Introduction.
Teeth with severe pulpal or periapical inflammation can be successfully treated with the established techniques of cleaning and shaping the root canals, followed by obturation of the root canal system. A large variety of root canal sealers have been advocated for use, in conjunction with solid or semisolid filling materials. Currently, root canal sealers are available based on various formulas such as expoxy resin, calcium hydroxide and zinc oxide–eugenol. Ideally, root canal sealers should be biocompatible and have satisfactory physico-chemical properties. They should also be well tolerated by the periradicular tissues. Indeed, since these materials will be in direct contact with periapical tissues for prolonged periods of time, their biocompatibility is of primary importance. A biocompatible sealer should neither prevent nor hinder tissue repair, but should aid or stimulate the reorganization of injured structures. One method of testing the biological compatibility of root canal sealers is to use an in vitro model to determine the cellular response. This has the advantage that many factors and variables can be controlled (Arenholt-Bindslev & Bleeg 1990, Barbosa et al . 1993) and the cytotoxicity can be determined with reliability and reproducibility (Arenholt-Bindslev & Horsted-Bindslev 1989, Beltes et al . 1995). Numerous permanent cell lines and oral primary fibroblasts derived from humans or animals have been used to evaluate cytotoxic effects of extracts, as well as solid specimens of root canal sealers (Arenholt-Bindslev & Horsted-Bindslev 1989, Matsumoto et al . 1989, Briseno & Willershausen 1991, Gerosa et al . 1995, Beltes et al . 1995, Vajrabhaya et al . 1997, Osorio et al . 1998, Koulaouzidou et al . 1998, Geurtsen et al . 1998, Guigand et al . 1999, Cohen et al . 2000, Leonardo et al . 2000). Although test systems vary considerably in the way cytotoxicity is measured, most employ cells that are transformed or of tumour origin as the model for cell response. However, normal diploid cells differ from established or transformed cells in many ways such as: mitotic rate, density-dependent regulation of growth, mitochondrial function, and media selection (Holley 1975, Lechner & Kaighn 1979, Feigal et al . 1985). To date, there has been very little data on the cytotoxicity of various types of root canal sealer in different culture systems. The purposes of this investigation were to study the cytotoxic effects and the long-term biocompatibility of elutes of three types of root canal sealer (resin-based, zinc oxide–eugenol-based, and calcium hydroxide-based) on human primary periodontal ligament (PDL) cells and a permanent hamster cell line (V79 cells).

Materials and methods.

Sealers.
Six root canal sealers were evaluated: N2, Endomethansone, AH26, AHPlus, Canals and Sealapex. The materials tested were resin-based (Table 1), zinc oxide–eugenol-based- (Table 2), or calcium hydroxide-based (Table 3).

Sample fabrication.
The cements were mixed according to the manufacturers’ instructions. Triplicate sample disks of the root canal sealers were fabricated in sterile cylindrical glass moulds 10 mm in height and 3 mm in diameter. Excess flash was removed with a sterile scalpel. The specimens were placed in polyethylene vials directly after mixing.

Elute preparation.
Immediately after the initial setting period, each specimen was placed in 10 mL of fresh culture medium and then transferred into fresh media after 24 h, 48 h, 72 h and 7 days. After each elution period, the medium was removed and cytotoxicity determined after the cells and elutes were incubated for 24 h. Cells without addition of elutes acted as untreated control.

Table 1. Resin-based root canal sealers tested.


Table 2. Zinc oxide-eugenol-based root canal sealers tested.


Table 3. Calcium hydroxide-based root canal sealers tested.

Cell cultures.
The PDL cells were cultured from healthy premolars (Chang et al . 1999, Tai & Chang 2000). Explants were cultured in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% foetal calf serum and antibiotics (100 units mL 1 penicillin, 100 g mL 1 streptomycin, and 0.25 g mL 1 of fungizone). To avoid contamination from gingival tissue, the periodontal ligament was carefully removed from the middle third of the root with a scalpel. The fragments were grown in DMEM supplemented with 10% foetal calf serum and antibiotics. Cells from passages 3–8 were used in this study. A permanent cell line derived from Chinese hamster lung fibroblasts (V79) was cultivated in minimal essential medium, supplemented with 10% foetal calf serum and 1% penicillin, streptomycin, and neomycin. Subcultivation was performed on confluent cultures.

