Na+/Ca2+ Exchanger Activity Modulates Connective Tissue Growth Factor mRNA Expression in Transforming Growth Factor β1- and Des-Arg10-kallidin-stimulated Myofibroblasts*

Transforming growth factor (TGF)-β and des-Arg10-kallidin stimulate the expression of connective tissue growth factor (CTGF), a matrix signaling molecule that is frequently overexpressed in fibrotic disorders. Because the early signal transduction events regulating CTGF expression are unclear, we investigated the role of Ca2+ homeostasis in CTGF mRNA expression in TGF-β1- and des-Arg10-kallidin-stimulated human lung myofibroblasts. Activation of the kinin B1 receptor with des-Arg10-kallidin stimulated a rise in cytosolic Ca2+ that was extracellular Na+-dependent and extracellular Ca2+-dependent. The des-Arg10-kallidin-stimulated increase of cytosolic Ca2+ was blocked by KB-R7943, a specific inhibitor of Ca2+ entry mode operation of the plasma membrane Na+/Ca2+ exchanger. TGF-β1 similarly stimulated a KB-R7943-sensitive increase of cytosolic Ca2+ with kinetics distinct from the des-Arg10-kallidin-stimulated Ca2+ response. We also found that KB-R7943 or 2′,4′-dichlorobenzamil, an amiloride analog that inhibits the Na+/Ca2+ exchanger activity, blocked the TGF-β1- and des-Arg10-kallidin-stimulated increases of CTGF mRNA. Pretreatment with KB-R7943 also reduced the basal and TGF-β1-stimulated levels of α1(I) collagen and α smooth muscle actin mRNAs. These data suggest that, in addition to regulating ion homeostasis, Na+/Ca2+ exchanger acts as a signal transducer regulating CTGF, α1(I) collagen, and α smooth muscle actin expression. Consistent with a more widespread role for Na+/Ca2+ exchanger in fibrogenesis, we also observed that KB-R7943 likewise blocked TGF-β1-stimulated levels of CTGF mRNA in human microvascular endothelial and human osteoblast-like cells. We conclude that Ca2+ entry mode operation of the Na+/Ca2+ exchanger is required for des-Arg10-kallidin- and TGF-β1-stimulated fibrogenesis and participates in the maintenance of the myofibroblast phenotype.

Connective tissue growth factor (CTGF), 1 a matrix signaling molecule of the Cyr61/connective tissue growth factor/nephroblastoma-overexpressed (CCN) family (1), promotes extracellular matrix production by fibroblasts. The CCN family of matricellular proteins is distinguished by a high degree of amino acid homology (50 -90%) and conservation of 38 cysteine residues. Like other CCN family members, CTGF is comprised of a secretory signal and four distinct protein modules: an insulinlike growth factor-binding domain, a von Willebrand factor type C repeat, a thrombospondin type 1 repeat, and a C-terminal module (1). CTGF is a cysteine-rich, heparin-binding, 349amino acid protein (2)(3)(4) expressed at high levels during wound repair and at sites of connective tissue formation in a variety of fibrotic disorders (5)(6)(7)(8)(9). Recombinant CTGF stimulates fibroblast proliferation and extracellular matrix protein synthesis (10,11). Transforming growth factor ␤ (TGF-␤) stimulates CTGF transcription in normal rat kidney fibroblasts, which is not observed following epidermal growth factor, fibroblast growth factor, or platelet-derived growth factor stimulation (11). We have shown that TGF-␤1 stimulates an increase in CTGF transcription in human myofibroblasts, suggesting a role for CTGF in the pathogenesis of fibrosis (12).
