+ MODEL

Journal of the Formosan Medical Association (2014) xx, 1e5

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.jfma-onl ine.com

ORIGINAL ARTICLE

Cyclosporine A induces connective tissuegrowth factor expression in human gingivalfibroblasts: Suppression by epigallocatechin-3-gallate

King-Jean Wu a,c, Guay-Fen Huang b,c, Chun-Hao Chen b,Hao-Hueng Chang b, Yi-Ting Deng a,*

a Department of Dentistry, National Taiwan University Hospital Hsinchu Branch, Hsinchu, Taiwanb Department of Dentistry, National Taiwan University Hospital, Taipei, Taiwan

Received 13 December 2013; received in revised form 9 January 2014; accepted 14 January 2014

KEYWORDSconnective tissuegrowth factor/CCN2;

cyclosporine;epigallocatechin-3-gallate;

fibroblast;gingival overgrowth

Conflicts of interest: The authors h* Corresponding author. Department

Jingguo Road, Hsinchu City, Taiwan.E-mail address: q95422002@ntu.ed

c K.-J.W. and G.-F.H. contributed eq

Please cite this article in press as: Wfibroblasts: Suppression by epigallocj.jfma.2014.01.008

0929-6646/$ - see front matter Copyrhttp://dx.doi.org/10.1016/j.jfma.201

Background/Purpose: Transforming growth factor-b (TGF-b) plays an important role in thepathogenesis of cyclosporine A (CsA)-induced gingival overgrowth (GO). Connective tissuegrowth factor (CTGF/CCN2) acts as a cofactor with TGF-b to induce the maximal profibroticeffects of TGF-b. We investigated the effects of CsA on CCN2 expression in human gingival fi-broblasts (HGFs) and the potential chemopreventive agent for CsA-induced GO.Methods: Western blot analyses were used to examine the signaling pathways of CsA-inducedCCN2 expression in HGFs and whether epigallocatechin-3-gallate (EGCG), curcumin, or lova-statin can inhibit CsA-induced CCN2 expression.Results: CsA significantly stimulated CCN2 synthesis in HGFs. This effect can be inhibited by c-Jun NH2-terminal kinase (JNK) and Smad3 inhibitors but not by TGF-b neutralizing antibody andTGF-b type I receptor inhibitor. Furthermore, EGCG completely blocked CsA-induced CCN2expression.Conclusion: CsA-induced CCN2 protein expression is mediated through JNK and Smad signaling.CsA may contribute to the pathogenesis of GO through upregulation of CCN2 expression inHGFs. EGCG could be an adjuvant for the prevention of CsA-induced GO.Copyright ª 2014, Elsevier Taiwan LLC & Formosan Medical Association. All rights reserved.

ave no conflicts of interest relevant to this article.of Dentistry, National Taiwan University Hospital Hsinchu Branch, Number 25, Lane 442, Section 1,

u.tw (Y.-T. Deng).ually to this work.

u K-J, et al., Cyclosporine A induces connective tissue growth factor expression in human gingivalatechin-3-gallate, Journal of the Formosan Medical Association (2014), http://dx.doi.org/10.1016/

ight ª 2014, Elsevier Taiwan LLC & Formosan Medical Association. All rights reserved.4.01.008

2 K.-J. Wu et al.

+ MODEL

Introduction

Cyclosporine A (CsA), a potent immunosuppressant, is awidely prescribed drug for the prevention of graft rejectionin many transplantation surgery but it has a number of sideeffects.1 One of these side effects is gingival overgrowth(GO) characterized by the accumulation of excessiveextracellular matrix proteins such as collagens, structuralglycoproteins, and proteoglycans in gingival tissues.2 Manystudies have been carried out on the mechanisms of GO,and the results suggest that increased collagen synthesisand reduced collagen degradation are the major causes inthe development of this disease.2e4 However, the exactmechanisms by which CsA induced GO are yet to be fullyidentified. Better understanding of the mechanismsinvolved in the development of CsA-induced GO is neededto develop new therapeutic strategies for the treatment ofthis disease.

