Angiogenic signaling triggered by cariogenic bacteria in pulp cells.

The inflammation observed in the dental pulp of teeth with deep caries lesions is characterized by a significant increase in blood vessel density. It is known that lipoteichoic acid (LTA) from Gram-positive cariogenic bacteria induces expression of vascular endothelial growth factor (VEGF) in dental pulp cells. The hypothesis underlying this study was that LTA induces VEGF expression in dental pulp cells through TLR2 and PI3k/Akt signaling. Odontoblast-like cells (MDPC-23) and undifferentiated pulp cells (OD-21) were exposed to LTA from Streptococcus sanguis, and the role of TLR2, PI3K/Akt, and IKK signaling in LTA-induced VEGF expression was evaluated. These studies demonstrated that TLR2 signaling through the PI3K-Akt pathway is necessary for LTA-induced VEGF expression in pulp cells. In contrast, inhibition of IKK signaling did not prevent VEGF up-regulation in response to LTA. Understanding signaling pathways triggered by cariogenic bacteria may reveal novel therapeutic targets for the clinical management of pulpitis.

Adhesive resin induces apoptosis and cell-cycle arrest of pulp cells.

The application of an adhesive resin near or directly over the pulp was shown to induce pulp inflammation and lack of dentin regeneration. We hypothesize that the absence of dentin bridging is due to adhesive-resin-induced apoptosis of cells responsible for pulp healing and dentin regeneration. Mouse odontoblast-like cells (MDPC-23), undifferentiated pulp cells (OD-21), or macrophages (RAW 264.7) were exposed to SingleBond polymerized for 0-40 seconds. Annexin V and propidium iodide assays demonstrated that SingleBond induced apoptosis of MDPC-23, OD-21, and macrophages. The proportion of apoptotic cells was dependent on the degree of adhesive resin polymerization. Adhesive-resin-induced death of pulp cells was associated with activation of the pro-apoptotic cysteine protease Caspase-3. Interestingly, most cells exposed to adhesive resin that did not undergo apoptosis showed cell-cycle arrest. We conclude that an adhesive resin induces apoptosis and cell-cycle arrest of cells involved in the regeneration of the dentin-pulp complex in vitro.

Lipoteichoic acid up-regulates VEGF expression in macrophages and pulp cells.

Vascular endothelial growth factor (VEGF) is a potent inducer of angiogenesis, vascular permeability, and edema. Up-regulation of VEGF expression in the dental pulp may result in increased intra-pulpal pressure, and contribute to pain and irreversible tissue damage. Lipoteichoic acid (LTA) is an amphiphilic molecule from Gram-positive bacteria that has been associated with the pathogenesis of pulpitis. To investigate if LTA regulates expression of VEGF, we exposed mouse odontoblast-like cells (MDPC-23), undifferentiated pulp cells (OD-21), fibroblasts, or macrophages to streptococcal LTA, and evaluated VEGF expression by ELISA and RT-PCR. LTA induced up to a nine-fold increase in VEGF protein expression in macrophages, a 2.4-fold increase in MDPC-23, and a 1.6-fold increase in OD-21 as compared with controls. In contrast, LTA did not induce VEGF expression in fibroblasts. VEGF mRNA expression remained constant upon exposure to LTA, which suggests that VEGF regulation in these cells is primarily post-transcriptional. This work constitutes the first demonstration that lipoteichoic acid is sufficient to induce expression of a pro-angiogenic factor.

The conservation and regulation of rat DSP-PP gene.

Two highly expressed noncollagenous proteins associated with dentin mineralization, dentin sialoprotein (DSP) and phosphophoryn (PP), are encoded by a single DSP-PP transcript. To better understand how DSP-PP transcripts are regulated, we have determined the DSP-PP transcription start site, sequenced its 5' flanking region, and analyzed the transcriptional activity of the gene promoter out to -1615 bp. Comparison of the rat cDNA sequence with the mouse, rat and human genes clearly indicates high sequence conservation within the DSP-PP 5' flanking region, implicating the possible presence of highly conserved gene regulatory cis elements. Among a number of conserved transcription sites identified in the 5' flanking region, we demonstrate that the conserved Y box sequence (ATTGG) can specifically bind nuclear extracts from mouse MDPC23 cells. This sequence (located within the -57 bp/-52 bp 5' flanking region) therefore likely represents one DSP-PP transcriptional regulatory sequence.

Cytotoxic effects of cleansing solutions recommended for chemical lavage of pulp exposures.

