SPARC Suppresses Apoptosis of Idiopathic PulmonaryFibrosis Fibroblasts through Constitutive Activationof �-Catenin*□S

Received for publication, May 26, 2009, and in revised form, January 8, 2010 Published, JBC Papers in Press, January 8, 2010, DOI 10.1074/jbc.M109.025684

Wenteh Chang, Ke Wei, Susan S. Jacobs, Daya Upadhyay, David Weill, and Glenn D. Rosen1

From Division of Pulmonary and Critical Care Medicine, Department of Medicine, Stanford University Medical Center,Stanford, California 94305-5236

Idiopathic pulmonary fibrosis (IPF) is a poorly understoodprogressive disease characterized by the accumulation of scartissue in the lung interstitium. A hallmark of the disease is areasof injury to type II alveolar epithelial cells with attendant accu-mulation of fibroblasts in areas called fibroblastic foci. In aneffort to better characterize the lung fibroblast phenotype in IPFpatients, we isolated fibroblasts from patients with IPF andlooked for activation of signaling proteins, which could helpexplain the exaggerated fibrogenic response in IPF. We foundthat IPF fibroblasts constitutively expressed increased basal lev-els of SPARC, plasminogen activator inhibitor-1 (PAI-1), andactive �-catenin compared with control cells. Control of basalPAI-1 expression in IPF fibroblasts was regulated by SPARC-mediated activation of Akt, leading to inhibition of glycogensynthase kinase-3� and activation of �-catenin. Additionally,IPF fibroblasts (but not control fibroblasts) were resistant toplasminogen-induced apoptosis andwere sensitized toplasmin-ogen-mediated apoptosis by inhibition of SPARC or �-catenin.These findings uncover a newly discovered regulatory pathwayin IPF fibroblasts that is characterized by elevated SPARC, giv-ing rise to activated �-catenin, which regulates expression ofdownstream genes, such as PAI-1, and confers an apoptosis-resistant phenotype. Disruption of this pathwaymay represent anovel therapeutic target in IPF.

Idiopathic pulmonary fibrosis (IPF)2 is a progressive and fatallung disease of unknown cause. Current estimates of diseaseincidence are 40–50 per 100,000 and �125,000 cases in theUnited States. Most patients are 50–70 years old, but patientswith familial IPF tend to present earlier. Patients are usuallysymptomatic for 6–24months before diagnosis but often pres-

ent with advanced fibrotic disease. Despite therapy, IPF has amedian survival of 4–5 years (1). Most of the current therapytargeted at IPF is anti-inflammatory. These treatments haveyielded few durable responses. Their failure is due in great partto the unique pathogenesis of IPF. Our understanding of thepathogenesis of IPF is evolving, withmore recent evidence sug-gesting a process of alveolar epithelial injury, possibly ongoing,and dysregulated repair, leading to proliferation of myofibro-blasts and fibrotic scarring in the lung.Myofibroblasts in IPF are thought to arise, at least in part,

from differentiation of fibroblasts in response to transforminggrowth factor-� (TGF-�). They are characterized by expressionof �-smooth muscle actin (�-SMA) and synthesis of matrix-modifying factors such as collagen and fibronectin. The prolif-eration of myofibroblasts is a critical step in the generation offibrotic scarring in IPF (2). This process is very similar to whatoccurs during normal wound healing. Myofibroblasts areresponsible for pulling on a wound in order for its edges toapproximate and close. They also lay down the extracellularmatrix (ECM) that is part of the wound scar. Once woundrepair is complete, myofibroblasts normally undergo apoptosis,and other cells in the vicinity break down the ECM. In IPF,however, apoptosis of myofibroblasts is impaired, and theymaintain an environment that is non-degradative, thereby pre-serving and extending the ECM. A critical regulator of fibro-blast differentiation into myofibroblasts and the genesis ofmatrix components is isoform 1 of TGF-�, a powerful mitogenthat is secreted by type II alveolar epithelial cells (AEC), mac-rophages, platelets, and myofibroblasts themselves in responseto injury (3).TGF-� regulates multiple signaling pathways that coordi-

nate cellular responses to injury, tissue repair, and fibrosis.Recent studies have elucidated that TGF-� activates Wnt/�-catenin. The Wnt/�-catenin pathway has been shown to orga-nize diverse regulatory pathways during development, cellgrowth and differentiation, tissue remodeling, and tumorigen-esis (4). The Wnt/�-catenin or canonical Wnt signaling path-way is characterized by the nuclear accumulation of �-catenin,which forms a complex with members of the T cell factor/lymphoid enhancer factor-1 family of transcription factors (4).Many of these genes are involved in matrix remodeling andfibrogenesis, such as SPARC (secreted protein acidic and rich incysteine); matrix metalloproteinase-2, -3, and -9; cyclin D1;matrilysin; and fibronectin. In recent studies, TGF-� has beenshown to induce rapid nuclear translocation of �-catenin in

* This work was supported by SAMFUND, Division of Pulmonary and CriticalCare, Department of Medicine, Stanford University.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. 1S– 4S.

1 To whom correspondence should be addressed: Div. of Pulmonary and Crit-ical Care Medicine, Dept. of Medicine, Stanford University Medical Center,MC 5236, 300 Pasteur Dr., Stanford, CA 94305-5236. Tel.: 650-725-9536;Fax: 650-725-4071; E-mail: grosen@stanford.edu.

2 The abbreviations used are: IPF, idiopathic pulmonary fibrosis; TGF-�, trans-forming growth factor-�; �-SMA, �-smooth muscle actin; ECM, extracellu-lar matrix; AEC, alveolar epithelial cells; GSK-3�, glycogen synthase kinase-3�; PI3K, phosphoinositide 3-kinase; PAI-1, plasminogen activatorinhibitor-1; DMEM, Dulbecco’s modified Eagle’s medium; PBS, phosphate-buffered saline; qPCR, quantitative PCR; shRNA, small hairpin RNA; DEVD-AFC, acetyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethylcoumarin; FACS,fluorescence-activated cell sorter; EMT, epithelial-mesenchymal transition.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 285, NO. 11, pp. 8196 –8206, March 12, 2010© 2010 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

8196 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

mesenchymal stem cells in a Smad-3-dependent manner (5).Also, a recent study showed that TGF-� activates the�-cateninpathway in lung fibroblasts through inhibition of glycogen syn-thase kinase-3� (GSK-3�) (6). In the lung, a Gata6-Wnt/�-catenin pathway was recently shown to be required for epithe-lial stem cell development and airway regeneration (7). Recentstudies also point to a role for the Wnt/�-catenin pathway inpulmonary fibrosis (8). Chilosi et al. (9) reported the accumu-lation of nuclear �-catenin in damaged type II AEC and myofi-broblasts in fibroblastic foci in IPF lung but not in other idio-pathic interstitial pneumonias such as nonspecific interstitialpneumonia. A more recent study revealed activation of theWnt/�-catenin pathway in type II AEC from IPF patients,which appeared to play a role in mediating epithelial injury andhyperplasia (8). Furthermore, it was suggested that activation ofWnt signaling in adjacent lung mesenchyme may prevent theproper differentiation of the alveolar epithelium (10).SPARC, a matricellular protein that regulates tissue repair

and wound healing, is a known target of TGF-� (reviewed inRef. 36). It is known to accumulate in myofibroblasts in fibro-blastic foci in IPF (11). It also plays a role in the assembly offibrillar collagen in the ECM (44). Recently, a study by Nie andSage (12) demonstrated that SPARC induces the accumulationand activation of �-catenin in preadipocytes, leading to anenhanced association of �-catenin with T cell factor/lymphoidenhancer factor and inhibition of adipogenesis. They showedthat integrin-linked kinase (ILK), but not Akt, is required forSPARC activation of �-catenin and that LiCl mimics the effectsof SPARC. They also revealed that SPARC regulates expressionof �5- and �6-integrins through �-catenin. Their laboratoryalso recently showed that SPARC mediates cell survivalthrough its interaction with �1-integrin and activation of ILK(45).We were interested in identifying genes and signaling path-

