Basic Research—Technology

Mineral Trioxide Aggregate–based Endodontic SealerStimulates Hydroxyapatite Nucleation in HumanOsteoblast-like Cell CultureLoise Pedrosa Salles, MSc,*† Ana L�ıvia Gomes-Corn�elio, MSc,* Felipe Coutinho Guimar~aes, MSc,†

Bruno Schneider Herrera, PhD,‡ Sonia Nair Bao, PhD,† Carlos Rossa-Junior, PhD,‡

Juliane Maria Guerreiro-Tanomaru, PhD,* and Mario Tanomaru-Filho, PhD*

Abstract

Introduction: The main purpose of this study was toevaluate the biocompatibility and bioactivity of a newmineral trioxide aggregate (MTA)-based endodonticsealer, MTA Fillapex (MTA-F; Angelus, Londrina, Brazil),in human cell culture. Methods: Human osteoblast-likecells (Saos-2) were exposed for 1, 2, 3, and 7 days toMTA-F, Epiphany SE (EP-SE; SybronEndo, Orange, CA),and zinc oxide–eugenol sealer (ZOE). Unexposedcultures were the control group (CT). The viability ofthe cells was assessed by MTT assay and themorphology by scanning electron microscopy (SEM).The bioactivity of MTA-F was evaluated by alkalinephosphatase activity (ALP) and the detection of calciumdeposits in the culture with alizarin red stain (ARS).Energy-dispersive X-ray spectroscopy (EDS) was usedto chemically characterize the hydroxyapatite crystallites(HAP). Saos-2 cells were cultured for 21 days for ARSand SEM/EDS. ARS results were expressed as thenumber of stained nodules per area. Statistical analysiswas performed with analysis of variance and Bonferronitests (P < .01). Results:MTA-F exposure for 1, 2, and3 days resulted in increased cytotoxicity. In contrast,viability increased after 7 days of exposure to MTA-F.Exposure to EP-SE and ZOE was cytotoxic at all timepoints. At day 7, ALP activity increase was significantin the MTA-F group. MTA-F presented the highestpercentage of ARS-stained nodules (MTA-F > CT >EP-SE > ZOE). SEM/EDS analysis showed hydroxyap-atite crystals only in the MTA-F and CT groups. In theMTA-F group, crystallite morphology and chemicalcomposition were different from CT. Conclusions:After setting, the cytotoxicity of MTA-F decreasesand the sealer presents suitable bioactivity to stimulateHAP crystal nucleation. (J Endod 2012;38:971–976)

From the *Department of Restorative Dentistry, Dental School ofBiological Sciences, University of Bras�ılia, Distrito Federal, Brazil; aAraraquara, S~ao Paulo, Brazil.

To CNPQ (Brazil) and CAPES (Brazil) for the fellowship grants (tTo CNPq, FAPDF (Brazil), FINEP (Brazil), and FAPESP (Brazil, graAddress requests for reprints to Dr Mario Tanomaru-Filho, Ru

tanomaru@uol.com.br0099-2399/$ - see front matter

Copyright ª 2012 American Association of Endodontists.doi:10.1016/j.joen.2012.02.018

JOE — Volume 38, Number 7, July 2012

Key WordsBioactivity, biocompatibility, hydroxyapatite, mineral trioxide aggregate sealer

Mineral trioxide aggregate (MTA) emerged as the material of choice for root perfo-ration repairs and root-end fillings in the 90s, a revolutionary period marked by

many advances in endodontics (1). MTA was developed at Loma Linda University andreceived approval from the Food and Drug Administration for human use in 1998 (2,3). Since then, MTA has shown excellent biological properties in several in vivo andin vitro studies (4–9). In cell culture systems, for example, MTA has been shown toenhance proliferation of periodontal ligament fibroblasts (6), to induce differentiationof osteoblasts (7, 8), and to stimulate mineralization of dental pulp cells (9). Thisbiocompatibility and bioactive potential raised the interest of scientists worldwide toimprove the handling characteristics and some physicochemical properties of MTAwith the intention of expanding its applicability in endodontics. Consequently, newMTA-based root-end filling cements and root canal sealers have been proposed(10–12), such as MTA Fillapex (MTA-F; Angelus, Londrina, Brazil).

