Pirfenidone exerts antifibrotic effects through ...

The FASEB Journal article fj.201600892RR. Published online February 1, 2017. THE

JOURNAL

• RESEARCH •

www.fasebj.org

Pirfenidone exerts antifibrotic effects through inhibition of GLI transcription factors Miroslava Didiasova,*,1 Rajeev Singh,†,1 Jochen Wilhelm,‡,§ Grazyna Kwapiszewska,{ Lukasz Wujak,* Dariusz Zakrzewicz,* Liliana Schaefer,k Philipp Markart,‡,§ Werner Seeger,‡,§ Matthias Lauth,†,1 and Malgorzata Wygrecka*,§,1,2

*Department of Biochemistry and ‡Department of Internal Medicine, Universities of Giessen and Marburg Lung Center, Giessen, Germany; † Institute of Molecular Biology and Tumor Research (IMT), Center for Tumor Biology and Immunology, Philipps University, Marburg, Germany; §German Center for Lung Research, Justus-Liebig University, Giessen, Germany; {Ludwig Boltzmann Institute for Lung Vascular Research, Graz, Austria; and kInstitute of Pharmacology and Toxicology, Goethe University School of Medicine, Frankfurt am Main, Germany

Pirfenidone is an antifibrotic drug, recently approved for the treatment of patients suffering from idiopathic pulmonary fibrosis (IPF). Although pirfenidone exhibits anti-inflammatory, antioxidant, and antifibrotic properties, the molecular mechanism underlying its protective effects remains unknown. Here, we link pirfenidone action with the regulation of the profibrotic hedgehog (Hh) signaling pathway. We demonstrate that pirfenidone selectively destabilizes the glioma-associated oncogene homolog (GLI)2 protein, the primary activator of Hhmediated gene transcription. Consequently, pirfenidone decreases overall Hh pathway activity in patients with IPF and in patient-derived primary lung fibroblasts and leads to diminished levels of Hh target genes such as GLI1, Hh receptor Patched-1, a-smooth muscle actin, and fibronectin and to reduced cell migration and proliferation. Interestingly, Hh-triggered TGF-b1 expression potentiated Hh responsiveness of primary lung fibroblasts by elevating the available pool of glioma-associated oncogene homolog (GLI)1/GLI2, thus creating a vicious cycle of amplifying fibrotic processes. Because GLI transcription factors are not only crucial for Hh-mediated changes but are also required as mediators of TGF-b signaling, our findings suggest that pirfenidone exerts its clinically beneficial effects through dual Hh/TGF-b inhibition by targeting the GLI2 protein.—Didiasova, M., Singh, R., Wilhelm, J., Kwapiszewska, G., Wujak, L., Zakrzewicz, D., Schaefer, L., Markart, P., Seeger, W., Lauth, M., Wygrecka, M. Pirfenidone exerts antifibrotic effects through inhibition of GLI transcription factors. FASEB J. 31, 000–000 (2017). www.fasebj.org

ABSTRACT:

KEY WORDS:

idiopathic pulmonary fibrosis

•

hedgehog signaling

Idiopathic pulmonary fibrosis (IPF) represents one of the most common forms of diffuse parenchymal lung diseases and is characterized by the histologic appearance of the usual interstitial pneumonia pattern. IPF is characterized by injury and activation of alveolar epithelial cells, (myo)fibroblast proliferation with formation of fibroblast foci, and excessive deposition of extracellular matrix proteins (1, 2). The overall prognosis of IPF is very poor, with a ABBREVIATIONS: FN, fibronectin; GLI, glioma associated oncogene homo-

log; Hh, hedgehog; HLF, human lung fibroblast; IPF, idiopathic pulmonary fibrosis; MEF, mouse embryonic fibroblast; PBGD, porphobilinogen deaminase gene; PTCH, protein patched homolog; qRT-PCR, quantitative RT-PCR; SAG, smoothened agonist; SHH, sonic hedgehog; SMA, smooth muscle actin; SMO, smoothened; SUFU, suppressor of fused homolog 1 2

These authors contributed equally to this work. Correspondence: Universities of Giessen and Marburg Lung Center, Department of Biochemistry, Friedrichstrasse 24, 35392 Giessen, Germany. E-mail: [email protected]

doi: 10.1096/fj.201600892RR This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

0892-6638/17/0031-0001 © FASEB

•

TGF-b signaling

median survival of 2–3 yr after diagnosis; this survival rate is worse than for many types of cancer (3). Pirfenidone [5-methyl-1-phenyl-2-(1H)-pyridone], an orally available synthetic drug, was approved for the treatment of mild to moderate IPF in 2011 in the European Union and in 2014 in the United States. To date, 4 randomized, placebo-controlled, phase III studies have demonstrated that pirfenidone significantly slows disease progression (4–7). Combined data from the Pirfenidone in Patients with Idiopathic Pulmonary Fibrosis (CAPACITY) and Assessment of Pirfenidone to Confirm Efficacy and Safety in Idiopathic Pulmonary Fibrosis (ASCEND) trials showed a reduced mortality in patients with IPF receiving pirfenidone (4, 5). On the basis of available clinical trial evidence, a (conditional) positive recommendation has been made for the treatment of patients with IPF with pirfenidone in the 2015 international IPF guidelines (8). Pirfenidone demonstrates anti-inflammatory, antioxidant, and antifibrotic effects, as shown in several in vitro studies and animal models of pulmonary fibrosis (9–11). Although it has been demonstrated that pirfenidone exerts

Downloaded from www.fasebj.org to IP 137.248.205.183. The FASEB Journal Vol., No. , pp:, February, 2017

1

antifibrotic effects through blockade of TGF-b signaling (12), the molecular mechanism of its action remains unknown. Accumulating evidence suggests that the hedgehog (Hh) pathway, a major regulator of many processes in embryonic development, is reactivated in the adult lung after injury (13, 14). The Hh pathway may be activated via canonical or noncanonical mode of action (15). In canonical Hh signaling, binding of the Hh ligands to their receptor Patched (PTCH-1, PTCH-2) leads to the derepression of smoothened (SMO), resulting in the activation of gliomaassociated oncogene homolog (GLI)1, GLI2, and GLI3 transcription factors (16). In noncanonical Hh signaling, GLI activity is regulated independently of SMO by other signaling pathways, including TGF-b (15). In IPF lungs, the Hh ligand sonic hedgehog (SHH) and the transcription factor GLI2 were found to be elevated in epithelial cells, whereas the SHH receptor PTCH-1, SMO, and GLI1 were shown to be up-regulated in fibroblasts/myofibroblasts (13, 14). Increased expression of SHH and GLI1 was also observed in the lungs of bleomycin-treated mice (14). Therapeutic interference with upstream elements of the Hh cascade, such as SHH or SMO, demonstrated little effect, whereas the blockade of downstream components of the GLI family inhibited experimental lung fibrosis (17). These findings argue for a role of noncanonical Hh pathway modulation and stress the importance of GLI function in IPF. Based on these findings, we investigated whether pirfenidone might exert antifibrotic effects via interference with the Hh pathway.

