MACITENTAN SLOWS DOWN THE DERMAL FIBROTIC PROCESS IN SYSTEMICSCLEROSIS: IN VITRO FINDINGS

C. CORALLO!, G. PECETTF, M. IGLARZ3, N. VOLPI\ D. FRANCI\A. MONTELLAl,F. D'ONOFRIOl, R. NUTP and N. GIORDANOl

1Department of Internal Medicine, Endocrine-Metabolic Sciences and Biochemistry, University ofSiena, Siena, Italy; 2Medical and Scientific Direction, Actelion Pharmaceuticals, Imola, Italy; 3DrugDiscovery Department, Actelion Pharmaceuticals Ltd, Allschwil, Switzerland; 4Department of

Biomedical Sciences, University of Siena, Siena, Italy

Systemic sclerosis (or scleroderma) is an autoimmune disease characterized by skin and internal organfibrosis, caused by microvascular dysfunction. The microvascular damage seems to be a consequence ofan endothelial autoimmune response, followed by activation of the inflammatory cascade and massivedeposition of collagen. Endothelin-1 (ET-1) contributes to the inflammatory and fibrotic processes byincreasing the concentration of pro-inflammatory and pro-fibrotic cytokines, and it is considered oneof the most relevant mediators of vascular damage in scleroderma. It is indeed found in very highconcentration in serum of sclerodermic patients. Moreover, in these pathological conditions there isan increased expression of ET-1 receptors (ETAand ETB), which mediate the detrimental action of ET-1, and often a change of ET)ETB ratio. The aim of the present study is to evaluate the in vitro effectof macitentan, an orally active tissue-targeting dual endothelin receptor antagonist, and its majormetabolite (ACT-132577) on alpha smooth muscle actin (<<SMA) expression, evaluated on dermalfibroblasts from healthy subjects and on dermal fibroblasts from lesional and non-Iesional skin fromsclerodermic patients. The combination of macitentan and its major metabolite reduced the levels of«SMA after 48 h in sclerodermic fibroblasts from lesional skin. No relevant changes in «SMA levels werefound in fibroblasts from non-Iesional skin, whose behavior is similar to that of dermal fibroblasts fromhealthy patients.

compound (18). Clinical studies showed that at steady-state the concentration of the metabolite is 6 times higherthan that of macitentan (19). Therefore, to perform ourin vitro experiments, we used a 1 to 6 ratio betweenmacitentan (MAC) and its metabolite (ACT) at twodifferent concentrations:

1) MAC 5xlO-8M + ACT 3xlO-7M; 2) MAC lxlO-6M+ ACT 6xlO-6M. The first combination of concentrationsis called MC1, while the second one is called MC2.

MacitentanMacitentan and ACT-132577 were provided by

Actelion Pharmaceuticals Ltd. In vivo, macitentan ismetabolized into a major and pharmacologically activemetabolite, ACT-132577 (Fig. 1), which is approximately5 to I0-fold less potent than macitentan on ET receptorsbut significantly contributes to the activity of the parent

Mailing address: Prof. Nicola Giordano, MD,Associate professor, Department of Internal Medicine,Endocrine and Metabolic Sciences and Biochemistry,University of Siena, Ospedale S. Maria aile Scotte,Viale Bracci, 53100 Siena, ItalyTel.: +39 0577 233383 Fax: +39 0577 233446e-mail: giordanon@unisi.it

0393-974X (2013)Copyright © by BIOLIFE, s.a.s.

This publication and/or article is fOf individual use only and may not be furtherreproduced without written permission from the copyright holder.

Unauthorized reproduction may result in financial and other penaltiesDISCLOSURE: ALL AUTHORS REPORT NO CONFLICTS OF

INTEREST RELEVANT TO THIS ARTICLE.