The new MTA-based sealers reflect a current requirement to have materials forendodontic therapy that are able to stimulate the healing process of periapical tissues,instead of merely biocompatible or inert materials. As a result, MTA-F represents theeffort in combining a material of excellent biological properties as MTA with resinsand other components to improve diverse required properties of an endodontic sealerincluding adhesiveness, dimensional stability, working time, radiopacity, flow, and anti-bacterial effects. According to the manufacturer’s information, MTA-F is composed ofsalicylate resin, resin diluent, natural resin, bismuth oxide as radiopacifying agent, silicananoparticles, MTA, and pigments. The MTA itself consists of fine hydrophilic particlesof tricalcium silicate, tricalcium aluminum oxide, tricalcium oxide, gypsum (calciumsulfate dihydrate), and other mineral oxides (3). Gypsum is an important determinantof setting time. MTA cements generally contain less gypsum to allowmore handling time.Unfortunately, MTA-F data sheet lacks details about the natural resin, pigments, anddiluents composition.

It is important to investigate if the combination of these resins and other constit-uents influence the bioactive potential of MTA in the new endodontic sealer. Therefore,the main purpose of this study was to evaluate the biocompatibility and the bioactivity ofMTA-F in stimulating mineralization in Saos-2 cell culture compared with Epiphany SE

S~ao Paulo State University, Araraquara, S~ao Paulo, Brazil; †Cellular Biology Department, Institute ofnd ‡Department of Diagnosis and Surgery, Araraquara Dental School, S~ao Paulo State University,

o L.P.S. and A.L.G.-C.).nt 2010/10769-1) for supporting this study.a Humait�a, 1680, Caixa Postal 331, Centro, 14801-903 Araraquara, SP, Brazil. E-mail address:

MTA-based Endodontic Sealer Stimulates Hydroxyapatite Nucleation 971

Basic Research—Technology

(EP-SE; SybronEndo, Orange, CA) and zinc oxide–eugenol endodonticsealers. Saos-2 is a human osteoblast-like cell that provides a suitablemodel for studying late events of osteoblast differentiation (13).EP-SE is a self-etch, dual-cure, resin-based root canal filling system(14). Zinc oxide–eugenol (ZOE) is a commonly used endodontic sealerknown to present cytotoxicity (15).

Materials and MethodsPreparation of the Endodontic Sealers

MTA-F, EP-SE, and EndoFill (prepared according to the manufac-turer’s recommendations; Petropolis, RJ, Brazil) were inserted intosterile polythenemoldsmeasuring 6mm in diameter and 2mm in thick-ness (microcentrifuge tube caps). After 1 day of initial set at 37�C, 95%humidity, and 5% CO2, the sealer pellets were removed from the moldsand placed on the bottom of transwell permeable supports (0.4-mmmembranes; Corning, Union City, CA) on 12-well culture plates withculturemedium for 1, 2, 3, and 7 days of cell exposure. Before perform-ing the mineralization assays (alizarin red staining and scanningelectron microscopy [SEM]), sealer samples were set for 7 days inthe same conditions described previously before cell exposure. The7 days setting time for mineralization assays was established in accor-dance with MTT results.

Cell CultureHuman osteoblast cells (Saos-2 line ATCC HTB-85) were grown as

a monolayer culture in T-75 flasks (Corning, Union City, CA) containingDulbecco modified Eagle medium (DMEM) supplemented with 10%fetal bovine serum (FBS), penicillin (100 IU/mL), and streptomycin(100 mg/mL) until confluent. The cells were subcultured twicea week at 37�C, 95% humidity, 5% CO2 (all culture supplies fromGibco-BRL, Gaithersburg, MD). Adherent cells in logarithmic growthphase were detached by a mixture of trypsin/EDTA (0.25%) at 37�Cfor 2 minutes. The collected cells were seeded on 12-well plates(Corning) at a density of 2� 105 per well determined by hemocytom-etry. Then, cells were incubated in the same conditions describedearlier for 24 hours to obtain exponential cell growth before exposureto the endodontic sealers. Unexposed cells were the positive control.The culture media was renewed every 3 days. To investigate the effectof endodontic sealers on the Saos-2 phenotype and analyze formationof mineralized nodules by scanning electron microscopy and energydispersive X-ray spectroscopy (EDS), the cells were seeded over glassslides on 12-well plates (n = 3 slide samples/group for culture in oste-ogenic or nonosteogenic medium). After 1 day in culture, the cells wereexposed to the endodontic sealers deposited on the transwell inserts for21 days in osteogenic medium (DMEM, 10% FBS, 100 IU/mL penicillin,100 mg/mL streptomycin, 0.0023 g/mL b-glycerophosphate, and 0.055mg/mL L-ascorbate; Sigma Chemicals, St Louis, MO) or nonosteogenicmedium (DMEM, 10% FBS, 100 IU/mL penicillin, and 100mg/mL strep-tomycin). During the assays, the culture medium was renewed every 24hours for 1 week and later on every 2 days.