Transforming growth factor-b (TGF-b), the most fibro-genic mediator characterized to date, has been suggestedas the main etiologic agent for CsA-induced GO. It is asso-ciated with GO in patients treated with CsA after a hearttransplantation procedure.5 The level of TGF-b in gingivalcrevicular fluid samples from CsA-treated patients with GOwas found to be significantly higher than that of the controlgroup.6 Cotrim et al7 demonstrated that concentrations ofCsA that are similar to those found in the serum of patientsundergoing CsA treatment significantly stimulate TGF-b1expression by normal human gingival fibroblasts (HGFs).TGF-b has been shown to increase collagen synthesis andreduce collagen degradation in various tissues.8 Inhibitionof TGF-b has been shown to be effective in preventing andreversing pre-existing clinically comparable hepatic fibrosisin rodent models.9,10 Therefore, neutralization of TGF-b1and its associated signaling molecules have been proposedas targets for the prevention or treatment of drug-inducedGO.11 However, TGF-b is essential for many physiologicalprocesses, including immune suppression, tumor suppres-sion, apoptosis, and wound healing.8 A recent clinical trialusing a neutralizing anti-TGF-b antibody (CAT-192) to treatsystemic sclerosis patients showed no evidence of efficacy,but had increased mortality and serious adverse effects.12

Connective tissue growth factor (CTGF/CCN2) is a 38-kDa secreted protein, which was originally identified as agrowth factor secreted by human vascular endothelialcells.13 It is now recognized as a growth factor that createsan environment favorable for fibrosis and is responsible forthe maximal profibrotic effects of TGF-b.14,15 Over-expression of CCN2 is found in many fibrotic disordersincluding GO14,15 and is positively related to the degree offibrosis in GO.16 A neutralizing monoclonal antibody tohuman CCN2 has been shown to ameliorate TGF-b-inducedmouse fibrosis17 and bleomycin-induced lung fibrosis.18 Inadults, CCN2 is only produced by kidney mesangial andhepatic stellate cells.14 Therefore, CCN2 may be a potentialand specific therapeutic target for the treatment offibrosis.

Recently, we found that epigallocatechin-3-gallate(EGCG) and curcumin completely inhibited lysophosphati-dic acid (LPA)-induced CCN2 expression in HGFs.19 Black andTrackman20 showed that lovastatin reduces approximately

Please cite this article in press as: Wu K-J, et al., Cyclosporine A indufibroblasts: Suppression by epigallocatechin-3-gallate, Journal of thej.jfma.2014.01.008

75% of TGF-b1-induced CCN2 protein expression. The aim ofthis study is to investigate the effect of CsA on CCN2expression in HGFs and the possible effects that EGCG, cur-cumin, or lovastatin can have on CsA-induced CCN2 expres-sion. Our findings may provide more useful information forthe treatment of GO.

Methods

Reagents

CsA, curcumin, EGCG, N-acetylcysteine (NAC), a selectiveSrc kinase inhibitor (Src inhibitor-1), and TGF-b type I re-ceptor/activin receptor-like kinase 5 (ALK5) inhibitorSB431542 were purchased from Sigma-Aldrich (St. Louis,MO, USA). Lovastatin was purchased from Standard Chem.& Pharm. Co. Ltd. (Tainan, Taiwan). Specific inhibitor ofthe Src-family kinases inhibitor PP2, extracellular signal-regulated kinase (ERK) inhibitor PD98059, c-Jun NH2-ter-minal kinase (JNK) inhibitor SP600125, p38 mitogen-activated protein kinase (MAPK) inhibitor SB203580, andSmad3 phosphorylation inhibitor SIS3 were purchased fromCalbiochem (San Diego, CA, USA). TGF-b neutralizing anti-body 1D11 was purchased from R&D Systems (Minneapolis,MN, USA). Goat anti-CCN2 polyclonal antibody, mouse anti-b-actin monoclonal antibody, donkey antigoat, and goatantimouse horseradish peroxidase-linked secondary anti-bodies were obtained from Santa Cruz Biotechnology (SantaCruz, CA, USA). All biological tissue cultures were pur-chased from Invitrogen (Carlsbad, CA, USA).

