK activation through Rac-dependent ROS production (17,25). Since it was reported that c-Jun, one of the target molecules of JNK, is required for agonist-induced up-regulation of TRPC6 proteins in pulmonary vascular smooth muscle cells (36), G␣ 12/13 -mediated JNK activation may participate in the increased expression of TRPC6 in cardiac fibroblasts. Furthermore, we clearly demonstrated that G␣ 12/13 mediate ET-1induced myofibroblast formation, since p115-RGS completely suppressed the increase in ␣-SMA expression and [ 3 H]proline incorporation by ET-1 treatment (Fig. 6). Since ET-1-induced myofibroblast formation was also suppressed by DN Rac and SP600125 (Fig. 6C), the G␣ 12/13 -Rac-JNK signaling pathway may be involved in ET-1-induced myofibroblast formation. However, ET-1-induced fibrotic responses were inhibited by up-regulation of TRPC6 expression and resultant NFAT activation through the G␣ 12/13 -mediated pathway (Fig. 6). Therefore, ET-1 treatment caused fibrotic responses through the Rac-JNK pathway, and at the same time ET-1 treatment negatively regulated fibrotic responses through NFAT activation by the increased expression of TRPC (Fig. 10). The concentration of ET-1 to induce maximal ␣-SMA expression was about 1 nM, which was about 10 times lower than that of ET-1 to induce maximal expression of TRPC6 proteins (Figs. 3B and 6B). Thus, cardiac fibroblasts may turn on their own protective system against excess progression of fibrotic responses when the cells are exposed to a high concentration of ET-1.
Although we cannot explain in detail the mechanism of how NFAT inhibits JNK activation (Fig. 8A) and TRPC6 expression (supplemental Fig. 2) induced by ET-1 stimulation, there are a couple of possibilities. Since DPI and SP600125 suppressed TRPC6 promoter activity by ET-1 treatment, the ROS-JNK pathway plays an important role in TRPC6 expression (Fig. 3E). Expression of CA NFAT suppressed ET-1-stimulated Rac activation, and we previously reported that Rac mediates JNK activation (17,25). Therefore, NFAT inhibits TRPC6 expression through inhibition of the Rac-JNK pathway. We can exclude down-regulation of ET receptors as a mechanism of NFATmediated inhibition of TRPC6 expression, since the ET-1-induced ERK activation was not affected by CA NFAT and TRPC6 WT (Fig. 8B). Since NFAT acts as a transcriptional factor, one possibility is that some factor(s) produced by NFAT activation negatively regulates JNK activity through inhibition of Rac activation in cardiac fibroblasts. For example, NFAT is reported to regulate the expression levels of BMP2 (bone morphogenetic protein 2), an antifibrotic factor (38). Further studies are required for understanding the mechanism of NFAT-dependent inhibition of JNK activation.
It has been reported that TRPC6 channel activity is positively regulated by various stimuli, such as DAG (28), tyrosine phosphorylation (29), and mechanical stress (39). We found that the increase in [Ca 2ϩ ] i of CA G␣ 13 -expressing cells was completely inhibited by PP2 but not by U73122 (Fig. 2D), suggesting that Src tyrosine kinase but not PLC-generated DAG is involved in CA G␣ 13 -mediated Ca 2ϩ influx. It also suggests that CA G␣ 13 does not activate PLC in cardiac fibroblasts. We also found that AG1478 inhibits CA G␣ 13 -induced Src activation (Fig. 2F), suggesting the involvement of EGFR kinase in G␣ 13 -mediated Src activation. Our results suggest that CA G␣ 13 -mediated Src activation is involved in CA G␣ 13 -induced Ca 2ϩ influx.
We also demonstrated that Ca 2ϩ influx-induced increase in [Ca 2ϩ ] i evoked by ionomycin is not enhanced by TRPC6 overexpression (Fig. 4C). This result suggests that TRPC6 does not participate in store-operated Ca 2ϩ entry in cardiac fibroblasts. TRPC1 is often referred to as a SOC, but it is unknown whether TRPC1 works as a true SOC, because this can also be activated by DAG and membrane stretch (19,40). In addition to TRPC1, it has been recently reported that STIM1, located predominantly in the endoplasmic reticulum, acts as a Ca 2ϩ sensor for store-operated Ca 2ϩ entry (41,42) and that Orai1/CRACM1 is identified as the Ca 2ϩ release-activated Ca 2ϩ channel gene (43,  44). Since TRPC1, STIM1, and Orai1/CRACM1 are ubiquitously expressed, these molecules may participate in store-operated Ca 2ϩ entry in cardiac fibroblasts.
A variety of evidence has implicated that the sustained increase in [Ca 2ϩ ] i is involved in the pathogenesis of heart failure (45). For example, up-regulation of TRPC proteins are associated with the reduction of sarcoplasmic reticulum Ca 2ϩ -ATPase (20), suggesting the importance of TRPC in remodeling of the Ca 2ϩ signaling mechanism in cardiac myocytes. Furthermore, it has been recently reported that diverse signals for cardiac hypertrophy stimulate the expression of TRPC channels and increase NFAT activity (21). Other groups demonstrated that the cardiac myocytes-specific expression of TRPC6 or CA calcineurin (upstream activator of NFAT) induces cardiac hypertrophy, and NFAT is activated in pathological but not physiological hypertrophy (46 -48). We found in this study that G␣ 12/13 -mediated up-regulation of TRPC6 protein participates in ET-1-induced cardiomyocyte hypertrophy (Fig. 9). Thus, the TRPC-NFAT signaling pathway in myocytes plays an essential role in cardiac hypertrophy. In contrast to cardiac myocytes, we demonstrated in cardiac fibroblasts that up-regulation of TRPC6 expression and resultant activation of NFAT inhibit myofibroblast formation and synthesis of ECM components induced by ET-1 treatment. Therefore, we suggest that TRPC6 and TRPC6-mediated NFAT activation in cardiac fibroblasts work as negative regulators against cardiac fibrosis. Many lines of evidence have indicated that TRPC channels represent novel pharmacologic targets, since NFAT is activated by Ca 2ϩ influx though TRPC channels in pathological hypertrophy. However, the role of TRPC channels in cardiac fibroblasts and cardiac fibrosis does not support this idea, since the TRPC6-NFAT signaling pathway works as a negative regulator of fibrotic responses. Inhibitor of TRPC channels may worsen the fibrotic process accompanied with hypertrophy. Since cardiac fibrosis leads to myocardial stiffness and diastolic dysfunction, future in vivo work will be required for understanding the role of TRPC proteins in the cardiac function and development of cardiac fibrosis. The present findings also indicate that activators of TRPC channel-NFAT signaling or downstream signaling components are novel anti-fibrotic drugs.
In summary, we have demonstrated a signal transduction pathway of ET-1-induced up-regulation of TRPC6 expression and NFAT activation. As the sustained increase in [Ca 2ϩ ] i through TRPC6 activates NFAT, the G␣ 12/13 -mediated TRPC6 expression and NFAT activation may act as a negative feedback regulator against ET-1-mediated fibrotic responses.