J Periodontol • December 2003

1747

Gene Expression of Extracellular MatrixProteoglycans in Human Cyclosporin-Induced Gingival OvergrowthNelson Gnoatto,* Roberto F.M. Lotufo,* Odaly Toffoletto,† and Mônica V. Marquezini‡

Background: Gingival overgrowth is one of the side effectsassociated with the systemic use of cyclosporin A (CsA). In vitrostudies on the extracellular matrix of gingival tissues have demon-strated an altered composition, particularly an accumulation ofproteoglycans and collagen. We investigated the gene expressionof extracellular matrix proteoglycans in CsA-induced gingival tis-sue alterations.

Methods: mRNA expression of the proteoglycans perlecan,decorin, biglycan, and versican was analyzed by reverse transcrip-tion polymerase chain reaction (RT-PCR) in gingival samplesobtained from 12 individuals, six with CsA-induced gingival over-growth (CsA group) and six with a normal gingiva (controlgroup). The RT-PCR products were subjected to 1% agarose gelelectrophoresis containing ethidium bromide and analyzed qual-itatively and semiquantitatively by densitometry. Density valueswere normalized by determining the expression of the house-keeping gene ß-actin in the same sample. Groups were comparedby the Student’s t test.

Results: Perlecan expression showed a marked increase (54%)in the CsA group compared to the control group (P <0.01), whileno significant differences were observed for the other proteoglycans.

Conclusion: CsA-induced gingival overgrowth seems to beassociated with increased expression of perlecan, a typical base-ment membrane proteoglycan, but not decorin, biglycan, or ver-sican. J Periodontol 2003;74:1747-1753.

KEY WORDSCyclosporin A/adverse effects; gene expression; gingivalhyperplasia/pathogenesis; kidney transplantation/complications; liver transplantation/complications;proteoglycans.

* University of São Paulo School of Dentistry, Department of Periodontology, São Paulo, Brazil.† Federal University of São Paulo, Oswaldo Ramos Foundation, Kidney and Hypertension

Hospital, São Paulo.‡ São Paulo Blood Center, Pró-Sangue Foundation, Laboratory of Tumor Biology, São

Paulo.

Gingival overgrowth (GO) is a com-mon side effect associated with thesystemic use of cyclosporin A (CsA),

an immunosuppressant extensively usedin transplant patients to prevent graft rejec-tion1 as well as in treatment of immuno-logical diseases.2

Kimball3 reported the first case of drug-induced GO associated with the prolongeduse of the anticonvulsant phenytoin. Sub-sequently, similar gingival side effects wereobserved with the use of CsA,1,4 the anti-convulsant sodium valproate in isolatedcases,5 and the calcium channel blockersnifedipine,6,7 verapamil,8 and diltiazem.9

Despite their distinct pharmacodynamicproperties, the GO induced by these drugshas been considered a similar entity interms of its clinical and histological aspects,and a unifying hypothesis of its etiopatho-genesis has been proposed by someauthors.10,11 It is important to validate thisassumption based on studies on the patho-genesis of gingival alterations.

Clinical studies on CsA-immunosup-pressed patients suggest a positive corre-lation between the drug dose and theincidence and severity of GO.12,13 How-ever, no correlation between plaque indexand these parameters was observed inmost studies.14 Insights into GO gainedduring the last 2 decades point to tissueresponses that differ from those attributedto bacterial aggression under certainaspects. There is a growing interest ininvestigating CsA-induced GO based oninitial clinical studies on its prevalence,15,16

those trying to understand its morphol-ogy,17 and the in vitro behavior of human

2347.qxd 1/9/04 10:51 AM Page 1747

1748

Proteoglycans in Cyclosporin-Induced Gingival Overgrowth Volume 74 • Number 12

Figure 1.Typical clinical aspect of moderate (A) and severe (B) CsA-inducedgingival overgrowth.

fibroblasts in proliferative and collagen biosynthesisassays.18-23 So far, such studies have generated con-troversial interpretations. Some clinical laboratory stud-ies analyzing gene expression in GO have resulted ina partial elucidation of the pathogenic mechanismsinvolved.24,25

Signaling involved in the interaction between cellsand the extracellular matrix in CsA-induced gingivaltissue alterations has also been studied, with the obser-vation of an increase in the levels of the cytokines inter-leukin (IL)-626,27 and IL-1β,28 the fibrogenic growthfactors platelet-derived growth factor B (PDGF-B),29

and transforming growth factor β1 (TGF-β1).30 How-ever, TGF-β1 levels in the gingival crevicular fluid didnot show differences compared to the normal gingiva.31