ways in lung fibroblasts from IPF patients that contribute to thefibrogenic phenotype through promoting ECM depositionand/or inhibition of apoptosis. Recently published supportingstudies reveal that TGF-� induces protection from serum star-vation-mediated apoptosis through the p38 MAPK (mitogen-activated protein kinase) and phosphoinositide 3-kinase(PI3K)/Akt pathways (13, 14). Also, endothelin-1 and TGF-�appear to independently protect lung fibroblasts from apopto-sis via these pathways (13, 14). The same laboratory alsoshowed that TGF-�-mediated induction of plasminogen acti-vator inhibitor-1 (PAI-1) protects lung fibroblasts from plas-minogen-induced apoptosis (15). Several previous studies showthat PAI-1 promotes fibrosis in lung and other tissues and thatplasminogen activation is anti-fibrogenic (16, 17). Additionally,PAI-1 has been shown to impair alveolar epithelial repairthrough binding to vitronectin, and TGF-� requires PAI-1 forits cytostatic effect on epithelial cells (18, 19).To further characterize the interplay between these matrix-

modifying pathways in IPF fibroblasts, we isolated fibroblastsfrom IPF patients and used fibroblasts isolated from patientsundergoing resection for lung cancer as our control. Like otherspreviously, we observed increased expression of total�-SMA inIPF fibroblast samples versus control fibroblasts, which is con-sistent with an increase in the number of myofibroblasts in IPF,

as has been observed in IPF. However, there was significantlymore heterogeneity in the level of total �-SMA expression inIPF fibroblasts versus control fibroblasts.We report, for the firsttime, that IPF fibroblasts constitutively express significantlymore SPARC and nuclear, i.e. active, �-catenin than controlfibroblasts. Because the resistance of IPF fibroblasts to normalapoptotic signals may play a role in disease pathogenesis, welooked for targets of SPARC/�-catenin, whichmaymediate thisprotection. Because PAI-1 has been shown to mediate apopto-tic resistance in lung fibroblasts, we compared the levels ofPAI-1 in control fibroblasts versus IPF fibroblasts and foundsignificantly higher basal PAI-1 expression in IPF fibroblasts.Also, IPF fibroblasts were significantlymore resistant than con-trol fibroblasts to plasminogen-induced apoptosis. Finally, weshow that SPARC mediates activation of �-catenin throughactivation of Akt, leading to inhibition of GSK-3�. Also, thispathway regulates PAI-1 expression in IPF fibroblasts. Thesedata describe, for the first time, a constitutive signaling pathwayregulated by SPARC/�-catenin in IPF fibroblasts that leads toincreased basal expression of PAI-1, which mediates resistanceto plasminogen-induced apoptosis. This may contribute toboth impairment of epithelial repair and fibrosis in IPF.

EXPERIMENTAL PROCEDURES

Cell Culture and Reagents—Lung tissue was obtained frompatients undergoing surgical biopsy for the diagnosis of inter-stitial lung disease or lung transplant for IPF, and non-neoplas-tic tissue was obtained from patients undergoing surgical lungcancer resection. The tissue was minced into small pieces witha scalpel and digested with type I collagenase (1mg/ml; Invitro-gen) and hyaluronidase (125 units/ml; Sigma) at 37 °C with agi-tation for 18 h inDulbecco’smodified Eagle’smedium (DMEM;Invitrogen) supplemented with 10% fetal bovine serum. Thedissociated tissues were incubated without shaking for 5min atroom temperature, followed by the separation of cell-enrichedsupernatant into a new tube. The cell fraction was centrifugedat 250 � g for 5 min, and the pellet was then resuspended inDMEMwith 10% fetal bovine serum. Epithelial cells did not, ingeneral, survive more than one passage and were in large parteliminated through trypsinization. Surviving fibroblasts werecultured in DMEM supplemented with 10% fetal bovine serumat 5% CO2 at 37 °C. Each fibroblast culture was frozen at itsearliest available passage and was used for studies for up to fivepassages. For each experiment, cells were plated in culture ves-sels in DMEM and cultured until 70–80% confluent, unlessindicated otherwise. Cells were subjected to starvation bywashing cells twice with 1� phosphate-buffered saline (PBS),followed by the addition of DMEMand 0.1% serum to eachwelland incubation for an additional 24 h. In the experiments forcell death induced by Glu-plasminogen (American DiagnosticaInc., Stamford, CT), phenol red-free DMEM was used.Western Blot Analysis—Western blot analysis and band

intensity quantitation were performed as described previously(20). Briefly, the protein concentration was measured by Brad-ford assay (Bio-Rad) according to the manufacturer’s instruc-tions. An equal amount of protein was separated by SDS-PAGEand transferred to polyvinylidene difluoridemembrane. Immu-noblotting was performed using primary antibodies for �-SMA

Activation of �-Catenin in IPF Fibroblasts

MARCH 12, 2010 • VOLUME 285 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 8197

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

(American Research Products, Inc., Belmont, MA); SPARC(Biodesign International, Saco, ME); cleaved caspase-3(Asp175), phospho-Akt (Ser473), phospho-GSK-3� (Ser9), andGSK-3� (Cell Signaling Technology, Danvers, MA); active�-catenin (8E7; Millipore, Billerica, MA); total �-catenin (BDBiosciences); PAI-1 (Santa Cruz Biotechnology, Santa Cruz,CA); and�-tubulin (used as a loading control; Sigma) overnightat 4 °C, followed by incubation with the appropriate horserad-ish peroxidase-conjugated secondary antibody (GE Health-care). The blot was visualized by enhanced chemiluminescence(GE Healthcare) and analyzed using a Kodak Image Station4000R system (Carestream Health, Rochester, NY).To detect secreted PAI-1, the culture medium was first

cleared by centrifugation, and proteins were precipitated in thepresence of ammonium sulfate at 50% saturation overnight at4 °C with gentle agitation. The excess salts were removed bydialysis against lysis buffer used for total lysate preparation, andprotein quantitation was performed by the Bradford assay asdescribed above. To suppress endogenous PI3K activity, 10 �M

LY293002 or 1�Mwortmannin (Sigma) was used, and dimethylsulfoxide (Sigma) was used as a control.Immunofluorescent Staining and Nuclear Localization of

�-Catenin—Fibroblasts were grown on 2-well chamber slides(Thermo Fisher Scientific) in complete culture medium until50–60% confluent. Cells were subjected to starvation (0.1%serum) for 48 h. After washing with 1� PBS, cells were treatedwith 100% methanol at �20 °C for 5 min. Cells were thenblocked in 1� PBS and 3% normal goat serum (Sigma) for 30min at room temperature, followed by incubation with anti-active�-catenin at 1:100 dilution in blocking solution for 16 h at4 °C. The target proteins were visualized with fluorescein iso-thiocyanate-conjugated secondary antibody (Calbiochem), and4�,6-diamidino-2-phenylindole (Molecular Probes, Eugene,OR) was used for nuclear counterstaining. Fluorescent imageswere taken and processed using a Labophot-2 microscopeequipped with an episcopic fluorescence attachment (EFD-3;Nikon Instruments Inc., Melville, NY). For the quantitation ofnuclear localized �-catenin, at least total 100 cells were ran-domly selected from five fields of each stained sample to obtainthe percentage of nuclear localization of �-catenin.Nuclear Isolation, Staining, and Fluorescence-activated Cell