We have previously shown that activation of the kinin B1 receptor by des-Arg 10 -kallidin enhances CTGF mRNA stability in human myofibroblasts (12). In contrast, activation of the kinin B2 receptor by bradykinin or kallidin does not alter CTGF mRNA levels. The B1 receptor is a G protein-coupled receptor that is expressed following injury or exposure to pro-inflammatory agents such as interleukin-1␤ (13)(14)(15)(16)(17). Des-Arg 10 -kallidin, the carboxypeptidase metabolite of kallidin, is the only known natural ligand for the human kinin B1 receptor (K D ϭ 0.2 nM), whereas des-arg 9 -bradykinin activates the B1 receptor in rodents, cows, sheep, guinea pigs, dogs, and cats (18). The B2 receptor agonists kallidin and bradykinin bind to the B1 receptor with relatively low affinity (K D Ͼ 100 nM) (18). B1 receptor activation induces a rise in cytosolic Ca 2ϩ with kinetics distinct from those following B2 receptor activation (6); however, the mechanisms for these transient increases of cytosolic Ca 2ϩ are unclear, and the relationship between the Ca 2ϩ transients and CTGF mRNA stability has not been determined.
Cytosolic Ca 2ϩ levels are maintained by various transporters in the plasma membrane and intracellular organelles. Among the different Ca 2ϩ transporters, the electrogenic Na ϩ /Ca 2ϩ exchanger is unique because it moves Ca 2ϩ across the membrane bidirectionally depending upon the Na ϩ , Ca 2ϩ , and K ϩ gradients and the membrane potential (19). In most cells, the stoichiometry for this transporter is three Na ϩ for one Ca 2ϩ , although this point is controversial (20,21). Major regulatory properties of the exchanger include its Na ϩ dependence, Ca 2ϩ modulation, and protein phosphorylation (22,23). Three genes code for three exchanger isoforms. Na ϩ /Ca 2ϩ exchanger isoform 1 is ubiquitously expressed and has recently been shown to exist in a macromolecular complex that includes PKC, cAMPdependent protein kinase, cAMP-dependent protein kinaseanchoring protein, and protein phosphatases 1 and 2A, suggesting that the Na ϩ /Ca 2ϩ exchanger activity may be rapidly regulated by phosphorylation/dephosphorylation events (22).
Fibrotic disorders such as idiopathic pulmonary fibrosis are characterized by excess deposition of extracellular matrix, a chronic process that can lead to organ dysfunction or death. Currently, there is no efficacious intervention that targets fibrogenesis. The foci of myofibroblasts, which are responsible for the synthesis and remodeling of extracellular matrix, represent a hallmark of the histopathology of interstitial pulmonary fibrosis (2, 3, 24 -26). Cells from the human embryonic lung cell line IMR-90 are classified as myofibroblasts and express metavinculin, sm22, ␣ smooth muscle actin, and calponin as well as extracellular matrix proteins including CTGF and type I collagen (27,28). In this report, we examine the role of Ca 2ϩ homeostasis in the regulation of expression of CTGF, ␣1(I) collagen, and ␣ smooth muscle actin in IMR-90 myofibroblasts. We report that Ca 2ϩ entry mode operation of the Na ϩ / Ca 2ϩ exchanger is required for des-Arg 10 -kallidin and TGF-␤1 to stimulate increases of CTGF mRNA. In addition, we demonstrate that Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger is required for basal expression of ␣1(I) collagen and ␣ smooth muscle actin, genes characteristically expressed in myofibroblasts.
Human microvascular endothelial cells (HMEC-1), a kind gift from Dr. Edwin Ades and Francisco J. Candal (Center for Disease Control) and Dr. Thomas Lawley (Emory University), were maintained on MCDB 131 medium supplemented with 10% FBS and 10 mM glutamine. The human osteoblast-like cell line MG-63 was maintained on DMEM supplemented with 10% FBS.
Cytosolic Ca 2ϩ Measurements-Steady-state levels of cytosolic Ca 2ϩ were determined with the fluorescent probe, fura-2 acetoxymethyl ester (fura-2 AM; Molecular Probes, Inc., Eugene, OR). Myofibroblasts were grown on coverslips to confluence and placed in DMEM supplemented with 0.4% FBS for 16 h. The cultures were preincubated for 15 min in physiologic saline solution containing 140 mmol/liter NaCl, 5 mmol/liter KCl, 10 mmol/liter HEPES (pH 7.4), 1 mmol/liter NaHPO 4 , 1 mmol/liter CaCl 2 , 0.5 mmol/liter MgSO 4 , and 5 mmol/liter glucose, followed by the addition of 5 M fura-2 AM for 45 min at 37°C. The cells were washed twice with physiologic saline solution and immediately placed in a cuvette containing physiologic saline solution at 37°C. Fura-2 fluorescence was monitored at 510-nm emission wavelength during alternate excitation at 340 and 380 nm with a dual wavelength excitation light source spectrofluorometer (DeltaScan; Photon Technology International, Princeton, NJ) equipped with a magnetic stirrer and temperature control. Cell fluorescence was corrected for autofluorescence and fura-2 leakage with Mn 2ϩ .
