Bisgma Stimulates Prostaglandin E2 Production in Macrophages via Cyclooxygenase-2, Cytosolic Phospholipase A2, and Mitogen-Activated Protein Kinases Family

Background Bisphenol A-glycidyl-methacrylate (BisGMA) employs as a monomer in dental resins. The leakage of BisGMA from composite resins into the peripheral environment can result in inflammation via macrophage activation. Prostaglandin E2 (PGE2) is a key regulator of immunopathology in inflammatory reactions. Little is known about the mechanisms of BisGMA-induced PGE2 expression in macrophage. The aim of this study was to evaluate the signal transduction pathways of BisGMA-induced PGE2 production in murine RAW264.7 macrophages. Methodology/Principal Findings Herein, we demonstrate that BisGMA can exhibit cytotoxicity to RAW264.7 macrophages in a dose- and time-dependent manner (p<0.05). In addition, PGE2 production, COX-2 expression, and cPLA2 phosphorylation were induced by BisGMA on RAW264.7 macrophages in a dose- and time-dependent manner (p<0.05). Moreover, BisGMA could induce the phosphorylation of ERK1/2 pathway (MEK1/2, ERK1/2, and Elk), p38 pathway (MEK3/6, p38, and MAPKAPK2), and JNK pathway (MEK4, JNK, and c-Jun) in a dose- and time-dependent manner (p<0.05). Pretreatment with AACOCF3, U0126, SB203580, and SP600125 significantly diminished the phosphorylation of cPLA2, ERK1/2, p38, and JNK stimulated by BisGMA, respectively (p<0.05). BisGMA-induced cytotoxicity, cPLA2 phosphorylation, PGE2 generation, and caspases activation were reduced by AACOCF3, U0126, SB203580, and SP600125, respectively (p<0.05). Conclusions These results suggest that BisGMA induced-PGE2 production may be via COX-2 expression, cPLA2 phosphorylation, and the phosphorylation of MAPK family. Cytotoxicity mediated by BisGMA may be due to caspases activation through the phosphorylation of cPLA2 and MAPKs family.


Introduction
Bisphenol A-glycidyl-methacrylate (BisGMA) is synthesized from diglycidyl ether and methacrylic acid of bisphenol-A type epoxy resin [1]. The most commonly composite resins are composed of BisGMA monomers or its derivatives. BisGMAbased resins are used to restore hard tissue, such as teeth and bone. The advantages of BisGMA-based resins include higher modulus, less shrinkage, and lower diffusivity [2]. The commercial composite resins could release BisGMA into peripheral environment. BisGMA, incubated with water-or organic-based medium for 1 to 180 days, was leachable at a concentration range about 10 23 to 10 21 or 10 21 to 10 mM, respectively [3]. Yap et al. have purposed that the leachable BisGMA monomers may result in tissue damage through inflammatory reactions [4]. The activation of innate immune cells, especially macrophages, play a key regulator leading to inflammation [5]. Recently, we have demonstrated that BisGMA could induce cytotoxicity and genotoxicity in macrophages [6]. BisGMA could induce macrophage activation, such as the expression of surface antigens and the generation of proinflammatory mediators, including tumor necrosis factor (TNF)-a, interleukin (IL)-1b, IL-6 nitric oxide, and reactive oxygen species via the phosphorylation of PI3K/Akt, the degradation of IkB, and the activation of NFkB [7,8].
Previously, BisGMA was found to generate PGE2 via COX2 protein expression in human pulp cells [14]. However, the role of cPLA2 activation on BisGMA-induced PGE2 generation and cytotoxicity in macrophage still remains to be elucidated. In this study, the effects of BisGMA on murine macrophage RAW264.7 cells were determined through measuring the production of PGE2 by enzyme-linked immunosorbent assay (ELISA) and cytotoxicity. Western blot was used to evaluate COX-2 expression, the phosphorylation of cPLA2, and the phosphorylation of MAPKs family to clarify the signal transduction pathways.