Patients and cell culturesTwenty women affected by limited SSc (lSSc)

diagnosed in accordance with LeRoy and Colleagues (21)(age in yrs 59±6; disease duration in yrs 8±3.5) underwentunaffected and affected skin biopsy with a 4 mm disposableskin biopsy punch. Two skin specimens for each patientwere collected. The research protocol was approved by ourlocal Ethics Committee and each donor provided informedwritten consent. The fibroblasts were isolated accordingto Moulin et. a!' (22) and were compared to that of 20control subjects (age in yrs 55±2.4). Isolated fibroblasts(lx106 cells) were cultured in DMEM (Sigma-Aldrich. St.Louis, MO, USA) supplemented with penicillin (l00 U/mL) (Sigma-Aldrich. St. Louis, MO, USA), streptomycin(100 flg/mL) (Sigma-Aldrich. St. Louis, MO, USA), 0.25flg/ml amphotericin B (Sigma-Aldrich. St. Louis, MO,USA), 2mM glutamine (Sigma-Aldrich. St. Louis, MO,USA) and 10% FBS (Sigma-Aldrich. St. Louis, MO,USA) and incubated at 37°C in a atmosphere of 5% CO2

95% air, till confluence (l week) onto 75 cm2 flasks (BDCostar Cambridge, MA, USA).Viability was estimated byvital staining using trypan blue (Sigma-Aldrich. St. Louis,MO, USA). One day before the experiments, cells wereharvested in a serum-fi'ee medium and cultured in 75 cm2

flasks (BD Costar Cambridge, MA, USA) for Westernblotting and MTT assay, and on Lab-Tek II chamberslides (Nalge Nunc International, Rochester, NY, USA)for immunocytochemistry analysis. Both healthy (CTR),not-affected sclerodermic fibroblasts (NA) and affectedsclerodermic fibroblasts (SSc) received the followingtreatments for 24 and 48 h: a) S (serum free medium); b)S+ET-l (serum free medium plus 100nM ET-l (Sigma-Aldrich. St. Louis, MO, USA); c) S+ET-1+MCI (serumfree medium plus ET-I 100nM plus macitentan andACT-132577 at lower concentration); d) S+ET-I +MC2(serum free medium plus ET-l 100nM plus macitentanand ACT-132577 at higher concentration); e) S+MCI(serum free medium plus macitentan and ACT-132577 atlower concentration); f) S+MC2 (serum free medium plusmacitentan andACT-132577 at higher concentration). Cellproliferation was assessed by MTT assay (23) (Sigma-Aldrich. 8t. Louis, MO, USA). All the experiments forwestern blot and for immunocytochemistry were repeatedthree times.

ImmunocytochemistryOn days 1 and 2, fibroblasts grown on Lab- Tek II

chamber slides (~5xl 04 cells/chamber) were fixed in a4% paraformaldehyde solution (Sigma-Aldrich. St. Louis,MO, USA) for 10 min at room temperature. Then, toreduce nonspecific background staining due to endogenousperoxidase, cells were incubated in hydrogen peroxidesolution for 10 min at room temperature. EnVision+

System-HRP Labelled Polymer Anti-Rabbit kit (DakoItalia S.p.A. Milano, Italy) was used to perform theimmunocytochemical analysis. Specifically, anti-humanalpha smooth muscle actin (aSMA) (Abcam, Cambridge,UK) diluted 1:200 in PBS (Sigma-Aldrich. St. Louis,MO, USA) containing 2% bovine serum albumin (BSA)(Sigma-Aldrich. St. Louis, MO, USA) was incubated withcells at 4°C overnight. Then, cells were washed twice inPBS and incubated with an anti-rabbit labeled polymersolution as secondary antibody (Dako Italia S.p.A.Milano, Italy) for I h at room temperature. Finally, cellswere washed twice in PBS and incubated with a substrate-chromogen solution (Dako Italia S.pA Milano, Italy)for 10 min at room temperature. Negative control wasperformed using PBS buffer instead of primary antibody.

Western blottingOn days I and 2, fibroblasts grown in 75 cm2 flasks

were detached using trypsin-EDTA 0.25% solution(Sigma-Aldrich. St. Louis, MO, USA) for 3 min at37°C, collected and centrifuged at 1500 rpm for 10 minat room temperature. Pellets were then resuspended inPBS buffer and centrifuged at 3000 rpm for 20 min at4°C. The obtained pellets were then re-suspended in RIPAbuffer containing protease inhibitor cocktail (Sigma-Aldrich. St. Louis, MO, USA), incubated for 40 min inice and then centrifuged at 15000 rpm for 30 min at 4°C.After centrifugation, supernatants were collected andprotein concentration was measured using the Bradfordassay (24). About 20 flg of proteins for each treatmentwere boiled for 5 min and submitted to 12% SDS-PAGEaccording to Laemmli (25). The bands were transferredfrom the gels to PVDF membranes using iBlot™ DryBlotting System (Invitrogen, Carlsbad, CA, USA) andthen stained using Ponceau-S solution (Sigma-Aldrich.St. Louis, MO, USA). The incubation with anti-humanaSMA (diluted 1:2500) (Abcam, Cambridge, UK) wasmade using iBlot® Western Detection Kit (Invitrogen,Carlsbad, CA, USA). The bound primary antibodies weredetected using anti-rabbit IgG ALP-conjugate (l :5000)(Invitrogen, Carlsbad, CA, USA), visualized using aChemidoc™ XRS 170-870 molecular imager (Bio-Rad,Hercules, CA, USA) and quantified by Quantity Onesoftware (© 2006, Bio-Rad Laboratories, Inc).