Many questions regarding the descriptive profile of dif-ferent drug-induced GO remain unanswered. The mainquestion concerns the action of the drug on the cells,since this is part of a multifactorial tissue response,including plaque-induced inflammation. CsA-inducedGO is caused apparently by the accumulation of extra-cellular matrix, which is characterized by deficient col-lagen degradation32 accompanied by an increasedsynthesis of non-collagenous matrix components, partic-ularly proteoglycans and glycosaminoglycans.23,33 Defi-cient collagenolytic activity has been shown to be theresult of decreased gene expression;25 however, noinformation exists regarding the expression of proteo-glycans in this lesion.

Therefore, the aim of the present study was to ana-lyze the gene expression of the extracellular matrixproteoglycans perlecan, decorin, biglycan, and versicanin CsA-induced human gingival overgrown tissue com-pared to clinically normal gingiva from individuals notusing any drug.

MATERIALS AND METHODSStudy Population and Clinical ProceduresThe protocol and the informed consent were approvedby the Ethics Committees of the University of São Pauloand São Paulo Blood Center. Twelve individuals parti-cipated in this study: six transplant patients aged 20 to50 years (median: 32 years) (CsA group, four kidneytransplants, two liver transplants; three males, threefemales) receiving CsA therapy (daily dose of 2.5 to5.0 mg/kg) for 3 to 12 months and exhibiting GO andsix non-drug using individuals (control, two males, fourfemales), aged 38 to 45 years (median: 40 years).Patients recently receiving phenytoin, sodium valproate,nifedipine, diltiazem, verapamil, and/or azalide antibi-otics, which are known to interfere with extracellularmatrix composition, were excluded, as were clinicalcases of mild CsA-induced GO. All participants signedinformed consent forms.

During the dental examination and periodontal eval-uation, GO severity was determined for the CsA group

based on a modification of the semiquantitative indexdeveloped by Aas34 for patients receiving phenytoin: 0= no overgrowth; 1 = blunting of gingival margin; 2 =moderate GO (<one-third of crown length); or 3 =severe GO (>one-third of crown length). Gingival sam-ples of the area with moderate or severe GO were col-lected during corrective surgery of the anatomicaldeformity. Control samples were obtained from gingi-val tissue that was free of clinical signs of inflamma-tion as determined during routine preprosthetic surgicalprocedures (for example, resective surgery for crownlengthening). The typical clinical condition of CsAgroup is illustrated in Figure 1 (1A, moderate GO; 1B,severe GO). Four patients presented moderate andtwo presented severe GO.

Total RNA IsolationTotal RNA was extracted from frozen tissues using phe-nol and guanidine isothiocyanate reagent§ according tomanufacturer’s protocol. Briefly, we used 1 ml of reagentper 100 mg tissue, and processed the mixture with apower homogenizeri(2 pulses per second at 4°C). Then

§ TriZOL, Gibco BRL Life Technologies, Rockville, MD.i Ultra-Turrax T25, IKA Labortechnik, IKA-Werke GmbH & Co. KG, Staufen,

Germany.

2347.qxd 1/9/04 10:51 AM Page 1748

1749

J Periodontol • December 2003 Gnoatto, Lotufo, Toffoletto, Marquezini

0.2 mL chloroform per 1 ml phenol and guani-dine isothiocyanate was added and the sampleswere centrifuged at 12,000 × g for 15 minutesat 4°C. Isopropanol (0.5 ml per 1 ml phenoland guanidine isothiocyanate) was added tothe aqueous phase and the samples were cen-trifuged at 12,000 × g for 15 minutes at 4°C.The pellet (total RNA) was washed with 75%ethanol and resuspended in distilled watertreated with diethyl-pyrocarbonate.¶ The RNAcontent of each sample was quantified by mea-suring absorbance with a spectrophotometer.#

One unit of absorbance at 260 nm was con-sidered to correspond to 40 µg/ml RNA. Theabsorbance ratio at 260 and 280 nm was cal-culated to confirm the absence of protein con-tamination. RNA integrity of each sample wasdetermined electrophoretically on 1% agarosegels in tris-borate-ethylenediaminetetraeceticacid (EDTA) buffer (TBE; 50 mM Tris base,50 mM boric acid, 1 mM EDTA, pH 8).