Cell culture

After receiving the approval from the Research EthicsCommittee of National Taiwan University Hospital HsinchuBranch, human gingival tissues were obtained from fourpatients during periodontal surgery with informed consentfrom the patients. Four primary HGF cultures were estab-lished by an explant technique as previouslydescribed.19,21,22 Cells between passages 4 and 10 wereplated on Petri dishes (60 mm) at a density of 2 � 105 cells/dish and subjected to various treatments. Before treat-ment, the cells were incubated in serum-free media for 18hours. In selected experiments, HGFs were pretreated with10 mM of ERK inhibitor PD98059, 10 mM of JNK inhibitorSP600125, 10 mM of p38 MAPK inhibitor SB203580, 2 mMantioxidant NAC, 10 mM of Smad3 inhibitor SIS3, 1 mM of Src-family kinases inhibitor PP2, 1 mM of Src inhibitor-1, 10 mMof ALK5 inhibitor SB431542, or 10 mg/mL of TGF-b neutral-izing antibody 1D11 for 1 hour, and then treated with250 ng/mL CsA for 2 hours. Data given herein are the meansof four experiments.

Western blot analysis

Western blot analysis was performed as previouslydescribed.19,21,22 In brief, cells were harvested, lysed,separated by sodium dodecyl sulfateepolyacrylamide gelelectrophoresis, and transferred to polyvinylidene fluoridemembranes. The membrane was then incubated with

ces connective tissue growth factor expression in human gingivalFormosan Medical Association (2014), http://dx.doi.org/10.1016/

Figure 1 Cyclosporine A (CsA) stimulated CCN2 protein andmessenger RNA (mRNA) expression in human gingival fibro-blasts (HGFs). (A) Cells were treated with 500 ng/mL CsA forvarious periods. (B) HGFs were treated with various concen-trations (0e1000 ng/mL) of CsA for 2 hours. CCN2 protein levelswere measured by Western blot analysis. (C) HGFs weretreated with 250 ng/mL CsA for 0e4 hours and the CCN2 mRNAlevels in HGFs were measured by TaqMan gene expression as-says. Data are presented as means � standard error of themean. *p < 0.05 when compared with the control.

EGCG inhibits CsA-induced CCN2 in gingival fibroblasts 3

+ MODEL

antibodies against CCN2 (1:1000) and b-actin (1:1000; SantaCruz, Santa Cruz, CA, USA). Bound antibodies were detec-ted with horseradish peroxidase-linked secondary anti-bodies (Jackson ImmunoResearch Laboratories, WestGrove, PA, USA). We then added the Western LightingChemiluminescence Reagent (PerkinElmer, MA, USA) on themembrane and analyzed with a Fuji LAS-4000 lumino imageanalyzer (Fuji Photo Film Co., Tokyo, Japan).

Quantitative real-time polymerase chain reaction

After treatment, total cellular RNA was extracted fromcells, reverse transcribed, and analyzed using the TaqManGene Expression Assays (Applied Biosystems, Foster City,CA, USA) for CCN2 (ID: Hs00170014_m1) and glyceraldehyde3-phosphate dehydrogenase (ID: Hs99999905_m1) as previ-ously described.21

Statistical analysis

Data for each group are expressed as the means � standarderror of the mean. The unpaired Student t test was appliedto compare group differences. A p value < 0.05 was takento be significant.