RNA Isolation and Northern Blot Analysis-Total cellular RNA was isolated from confluent cultures by the single-step method employing guanidine thiocyanate/phenol/chloroform extraction as described by Chomczynski and Sacchi (29). RNA purity and quantity were determined spectrophotometrically. RNA (10 g) was electrophoresed through a 1% agarose, 6% formaldehyde gel and transferred to a nylon membrane (Hybond-Nϩ; Amersham Biosciences). RNA loading was assessed by ethidium bromide staining of ribosomal bands and by cohybridization with glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The membranes were exposed to x-ray film for autoradiography at several different times to ensure that the bands could be quantified by densitometry. The ␣1(I) collagen probe came from a rat ␣1(I) collagen cDNA that specifically binds human ␣1(I) collagen mRNA (12). The CTGF probe was generated by reverse transcription-PCR as described (12).
Luciferase Assay-IMR-90 myofibroblasts were plated in 6-well plates (350,000 cells/well) in DMEM supplemented with 10% FBS. After 24 h the cultures were washed with phosphate-buffered saline and incubated with the 3TP-LUX reporter (1 g/well) mixed with Lipofectamine 2000 according to the manufacturer's instructions (Invitrogen). After 4 h, DMEM supplemented with 10% FBS without antibiotics (3 ml/well) was added to the transfection medium, and the cultures were incubated for an additional 20 h. Before experimentation, the cultures were incubated in DMEM supplemented with 0.4% FBS without antibiotics for 16 h.

Role of Plasma Membrane Na ϩ /Ca 2ϩ Exchanger in B1
Receptor-mediated Ca 2ϩ Transients-We have previously reported that kinin B1 receptor activation increases cytosolic Ca 2ϩ in a manner that is kinetically distinct from the Ca 2ϩ increase following B2 receptor activation (12). However, the mechanisms that mediate the kinin-induced Ca 2ϩ responses remain uncharacterized. To determine the contribution of extracellular Ca 2ϩ to the kinin-stimulated increase of cytosolic Ca 2ϩ , myofibroblasts were assayed for changes in cytosolic Ca 2ϩ in medium prepared without Ca 2ϩ (nominally Ca 2ϩ -free). As we previously reported, B1 receptor activation with des-Arg 10 -kallidin induced a rise in cytosolic Ca 2ϩ . In nominally Ca 2ϩ -free medium, activation of the B1 receptor did not increase cytosolic Ca 2ϩ levels (Fig. 1A). These data strongly suggest that the des-Arg 10 -kallidin-stimulated increase of cytosolic Ca 2ϩ is dependent on entry of extracellular Ca 2ϩ . In contrast, B2 receptor activation with kallidin stimulates an increase of cytosolic Ca 2ϩ in the presence or absence of extracellular Ca 2ϩ (Fig. 1B).
Many mechanisms may mediate entry of extracellular Ca 2ϩ such as Ca 2ϩ channels and Na ϩ /Ca 2ϩ exchange. We initially examined the activity of L-type channels and found that the des-Arg 10 -kallidin-stimulated increase of cytosolic Ca 2ϩ was not sensitive to inhibition of L-type Ca 2ϩ channels with verapamil (data not shown). The Na ϩ dependence of the des-Arg 10kallidin-stimulated Ca 2ϩ response was examined in myofibroblasts incubated in nominally Na ϩ -free medium in which the Na ϩ content was iso-osmotically replaced with N-methyl D-glucamine and immediately assayed for changes in cytosolic Ca 2ϩ . The rise in cytosolic Ca 2ϩ following B1 receptor activation by des-Arg 10 -kallidin was abrogated in myofibroblasts incubated in nominally Na ϩ -free medium ( Fig. 2A). In contrast, B2 receptor activation in nominally Na ϩ -free medium induces a rapid and sustained rise in cytosolic Ca 2ϩ (Fig. 2B). Altogether, these observations indicate that des-Arg 10 -kallidin stimulates a Na ϩ -dependent Ca 2ϩ influx, suggesting that des-Arg 10 -kallidin activates Ca 2ϩ entry mode operation of the Na ϩ / Ca 2ϩ exchanger. Expression of Na ϩ /Ca 2ϩ exchanger isoforms 1 and 2 in IMR-90 myofibroblasts was determined by reverse transcription-PCR (data not shown).