Cell Culture
Murine macrophage cell line, RAW264.7, was obtained from Bioresource Collection and Research Center (BCRC 60001; Hsinchu, Taiwan). Cells were cultured in DMEM containing 10% FBS, 100 mg/ml streptomycin, and 100 U/ml penicillin. RAW 264.7 cells were maintained at sub-confluence in a 95% air and 5% CO 2 humidified atmosphere at 37uC. To investigate the effects of BisGMA on RAW264.7 macrophages, cells were seeded on the plates and cultured for 24 h. After cell attachment, the medium were replaced with serum-free medium. Cells pretreated with or without 1 mM ERK1/2 inhibitor U0126, 30 mM p38 inhibitor SB203580, 30 mM JNK inhibitor SP600125, and 30 mM cPLA2 inhibitor AACOCF3 for 30 min. And then, RAW264.7 Figure 1. Cytotoxicity and effects of BisGMA on the morphology of RAW264.7 macrophages. (A) Cells were exposed to 3 mM BisGMA for the indicated treatment periods. (B) Cells were exposed to the indicated concentrations of BisGMA for 2 h. The morphology exchange were observed with an inverted microscope (Original magnification6200). (C) Cells were exposed to 3 mM BisGMA for the indicated treatment periods measured by LDH assay. (D) Cells were exposed to the indicated concentrations of BisGMA for 2 h. Values are expressed as mean 6 SD (n = 5). *represents significant difference from control values with p,0.05.

Cytotoxicity Assay
Cytotoxicity was measured by the release of lactate dehydrogenase (LDH). Briefly, 1610 5 cells were treated with various concentrations (0, 0.1, 0.3, 1, and 3 mM) of BisGMA for the indicated treatment period. Then, 1% Triton-X 100 was added to another well and incubated for 45 min. 50 ml supernatants of each well was transferred to new 96-well plate and added 50 ml of reconstituted substrate mix from CytoToxH 96 nonradioactive assay kit (Promega, Sunnyvale, CA, USA). After 30 min, the stop solution was added to each well. Absorbance was recorded at 490 nm by using a microplate reader (Dynatech MR 4000; Dynatech, Boston, MA, USA). Cell viability was calculated according to our perious study [6].

Measurement of PGE2
The protein concentrations of PGE2 were measured by using ELISA [15]. Briefly, 5610 4 cells were incubated with different concentrations (0, 0.1, 0.3, 1, and 3 mM) of BisGMA for 2 h. The PGE2 levels in culture medium were measured by ELISA kit (eBiosciences, Diego, CA, USA) according to the manufacturer's instructions.

Western Blotting
To investigate the expression of COX2, phosphorylation and non-phosphorylation types of cPLA2, cells were treated with BisGMA for 2 h. To evaluate the phosphorylation and nonphosphorlation types of MEK1/2, ERK1/2, Elk, MEK3/6, p38, MAPKAPK2, MEK4, JNK, and cJUN, cells were treated with  BisGMA for 15 min. After treatment, 50 mg of cell lysate protein, was separated by SDS-PAGE, and electrophoretically transferred to polyvinylidene difluoride membrane. The membranes were blocked with 5% (w/v) nonfat dried milk for 1 h at room temperature to reduce nonspecific binding, washed with PBS  containing 0.1% Tween-20 (PBST), then probed with antibodies including COX-2, phosphorylation and non-phosphorylation of cPLA2, MEK1/2, ERK1/2, Elk, MEK3/6, p38, MAPKAPK2, MEK4, JNK, and cJUN. After membranes were washed again with PBST, a 1:10,000 (v/v) dilution of horseradish peroxidaselabeled IgG was added at room temperature for 1 h, and the blots were developed using enhanced chemiluminescence reagents. The phosphorylation and expression of proteins were quantified with a LAS-4000 mini luminescent image analyzer (GE Healthcare Life Sciences, Tokyo, Japan) [16].

Fluorometric Assay for Caspases Activities
Caspases activities were analyzed with a caspases fluorometric assay kits (Enzo Life Sciences, Plymouth Meeting, PA, USA) according to our recent study [6]. Briefly, cell lysates from each sample was mixed with a reaction buffer containing the fluorogenic substrates of caspase-3, -8, and -9, which are DEVD-AFC, IETD-AFC, and LEHD-AFC. The data were collected by using a fluorescence microplate reader (Molecular Devices, CA, USA) at excitation/emission wavelengths of 400/ 505 nm.

Statistical Analysis
At least three independent experiments were performed as indicated in the figure legends. All data were expressed as mean 6 standard deviation (SD). Statistical analysis was performed by using one-way analysis of variance (ANOVA) with Bonferroniadjusted post hoc tests for multigroup comparisons; p,0.05 was considered significantly for each test.