Results

To examine the effect of CsA on the expression of CCN2,HGFs were treated with CsA and the levels of CCN2 proteinexpression were assessed by Western blot analysis. Asshown in Fig. 1A, CsA significantly increased CCN2 synthesisafter a 2-hour exposure to 500 ng/mL CsA. We then treatedHGFs with various concentrations of CsA. Fig. 1B shows thatthe levels of CCN2 protein increased by approximatelyfivefold after treatment with CsA at a concentrationgreater than 250 ng/mL for 2 hours. Therefore, 250 ng/mLCsA was used in subsequent experiments. We further usedquantitative real-time polymerase chain reaction toexamine whether the effect of CsA on CCN2 expressionoccurred at the transcription level. As shown in Fig. 1C,CCN2 messenger RNA expression increased by approxi-mately 2.5-fold after 1-hour exposure to 250 ng/mL CsA.

To investigate the signal transduction pathways involvedin CsA-induced CCN2 expression in HGFs, cells were pre-treated with 10 mM PD98059, 10 mM SP600125, 10 mMSB203580, 2 mM NAC, 10 mM SIS3, 1 mM PP2, 1 mM of Srcinhibitor-1, 10 mM SB431542, or 10 mg/mL of TGF-bneutralizing antibody for 1 hour before exposure to 250 ng/mL CsA. Results revealed that SP600125 and SIS3 reduced toapproximately 84.6% and 98.4% of the CsA-induced levels ofCCN2 protein (Fig. 2A and B). PD98059, SB203580, NAC,SB431542, two Src inhibitors, and TGF-b neutralizing anti-body had no significant effect on the CsA-induced CCN2protein expression. These results indicated that CsA-induced CCN2 protein expression is mediated through JNKand Smad signaling. We further investigated whether cur-cumin, EGCG, or lovastatin could affect CsA-induced CCN2expression in HGFs. As shown in Fig. 3A, 10 mM EGCG and20 mM lovastatin significantly inhibited CsA-induced CCN2expression, but curcumin did not show this effect.Furthermore, EGCG inhibited CsA-induced CCN2 expression

Please cite this article in press as: Wu K-J, et al., Cyclosporine A indufibroblasts: Suppression by epigallocatechin-3-gallate, Journal of thej.jfma.2014.01.008

in a dose-dependent manner (Fig. 3B). At concentrationsgreater than 10 mM, EGCG almost completely inhibited CsA-induced CCN2 expression in HGFs.

Discussion

It was shown that TGF-b plays an important role in thepathogenesis of CsA-induced GO. However, there isincreasing evidence that profibrotic activity of CsA is notonly due to enhanced TGF-b expression.3 Interactions be-tween various mediators might be involved in the patho-genesis of this disease.3 In this study, we found that CsAstimulated CCN2 production in HGFs. CCN2 can act togetherwith TGF-b to induce the maximal profibrotic effects ofTGF-b and thereby it contributes to the pathogenesis ofCsA-induced GO.

The GO was reported to be as high as 70% in patients whounderwent transplantation.3 It impaired the efficiency tomaintain good oral hygiene and may lead to accumulationof microorganisms. These microorganisms may compromisethe systemic health of the patients undergoing CsA therapy.It is of interest to note that there is no agreement in theliterature about the effects of plaque control on CsA-induced GO.23 Therefore, continued investigation of newstrategies for the treatment of GO is needed.

The gene for CCN2 has been shown to be regulated bymultiple signaling pathways in HGFs. Identifying moleculescontrolling signaling pathways associated with CsA-inducedCCN2 expression is important in the development of good

ces connective tissue growth factor expression in human gingivalFormosan Medical Association (2014), http://dx.doi.org/10.1016/

Figure 2 Effects of various inhibitors on the cyclosporine A(CsA)-induced CCN2 expression in human gingival fibroblasts(HGFs). Cells were pretreated with (A) 10 mM of extracellularsignal-regulated kinase (ERK) inhibitor PD98059, 10 mM of c-JunNH2-terminal kinase (JNK) inhibitor SP600125, 10 mM of p38mitogen-activated protein kinase (MAPK) inhibitor SB203580,2 mM antioxidant N-acetylcysteine (NAC), or (B) 10 mM ofSmad3 inhibitor SIS3, 1 mM of Src-family kinases inhibitor PP2,1 mM of Src inhibitor-1, 10 mM of activin receptor-like kinase 5(ALK5) inhibitor SB431542, 10 mg/mL of transforming growthfactor-b (TGF-b) neutralizing antibody 1D11 for 1 hour, andthen treated with 250 ng/mL CsA for 2 hours. CCN2 proteinlevels were measured by Western blot analysis.

