Preferential Attachment of Human Gingival Fibroblasts tothe Resin Ionomer GeristoreFuwad Al-Sabek, DDS, MS, Sandra Shostad, DDS, MS, and Keith L. Kirkwood, DDS, PhD

AbstractThe resin ionomer Geristore has been used extensivelyfor root perforation repairs. The purpose of this studywas to evaluate oral in vitro biocompatibility of theresin ionomer Geristore compared to two other dentalperforation repair materials, Ketac-Fil and ImmediateRestorative Material (IRM). Growth and morphology ofhuman gingival fibroblasts (HGFs) was determined us-ing scanning electron microscopy (SEM) of HGFs cellsgrown on test materials as well as cytotoxicity assaysusing eluates from test materials. SEM analysis showedthat HGFs attached and spread well over Geristore withrelatively normal morphology. SEM showed that fibro-blasts did not attach and spread well over Ketac-Fil orIRM as cells appeared much fewer with rounded anddifferent morphology than fibroblasts grown on Geris-tore. Cytotoxicity assays indicated that HGFs prolifer-ated in the presence of Geristore eluates and not in thepresence of Ketac-Fil or IRM eluates. In vitro interpre-tation indicates that Geristore is less cytotoxic to gin-gival fibroblasts.

From the Departments of Endodontics and Periodontics,Oral Biology, and Pharmacology & Toxicology, State Universityof New York at Buffalo, Buffalo, NY.

Address request for reprints to Keith L. Kirkwood, DDS,PhD, Assistant Professor, Department of Periodontics/Preven-tion/Geriatrics, University of Michigan 1011 N. University Ave.,Ann Arbor, MI 48109-1078; E-mail address: klkirk@umich.edu.

Copyright © 2005 by the American Association ofEndodontists

Radicular perforations can occur during Endodonontic treatment, postspace prep-aration, postremoval, and operative procedures. Perforations that occur below the

level of the epithelial attachment can cause infection, which may result in attachmentloss. These defects have been treated in the past with a variety of different materialsincluding gutta-percha, calcium hydroxide products, amalgam, Immediate RestorativeMaterial (IRM), cavit, and super Epoxy Benzoic Acid (EBA). Major problems relative tothese materials include biocompatibility and inadequate sealing. The use of compositeresins, glass ionomer cement and Mineral Trioxide Aggregate (MTA) has been sug-gested to circumvent these problems. Studies have shown that MTA is the most suitablematerial for repairing sub-osseous perforations especially furcation perforations, be-cause of its superior sealing and biocompatibility characteristics (1, 2), as well aspermitting mineralized matrix formation (3).

Subgingival root perforations that are supraosseous have been repaired usingglass ionomers and composite resins. However, all studies have demonstrated somedegree of cytotoxicity through in vitro and in vivo studies (4 –14). Recently, resinionomers such as Geristore, originally designed for restorative procedures, have beenused in treating subgingival defects such as root resorption and perforation. Geristoreis a dual cure (both self and light-curing), hydrophilic, nonaqueous polyacid modifiedcomposite resin composed of fluoride releasing glass, mainly barium fluorosilicate, anda polymerizable organic matrix (Modified Bis-GMA, including 2-HEMA) combined witha photoinitiator. Advantages of these materials include insolubility in oral fluids, in-creased adhesion to tooth structure, dual cure capabilities, low cure shrinkage, lowcoefficient of thermal expansion, radiopacity, fluoride release, and biocompatibility(15). Relatively few clinical studies have addressed biocompatibility (15), althoughseveral clinical studies have demonstrated Geristore could repair subgingival and sub-osseous defects and could be used as a barrier for guided tissue regeneration (16 –22).Cytotoxicity of resin ionomers (Geristore) towards oral tissues has only started to beinvestigated (23). The aim of the present study is to evaluate in vitro biocompatibility ofhuman gingival fibroblasts with Geristore using scanning electron microscopy andnucleic acid fluorescent staining to measure cell attachment and morphology as well ascell cytotoxicity assays to eluates from root perforation repair materials namely Geris-tore, Ketac-Fil, and IRM.