after histopathological evaluation. Table 1 provides demographic and clinical characteristics of the patient cohort. Cell isolation Primary human lung fibroblasts (HLFs) were isolated from donor (n = 10) and IPF (n = 10) lungs as previously described (18). Cell culture and cell stimulation Human embryonic kidney cells HEK-293A and mouse NIH3T3 fibroblasts carrying a stably integrated Hh reporter plasmid ShhL2 were purchased from American Type Culture Collection (Manassas, VA, USA). Mouse embryonic fibroblasts (MEFs) genetically depleted of Suppressor of Fused (Sufu2/2) were a kind gift from Rune Toftgard (Karolinska Institute, Stockholm, Sweden). Gli32/2 MEFs and MEF[SHH] were kindly provided by Wade Bushman (Molecular and Environmental Toxicology Center, University of Wisconsin Medical School, Madison, WI, USA). Ptch-12/2 MEFs were kindly provided by Jussi Taipale and Philip Beachy (Johns Hopkins University, Baltimore, MD, USA). All cells were cultured in a 5% CO2 incubator at 37°C in DMEM (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum (Hyclone, Cramlington, United Kingdom) and 1% penicillin/streptomycin (Thermo Fisher Scientific). HLFs and MEFs were stimulated with following agents: smoothened agonist (SAG) (Calbiochem, Darmstadt, Germany); pirfenidone, SB431542, SANT (all from Sigma-Aldrich, Hamburg, Germany); JQ1 (Cayman Chemical, Ann Arbor, MI, USA); cycloheximide (U.S. Biologic, Salem, MA, USA); GANT61 (Anthem Bioscience, Bengaluru, India); or JIB-04 (Axon Medchem, Groningen, The Netherlands). RNA isolation and quantitative RT-PCR

MATERIALS AND METHODS Study population The investigations were in accordance with the Declaration of Helsinki principles and were approved by the local institutional review board and ethics committee. Informed consent was obtained from either the patients or their next-ofkin. Lung tissue was obtained from 29 patients with IPF who underwent lung transplantation. Six of these patients received pirfenidone until lung transplantation for a mean period of 10.9 6 8.0 mo. IPF diagnosis was based on clinical criteria and proof of a usual interstitial pneumonia pattern in histopathological specimens from the explanted lungs after multidisciplinary discussion according to recent guidelines (8). Unused donor lungs served as a control (n = 17). Inflammatory processes were not observed in the donor lungs

Isolation of RNA was performed using a peqGold Total RNA Kit (Peqlab, Erlangen, Germany) according to the manufacturer’s instructions. The primers used for quantitative RT-PCR (qRT-PCR) were as follows: 59-TCTGGACATACCCCACCTCCCTCTG-39 and 59-ACTGCAGCTCCCCCAATTTTTCTGG-39 for human GLI1, 59-TGGCCGCTTCAGATGACAGATGTTG-39 and 59-CGTTAGCCGAATGTCAGCCGTGAAG-39 for human GLI2, 59-CCGCCTTCGCTCTGGAGCAGATT-39 and 59-TCTGAAACTTCGCTCTCAGCCACAGC-39 for human protein patched homolog (PTCH)1, 59-CGACTACTACGCCAAGGAGGT-39 and 59-CGGAGCTCTGATGTGTTGAA-39 for human TGF-b1, 59-ATTGCCGACAGGATGCAGGAA-39 and 59GCTGATCCACATCTGCTGGAA-39 for human b-actin (b-ACT), 59-CCCACGCGAATCACTCTCAT-39 and 59-TGTCTGGTAACGGCAATGCG-39 for human porphobilinogen deaminase (PBGD), 59-CCCATAGGGTCTCGGGGTCTCAAAC-39 and

TABLE 1. Demographic characteristics and clinical data of the patient cohort Variable

Subjects(n) Age;means 6 SD (yr) Sex, male/female (n) Smoking status, never/former/current (n) FVC predicted; means 6 SD (%) Histologic confirmation of a UIP pattern (%) Daily dose of pirfenidone (g) Duration of pirfenidone treatment (mo)

IPF

IPF + pirfenidone

23 53.8 6 12.9 15/7 6/16/0 50.1 6 15.0 100 —

6 57.2 6 10.4 2/4 2/4/0 46.3 6 12.4 100 2.1 6 0.4 10.9 6 8.0

FVC, forced vital capacity; IPF, idiopathic pulmonary fibrosis; UIP, usual interstitial pneumonia. 2

Vol. 31

May 2017

The FASEB Journal x www.fasebj.org


DIDIASOVA ET AL.

59-GGAGGACCTGCGGCTGACTGTGTAA-39 for mouse Gli1, 59-TGAGGAGAGTGTGGAGGCCAGTAGCA-39 and 59-CCGGGGCTGGACTGACAAAGC-39 for mouse Gli2, 59-AAAGCGGGAAGAGTGCCTCCAGGT-39 and 59-TGGCTGCTGCATGAAGACTGACCAC-39 for mouse Gli3, 59CGCCTTCGCTCTG-GAGCAGATTTC-39 and 59-TGAGGAGACCCACAACCAAAAACTTGC-39 for mouse Ptch1, 59CCCGTGGTAATCCTCGTGGCCTCTAT-39 and 59-TCCATCAGTCACAGGGGCAAAGGTC-39 for mouse Ptch2, and 59TGCACTCTCGCTTTCTGGAGGGTGT-39 and 59-AATGCAGATGGATCAGCCAGGAAGG-39 for mouse 60S ribosomal protein (Rplp0). b-ACT, PBGD, and Rplp0 were used as reference genes. Results are presented either as d Ct (DCt), calculated by subtracting the Ct value of the target gene from the Ct value of the reference gene, or as a relative mRNA expression (2DDCt). Protein isolation and Western blotting Lung tissue specimens or cells were lysed in ice-cold lysis buffer (15 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 1 mM Na3VO4, 1 mM PMSF, protease inhibitor cocktail) (Complete Mini; Roche, Mannheim, Germany), incubated on ice for 40 min, and centrifuged at 8000 g for 10 min at 4°C. The supernatants were collected, and the protein content was determined (Pierce BCA Protein Assay Kit; Thermo Fisher Scientific). Proteins (10–100 mg) were separated on a 10% SDS-PAGE followed by electrotransfer to a PVDF membrane (GE Healthcare, Munich, Germany). After blocking the membrane with 5% nonfat milk (Sigma-Aldrich) in TBS-T (5 mM Tris, pH 7.5, 150 mM NaCl, 0.1% Tween 20), the membrane was probed with one of the following antibodies: mouse anti–a-smooth muscle actin (a-SMA) (Sigma-Aldrich), ¨ mouse antifibronectin (Enzo Life Sciences GmbH, Lorrach, Germany), mouse anti-GLI1, rabbit anti–a-tubulin (both from Cell Signaling Technology, Frankfurt am Main, Germany), mouse anti–Hedgehog-interacting protein, goat anti-GLI3 (both from R&D Systems, Wiesbaden, Germany), mouse anti-GLI2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The membrane was incubated with a peroxidase-labeled secondary antibody (all from Dako, Gostrup, Denmark). To determine the amounts of protein loaded on the gel, the membrane was stripped and reprobed using a mouse anti–b-actin antibody (Sigma-Aldrich).

stimulated for 8 h with 100 nM SAG alone or in the combination with pirfenidone or 10 mM SB431542. Subsequently, cells were pulsed with 0.2 mCi/ml [3H]thymidine for 16 h. Cells were solubilized in 0.5 M NaOH, and [3H]thymidine incorporation was determined by liquid scintillation spectrometry.