PURPOSE: To evaluate the in vitro cytotoxic effects of three cleansing solutions used for chemical lavage of pulp exposures. MATERIALS AND METHODS: The immortalized odontoblast cell line (MDPC-23) was plated (30,000 cells/cm2) and incubated for 72 hrs in 24-well dishes. After counting the cell number under inverted light microscopy, 20 microl of the experimental and control solutions were added to 980 microl of fresh culture medium. Then, hydrogen peroxide (3%, H2O2), sodium hypochlorite (6%, NaOCl) or calcium hydroxide-saline solution (5g of Ca(OH)2 in 10 ml of sterile distilled water) were added to wells for experimental Groups 1, 2 and 3, respectively. The positive and negative control groups received Syntac Sprint bonding agent (SS) and phosphate buffered saline (PBS), respectively. Following incubation for 120 min the cell number was counted again, the cell morphology was evaluated by scanning electron microscopy (SEM) and the cell metabolism was determined by the methyltetrazolium (MTT) assay. The scores obtained from cell counting and MTT assay were analyzed with an ANOVA followed by Fisher's PLSD tests. RESULTS: H2O2, NaOCl solutions, and SS bonding agent were more cytotoxic than Ca(OH)2 or PBS. In the groups with H2O2 or SS, only a few cells remained attached to the bottom of wells. The difference between these two groups was not statistically significant. H2O2, NaOCl and SS depressed the mitochondrial enzyme response by 97.7%, 97.3%, and 95.0%, respectively. On the other hand, Ca(OH)2 depressed the metabolic activity of cells by only 5%. While H2O2, NaOCl and SS caused extreme changes on the cell morphology, neither Ca(OH)2 nor PBS promoted dramatic changes in the cell morphology.

Current status of pulp capping with dentin adhesive systems: a review.

Several studies have assessed the morphology and thickness of hybrid layer, the dentin bond strengths as well as sealing ability of dentin adhesive systems. However, few in vivo studies have evaluated the biocompatibility of the adhesive systems following application to deep dentin or directly to the pulp of human teeth. Many studies performed in non-human primate teeth or teeth of rats have reported pulp healing and dentin bridging following pulp capping with bonding agents. In addition, a few clinical and radiographical reports of the success of resin pulp capping have been described in the dental literature. OBJECTIVES: The aim of this review was to evaluate the literature on pulp responses following total acid etching and application of adhesive resins on deep cavities or pulp exposures. In addition, the clinical/radiographical evidence for the apparent success of vital pulp therapy and results obtained from animal and human studies were compared and discussed. SIGNIFICANCE AND CONCLUSIONS: The self-etching adhesive systems may be useful and safe when applied on dentin. In contrast, persistent inflammatory reactions as well as delay in pulpal healing and failure of dentin bridging were seen in human pulps capped with bonding agents. The results observed in animal teeth cannot be directly extrapolated to human clinical conditions. Consequently, vital pulp therapy using acidic agents and adhesive resins seems to be contraindicated.

Cytotoxic effects of current dental adhesive systems on immortalized odontoblast cell line MDPC-23.

OBJECTIVES: Evaluate the cytotoxic effect of the three dental adhesive systems. METHODS: The immortalized mouse odontoblast cell line (MDPC-23) was plated (30,000 cell/cm2) in 24 well dishes, allowed to grow for 72 h, and counted under inverted light microscopy. Uncured fresh adhesives were added to culture medium to simulate effects of unset adhesive. Three adhesives systems were applied for 120 min to cells in six wells for each group: Group 1) Single Bond (3M), Group 2) Prime & Bond 2.1 (Dentsply), and Group 3) Syntac Sprint (Vivadent). In the control group, PBS was added to fresh medium. The cell number was counted again and the cell morphology was assessed under SEM. In addition, the adhesive systems were applied to circles of filter paper, light-cured for 20 s, and placed in the bottom of 24 wells (six wells for each experimental materials and control group). MDPC-23 cells were plated (30,000 cell/cm2) in the wells and allowed to incubate for 72 h. The zone of inhibition around the filter papers was measured under inverted light microscopy; cell morphology was evaluated under SEM; and the MTT assay was performed for mitochondrial respiration. RESULTS: The fresh adhesives exhibited more toxic (cytopathic effects) to MDPC-23 cells than polymerized adhesives on filter papers, and as compared to the control group. The cytopathic effect of the adhesive systems occurred in the inhibition zone around the filter papers, which was confirmed by the MTT assay and statistical analysis (ANOVA) combined with Fisher's PLSD test. In the control group, MDPC-23 cells were dense on the plastic substrate and were in contact with the filter paper. In the experimental groups, when acid in the adhesive systems was removed by changing the culture medium, or when the adhesives were light-cured, some cells grew in the wells in spite of the persistent cytotoxic effect. SIGNIFICANCE: All dentin adhesive systems were cytotoxic odontoblast-like cells. Both acidity and non-acidic components of these systems were responsible for the high cytopathic effect of those dental materials.