We investigated the role of the Na ϩ /Ca 2ϩ exchanger in the kinin-stimulated Ca 2ϩ transients using KB-R7943, a novel isothiourea derivative that inhibits Ca 2ϩ entry mode operation of Na ϩ /Ca 2ϩ exchanger by interacting with an extracellular region of the exchanger (30 -32) but does not modulate Na ϩ /H ϩ exchanger activity (33). In KB-R7943-treated myofibroblasts, basal Ca 2ϩ levels are unchanged; however, des-Arg 10 -kallidin does not induce an increase of cytosolic Ca 2ϩ (Fig. 3). In contrast, B2 receptor activation with kallidin in the presence of KB-R7943 induces a rapid and sustained cytosolic Ca 2ϩ transient (data not shown). These results suggest that the increase of cytosolic Ca 2ϩ observed in des-Arg 10 -kallidin-stimulated myofibroblasts is mediated by Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger.
12-O-Tetradecanoylphorbol 13-Acetate (TPA) Attenuates B1 Receptor-mediated Ca 2ϩ Transients-We previously reported that agonist-stimulated levels of CTGF mRNA are reduced in TPA-treated IMR-90 myofibroblasts (12) presumably through modulation of PKC activity. Direct phosphorylation of Na ϩ / Ca 2ϩ exchanger isoform 1 by PKC has recently been reported in rat myocytes (22). We now report that basal Ca 2ϩ levels are not changed, and B1 receptor activation with des-Arg 10 -kallidin does not induce an increase of cytosolic Ca 2ϩ in TPA-treated myofibroblasts (Fig. 4A). In contrast, the B2 receptor-mediated increase of cytosolic Ca 2ϩ is only slightly diminished in TPAtreated myofibroblasts (Fig. 4B). We also examined the effects of the PKC inhibitors calphostin C and chelerythrine on the des-Arg 10 -kallidin-stimulated increase of cytosolic Ca 2ϩ . Fifteen-minute preincubation with either 500 nM calphostin C or 1 M chelerythrine does not alter basal cytosolic Ca 2ϩ levels and does not affect the B1 receptor-mediated increases of cytosolic Ca 2ϩ (data not shown). These findings suggest that PKC modulates Na ϩ /Ca 2ϩ exchanger activity and are consistent with other observations demonstrating down-regulation of Na ϩ /Ca 2ϩ exchanger activity by PKC (34,35). Thus, it appears that PKC regulates the exchanger, which in turn regulates CTGF production. However, at the moment we do not know which PKC isoform(s) regulate the exchanger nor which Na ϩ / Ca 2ϩ exchanger isoform(s) mediate the observed Ca 2ϩ fluxes.
Regulation of CTGF and ␣1(I) Collagen mRNAs by Extracellular Ca 2ϩ -The involvement of Ca 2ϩ influx pathways in TGF-␤1 signaling pathways has been proposed. Nesti et al. (36) described TGF-␤1-stimulated, nifedipine-sensitive Ca 2ϩ influx in human osteoblasts. McGowan et al. (37) reported that TGF-␤1 stimulated a Ca 2ϩ influx in SV40-transformed murine mesangial cells via cell membrane-associated inositol 1,4,5trisphosphate receptors. We hypothesized that Ca 2ϩ metabolism modulates TGF-␤1 signaling in human lung myofibroblasts. In Fig. 5, we demonstrate that TGF-␤1 induces a gradual rise in cytosolic Ca 2ϩ that continues to increase for as long as 40 min after stimulation (filled triangles), at which time the data accumulation was terminated. In contrast, there is no increase of cytosolic Ca 2ϩ in untreated myofibroblasts (filled diamonds). The rise in cytosolic Ca 2ϩ induced by TGF-␤1 is abrogated in the absence of extracellular Ca 2ϩ (data not shown). Although the kinetics of the TGF-␤1-induced increase in cytosolic Ca 2ϩ are distinct from those observed following B1 receptor activation, we examined the effects of KB-R7943 treatment on TGF-␤1-induced increase of cytosolic Ca 2ϩ . KB-R7943 blocked the rise in cytosolic Ca 2ϩ in response to TGF-␤1 (open circles). These data strongly suggest that the TGF-␤1-stimulated increase of cytosolic Ca 2ϩ is also mediated by Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger.