Statistical analysisStatistical analysis was performed using GraphPad

Prism® 5 for Windows. To analyze differences in proteinexpression, one-way ANOVA and Tukey's MultipleComparison test were performed. For intra-classstatistics (same cell phenotype but different treatments)data are reported as the percentage increment comparedto untreated S (100%). For inter-class statistics (same

treatment but different cell phenotypes) data are reportedas the percentage increment compared to untreatedcontrol CTR S (100%). Moreover, a linear regression hasbeen found between aSMA expression levels and cellproliferation index.

First, we want to underline that we performed theexperiments considering two variables: the time ofincubation (24 h-48 h) and the drug concentration(MCl and MC2). We had statistically consistentresults with an incubation time of 48 h and with thehigher concentration of the drug (MC2). Therefore,from now on, for Western blotting analysis, wewill report the results obtained at 48 h and usingthe drug as MC2, while for immunocytochemicalanalysis we will report the results obtained at 48 hwith both MCI and MC2. In basal conditions (S), anoverexpression of aSMA was found in sclerodermicfibroblasts compared to unaffected fibroblastsfrom ISSc patients and healthy controls (Fig. 2a)(p<O.05). Moreover, the morphology of sclerodermicfibroblasts showed a myofibroblast phenotype with asmooth profile compared to unaffected and controlfibroblasts, as shown in Fig. 3. After the additionof ET-l (Fig. 2a), we noted a significant increase inET-l-induced aSMA expression, paIticularly fromsclerodermic fibroblasts (p<O.OI). Simultaneously,there was a relevant increment (p<O.Ol) in SScfibroblast proliferation, as shown in Fig. 2c. It isvery important to underline that there is a directcorrelation between aSMA concentration and thefibroblast proliferation index. Therefore, it could behypothesized that ET-l is one of the key mediatorsof fibroblast activation and proliferation. Whenmacitentan and its metabolite (MC2) were addedtogether with ET-l to the cultures, a decrease ofaSMA expression and proliferation were found inSSc fibroblasts (Fig. 2a, p<O.OI; Fig. 2c, p<O.OI):in particular, aSMA levels from SSc fibroblastswere somewhat lower than the levels recorded inbasal conditions (S) (Fig. 2b, p<O.05). This mightexplain the strength of macitentan and its metabolitein reducing the predominant role of ET-l. In fact,when sclerodermic fibroblasts were incubated withMC2 alone (without ET-I), a statistically relevantdecrease in aSMA expression (Fig. 2b) (p<O.Ol)and proliferation (Fig. 2d) (p<O.05) were found in

N0yBr

o A.J~o N

Fig. 1. Chemical structures of macitentan and its activemetabolite ACT-132577.

In this study, we have used primary humandermal fibroblasts from sclerodermic patients andcontrol subjects as an in vitro model to examinethe role of ET-l and macitentan (dual ET receptorsantagonist) in ECM remodeling and fibrosis. Manyprevious in vivo and in vitro studies reported thatsclerodermic fibroblasts directly contribute to theexcessive scarring observed in fibrosis, as a result ofthe enhanced production of ECM components (26).In fibrotic tissue the presence of high levels of aSMAwas identified. aSMA is a protein able to enhancecontraction of the fibroblasts around the lesion(27). Previous studies demonstrated that levels ofcirculating ET-l are elevated in patients with skin andlung fibrosis, and that these levels are correlated withthe severity of the fibrotic development (28). ET-lcould have a central role in fibrotic evolution and, asa consequence, the blockade ofET-1 receptors couldalleviate the persistent fibroblast activation as wellas the profibrotic responses to TGF-~ (28). Previous

c: 150o'0;1/1

e 100a.><w~ 50 if

§] CTR00 NAg SSC

c: 200.2~ 150~e10011.