Cytotoxicity assay.
A simple colorimetric assay developed by Mosmann (1983), as a test for cell proliferation and survival, has been adapted for the measurement of cytotoxicity. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) (Sigma, St. Louis, MO, USA) solution was prepared as 1 mg mL –1 in complete medium just before use. Cells were diluted in fresh complete medium and seeded in 96-well plates (V79: 1 10 4 cells well 1 , PDL: 2 10 4 cells well 1 ). After overnight attachment, cells were treated with various extracts of sealers (200 L well 1 ) for 20 h, then 50 L MTT dye was added to each well. Plates were incubated in a CO 2 incubator for 4 h. Optical density was determined by eluting the dye with dimethyl sulfoxide (Sigma, St. Louis, MO, USA), and the spectrophotometric absorbance was measured at 550 nm using a spectrophotometer (Hitachi, Tokyo, Japan).

Statistical analysis.
Five replicates of each concentration were performed in each test. All assays were repeated three times to ensure reproducibility. Statistical analysis was conducted by one-way analysis of variance. Tests of differences of the treatments were analysed by Duncan’s test and a value of P < 0.05 was considered statistically significant.

The results showed that resin-based, zinc oxide–eugenolbased, and calcium hydroxide-based root canal sealers were cytotoxic to primary human PDL cultures and V79 cells by MTT assay. The sensitivity of toxicity depended on the materials tested and the cell culture system used. Figures 1 and 2 show the cytotoxic effects of three different sealers on PDL cells and V79 cells, respectively.
The six sealers clearly induced cytotoxic effects. N2 was significantly more cytotoxic ( P < 0.05) than the other sealers in both cultures (Figs 1, 2). As shown in Figs 1 and 2, Sealapex initially resulted in a good biocompatibility to PDL cells or V79 cells compared with the control ( P > 0.05). Day 1 elutes of AH26 and AHPlus significantly inhibited growth of cells, but the leaching of toxic substances diminished from day 2 (Figs 1, 2). Extracts from day 1 through 3 of AH26 and AHPlus inhibited viability of PDL significantly more in comparison to V79 cells ( P < 0.05), whereas the day 7 elution of AH26 and AHPlus on PDL cells was not significantly different from control values (Fig. 2).

Figure 1. Effect of elutes of three types of root canal sealer on V79 cells by MTT assay. Percentage of absorbance at each elute compared with that of control was calculated. Each bar represents a mean SD. *, ** denote significant differences from control values with P <0.05 and P <0.001, respectively.


Figure 2. Effect of elutes of three types of root canal sealer on human PDL cells by MTT assay. Percentage of absorbance at each elute compared with that of control was calculated. Each bar represents a mean SD. *, ** denote significant differences from control values with P <0.05 and P <0.001, respectively.

Canals and Endomethansone also demonstrated cytotoxic effects on both cultures (Figs 1, 2). However, human PDL cells were more sensitive to Canals than V79 cells ( P < 0.05). In contrast, Endomethansone significantly inhibited V79 cell viability compared to PDL cells ( P < 0.05).
As shown in Fig. 1, the reduction of cell viability from the final extracts of Sealapex was not significantly different from control values in both cultures. In addition, the final extracts of Canals, AHPlus, and AH26 did not exert cytotoxic effects on PDL cells (Fig. 2).
In general, the rank order for both cultures demonstrated the same pattern. The rank orders with respect to cytotoxicity were found to be as follows: N2 > Endomethansone > AH26 > AHPlus > Canals > Sealapex.