Cell ViabilityCell proliferation was determined by MTT assay. This assay is

based on the ability of the mitochondrial enzyme to convert the yellowwater-soluble tetrazolium salt, 3-(4,5-dimethyl-thiazoyl)-2,5-diphenyl-tetrazolium bromide (MTT, Sigma Chemicals) into purple-coloredformazan compounds. The absorbance measured is proportional tothe amount of viable cells. After 0, 1, 2, 3, and 7 days of cell exposureto endodontic sealers (or nonexposure in the control group), the trans-wells with dental material samples were removed, and the culturemediumwas changed to DMEM containing 0.55mg/mLMTT. The plates

972 Salles et al.

were incubated for an additional 4 hours in the same conditionsdescribed previously. Thereafter, each well was washed with 1 mL ofphosphate-buffered saline (PBS 1x) and 500mL acid-isopropyl alcohol(isopropyl alcohol, 0.04 N) were added to extract and solubilize the for-mazan. Aliquots of 150 mL of the formazan solution from each samplewere transferred to a 96-well plate (Corning) and the optical density(OD = 570 nm) was measured using an automated microplate reader(ELx800; BioTek Instruments, Winooski, VT). Duplicate samples wereprepared for each test and control groups at the different exposuretimes. The experiment was repeated three independent times(n = 6/group). Data were then exported to Excel spreadsheets (Office2007; Microsoft Corporation, Redmond, WA) and subjected to statis-tical analysis.

Alkaline Phosphatase ActivityAlkaline phosphatase (ALP) activity was determined using

a commercial kit (Labtest; Lagoa Santa, MG, Brazil). Briefly, ALP hydro-lyzes thymolphthalein monophosphate, releasing thymolphthalein inalkaline medium. After 1, 2, 3, and 7 days of exposure to the sealers,the attached Saos-2 cells were rinsed with PBS 1x and immersed in1 mL sodium lauryl sulfate (1 mg/mL, Sigma Chemicals) for 30 minutesat room temperature. Aliquots of each sample solution (50 mL) wereadded to the kit contents according to the manufacturer’s instructions.Absorbance was spectrophotometrically measured at 590 nm, and ALPactivity was calculated as mmol of thymolphthalein/min/L. Duplicatesamples were prepared for each test and control group at the differentexposure times. The experiment was repeated three times (n = 6/group). Data were expressed as ALP activity normalized by the numberof viable cells at the respective culture period (OD = 570 nm) (16).

Mineralization and Alizarin Red StainingAfter 21 days of cell exposure to endodontic sealers, adherent

Saos-2 was washed three times with PBS 1� and fixed in 10% (v/v)formaldehyde (Sigma) at room temperature for 15 minutes. The mono-layers were then washed twice with dH2O before the addition of 1 mLARS (2%, pH = 4.1) per well. The plates were incubated again atroom temperature for 20 minutes. Then, the wells were washed fivetimes with 2 mL dH2O. Stained monolayers were observed using an in-verted microscope (Axiovert 100, Carl Zeiss, Jena, Germany) with 40�magnification. The wells were photographed (Canon EOS-1D, CanonInc, Tokyo, Japan), and the digital images were processed using ImageJ1.45 software (National Institutes of Health, Bethesda, MD). Two exam-iners individually counted the stained nodules in each macroscopicimage (n = 6/group, including control).

