Transforming Growth Factor- Receptor Type 1 (TGF RI) Kinase Activity but Not p38 Activation Is Required for TGF RI-Induced Myofibroblast Differentiation and Profibrotic Gene Expression

Transforming growth factor(TGF ) is a major mediator of normal wound healing and of pathological conditions involving fibrosis, such as idiopathic pulmonary fibrosis. TGF also stimulates the differentiation of myofibroblasts, a hallmark of fibrotic diseases. In this study, we examined the underlying processes of TGF RI kinase activity in myofibroblast conversion of human lung fibroblasts using specific inhibitors of TGF RI (SD-208) and p38 mitogen-activated kinase (SD-282). We demonstrated that SD-208, but not SD-282, inhibited TGF -induced SMAD signaling, myofibroblast transformation, and collagen gel contraction. Furthermore, we extended our findings to a rat bleomycin-induced lung fibrosis model, demonstrating a significant decrease in the number of myofibroblasts at fibroblastic foci in animals treated with SD-208 but not those treated with SD-282. SD-208 also reduced collagen deposition in this in vivo model. Microarray analysis of human lung fibroblasts identified molecular fingerprints of these processes and showed that SD-208 had global effects on reversing TGF -induced genes involved in fibrosis, inflammation, cell proliferation, cytoskeletal organization, and apoptosis. These studies also revealed that although the p38 pathway may not be needed for appearance or disappearance of the myofibroblast, it can mediate a subset of inflammatory and fibrogenic events of the myofibroblast during the process of tissue repair and fibrosis. Our findings suggest that inhibitors such as SD-208 may be therapeutically useful in human interstitial lung diseases and pulmonary fibrosis. Pulmonary fibrosis is a pathological hallmark of interstitial lung diseases (Green, 2002). A common feature of fibrosis is the increased deposition of the extracellular matrix (ECM). Myofibroblasts are instrumental in fibrogenic processes in pulmonary fibrosis, in that they are the major producers of ECM proteins. Commonly identified by the expression of -smooth muscle actin, myofibroblasts are considered an intermediate between a fibroblast and a true smooth muscle cell. As such, they are involved in wound healing and contraction. There are substantial data showing the presence of myofibroblasts in lung tissues from patients with pulmonary fibrosis (Phan, 2002). TGF , a major player in the conversion of fibroblasts to myofibroblasts, is a pleiotropic growth factor involved in multiple biological processes including cell proliferation, fibrosis, tissue repair, inflammation, apoptosis, cell differentiation, cell adhesion, and motility (Sporn and Roberts, 1992; Massague et al., 1994; Grande, 1997). TGF levels are elevated in the bronchoalveolar lavage fluid of patients with pulmonary fibrosis (Kuroki et al., 1995; Ludwicka et al., 1995) and in the N.J.G. and Y.W. contributed equally to this work □S The online version of this article (available at http://molpharm. aspetjournals.org) contains supplemental material. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.105.021600. ABBREVIATIONS: ECM, extracellular matrix; TGF , transforming growth factor; TGF R, transforming growth factorreceptor; MAPK, p38 mitogen-activated kinase; HLF, human lung fibroblast; RT-PCR, reverse transcription-polymerase chain reaction; ELISA, enzyme-linked immunosorbent assay; FGM, fibroblast growth medium; BSA, bovine serum albumin; SMAD, mothers against DPP homolog; CTGF, connective tissue growth factor; PAI-1, plasminogen activator inhibitor, type I; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PBS, phosphate-buffered saline; DMEM, Dulbecco’s modified Eagle’s medium; PEG, polyethylene glycol. 