Sorting—To isolate stable cell nuclei from fibroblasts for thestaining of intranuclear �-catenin and for analyzing on a flowcytometer, a protocol disrupting cell membrane by detergent(TritonX-100) andmaintaining nuclearmembrane integrity bymagnesium was adopted from the Flow Cytometry Core Labo-ratory at the NCI ETI Branch (home.ncifcrf.gov/ccr/flowcore/nuclei.pdf). Briefly, cells were collected by gentle scraping andwashed twice with cold PBS. Cells were then resuspended incold nuclear isolation buffer (320 mM sucrose, 5 mM MgCl2, 10mMHEPES, and 1% Triton X-100, pH 7.4) and allowed to incu-bate on ice for 10 min. Nuclear yield and integrity were con-firmed by microscopic examination with trypan blue staining.We routinely observed �98% nuclear isolation efficiency (datanot shown). Nuclei were pelleted by centrifugation at 2000 � gand washed twice with nuclear wash buffer (320 mM sucrose,5 mM MgCl2, and 10 mM HEPES, pH 7.4). Isolated nuclei werethen incubated overnight with anti-�-catenin antibody (5 �g/

ml) or normalmouse IgG (Santa Cruz Biotechnology), followedby a 1-h incubation with fluorescein isothiocyanate-conjugatedanti-mouse IgG (2 �g/ml) in nuclear wash buffer plus 1%bovine serum albumin and 0.1% sodium azide. All stepsdescribed above were done at 4 °C. After extensive washing,nuclei were resuspended in 250�l of nuclearwash buffer beforeflow cytometry analysis. Flow cytometry was performed on anAccuri C6 flow cytometer system (Accuri Cytometers, Inc.,Ann Arbor, MI) using 488 nm excitation and standard fluores-cein isothiocyanate emission optics with 10,000 events fromeach sample, and analysis was performed using FlowJo software(Tree Star Inc., Ashland, OR).Analysis of Gene Expression—RNAwas extracted from fibro-

blasts using TRIzol (Invitrogen), and cDNA converted from 5�g of total RNA was obtained using a SuperScript first-strandsynthesis system for reverse transcription-PCR kit (Invitrogen).To control for genomic DNA contamination, additional RNAsamples were processed without reverse transcriptase. Thereverse transcription product equivalent to 25 ng of total RNAwas then added to a real-time quantitative PCR (qPCR) using aDynamo SYBR Green qPCR kit (Finnzymes, Espoo, Finland)according to the manufacturer’s protocol. The following prim-ers were used: human PAI-1, 5�-TGGAACAAGGATGAGAT-CAG-3� (sense) and 5�-CCGTTGAAGTAGAGGGCATT-3�(antisense); human �-SMA, 5�-CTGTTCCAGCCATCCT-TCAT-3� (sense) and 5�-CCGTGATCTCCTTCTGCATT-3�(antisense); and glyceraldehyde-3-phosphate dehydrogenase,5�-GACCCCTTCATTGACCTCAAC-3� (sense) and 5�-CTT-CTCCATGGTGGTGAAGA-3� (antisense). The annealingand amplification temperature was 60 °C. Real-time qPCR wasperformed in strip tubes using a StepOne PCR system (AppliedBiosystems, Foster City, CA) according to the manufacturer’sinstructions. The specificity of amplified products was sug-gested by a melting curve resulting in only one peak. This wasfurther confirmed by agarose gel electrophoresis of the PCRproducts visualized under ethidium bromide/UV illumination.Target amplificationswere comparedwith the reference ampli-fications (glyceraldehyde-3-phosphate dehydrogenase) in thesame experiment for each reverse transcription product tested.All reactions were carried out in duplicate, and the thresholdcycle (Ct) valueswere determined by automated threshold anal-ysis with the StepOne software. The final results are presentedas relative -fold change in target gene expression comparedwith the reference based on the comparative or ��Ct method.The efficiency of each primer pair was determined by the qPCRprocedure from standard dilutions of cDNA (equivalent to 10pg to 10 ng of total RNA in the reverse transcription reaction).To assess the effect of an inhibitor of the TGF-�1 receptor/ALK5 in suppressing PAI-1 expression, 10 �M SB431542 (Toc-ris Bioscience, Ellisville, MO) was added 1 h prior to the addi-tion of human recombinant TGF-�1 at 2 ng/ml overnight.Enzyme-linked Immunosorbent Assay for Secreted PAI-1—

To determine the concentration of PAI-1 in the culturemedium of human lung fibroblasts, cells were plated in a24-well plate at 5� 105 cells/well overnight in DMEM and 10%fetal bovine serum.After serumdeprivation for 24 h, the culturemedium was collected and spun at 10,000 � g for 5 min at 4 °Cto remove any cell debris. The PAI-1 concentration of each

Activation of �-Catenin in IPF Fibroblasts

8198 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

supernatant was determined using a Quantikine human serpinE1/PAI-1 immunoassay kit (R&D Systems, Minneapolis, MN)according to the manufacturer’s instructions.Determination of TGF-�Bioactivity—The activeTGF-� con-

centration in the cultured fibroblasts was assayed using the co-culture method with mink lung epithelial cells transfected witha truncated PAI-1 promoter fused to a firefly luciferase gene(referred to as MLEC-PAI-1-Lux cells; a kind gift from Dr.George Yang, Stanford University) (21). Fibroblasts wereseeded in 96-well plates at 2.5 � 104 cells/well in triplicates,along withMLEC-PAI-1-Lux cells at 1.5� 104 cells/well in lowserum (1%) medium. The use of 1% serum instead of starvationmediumwith 0.1% serumwas needed to support survival of theMLEC-PAI-1-Lux cells. A standard curve of active TGF-�1wasgenerated in MLEC-PAI-1-Lux cells with serial dilutions ofhuman recombinant TGF-�1 (0–10 ng/ml; Sigma). After incu-bation for 24 h, the viability of cells was checked microscopi-cally beforewashing twicewith 1�PBS.Cellswere then lysed in1� passive lysis buffer at 50 �l/well (luciferase assay kit, Pro-mega,Madison,WI) and incubatedwith agitation at room tem-perature for 20 min. 10 �l of cell lysate was analyzed for lucif-erase activity according to the manufacturer’s instructions.Cells from fibroblast-only wells were used for the cell count.Themean values of luciferase activity from triplicateswere thenconverted into concentrations of TGF-� in picograms/numberof cells using a standard curve obtained with human recombi-nant TGF-�1, normalized with cell numbers. The concentra-tion of total TGF-�1 in the culture medium for correspondingfibroblasts was determined using a Quantikine human TGF-�1immunoassay kit (R&D Systems) according to the manufac-turer’s instructions.RNA Interference—We constructed a lentivirus-driven �-

catenin small hairpin RNA (shRNA) expression plasmid fromthe pLKO.1 vector (22), targeting to �-catenin (5�-CGGGAT-GTTCACAACCGAATT-3�; pLKO.1-sh�-Cat), SPARC (5�-AACAAGACCTTCGACTCTTCC-3�; pLKO.1-shSPARC), or ascrambled sequence (5�-GTTCTCCGAACGTGTCAC-GTT-3�; pLKO.1-Scr). pLKO.1-sh�-Cat, pLKO.1-shSPARC, orpLKO.1-Scr was transduced into cells, followed by puromycinselection at 2 �g/ml for at least 48 h. The efficiency of shRNAknockdown of endogenous �-catenin or SPARC was assessedby Western blot analysis.Quantitation of Cell Viability and Caspase-3 Activity Assay—

Cell viability was determined by an alamarBlue assay. Briefly,primary fibroblasts were seeded overnight in 96-well plates intriplicates and then subjected to serum starvation (0.1% serum)for 24 h. Cells were left untreated or were treated with Glu-plasminogen at the indicated concentrations for 48 h. alamar-Blue (resazurin from Sigma) was added to each well at 1.25�g/ml for 2–4 h, and the fluorometric assay was done withexcitation wavelength at 560 nm and emission wavelength at590 nm with a fluorescence plate reader (FLUOstar Omega,BMG Labtech, Durham, NC). For each assay, data were col-lected from triplicates and analyzed and are represented as thepercentage of viable cells relative to the untreated sample.Caspase-3 protease activity inGlu-plasminogen-treated lung

fibroblasts was determined by the fluorometric reaction usingacetyl-Asp-Glu-Val-Asp-7-amino-4-trifluoromethylcoumarin

(DEVD-AFC; R&DSystems) as substrate according to theman-ufacturer’s instructions. Briefly, following the induction of celldeath by Glu-plasminogen for 24 h, cells were collected by cen-trifugation and lysed in lysis buffer on ice for 10 min. 50 �l ofcell lysate (from 2 � 106 cells) was mixed with 50 �l of 2�reaction buffer and 10 mM dithiothreitol and 5 �l of 1 mM

DEVD-AFC in a 96-well plate. After incubation at 37 °C for 90min, the release of free AFC cleaved by active caspase-3 pro-teases was determined using the fluorescence microplatereader with excitation at 400 nm and emission at 505 nm. Thelevel of caspase-3 enzymatic activity in the cell lysate is directlyproportional to the fluorescent signal of cleaved AFC.Statistical Analysis—Data are expressed as the mean � S.D.

One-way analysis of variance and Student’s t test were used forintergroup comparison. A probability level of 0.05 (p � 0.05)was considered significant.