Luciferase reporter assays ShhL2 cells were plated in a solid white, 96-well, clear-bottom plate and grown to full confluency. Subsequently, cells were treated in full growth medium with 100 nM SAG in the absence or presence of the indicated compounds for 48 h. Finally, cells were lysed in Passive Lysis Buffer (Promega, Madison, WI, USA), and firefly and Renilla luciferase activity were measured by an Orion L microplate luminometer (Berthold Detection Systems, Pforzheim, Germany) using Beetle- and Renilla-Juice reagents (both from PJK, Kleinblittersdorf, Germany). TGF-b1 immunoassay HLFs were treated with 100 nM SAG in the absence or presence of pirfenidone for 16 h. At the indicated time points, supernatants were collected and analyzed for active TGF-b1 using the TGF-b1 Emax Immunoassay System (Promega) according to the manufacturer’s instructions. Short interfering RNAs Commercially available short interfering RNA (siRNA) against human GLI2 (Thermo Fisher Scientific) was used. Cells were starved overnight and then treated with 100 nM siRNA for 72 h using the siLentFect Lipid transfection reagent (Bio-Rad Laboratories, Munich, Germany) according to the manufacturer’s instructions. For experiments targeting mouse Sufu, a pool of 4 different siRNA sequences was used (Dharmacon, Lafayette, CO, USA). Control siRNA was obtained from Qiagen (Hilden, Germany). Statistics

Lactate dehydrogenase release Lactate dehydrogenase release was assessed using a Cytotoxicity Detection Kit (Roche Diagnostics, Mannheim, Germany) according to the manufacturer’s instruction. Scratch wound-healing assay HLFs were seeded onto 6-well tissue culture plates into silicone inserts (Ibidi, Planegg-Martinsried, Germany) and serumstarved overnight. Cells were stimulated with either 100 nM SAG alone or with the combination of increasing concentrations of pirfenidone or 10 mM SB431542 (Sigma-Aldrich). Images of the gap area were taken after removing the inserts at time points 0 or 16 h after stimulation. The rate of the wound closure was assessed by counting the cells that had migrated into the same-sized square fields. Cell proliferation assay Proliferation of HLFs was determined by the uptake of [3H]thymidine (PerkinElmer, Waltham, MA, USA). Cells were cultured in 48-well plates, growth-arrested in serum-free medium, and PIRFENIDONE INHIBITS GLI

Statistical analysis was performed in GraphPad Prism 5.02 (GraphPad Software Inc., La Jolla, CA, USA). Data are presented as means 6 SEM. Differences between 2 groups were assessed using Student’s t test. When 3 or more groups were compared, ANOVA followed by the Tukey’s post hoc test was applied. All tests were performed with an undirected hypothesis. The level of statistical significance was set at P $ 0.05.

RESULTS Pirfenidone decreases Hh pathway activity in the lungs of patients with IPF and in patient-derived lung fibroblasts We tested the hypothesis that pirfenidone displays antifibrotic activities via its ability to interfere with the Hh signaling pathway. To this end, we examined the expression of the well-established Hh pathway target genes GLI1, GLI2, and PTCH-1 in the lungs of donors and of patients with IPF having or not having received pirfenidone treatment. Expression of these target genes correlates with active Hh signaling and can thus be used as a readout for


3

pathway activity (16). Compared with nonfibrotic donor lung tissue, lungs from patients with IPF displayed markedly elevated GLI1 mRNA levels (Fig. 1A). The GLI1 mRNA expression was significantly reduced in the lungs of patients with IPF who received pirfenidone when

compared with the lungs of patients with IPF without pirfenidone treatment (Fig. 1A). GLI1 protein levels mirrored the mRNA expression pattern (Fig. 1B, C). Expression of GLI2 and PTCH-1 was also up-regulated in the lungs of untreated patients with IPF as compared with

Figure 1. Expression of Hh signaling components is down-regulated in the lungs of patients with IPF treated with pirfenidone. A) qRT-PCR of GLI1 mRNA expression in the lungs of donors (n = 17), patients with IPF (n = 23), and pirfenidone-treated patients with IPF (n = 6). Data are expressed as DCt using b-actin as a reference gene. **P , 0.01; ***P # 0.001. B) GLI1 protein level in the lung homogenates of donor, patients with IPF, and pirfenidone-treated patients with IPF. Representative donors (5/17), patients with IPF (5/23), and pirfenidone-treated patients with IPF (5/6) are shown. a-Tubulin served as a loading control. C ) Densitometry analysis of B. *P # 0.05. D, E ) qRT-PCR analysis of GLI2 (D) and PTCH-1 (E ) mRNA expression in the lungs of donors (n = 17), patients with IPF (n = 23), and pirfenidone-treated patients with IPF (n = 6). Data are expressed as DCt using b-actin as a reference gene. **P , 0.01; ***P # 0.001. F–H ) qRT-PCR analysis of GLI1 (F ), GLI2 (G), and PTCH-1 (H ) mRNA expression in HLFs from donors (n = 10) and patients with IPF (n = 10) either treated or untreated with 1.2 mg/ml of pirfenidone. Data are expressed as DCt using PBGD as a reference gene. *P , 0.05; ***P # 0.001. I ) a-SMA and FN protein level in donor and IPF HLF either treated or untreated with 1.2 mg/ml of pirfenidone. Representative donor (5/10) and IPF (5/10) lung fibroblasts are shown. b-actin served as a loading control. J, K ) Densitometry analysis of I. *P # 0.05; ***P # 0.001. L) Number of migrating HLFs from donors (n = 10) and patients with IPF (n = 10) either treated or untreated with 1.2 mg/ml of pirfenidone as assessed by the wound healing assay. **P , 0.01; ***P # 0.001. M ) Proliferation of HLFs from donors (n = 10) and patients with IPF (n = 10) either treated or untreated with 1.2 mg/ml of pirfenidone as measured by [3H]thymidine incorporation. Data represent means 6 SEM. 4

Vol. 31

May 2017



DIDIASOVA ET AL.

donor lungs, and pirfenidone lowered the expression of these genes in IPF lungs, although not significantly (Fig. 1D, E). Next, we studied the effect of pirfenidone on HLFs obtained from patients with IPF. Treatment of HLFs from patients with IPF with pirfenidone markedly diminished GLI1, a-SMA, and fibronectin (FN) expression (Fig. 1F, I–K) and tended to decrease the expression of GLI2 and PTCH-1 (Fig. 1G, H) when compared with untreated cells. Moreover, exposure of IPF HLFs to pirfenidone significantly reduced their migration and proliferation (Fig. 1L, M). Collectively, our results demonstrate that Hh signaling is active in IPF lungs and in primary IPF fibroblasts and that pirfenidone is capable of reducing Hh pathway activity. Hh pathway activation induces profibrotic responses in HLFs To investigate the relationship between Hh signaling and profibrotic responses, we incubated donor HLFs with

SAG, a widely used synthetic SMO agonist. As a read-out for profibrotic effects, we determined the a-SMA and FN expression, the proliferation rate, and the migratory potential. Exposure of HLFs to SAG stimulated expression of a-SMA and FN in a time-dependent manner (Fig. 2A, B). Furthermore, SAG induced migration and proliferation of HLFs (Fig. 2C–E) and augmented the expression of GLI1 (Fig. 2F) and PTCH-1 (Fig. 2G). mRNA levels of GLI2 remained unchanged (Fig. 2H). These results suggest that the Hh pathway is operational in HLFs and that Hh pathway activation results in profibrotic changes. Pirfenidone inhibits Hh-triggered profibrotic effects Next, we examined whether pirfenidone is able to block SAG-induced profibrotic activities in donor HLFs. The concentrations of pirfenidone applied were selected on the basis of previous studies (19, 20). No cellular toxicity was

Figure 2. Activation of the Hh pathway induces profibrotic responses in human lung fibroblasts. A) Time course of a-SMA and FN protein expression after treatment of HLFs with 100 nM SAG. b-actin served as loading control. Representative Western blots are shown. B) Densitometry analysis of A (n = 3). C ) Impact of SAG on nondirectional migration of HLFs. Representative pictures from the wound healing assay at time 0 and 8 h are shown. D) The rate of wound closure was assessed by counting the cells that migrated into the same-sized square fields. Data represent means 6 SEM. ***P # 0.001 (n = 4). E ) Proliferation of HLFs stimulated for 8 h with 100 nM SAG as assessed by [3H]thymidine incorporation. Data represent means 6 SEM. *P # 0.05 (n = 3). F–H ) Time course of GLI1 (F ), PTCH-1 (G), and GLI2 (H ) expression after treatment of HLFs with 100 nM SAG. Data are expressed as fold change in the expression of the target gene. PBGD served as a reference gene. *P # 0.05 (n = 3). PIRFENIDONE INHIBITS GLI