0026-895X/06/7002-518–531$20.00 MOLECULAR PHARMACOLOGY Vol. 70, No. 2 Copyright © 2006 The American Society for Pharmacology and Experimental Therapeutics 21600/3129610 Mol Pharmacol 70:518–531, 2006 Printed in U.S.A. 518 http://molpharm.aspetjournals.org/content/suppl/2006/05/18/mol.105.021600.DC1 Supplemental material to this article can be found at: at A PE T Jornals on Jne 1, 2017 m oharm .aspeurnals.org D ow nladed from lungs of animals with experimentally induced lung fibrosis (Coker et al., 1997). TGF signals through heteromeric receptor complexes consisting of type I (TGF RI) and type II (TGF RII) serine-threonine kinase receptors. Upon ligand binding, TGF RII transphosphorylates TGF RI. The activated receptor then phosphorylates and activates members of the SMAD family of proteins. Receptor SMADs subsequently bind to coSMAD4 and translocate into the nucleus, where they regulate the transcriptional response of target genes. Besides signaling through SMADs, several additional pathways have been shown to mediate the downstream signaling from TGF RI. These include the p38 mitogen-activated kinase (MAPK) (Bhowmick et al., 2001; Bakin et al., 2002), the c-Jun NH2-terminal kinase (Atfi et al., 1997), the extracellular-regulated kinase (Frey and Mulder, 1997), and the phosphatidylinositol kinase pathways (Bakin et al., 2000). The p38 pathway is involved in a variety of biological responses to the TGF signaling, such as migration of smooth muscle cells (Hedges et al., 1999), neuronal differentiation of PC-12 cells (Iwasaki et al., 1999), chondrogenesis of ATDC-5 cells (Nakamura et al., 1999), and cardiomyocyte differentiation (Monzen et al., 1999). Despite the numerous implications of p38 signaling in the TGF pathway, the significance of TGF -mediated p38 signaling in pulmonary fibrosis and myofibroblast transformation is not well defined. In this study, we used a combination of cDNA arrays and small molecule inhibitors of TGF -R1 and p38 MAP kinase to begin to delineate the molecular events downstream of TGF -R1. We identify a number of TGF -regulated genes and assess which genes are regulated by p38. In addition, we show that small molecule inhibitors of TGF RI, but not p38, block SMAD signaling, inhibit an in vitro model of fibrotic tissue remodeling (fibroblast mediated collagen gel contraction), and prevent the transformation of fibroblasts to myofibroblasts in vivo and in vitro, as measured by expression of -smooth muscle actin. Materials and Methods Cell Culture and Inhibitor Reagents. Three lots of primary human lung fibroblast (HLF) cells, derived from a 40-year-old woman (donor A, lot 8F162), a 10-year-old boy (donor B, lot 17423), and a 51-year-old woman (donor C, lot 4F0512), were provided by Cambrex Bio Science Walkersville, Inc. (Walkersville, MD). Realtime RT-PCR experiments were performed on samples from all donors. The microarray and gel contraction experiments were performed on cells from donor A, and the ELISA and Western blot data were assayed using samples from donors A and C. Cells were seeded at 4 10 cells (passage 4) in 100-mm dishes and cultured in complete fibroblast growth medium (FGM; Cambrex Bio Science Walkersville, Inc.). The next day, medium was changed to FGM without serum or fibroblast growth factor but supplemented with 0.2% bovine serum albumin and 50 g/ml vitamin C. Cells for realtime RT-PCR and microarray were serum-deprived for 24 h before treating with 5 ng/ml TGF 1 (R&D Systems, Minneapolis, MN) in the presence or absence of 400 nM SD-208 or 400 nM SD-282. Concentrations of inhibitors were chosen based on IC50s in HLF cell-based assays (data not shown and Fig. 2; IC50, 70–80 nM). Induction with TGF was carried out for various times (7.5, 24, and 72 h) after which RNA was harvested by lysing the cells in RLT buffer (QIAGEN, Valencia, CA) and frozen at 80°C. SD-208 and SD-282 Kinase Assays. TGF RI kinase inhibitor SD-208 and p38 MAP kinase inhibitor SD-282 were synthesized by the Medicinal Chemistry Department at Scios, Inc. and dissolved in dimethyl sulfoxide. They were designed from medicinal chemistry efforts optimizing high-throughput screening leads into potent, selective inhibitors with acceptable pharmaceutical properties. SD-282 is an indole-5-carboxamide, ATP-competitive inhibitor of p38 MAP kinase, and SD-208 is a 2,4-disubstituted pteridine, ATP-competitive inhibitor of TGF RI kinase (Fig. 1). SD-282 was prepared by functionalizing the 3-position of the corresponding indole-5-carboxamide through treatment with oxalyl chloride in methylene chloride followed by addition of dimethylamine. The resulting material was purified by silica gel chromatography and then converted to its hydrochloride salt form. SD-208 was prepared by heating 4-chloro2-(5-chloro-2-fluorophenyl)pteridine with 4-aminopyridine and triethylamine in dimethylformamide. The resulting material was purified by silica gel chromatography