RESULTS

Increased Expression of SPARC in IPF Fibroblasts Leads toActivation of�-Catenin—To further elucidate the phenotype offibroblasts originating from IPF lung, we isolated fibroblastsfrom IPF patients and compared them with control fibroblasts,whichwere isolated from tissue taken at the time of lung cancerresection surgery and remote from the cancer. As others havereported previously, we found an increase in total�-SMA in IPFfibroblasts versus control fibroblasts (Fig. 1). This agrees withthe observation that increased numbers of myofibroblasts arepresent in IPF, but it is noteworthy that they also retain thisphenotypic identity in culture (23). There was also greater het-erogeneity in the expression of �-SMA in IPF fibroblasts com-pared with control fibroblasts, which likely reflects varyingnumbers of myofibroblasts in IPF samples. While examiningthe expression of known matrix regulatory proteins in fibro-blasts isolated fromcontrol or IPF lung, which have been shownto regulate the wound-healing response, we found SPARCexpression increased 8-fold in fibroblasts from IPF lung fibro-blasts versus control fibroblasts (Fig. 1). SPARC was more con-sistently expressed at higher levels in IPF fibroblasts versuscontrol fibroblasts than fibronectin. Type I collagen was con-sistently expressed at higher levels in IPF fibroblasts, but unlikeSPARC, levels variedmorewith serial passage (data not shown).Also, serial passage of cells did not diminish the percentage of

FIGURE 1. �-SMA and SPARC are highly expressed in IPF fibroblasts.A, Western blot analysis of lysate from control (n 4) and IPF (n 4) fibroblastcultures showing expression of �-SMA and SPARC, with �-tubulin as a loadingcontrol. B, quantitation of �-SMA and SPARC in control (n 4) and IPF (n 4)fibroblasts.

Activation of �-Catenin in IPF Fibroblasts

MARCH 12, 2010 • VOLUME 285 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 8199

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

fibroblasts in the culture, but we found that both populations offibroblasts began to senesce after nine passages (data notshown). There was no significant difference, however, in prolif-eration or survival of control fibroblasts versus IPF fibroblasts(supplemental Fig. 1S).

In view of recent studies demon-strating the activation of �-cateninin lung epithelial cells and fibro-blasts in IPF lung and a study show-ing that SPARC activates �-catenin(8, 12), we examined whether�-catenin is activated in IPF fibro-blasts. Because �-catenin regulatestranscriptional responses in thenucleus, nuclear localization of�-catenin is considered synony-mous with activated �-catenin.Using fluorescence-activated cellsorter (FACS) analysis to detectnuclear �-catenin, we discoveredthat 25–55% of IPF fibroblastsexpressed nuclear �-catenin com-pared with �5% of control fibro-blasts (Fig. 2A). The same four IPFand control samples were used atearly passage (lower than six) forthis and all subsequent experimentsshown. There was heterogeneity inthe number of fibroblasts express-ing nuclear �-catenin, like �-SMA,between IPF patients, but controlfibroblasts from different patientsuniformly expressed little nuclear�-catenin (Fig. 2A). In Fig. 2B, a rep-resentative immunostaining fromtwo of these IPF patients and twocontrol patients shows nuclearlocalization of �-catenin in IPFfibroblasts versus cytoplasmic local-ization in control fibroblasts. Anal-ysis of nuclear �-catenin expressionby immunostaining in IPF and con-trol fibroblasts revealed similarresults as seen with FACS analysis(Fig. 2B). Because SPARC can acti-vate �-catenin in adipocytes (12),we investigated whether this wasalso the case in IPF fibroblasts. Fol-lowing successful down-regulationof SPARC by RNA interference(Fig. 2C), we observed that 40% ofcontrol shRNA cells versus 12% ofSPARC shRNA cells expressednuclear �-catenin as detected byimmunohistochemical analysis(Fig. 2D).PAI-1 Expression Is Increased in

IPF Fibroblasts and Is Regulatedby SPARC/�-Catenin—We then sought to identify potentialtargets of SPARC and �-catenin, which would play a role inpromoting fibrosis by inducing matrix production or reducingmatrix turnover and/or suppressing apoptosis of IPF fibro-blasts. We identified PAI-1 as a potential target because the

FIGURE 2. Increased active �-catenin in IPF fibroblasts and its regulation by SPARC. A, quantitation ofnuclear, i.e. active, �-catenin in control and IPF fibroblast cultures by FACS analysis with anti-�-catenin anti-body. Results are shown from two control and two IPF samples. Statistical analysis of IPF (n 4) and control(n 4) fibroblast cultures was performed using Student’s t test and is presented as the mean � S.D. in the rightpanel. FITC, fluorescein isothiocyanate. B, quantitation of nuclear localization of active �-catenin in control andIPF fibroblasts by immunostaining with �-catenin antibody and 4�,6-diamidino-2-phenylindole (DAPI) fornuclear staining. Representative control cells show peripheral cytoplasmic localization of �-catenin, whereas�-catenin is expressed in the nuclei of IPF cells. Results are shown in higher magnification (lower panels). Thepercentages of nuclear localized �-catenin in control (n 4) and IPF (n 4) fibroblast cultures are presented asthe mean � S.D. in the right panel. Statistical analysis was performed using Student’s t test. C and D, SPARCregulates nuclear localization of �-catenin. Following lentiviral transduction of control (Ctrl; scrambled) orSPARC shRNA, IPF fibroblasts were harvested for Western blot analysis with anti-SPARC antibody (C). Thenuclear localization of active �-catenin in control or SPARC shRNA of four IPF samples was assessed asdescribed for B. Representative results from two samples are shown in the left panel of D. In the right panel, thepercentage of nuclear localized �-catenin in control shRNA (n 4) versus SPARC shRNA in IPF fibroblasts ispresented as the mean � S.D. Statistical analysis was performed using Student’s t test.

Activation of �-Catenin in IPF Fibroblasts

8200 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

balance of expression of plasminogen versus PAI-1 regulatesmatrix production and turnover during wound healing, PAI-1expression is increased in IPF lung, deficiency in PAI-1 protectsmice from bleomycin-induced fibrosis, and it has been shownto mediate resistance of lung fibroblasts to plasminogen-in-duced apoptosis in response to TGF-� and endothelin-1 (Ref.15; reviewed in Ref. 17). We found mean basal PAI-1 mRNA tobe increased 6-fold in IPF fibroblasts versus control fibroblasts,but again therewasmore heterogeneity in its level of expressionin IPF fibroblasts versus control fibroblasts (Fig. 3A). Addition-ally, basal levels of extracellular secreted PAI-1 were 4.7 timeshigher in IPF cells versus control cells (Fig. 3B). Interestingly,SPARC, �-SMA, PAI-1, and �-catenin are all targets of TGF-�and are induced or activated during the differentiation of fibro-blasts into myofibroblasts. It was therefore possible that theincreased expression or activity of PAI-1 resulted from an

increase in basal TGF-� levels and/or activity, which led toautocrine induction of PAI-1. However, SB431542, a TGF-�1receptor/ALK5-selective inhibitor, did not suppress basal levelsof secreted PAI-1 in either IPF or control fibroblasts (Fig. 3C). Itdid, however, block TGF-�-mediated induction of PAI-1 incontrol fibroblasts (Fig. 3D), suggesting that basal PAI-1 in IPFand control fibroblasts is not regulated by the TGF-�1 receptorpathway. However, the addition of exogenous TGF-�1 stillrequires its cognate receptor to induce PAI-1. Also, we exam-ined whether total or active secreted TGF-� levels were differ-ent in IPF fibroblasts versus control fibroblasts, but neither totalnor active TGF-� differed significantly in control or IPF fibro-blasts (Table 1).Because basal SPARC was elevated in IPF fibroblasts and

accompanied by activated �-catenin, we hypothesized thatPAI-1 is a target of SPARC/�-catenin. To examine this, we firstdown-regulated SPARC or �-catenin by RNA interference inIPF fibroblasts. This resulted in significantly reduced basalPAI-1mRNAand secreted PAI-1 comparedwith the scrambledcontrol (Fig. 4, A and B). Total �-catenin was reduced by 70%following transduction of the �-catenin shRNA compared withthe scrambled control shRNA, and SPARCwas reduced by 90%following transduction of a SPARC shRNA (Fig. 4C). Also, asimilar result was observed in 293T cells, where we first down-regulated endogenous PAI-1 in 293T cells by transduction of a�-catenin shRNA, followed by overexpression of constitutivelyactive �-catenin (supplemental Fig. 2S). This construct dis-played nuclear localization of the fusion protein, in contrast to

FIGURE 4. SPARC and �-catenin regulate expression of PAI-1. Down-reg-ulation of SPARC or �-catenin suppressed PAI-1 in IPF fibroblasts. A, PAI-1mRNA expression was assessed by real-time qPCR in IPF fibroblasts followingtransduction of control (Ctrl; scrambled), SPARC, or �-catenin (�-cat) shRNA.Data are from duplicate samples (mean � S.D., n 4). B, secreted PAI-1 in theculture medium from IPF fibroblasts was determined by enzyme-linkedimmunosorbent assay (done in triplicate), and data are the mean � S.D. (n 4). Statistical analysis was performed using Student’s t test. C, Western blotanalysis shows down-regulation of endogenous SPARC (upper panel) and�-catenin (lower panel) proteins in IPF cells following transduction of therespective shRNA constructs.