Materials and MethodsCell Culture Conditions

Human gingival fibroblasts (HGF-1; American Type Culture Collection, Manassas,VA; #ATCC CRL-2014) were obtained through commercial sources for these studies.Cells utilized for these studies were between passages 14 and 21. Cells were cultured inDMEM (Invitrogen Life Technologies, Grand Island, NY), supplemented with 10% fetalbovine serum (Sigma, St. Louis, MO), penicillin (100 U/ml; Sigma), and streptomycin(100 �g/ml; Sigma) at 37°C in a humidified atmosphere of 5% CO2 in air. The culturemedium was changed every 3 to 4 days.

Sample Preparation and SterilizationGeristore (DEN-MAT Corporation, Santa Maria, CA.), Ketac Fil (ESPE, Seefeld/

Oberbay, Germany), and IRM (Caulk/Dentsply, Milford, DE) were obtained from com-mercial sources. Discs (6 � 2 mm) from the three materials were fabricated underaseptic conditions by packing the material after mixing in a Teflon washer (internaldiameter of 6 � 2 mm), and compressed between two glass slides to generate even

Basic Research—Technology

JOE — Volume 31, Number 3, March 2005 Gingival Fibroblast Attachment to Geristore 205

thickness of material. Glass cover slips were used as positive controls inall experimental conditions throughout this study. Extracts from allthree materials were prepared by preincubating sterilized discs in 50cm3 conical tubes with 5 ml of DMEM (Life Technologies) supple-mented with 100 U/ml penicillin and 100 �g/ml streptomycin at 37°C ina shaker (Labline, Barnstead Int., Dubuque, IA) at 150 rpm.

Scanning Electron Microscopy Growth AssayAttachment, growth and morphology of human gingival fibroblasts

over Geristore were evaluated by scanning electron microscopy. Geristore,Ketac Fil and IRM discs and control glass cover slips were sterilized asdescribed above and placed in the bottom of a 12-well culture plate. HGF-1cells were seeded into the wells at density (3�104 cells per well) in DMEMmedium containing 10% fetal bovine serum. After incubation, cells werewashed three times with phosphate buffered saline and fixed in 2.5% glu-taraldehyde in 1 mmol cacodylate buffer (pH 7.4). Samples were thenobserved using a JEOL JSM 6400 scanning electron microscope (Jeol USA,Peabody, MA). Cell attachment and viability in each experimental conditionwere assessed by qualitatively comparing morphology of cells over testedmaterials with that of cells over glass coverslips, which were consideredcells with normal morphology.

Cytotoxicity AssaysHuman gingival fibroblasts cytotoxicity towards Geristore, Ketac-

fil, and IRM material extracts was determined by means of using theCellTiter Aqueous Non-Radioactive Cell Proliferation Assay (Promega,Madison, WI). HGF-1 cells were seeded into 96-well culture plates at1 � 104 cells per well in DMEM supplemented with 10% fetal bovineserum. After 24h cells were rinsed three times with PBS and treated with100 �l of the following material eluates in serum-free media: Geristore(24 and 72 hr extracts), IRM (24 and 72 hr extracts), and Ketac-fil (24and 72 hr extracts). Controls included media only (no material ex-

tracts) and media without cells present. Soluble formazan absorbancewas recorded using an ELISA plate reader at 490 nm. These experimentwere repeated four times (n � 4). Cytotoxicity was calculated by ex-pressing absorbance values as a percentage of control values (100%represented zero cytotoxicity). ANOVA statistical analysis was used toassess if differences exist between controls and treatment groups andthe Tukey-Kramer test was used for multiple comparisons of means todetermine differences between treatment groups.