and then converted to its hydrochloride salt form. Kinase assays were performed by standard methods (Davies et al., 2000). In brief, kinases were assayed by measuring the incorporation of radiolabeled ATP into a peptide or a protein substrate. Reactions were performed in 96-well plates and included the relevant kinase, substrate, ATP, and appropriate cofactors. The reactions were incubated and then stopped by the addition of phosphoric acid. Substrate was captured onto a phosphocellulose filter, which was washed free of unreacted [ -P]ATP from PerkinElmer Life and Analytical Sciences (Boston, MA). The counts incorporated were determined by counting on a microplate scintillation counter (TopCount; PerkinElmer Life and Analytical Sciences). The ability of compounds to inhibit the kinase was determined by comparing the counts incorporated in the presence of compound to those in the absence of compound. Western Blot and ELISA Analysis. HLF cells were serumstarved for 24 h in FGM (Cambrex Bio Science Walkersville, Inc.) containing no serum or FGF but supplemented with 0.2% BSA and 50 g/ml vitamin C. To analyze SMAD phosphorylation, cells were treated with 2 ng/ml TGF 1 for 30 min in the presence of increasing concentrations of SD-208 or SD-282 inhibitor as indicated. To analyze induction of -smooth muscle actin, CTGF, and PAI-1, cells were treated with 2 ng/ml TGF 1 for 7, 24, or 72 h in the presence of 400 nM SD-208 or SD-282. Total cell lysates were prepared in MPER buffer (mammalian protein extraction reagent; Pierce, Rockford, IL) or in radioimmunoprecipitation assay (10 nM Tris, pH 8.0, 150 mM NaCl, 1% Triton X-100, 1% deoxycholate, and 0.1% SDS; Roche, Indianapolis, IN) containing protease and phosphatase inhibitor cocktails. Aliquots of the cell lysates were fractionated on SDSpolyacrylamide gels and transferred to nitrocellulose membranes (Invitrogen, Carlsbad, CA). The membranes were blocked against nonspecific binding using 4% skim milk. Proteins were detected using specific primary antibodies and peroxidase-conjugated secondary antibodies. The antibody against phospho-SMAD2 was from Cell Signaling Technology (Beverly, MA); the mouse monoclonal antibody against SMAD2 was purchased from BD Biosciences (Palo Alto, CA); the CTGF polyclonal antibody was custom made (U. Schellenberger, Scios Inc.); the monoclonal antibody against -smooth muscle actin Fig. 1. Structures of SD-208 and SD-282. TGF RI and p38 Signaling in Human Lung Fibroblasts 519 at A PE T Jornals on Jne 1, 2017 m oharm .aspeurnals.org D ow nladed from was purchased from Sigma (St. Louis, MO); the mouse monoclonal antibody against GAPDH was from Biogenesis (Poole, Dorset, UK); and the mouse monoclonal antibody against vimentin was from Affinity Bioreagents (Golden, CO). The blots were visualized by the SuperSignal West Femto detection system (Pierce) or the Amersham ECL detection system followed by quantitation using Image Quant 5.2 (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Phospho-SMAD ELISA assays were performed in 96-well plates coated with anti-SMAD2/3 monoclonal antibodies (100 ng/well; BD Biosciences) for 18 h at 4°C or 2 h at room temperature. To block nonspecific binding, excess antibody was removed and the wells were treated with blocking buffer (0.3% BSA/PBS) for 2 h at room temperature. The wells were rinsed three times with wash buffer (0.5% Tween 20/PBS), and cell lysates (10 g of total protein) were added to each well and incubated overnight at 4°C. Wells were rinsed three times with wash buffer before adding a polyclonal anti-phosphoSMAD2/3 antisera diluted in 2% BSA/0.5% Tween 20/PBS. After a 2-h incubation at room temperature, wells were washed three times with wash buffer before developing with tetramethyl benzidine (Sigma). Reactions were stopped with 0.5 N H2SO4 after a 5to 30-min incubation, and plates were read at an optical density of 450 nm in a SpectraMax 250 plate reader (Molecular Devices, Sunnyvale, CA). For PAI-1 ELISA analysis, Imulyse (Biopool International, Ventura, CA) kits were used according to the manufacturer’s instructions. In Vitro Immunofluorescence Analysis. To examine the TGF -induced nuclear translocation of SMAD2, HLF cells were grown to 50 to 80% confluence in Lab-Tek Chamber Slides (Nalge Nunc International, Naperville, IL). Cells were serum-starved and treated with