TABLE 1TGF-� activity in control and IPF fibroblasts

Fibroblastsp valueaControl

(n � 4)IPF

(n � 4)

Active TGF-� (pg/250,000 cells) 49.8 � 29.7 63.9 � 41.5 0.12 (NSb)Total TGF-� (pg/100,000 cells) 523.8 � 166.6 570.6 � 177.0 0.35 (NS)a Statistical analysis was performed using Student’s t test.b NS, not significant.

FIGURE 3. Elevated basal PAI-1 expression in IPF fibroblasts does notrequire the TGF-�1 receptor/ALK5. A, PAI-1 mRNA expression was mea-sured by real-time qPCR in control and IPF cells after serum starvation.B, secreted PAI-1 was determined by enzyme-linked immunosorbent assay(average of triplicates) in serum-starved control or IPF cells. Data representthe mean � S.D. (n 4, control and IPF samples). Statistical analysis wasperformed using Student’s t test. C, human lung fibroblasts from control (n 4) and IPF (n 4) cultures were incubated overnight with the TGF-�1 recep-tor/ALK5 inhibitor SB431542 at 10 �M. Secreted PAI-1 was determined byenzyme-linked immunosorbent assay as described for B. Data represent themean � S.D. Statistical analysis was performed using Student’s t test.D, SB431542 suppressed the induction of PAI-1 mRNA expression by exoge-nous TGF-�1 as measured by real-time qPCR. Human control fibroblasts (n 3) were preincubated with SB431542 at 10 �M for 1 h before the addition ofTGF-�1 overnight at 2 ng/ml. The PAI-1 mRNA levels were measured by real-time qPCR. Results are shown as the mean � S.D. with statistical analysis byStudent’s t test.

Activation of �-Catenin in IPF Fibroblasts

MARCH 12, 2010 • VOLUME 285 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 8201

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

the diffused cytoplasmic distribution of green fluorescent pro-tein alone (supplemental Fig. 2S).Elevated PAI-1 in IPF Fibroblasts Confers Resistance to Plas-

minogen-induced Apoptosis—A prevailing hypothesis is thatfibrosis in IPF is due, at least in part, to inadequate apoptosis ofinterstitial fibroblasts. A recent study showed, for example, thatTGF-� induction of PAI-1 in lung fibroblasts inhibits plasmin-ogen-induced apoptosis of normal lung fibroblasts (15). Fol-lowing our observation that IPF fibroblasts express significantly

higher basal PAI-1 than controlfibroblasts, we speculated that IPFfibroblasts would be more resistantthan control fibroblasts to plasmin-ogen-induced apoptosis. To exam-ine this, exogenous plasminogen(10–50�g/ml)was added to controlor IPF fibroblasts, followed by analamarBlue assay to analyze cell via-bility. Plasminogen caused a dose-dependent reduction in cell viabil-ity, with amaximal 42% reduction incell viability at 50 �g/ml in controlfibroblasts; this compared with a10% decrease in cell viability of IPFfibroblasts at the same concentra-tion (Fig. 5A). Cell death followingexposure to plasminogen wascaspase-mediated in control fibro-blasts, as plasminogen induceda dose-dependent activation ofcaspase-3 as evidenced by theappearance of cleaved caspase-3.The activation of caspase-3 was notobserved in plasminogen-treatedIPF fibroblasts (Fig. 5B and data notshown). Using a more sensitive

caspase-3 activity assay, we did detect a 1.5-fold increase incaspase-3 activity in IPF fibroblasts after exposure to the max-imal plasminogen concentration (50 �g/ml), but this was sig-nificantly less than the 3.5-fold increase in control fibroblasts(Fig. 5B, lower panel). Because IPF fibroblasts are relativelyresistant to plasminogen-induced apoptosis, which is likely aresult of elevated PAI-1, and SPARC and �-catenin regulatePAI-1 expression, it is possible that down-regulation of SPARCor �-catenin would sensitize IPF fibroblasts to plasminogen-induced apoptosis. This was, in fact, what we observed, as len-tiviral transduction of either SPARC shRNA or �-cateninshRNA into IPF fibroblasts caused a significant 46% (SPARCshRNA) or 48% (�-catenin shRNA) reduction in cell viability,compared with a 5–7% reduction in viability in the scrambledcontrol, and a 2-fold increase in caspase-3 activity followingexposure to plasminogen (Fig. 5C).To further validate that PAI-1 is regulated by �-catenin, we

examined whether LiCl, a known activator of the �-cateninpathway through phosphorylation and inactivation of GSK-3�(reviewed in Ref. 24), induces PAI-1 in control fibroblasts. Theaddition of LiCl to control fibroblasts caused a dose-dependentincrease in PAI-1 mRNA, with a 4-fold maximal increase at 50mM LiCl and coincident nuclear translocation of �-catenin(supplemental Fig. 3S). Additionally, LiCl (25 or 50 mM) pro-tected control fibroblasts fromplasminogen-induced apoptosis(supplemental Fig. 4S).SPARC Activates �-Catenin in IPF Fibroblasts through Akt

and GSK-3�—To elucidate the mechanism of SPARC-medi-ated activation of �-catenin in IPF fibroblasts, we examined theeffect of attenuating SPARC in IPF fibroblasts onAkt andGSK-3�, known signaling targets of SPARC. Fig. 6 shows that down-

FIGURE 5. IPF fibroblasts are resistant to plasminogen-induced apoptosis. A, cell survival determined bythe alamarBlue assay following overnight incubation with various concentrations of plasminogen is shown.Data are from experiments repeated twice with control (n 4) and IPF (n 4) fibroblast cultures and are shownas the mean � S.D. Statistical analysis was performed using a one-way analysis of variance. B, Western blotanalysis using an antibody specific for cleaved caspase-3 shows that plasminogen caused a concentration-de-pendent increase in cleaved caspase-3 in control cells. Results are representative of three independent exper-iments from control cell lines (n 4). In the lower panel, a significant increase in caspase-3 activity is shown incontrol fibroblasts (n 4) treated with various concentrations of plasminogen for 24 h but not in IPF fibroblasts(n 4). Data are presented as the mean � S.D. from triplicates for each sample. Statistical analysis was per-formed using a one-way analysis of variance. C, SPARC and �-catenin regulate plasminogen-induced apopto-sis. Endogenous SPARC or �-catenin in IPF cells was attenuated by transduction of SPARC or �-catenin (�-Cat)shRNA. Left panel, following the addition of plasminogen, an alamarBlue assay revealed a significant decreasein cell viability in SPARC or �-catenin shRNA cells but not in control (Ctrl) shRNA cells. Right panel, under thesame treatment conditions, a significant increase in caspase-3 activity was observed following attenuation ofSPARC or �-catenin. Data are the mean � S.D. (n 4). Statistical analysis was performed using Student’s t test.

FIGURE 6. SPARC regulates �-catenin activity thorough Akt. Akt andGSK-3� activity was reduced in SPARC shRNA IPF fibroblasts with a coincidentdecrease in expression of PAI-1 and �-SMA. IPF fibroblast cultures (n 4)were transduced with control (Ctrl) or SPARC shRNA, followed by Westernblot analysis for the indicated proteins. Representative results from threeindependent experiments are shown.