Cloned 3T6 cell line from CD-1 mouse fetal molar dental papillae.

Only primary pulpal cell cultures and one virally transformed mouse cell culture have been formally reported in the literature to synthesize proteins such as phosphophoryn which are unique to dentin matrix. In the present study, a mixed culture was derived from dental papilla cells of 18-19 fetal day CD-1 mouse mandibular first molars, maintained on a 3T6 plating regimen, and subsequently cloned after 28 passages. This cloned cell line (MDPC-23) exhibited several unique features, some of which were characteristic of odontoblasts in vivo. The features of this cell line included (1) epithelioid morphology of all cells with multiple cell membrane processes, (2) high alkaline phosphatase activity in all cells, (3) formation of multilayered nodules and multilayered cultures when maintained in ascorbic acid and beta-glycerophosphate, and (4) expression of two markers for odontoblast differentiation, i.e. dentin phosphoprotein and dentin sialoprotein.

Expression of dentin sialoprotein (DSP) and other molecular determinants by a new cell line from dental papillae, MDPC-23.

The purpose of this study was to characterize the molecular expression of a spontaneously immortalized and cloned cell line (MDPC-23) derived from 18-19 day CD-I fetal mouse molar dental papillae to determine if these cells were odontoblast-like. Western blots showed that a protein band, at approximately 105 kDa, reacting positively with anti-DSP antibodies and co-migrating with mouse DSP, was present in lysates of cells from passages 7, 37 and 77, in serum-free conditioned medium from passage 37 cells, and in mouse dentin extract. A minor band at 55 kDa was also apparent in cell lysates. Using a cDNA probe for a 486bp mouse DSP coding sequence, DSP or DSP-PP mRNA expression was detected by Northern analysis as well as Southern analysis after RT-PCR in all three passages. It was also shown that in these cells 1,25 (OH)2 vitamin D3 upregulated both osteopontin and osteocalcin mRNA, and dexamethasone downregulated alkaline phosphatase and alpha2(I) collagen mRNA. Thus, MDPC-23 cells express proteins which are common to mineralizing tissue. The expression of DSP and DSP-PP strongly suggests that this cell line is from the odontoblast lineage.

Dentin-specific proteins in MDPC-23 cell line.

Only four established odontoblast-like cell lines have been reported in the literature (1-6). Of the four, only two synthesize dentin-specific proteins. These studies report that the cell line MO6-G3 synthesizes phosphophoryn (DPP), dentin sialoprotein (DSP) and dentin matrix protein-1 (DMP-1) while MDPC-23 synthesizes DSP, but not DMP-1. The objective of the present study was to determine whether polyclonal antibodies to rat DSP and DPP would label odontoblasts on microscopic sections of day-19 fetal mouse incisor odontoblasts as well as cultured cells of the MDPC-23 cell line. The spontaneously immortalized MDPC-23 cell line was derived from fetal mouse molar papillae, made continuous by the 3T6 method and cloned by dilution. These cultures have been passaged 77 times after cloning, form multilayered nodules, and have high alkaline phosphatase activity. The data show positive reactivity in odontoblasts in 19-d mouse fetal incisors as well as in cultures of MDPC-23 cells by fluorescence and confocal microscopy. In addition, these cultures were characterized by phase microscopy and scanning and transmission electron microscopy. These findings suggest that MDPC-23 cells are of the odontoblast lineage.

Uptake and reversibility of uptake of nickel by human macrophages.

The use of nickel-containing alloys in dentistry has been questioned because of the biological liabilities of nickel and the release of nickel ions from dental appliances into the oral cavity. The uptake of nickel by cells in the oral tissues is a critical factor in assessing the biological liabilities of nickel. Nickel uptake by macrophages may be particularly important because of the central role of macrophages in the inflammatory process and the known role of the macrophage in orchestrating the response to biomaterials. The aims of the current study were to assess the reversibility of the uptake of nickel from human macrophages and determine the portion of the nickel which reaches the nuclei as a function of time. Cellular nickel content was measured by means of atomic absorption spectrometry. Nuclear nickel content was assessed after fractionating cells. The results showed that nickel was rapidly taken up by macrophages and that the nickel accumulated in the nucleus in as little as 8 h. After 48 h, over 60% of the cellular nickel was in the nucleus. Once taken up, the nickel was lost at a lower rate. The rate of loss decreased as the initial exposure time to the nickel increased. Thus, the results indicated that macrophages may accumulate nickel if the time between exposures is insufficient to reverse the uptake. Further studies are necessary to correlate the retention of nickel with impaired function of macrophages and to further define the biological risks of using nickel in dental alloys.