Morphological Analysis of Mineralized Noduleson Saos-2 by SEM and EDS

After exposure to the endodontic sealers for 21 days in osteogenicor nonosteogenic medium, the cell slides were washed three times inPBS 1x and fixed in 2.5% glutaraldehyde for 2 hours. Specimenswere then dehydrated in ethanol series (30%, 50%, 70%, 90%, and100%) for 20 minutes at each concentration and dried in a critical-point dryer (LADD 28000; LADD, Williston, VT). Dried specimenswere mounted on stubs, sputter coated with gold, and observed bySEM for morphological characterization of the Saos-2 cells (n = 3 slidesamples/group). The crystallites on cells exposed or not (control) toendodontic sealers were screened by EDS for the presence of chemicalelements (n = 30 nodules/group). The systemwas operated at an accel-erating voltage of 15 kV (SEM-EDS JSM-7001F; JEOL, Tokyo, Japan).

JOE — Volume 38, Number 7, July 2012

Figure 1. (A) Saos-2 viability and (B) ALP activity after cells exposure todifferent sealers: MTA-F, EP-SE, and ZOE. Unexposed cells were the controlgroup.Mean� SEM (n= 6/group). *A significant difference between endodonticsealer treatment and control group. #A significant difference comparing toother groups of treatment and control. Analysis of variance, Bonferroni (P< .01).

Figure 2. (A) Micrographs of calcium nodules on Saos-2 culture stained withARS (2%). CT, control group; MTA-F, Saos-2 exposed to MTA Fillapex; EP-SE,Saos-2 exposed to Epiphany SE and ZOE, Saos-2 exposed to zinc eugenolendodontic sealer. The stain of MTA-F mineralized nodules was the mostintense (bar = 100 mm). (B) Quantitative analysis of mineralized nodules after21 days in culture. Data reported as mean � SEM (n = 6/group); differentletter represents significant difference between groups. Analysis of variance,Bonferroni post test (P < .01).

Basic Research—Technology

Statistical AnalysisThe MTT results, ALP activity data, alizarin-stained nodules, and

atom percentage of chemical elements (EDS of mineralized nodules)were evaluated by one-way analysis of variance. The mean differencesbetween all cell treatment groups were compared by Bonferroni posthoc test and were considered to be significant at P < .01.

ResultsCell Viability

The MTT assay showed significant decrease in cell viability for allendodontic sealers after 1, 2, and 3 days of cell exposure whencompared with the control group. At day 7, Saos-2 exposed to MTA-Frevealed a recovery from cytotoxicity (Fig. 1A). ZOE and EP-SE pre-sented cytotoxic effects at all time points. The groups of both dentalmaterials showed a significant decline in cell viability. Exposure toEP-SE significantly reduced cell viability in a time-dependent manner.

ALP ActivitySaos-2 treated with all test materials had significantly lower meta-

bolic activity and reduced levels of ALP activity compared with thecontrol group after 1 and 2 days of exposure (Fig. 1B). After 7 daysof exposure, only the MTA-F group showed enhancement of ALP activity(about two-fold compared with the control, P < .01).

Mineralization and ARSAfter 21 days of cell exposure, only MTA-F had a significant stim-

ulatory effect on the formation of a larger number of mineralizednodules than the control group. Moreover, the nodules in the MTA-Fgroup appeared intensely stained in the microscopic analysis(Fig. 2A). Statistical analysis of ARS data showed significant differencesamong the sealers (Fig. 2B). ARS-stained nodules were observed in theEP-SE cell culture although in a significantly lower quantity than in the

JOE — Volume 38, Number 7, July 2012

MTA-F and control groups. Sparse mineralized nodules were observedin ZOE-treated cells at all time points (MTA-F > CT > EP-SE > ZOE).

Morphological Analysis of Saos-2 and MineralizedNodules by SEM

In EP-SE and ZOE groups, no difference was observed in cellmorphology between cultures in nonosteogenic (Fig. 3A and B, respec-tively) or osteogenic medium (Fig. 3E and F). However, Saos-2 exposedto EP-SE and ZOE displayed morphology different from that of thecontrol, and no mineralized nodules were observed at all. Briefly, weobserved smaller cells with a loss of plasma membrane integrity, whichsuggested that cell necrosis was taking place (Fig. 3A, B, E, and F). In theMTA-F/nonosteogenic medium group, we observed a gel-like layer withnanoparticle precipitate (Fig. 3D). SEM cell morphology analysisshowed plasmatic membrane integrity and adhesion to glass slides after