§] CTR00 NAg SSC

~I'cfl. 50 ...l.? : -

"/ 0'>x«,. x~C:) ,,';\

x«,.C:)

Treatments

§]S

00 S+ET-1g S+ET1+MC2[]IIJ S+MC2

§]S

00 S+ET-1g S+ET-1+MC2[]IIJ S+MC2

Fig. 2. Western blot analysis of cell lysates. A: aSMA levels reported as the percentage increment respect to untreatedcontrol CTR S (100%). The inter-class statistics (same treatment different cell phenotypes) is represented by: *p<0.05,**p<0.01. B: aSMA levels reported as the percentage increment respect to untreated S in the three phenotypes (100%).The intra-class statistics (same cell phenotype different treatments) is represented by: *p<0.05, **p<0.01. C: MTTproliferation results reported as the percentage increment respect to untreated control CTR S (100%). The inter-classstatistics is represented by: *p<0.05, ** p<0.01. D: MTT proliferation results reported as the percentage incrementrespect to untreated S in the three phenotypes (100%). The inter-class statistics is represented by: *p<0.05, **p<O.Ol.

studies focused their attention on ET-l signaling andshowed that in scleroderma patients with interstitialpneumonia and severe pulmonary hypertension,blocking both ET receptors with bosentan improvedpulmonary arterial hypertension (PHA) and led toa significant decrease in skin thickness and digitalulcers (29). Findings on skin fibrosis, combined withthe previous evidence in PAH, support the idea thatbosentan inhibits the sclerotic process in differentorgan systems.

In this study we have shown that the interference

c: 150o'in1/1

e 100a. I

tij 1·~ 50 ~~;.

i. ~:

0"~ ~~Cell Groups

g 150:;:;••..: 100e11.

'iF- 50

~~Cell Groups

with ET-l signaling could represent a potentiallynew therapeutic strategy aiming at slowing downfibrosis development in SSc patients. Moreover,for the first time, we studied the effect of a newdual ET receptor antagonist, macitentan, on dermalfibroblasts from healthy subjects and sclerodermicpatients. In our case, macitentan has been testedin vitro as a "fibrosis-slowing agent" in the skin ofpatients affected by ISSc. In particular, we focusedour attention on aSMA levels before and afterthe treatment with macitentan. We have chosen

Fig. 3. Immunocytochemistry for aSMA at 48 h of control fibroblasts (left column), unaffected sclerodermic fibroblasts(central column) and affected sclera dermic fibroblasts (right column). Each cell phenotype was stained untreated(S), treated with ET-l (S+ET-l), treated with ET-l plus macitentan and ACT in lower concentration (S+ET-l+MC1),treated with ET-l plus macitentan and ACT in higher concentration (S+ET-l +MC2), treated only with macitentan andACT-132577 in lower concentration (S+MC1) and finally with macitentan and ACT in higher concentration (S+MC2).Negative control with a non-specific antibody is also reported (NC).

to monitor aSMA levels, because it is one of themain proteins involved in the activation and trans-differentiation of the fibroblast from the regularphenotype to the myofibroblast phenotype (9, 27).First of all, our results showed that control fibroblasts(healthy subjects) and unaffected fibroblasts fromsclerodermic patients had similar aSMA levels beforethe treatment. This may suggest that, despite the term"systemic", the lesional environment, at least in ISScpatients, can be responsible for a different pattern offibroblast activation and differentiation within thesame patient. The second interesting result is that,in basal conditions, aSMA levels appeared to behigher in affected fibroblasts than in unaffected andcontrol fibroblasts. As a consequence of this bias, theincrease in aSMA levels was substantially higherin affected fibroblasts compared to unaffected andcontrol fibroblasts, after the addition of ET-l: thislast finding suggests that ET-l plays a key role morein the myofibroblast phenotype than in the normalfibroblast phenotype. In fact, Shi-Wen et al. (30),showed that healthy fibroblasts were not activated inpresence of endogenous 10 pg/ml ET-I whereas theybecame activated at concentrations around 40 pg/ml. This activation could be reversed by addition ofbosentan. Therefore, we also expect that macitentan isnot so effective on healthy and unaffected fibroblastsin which the endogenous ET-I production is too lowto activate the cells whereas it will be efficacious inaffected sclerodermic fibroblasts. Indeed, after theaddition of macitentan, aSMA levels decreased inthe affected fibroblasts, much more conspicuouslywhen the cells were not stimulated simultaneouslywith ET-l. This may indicate that macitentan is ableto decrease aSMA levels in affected sclerodermicfibroblasts by antagonizing the ET-l signaling. Whenwe analyzed the statistical relationship betweenaSMA and the fibroblast replication index, we founda direct correlation (r2=0.83). Macitentan seems toeffectively antagonize ET-l in dermal sclerodermicfibroblasts, resulting in a reduced expression ofaSMA. To conclude, although ET-l is one of theseveral factors involved in fibrosis evolution,probably it represents one of the most importantmediators. ET-l seems to be a central peptide in thepathogenesis of SSc, joining together vascular andfibrotic manifestations. Consequently, endothelinreceptor antagonists could be considered as disease