Effects of BisGMA on RAW 264.7 Macrophages
The changes of RAW264.7 macrophage morphology and cell density after exposure to 3 mM BisGMA for the indicated treatment periods and the indicated concentrations of BisGMA for 2 h were shown in Fig. 1A and 1B, respectively. In addition, cytotoxicity of BisGMA was measured by the release of cytoplasmic enzyme LDH. As shown figures 1C and 1D, BisGMA exhibited cytotoxicity to RAW264.7 macrophage in a dosedependent and a time-dependent manner (p,0.05), respectively.

Effects of BisGMA on the Expression of PGE2 and COX-2
The effect of BisGMA on the generation of PGE2 in RAW264.7 macrophages by ELISA assay was shown in figure 2A. The production of PGE2 was demonstrated in a dose-dependent manner (p,0.05). The levels of PGE2 were 0.71, 0.65, 1.02, 1.66, and 2.25 ng/ml at the concentrations of 0, 0.1, 0.3, 1, and 3 mM BisGMA for 2 h, respectively. As shown in figure 2B, BisGMA was demonstrated to increase COX-2 protein expression in a dosedependent manner (p,0.05). The expressions of COX-2 were 1.00, 1.19, 1.62, 1.99, and 2.00 fold at the concentrations of 0, 0.1, 0.3, 1, and 3 mM BisGMA for 2 h, respectively. In addition, using antibodies against phosphorylated and total ERK1/2. (B) Cells were stimulated with the indicated concentrations BisGMA for 15 min. (C) After treatment with 1 mM U0126 for 30 min, cells were stimulated with 3 mM BisGMA for 15 min. Cells were harvested and protein extracts were subjected to SDS-PAGE Western blot analysis using antibodies against the phosphorylated and total MEK1/2, ERK1/2, and Elk. The ratio of immunointensity between the phosphorylation and total protein was calculated. Values are expressed as mean 6 SD (n = 3). *represents significant difference from control values with p,0.05. #represents significant difference from BisGMA values with p,0.05. doi:10.1371/journal.pone.0082942.g005