MTA-based Endodontic Sealer Stimulates Hydroxyapatite Nucleation 973

Figure 3. Scanning electron micrographs of Saos-2 slides in (A-D) nonosteogenic medium and (E-H) osteogenic medium. (A and E) Saos-2 with disruption ofplasma membrane after exposure to EP-SE. (B and F) Dead Saos-2 of ZOE group. (C and G) Saos-2 and HAP crystallites of the control group. (D and H) Micro-graphs of the gel-like layer and Saos-2 membrane with crystallites after exposure to MTA-F, respectively. (I) SEM of a MTA-F pellet sample surface, which wasincubated in osteogenic medium at the same conditions of the cells for comparative purpose (bar = 1 mm [�20,000]).

Basic Research—Technology

21 days of MTA-F treatment in osteogenic media (Fig. 3H) although weobserved a lower number of cells in the MTA-F groups when comparedwith control cells. SEM analysis of MTA-F group confirmed the presenceof numerous hydroxyapatite-like crystals (0.2-0.8 mm) deposited overthe slide and attached to the cell membranes of Saos-2 in osteogenicmedium, both unexposed and exposed to MTA-F (Fig. 3G and H,respectively). However, the crystals in MTA-F group had differentmorphological features from those observed in the control group.Mineralized nodules in the MTA-F group showed an irregularly smoothsurface formed by compact spherical nanoparticles, whereas thenodules in the control cultures showed a rough surface formed by anaggregate of rod-shaped nanoparticles.

EDS Analysis of Mineralized NodulesEDS analysis of crystallites in the control and MTA-F groups of

Saos-2 cells cultured in osteogenic media confirmed hydroxyapatite(HAP) formation and differences in chemical element composition(Fig. 4A and B). EDS on HAP crystallites from the MTA-F/osteogenicmedium group displayed prominent peaks of Ca, P, O, and Si. Tracesof K, Z, N, Na, andMgwere also detected. Cells exposed toMTA-F showed

974 Salles et al.

nanocrystallite deposits even in nonosteogenic media. However, themain minerals in this group were Ca and Si (Fig. 4C); no P was detected.Only EDS of the MTA-F pellet sample displayed a peak of Bi (Fig. 4D).The statistical analysis of the chemical elements showed significantdifferences in the percentages of Si, P, and Ca atoms in the MTA-F groupcrystals compared with the control (Fig. 5). The Ca/P atomic ratios of thecontrol andMTA-F HAP crystallites in osteogenic mediumwere 2.01 and1.5, respectively. The Si/Ca ratios were 0.18, 1.18, and 2.43 for thecontrol, MTA-F/Saos-2 HAP crystals in osteogenic medium, and MTA-F/Saos-2 nano-crystallites in nonosteogenic medium, respectively.

DiscussionNew MTA-based sealers and cements have been the focus of many

studies in the endodontic field since the early investigations of Torabi-nejad et al (1, 2). The strong interest in developing MTA-basedendodontic materials is because of the excellent biocompatibility,bioactivity, and osteoconductivity of MTA (17). In this study, we evalu-ated the bioactivity of a novel endodontic sealer, MTA Fillapex, to inducemineralization in human Saos-2 cell culture system. Saos-2 cell line haswell-documented characterization, especially with regards to its high

JOE — Volume 38, Number 7, July 2012

Figure 4. EDS of hydroxyapatite crystallites from Saos-2 culture in osteogenic media (A) unexposed and (B) exposed to MTA-F displaying high peaks of Ca, P, andO. (C) EDS of nanocrystallites from the MTA-F group in nonosteogenic media. (D) EDS of a MTA-F pellet sample for comparative purpose.

Basic Research—Technology

expression of bone ALP and its ability to deposit mineralization-competent extracellular matrix (13).