We then examined the role of extracellular Ca 2ϩ in the TGF-␤1-stimulated increase of CTGF mRNA. Myofibroblasts were incubated in nominally Ca 2ϩ -free DMEM and stimulated with TGF-␤1. Basal expression of CTGF is not altered in Ca 2ϩfree DMEM; however, both the des-Arg 10 -kallidin-and the TGF-␤1-stimulated increases of CTGF mRNA are attenuated (Fig. 6). These findings are consistent with a role for the Na ϩ / Ca 2ϩ exchanger mediating CTGF mRNA stimulation by both des-Arg 10 -kallidin and TGF-␤1. We also examined the role of extracellular Ca 2ϩ in ␣1(I) collagen mRNA expression by incubating the myofibroblasts in nominally Ca 2ϩ -free DMEM. As observed with CTGF mRNA, the TGF-␤1-stimulated increase of ␣1(I) collagen mRNA is attenuated in myofibroblasts incubated in nominally Ca 2ϩ -free DMEM (Fig. 7).
To determine the effect of removal of extracellular Ca 2ϩ on TGF-␤-stimulated responses, we transfected IMR-90 myofibroblasts with the TGF-␤-responsive, Smad-mediated luciferase reporter 3TP-LUX (38). The myofibroblasts were stimulated with TGF-␤1 in DMEM or in nominally Ca 2ϩ -free DMEM. TGF-␤1 stimulated an increase of luciferase expression in both DMEM and Ca 2ϩ -free DMEM (Fig. 8). These findings suggest that removal of extracellular Ca 2ϩ does not alter Smad-mediated TGF-␤1 signaling. We conclude that the attenuation of the TGF-␤-stimulated increase of CTGF mRNA by the absence of extracellular Ca 2ϩ is not Smad-mediated.
Effect of Na ϩ /Ca 2ϩ Exchanger Inhibitors on CTGF mRNA in Human Myofibroblasts-To demonstrate that the B1 receptorinduced stabilization of CTGF mRNA is mediated through the Na ϩ /Ca 2ϩ exchanger, myofibroblasts were treated with dichlorobenzamil (DCB) or KB-R7943. DCB did not change basal expression of CTGF mRNA. However, DCB blocked the des-Arg 10kallidin-stimulated increase of CTGF mRNA and attenuated the TGF-␤1-stimulated increase of CTGF mRNA (Fig. 9). Because DCB is an amiloride analog, albeit with low activity for blocking the Na ϩ /H ϩ exchanger, the effects of more specific Na ϩ /H ϩ exchanger inhibitors were also assessed. Basal or TGF-␤1-stimulated levels of CTGF mRNA were not altered in myofibroblasts treated with amiloride or the more specific Na ϩ /H ϩ exchange inhibitor, ethyl-isopropyl amiloride (data not shown). In KB-R7943-treated myofibroblasts, basal expression of CTGF mRNA did not change. However, activation of the B1 receptor with des-Arg 10 -kallidin failed to increase CTGF mRNA, and the TGF-␤1-stimulated increase of CTGF mRNA was attenuated.