Activation of �-Catenin in IPF Fibroblasts

8202 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

regulation of SPARC suppressed both basal phospho-Akt andphospho-GSK-3�. Interestingly, SPARC down-regulation alsoattenuated �-SMA in IPF fibroblasts (Fig. 6), which suggeststhat SPARC may be required to maintain a myofibroblastphenotype.

Our results revealing a mecha-nism of �-catenin activation bySPARC through Akt differ some-what from those of Nie and Sage(12), who found that SPARC activa-tion of �-catenin in adipocytes doesnot involve Akt but instead is regu-lated by integrin-linked kinase. It ispossible that integrin-linked kinasealso regulates �-catenin activity inIPF fibroblasts, and studies toaddress this are planned. To validatethat Akt mediates, at least in part,the activation of �-catenin in IPFfibroblasts, we examined the effectof LY294002 and wortmannin,PI3K/Akt-selective inhibitors, on�-catenin activation in IPF fibro-blasts. Both inhibitors functioned asexpected to suppress basal phos-pho-Akt and phospho-GSK-3� lev-els. Additionally, each reducednuclear�-catenin to a similar extentin IPF fibroblasts (Fig. 7). Both alsoinhibited PAI-1mRNAand reducedbasal �-SMA in IPF fibroblasts,

albeit LY294002 to a greater extent (Fig. 7,A andC). These datasuggest that SPARC causes activation of �-catenin in IPF fibro-blasts by first activating Akt, which in turn phosphorylates, i.e.inactivates, GSK-3� and is then accompanied by the nuclearimport of �-catenin (see schematic in Fig. 8).

DISCUSSION

IPF is a devastating progressive fibrotic disease of the lung.The pathogenesis is unknown, although some recent studieshave identified telomerase mutations in almost 10% of patientswith familial IPF, which suggests that telomere shortening inepithelial and mesenchymal cell compartments may limit lungregenerative capacity (25–28). A recent study by Larsson et al.(23) showed that myofibroblasts are in greater abundance inIPF fibroblast populations versus control fibroblast populationsand exhibit genome-wide abnormalities in translational con-trol, which may confer an anti-apoptotic phenotype. We, too,found an increase in�-SMAexpression in IPF fibroblasts versuscontrol fibroblasts, consistent, at least in part, with the exist-ence of an increased number of myofibroblasts in IPF samplesversus control samples. There was, however, significantly moreheterogeneity in the level of expression of markers of fibroblastactivation, i.e. SPARC, type I collagen, and �-SMA, and activa-tion of signaling pathways, i.e. Akt and �-catenin, in IPF sam-ples versus control samples, which in part likely reflects differ-ences in the numbers of myofibroblasts. This is not surprisingbecause our patient samples were taken from patients at vari-ous stages of disease, and IPF patients exhibit profound varia-bility in how they present and progress. Also, we found no sig-nificant differences in base-line proliferation or apoptosis incontrol fibroblasts versus IPFmyofibroblasts, which agreeswiththe findings of Huang et al. (29). For the first time, we have

FIGURE 7. Reduced nuclear �-catenin and PAI-1 expression in the presence of PI3K/Akt inhibitors. A, Aktand GSK-3� activity was reduced in IPF fibroblasts by PI3K inhibitor LY294002 (10 �M) or wortmannin (1 �M)treatment for 24 h, in concert with a decrease in �-SMA expression. �-Tubulin was used as a loading control.Representative results of Western blot analysis from four IPF samples are shown with specific antibodiesagainst the indicated proteins. DMSO, dimethyl sulfoxide. B, shown is the decrease in nuclear �-catenin local-ization in IPF fibroblasts treated with PI3K inhibitors. A decrease in nuclear �-catenin in inhibitor-treated IPFfibroblasts was assessed by FACS analysis as described in the legend to Fig. 2. Representative results are shownin the upper panel, and the results from statistical analysis of four IPF samples performed using Student’s t testshown are shown in the lower panel. FITC, fluorescein isothiocyanate. C, suppression of basal PAI-1 mRNAexpression in IPF fibroblasts by PI3K inhibitors was demonstrated by real-time qPCR. Data are from duplicatefour IPF samples and are represented as the mean � S.D.

FIGURE 8. Schematic of signaling through SPARC in IPF fibroblasts.Enhanced expression of SPARC in IPF fibroblasts stimulates PI3K activity, fol-lowed by Akt activation. Akt-mediated phosphorylation of GSK-3� releases�-catenin for nuclear translocation and transactivation of downstream geneexpression, e.g. PAI-1. The accumulation of secreted PAI-1 protects IPF fibro-blasts from plasminogen-induced apoptosis.

Activation of �-Catenin in IPF Fibroblasts

MARCH 12, 2010 • VOLUME 285 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 8203

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

shown that IPF fibroblasts express higher levels of SPARC com-pared with control fibroblasts, which mediates activation of�-catenin through activation of Akt and inhibition of GSK-3�.This results in an increase in basal PAI-1 expression, whichrenders IPF myofibroblasts resistant to plasminogen-inducedapoptosis. SPARC has been shown to induce PAI-1 expressionin aortic endothelial cells, but the mechanism was not eluci-dated (30). Our findings of elevated basal PAI-1 in IPF fibro-blasts complement those ofHorowitz et al. (15), who found thatTGF-� induces PAI-1 in lung fibroblasts, which confers resist-ance to anoikis and plasminogen-induced apoptosis. Our find-ings suggest that SPARC, but not TGF-�, regulates basalexpression of PAI-1; we did find, however, that TGF-�-depen-dent induction of PAI-1 in control fibroblasts requires theTGF-�1 receptor/ALK5 (Fig. 3). Also, TGF-� induced PAI-1 inIPF fibroblasts, but down-regulation of SPARC did not abro-gate TGF-�-mediated induction of PAI-1 in control or IPFfibroblasts (data not shown). This suggests that regulation ofbasal PAI-1 in lung fibroblasts is controlled by a pathway dis-tinct from the TGF-�-mediated induction of PAI-1. Our find-ings and published reports suggest that myofibroblasts, whichare increased in number in IPF, acquire an apoptosis-resistantphenotype, which is characterized in part by elevated basalexpression of PAI-1.A recent study revealed that IPF fibroblasts are more resist-

ant to prostaglandin E2-induced apoptosis compared with con-trol fibroblasts, which, as speculated, may be due to reducedlevels of prostaglandin E2 receptors, diminished protein kinaseA levels and activity, and reduced expression of PTEN (phos-phate and tensin homologue) in IPF lung (31). The authorsrevealed that prostaglandin E2 requires PTEN-mediated inhi-bition ofAkt to induce apoptosis in fibroblasts. Also, pathologicsignaling through�1-integrin in IPF fibroblasts has been shownto generate defective PTEN function (32). Moreover, a recentstudy showed that bothTGF-� and endothelin-1 activateAkt inlung fibroblasts and that Akt promotes fibroblast resistance toapoptosis (14); endothelin-1 levels are known to be increased inIPF, and clinical trials in IPF with endothelin receptor antago-nists are ongoing (33). Increased levels of SPARC in IPF lungand lung fibroblasts would activate Akt in IPF lung fibroblasts,whichmay also play a role in the resistance of IPF fibroblasts toprostaglandin E2-induced apoptosis.