RT-PCRTotal RNA samples whose 28S and 18S ribo-somal subunits were found to be intact were reverse tran-scribed using a commercially available RNA kit.** TotalcDNA was made from 3 µg each RNA sample using 0.5U reverse transcriptase and random hexamer primers ina final reaction volume of 20 µl at 42°C for 50 minutes,70°C for 15 minutes, and 4°C for 1 minute.

PCR was carried out with 1 µl cDNA, 0.2 mM eachdGTP, dATP, dTTP, and dCTP nucleotide, 0.2 µM of eachprimer (Table 1), 1.5 mM MgCl2 and 2.5 U DNA Taqpolymerase in PCR buffer, in a reaction volume of 50 µl.Table 1 shows the primer sequences. The reaction mix-ture was amplified in a thermal cycler†† using the fol-lowing program: denaturing for 5 minutes at 94°C, 35cycles of 1 minute at 94°C (denaturing) 1 minute at58°C (annealing), 1 minute at 72°C (extension), and afinal extension step at 72°C for 10 minutes. The PCRproducts were then subjected to 1% agarose gel electro-phoresis in TBE buffer, stained with ethidium bromide,sized against a 100 pb DNA ladder as standard,‡‡ andband densities were determined using an image analysissystem.§§ The identity of each PCR product was con-firmed by digestion with restriction enzymes.‡‡

Semiquantitative PCRFor this experiment, increasing amounts of cDNA (10to 700 ng) were subjected to PCR amplification underthe conditions described above and the gels were ana-lyzed for band densities.

Statistical AnalysisAll experiments were repeated at least three times. Theband densities corresponding to each primer studied

were divided by the density of the standard DNA bandwith a similar size in the gel (relative density) and thendivided by the relative density for the housekeepinggene β-actin calculated in the same sample (normal-ized density, which was expressed for β-actin as 100%).The normalized density values were used in the con-struction of a semilogarithmic plot as a function of thelogarithm of the amount of cDNA used in the reactions.Results were expressed as the mean value (± standarderror) of normalized densities identified in the linearprogressive phase of each plot and then as a percent-age of control (itself expressed as 100%). Once a nor-mal distribution was confirmed, the difference betweenCsA and control groups was analyzed by Student’s ttest, and statistical significance was set at the 95% con-fidence level. Computation was performed using twosoftware programs.ii ¶¶

RESULTSTotal RNA IntegrityTotal RNA extracted from the gingiva samples of bothgroups was free of protein, as demonstrated by a260/280 nm absorbance ratio higher than 1.6. Electro-phoresis of total RNA in 1% agarose gel showed theintegrity of 28S and 18S ribosomal RNA, indicatingthat the RNA preparations were intact.¶ Sigma Chemical Company, St. Louis, MO.# GeneQuant pro RNA/DNA Calculator, Amersham Pharmacia Biotech,

Uppsala, Sweden.** Superscript II kit, Gibco BRL Life Technologies.†† PCR PTC-200 Peltier thermal cycler, MJ Research, Watertown, MA.‡‡ Gibco BRL Life Technologies.§§ EagleEye II, Stratagene, La Jolla, CA.ii Microsoft Excel, Version 5.0, Microsoft Corporation, Redmond, WA.¶¶ GraphPad Prism for Windows, GraphPad Software Inc., San Diego, CA.

Table 1.

Primer Sequences and Expected Product Length

Gene Primer Sequences Length

β-actin

825 bp*Sense 5′ATC ATG TTT GAG ACC TTC AAC AC 3′Antisense 5′ TCT GCG CAA GTT AGG TTT TGT C 3′

Perlecam

503 pbSense 5′CAT GGG CTG AGG GCC TAC G 3′Antisense 5′ TGT GCC CAG GCG TCG GAA C 3’

Decorin

303 bpSense 5′ CTT ACG GAA TTA CAT CTT GA 3′Antisense 5′ AGA AGC CTT TTT GGT GTT GT 3′

Biglycan

357 bpSense 5′ ACA CAC CGG ACA GAT AGA 3′Antisense 5′ CTC TTT GGG CAC AGA CTT 3′

Versican

304 bpSense 5′ AAC ATT TTT CAG GTG GTG AG 3′Antisense 5′ GAT GCA GAA CTT GGA ACT AT 3′

* Base pairs.

2347.qxd 1/9/04 10:51 AM Page 1749

1750

Proteoglycans in Cyclosporin-Induced Gingival Overgrowth Volume 74 • Number 12

Identification of PCR ProductsPCR products using the primers for perlecan, decorin,biglycan, and versican were digested with NciI, HincII,AluI and RsaI, respectively. The fragments founded afterdigestion with the appropriate enzyme were of theexpected length, confirming their identity (Fig. 2).