Discussion.
Many cell culture techniques have been applied to assess the cytotoxicity of root canal sealers. These methods are based on cell cultures with established or diploid cell lines and a few tissue explant techniques. Any material used for obturation will come into contact with, or close proximity to the periodontal ligament. Various cell lines (normal diploid or transformed) are commonly used in cytotoxicity evaluations, and since PDL cells are critical to a healthy periodontium, they were used in this study. In addition, the selection of a permanent cell line derived from the Chinese hamster was used because they are easily maintained in culture. Furthermore, donor biopsy variability was eliminated and greater reproducibility was possible.
The MTT assays are colorimetric methods for quantifying viable cell numbers. The methyl-tetrazolium ring is cleared by mitochondrial dehydrogenouses in viable cells to formazan, which has a blue colour and can be measured with a spectrophotometer (Mosmann 1983). The amount of formazan produced is directly proportional to the total viable cell number over a wide range of cell numbers. Whilst in proliferating cells, comparison of bioassays performed with 3 H-thymidine incorporation versus the MTT assay utilized in this investigation display similar results, the MTT assay reflects cell numbers at any stage in their growth cycle. Since dead cells are unable to produce the coloured formazon product, this assay can be distinguished from dead cells (Mosmann 1983). The advantages of this method are its simplicity, rapidity, and precision, in addition, it does not require radioisotopes.
In this study, the freshly prepared root canal sealers were placed immediately into medium. Clearly, canal sealer should be tested immediately after mixing and also after a period of time when it is assumed that they have reached their final chemical structure. Root canal sealers are inserted into the mouth in a freshly mixed, incompletely polymerized stage, and thus it is probable that, during a relatively short period after clinical application of the material, local responses are provoked by unreacted or only partially reacted components. After setting, it is possible that potentially toxic constituents may be released from the materials. The difference in toxicity patterns at the various elution times may be related to the degree of setting. This would be reflected in the rate of component leaching. Thus, the different time extracts might be important to determine long-term cytotoxicity of root canal sealers.
The cytotoxicity of extracts of six root canal sealers was evaluated using MTT assay in human PDL cells and V79 cells. Our results agree with previous reports (Arenholt-Bindslev & Horsted-Bindslev 1989, Matsumoto et al . 1989, Briseno & Willershausen 1991, Gerosa et al . 1995, Beltes et al . 1995, Vajrabhaya et al . 1997, Osorio et al . 1998, Koulaouzidou et al . 1998, Geurtsen et al . 1998, Guigand et al . 1999, Cohen et al . 2000, Leonardo et al . 2000) that all materials tested were cytotoxic. In the longterm extracts of sealers, human PDL cells were more sensitive than V79 cells, except for Endomethansone and N2. However, the final extracts of Sealapex, Canals, AHPlus, and AH26 did not exert cytotoxic effects on PDL cells. The different responses between PDL and V79 cells to various sealers is difficult to explain; it is probably due to differences in the origin of cells.
Both resin-based sealers, AH26 and AHPlus, were cytotoxic. The cytotoxicity of resin-based sealers may be related to the release of formaldehyde (Spångberg et al . 1993, Cohen et al . 1998). In addition, bisphenol A diglycidyl ether was identified as a mutagenic component of resin-based materials, which may also be cytotoxic (Heil et al . 1996). AHPlus exhibited a lower cytotoxic potential compared to AH26 in this study, to confirm previous reports (Koulaouzidou et al . 1998, Cohen et al . 2000). Moreover, the amount of formaldehyde released by AH26 is significantly higher than AHPlus (Spångberg et al . 1993, Cohen et al . 1998, Leonardo et al . 1999).
In the case of zinc oxide–eugenol-based sealers, moderate to severe cytotoxicity was observed. The toxicity decreased in an order of N2 > Endomethansone > Canals. It was suggested by Lindqvist & Otteskog (1981) that the cytotoxicity of zinc oxide–eugenol root canal sealers was attributable to free eugenol liberated from the set material. A previous study has shown that both N2 and Endomethansone sealers can release formaldehyde after setting (Leonardo et al . 1999). The combined effects of eugenol and formaldehyde might explain why N2 and Endomethansone were highly toxic. Apart from formaldehyde and eugenol, N2 also contains a variety of aromatic oils that are cytotoxic (Table 2). These might be the reason why N2 is the most toxic sealer.
The use of so called ‘biological’ sealers based on calcium hydroxide has been proposed for the permanent obturation of the root canal system. Sealapex is primarily made of calcium hydroxide, and it demonstrated only slight toxicity in the fresh state. However, it exhibited increasing toxicity when set to confirm the results of previous studies that reported considerable leakage of cytotoxic substances from the disintegrating sealer (Gerosa et al . 1995, Beltes et al . 1995, Leonardo et al . 2000). This instability in an aqueous environment might enhance the release of substances from set Sealapex.

References.