Cytotoxicity of urethane dimethacrylate composites before and after aging and leaching.

The in vitro cytotoxicity of urethane dimethacrylate composites cured at different times by visible light and after different aging times and extraction treatments was evaluated using succinic dehydrogenase activity in the mitochondria of a fibroblastic cell line to reflect cell viability. In addition, extractable chemicals associated with cell response were identified. The composite samples were tested untreated, polished, or extracted with water or 75% ethanol-water. Balb/c 3T3 fibroblasts were used as the cell culture system while MTT-formazan production was used as the toxicity parameter. Cell viability was calculated as a percentage of Teflon controls. Identification of the chemicals was measured by extracting the composites with 75% ethanol-water, separating the extract by HPLC, and identifying the fractions with mass spectroscopy. In general, cell viability increased continuously with curing time for differently treated samples at high aging times (288 h) while it decreased when the composites were not aged (0 h). In addition, for 75% ethanol or water-extracted composites, cell viability increased within the first 24 h of aging and reached a plateau after 72 h. Lowest cytotoxicity occurred when the samples were extracted with the 75% ethanol solution. The highest cytotoxic effects were found when the samples were untreated. Slightly reduced cytotoxic effects were seen with polished composites. The results suggest that curing the light-activated composites for a minimum of 150 s and post-curing for 24 h is required to attain comparable biocompatibility with the Teflon control. Removing the oxygen-inhibited layer from these composites decreased the cytotoxicity by 33% while extracting the composites with 75% ethanol-water decreased it by 77%. Chemicals released from the surface accounted for approximately 40% of cellular response while about 60% of the response was due to chemical components released from the bulk. The primary leachable component from the composites was UDMA monomer. Small quantities of 1,6 hexane diol methacrylate, camphoroquinone, and 2,4,6-tritertiarybutyl phenol also were found.

Dentin sialoprotein (DSP) transcripts: developmentally-sustained expression in odontoblasts and transient expression in pre-ameloblasts.

Dentin sialoprotein (DSP), a 53 kDa glycoprotein, is believed to be present exclusively in dentin. Using rat and mouse digoxigenin labeled (DIG)-DSP and 35S-DSP riboprobes, and in situ hybridization techniques, we have studied the presence of DSP mRNA at specific developmental stages of dentinogenesis. In mouse and rat molars and incisors, DSP transcripts were localized in young odontoblasts associated with early stages of predentin formation, as well as in mature odontoblasts, cells with cytoplasmic extensions embedded in the forming dentin. No DSP transcripts were detected in dental pulp, enamel organ, ameloblasts, epithelial root sheath, Meckel's cartilage, alveolar bone or tibia. Furthermore, no DSP mRNA was observed in other soft tissues including heart, lung, kidney, intestine, eye, and muscle. In addition to the intense and prolonged expression by odontoblasts, DSP mRNA was transiently expressed by pre-ameloblasts in both developing molars and incisors. These observations are consistent with the results of previous immunohistochemical studies (1). The transient expression of DSP in pre-ameloblasts across from young odontoblasts suggests an involvement of DSP in epithelial-mesenchymal interactions that are crucial to later stages of tooth development.

Effect of Ni ions on expression of intercellular adhesion molecule 1 by endothelial cells.