ResultsCellular Attachment and Growth Assays

For direct visual comparisons of gingival fibroblasts grown onGeristore, Ketac-Fil, and IRM, we employed SEM analysis to view cellmorphology and spreading over 72 h in culture under serum-free con-ditions. As a control, HGF-1 cells attached and spread well over glasscoverslips in these studies (Fig. 1A). Compared to glass coverslips,HGF-1 cells grew and spread qualitatively equally well over the surfaceof Geristore exhibiting characteristic elongated fibroblastic morphology(Fig. 1B). However, HGF-1 cells grown on coverslips appeared to havea smoother cell membrane compared to cells on Geristore. However,human gingival fibroblasts attached poorly to Ketac-Fil (Fig. 1C) andIRM (Fig. 1D) under the same experimental concentrations. In addi-tion, these experiments were repeated in the presence of serum (2 and10%) to rule out any deleterious effects on HGF-1 cells because of theabsence of serum. As shown in Fig. 2, HGF-1 cells could attach toGeristore in the presence of serum (Fig. 2A, B; 10 and 2% FCS, respec-tively) as well as in the absence of FCS (Fig. 2C). Cells grown on Geris-tore also had similar cellular morphology as glass coverslip controls(Fig. 2D). In parallel experiments, HGF-1 cells failed to attach to eitherKetac-Fil or IRM in the presence of serum (data not shown).

Figure 1. Scanning Electron Microscopic analysis of human gingival fibroblasts attachment and growth following 72-h incubation on (A) glass coverslip controls, (B)Geristore, (C) Ketac Fil, or (D) IRM. SEM views are shown at 100� with reference bar length indicated.

Basic Research—Technology

206 Al-Sabek et al. JOE — Volume 31, Number 3, March 2005

Cytotoxicity AssayTo determine if Geristore was less cytotoxic to human gingival

fibroblasts, in vitro cytotoxicity (MTS) assays were performed usingmaterial extracts with HGF-1 cells. Results obtained with the MTS assayindicated that Geristore was less toxic to HGF-1 cells after 24 h incuba-tion period than toxicity observed with Ketac Fil or IRM extracts (Fig. 3).In these studies, we used serum free media as the control since 24 and72 h material extracts were also prepared in serum free media (as-cribed 100% viability). Normalized percent viability values from multi-ple experimental samples (n � 4) were subjected to statistical analysis.ANOVA statistical analysis indicated that material extracts had a signifi-cant effect on cell proliferation (cell cytotoxicity) with a p � 0.0001(F � 13.105). Posthoc pair wise comparison using Tukey-Kramer mul-tiple comparison test indicated that significant differences exist betweenGeristore and both IRM and Ketac Fil. Results from Tukey-Kramer sta-tistical analysis indicated that Geristore was significantly different than0% FCS at 72 h (p � 0.05) but not at 24 h. However, IRM effects on cellviability was not significantly different from 0% FCS control at either 24or 72 h time points (p � 0.05). The effects of Ketac Fil extracts onHGF-1 cell viability was not significantly different from controls at 24 h(p � 0.05) but was significantly different at 72 h (p � 0.001). Signif-icant differences between Geristore and the other materials existed aswell. See Fig. 3 to compare Geristore with Ketac Fil and IRM at 24 and72 h time points. Significant differences between the same materialsusing 24 and 72 h extracts were compared using Student t test. Geristore(p � 0.04) and Ketac Fil (p � 0.0008) indicated significant differencesbetween material extracts but not for IRM (p � 0.057).

DiscussionMaterials used adjacent to oral tissues should have minimal cyto-

toxicity towards oral cells. Geristore is currently being used to repair

subgingival defects and the interaction between Geristore in the defectwith adjacent periodontal tissues must be critical in the wound healingprocess. Evaluation of cellular growth and attachment has been used totest cytotoxicity of dental materials (4, 6, 10, 24, 25). It has beensuggested that cell growth on the surface of a material is a more sensitiveindicator of cytotoxicity than surrounding cell growth (5, 26). Scanningelectron microscopy has been used to evaluate adhesion of cells onmaterials used in a proximity to periodontal tissues as a part of evalu-ating the cytotoxicity of these materials (27).

Geristore did not inhibit growth of gingival fibroblasts as evaluatedby scanning electron microscopy. Fibroblasts in this study grew andspread well over Geristore with a morphology close to that of the con-trols. Interestingly, we found that fibroblasts attached well to Geristoreeven in the absence of serum in the tissue culture media. When evalu-ated by the scanning electron microscopy, cells did not grow or spreadwell over Ketac Fil or IRM. These cells appeared rounded and balled upwhen compared with cells over Geristore and the controls in the pres-ence and absence of serum (Figs. 1 and 2). Adhesion and spreading ofcells on a material surface are the initial phase for cellular function. Thepersistence of rounded cells with little or no spreading suggests thesurface material may be toxic (28).