Arenholt-Bindslev D, Horsted-Bindslev P (1989) A simple model for evaluating relative toxicity of root filling materials in cultures of human oral fibroblasts.Endodontics and Dental Traumatology 5, 219-26.
Arenholt-Bindslev D, Bleeg H (1990) Characterization of two types of human oral fibroblast with a potential application to cellular toxicity studies: tooth pulp fibroblasts and buccal mucosa fibroblasts.International Endodontic Journal 23, 84-91.
Barbosa SV, Arakik Spångberg LSW (1993) Cytotoxicity of some modified root canal sealers and their leachable components.Oral Surgery, Oral Medicine and Oral Pathology 75, 357-61.
Beltes P, Koulaouzidou E, Kotoula V, Kortsaris AH (1995) In vitro evaluating of the cytotoxicity of calcium hydroxidebased root canal sealers.Endodontics and Dental Traumatology 11, 245-9.
Briseno BM, Willershausen B (1991) Root canal sealer cytotoxicity on human gingival fibroblasts: II. Silicone- and resinbased sealers.Journal of Endodontics 17, 537-40.
Chang YC, Tai KW, Chou LSS, Chou MY (1999) Effects of Camphorated parachlorophenol on human periodontal ligament cells in vitro.Journal of Endodontics 25, 779-81.
Cohen BI, Pagnillo MK, Musikant BL, Dentsch AS (1998) Evaluation of the release of formaldehyde for three endodontic filling materials.Oral Health 88, 37-9.
Cohen BI, Pagnillo MK, Musikant BL, Deutsch AS (2000) An in vitro study of the cytotoxicity of two root canal sealers.Journal of Endodontics 26, 228-9.
Feigal RJ, Yesilsoy C, Messer HH, Nelson J (1985) Differential sensitivity of normal human pulp and transformed mouse fibroblasts to cytotoxic challenge. Archives of Oral Biology 30, 609-13.
Gerosa R, Menegazzi G, Borin M, Cavallevi G (1995) Cytotoxicity evaluation of six root canal sealers.Journal of Endodontics 21, 446-8.
Geurtsen W, Leinenbach F, Krage T, Leyhausen G (1998) Cytotoxicity of four root canal sealers in permanent 3T3 cells and primary human periodontal ligament fibroblast cultures.Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 85 , 592-7.
Guigand M, Pellen-Mussi P, Legoff A, Vulcain JM, Bonnaure-Mallet M (1999) Evaluation of cytocompatibility of three endodontic materials.Journal of Endodontics 25, 419-23.
Heil J, Reifferscheid G, Waldmann P, Leyhausen G, Geurtsen W (1996) Genotoxicity of dental materials.Mutation Research 368, 181-94.
Holley RW (1975) Control of growth of mammalian cells in cell culture.Nature 258, 478-90.
Koulaouzidou EA, Papazisis KT, Beltes P, Geromichelos GD, Kortsaris AH (1998) Cytotoxicity of three resin-based root canal sealers: an in vitro evaluation. Endodontics and Dental Traumatology 14, 180-5.
Lechner JF, Kaighn ME (1979) Reduction of the calcium requirement of normal human epithelial cells by EGF.Experimental Cell Research 121, 432-5.
Leonardo MR, da Silva LAB, Filho MT, da Silva RS (1999) Release of formaldehyde by 4 endodontic sealers.Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 88, 221-5.
Leonardo RT, Consolaro A, Carlos IZ, Leonardo MR (2000) Evaluation of cell culture cytotoxicity of five root canal sealers.Journal of Endodontics 26, 328-30.
Lindqvist L, Otteskog O (1981) Eugenol: liberation from dental materials and effect on human diploid fibroblast cells.Scandinavian Journal of Dental Research 89, 552-6.
Matsumoto K, Inoue K, Matsumoto A (1989) The effect of newly developed root canal sealers on rat dental pulp cells in primary culture.Journal of Endodontics 15, 60-7.
Mosmann T (1983) Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays.Journal of Immunologic Methods 65, 55-63.
Osorio RM, Hefti A, Vertucci FJ, Shawley AL (1998) Cytotoxicity of endodontic materials.Journal of Endodontics 24, 91-6.
Spångberg L, Barbosa SV, Lavigne GD (1993) AH26 releases formaldehyde.Journal of Endodontics 19, 576-7.
Tai KW, Chang YC (2000) Cytotoxicity evaluation of perforation repair materials on human periodontal ligament cells in vitro.Journal of Endodontics 26, 395-7.
Vajrabhaya LO, Sithisarn P, Wilairat P, Leelaphiwat S (1997) Comparison between sulphorhodamin-R dye staining and 51 Cr-release method in cytotoxicity assay of endodontic sealers.Journal of Endodontics 23, 355-7.