5

observed at the concentrations used (Supplemental Fig. 1). Incubation of the cells with pirfenidone did decrease SAGtriggered a-SMA and FN expression (Fig. 3A–C). Moreover, pirfenidone inhibited SAG-induced migration of HLFs (Fig. 3D, E) and reduced, at the highest concentration used, SAG-triggered cell proliferation (Fig. 3F). Treatment of cells with pirfenidone also diminished the expression of GLI1 and PTCH-1 in response to SAG stimulation (Fig. 3G, H). Taken together, our results show that pirfenidone suppresses Hh signaling in HLFs and that it blocks Hhinduced profibrotic changes. To investigate whether pirfenidone inhibits the Hh pathway in mouse cells, we treated mouse NIH3T3 fibroblasts carrying a stably integrated Hh reporter plasmid (ShhL2 cells) with SAG and increasing concentrations of pirfenidone. In line with the observations made in human cells, pirfenidone reduced luciferase signals in a dosedependent manner in mouse fibroblasts, indicative of Hh pathway inhibition (Fig. 3I). This finding was supported by a pirfenidone-induced decrease in expression of Hh target genes on the mRNA (Gli1, Ptch-1, Ptch-2) and protein (GLI1) level (Fig. 3J–M). The small molecule BRD inhibitor JQ1, which antagonizes Hh signaling at the level of GLI1/2 gene transcription, was used as a positive control (Fig. 3J–M). Pirfenidone interferes with the activity of GLI transcription factors We performed an epistatic analysis to determine at which step pirfenidone inhibits the Hh cascade. Pirfenidone significantly inhibited Hh target gene expression (Gli1, Ptch-1, Ptch-2) in cells in which the pathway is initiated at the level of Hh ligands [mouse embryonic fibroblasts stably expressing SHH (MEF[SHH]) cells] (Fig. 4A) (21). This finding was confirmed by detecting the reduced protein levels of the target genes GLI1 and Hedgehoginteracting protein 1 in MEF[SHH] cells upon pirfenidone treatment (Fig. 4B). In addition, the constitutive and ligand-independent Hh signaling in cells that genetically lack the negative receptor Ptch-1 (Ptch-12/2 MEFs) could also be blocked by the administration of pirfenidone (Fig. 4C). These data suggest that pirfenidone inhibits the pathway at the level of SMO or downstream of it. We next tested this hypothesis by exposing Gli32/2 MEFs to pirfenidone. Because GLI3 acts as an important negative regulator of Hh signaling, these cells possess de-repressed Hh pathway activity, which solely relies on GLI1 and GLI2. Pirfenidone was able to reduce Hh signaling activity in these cells, in contrast to the ineffective upstream SMO inhibitor SANT (Fig. 4D). Taken together, these data suggest that pirfenidone blocks endogenous Hh signaling at a step downstream of SMO at the level of GLI1/2. Pirfenidone destabilizes the GLI2 protein We next aimed at elucidating the Hh-related mechanism of action of pirfenidone in more depth. To this end, we tested the effects of pirfenidone in cells having a downstream activation of Hh signaling due to the genetic loss of Suppresor of Fused (Sufu), a major negative regulator of the 6

Vol. 31

May 2017

pathway (22). Pirfenidone did not display noticeable inhibitory activity in Sufu2/2 MEFs (Fig. 5A). Of note, it has recently been shown that Sufu is capable of stabilizing GLI2 and GLI3 proteins and that Sufu-null MEFs therefore only possess low levels of endogenous GLI2 and GLI3 protein but high GLI1 levels (22, 23). Therefore, suggest that pirfenidone preferentially interferes with GLI2 (and not GLI1) function and that SUFU might be involved in this process. To corroborate this hypothesis, we knocked down Sufu in MEF[SHH] cells and investigated the impact of pirfenidone on GLI2 levels. Indeed, pirfenidone significantly reduced GLI2 levels in wild-type cells, an effect that was abrogated under conditions of SUFU depletion (Fig. 5B). No significant changes in SUFU protein expression were detected upon pirfenidone treatment (Supplemental Fig. 2). To investigate the effect of pirfenidone on endogenous GLI2 in another fibroblast cell line, we continued with the analysis of the protein levels of this transcription factor in NIH3T3 cells. Pirfenidone significantly decreased GLI2 protein levels, which was particularly evident under SAG stimulation (Fig. 5C, D). As a control, we also investigated the effects of pirfenidone on the amounts of GLI3 protein. GLI3 is truncated to a repressor form (GLI3R) in the absence of Hh signaling and accumulates as a full-length GLI3 activator (GLI3A) upon pathway induction. Pirfenidone exposure did not show a pronounced impact on GLI3R levels, but it prevented the appearance of GLI3A upon pathway activation (Fig. 5C, E). Because a similar result was obtained with the GLI1/2-targeting inhibitor JQ1, we considered this outcome to be the indirect result of general Hh pathway inhibition. Assuming that pirfenidone might affect the stability of GLI2, we recorded the protein levels of GLI2 by treating NIH3T3 (ShhL2) fibroblasts with cyloheximide in order to block de novo protein synthesis. As predicted, simultaneous pirfenidone exposure shortened the half-life of endogenous full-length GLI2 from approximately 3 h to less than 1 h (Fig. 5F, G). In contrast, the half-lives of GLI3A and GLI3R were only slightly affected by the exposure of cells to pirfenidone (Fig. 5H–K). Supplemental Fig. 3 shows specificity of GLI2 and GLI3 antibody (Supplemental Fig. 3). In summary, our data suggest that GLI2 represents the primary physiologic target of pirfenidone within the Hh pathway. Mechanistically, pirfenidone negatively affects the stability of the GLI2 protein, a central activator of Hh signaling, and an integration point for noncanonical input through TGF-b signaling.

Hh signaling induces a fibrosis-amplifying auto/paracrine TGF-b loop Because Hh signaling was reported to regulate TGF-b expression (24) and TGF-b in turn was found to activate Hh/GLI (25), we investigated the inhibitory efficacy of pirfenidone toward the Hh pathway in the presence of the TGF-b–type I receptor kinase blocker SB431542. Longterm exposure (72 h) of HLFs to SAG markedly upregulated a-SMA and FN expression (Fig. 6A). This effect was reduced by SB431542 and strongly suppressed by pirfenidone (Fig. 6A), implying that SAG-triggered a-SMA



DIDIASOVA ET AL.

Figure 3. Pirfenidone inhibits Hh-driven cellular responses. A) a-SMA and FN protein expression in HLFs stimulated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone. b-Actin served as a loading control. Representative Western blots are shown. B, C ) Densitometry analysis of A. *P # 0.05; **P , 0.01 (n = 3). D) HLFs were stimulated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone and subjected to a wound healing assay. Representative pictures from the wound healing assay at time 0 and 8 h are shown. E ) The rate of wound closure was assessed by (continued on next page)

PIRFENIDONE INHIBITS GLI


7

Figure 4. Pirfenidone inhibits GLI transcription factors. A) qRT-PCR analysis of Gli1, Ptch-1, and Ptch-2 mRNA expression in mouse embryonic fibroblasts stably overexpressing SHH (MEF[SHH]) untreated or treated with 200 nM SANT, 1 mM JQ1, or increasing concentrations of pirfenidone. Data are expressed as fold change in the expression of the target gene. Rplp0 served as a reference gene. **P , 0.01; ***P # 0.001 (n = 3). B) GLI1 and Hedgehog-interacting protein (HIP1) expression in MEF[SHH] cells treated with 200 nM SANT, 1 mM JQ1, or increasing concentrations of pirfenidone as assessed by Western blotting. b-actin served as a loading a control. Representative Western blots are shown (n = 3). C, D) Mouse embryonic fibroblasts that lack Ptch-1 (Ptch-12/2 MEFs) (C ) or Gli3 (Gli32/2 MEFs) (D) were treated with 200 nM SANT, 1 mM JQ1, or increasing concentrations of pirfenidone. Gli1, Ptch-1, and Ptch-2 mRNA levels were analyzed by qRT-PCR. Data are expressed as a fold change in the expression of the target gene. Rplp0 served as a reference gene. ***P # 0.001 (n = 3).