Semiquantitative PCRFigure 3 shows a typical run of a semiquantitative PCRassay and the semilogarithm plot of the relative densitiesversus cDNA concentration, using β-actin (3A and 3D)and perlecan primers (3B, 3C, and 3D). For perlecan andall other primers, the relative density of each band wasnormalized against the relative density of β-actin bandin the same sample. The expression of perlecan in CsAgroup was 54% greater than in the controls (P ≤0.01;Table 2). Regarding decorin, biglycan, and versican,there was no difference between the groups (108%, 97%,and 108%, respectively; Table 2).

DISCUSSIONTo our knowledge, no study has thus far analyzed geneexpression of proteoglycans in CsA-induced GO. Westudied the expression of proteoglycans that predom-inate in the extracellular matrix of gingival tissue35-37

and identified an increased transcription of perlecancompared to the control group. The higher perlecanexpression might be related to the increase in subep-ithelial and vascular basement membranes reported inhistological studies on cyclosporin-induced GO.38,39

The functions of perlecan in the extracellular matrixinvolve important interactions between this macro-molecule and important signaling with other matrix

proteoglycans. These interactions are mediated bygrowth factors and have repercussions on collagen fib-rillogenesis, as also observed for other connective tis-sues.40 A high affinity for growth factors has beenattributed to this proteoglycan, which is able to retainthese factors inside the basement membrane. White-lock et al.40 confirmed the role of perlecan as a majorreservoir of fibroblast growth factor-2 (FGF-2) in vas-cular basement membranes. The great expansion ofthe vascular component observed in the lamina propriain cases of cyclosporin-induced GO17,41 probably leadsto a variety of events with repercussions on tissue over-growth. In certain stages of vascular development, per-lecan also functions as a regulator of endothelialproliferation, inhibiting vascular growth, as shown inthe rat experimental model.42 Handler et al.43 suggestedthat this regulating function of vasculogenesis is prob-ably modulated by perlecan since this proteoglycanreleases bound growth factors. The increased perlecanexpression observed in the CsA group may suggest arole of this proteoglycan in the process of gingival GO,possibly through FGF-2 metabolism. Further studiesare needed to better characterize the tissue interactionsinvolving perlecan, which, in turn, may contribute toelucidating the pathogenesis of GO.

The expression of decorin and biglycan, which wasapparently unaltered by CsA, can clarify many impor-tant questions regarding the pathogenesis of this lesion.These proteoglycans have been studied in various con-nective tissues and seem to play an important modu-lating role in collagen fibrillogenesis and organization.44

Decorin and biglycan are related to the amount of fib-rillar collagen synthesized in the extracellular matrix.45

These macromolecules form asystem of fibrogenic response inwhich both are able to respond toa variety of signals in a similarmanner.46,47 Decorin blocksgrowth factors under certain cir-cumstances. It shows high affinityfor different isoforms of TGF-β andhas been characterized as aninhibitor of cell proliferation, block-ing the activity of these fac-tors.48,49 On the other hand,biglycan releases TGF-β1, thusantagonizing the decorin-mediatedblockade.50,51 The unaltered geneexpression of decorin and bigly-can observed in this study is inaccordance with the hypothesisthat the large accumulation of col-lagen in tissue is possibly relatedto deficient collagen degradation.If, in contrast, collagen synthesiswere increased we would expect

Figure 2.One percent agarose gel electrophoresis of PCR amplification products of one CsA sample usingprimers for A) perlecan; B) decorin; C) biglycan; and D) versican. In all gels, lane L corresponds tothe DNA ladder (100 bp).A1, 503 pb product;A2, Nci I restriction products (309 and 194 bp); B1,303 pb product; B2, Hinc II restriction products (191 and 111 bp); C1, 357 pb product; C2,Alu Irestriction products (215 and 141 bp); D1, 304 pb product; D2, Rsa I restriction product (260 bp).