The MTT results revealed that all materials, MTA-F, EP-SE, andZOE, exhibited a cytotoxic effect. The cytotoxicity of MTA-F was likelyrelated to the presence of resin in this material or the release of arsenic,a heavy metal that may be found as a contaminant in MTA (18). Arsenicreacts with protein thiols, and exposure to high concentrations of thiselement may induce genotoxicity. After 7 days of exposure to MTA-F, thecell culture showed evident recovery of viability. At this time point, theMTA-F sample was totally set. This suggests that the long setting time ofMTA-F may play a role in its cytotoxicity because of the leakage of toxiccompounds for a long drawn-out period. Interestingly, MTA-F main-tained antibacterial activity during 7 days after mixture in a differentstudy (19). Later, however, no antibacterial effect was detected. Theantibacterial effect of MTA-F was related to its resin component andto the pH range of 10.14 to 10.5 that MTA-F promoted in suspension.The cytotoxicity observed for EP-SE and ZOE in the present study hasbeen described (14, 20, 21). Two possible reasons for the highcytotoxicity of EP-SE include residual unreacted monomers such as

Figure 5. Atoms percentage in crystallites; * and # represent significantdifference in P, Ca, and Si content between crystals of the different groups.Analysis of variance, Bonferroni post test (P < .01).

JOE — Volume 38, Number 7, July 2012

2-hydroxyethyl methacrylate (HEMA) release from the dual-curecomposite and the filler particles leaching out (21). ZOE cements areknown to cause inflammation and bone resorption in vivo. Further-more, eugenol was shown to activate nuclear factor kappa B and toinduce cyclooxygenase-2 expression, vacuolization, and toxicity inhuman osteosarcoma cells in vitro (15).

ALP is a recognized marker of osteoblast differentiation and anessential enzyme in hydroxyapatite nucleation process. The initial cyto-toxic effect of MTA-F possibly caused a decrease in ALP activity at the firstdays of exposure to the dental material. However, Saos-2 showed signif-icant increase of ALP after 7 days of exposure to MTA-F, which supportsthe assumption of biocompatibility and bioactivity recovery once MTA-Fis completely set. ALP activity in this study was similar to that of Saos-2when exposed to MTA-based root-end filling materials after 3 days oftreatment (22).

The increase of ALP activity was consistent with the significantlyhigh number of ARS-stained nodules observed in Saos-2 culture thathad been exposed to MTA-F. In addition, the stain of mineralizednodules of MTA-F group appeared more intense under light micros-copy. These findings suggested that MTA-F had the potential to induceformation of mineralized nodules. Furthermore, this possibility led tothe characterization of mineralized nodules by SEM/EDS.

Surprisingly, most of the mineralized nodules observed in SEM ofthe MTA- F group had morphology different from mineralized nodulesin the control group. The diverse conditions of hydroxyapatite nucle-ation can explain this difference in nanocrystallites morphology. Nucle-ation is the first step in biomineralization and requires that a free energybarrier be overcome, which results in size and pattern formation ofmineral crystallites through the mediating role of many organic andinorganic substrates (23). Super-saturation, foreign ions, cellmembrane, and biomolecules, such as bone matrix proteins, may actas regulating agents in controlling the pattern of HAP crystalliteassembly. In the control group, the confluence of Saos-2 associatedwith themassive presence of matrix biomolecules provided the templateand site for a heterogeneous nucleation process. The presence of

MTA-based Endodontic Sealer Stimulates Hydroxyapatite Nucleation 975

Basic Research—Technology

biomolecules lowers the energy barrier against nucleation andimproves the structural match between crystalline phase and substrate,promoting structural synergy of the HAP crystals. On the other hand, it ispossible that MTA-F promoted an environment highly saturated withforeign ions, a lower number of cells, and a Ca-Si hydrogel nucleationsite. It was shown that relatively high super-saturation gives rise toa compact and random assembly of HAP crystallites (23). In addition,calcium silicate hydrate is the main byproduct of the reaction of MTAwith calcium silicate cements (24). Therefore, the SEM characterizationof the different morphology of HAP crystals in theMTA-F group, togetherwith EDS analysis of the crystallites, showed that Ca and Si leached fromthe dental material participated in the nucleation process of HAP crys-tals, probably through formation of Ca-Si hydrogel nucleation sites.Previous studies have shown that Ca and Si ions readily leach outfrom MTA and calcium silicate cements in solution (24). Bioglassand tricalcium silicate were shown to release silicon to a level of 50to 100 mg/L in cell culture medium (25).