A recent study has also evaluated gingival fibroblast and periodon-tal ligament fibroblast attachment to Geristore (23). These investigatorsdetermined that cellular attachment occurred significantly greater thanother endodontic root-end filling materials tested with approximately90% attachment after 72 h. These studies are consistent with ourpresent study where we observed that gingival fibroblasts attached morereadily to Geristore than even glass cover slip controls. The same inves-tigators also determined that gingival fibroblasts attached to Geristore inan integrin-independent manner (23) indicating that integrins do notdirectly mediate attachment to this material.

Figure 2. Scanning Electron Microscopic analysis of human gingival fibroblasts attachment and growth following 72-h incubation on Geristore under different serumconditions. HGF-1 cells grown on Geristore in media containing (A) 10%, (B) 2%, or (C) 0% fetal calf serum (FCS) in tissue culture media. Panel D indicates HGF-1cells grown on glass coverslip controls in 10% FCS containing media. SEM views are shown at 500� with reference bar length indicated.

Basic Research—Technology

JOE — Volume 31, Number 3, March 2005 Gingival Fibroblast Attachment to Geristore 207

Regardless of the mechanism of cellular attachment to Geristore,gingival fibroblasts appeared to have less cellular cytotoxicity in thepresence of extracts prepared from Geristore. We observed that gingivalfibroblasts proliferated significantly better (p � 0.05) in the presenceof Geristore extracts compared to controls after 72 h in culture. Gersi-tore extracts were significantly less toxic to gingival fibroblasts than IRM(p � 0.01) or Ketac-Fil (p � 0.001) with 72-h extracts over the sameperiod of time.

It is unclear why Geristore has more favorable cellular response. Itmight be a result of certain surface characteristics that Geristore, espe-cially because cellular attachment is not dependent upon integrins(23). Another reason could be that Geristore elutes less toxic materialsinto the medium. Our current results support this conclusion, however,the exact component differences are not known at the present time.Several components of dental resin composite monomers or additivesare cytotoxic to fibroblasts, including gingival fibroblasts (10). There-fore, future studies will be conducted to evaluate the organic constitu-ents present in Gersitore and compare it to other resin or perforationrepair materials and test their effects on different oral cells.

References1. Osorio RM, Hefti A, Vertucci FJ, Shawley AL. Cytotoxicity of Endodonontic materials.

J Endod 1998;24:91– 6.2. Torabinejad M, Hong CU, McDonald F, Pitt Ford TR. Physical and chemical properties

of a new root-end filling material. J Endod 1995;21:349 –53.3. Thomson TS, Berry JE, Somerman MJ, Kirkwood KL. Cementoblasts maintain expres-

sion of osteocalcin in the presence of mineral trioxide aggregate. J Endod 2003;29:407–12.

4. Willershausen B, Schafer D, Pistorius A, Schulze R, Mann W. Influence of resin-basedrestoration materials on cytotoxicity in gingival fibroblasts. Eur J Med Res 1999;4:149 –55.

5. Hensten-Pettersen A, Helgeland K. Evaluation of biologic effects of dental materialsusing four different cell culture techniques. Scand J Dent Res 1977;85:291– 6.

6. Caughman WF, Caughman GB, Dominy WT, Schuster GS. Glass ionomer and compos-ite resin cements: effects on oral cells. J Prosthet Dent 1990;63:513–21.

7. Caughman WF, Caughman GB, Shiflett RA, Rueggeberg F, Schuster GS. Correlation ofcytotoxicity, filler loading and curing time of dental composites. Biomaterials 1991;12:737– 40.

8. Makkawy HA, Koka S, Lavin MT, Ewoldsen NO. Cytotoxicity of root perforation repairmaterials. J Endod 1998;24:477–9.

9. Franz A, Konig F, Anglmayer M, et al. Cytotoxic effects of packable and nonpackabledental composites. Dent Mater 2003;19:382–92.