We identified a constitutively active SPARC/Akt/�-cate-nin signaling cascade in IPF fibroblasts, which does notappear to be regulated by autocrine activation by TGF-�signaling because neither total nor active TGF-� was signif-icantly increased in IPF myofibroblasts compared with con-trol fibroblasts. In a previous study by Koli et al. (34), cul-tured IPF fibroblasts secreted �3-fold higher levels of activeTGF-�, whereas the total TGF-� levels increased only �1.5-fold. Only two IPF fibroblast samples and one control fibro-blast sample were used in this study, and the activity ofTGF-� was measured in conditioned medium at 24 h with amink lung epithelial cell reporter assay. Because activeTGF-� binds to its receptors only after being liberated fromthe latent complex, TGF-� is probably not formed in solu-tion and may not be stable in its soluble phase (35). There-fore, the measurement of the activity of TGF-� in condi-

tioned medium at 24 h may underestimate its biologicalactivity, whereas our study is based on the co-culture of minklung epithelial cell reporter cells with fibroblasts, whichyields a continuous measurement of the basal activity ofTGF-�. A plausible model in IPF may be that lung epithelialinjury is followed by a TGF-�-dependent phase, whichcauses differentiation of fibroblasts into myofibroblasts andcontributes to persistent epithelial injury and possiblyepithelial-mesenchymal transition (EMT). Pathways thatbecome active in myofibroblasts, such as SPARC/Akt/�-catenin, may be autonomous and no longer require TGF-� toperpetuate the fibrotic response, but fibrosis could be fur-ther exacerbated by exposure of myofibroblasts to TGF-�,e.g. accentuated production of PAI-1 and other fibrogenicfactors. In the absence of augmented TGF-� signaling, it isunclear why SPARC levels are elevated in IPF fibroblasts, andstudies are ongoing to investigate this further. In an intratra-cheal bleomycin model, Strandjord et al. (37) showed thatcollagen accumulation is decreased in SPARC-null mice.Strandjord et al. showed a decrease in collagen accumulationin the lungs of SPARC-null mice following bleomycin injury,a finding supported by recent studies such as those of Zhouet al. (38) showing that SPARC regulates collagen expres-sion. The bleomycin model in mice clearly does not com-pletely mirror IPF, especially in regard to chronicity and thepresence of fibroblastic foci in IPF but not in the bleomycinmodel. Myofibroblasts do accumulate in bleomycin injurybut not in defined fibroblastic foci. Also, SPARC is clearlyexpressed in myofibroblasts in fibroblastic foci in IPF lung(11). Chronic models of lung injury and fibrosis may be moreinformative in discerning a role for SPARC in the regulationof lung fibrosis.It is interesting to speculate that, because type II AEC in

fibroblastic foci constitutively express active �-catenin,SPARC secretion frommyofibroblasts in these foci may con-tribute to activation of �-catenin in AEC. It is also possiblethat trophic factors secreted from IPF myofibroblasts, inaddition to SPARC, are responsible for the constitutive�-catenin activity. For example, Konigshoff et al. (8) demon-strated recently that Wnt1, Wnt7b, Wnt10b, Fzd2, Fzd3,�-catenin, and lymphoid enhancer factor-1 are significantlyincreased in IPF lung tissue with similar elevations in cul-tured type II AEC from IPF patients. Interestingly, in a recentstudy, the same laboratory showed that WISP1 (Wnt1-in-ducible signaling protein-1) is increased in type II AEC inbleomycin lung injury and in patients with IPF (39). Theyalso showed that WISP1 activates Akt, induces PAI-1, andpromotes EMT in cultured type II AEC and enhances myo-fibroblast activation and ECM expression in lung fibroblasts;neutralization of WISP1 inhibited WISP1-induced EMT incultured type II AEC and reduced expression of matrix pro-teins in lung fibroblasts, suggesting that both WISP-1 andSPARC may contribute to increased Akt activity in IPF lungfibroblasts and Akt-mediated resistance of lung fibroblaststo apoptosis by TGF-� and endothelin (13, 14, 23, 32). Also,the inhibition of WISP1 in a murine bleomycin lung injurymodel suppressed lung fibrosis and improved survival (39).We did not observe, however, increased expression of

Activation of �-Catenin in IPF Fibroblasts

8204 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

WISP1 mRNA in IPF fibroblasts versus control fibroblasts(data not shown). A preliminary analysis in our laboratoryalso did not reveal elevation ofWnt1b,Wnt3a, orWnt antag-onist Fzd2 or Fzd3 mRNA in IPF fibroblasts (data notshown). However, Wnt proteins and WISP1 originatingfrom type II AEC could act in a paracrine manner to activate�-catenin in IPF fibroblasts; selective perturbation of Wntproteins or �-catenin in murine models of lung injury willhelp to decipher the contribution of �-catenin to lung injuryand fibrosis.The cellular origin of myofibroblasts in IPF is as yet

unknown. Fibrocytes, circulating progenitors of fibroblasts,have been identified in fibroblastic foci in IPF lung but not innormal lung (40). In a recent study by Larsson et al. (23), fibro-blasts from IPF patients were shown to express keratin 18, anepithelium-specific marker, suggesting that these fibroblastsmay arise through EMT. TheWnt/�-catenin pathway is knownto regulate EMT in other cell types (41, 42); given that both typeII AEC and myofibroblasts constitutively express �-catenin inIPF, it is possible that �-catenin mediates a transition of type IIAEC to myofibroblasts in IPF.Our results in IPF fibroblasts reveal that SPARC activates

Akt, which then phosphorylates GSK-3�, leading to its inac-tivation, which results in nuclear translocation and activa-tion of �-catenin. GSK-3� has recently been shown to regu-late heart development through a �-catenin-dependent butWnt-independent pathway (43). Kerkela et al. (43) showedthat deletion of GSK-3� leads to cardiac hypertrophy. Ourfindings reveal a central role for SPARC/�-catenin in medi-ating the maintenance of a myofibroblast phenotype, whichis supported by our observation that down-regulation ofeither SPARC or �-catenin suppresses �-SMA and confersresistance to plasminogen-induced apoptosis. Coupled withrecently published data (8), our findings suggest that inhibi-tion of SPARC/�-catenin may represent a novel therapeutictarget in IPF.

Acknowledgments—We are very grateful to Dr. Helene Sage (Ben-aroya Research Institute, Seattle, WA) for critical reading of themanuscript. The lentiviral construct expressing constitutively active�-catenin fused with a 3�-green fluorescent protein was a kind giftfrom Dr. Irving Weissman (Stanford University).

REFERENCES1. Zisman, D. A., Keane,M. P., Belperio, J. A., Strieter, R.M., and Lynch, J. P.,

3rd (2005)Methods Mol. Med. 117, 3–442. Phan, S. H. (2002) Chest 122, 286S–289S3. Kapanci, Y., Desmouliere, A., Pache, J. C., Redard, M., and Gabbiani, G.

(1995) Am. J. Respir. Crit. Care Med. 152, 2163–21694. Cadigan, K. M. (2008) Dev. Cell 14, 322–3235. Jian, H., Shen, X., Liu, I., Semenov, M., He, X., and Wang, X. F. (2006)

Genes Dev. 20, 666–6746. Caraci, F., Gili, E., Calafiore, M., Failla, M., La Rosa, C., Crimi, N., Sortino,

M. A., Nicoletti, F., Copani, A., and Vancheri, C. (2008) Pharmacol. Res.57, 274–282

7. Zhang, Y., Goss, A. M., Cohen, E. D., Kadzik, R., Lepore, J. J., Muthuku-maraswamy, K., Yang, J., DeMayo, F. J., Whitsett, J. A., Parmacek, M. S.,and Morrisey, E. E. (2008) Nat. Genet. 40, 862–870

8. Konigshoff, M., Balsara, N., Pfaff, E. M., Kramer, M., Chrobak, I., Seeger,W., and Eickelberg, O. (2008) PLoS ONE 3, e2142

9. Chilosi, M., Poletti, V., Zamo, A., Lestani, M., Montagna, L., Piccoli, P.,Pedron, S., Bertaso, M., Scarpa, A., Murer, B., Cancellieri, A., Maestro, R.,Semenzato, G., and Doglioni, C. (2003) Am. J. Pathol. 162, 1495–1502

10. Ramasamy, S. K., Mailleux, A. A., Gupte, V. V., Mata, F., Sala, F. G., Velt-maat, J. M., DelMoral, P. M., De Langhe, S., Parsa, S., Kelly, L. K., Kelly, R.,Shia,W., Keshet, E., Minoo, P.,Warburton, D., and Bellusci, S. (2007)Dev.Biol. 307, 237–247

11. Kuhn, C., and Mason, R. J. (1995) Am. J. Pathol. 147, 1759–176912. Nie, J., and Sage, E. H. (2009) J. Biol. Chem. 284, 1279–129013. Horowitz, J. C., Lee,D. Y.,Waghray,M., Keshamouni, V.G., Thomas, P. E.,

Zhang, H., Cui, Z., and Thannickal, V. J. (2004) J. Biol. Chem. 279,1359–1367

14. Kulasekaren, P., Scavone, C. A., Rogers, D. S., Arenberg, D. A., Thannickal,V. J., and Horowitz, J. C. (2009) Am. J. Respir. Cell Mol. Biol. 42, 484–493