The strong peaks for Ca, P, and Si in EDS analysis of crystallites inthe control and MTA-F groups are indicative of calcium phosphatedeposition. The Ca/P ratio of 1.5 to 2.0 observed in the elemental anal-ysis of crystallites from the MTA-F and control groups, respectively, isconsistent with the apatite maturation stage in which carbonate ionsmay replace phosphate ions (type B carbonate apatite, type B CAp)or hydroxyl ions (type A carbonate apatite, type A CAp) in the apatitestructure (26). Significant differences in the Si/Ca atomic ratios wereobserved between the control, MTA-F/osteogenic medium, and MTA-F/nonosteogenic medium groups. Two distinct possibilities couldexplain the high Si/Ca ratio in the crystallites from the MTA-F/osteogenic group: one is the occurrence of silicate coprecipitationwith apatite. Metasilicate is soluble in an aqueous environment in whichit reacts with calcium-forming polymerized gel-like insoluble silicate(25). The other possibility is the substitution of silicon for phosphorusin the apatite lattice of some crystallites (27) although there was nostatistical difference in P content between crystallites of the controland MTA-F/osteogenic groups in this study. In the MTA-F/nonosteogenic medium group, the 2.43 Si/Ca ratio reflects the presenceof calcium-silicate hydrate gel, a fine-grained and highly disorganizedhydrated silicate gel containing Si-OH groups that may provide nucle-ation sites for apatite formation (28, 29). Interestingly, the presenceof silicon from the MTA-F sealer within the HAP crystallites enhancesthis sealer’s bioactive potential because silicon is an essential elementfor the normal growth of bone and connective tissues. Silicon wasproven to stimulate DNA synthesis, ALP activity, osteocalcin expression,and fibroblast proliferation (30, 31).

Despite the initial cytotoxic effect during setting, the endodonticsealer MTA Fillapex can be considered a promising material for rootcanal treatment, considering its bioactive potential. In this study,MTA-F clearly showed the ability to stimulate nucleation sites for theformation of apatite crystals in human osteoblast-like cell culture.Further analysis is needed to elucidate whether silicon and othermineral components of this material are incorporated into the apatitecrystal lattice.

AcknowledgmentsThe authors thank to the Brasilia University Microscopy Lab

group for their cooperation with the micrograph images.The authors deny any conflicts of interest related to this study.

References1. Torabinejad M, Hong CU, McDonald F, Pitt Ford TR. Physical and chemical prop-

erties of a new root-end filling material. J Endod 1995;21:349–53.

976 Salles et al.

2. Torabinejad M, White DJ. United States Patent 5, 415, 547 USPTO. Patent Full Textand Image Database 1995.

3. Parirokh M, Torabinejad M. Mineral trioxide aggregate: a comprehensive literaturereview—part I: chemical, physical, and antibacterial. J Endod 2010;36:16–27.

4. Holland R, Filho JA, de Souza V, Nery MJ, Bernabe PF, Junior ED. Mineral trioxideaggregate repair of lateral root perforations. J Endod 2001;27:281–4.

5. De Deus G, Petruccelli V, GurgelFilho E, Coutinho-Filho T. MTA versus Portlandcement as repair material for furcal perforations: a laboratory study using a polymi-crobial leakage model. Int Endod J 2006;39:293–6.

6. Bonson S, Jeansonne BG, Laillier TE. Root-end filling materials alter fibroblastdifferentiation. J Dent Res 2004;83:408–13.

7. Nakayama A, Ogiso B, Tanabe N, Takeichi O, Matsuzaka K, Inoue T. Behavior of bonemarrow osteoblast-like cells on mineral trioxide aggregate: morphology and expres-sion of type I collagen and bone-related protein mRNAs. Int Endod J 2005;38:203–10.

8. Gomes-Filho JE, de FariaMD, Barnab�e PF, et al. Mineral trioxide aggregate but not light-cure mineral trioxide aggregate stimulated mineralization. J Endod 2008;34:62–5.

9. Yasuda Y, Ogawa M, Arakawa T, Kodowaki T, Takashi S. The effect of mineraltrioxide aggregate on the mineralization ability of rat dental pulp cells: an in vitrostudy. J Endod 2008;34:1057–60.

10. Bortoluzzi EA, Guerreiro-Tanomaru JM, Tanomaru-Filho M, Duarte MAH. Radio-graphic effect of different radiopacifiers on a potential retrograde filling material.Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009;108:628–32.