2347.qxd 1/9/04 10:51 AM Page 1750

1751

J Periodontol • December 2003 Gnoatto, Lotufo, Toffoletto, Marquezini

Figure 3.Semiquantitative PCR.Typical 1% agarose gel of PCR products usingprimers for the β-actin gene (A) and perlecan (B, CsA; C, control) andincreasing amounts of cDNA (10 to 700 ng). L, 100 pb standard. D.Semilogarithmic plot of relative density versus cDNA amount. Closedcircles, CsA cDNA, perlecan primers; open squares, control cDNA,perlecan primers; open circles, control cDNA, β-actin primers.

an accumulation of these two proteoglycans. The pre-sent results need to be confirmed by the determina-tion of the tissue levels of these proteoglycans andtheir glycosaminoglycan components (dermatan andchondroitin sulfate). Our findings are compatible withthe periodontal literature, which suggests that theexpression of procollagen types I and III is unaltered,

or even reduced, under the action of CsA24 and whichrelates collagen accumulation to reduced collagenaseactivity.25,32,52

In the present study, no difference in versican expres-sion was observed between the CsA and control groups.An increase in versican is often associated with hyper-cellularity,37,53 which is apparently not the case incyclosporin-induced GO.18,19,54-56 Within this context,the absence of increased versican expression in theCsA group is compatible with the cellular profile attrib-uted to the tissue analyzed. Versican is the largest inter-fibrillar proteoglycan and predominates in gingivalconnective tissue.35,37 This proteoglycan establishesan efficient molecular bridge between the fibroblast sur-face and the extracellular matrix stabilizing a supramo-lecular complex around the plasma membrane.

The analysis of non-collagenous matrix in cyclosporin-induced GO needs to be extended in order to betterunderstand the mechanisms of its pathogenesis. Someauthors defend the unifying hypothesis of GO inducedby different drugs based on an apparently commonmechanism of action that involves alterations in intra-cellular calcium.10,11 On the other hand, analysis of theproteoglycans/glycosaminoglycans in phenytoin-inducedGO showed an approximately 2-fold increase in thesemacromolecules in the tissue affected by the drug.57-60

In contrast, studies on the action of nifedipine showedpeculiar results in terms of both an increase10 and adecrease61 in these molecules, suggesting mechanismsof pathogenesis different from those acting in cyclosporin-induced GO. Only one biochemical study analyzed semi-quantitatively the glycosaminoglycans extracted directlyfrom four cyclosporin-induced GO biopsies.62 This analy-sis did not permit the identification of tissue proteogly-cans but rather allowed the quantification of the fourglycosaminoglycan family components (chondroitin sul-fate, dermatan sulfate, heparan sulfate, and hyaluronan).In contrast to the literature, the authors did not find anydifference in the glycosaminoglycan content comparedto normal controls. Wondimu et al.41 and Mariani et al.17

have observed an increase in the proportion of these

* Means ±± SE of six individual samples in each group.

Table 2.

Comparative Mean Normalized Density (nd)*

CsA Control CsA/ControlProteoglycan (nd) (nd) (%) P Value

Perlecan 106.3 ± 11.0 69.0 ± 8.0 154 <0.01

Decorin 213.0 ± 5.1 199.0 ± 1.4 108 NS

Biglycan 203.4 ± 5.8 210.0 ± 4.1 97 NS

Versican 343.8 ± 14.1 318.4 ± 11.6 108 NS

2347.qxd 1/9/04 10:51 AM Page 1751

1752

Proteoglycans in Cyclosporin-Induced Gingival Overgrowth Volume 74 • Number 12

extracellular matrix components using semiquantitativehistological techniques. It should be noted that all studiesonly employed small series. Other studies analyzed semi-quantitatively the biosynthesis of glycosaminoglycansin a generic manner (sulfate compounds or hyaluronan)in in vitro assays of gingival fibroblasts stimulated withCsA. The results suggest an increase of these com-pounds in cyclosporin-induced GO,33,63 although thesestudies used few cell lines showing great heterogeneity.

It should be noted that multiple mechanisms of acti-vation/inhibition of inflammatory mediators and matrixmetalloproteinases are involved in the selective poly-merization of proteoglycans. Small qualitative or quan-titative alterations in specific proteoglycans and growthfactors may lead to large changes in tissue architectureand function. The knowledge about the extracellularmatrix and its interactions thus far does not permit theunderstanding of the mechanism of pathogenesis ofcyclosporin-induced GO as a whole, but focuses atten-tion on the investigation of important components thatare affected. More detailed studies regarding mRNAexpression and the corresponding protein response arenecessary. This investigation may contribute to a bet-ter understanding of the etiopathogenesis of the lesionand may encourage the search for regulatory measuresof some local factors involved in its clinical course.

ACKNOWLEDGMENTThis study was supported by Fundação de Amparo àPesquisa do Estado de São Paulo-FAPESP Grant01/00392-9.