15. Horowitz, J. C., Rogers, D. S., Simon, R. H., Sisson, T. H., and Thannickal,V. J. (2008) Am. J. Respir. Cell Mol. Biol. 38, 78–87

16. Kortlever, R. M., and Bernards, R. (2006) Cell Cycle 5, 2697–270317. Liu, R. M. (2008) Antioxid. Redox Signal. 10, 303–31918. Lazar, M. H., Christensen, P. J., Du, M., Yu, B., Subbotina, N. M., Hanson,

K. E., Hansen, J. M., White, E. S., Simon, R. H., and Sisson, T. H. (2004)Am. J. Respir. Cell Mol. Biol. 31, 672–678

19. Kortlever, R. M., Nijwening, J. H., and Bernards, R. (2008) J. Biol. Chem.283, 24308–24313

20. Chang, W. T., Kang, J. J., Lee, K. Y., Wei, K., Anderson, E., Gotmare, S.,Ross, J. A., and Rosen, G. D. (2001) J. Biol. Chem. 276, 2221–2227

21. Abe, M., Harpel, J. G., Metz, C. N., Nunes, I., Loskutoff, D. J., and Rifkin,D. B. (1994) Anal. Biochem. 216, 276–284

22. Moffat, J., Grueneberg, D. A., Yang, X., Kim, S. Y., Kloepfer, A.M., Hinkle,G., Piqani, B., Eisenhaure, T. M., Luo, B., Grenier, J. K., Carpenter, A. E.,Foo, S. Y., Stewart, S. A., Stockwell, B. R., Hacohen, N., Hahn, W. C.,Lander, E. S., Sabatini, D. M., and Root, D. E. (2006) Cell 124, 1283–1298

23. Larsson, O., Diebold, D., Fan, D., Peterson,M., Nho, R. S., Bitterman, P. B.,and Henke, C. A. (2008) PLoS ONE 3, e3220

24. Medina, M., and Castro, A. (2008) Curr. Opin. Drug Discov. Devel. 11,533–543

25. Alder, J. K., Chen, J. J., Lancaster, L., Danoff, S., Su, S. C., Cogan, J. D.,Vulto, I., Xie, M., Qi, X., Tuder, R. M., Phillips, J. A., 3rd, Lansdorp, P. M.,Loyd, J. E., and Armanios, M. Y. (2008) Proc. Natl. Acad. Sci. U.S.A. 105,13051–13056

26. Armanios, M. Y., Chen, J. J., Cogan, J. D., Alder, J. K., Ingersoll, R. G.,Markin, C., Lawson, W. E., Xie, M., Vulto, I., Phillips, J. A., 3rd, Lansdorp,P. M., Greider, C. W., and Loyd, J. E. (2007) N. Engl. J. Med. 356,1317–1326

27. Cronkhite, J. T., Xing, C., Raghu, G., Chin, K.M., Torres, F., Rosenblatt,R. L., and Garcia, C. K. (2008) Am. J. Respir. Crit. Care Med. 178,729–737

28. Tsakiri, K. D., Cronkhite, J. T., Kuan, P. J., Xing, C., Raghu, G., Weissler,J. C., Rosenblatt, R. L., Shay, J. W., and Garcia, C. K. (2007) Proc. Natl.Acad. Sci. U.S.A. 104, 7552–7557

29. Huang, S. K., Wettlaufer, S. H., Hogaboam, C. M., Flaherty, K. R., Mar-tinez, F. J., Myers, J. L., Colby, T. V., Travis, W. D., Toews, G. B., andPeters-Golden, M. (2008) Am. J. Respir. Crit. Care Med. 177, 66–74

30. Hasselaar, P., Loskutoff, D. J., Sawdey, M., and Sage, E. H. (1991) J. Biol.Chem. 266, 13178–13184

31. Huang, S. K., White, E. S., Wettlaufer, S. H., Grifka, H., Hogaboam, C. M.,Thannickal, V. J., Horowitz, J. C., and Peters-Golden, M. (2009) FASEB J.23, 4317–4326

32. Xia,H., Diebold,D.,Nho, R., Perlman,D., Kleidon, J., Kahm, J., Avdulov, S.,Peterson, M., Nerva, J., Bitterman, P., and Henke, C. (2008) J. Exp. Med.205, 1659–1672

33. Shi-Wen, X., Rodríguez-Pascual, F., Lamas, S., Holmes, A., Howat, S.,Pearson, J. D., Dashwood,M. R., du Bois, R.M., Denton, C. P., Black, C.M.,Abraham, D. J., and Leask, A. (2006)Mol. Cell. Biol. 26, 5518–5527

34. Koli, K., Myllarniemi, M., Vuorinen, K., Salmenkivi, K., Ryynanen, M. J.,Kinnula, V. L., and Keski-Oja, J. (2006) Am. J. Pathol. 169, 61–71

35. Jurukovski, V., Dabovic, B., Todorovic, V., Chen, Y., and Rifkin, D. B.(2005)Methods Mol. Med. 117, 161–175

36. Clark, C. J., and Sage, E. H. (2008) J. Cell. Biochem. 104, 721–732

Activation of �-Catenin in IPF Fibroblasts

MARCH 12, 2010 • VOLUME 285 • NUMBER 11 JOURNAL OF BIOLOGICAL CHEMISTRY 8205

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

37. Strandjord, T. P., Madtes, D. K., Weiss, D. J., and Sage, E. H. (1999) Am. J.Physiol. Lung Cell. Mol. Physiol. 277, L628–L635

38. Zhou, X., Tan, F. K., Guo, X., Wallis, D., Milewicz, D. M., Xue, S., andArnett, F. C. (2005) Arthritis Rheum. 52, 257–261

39. Konigshoff, M., Kramer, M., Balsara, N., Wilhelm, J., Amarie, O. V., Jahn,A., Rose, F., Fink, L., Seeger, W., Schaefer, L., Gunther, A., and Eickelberg,O. (2009) J. Clin. Invest. 119, 772–787

40. Andersson-Sjoland, A., de Alba, C. G., Nihlberg, K., Becerril, C., Ramírez,R., Pardo, A., Westergren-Thorsson, G., and Selman, M. (2008) Int. J. Bio-chem. Cell Biol. 40, 2129–2140

41. Medici, D., Hay, E. D., and Olsen, B. R. (2008) Mol. Biol. Cell 19,

4875–488742. Dissanayake, S. K., Wade, M., Johnson, C. E., O’Connell, M. P., Leotlela,

P. D., French, A. D., Shah, K. V., Hewitt, K. J., Rosenthal, D. T., Indig, F. E.,Jiang, Y., Nickoloff, B. J., Taub, D. D., Trent, J.M.,Moon, R. T., Bittner,M.,and Weeraratna, A. T. (2007) J. Biol. Chem. 282, 17259–17271

43. Kerkela, R., Kockeritz, L., Macaulay, K., Zhou, J., Doble, B. W., Beahm, C.,Greytak, S., Woulfe, K., Trivedi, C. M., Woodgett, J. R., Epstein, J. A.,Force, T., and Huggins, G. S. (2008) J. Clin. Invest. 118, 3609–3618

44. Bradshaw, A. D. (2009) J. Cell Commun. Signal. 3, 239–24645. Weaver, M. S., Workman, G., and Sage, E. H. (2008) J. Biol. Chem. 283,

22826–22837

Activation of �-Catenin in IPF Fibroblasts

8206 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 11 • MARCH 12, 2010

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from

RosenWenteh Chang, Ke Wei, Susan S. Jacobs, Daya Upadhyay, David Weill and Glenn D.

-Cateninβthrough Constitutive Activation of SPARC Suppresses Apoptosis of Idiopathic Pulmonary Fibrosis Fibroblasts

doi: 10.1074/jbc.M109.025684 originally published online January 8, 20102010, 285:8196-8206.J. Biol. Chem. 

  10.1074/jbc.M109.025684Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2010/01/08/M109.025684.DC1

  http://www.jbc.org/content/285/11/8196.full.html#ref-list-1

This article cites 45 references, 13 of which can be accessed free at

by guest on August 21, 2019

http://ww

w.jbc.org/

Dow

nloaded from