11. Corn�elio ALG, Salles LP, Campos da Paz M, Cirelli JA, Guerreiro-Tanomaru JM,Tanomaru-Filho M. Cytotoxicity of Portland cement with different radiopacifyingagents: a cell death study. J Endod 2011;37:203–10.

12. Camilleri J. The physical properties of accelerated Portland cement for endodonticuse. Int Endod J 2008;41:151–7.

13. McQuillan DJ, Richardson MD, Bateman JF. Matrix deposition by a calcifying humanosteogenic sarcoma cell line (SAOS-2). Bone 1995;16:415–26.

14. Al-Hiyasat AS, Tayyar M, Darmani H. Cytotoxicity evaluation of various resin-basedroot canal sealers. Int Endod J 2010;43:148–53.

15. Lee Y, Yang S, Ho W, Lee YH, Hung S. Eugenol modulates cyclooxygenase-2 expres-sion through the activation of nuclear factor kappa B in human osteoblasts. J Endod2007;33:1177–82.

16. Westgard JO, Barry PL, Hunt MR, Groth T. A multi-rule Shewhart chart for qualitycontrol in clinical chemistry. Clin Chem 1981;27:493–501.

17. Torabinejad M, Parirokh M. Mineral trioxide aggregate: A comprehensive literaturereview—part II: leakage and biocompatibility investigations. J Endod 2010;36:190–202.

18. Bramante CM, Demarchi AC, Moraes IG, et al. Presence of arsenic in different typesof MTA and white and gray Portland cement. Oral Surg Oral Med Oral PatholOralRadiol Endod 2008;106:909–13.

19. Morgental RD, Vier-Pelisser FV, Oliveira SD, Antunes FC, Cogo DM, Kopper PM. Anti-bacterial activity of two MTA-based root canal sealers. Int Endod J 2011;44:1128–33.

20. Versiani MA, Carvalho-Junior JR, Padilha MI, Lacey S, Pascon EA, Sousa-Neto MD. Acomparative study of physicochemical properties of AHPlus and Epiphany root canalsealants. Int Endod J 2006;39:464–71.

21. Yamanaka Y, Shigetani Y, Yoshiba K, Yoshiba N, Okiji T. Immunohistochemical anal-ysis of subcutaneous tissue reactions to methacrylate resin-based root canal sealers.Int Endod J 2011;44:669–75.

22. Gandolfi MG, Perut F, Ciapetti G, Mongiorgi R, Prati C. New Portland cement-basedmaterials for endodontics mixed with articaine solution: a study of cellularresponse. J Endod 2008;34:39–44.

23. Jiang H, Liu XY. Principles of mimicking and engineering the self-organized struc-ture of hard tissues. J BiolChem 2004;279:41286–93.

24. Camilleri J. Hydration characteristics of calcium silicate cements with alternative ra-diopacifiers used as root-end filling materials. J Endod 2010;36:502–8.

25. Wang X, Ito A, Sogo Y, Li X, Oyane A. Silicate-apatite composite layers on externalfixation rods and in vitro evaluation using fibroblast and osteoblast. J Biomed MaterRes 2010;92A:1181–9.

26. Gandolfi MG, Taddei P, Tinti A, Prati C. Apatite-forming ability (bioactivity) of Pro-Root MTA. Int Endod J 2010;43:917–29.

27. Gibson IR, Best SM, Bonfield W. Chemical characterization of silicon-substitutedhydroxyapatite. J Biomed Mater Res 1999;44:422–8.

28. Oliveira AL, Malafaya PB, Reis RL. Sodium silicate gel as a precursor for the in vitronucleation and growth of a bone-like apatite coating in compact and porous poly-meric structures. Biomaterials 2003;24:2575–84.

29. Carlisle EM. Silicon: a possible factor in bone calcification. Science 1970;167:279–80.

30. Reffitt DM, Ogston N, Jugdaohsingh R, et al. Orthosilicic acid stimulates collagentype 1 synthesis and osteoblastic differentiation in human osteoblast-like cellsin vitro. Bone 2003;32:127–35.

31. Hanasono MM, Lum J, Carroll LA, Mikulec AA, Koch RJ. The effect of silicone gel onbasic fibroblast growth factor levels in fibroblast cell culture. Arch Facial PlastSurg2004;6:88–93.

JOE — Volume 38, Number 7, July 2012