and FN expression depends, at least in part, on TGF-b receptor activity. Consequently, we examined the expression of TGF-b1 in HLFs stimulated for 8 h (mRNA) or 16 h (protein) with SAG. SAG induced TGF-b1 expression, and this effect was significantly inhibited by pirfenidone (Fig. 6B, C). Strikingly, whereas short-term incubation (8–12 h) of HLFs with SAG did not increase a-SMA and FN expression (data not shown), it resulted in the up-regulation of GLI1 and PTCH-1 mRNA levels (Fig. 6D, E) as well as in the induction of cell proliferation and migration (Fig. 6F, G). Although the latter events were blocked by pirfenidone, they were unaffected by SB431542, suggesting that some of the Hh-triggered profibrotic effects are independent of TGF-b1 receptor signaling. We next investigated whether the SAG-triggered increase in a-SMA and FN levels depends on the ability of TGF-b1 to elevate GLI2/GLI1 expression and thus amplifies Hh signaling. Exposure of HLFs to TGF-b1 indeed induced GLI2 and GLI1 transcription, and this effect was abolished by pirfenidone (Fig. 7A, B). Similar results were obtained when conditioned media from SAG-treated cells

were used (Fig. 7C, D). Notably, addition of the neutralizing anti2TGF-b1 antibody blocked the induction of GLI1/GLI2 expression, ultimately proving that autocrine TGF-b1 signaling is sufficient to up-regulate the GLI1/ GLI2 levels (Fig. 7C, D). Furthermore, the TGF-b1– stimulated increase in a-SMA and FN expression was blocked by the small-molecule GLI antagonists JIB-04 and GANT61 but not by the SMO inhibitor SANT (Fig. 7E). In line with these findings, RNA interference–mediated knockdown of GLI2 (Supplemental Fig. 4) reduced TGFb–mediated induction of a-SMA by 2.1-fold and of FN by 1.7-fold (Fig. 7F, H). At the same time, pirfenidone reduced expression of a-SMA by 5.1-fold in control siRNA–treated cells, whereas only a 2-fold reduction was observed in cells exposed to siRNA against GLI2. Similarly, pirfenidone impaired expression of FN by 5.8-fold in control siRNA– treated cells and only by 3.4-fold in cells exposed to siRNA against GLI2 (Fig. 7F–H). These results indicate that GLI2 not only regulates TGF-b–triggered expression of profibrotic markers but also that it is important for pirfenidone inhibitory activities. The efficacy of GLI2 knockdown is

counting the cells that migrated into the same-sized square fields. Data represent means 6 SEM. **P , 0.01; ***P # 0.001 (n = 4). F ) Proliferation of HLFs treated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone as assessed by [3H]thymidine incorporation. Data represent means 6 SEM. *P # 0.05 (n = 3). G, H ) GLI1 (G) and PTCH-1 (H ) mRNA level in HLFs treated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone. Data are expressed as fold change in the expression of the target gene. PBGD served as a reference gene. *P # 0.05 (n = 3). I ) Luciferase activity in NIH3T3 cells stably transfected with Hh reporter plasmid (ShhL2) and stimulated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone. Data are expressed as fold change. ***P # 0.001 (n = 3). J–L) qRT-PCR analysis of Gli1 (J ), Ptch-1 (K ), and Ptch-2 (L) mRNA expression in NIH3T3 ShhL2 cells either untreated or treated with 100 nM SAG in the absence or presence of 1 mM JQ1 or increasing concentrations of pirfenidone. Data are expressed as fold change in the expression of the target gene. Rplp0 served as a reference gene. **P , 0.01; ***P # 0.001 (n = 3). M ) GLI1 protein level in NIH3T3 ShhL2 cells untreated or treated with 100 nM SAG in the absence or presence of increasing concentrations of pirfenidone or 1 mM JQ1. b-actin served as a loading control. Representative Western blots are shown (n = 3). 8

Vol. 31

May 2017



DIDIASOVA ET AL.

Figure 5. Pirfenidone affects the protein stability of the GLI2 transcription factor. A) Real-time qPCR analysis of Gli1, Ptch-1, and Ptch-2 mRNA expression in MEFs lacking SUFU (Sufu2/2 MEFs). Cells were left untreated or were treated with 200 nM SANT, 1 mM JQ1, or increasing concentrations of pirfenidone. Data are expressed as fold change in the expression of the target gene. Rplp0 served as a reference gene. *P # 0.05; ***P # 0.001 (n = 3). Ns, not significant. B) GLI2 and SUFU protein expression in MEF[SHH]-depleted of SUFU and stimulated for 24 h with 1.2 mg/ml pirfenidone. b-actin served as a loading control. Representative Western blots are shown (n = 3). siCtrl, control siRNA; siSUFU, siRNA against Sufu. C ) GLI2, GLI3A, and GLI3R protein expression in NIH3T3 (ShhL2) cells treated either with 100 nM SAG alone or with the combination of 1 mM JQ1 or increasing concentrations of pirfenidone as assessed by Western blotting. b-Actin served as loading a control. Representative Western blots are shown. D, E ) Densitometry analysis of C. ***P # 0.001 (n = 3). F ) Time course of GLI2 protein stability in NIH3T3 (ShhL2) cells exposed to 100 mg/ml cycloheximide (CHX) alone or in combination with 1.2 mg/ml of pirfenidone as assessed by Western blotting. b-actin served as a loading control. Representative Western blots are shown. G) Densitometry analysis of F. ***P # 0.001 (n = 3). H, J ) Time course of GLI3A (G) and GLI3R (I ) protein stability in NIH3T3 (ShhL2) cells exposed to 100 mg/ml CHX alone or in combination with 1.2 mg/ml of pirfenidone as assessed by Western blotting. b-actin served as a loading control. Representative Western blots are shown. I, K ) Densitometry analysis of H and J. *P # 0.05 (n = 3).



9

Figure 6. Hh-mediated profibrotic activities in HLFs are in part mediated by TGF-b1. A) a-SMA and FN protein expression in donor HLFs stimulated for 72 h with 100 nM SAG alone or with the combination of 1.2 mg/ml pirfenidone and/or 10 mM SB431542 (SB). b-actin served as a loading control. Representative Western blots are shown (n = 3). B) Real-time qPCR analysis of TGF-b1 mRNA expression in donor HLFs stimulated for 8 h with 100 nM SAG in the absence or presence of 1.2 mg/ml of pirfenidone. Data are expressed as fold change in the expression of the target gene. PBGD served as a reference gene. **P , 0.01 (n = 6). C ) Active TGF-b1 in cell culture medium of HLFs treated for 16 h with 100 nM SAG in the absence or presence of 1.2 mg/ml pirfenidone (n = 4). D, E ) Real-time qPCR analysis of GLI1 (D) and PTCH-1 (E ) mRNA expression in donor HLFs stimulated for 8 h with 100 nM SAG in the absence or presence of 1.2 mg/ml of pirfenidone and/or 10 mM SB431542. Data are expressed as fold change in the expression of the target gene. PBGD served as a reference gene. *P , 0.05; **P , 0.01; ***P # 0.001 (n = 3). F ) Donor HLFs were stimulated for 8–12 h with 100 nM SAG in the absence or presence of 1.2 mg/ml pirfenidone and/or 10 mM SB431542 and subjected to a wound healing assay. The rate of wound closure was assessed by counting the cells that migrated into the same-sized square fields. Data represent means 6 SEM. ***P # 0.001 (n = 3). G) Proliferation of donor HLFs treated for 16 h with 100 nM SAG in the absence or presence of 1.2 mg/ml pirfenidone and/or 10 mM SB431542 as assessed by [3H]thymidine incorporation. Data represent means 6 SEM. *P # 0.05; **P , 0.01; ***P # 0.001 (n = 3).

shown in Supplemental Fig. 4. Together, our results show that pirfenidone, through selective targeting of GLI2 in a SUFU-dependent mechanism, is able to inhibit canonical/ noncanonical Hh signaling pathway and TGF-b signaling.