TGF 1 (2 ng/ml) and inhibitors as described above. After treatment, cells were washed with PBS and fixed for 15 min with 4% paraformaldehyde in PBS. Cells were then treated with 0.1% saponin in PBS for 10 min. After washing off the detergent, fixed cells were incubated with primary antibodies overnight at 4°C. Specific antibodies for SMAD2 were from Zymed Laboratories (South San Francisco, CA). After extensive washing, the biotinylated antirabbit IgG and fluorescein avidin D secondary antibodies (Vector Laboratories, Burlingame, CA) were added. The fluorescence images of SMAD2 were visualized by a Nikon microscope using Image-Pro plus 4.5 software (Media Cybernetics, Inc., Silver Spring, MD). Collagen Gel Contraction. Native type I collagen (rat tail tendon collagen) was extracted from rat tail tendons by a method published previously (Bell et al., 1979; Mio et al., 1996). In brief, tendons were excised from rat tails without tendon sheath and other connective tissues. Repeated washing with Tris-buffered saline was followed by dehydration and sterilization with 50%, 75%, 95% and pure ethanol. Type I collagen was then extracted in 6 mM hydrochloric acid at 4°C. The supernatant was harvested by centrifugation at 3000g for 1 h at 4°C. Collagen concentration was determined by weighing a lyophilized aliquot from each lot of collagen solution. Gels were prepared using a method described previously (Mio et al., 1996, 1998) that involves mixing rat tail tendon collagen, distilled water, and 4 Dulbecco’s modified Eagle’s medium so that the final mixture resulted in a physiologic ionic strength, 1 Dulbecco’s modified Eagle’s medium, and a pH of 7.40. Cells were trypsinized (trypsin-EDTA; 0.05% trypsin, 0.53 mM EDTA-4Na; Invitrogen) and suspended in 10 ml of complete FGM and counted with a Coulter Counter (Beckman Coulter, Fullerton, CA). HLF cells were pelleted and resuspended in basal FGM without serum or other growth factors at a density of 10 cells/ml. Cells were then mixed with the neutralized collagen solution so that the final cell density in the collagen solution was 5 10 cells/ml, and the final concentration of collagen was 0.75 mg/ml. Aliquots (0.5 ml/well) of the mixture of cells in collagen were cast into each well of 24-well tissue culture plates (BD Biosciences). After gelation was completed, within 20 min at room temperature, the gels were gently released from the 24-well tissue culture plates and transferred into 60-mm tissue culture dishes (three gels in each dish) that contained 5 ml of freshly prepared basal FGM with or without TGF1 (200 pmol/l) or inhibitors. The gels were then incubated at 37°C in a 5% CO2 atmosphere for 2 days, and the area of each gel was measured with an Optomax V image analyzer (Optomax, Burlington, MA) daily. Data were expressed as the percentage area compared with the original gel size. Histology and Immunohistochemistry in the Rat Bleomycin Model. Sprague-Dawley rats were administered bleomycin for 14 days. Rats were intubated and aerosolized with 200 l of saline or 1.0 unit of bleomycin per rat (three to four rats per group). One day after bleomycin challenge rats were treated twice daily with vehicle or vehicle plus SD-208 (60 mg/kg in 1% methyl cellulose) or SD-282 (60 mg/kg in 1% PEG-400). Inhibitor doses were chosen based on efficacious doses from our previous studies (Li et al., 2004; Bonniaud et al., 2005). In addition, the inhibitor doses used in this study have been shown to have pharmacological effects in the lungs of bleomycin-challenged rats as demonstrated by the inhibition of CTGF mRNA by SD-208 and of COX2 mRNA by SD-282 (data not shown). Lungs were removed en bloc; they were inflated with 4% formalin at a constant pressure of 15 cm of water and then fixed in 10% formalin for 48 h. The left lung was cut perpendicular to the tracheobronchial tree into sections. Tissue sections were processed, embedded in paraffin, cut into 5m sections, and immunohistochemically stained using mouse antibodies against -smooth muscle actin (Chemicon International, Inc., Temecula, CA). Negative control sections were run in parallel with normal mouse IgG2 diluted to the same concentration as the primary antibody (data not shown). The number of myofibroblasts was measured under the Nikon E600 light microscope equipped with a Spot digital camera at magnification of 