Figure 3 Epigallocatechin-3-gallate (EGCG) and lovastatininhibited cyclosporine A (CsA)-induced CCN2 expression inhuman gingival fibroblasts (HGFs). Cells were pretreated with(A) 20 mM curcumin, 10 mM EGCG, 20 mM lovastatin or (B)increasing concentrations of EGCG (0e20 mM) for 1 hour, andthen treated with 250 ng/mL CsA for 2 hours. CCN2 levels weremeasured by Western blot analysis.

4 K.-J. Wu et al.

+ MODEL

Please cite this article in press as: Wu K-J, et al., Cyclosporine A indufibroblasts: Suppression by epigallocatechin-3-gallate, Journal of thej.jfma.2014.01.008

strategies for the treatment of GO. It was shown that CsAactivates latent TGF-b through reactive oxygen speciesgeneration, thereby initiating the Smad and p38 MAPKsignaling pathways, which synergistically induce TGF-b-induced target genes, such as CCN2 and tissue inhibitor ofmetalloprotease-1 in rat renal mesangial cells.24 Further-more, CsA-increased HGF proliferation25 and reduced pro-teolytic activity of GF7 are mediated through TGF-b in anautocrine fashion. In this study, pretreatment of GFs withTGF-b neutralizing antibody or SB431542 did not signifi-cantly affect the expression of CCN2 protein in CsA-treatedHGFs, indicating that TGF-b receptor binding or ALK activityis not involved in the CsA-induced CCN2 expression.

Black et al26 found that TGF-b induced CCN2 expressionthrough JNK and Smad signaling pathways, but not Ras/MEK/ERK in HGFs. Lovastatin reduces approximately 75% ofTGF-b1-induced CCN2 protein expression.20 The LPAinduced CCN2 expression in HGFs also through JNK/Smadpathway.20 Recently, we found that 10 mM curcumin and10 mM EGCG inhibited the TGF-b1-induced CCN2 proteinexpression in HGFs by 71.5% and 67.7%, respectively.27 Inaddition, curcumin and EGCG completely inhibited LPA-induced CCN2 expression in HGFs.19 In this study, wefound that CsA-induced CCN2 expression in HGFs wasalmost completely inhibited at concentrations of EGCGgreater than 10 mM, but not 20 mM curcumin. These resultssuggested that EGCG is a better choice for the preventionof CsA-induced GO.

In conclusion, our study demonstrates, for the first timeto our knowledge, that CsA at similar concentrations foundin the serum of patients undergoing CsA treatment(80e300 ng/mL)28 stimulated CCN2 production in HGFs.CCN2 can act as a cofactor with TGF-b to induce themaximal profibrotic effects of TGF-b and thereby contrib-utes to the pathogenesis of CsA-induced GO. The effects ofCsA-induced CCN2 expression can be completely inhibitedby EGCG. These results suggest that EGCG may be useful forthe prevention of CsA-induced GO. Nevertheless, furtherendeavors involving animal studies29 are required toconfirm the role of EGCG in preventing the CsA-induced GO.

Acknowledgments

This study was supported by grants from the NationalTaiwan University Hospital Hsinchu Branch and ProfessorKuo Shui Foundation, Taiwan.

References

1. Ting LS, Villeneuve E, Ensom MH. Beyond cyclosporine: a sys-tematic review of limited sampling strategies for other im-munosuppressants. Ther Drug Monit 2006;28:419e30.