DISCUSSION Pirfenidone is an orally applied drug that has been approved for the treatment of patients suffering from IPF. In 4 double-blind, placebo-controlled, randomized phase III clinical trials, it was shown that pirfenidone slows the decline of forced vital capacity, and combined data suggest that pirfenidone may reduce IPF-related and all-cause mortality in patients with IPF (4–6). In vitro, it has been demonstrated that pirfenidone blocks many of TGF-b– triggered cellular events, such as fibroblast proliferation and migration, extracellular matrix protein production, and epithelial to mesenchymal transition (12, 26). In addition, it has been reported that pirfenidone downregulates the synthesis of proinflammatory mediators and scavenges reactive oxygen species (10, 11). In vivo pirfenidone was found to attenuate bleomycin-induced pulmonary fibrosis in mice and rats and to reduce 10

Vol. 31 May 2017

bleomycin- and amidarone-triggered pulmonary fibrosis in hamsters (9, 27). Furthermore, some beneficial effects of pirfenidone were observed in the treatment of fibrotic disease of other organs, such as liver and kidney, and in the treatment of multiple sclerosis (28). The above-mentioned pathologic conditions are characterized by abnormal deposition of collagen, which is mainly driven by the high expression and activity of TGF-b (29). Thus, the ability of pirfenidone to inhibit TGF-b–induced events may explain the protective effects of this drug in numerous models of organ fibrosis, but it still leaves open the question of the molecular mechanism of its action. In the present study, we demonstrate that pirfenidone interferes with the Hh pathway. We show that: 1) increased expression of the main components of the Hh pathway is down-regulated in lung tissue of patients with IPF who received pirfenidone, 2) pirfenidone blocks expression of Hh target genes as well as Hh-triggered lung fibroblast migration and proliferation, 3) pirfenidonemediated Hh inhibition occurs on the level of the GLI2 transcription factor in a mechanism involving the GLIstabilizing protein SUFU. The involvement of the Hh pathway in the development/ progression of pulmonary fibrosis recently became



DIDIASOVA ET AL.

Figure 7. a-SMA and FN expression depends on noncanonical GLI2/GLI1 activity attributed to TGF-b1. A, B) Real-time qPCR analysis of GLI1 (A) and GLI2 (B) mRNA expression in donor HLFs stimulated for 8 h with 10 ng/ml of TGF-b1 in the absence or presence of 1.2 mg/ml of pirfenidone. Data are expressed as fold change in the expression of the target gene. PBGD served as a reference gene. *P , 0.05; ***P # 0.001 (n = 3). C, D) mRNA expression of GLI1 (C ) and GLI2 (D) in HLFs exposed for 8 h to conditioned medium from control or SAG-treated cells supplemented with the neutralizing anti–TGF-b1 antibody or IgG control (10 mg/ml each). *P , 0.05 (n = 4). E ) a-SMA and FN protein expression in donor HLFs stimulated for 24 h with 10 ng/ml of TGF-b1 alone or with the combination of 1.2 mg/ml pirfenidone, 200 nM SANT, 10 mM JIB-04, or 20 mM GANT61. b-actin served as a loading control. Representative Western blots are shown (n = 3). F ) a-SMA and FN protein expression in donor HLFdepleted of GLI2 and then stimulated for 24 h with 10 ng/ml of TGF-b1 in the absence or presence of 1.2 mg/ml pirfenidone. b-actin served as a loading control. Representative Western blots are shown (n = 3). siCtrl, control siRNA; siGLI2, siRNA against GLI2. H, G) Densitometry analysis of F. **P , 0.01; ***P # 0.001 (n = 4).

apparent. Lineage tracing studies demonstrated that tissue-resident Gli1 + cells expand and give rise to myofibroblasts in response to bleomycin and genetic ablation of Gli1+ cells or inhibition of the Hh pathway on the level of GLI reduces bleomycin-induced pulmonary fibrosis (30). Blockade of the Hh pathway using the clinically approved SMO antagonist Vismodegib did not influence the development of fibrosis in this model (17). These findings can be explained by noncanonical modes of pathway activation, which PIRFENIDONE INHIBITS GLI

have previously been described in fibroblasts and in pathologic settings such as cancer. Specifically, the TGF-b pathway has been shown to be involved in SMO-independent GLI2 activation (25). GLI proteins may contribute to the development of pulmonary fibrosis in several ways; namely, they may: 1) control the expression of proproliferative and antiapoptotic genes (31), 2) regulate collagen production (32), 3) promote epithelial to mesenchymal transition by inducing the expression of Snail (33), and 4) foster


11

differentiation of fibroblasts into myofibroblasts by modulating the transcriptional activity of a-SMA gene (34). Because GLI proteins may be regulated by growth factors/signaling cascades other than Hh ligands through noncanonical mechanisms (15), they are perfectly suited to fulfill a master regulator function in a vast array of profibrotic processes. In fact, various studies have demonstrated a crosstalk between the TGF-b and Hh pathway (15, 25). For example, in gastric cancer cells, activation of Hh signaling promoted the expression of TGF-b family members, and the latter were required in Shh-induced cell activities (35). Conversely, TGF-b–driven epithelialto-mesenchymal transition in non-small cell lung cancer was shown to be blocked by pharmacological inhibitors of the Hh pathway (36). These observations are in line with our results demonstrating that SAG triggers TGF-b expression, which in turn up-regulates GLI1/GLI2 levels and that TGF-b–induced a-SMA and FN expression is reduced by GLI antagonists as well as by GLI2 depletion. In view of these findings, it is tempting to speculate that pirfenidone, via its ability to target GLI2, disrupts a vicious cycle created by TGF-b and Hh signaling in order to promote and amplify profibrotic processes (Fig. 8). Considering the interconnectivity of Hh/GLI with additional signaling systems such as fibroblast growth factor, epidermal growth factor, TNF-a, or IL-6 (37), we suggest that the GLI transcription factors might represent key hubs governing numerous profibrotic signaling arms. Suppression of GLI activity may therefore represent a valuable therapeutic option for the treatment of pulmonary fibrosis. In light of the fact that GLI protein– regulated signaling pathways such as TGF-b and Hh are pathogenetically involved in numerous conditions, the GLI suppressive function of pirfenidone provides a rationale to apply this drug to other human lung diseases. For example, differential expression of TGF-b and Hh signaling components was also observed in diffuse parenchymal lung diseases, such as fibrotic nonspecific interstitial pneumonia and cryptogenic organizing pneumonia (38). Furthermore, TGF-b and Hh signaling were also found to play a role in several solid cancers, including small cell and nonsmall cell lung cancer (15). Thus, there is a rationale for using pirfenidone in patients with lung cancer or in patients affected by both IPF and lung cancer. Preliminary data suggest that pirfenidone may indeed affect the incidence of lung carcinoma in chronic interstitial pneumonia (39); however, trials or reports on the treatment with pirfenidone of patients with both lung cancer and IPF are lacking and urgently needed. In summary, our study demonstrates that pirfenidone interferes with the activity of the GLI2 transcription factor, impairing not only the Hh pathway but also other signaling systems (e.g., TGF-b) regulated by this protein. Targeting such a central regulatory unit might explain the clinical efficacy of this drug in the treatment of pulmonary fibrosis and might encourage its use in other GLI-driven diseases such as cancer. 12

Vol. 31 May 2017

Figure 8. Schematic model of pirfenidone action. Activation of the Hh pathway by SAG leads to the expression of TGF-b1 in a GLI2-dependent manner. TGF-b1 in an autocrine fashion upregulates GLI2 production thereby amplifying Hh-driven profibrotic responses. In addition, GLI2 regulates target gene expression and exerts profibrotic effects downstream of TGFb1 signaling. Pirfenidone decreases the stability of GLI2 transcription factor and thus inhibits both canonical and noncanonical Hh signaling pathway as well as TGF-b signaling.