Effect of KB-R7943 on CTGF mRNA in Human Osteoblasts and Microvascular Endothelial Cells-Recent findings reported by Stains and Gay (39) demonstrate a role for the Na ϩ /Ca 2ϩ exchanger in osteoblast collagen deposition. We hypothesized that CTGF would be produced in the human osteoblast-like cell line MG-63 in response to TGF-␤1. To demonstrate that TGF-␤1induced stabilization of CTGF mRNA is mediated through the Na ϩ /Ca 2ϩ exchanger, we treated MG-63 cells with KB-R7943. In KB-R7943-treated osteoblasts, basal expression of CTGF mRNA did not change. However, the TGF-␤1-stimulated increase of CTGF mRNA was attenuated in KB-R7943-treated cells (Fig.  10). Because CTGF was originally described in endothelial cells, we examined changes in CTGF mRNA in TGF-␤1-stimulated human microvascular endothelial cells that were treated with KB-R7943. As with the osteoblasts, treatment with KB-R7943 did not alter basal expression of CTGF mRNA, but the TGF-␤stimulated increase of CTGF was attenuated (data not shown).
Effect of Na ϩ /Ca 2ϩ Exchanger Inhibitors on ␣1(I) Collagen and ␣ Smooth Muscle Actin Expression-We also examined the effect of inhibiting Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger on ␣1(I) collagen mRNA. Northern blot analyses indicate that, in KB-R7943-treated myofibroblasts, basal expression of ␣1(I) collagen mRNA and the TGF-␤1-stimulated increase of ␣1(I) collagen mRNA were strongly down-regulated (Fig. 11). To further characterize this modulation of the myofibroblast phenotype, we examined the expression of ␣ smooth muscle actin mRNA in the presence of KB-R7943. IMR-90 myofibroblasts preincubated with KB-R7943 were stimulated with TGF-␤1, and changes in ␣ smooth muscle actin mRNA were monitored by real time PCR (Fig. 12) and confirmed by Northern blot analyses (data not shown). In untreated myofibroblasts, TGF-␤1 stimulated an increase of expression of ␣ smooth muscle actin mRNA. In KB-R7943-treated myofibroblasts, basal levels and TGF-␤-stimulated levels of ␣ smooth muscle actin mRNA were strongly down-regulated. These results strongly suggest that inhibition of Na ϩ /Ca 2ϩ exchanger activity down-regulates the myofibroblast phenotype.
The mechanisms by which Na ϩ /Ca 2ϩ exchanger activity regulates CTGF, ␣1(I) collagen, and ␣ smooth muscle actin mRNA levels are unclear. The Ca 2ϩ /calmodulin-dependent protein phosphatase, calcineurin, has been shown to participate in TGF-␤-stimulated extracellular matrix accumulation (42). In turn, calcineurin has been shown to regulate transcription of Na ϩ /Ca 2ϩ exchanger isoforms (43). However, we did not observe changes in CTGF, ␣1(I) collagen, and ␣ smooth muscle actin mRNA levels in quiescent and TGF-␤-stimulated myofibroblasts that were treated for 4 or 16 h with the calcineurin inhibitor, cyclosporin A (data not shown). Further experimentation is required to definitively conclude that calcineurin does not participate in these responses. DISCUSSION In this report, we characterize the role of the Na ϩ /Ca 2ϩ exchanger in fibrogenesis. Using des-Arg 10 -kallidin-and TGF- ␤1-stimulated human myofibroblasts, we investigated the role of the Na ϩ /Ca 2ϩ exchanger in early agonist-stimulated signal transduction events. Our results demonstrate that activation of the kinin B1 receptor induces an increase of cytosolic Ca 2ϩ that is mediated through Ca 2ϩ entry mode operation of the Na ϩ / Ca 2ϩ exchanger. We previously reported that activation of the B1 receptor (but not the B2 receptor) increases the stability of CTGF mRNA (12). The present results indicate that the B1 receptor-induced increase of CTGF mRNA is blocked by KB-R7943. From these results, we conclude that Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger is required for the B1 receptor-mediated increase in CTGF mRNA stability. These results suggest that the Na ϩ /Ca 2ϩ exchanger is functionally coupled to B1 receptor activation and that CTGF mRNA stability is modulated by cation homeostasis. It is important to note that a rise in cytosolic Ca 2ϩ per se is not sufficient to induce an increase in CTGF mRNA. Although increased cytosolic Ca 2ϩ is required for the TGF-␤1-and the B1 receptorstimulated increases of CTGF mRNA, kallidin, which also mobilizes intracellular Ca 2ϩ , does not induce an increase in CTGF mRNA (12).