REFERENCES1. Calne RY, Rolles K, White DJ, et al. Cyclosporin A ini-

tially as the only immunosuppressant in 34 recipients ofcadaveric organs: 32 kidneys, 2 pancreases, and 2 liv-ers. Lancet 1979;2:1033-1036.

2. Powles RL, Clink HM, Spence D, et al. Cyclosporin-A toprevent graft-versus-host disease in man after allogeneicbone-marrow transplantation. Lancet 1980;1:327-329.

3. Kimball OP. The treatment of epilepsy with sodiumdiphenyl hydantoinate. JAMA 1939;112:1244-1245.

4. Rateitschak-Plüss EM, Hefti A, Lörtscher R, Thiel G. Initialobservation that cyclosporin-A induces gingival enlarge-ment in man. J Clin Periodontol 1983;10:237-246.

5. Syrjanen SM, Syrjanen KJ. Hyperplastic gingivitis in achild receiving sodium valproate treatment. Proc Finn DentSoc 1979;75:95-98.

6. Ramon Y, Behar S, Kishon Y, Engelberg IS. Gingivalhyperplasia caused by nifedipine: A preliminary report.Int J Cardiol 1984;5:195-204.

7. Lederman D, Lumerman H, Reuben S, Freedman PD.Gingival hyperplasia associated with nifedipine therapy:Report of a case. Oral Surg Oral Med Oral Pathol1984;57:620-622.

8. Colvard MD, Bishop J, Weissman D, Gargiulo AV. Cardizeminduced gingival hyperplasia: A report of two cases. Peri-odontal Case Rep 1986;8:67-68.

9. Giustiniani S, Cuna Robustelli F, Marieni M. Hyperplasticgingivitis during diltiazem therapy. Int J Cardiol 1987;15:247-249.

10. Lucas RM, Howell LP, Wall BA. Nifedipine-induced gin-gival hyperplasia: A histochemical and ultrastructuralstudy in human gingiva. J Periodontol 1985;56:211-215.

11. Seymour RA, Thomason JM, Ellis JS. The pathogenesisof drug-induced gingival overgrowth. J Clin Periodontol1996;23:165-175.

12. Somacarrera ML, Hernandez G, Acero J, Moskow BS. Fac-tors related to the incidence and severity of cyclosporin-induced gingival overgrowth in transplant patients. Alongitudinal study. J Periodontol 1994;65:671-675.

13. Hefti AF, Eshenaur AE, Hassell TM, Stone C. Gingivalovergrowth in cyclosporine A treated multiple sclerosispatients. J Periodontol 1994;65:744-749.

14. Seymour RA, Smith DG, Rogers SR. The comparativeeffect of azathioprine and cyclosporin on some gingivalhealth parameters of renal transplant patients. J ClinPeriodontol 1987;14:610-613.

15. Daley TD, Wysocki GP, Day C. Clinical and pharmaco-logic correlations in cyclosporine-induced gingival hyper-plasia. Oral Surg Oral Med Oral Pathol 1986;62:417-421.

16. Pernu HE, Pernu LMH, Knuuttila MLE, Huttunen KRH. Gin-gival overgrowth among renal transplant recipients anduraemic patients. Nephrol Dial Transplant 1993;8:1254-1258.

17. Mariani G, Calastrini C, Carinci F, Bergamini L, CalastriniF, Stabellini G. Ultrastructural and histochemical featuresof the ground substance in cyclosporin A-induced gingi-val overgrowth. J Periodontol 1996;67:21-27.

18. Bartold PM. Regulation of human gingival fibroblastgrowth and synthetic activity by cyclosporine-A in vitro.J Periodont Res 1989;5:314-321.

19. Barber MT, Savage NW, Seymour GJ. The effect ofcyclosporin and lipopolysaccharide on fibroblasts: Impli-cations for cyclosporin-induced gingival overgrowth.J Periodontol 1992;63:397-404.

20. Schincaglia GP, Forniti F, Cavallini R, et al. Cyclosporin-A increases type I procollagen production and mRNAlevel in human gingival fibroblasts in vitro. J Oral PatholMed 1992;21:181-185.

21. Willerhausen B, Lemen C, Schumacher U. Influence ofcyclosporin-A on growth and extracellular matrix synthe-sis of human fibroblasts. J Cell Physiol 1992;152:397-402.

22. James JA, Irwin CR, Linden GJ. The effects of cultureenvironment on the response of human gingival fibro-blasts to cyclosporin A. J Periodontol 1995;66:339-344.