ACKNOWLEDGMENTS This study was funded by the Von-Behring-Röntgen Foundation (to M.L. and M.W.), the German Research Foundation (Grant WY119/1-3; to M.W.), the Oskar Helene Heim Foundation (to P.M.), and the German Center for Lung Research (to P.M. and M.W.). The authors thank Y. Horn, M. Schwinn, and H. Thiele (all from the Department of Biochemistry, Universities of Giessen and Marburg Lung Center, Giessen, Germany) for excellent technical assistance. J.W., P.M., W.S., and M.W. are members of the German Center for Lung Research. The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS P. Markart, W. Seeger, M. Lauth, and M. Wygrecka designed the research studies; M. Didiasova, R. Singh, J. Wilhelm, G. Kwapiszewska, L. Wujak, D. Zakrzewicz, and L. Schaefer conducted the experiments; M. Didiasova, M. Lauth, and M. Wygrecka analyzed the data; and M. Didiasova, P. Markart, W. Seeger, M. Lauth, and M. Wygrecka wrote the manuscript. REFERENCES 1. Pardo, A., and Selman, M. (2002) Idiopathic pulmonary fibrosis: new insights in its pathogenesis. Int. J. Biochem. Cell Biol. 34, 1534–1538 2. Selman, M., King, T. E., and Pardo, A.; American Thoracic Society; European Respiratory Society; American College of Chest Physicians. (2001) Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann. Intern. Med. 134, 136–151 3. Okamoto, T., Ichiyasu, H., Ichikado, K., Muranaka, H., Sato, K., Okamoto, S., Iyonaga, K., Suga, M., and Kohrogi, H. (2006) Clinical analysis of the acute exacerbation in patients with idiopathic pulmonary fibrosis [in Japanese]. Nihon Kokyuki Gakkai Zasshi 44, 359–367 4. King, T. E., Jr., Bradford, W. Z., Castro-Bernardini, S., Fagan, E. A., Glaspole, I., Glassberg, M. K., Gorina, E., Hopkins, P. M., Kardatzke, D., Lancaster, L., Lederer, D. J., Nathan, S. D., Pereira, C. A., Sahn, S. A., Sussman, R., Swigris, J. J., Noble, P. W., and Group, A. S.; ASCEND Study Group. (2014) A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N. Engl. J. Med. 370, 2083–2092 5. Noble, P. W., Albera, C., Bradford, W. Z., Costabel, U., Glassberg, M. K., Kardatzke, D., King, T. E., Jr., Lancaster, L., Sahn, S. A.,



DIDIASOVA ET AL.

6.

7.

8.

9.

10. 11. 12.

13. 14.

15. 16. 17.

18.

19.

20.

21. 22.

Szwarcberg, J., Valeyre, D., du Bois, R. M., and Group, C. S.; CAPACITY Study Group. (2011) Pirfenidone in patients with idiopathic pulmonary fibrosis (CAPACITY): two randomised trials. Lancet 377, 1760–1769 Taniguchi, H., Ebina, M., Kondoh, Y., Ogura, T., Azuma, A., Suga, M., Taguchi, Y., Takahashi, H., Nakata, K., Sato, A., Takeuchi, M., Raghu, G., Kudoh, S., and Nukiwa, T.; Pirfenidone Clinical Study Group in Japan. (2010) Pirfenidone in idiopathic pulmonary fibrosis. Eur. Respir. J. 35, 821–829 Azuma, A., Nukiwa, T., Tsuboi, E., Suga, M., Abe, S., Nakata, K., Taguchi, Y., Nagai, S., Itoh, H., Ohi, M., Sato, A., and Kudoh, S. (2005) Double-blind, placebo-controlled trial of pirfenidone in patients with idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 171, 1040–1047 Raghu, G., Rochwerg, B., Zhang, Y., Garcia, C. A., Azuma, A., Behr, J., Brozek, J. L., Collard, H. R., Cunningham, W., Homma, S., Johkoh, T., Martinez, F. J., Myers, J., Protzko, S. L., Richeldi, L., Rind, D., Selman, M., Theodore, A., Wells, A. U., Hoogsteden, H., and Schu¨ nemann, H. J.; American Thoracic Society; European Respiratory society; Japanese Respiratory Society; Latin American Thoracic Association. (2015) An Official ATS/ERS/JRS/ALAT clinical practice guideline: treatment of idiopathic pulmonary fibrosis. An update of the 2011 Clinical Practice Guideline. Am. J. Respir. Crit. Care Med. 192, e3–e19 Iyer, S. N., Wild, J. S., Schiedt, M. J., Hyde, D. M., Margolin, S. B., and Giri, S. N. (1995) Dietary intake of pirfenidone ameliorates bleomycin-induced lung fibrosis in hamsters. J. Lab. Clin. Med. 125, 779–785 Giri, S. N., Leonard, S., Shi, X., Margolin, S. B., and Vallyathan, V. (1999) Effects of pirfenidone on the generation of reactive oxygen species in vitro. J Environ Pathol Toxicol Oncol. 18, 169–177 Iyer, S. N., Hyde, D. M., and Giri, S. N. (2000) Anti-inflammatory effect of pirfenidone in the bleomycin-hamster model of lung inflammation. Inflammation 24, 477–491 Choi, K., Lee, K., Ryu, S. W., Im, M., Kook, K. H., and Choi, C. (2012) Pirfenidone inhibits transforming growth factor-b1-induced fibrogenesis by blocking nuclear translocation of Smads in human retinal pigment epithelial cell line ARPE-19. Mol. Vis. 18, 1010–1020 Coon, D. R., Roberts, D. J., Loscertales, M., and Kradin, R. (2006) Differential epithelial expression of SHH and FOXF1 in usual and nonspecific interstitial pneumonia. Exp. Mol. Pathol. 80, 119–123 Bolaños, A. L., Milla, C. M., Lira, J. C., Ram´ırez, R., Checa, M., Barrera, L., Garc´ıa-Alvarez, J., Carbajal, V., Becerril, C., Gaxiola, M., Pardo, A., and Selman, M. (2012) Role of Sonic Hedgehog in idiopathic pulmonary fibrosis. Am. J. Physiol. Lung Cell. Mol. Physiol. 303, L978–L990 Javelaud, D., Pierrat, M. J., and Mauviel, A. (2012) Crosstalk between TGF-b and hedgehog signaling in cancer. FEBS Lett. 586, 2016–2025 Wilson, C. W., and Chuang, P. T. (2010) Mechanism and evolution of cytosolic Hedgehog signal transduction. Development 137, 2079–2094 Moshai, E. F., Wémeau-Stervinou, L., Cigna, N., Brayer, S., Sommé, J. M., Crestani, B., and Mailleux, A. A. (2014) Targeting the hedgehogglioma-associated oncogene homolog pathway inhibits bleomycininduced lung fibrosis in mice. Am. J. Respir. Cell Mol. Biol. 51, 11–25 Wygrecka, M., Kwapiszewska, G., Jablonska, E., von Gerlach, S., Henneke, I., Zakrzewicz, D., Guenther, A., Preissner, K. T., and Markart, P. (2011) Role of protease-activated receptor-2 in idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 183, 1703–1714 Shi, Q., Liu, X., Bai, Y., Cui, C., Li, J., Li, Y., Hu, S., and Wei, Y. (2011) In vitro effects of pirfenidone on cardiac fibroblasts: proliferation, myofibroblast differentiation, migration and cytokine secretion. PLoS One 6, e28134 Kozono, S., Ohuchida, K., Eguchi, D., Ikenaga, N., Fujiwara, K., Cui, L., Mizumoto, K., and Tanaka, M. (2013) Pirfenidone inhibits pancreatic cancer desmoplasia by regulating stellate cells. Cancer Res. 73, 2345–2356 Lipinski, R. J., Bijlsma, M. F., Gipp, J. J., Podhaizer, D. J., and Bushman, W. (2008) Establishment and characterization of immortalized Glinull mouse embryonic fibroblast cell lines. BMC Cell Biol. 9, 49 Svärd, J., Heby-Henricson, K., Persson-Lek, M., Rozell, B., Lauth, M., Bergström, A., Ericson, J., Toftg˚ard, R., and Teglund, S. (2006) Genetic elimination of Suppressor of fused reveals an essential repressor function in the mammalian Hedgehog signaling pathway. Dev. Cell 10, 187–197