TGF-␤1 also induces KB-R7943-sensitive increases in cytosolic Ca 2ϩ and CTGF mRNA, suggesting that TGF-␤1 induces an increase of CTGF mRNA stability in addition to the previously reported increase in CTGF transcription. However, regulation of TGF-␤1-stimulated CTGF transcription by Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger cannot be ruled out at this moment. Furthermore, basal expression of CTGF mRNA is not sensitive to inhibition of Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger. Taken together, these results suggest that in quiescent human lung myofibroblasts expression of CTGF may be mediated through both TGF-␤-dependent and TGF-␤-independent mechanisms. These results are important because they show that two divergent agonists modulate CTGF mRNA levels via convergence on Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger, suggesting a commonality for this mechanism in CTGF mRNA stability. Furthermore Na ϩ / Ca 2ϩ exchanger activity appears to play a central role in the TGF-␤-stimulated increase of CTGF mRNA in human lung myofibroblasts, microvascular endothelial cells, and osteoblastlike cells.
We previously reported that TGF-␤1 stimulates CTGF transcription (12), and others have shown this process to be mediated via Smad signaling (40,41). The luciferase experiments demonstrate that, in TGF-␤1-stimulated myofibroblasts, Smad-mediated responses are not significantly affected by removal of extracellular Ca 2ϩ , suggesting that Smad-mediated signaling is not affected by changes in Na ϩ /Ca 2ϩ exchanger activity. In addition, these experiments demonstrate that the myofibroblasts are responsive to TGF-␤1 stimulation in the absence of extracellular Ca 2ϩ . Thus, it appears that both genomic and nongenomic mechanisms modulate the TGF-␤stimulated increase of CTGF mRNA.
Quiescent IMR-90 myofibroblasts express ␣1(I) collagen and ␣ smooth muscle actin, genes characteristically expressed in myofibroblasts. We show that inhibition of Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger attenuates basal and TGF-␤1-stimulated levels of ␣1(I) collagen and ␣ smooth muscle actin mRNAs, suggesting that Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger is required for maintenance of the myofibroblasts phenotype. Together, these findings suggest that the Na ϩ /Ca 2ϩ exchanger, functioning as a Ca 2ϩ influx pathway, plays an important and novel role in fibrogenesis. Consistent with this hypothesis, Stains and Gay (39) have recently shown by immunocytochemistry that the secretion of type (I) collagen and bone sialoproteins is reduced in KB-R7943-treated osteoblasts. A relationship among TGF-␤, fibrogenesis, and the Na ϩ /Ca 2ϩ exchanger has been suggested by prior studies reporting an increase in Na ϩ /Ca 2ϩ exchanger mRNA expression in the setting of fibrosis or pro-fibrotic agents. Nakamura et al. (44) reported that fibrosis is associated with increased expression of the Na ϩ /Ca 2ϩ exchanger in hepatic stellate cells. Carrillo et al. (45) showed that TGF-␤ increases Na ϩ /Ca 2ϩ exchanger mRNA levels via a PKC-sensitive pathway. In vivo induction of fibrosis in rats with CCl 4 was also associated with an increase in liver Na ϩ /Ca 2ϩ exchanger mRNA. Our studies extend these observations, highlighting the necessity of the Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger for CTGF mRNA expression and fibrogenesis in human lung myofibroblasts. Taken together, these results suggest that Na ϩ /Ca 2ϩ exchanger expression is a biomarker for fibrosis and a potential target for intervention.
In conclusion, our data suggest that Ca 2ϩ entry mode operation of the Na ϩ /Ca 2ϩ exchanger plays an integral role in the expression of CTGF, ␣1(I) collagen, and ␣ smooth muscle actin, implicating an important role for the Na ϩ /Ca 2ϩ exchanger in the maintenance of the myofibroblast phenotype. Furthermore, our data demonstrate that regulation of CTGF expression by the Na ϩ /Ca 2ϩ exchanger is a mechanism shared by various cell types and that calcium metabolism plays an important role in the expression of genes associated with fibrosis. Based on these findings we speculate that targeting Na ϩ /Ca 2ϩ exchanger activity may be an efficacious intervention for the treatment of fibrosis.