2. Trackman PC, Kantarci A. Connective tissue metabolism andgingival overgrowth. Crit Rev Oral Biol Med 2004;15:165e75.

3. Chae HJ, Ha MS, Yun DH, Pae HO, Chung HT, Chae SW, et al.Mechanism of cyclosporine-induced overgrowth in gingiva. JDent Res 2006;85:515e9.

4. Sume SS, Kantarci A, Lee A, Hasturk H, Trackman PC. Epithelialto mesenchymal transition in gingival overgrowth. Am J Pathol2010;177:208e18.

5. Pakosz K, Zakliczy�nski M, Krol W, Pyka L, Zakliczy�nska H,Trybunia D, et al. Association of transforming growth factor b1

ces connective tissue growth factor expression in human gingivalFormosan Medical Association (2014), http://dx.doi.org/10.1016/

EGCG inhibits CsA-induced CCN2 in gingival fibroblasts 5

+ MODEL

(TGF-b1) with gingival hyperplasia in heart transplant patientsundergoing cyclosporine-A treatment. Ann Transplant 2012;17:45e52.

6. Buduneli N, Kutukculer N, Aksu G, Atilla G. Evaluation oftransforming growth factor-beta 1 level in crevicular fluid ofcyclosporin A-treated patients. J Periodontol 2001;72:526e31.

7. Cotrim P, de Andrade CR, Martelli-Junior H, Graner E, Sauk JJ,Coletta RD. Expression of matrix metalloproteinases incyclosporin-treated gingival fibroblasts is regulated by trans-forming growth factor (TGF)-beta1 autocrine stimulation. JPeriodontol 2002;73:1313e22.

8. Leask A, Abraham DJ. TGF-beta signaling and the fibroticresponse. FASEB J 2004;18:816e27.

9. Yata Y, Gotwals P, Koteliansky V, Rockey DC. Dose-dependentinhibition of hepatic fibrosis in mice by a TGF-beta solublereceptor: implications for antifibrotic therapy. Hepatology2002;35:1022e30.

10. Ling H, Roux E, Hempel D, Tao J, Smith M, Lonning S, et al.Transforming growth factor b neutralization ameliorates pre-existing hepatic fibrosis and reduces cholangiocarcinoma inthioacetamide-treated rats. PLoS One 2013;8. e54499.

11. Thompson K, Hamilton DW, Leask A. ALK5 inhibition blocksTGFß-induced CCN2 expression in gingival fibroblasts. J DentRes 2010;89:1450e4.

12. Denton CP, Merkel PA, Furst DE, Khanna D, Emery P, Hsu VM,et al. Recombinant human anti-transforming growth factorbeta1 antibody therapy in systemic sclerosis: a multicenter,randomized, placebo-controlled phase I/II trial of CAT-192.Arthritis Rheum 2007;56:323e33.

13. Bradham DM, Igarashi A, Potter RL, Grotendorst GR. Connec-tive tissue growth factor: a cysteine-rich mitogen secreted byhuman vascular endothelial cells is related to the SRC-inducedimmediate early gene product CEF-10. J Cell Biol 1991;114:1285e94.

14. Leask A, Parapuram SK, Shi-Wen X, Abraham DJ. Connectivetissue growth factor (CTGF, CCN2) gene regulation: a potentclinical bio-marker of fibroproliferative disease? J Cell Com-mun Signal 2009;3:89e94.

15. Brigstock DR. Connective tissue growth factor (CCN2, CTGF)and organ fibrosis: lessons from transgenic animals. J CellCommun Signal 2010;4:1e4.

16. Kantarci A, Black SA, Xydas CE, Murawel P, Uchida Y, Yucekal-Tuncer B, et al. Epithelial and connective tissue cellCTGF/CCN2 expression in gingival fibrosis. J Pathol 2006;210:59e66.

17. Ikawa Y, Ng PS, Endo K, Kondo M, Chujo S, Ishida W, et al.Neutralizing monoclonal antibody to human connective tissuegrowth factor ameliorates transforming growth factor-beta-induced mouse fibrosis. J Cell Physiol 2008;216:680e7.