Discussion
BisGMA is the most common used resin monomer for dental materials such as bonding agents, sealants, cement, bite guards, and dentures. To restore the caries cavity on the teeth bottom, the unpolymerized monomers of BisGMA can leach into oral cavity and prompt pulpal imflammation, cytoxicity, and delay pulpal healing [4,17,18]. The concentrations of BisGMA has been reported to leach into peripheral environment from 0.001 to 10 mM [3], thus we employed the doses of BisGMA from 0.1 to 3 mM in the present study. BisGMA was found to induce cytotoxicity at concentration of 0.3 mM in RAW264.7 macrophages. Similar results were found in human pulp cells [18] and macrophages [6]. Pro-inflammatory mediators, such as nitric oxide, reactive oxygen species, TNF-a, IL-1b, and IL-6, generation from macrophages are induced by lipopolysaccharide, interferon-c, and BisGMA [7,8,19]. The moderate amount of proinflammatory mediators could kill or remove invasive microorganisms. However, over-expression of proinflammatory mediators would led to peripheral tissue damage and self-annihilation [20]. Macrophage activation is a hallmark of local inflammatory response. Previously, the simplified etch-and-rinse adhesive system, which contains BisGMA, could induce cytotoxicity and NO generation in alveolar macrophages [21]. In cell culture system, RAW264.7 macrophages have been used for the experiments of immunological function directly and as well as human U937 macrophages and peripheral boold mononuclear cells [22,23]. Therefore, we chose RAW264.7 macrophages to evaluate the effects cytotoxicitiy, generation of PGE2, activation of MAPKs and cPLA2 in this study.
PGE2 plays an important role in acute and chronic inflammation, infection, pain, fever, and cancer [9,24]. Macrophages activated by physiological or pathological stimuli could release PGE2 which increases vascular permeability, infiltration of neutrophils, macrophages, and mast cells into secretion sites [9]. COX is the rate-limiting enzyme that catalyzes the conversion of AA to PGE2. COX-2 is an inducible enzyme predominantly expressed in inflamed tissues [25]. In the present study, BisGMA was demonstrated to significantly evoke PGE2 secretion and COX-2 expression in RAW264.7 macrophages. Similar results were found in human pulp cells by Chang et al. [14]. Nuclear factor (NF)kB is an important transcription factor in the expression of COX-2 [26]. In perivous study, we have demonstrated BisGMA induced NFkB phosphorylation [7]. These results suggest that PGE2 generation was induced by BisGMA stimulated COX-2 expression via NFkB phosphorylation.
cPLA2 catalyzes the first step in hydrolysis of membrane phosphatidylcholine convert into AA, which is the precursor of proinflammatory lipid mediators such as PGs, thromboxane, and leukotrienes [27]. AA has been shown to induce chemotaxis, reactive oxygen species generation, and apototsis in neutrophils, epithelial cells, and macrophages [10,28]. In addition, cPLA2 and AA directly associate to COX-2-derived PGE2 synthesis correlation with inflammation [29]. To the best of our knowledge, we first found the phosphorylation of cPLA2 induced by BisGMA in RAW264.7 macrophages. The phosphorylation of cPLA2 and secretion of PGE2 by BisGMA exhibited apparently in a parallel concentration-dependent manner. Pretreatment with cPLA2 inhibitor AACOCF3 suppressed cPLA2 phosphorylation, COX-2 expression, PGE2 generation, and cytotoxicity in BisGMAstimulated RAW264.7 macrophages. These results indicated that the secretion of PGE2 and cytotoxicity induced by BisGMA may be via the phosphorylation of cPLA2.
Several studies have demonstrated the activation and phosphorylation of cPLA2 regulated by MAPKs in macrophage during stimulation [12,30]. In addition, MAPKs play an important role in the stimulation of pro-inflammatory genes and mediators in macrophages [31]. MAPKs pathways are organized by threetiered hierarchical kinases module. MAPKs are activated by MEKs, which is dual phosphorylation of threonine and tyrosine residues in the Thr-X-Tyr motif. MEK is activated by MEKs kinases (MEKKs) through serine and threonine phosphorylation [32]. ERK1/2 is activated and phosphorylated by MEK1/2. MEK1/2 is in turn regulated by Raf. The substrates of ERK1/2 are Elk and p90 RSK . p38 is activated and phosphorylated by MEK3/6. MEK3/6 is in turn regulated by MEKK1/2/3/4. The substrate of p38 is MAPKAPK2 and ATF2. JNK is activated and phosphorylated by MEK4/7. MEK4/7 is in turn regulated by MEKK1/2/3/4. The substrate of JNK is cJUN [33]. In human pulp cells, BisGMA was reported to induce the phosphorylation of MEK1/2-ERK1/2 signaling and the production of PGE2 [14]. In the present study, we first purposed that the phosphorylation of MEK1/2, ERK1/2, and Elk was triggered by BisGMA in RAW264.7 macrophages. Pretreatment with U0126 inhibited BisGMA-induced phosphorylation of cPLA2, generation of PGE2, and cytotoxicicty. In addition, we also first found BisGMAinduced the phosphorylation of p38 pathway, including MEK3/6, p38, and MAPKAPK2, and JNK pathway, including MEK4, JNK, and cJUN. The MAPKs inhibitors U0126, SB203580, and SP600125 also significantly inhibited BisGMA-induced the phosphorylation of cPLA2 and generation of PGE2. Taken together, these results indicated that cPLA2 could participate in BisGMA-induced PGE2 generation in RAW 264.7 macrophages via MAPKs signal transduction pathways.
Recently, we demonstrated that BisGMA-induced cytotoxicity and apoptosis may be via caspases 3, 8, and 9 activation in RAW264.7 macrophages [6]. Previously, studies have shown that BisGMA-induced cytotoxicity may be via the generation of reactive oxygen species and the expression of hemeoxygenase-1 and carboxylesterase could against BisGMA-induced cytotoxicity in human pulp cells [18,34]. cPLA2 has demonstrated to participate caspases activation and result in macrophage apoptosis [10]. Pretreatment with AACOCF3, U0126, SB203580, and SP600125 could reduce BisGMA-induced cytotoxicity. Furthermore, AACOCF3, U0126, SB203580, and SP600125 could suppress BisGMA-induced caspase 3, 8, and 9 activation. These results indicated that cPLA2 may participate in BisGMA-induced cytotoxicity and caspases activation in RAW 264.7 macrophages via MAPKs signal transduction pathways.