400 for 36 to 40 fields (for each animal). Positively stained smooth muscle cells surrounding major airways and blood vessels were excluded from the analysis. Measurement of Hydroxyproline Content. To quantify lung collagen content as an indicator of pulmonary fibrosis, the hydroxyproline content in whole lungs was measured in all animals (seven to eight animals per group) according to published methods (Woessner, 1961). In brief, lung samples were minced into fine pieces with a scissors and homogenized thoroughly in 15 ml of 1 PBS with a Polytron homogenizer (Kinematica, Basel, Switzerland). One milliliter of whole homogenate was precipitated with 0.25 ml of ice-cold 50% (w/v) trichloroacetic acid. The precipitate was hydrolyzed in 2.0 ml of 6 N HCL for 18 h at 110°C. Hydroxyproline content was determined after neutralization. The results were calculated as micrograms of hydroxyproline per whole lung using hydroxyproline standards from Sigma (St. Louis, MO). cDNA Microarray. Gene expression profiles were determined from cDNA microarrays as described previously (Kapoun et al., 2004). In brief, arrays containing 8600 elements were derived from clones isolated from normalized cDNA libraries or purchased from ResGen (Invitrogen). Clones were sequence verified, and differentially expressed genes were reconfirmed by NCBI BLAST analyses. Differential expression values were expressed as the ratio of the median of background-subtracted fluorescent intensity of the experimental RNA to the median of background-subtracted fluorescent intensity of the control RNA. For ratios greater than or equal to 1.0, the ratio was expressed as a positive value. For ratios less than 1.0, the ratio was expressed as the negative reciprocal (e.g., a ratio of 0.5 2.0). Differential expression ratios were determined as the mean of the two values from dye-swapped duplicates. Expression data were rejected if neither channel produced a signal of at least 2.0-fold over background. Statistically significant differential expression threshold values were determined according to the method of Yang et al. (2002). Hierarchical clustering was used to visualize the data and to group genes into similar expression patterns (Spotfire, Somerville, MA). The data were prepared for clustering by using the log base 10 of the median expression values and then normalized by the Z-score method within Spotfire. mRNA Isolation, Labeling, and Hybridizations. Total RNA was extracted from cells using QIAGEN s RNeasy kit (Valencia, CA). 520 Kapoun et al. at A PE T Jornals on Jne 1, 2017 m oharm .aspeurnals.org D ow nladed from RNA was amplified using a modified Eberwine protocol (Kapoun et al., 2004) that incorporated a poly(A) tail into the amplified RNA. Fluorescently labeled cDNA probes were generated by reverse transcription of 4 g of RNA with SuperScript II (Invitrogen) using anchored dT primers in the presence of Cy3 or Cy5 dUTP as described previously (Kapoun et al., 2004). Hybridization of each fluorophore was quantified using an Axon GenePix 4000A scanner. Real-Time RT-PCR. Real-time RT-PCR was performed in a twostep manner. cDNA synthesis and real-time detection were carried out in a PTC-100 Thermal Cycler (MJ Research Inc., Waltham, MA) and an ABI Prism 7900 Sequence Detection System (Applied Biosystems, Foster City, CA), respectively. Random hexamers (QIAGEN) were used to generate cDNA from 200 ng of RNA as described in Applied Biosystems User Bulletin 2. TaqMan PCR Core Reagent Kit or TaqMan Universal PCR Master Mix (Applied Biosystems) were used in subsequent PCR reactions according to the manufacturer’s protocols. Relative quantitation of gene expression was performed using the relative standard curve method. Sequence specific primers and probes were designed using Primer Express software (ver. 2; Applied Biosystems). Sequences of primers and probes can be found in Supplemental Table E1. Expression levels were normalized to 18 S rRNA. All real-time RT-PCR reactions were performed in triplicate on each of the three biological replicates from each donor. Statistical Analysis. The experiments were usually performed 3 to 4 times with similar results. Significance was tested by oneor two-tailed analysis of variance followed by Bonferroni’s correction using PRISM4 software (GraphPad Software Inc. San Diego, CA) unless otherwise indicated.