18. Ponticos M, Holmes AM, Shi-wen X, Leoni P, Khan K,Rajkumar VS, et al. Pivotal role of connective tissue growth

Please cite this article in press as: Wu K-J, et al., Cyclosporine A indufibroblasts: Suppression by epigallocatechin-3-gallate, Journal of thej.jfma.2014.01.008

factor in lung fibrosis: MAPK-dependent transcriptional acti-vation of type I collagen. Arthritis Rheum 2009;60:2142e55.

19. Wang CY, Deng YT, Huang SY, Liu CM, Chang HH, Wong MY.Epigallocatechin-3-gallate inhibits lysophosphatidic acid-stimulated connective tissue growth factor via JNK andSmad3 suppression in human gingival fibroblasts. J Formos MedAssoc 2014;113:50e5.

20. Black Jr SA, Trackman PC. Transforming growth factor-beta1(TGFbeta1) stimulates connective tissue growth factor(CCN2/CTGF) expression in human gingival fibroblasts througha RhoA-independent, Rac1/Cdc42-dependent mechanism:statins with forskolin block TGFbeta1-induced CCN2/CTGFexpression. J Biol Chem 2008;283:10835e47.

21. Deng YT, Chen HM, Cheng SJ, Chiang CP, Kuo MY. Arecoline-stimulated connective tissue growth factor production inhuman buccal mucosal fibroblasts: modulation by curcumin.Oral Oncol 2009;45:e99e105.

22. Chen YW, Yang WH, Wong MY, Chang HH, Yen-Ping Kuo M.Curcumin inhibits thrombin-stimulated connective tissuegrowth factor (CTGF/CCN2) production through c-Jun NH2-terminal kinase suppression in human gingival fibroblasts. JPeriodontol 2012;83:1546e53.

23. Montebugnoli L, Servidio D, Bernardi F. The role of time inreducing gingival overgrowth in heart-transplanted patientsfollowing cyclosporin therapy. J Clin Periodontol 2000;27:611e4.

24. Akool el-S, Doller A, Babelova A, Tsalastra W, Moreth K,Schaefer L, et al. Molecular mechanisms of TGF beta receptor-triggered signaling cascades rapidly induced by the calcineurininhibitors cyclosporin A and FK506. J Immunol 2008;181:2831e45.

25. Cotrim P, Martelli-Junior H, Graner E, Sauk JJ, Coletta RD.Cyclosporin A induces proliferation in human gingival fibro-blasts via induction of transforming growth factor-beta1. JPeriodontol 2003;74:1625e33.

26. Black Jr SA, Palamakumbura AH, Stan M, Trackman PC. Tissue-specific mechanisms for CCN2/CTGF persistence in fibroticgingiva: interactions between cAMP and MAPK signaling path-ways, and prostaglandin E2-EP3 receptor mediated activationof the c-JUN N-terminal kinase. J Biol Chem 2007;282:15416e29.

27. Yang WH, Kuo MY, Liu CM, Deng YT, Chang HH, Chang JZ.Curcumin inhibits TGFb1-induced CCN2 via Src, JNK, andSmad3 in gingiva. J Dent Res 2013;92:629e34.

28. Buduneli N, Buduneli E, Ciotanar S, Atilla G, Lappin D,Kinane D. Plasminogen activators and plasminogen activatorinhibitors in gingival crevicular fluid of cyclosporin A-treatedpatients. J Clin Periodontol 2004;31:556e61.

29. Chin YT, Tu HP, Chen YT, Dai NT, Shen EC, Chiang CY, et al.Expression and bioactivities of endothelin-1 in gingiva duringcyclosporine A treatment. J Periodontal Res 2009;44:35e42.

ces connective tissue growth factor expression in human gingivalFormosan Medical Association (2014), http://dx.doi.org/10.1016/