Previous studies have shown that Ni-based alloys implanted into soft tissues cause an infiltration of inflammatory cells around the implant. This phenomenon is potentially important to dental alloys which are adjacent to oral tissues. To help define the mechanisms by which Ni causes an infiltration of inflammatory cells, we exposed endothelial cells in vitro to Ni ions and measured the expression of intercellular adhesion molecule 1 (ICAM-1). ICAM-1 is known to be involved in the recruitment of inflammatory cells from the bloodstream. We also exposed macrophages to Ni ions to test the hypothesis that Ni might alter cytokine secretion and subsequently cause expression of ICAM-1 on endothelial cells. The results showed that Ni ions could promote the expression of ICAM-1 on endothelial cells, but only at concentrations which were high enough (850 mumol/L for 24 h) to suppress cell metabolic activity. Although we had previous evidence that Ni could cause macrophages to secrete cytokines such as interleukin 1 beta, Ni-exposed macrophage supernatants did not induce expression of ICAM-1 on endothelial cells at concentrations subtoxic to the macrophages (85 mumol/L). At subtoxic concentrations, Ni ions were able to suppress ICAM-1 expression on endothelial cells which were stimulated with lipopolysaccharide. Thus, Ni ions either promoted or suppressed the expression of ICAM-1 depending on their concentration. This dual action of Ni ions may be important in vivo where a gradient of concentrations of released ions is likely to exist around the implanted biomaterial. Further studies are necessary to determine the effect of time of exposure and the molecular mechanisms of increased ICAM-1 expression.

Effects of metal ions on osteoblast-like cell metabolism and differentiation.

The objective of this study was to evaluate the effects of metal ions, which may be released from orthopedic or dental implants, on osteoblast metabolism and differentiation. ROS 17/2.8 cells were cultured in F-12 medium for 7 days. Then Al+3, Co+2, Cr+3, Ni+2, Ti+4, and V+3 were added at concentrations less than their cytotoxic concentrations. After 3 days, DNA synthesis, succinate dehydrogenase activity, alkaline phosphatase (ALP) activity, and culture calcification were assessed. Northern blots were performed for ALP, osteocalcin (OCN), and osteopontin (OPN) mRNA transcription. The data indicated that Cr+3 and A1+3 had few inhibitory effects on ROS cell metabolism below their cytotoxic concentrations, Ni+2, Co+2, Ti+4, and V+3 affected all these parameters of ROS cell metabolism at concentrations below cytotoxic levels. For RNA analysis, A1+3 significantly suppressed the expression of ALP, OCN, and OPN at both cytotoxic and noncytoxic concentrations. Co+2 specifically suppressed ALP expression at cytotoxic concentrations. Cr+3 and Ni+2 inhibited OCN, OPN, and ALP gene expression only at cytotoxic concentrations. For Ti+4 and V+3 ions, gene expression at cytotoxic levels was not significantly affected as compared with the effects at noncytotoxic level. These results show that metal ions may alter osteoblast behavior even at subtoxic concentrations, but do not always affect the expression of all genes similarly.

In vitro IL-1 beta and TNF-alpha release from THP-1 monocytes in response to metal ions.

OBJECTIVES: Previous studies have shown that biomaterials can activate macrophages to produce cytokines and promote an inflammatory response. Although the toxicity of many metal ions has been extensively investigated, little is known about the ability of these ions to alter cytokine release from macrophages. Yet the release of these ions from biomaterials has been well documented. Previous studies indicated that alterations in cytokine release might be expected because metal ions alter protein production in macrophages at sub-toxic concentrations. Thus, the hypothesis of this study was that metal ions can alter the secretion of cytokines from macrophages at sub-toxic concentrations. METHODS: The release of interleukin-1 beta (IL-1 beta) and tumor necrosis factor-alpha (TNF-alpha) from macrophages was investigated when the macrophages were exposed to metal ions, with or without lipopolysaccharide (LPS), a component of dental plaque. Human THP-1 macrophages were exposed to ions of Ag, Au, Cu, Hg, and Ni for 24 h. In half of the cultures, LPS was added for the last 4 h. The release of IL-1 beta and TNF-alpha into the medium was measured using enzyme-linked immunosorbent assays. ANOVA and Tukey multiple comparison intervals were used to compare the various experimental conditions. RESULTS: None of the metal ions elevated the IL-1 beta or TNF-alpha levels after 24 h, but Ni ions significantly elevated the IL-1 beta and TNF-alpha levels after 72 h. With LPS added, Ag, Cu, and Ni significantly amplified the LPS-induced production of IL-1 beta but only Ni amplified the TNF-alpha response. These alterations in cytokine response occurred with metal ion concentrations which have been previously shown to be released from dental alloys in vitro and in vivo. SIGNIFICANCE: It appeared plausible that macrophage-cytokine mediated inflammatory responses may be altered by the presence of some metal ions in tissues, particularly Ni.

In vitro cytotoxicity and dentin permeability of HEMA.