23. Mariotti A, Hassel T, Jacobs D, Manning CJ, Hefti AF.Cyclosporin A and hydroxycyclosporine (M-17) affectthe secretory phenotype of human gingival fibroblasts.J Oral Pathol Med 1998;27:260-266.

24. Redlich M, Greenfeld Z, Cooperman H, Pisanty S, ShoshanS. Lack of influence of cyclosporin A on levels of gingi-val procollagen types I and III mRNAs in rats of differentages. Arch Oral Biol 1997;42:277-282.

25. Yamada H, Nishimura F, Naruishi K, et al. Phenytoin andcyclosporin A suppress the expression of MMP-1, TIMP-1,and cathepsin L, but not cathepsin B in cultured gingi-val fibroblasts. J Periodontol 2000;71:955-960.

26. Williamson MS, Miller EK, Plemons J, Rees T, Iacopino AM.Cyclosporine A upregulates interleukin-6 gene expressionin human gingiva: Possible mechanism for gingival over-growth. J Periodontol 1994;65:895-903.

27. Morton RS, Dongari-Bagtzoglou AI. Regulation of gingi-val fibroblast interleukin-6 secretion by cyclosporine A.J Periodontol 1999;70:1464-1471.

28. Myrillas TT, Kinden GJ, Marley JJ, Irwin CR. CyclosporinA regulates interleukin-1-beta and interleukin-6 expressionin gingiva: Implications for gingival overgrowth. J Peri-odontol 1999;70:294-300.

2347.qxd 1/9/04 10:51 AM Page 1752

1753

J Periodontol • December 2003 Gnoatto, Lotufo, Toffoletto, Marquezini

29. Nares S, Ng MC, Dill RE, Park B, Cutler CW, Iacopino AM.Cyclosporine A upregulates platelet-derived growth fac-tor B chain in hyperplastic human gingiva. J Periodontol1996;67:271-278.

30. James JA, Irwin CR, Linden GJ. Gingival fibroblastresponse to cyclosporin A and transforming growth fac-tor beta 1. J Periodont Res 1998;33:40-48.

31. Buduneli N, Kütükçüler N, Aksu G, Atilla G. Evaluationof transforming growth factor-beta 1 level in crevicularfluid of cyclosporin A-treated patients. J Periodontol2001;72:526-531.

32. Thomason JM, Sloan P, Seymour RA. Immunolocaliza-tion of collagenase (MMP-1) and stromelysin (MMP-3)in the gingival tissues of organ transplant patientsmedicated with cyclosporin. J Clin Periodontol 1998;25:554-560.

33. Newell J, Irwin CR. Comparative effects of cyclosporinon glycosaminoglycan synthesis by gingival fibroblasts.J Periodontol 1997;68:443-447.

34. Aas E. Hyperplasia gingivae diphenylhydantoinea: Aclinical, histological, and biochemical study. Acta Odon-tol Scand 1963;21(Suppl. 34):1-142.

35. Larjava H, Hakkinen L, Rahemtulla F. A biochemicalanalysis of human periodontal tissue proteoglycans.Biochem J 1992;284:267-274.

36. Bartold PM. Turnover in periodontal connective tissues:Dynamic homeostasis of cells, collagen and ground sub-stances. Oral Dis 1995;1:238-253.

37. Haase HR, Clarkson RW, Waters MJ, Bartold PM. Growthfactor modulation of mitogenic responses and proteo-glycan synthesis by human periodontal fibroblasts. J CellPhysiol 1998;174:353-361.

38. Tyldesley WR, Rotter E. Gingival hyperplasia inducedby cyclosporin-A. Br Dent J 1984;157:305-309.

39. Pisanty S, Rahamim E, Ben-Ezra D, Shoshan S. Pro-longed systemic administration of cyclosporin A affectsgingival epithelium. J Periodontol 1990;61:138-141.

40. Whitelock JM, Murdoch AD, Iozzo RV, Underwood PA.The degradation of human endothelial cell-derived per-lecan and release of bound basic fibroblast growth factorby stromelysin, collagenase, plasmin, and heparanases.J Biol Chem 1996;271:10079-10086.

41. Wondimu B, Reinholt FP, Modeer T. Stereologic study ofcyclosporin A-induced gingival overgrowth in renal trans-plant patients. Eur J Oral Sci 1995;103:199-206.

42. Benitz WE, Kelley RT, Anderson CM, Lorant DE, BernfieldM. Endothelial heparan sulfate proteoglycan. I. Inhibitoryeffects on smooth muscle cell proliferation. Am J RespirCell Mol Biol 1990;2:13-24.