modifiers in SSc. Finally, macitentan could act notonly as a pulmonary arterial hypeliension treatment,but could also reduce or slow down the developmentof dermal fibrosis in SSc patients.

This work was supported by Medical andScientific Direction of Actelion Pharmaceuticals(Italy) and by Drug Discovery Department ofActelion Pharmaceuticals (Switzerland).

1. Kom JR. Pathogenesis of systemic sclerosis.Arthritis and allied conditions, Koopman WJ editor2001; 1643-54.

2. Zamora MR, O'Brien RF, Rutherford RB, Wei1N. Serum endothe1in-1 concentrations and coldprovocation in primary Raynaud's phenomenon.Lancet 1990; 336(8724):1144-7.

3. Soldano S, Montagna P, Brizzolara R. Endothelinreceptor antagonists: effects on extracellular matrixsynthesis in primary cultures of skin fibroblastsfrom systemic sclerosis patients. Reumatismo 2012;64(5):326-34.

4. Kom JR, Mayes M, Matucci-Cerinic M, et al. Digitalulcers in systemic sclerosis: prevention by treatmentwith bosentan, an oral endothelin receptor antagonist.Arthritis Rheumatism 2004; 50:3985-93.

5. Finsnes F, Lyberg T, Christensen G, Skj0nsberg OR.Effect of Endothelin antagonism on the productionof cytokines in eosinophilic airway inflammation.Am J Physiol Lung Cell Mol Physio1200l; 280:659-65.

6. Shi-Wen X, Rodriguez-Pascual F, Lamas S.Constitutive ALK5-independent c-Jun N-terminalkinase activation contributes to endothelin-loverexpression in pulmonary fibrosis: evidence ofan autocrine endothelin loop operating through theendothelin A and B receptors. Mol Cell Bioi 2006;26:5518-27.

7. Iannone F, Bizzocca R, Chiala A, Lanciano E,Lapadula G. Inhibition of endothelin-l (ET-1) by thedual receptor antagonist, Bosentan, increases MMP-9 expression by sclerodermic fibroblasts in vitro.

Clinical and experimental rheumatology I Sl SystemicSclerosis World Congress 20 IO.

8. Dooley A, Shi-Wen X, Aden N, Tranah T, DesaiN, Denton CP, Abraham DJ, Bruckdorfer R.Modulation of collagen type I, fibronectin and dermalfibroblast function and activity, in systemic sclerosisby the antioxidant epigallocatechin-3-gallate.Rheumatology (Oxford) 2010; 49(11):2024-36.

9. Andrawis NS, Wang E, Abernethy DR. Endothelin-Iinduces an increase in total protein synthesis andexpression of the smooth muscle alpha-actin genein vascular smooth muscle cells. Life Sci 1996;59(7):523-8.

10. Hinz B, Phan SH, Thannickal VJ, Galli A, Bochaton-Piallat ML, Gabbiani G. The myofibroblast: onefunction, multiple origins. Am J Pathol 2007;170(6):1807-16Frommer KW, Muller-Ladner U.Expression and function of ETA and ETB receptorsin SSc. Rheumatology (Oxford) 2008; 47(5):27-8.