23. Humke, E. W., Dorn, K. V., Milenkovic, L., Scott, M. P., and Rohatgi, R. (2010) The output of Hedgehog signaling is controlled by the dynamic association between Suppressor of Fused and the Gli proteins. Genes Dev. 24, 670–682 24. Cigna, N., Farrokhi Moshai, E., Brayer, S., Marchal-Somme, J., Wémeau-Stervinou, L., Fabre, A., Mal, H., Lesèche, G., Dehoux, M., Soler, P., Crestani, B., and Mailleux, A. A. (2012) The hedgehog system machinery controls transforming growth factor-b-dependent myofibroblastic differentiation in humans: involvement in idiopathic pulmonary fibrosis. Am. J. Pathol. 181, 2126–2137 25. Dennler, S., André, J., Alexaki, I., Li, A., Magnaldo, T., ten Dijke, P., Wang, X. J., Verrecchia, F., and Mauviel, A. (2007) Induction of sonic hedgehog mediators by transforming growth factor-beta: Smad3dependent activation of Gli2 and Gli1 expression in vitro and in vivo. Cancer Res. 67, 6981–6986 26. Hisatomi, K., Mukae, H., Sakamoto, N., Ishimatsu, Y., Kakugawa, T., Hara, S., Fujita, H., Nakamichi, S., Oku, H., Urata, Y., Kubota, H., Nagata, K., and Kohno, S. (2012) Pirfenidone inhibits TGF-b1induced over-expression of collagen type I and heat shock protein 47 in A549 cells. BMC Pulm. Med. 12, 24 27. Schaefer, C. J., Ruhrmund, D. W., Pan, L., Seiwert, S. D., and Kossen, K. (2011) Antifibrotic activities of pirfenidone in animal models. Eur. Respir. Rev. 20, 85–97. 28. Cho, M. E., and Kopp, J. B. (2010) Pirfenidone: an anti-fibrotic therapy for progressive kidney disease. Expert Opin. Investig. Drugs 19, 275–283 29. Chen, S. J., Yuan, W., Mori, Y., Levenson, A., Trojanowska, M., and Varga, J. (1999) Stimulation of type I collagen transcription in human skin fibroblasts by TGF-beta: involvement of Smad 3. J. Invest. Dermatol. 112, 49–57 30. Kramann, R., Schneider, R. K., DiRocco, D. P., Machado, F., Fleig, S., Bondzie, P. A., Henderson, J. M., Ebert, B. L., and Humphreys, B. D. (2015) Perivascular Gli1+ progenitors are key contributors to injuryinduced organ fibrosis. Cell Stem Cell 16, 51–66 31. Regl, G., Neill, G. W., Eichberger, T., Kasper, M., Ikram, M. S., Koller, J., Hintner, H., Quinn, A. G., Frischauf, A. M., and Aberger, F. (2002) Human GLI2 and GLI1 are part of a positive feedback mechanism in Basal Cell Carcinoma. Oncogene 21, 5529–5539 32. Amano, K., Densmore, M., Nishimura, R., and Lanske, B. (2014) Indian hedgehog signaling regulates transcription and expression of collagen type X via Runx2/Smads interactions. J. Biol. Chem. 289, 24898–24910 33. Louro, I. D., Bailey, E. C., Li, X., South, L. S., McKie-Bell, P. R., Yoder, B. K., Huang, C. C., Johnson, M. R., Hill, A. E., Johnson, R. L., and Ruppert, J. M. (2002) Comparative gene expression profile analysis of GLI and c-MYC in an epithelial model of malignant transformation. Cancer Res. 62, 5867–5873 34. Hu, B., Liu, J., Wu, Z., Liu, T., Ullenbruch, M. R., Ding, L., Henke, C. A., Bitterman, P. B., and Phan, S. H. (2015) Reemergence of hedgehog mediates epithelial-mesenchymal crosstalk in pulmonary fibrosis. Am. J. Respir. Cell Mol. Biol. 52, 418–428 35. Yoo, Y. A., Kang, M. H., Kim, J. S., and Oh, S. C. (2008) Sonic hedgehog signaling promotes motility and invasiveness of gastric cancer cells through TGF-beta-mediated activation of the ALK5-Smad 3 pathway. Carcinogenesis 29, 480–490 36. Maitah, M. Y., Ali, S., Ahmad, A., Gadgeel, S., and Sarkar, F. H. (2011) Up-regulation of sonic hedgehog contributes to TGF-b1-induced epithelial to mesenchymal transition in NSCLC cells. PLoS One 6, e16068 37. Hanna, A., and Shevde, L. A. (2016) Hedgehog signaling: modulation of cancer properies and tumor mircroenvironment. Mol. Cancer 15, 24 [Erratum] 38. Fitch, P. M., Howie, S. E., and Wallace, W. A. (2011) Oxidative damage and TGF-b differentially induce lung epithelial cell sonic hedgehog and tenascin-C expression: implications for the regulation of lung remodelling in idiopathic interstitial lung disease. Int. J. Exp. Pathol. 92, 8–17 39. Miura, Y., Saito, T., Tsunoda, Y., Tanaka, T., Takoi, H., Yatagai, Y., Lin, S.-Y., Sekine, A., Hayashihara, K., Inomata, M., Nei, T., Saito, Y., Gemma, A., and Azuma, A. (2014) Clinical effect of pirfenidone on incidence of lung carcinoma in chronic interstitial pneumonia. Eur. Respir. J. 44, P3497 Received for publication August 8, 2016. Accepted for publication January 9, 2017.


13

Pirfenidone exerts antifibrotic effects through inhibition of GLI transcription factors Miroslava Didiasova, Rajeev Singh, Jochen Wilhelm, et al. FASEB J published online February 1, 2017 Access the most recent version at doi:10.1096/fj.201600892RR

Supplemental Material

http://www.fasebj.org/content/suppl/2017/02/01/fj.201600892RR.DC1

Subscriptions

Information about subscribing to The FASEB Journal is online at http://www.faseb.org/The-FASEB-Journal/Librarian-s-Resources.aspx

Permissions

Submit copyright permission requests at: http://www.fasebj.org/site/misc/copyright.xhtml

Email Alerts

Receive free email alerts when new an article cites this article - sign up at http://www.fasebj.org/cgi/alerts

© FASEB Downloaded from www.fasebj.org to IP 137.248.205.183. The FASEB Journal Vol., No. , pp:, February, 2017