43. Handler M, Yurchenco PD, Iozzo RV. Developmentalexpression of perlecan during murine embryogenesis.Dev Dyn 1997;210:130-145.

44. Iozzo RV. Matrix proteoglycans: From molecular designto cellular function. Annu Rev Biochem 1998;67:609-652.

45. Bianco P, Fisher LW, Young MF, Termine JD, Robey PG.Expression and localization of the two small proteogly-cans biglycan and decorin in developing human skele-tal and non-skeletal tissues. J Histochem Cytochem1990;38:1549-1563.

46. Vogel W, Gish GD, Alves F, Pawson T. The discoidin domainreceptor tyrosine kinases are activated by collagen. Mol Cell1997;1:13-23.

47. Shrivastava A, Radziejewski C, Campbell E, et al. Anorphan receptor tyrosine kinase family whose membersserve as nonintegrin collagen receptors. Mol Cell 1997;1:25-34.

48. Yamaguchi Y, Mann DM, Ruoslathi E. Negative regula-tion of transforming growth factor-beta by the proteo-glycan decorin. Nature 1990;346:281-284.

49. De Luca A, Santra M, Baldi A, Giordano A, Iozzo RV.Decorin-induced growth suppression is associated withup-regulation of p21, an inhibitor of cyclin-dependentkinases. J Biol Chem 1996;271:18961-18965.

50. Kresse H, Hausser H, Schonherr E, Bittner K. Biosyn-thesis and interactions of small chondroitin/dermatansulphate proteoglycans. Eur J Clin Chem Clin Biochem1994;32:259-264.

51. Takeuchi Y, Kodama Y, Matsumoto T. Bone matrix decorinbinds transforming growth factor-beta and enhances itsbioactivity. J Biol Chem 1994;269:32634-32638.

52. Tipton DA, Stricklin GP, Dabbous MK. Fibroblast hetero-geneity in collagenolytic response to cyclosporine. J CellBiochem 1991;46:152-165.

53. Tiedemann K, Malmstrom A, Westergren-Thorsson G.Cytokine regulation of proteoglycan production in fibro-blasts: Separate and synergistic effects. Matrix Biol 1996;15:469-478.

54. Tipton DA, Stricklin GP, Dabbous MK. Fibroblast hetero-geneity in collagenolytic response to cyclosporine. J CellBiochem 1991;46:152-165.

55. Rostock MH, Fry HR, Turner JE. Severe gingival over-growth associated with cyclosporin therapy. J Periodontol1986;57:294-299.

56. McGaw WT, Porter H. Cyclosporine-induced gingivalovergrowth: An ultrastructural stereologic study. OralSurg Oral Med Oral Pathol 1988;65:186-190.

57. Kantor ML, Hassell TM. Increased accumulation of sul-fated glycosaminoglycans in cultures of fibroblasts fromphenytoin-induced gingival overgrowth. J Dent Res 1983;62:383-387.

58. Dahllof G, Modéer T, Reinholt FP, Wikström B, Hjerpe A.Proteoglycans and glycosaminoglycans in phenytoin-induced gingival overgrowth. J Periodont Res 1986;21:13-21.

59. Bartold PM. Proteoglycans of the periodontium: Structure,role and function. J Periodont Res 1987;22:431-444.

60. Suresh R, Puvanakrishnan R, Dhar SC. Alterations inhuman gingival glycosaminoglycan pattern in inflam-mation and in phenytoin induced overgrowth. Moll CellBiochem 1992;115:149-154.

61. Tipton DA, Fry HR, Dabbous MK. Altered collagen metab-olism in nifedipine-induced gingival overgrowth. J Peri-odont Res 1994;29:401-409.

62. Rocha LA, Martins RC, Werneck CC, Feres-Filho EJ, SilvaLC. Human gingival glycosaminoglycans in cyclosporin-induced overgrowth. J Periodont Res 2000;35:158-164.

63. Zebrowski EJ, Pylypas SP, Odlum O, Johnson RB. Com-parative metabolism of 3H-glucosamine by fibroblastpopulations exposed to cyclosporine. J Periodontol1994;65:565-567.

Correspondence: Dr. Mônica V. Marquezini, Rua Dr. AlfredoEllis, 183, Apto 104, São Paulo, SP, 01322-050 Brazil. E-mail:marquezinissa@uol.com.br.

Accepted for publication May 2, 2003.

2347.qxd 1/9/04 10:51 AM Page 1753