An in vitro diffusion chamber was used to measure the diffusion of 2-hydroxyethyl methacrylate (HEMA) through etched human dentin disks. Concentrations of HEMA, which diffused through dentin, were measured by ultraviolet spectroscopy, and the effect of initial HEMA concentration, dentin thickness, and back pressure on diffusion were assessed. The cytotoxicity of HEMA was determined using BALB/c 3T3 mouse fibroblasts in direct contact with HEMA for 12 or 24 h. HEMA diffused rapidly through dentin under all conditions, but increased thickness, back pressure, or decreased initial concentration all reduced diffusion. The permeability coefficient of HEMA was approximately 0.0003 cm/min, and diffusion through 0.5 mm of dentin reduced the HEMA concentration by a factor of approximately 6,000 (with 10 cm of H2O back pressure). It was concluded that the risk of acute cytotoxicity to HEMA through dentin was probably low, but that decreased dentin thickness, lack of polymerization, or extended exposure times might increase the risk significantly.

Biological effects of palladium and risk of using palladium in dental casting alloys.

In dentistry, palladium is a very common component of dental casting alloys of all types, and its use has increased over the past several decades in response to the increased cost of gold. However, there have been recent controversies, particularly in Germany, over possible adverse biological effects of using palladium in dental alloys. Therefore, the purpose of this paper is to review the known biological effects of palladium and the likelihood that these effects can be caused by dental alloys which contain palladium. In an ionic form and at sufficiently high concentrations, palladium has toxic and allergic effects on biological systems. Palladium allergy almost always occurs in individuals who are sensitive to nickel. The carcinogenic potential of the palladium ion is still unclear, although there is some evidence that it is capable of acting as a mutagen. However, there are no well documented cases of adverse biological reactions to palladium in the metallic state. Furthermore, in spite of the potential adverse biological effects of palladium ions, the risk of using palladium in dental casting alloys appears to be extremely low because of the low dissolution rate of palladium ions from these alloys.

In vitro models of biocompatibility: a review.

The objectives of this paper were to define in vitro biocompatibility of materials, to discuss some of the issues concerning why conclusions from tissue culture are sometimes different from in vivo biocompatibility, to give highlights of the sequence of the development of these in vitro assays from the early 1950s to their present state of development, and to discuss possible future trends for in vitro testing. In vitro biocompatibility tests were developed to simulate and predict biological reactions to materials when placed into or on tissues in the body. Traditional assays have measured cytotoxicity by means of either an end-stage event, (i.e., permeability of cytoplasmic membranes of dead and dying cells, or some metabolic parameter such as cell division or an enzymatic reaction). In vitro assays for initiation of inflammatory and immune reactions to materials have also begun to appear in the literature. More recently, the concept of dentin barrier tests has been introduced for dental restorative materials. Four models which measure both permeability and biological effects of materials are compared and discussed. Future efforts may be directed toward development of materials which will allow or promote function and differentiation of tissues associated with materials. New analytical procedures and understanding of optimal characteristics of materials should improve our ability to develop more biocompatible materials. Both molecular biology techniques, and altered design of material surfaces may make the materials either more or less reactive to the biological milieu. These trends suggest a greater future role of the biological sciences in the development of biomaterials.

Cytotoxic interactive effects of dentin bonding components on mouse fibroblasts.

Previous studies have shown a wide range of pulpal reactions to dentin bonding systems and a poor correlation between in vitro and in vivo toxicity of dentin bonding agents. Because dentin bonding agents are composed of multiple components which may diffuse through dentin, we hypothesized that these components may cause cytotoxicity through interactive (synergistic) effects. We investigated the cytotoxicities of four dentin bonding components--HEMA, Bis-GMA, TEGDMA, and UDMA--and interactive effects for three binary combinations of the dentin bonding components--HEMA and Bis-GMA, Bis-GMA and TEGDMA, and TEGDMA and UDMA. Cytotoxicities to Balb/c 3T3 mouse fibroblasts were measured by the MTT assay. Concentrations which caused 50% toxicity compared with controls (TC50 values) were compared, and the interactive effects were determined by evaluation of the differences between observed and expected MTT activities of the cells. The ranks of toxicity of the dentin bonding components in terms of TC50 values were as follows: Bis-GMA > UDMA > TEGDMA >>> HEMA (least toxic) after 24- and 72-hour exposures. As binary combinations, the three combinations of dentin bonding components interacted in three ways--synergism, additivism, and antagonism--which were influenced by the concentrations of both components. The longer period of exposure resulted in a significant increase in the cytotoxicity of the dentin bonding components and combinations. The findings indicate that both exposure time and the interactions between the dentin bonding components may be important parameters in determining the cytotoxicity of dentin bonding agents in vivo.