II. Kakoki M, Hirata Y, Hayakawa H, et al.Effect of hypertension, diabetes mellitus, andhypercholesterolemia on endhotelin type B receptor-mediated nitric oxide release from rat kidney.Circulation 1999; 99:1242-8.

12. Abraham DJ, Vancheeswaran R, Dashwood MR,Rajkumar VS, Pantelides P, Xu SW, du Bois RM,Black CM. Increased levels of endothelin-l anddifferential endothelin type A and type B receptorexpression in scleroderma-associated fibrotic lungdisease. Am J Pathol 1997; 151:831-41.

13. Gregan B, Jurgensen J, Papsdorf G, Furkert J,Schaefer M, Beyermann M, Rosenthal W, Oksche A.Ligand-dependent differences in the internalization ofendothelin A and endothelin B receptor heterodimers.J Bioi Chern 2004; 279:27679-87.

14. Sauvageau S, Thorin E, Caron A, Dupuis J.Endothelin-l-induced pulmonary vasoreactivity isregulated by ET(A) and ET(B) receptor interactions.J Vasc Res 2007; 44:375-81.

15. Shi-Wen X, Denton CP, Dashwood MR, HolmesAM, Bou-Gharios G, Pearson JD, Black CM,Abraham DJ. Fibroblast matrix gene expression andconnective tissue remodeling: role of endothelin-I. JInvest Dermatol2001; 116:417-25.

16. Porter KE, Olojugba DH, Masood I, PembertonM, Bell PR, London NJ. Endothelin-B receptors

mediate intimal hyperplasia in an organ culture ofhuman saphenous vein. J Vasc Surg 1998; 28:695-701.

17. Clozel M, Gray GA. Are there different ETBreceptors mediating constriction and relaxation? JCardiovasc Pharmacol 1995; 26:262-4.

18. Iglarz M, Binkert C, Morrison K,. Pharmacologyof macitentan, an orally active tissue-targeting dualendothelin receptor antagonist. J Pharmacol ExpTher 2008; 327:736-45.

19. Sidhalia PN, Dingemanse J. Multiple-dosetolerability, safety, pharmacokinetics, andphannacodynamics of the dual endothelin receptorantagonist act-064992 in healthy human subjects. JClin Pharmacol 2008; 48: 1114.

20. LeRoy EC, Black C, Fleischmajer R, et al.Scleroderma (systemic sclerosis): classification,subsets and pathogenesis. J Rheumatol1988; 15:202-5.

21. Moulin V, Plamondon M. Differential expressionof collagen integrin receptor on fetal vs. adult skinfibroblasts: implications in wound contraction duringhealing. Br J Dermatol 2002; 5:886-92.

22. Van Meerloo J, Kaspers GJ, Cloos J. Cell sensitivityassay: the MTT assay. Methods Mol Bioi 2011;731:237-45.

23. Bradford MM. A rapid and sensitive method forthe quantitation of microgram quantities of proteinutilizing the principle of protein-dye binding. AnalBiochem 1976; 72:48-254.

24. Laemrnli UK. Cleavage of stmctural proteins duringthe assembly ofthe head of bacteriophage T4. Nature1970; 227 :680-5.

25. Harrison NK, Argent AC, McAnulty RJ, BlackCM, Corrin B, Laurent GJ. Collagen synthesis anddegradation by systemic sclerosis lung fibroblasts.Responses by transfOlming growth factor-beta.Chest 1991; 99:71-2.

26. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C,Brown RA. Myofibroblasts and mechano-regulationof connective tissue remodeling. Nat Rev Mol CellBioi 2002; 3:349-63.

27. Giordano N, Papakostas P, Pecetti G, Nuti R.Cytokine modulation by endothelin-l and possibletherapeutic implications in systemic sclerosis. J BioiRegul HomeostAgents 2011; 25:487-92.

28. Nishibu A, Sakai E, Oyama N, Yamamoto T.Endothelin receptor antagonist bosentan improvesthe dermal sclerosis in a patient with systemicsclerosis. Australas J Dermato12012; 53:32-3.

29. Shi-Wen X, Chen Y, Denton CP, et al. Endothelin-1

promotes myofibroblast induction through the ETAreceptor via a rac/phosphoinositide 3-kinase/ Akt-dependent pathway and is essential for the enhancedcontractile phenotype of fibrotic fibroblasts. MolBioI Cell 2004; 15(6):2707-19.