10. Geurtsen W, Lehmann F, Spahl W, Leyhausen G. Cytotoxicity of 35 dental resin com-posite monomers/additives in permanent 3T3 and three human primary fibroblastcultures. J Biomed Mater Res 1998;41:474 – 80.

11. Schwarze T, Fiedler I, Leyhausen G, Geurtsen W. The cellular compatibility of fiveEndodonontic sealers during the setting period. J Endod 2002;28:784 – 6.

12. Kan KC, Messer LB, Messer HH. Variability in cytotoxicity and fluoride release ofresin-modified glass-ionomer cements. J Dent Res 1997;76:1502–7.

13. Theilig C, Tegtmeier Y, Leyhausen G, Geurtsen W. Effects of BisGMA and TEGDMA onproliferation, migration, and tenascin expression of human fibroblasts and keratin-ocytes. J Biomed Mater Res 2000;53:632–9.

14. Hanks CT, Wataha JC, Sun Z. In vitro models of biocompatibility: a review. Dent Mater1996;12:186 –93.

15. Dragoo MR. Resin-ionomer and hybrid-ionomer cements: part I. Comparison ofthree materials for the treatment of subgingival root lesions. Int J Periodontics Re-storative Dent 1996;16:594 – 601.

16. Dragoo MR. Resin-ionomer and hybrid-ionomer cements: part II. Human clinicaland histologic wound healing responses in specific periodontal lesions. Int J Peri-odontics Restorative Dent 1997;17:75– 87.

17. Resillez-Urioste F, Sanandajt K, Davidson RM. Use of a resin-ionomer in the treatmentof mechanical root perforation: report of a case. Quintessence Int 1998;29:115– 8.

18. Shuman IE. Repair of a root perforation with a resin-ionomer using an intentionalreplantation technique. Gen Dent 1999;47:392–5.

19. Scherer W, Dragoo MR. New subgingival restorative procedures with Geristore resinionomer. Pract Periodontics Aesthet Dent 1995;7(Suppl):1– 4.

20. Abitbol T, Santi E, Scherer W. Use of a resin-ionomer in guided tissue regeneration:case reports. Am J Dent 1995;8:267–9.

21. Abitbol T, Santi E, Scherer W, Palat M. Using a resin-ionomer in guided tissue regen-erative procedures: technique and application— case reports. Periodontal Clin In-vestig 1996;18:17–21.

22. Behnia A, Strassler HE, Campbell R. Repairing iatrogenic root perforations. J Am DentAssoc 2000;131:196 –201.

23. Camp MA, Jeansonne BG, Lallier T. Adhesion of human fibroblasts to root-end-fillingmaterials. J Endod 2003;29:602–7.

24. Willershausen B, Marroquin BB, Schafer D, Schulze R. Cytotoxicity of root canal fillingmaterials to three different human cell lines. J Endod 2000;26:703–7.

25. Geurtsen W, Leinenbach F, Krage T, Leyhausen G. Cytotoxicity of four root canalsealers in permanent 3T3 cells and primary human periodontal ligament fibroblastcultures. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:592–7.

26. Hensten-Pettersen A. Comparison of the methods available for assessing cytotoxicity.Int Endod J 1988;21:89 –99.

27. Zhu Q, Haglund R, Safavi KE, Spangberg LS. Adhesion of human osteoblasts onroot-end filling materials. J Endod 2000;26:404 – 6.

28. Zmener O, Cabrini RL. Adhesion of human blood monocytes and lymphocytes todifferent Endodonontic cements. A methodological in vitro study. J Endod 1986;12:150 –5.

Figure 3. HGF cell viability assay. Geristore, IRM, and Ketac Fil material extractswere prepared following 24 (A) and 72 (B) hour incubation periods in serum-free tissue culture media. HGF-1 cells were incubated for 24 h in the presenceof material extracts and MTS cell proliferation assay was used to determine cellviability in the presence of material extracts. Control cells were incubated inserum free media in the absence of any material. Statistically significant differ-ences (ANOVA, p � 0.001) using posthoc Tukey multiple comparison areindicated between control and test materials. Refer to text for details.

Basic Research—Technology

208 Al-Sabek et al. JOE — Volume 31, Number 3, March 2005