The Effect of a Brominated Flame Retardant, Tetrabromobisphenol-A, on Free Radical Formation in Human Neutrophil Granulocytes: The Involveme
http://www.100md.com
《毒物学科学杂志》
ABSTRACT
This study investigates the effects of one of the most frequently used brominated flame-retardants (BFR), tetrabromobisphenol-A (TBBPA), on formation of reactive oxygen species (ROS) and calcium levels in human neutrophil granulocytes. TBBPA enhanced ROS production in a concentration-depended manner (1–12 μM), measured as 2,7-dichlorofluorescein diacetate amplified (DCF) fluorescence. The results on ROS production by TBBPA was confirmed by lucigenin-amplified chemiluminescence. The TBBPA induced formation of ROS was due to activation of respiratory burst, as shown by the NADPH oxidase inhibitor DPI (10 μM). TBBPA induced activation of respiratory burst was also inhibited by the MEK 1/2 inhibitor U0126 (10 μM), the PKC inhibitor BIM (0.25 μM), and the tyrosine kinase inhibitor erbstatin-A (25 μM). We also found a small reduction in ROS formation in the absence of extracellular calcium and when verapamil was added. The phosphorylation of ERK 1/2 was confirmed by Western blotting. TBBPA also induced a concentration dependent increase in intracellular free calcium measured with Fura-2/AM. We suggest that exposure of human neutrophil granulocytes to the brominated flame retardant TBBPA leads to an activation of the NADPH oxidase primarily by an ERK 1/2 stimulated pathway. The data also show that PKC, calcium, and tyrosine kinases may be involved in the activation
Key Words: brominated flame-retardants (BFR); tetrabromobisphenol-A; neutrophil granulocytes; reactive oxygen species (ROS); MAP kinase pathway; calcium; extracellular signal-regulated kinase (ERK).
INTRODUCTION
Brominated flame-retardants (BFRs) are a large group of compounds widely used to protect various products, such as plastics, textiles, and electronic equipment from catching fire. Several of the BFRs are persistent and lipophilic compounds. They may bioaccumulate and are thus regarded as a potential environmental health problem (de Wit, 2002). Within this group we find compounds such as polybrominated diphenyl ethers (PBDE), tetrabromobisphenol-A (TBBPA), hexabromocyclododecane (HBCD), and polybrominated biphenyls (PBBs).
The phenolic TBBPA is industrially the most important individual BFR used with an annual demand of approximately 120,000 metric ton (de Wit, 2002). TBBPA is a phenolic compound primarily used as a chemically bound flame retardant, which is supposed to limit its spread in the environment and reduce its accumulative properties. However, studies have shown that this compound may leak from treated products (Sellstrom and Jansson, 1995) and several recent reports on TBBPA in human and wildlife samples have shown the presence of this compound. Human TBBPA serum levels were measured by Thomsen et al. (2001), who found TBBPA in blood serum from electronic dismantlers, circuit board producers, and laboratory personnel. TBBPA has also been reported to be a contaminant in sediments and mussels (Saint-Louis and Pelletier, 2004; Watanabe et al., 1983) and in eggs from predatory bird species (Berger et al., 2004).
The majority of studies conducted to date on BFRs, both with regards to toxicology and environmental levels, have focused on the PBDEs (de Wit, 2002). There is little knowledge about the toxicity of TBBPA, which is of concern considering its extensive use and presence in the environment. A few toxicological studies have been carried out and the TBBPA can elicit thyroidogenic and estrogenic-like activity in vitro (Darnerud, 2003; Kitamura et al., 2002; Meerts et al., 2000), and has a neurotoxic potential (Mariussen and Fonnum, 2003; Reistad et al., 2002). A recent work by Fukuda et al. (2004) showed that TBBPA induces polycystic lesions in the kidney of exposed newborn rats. Pullen et al. (2003) showed that TBBPA suppresses the induction of interleukin-2 in murine splenocytes in vitro, indicating an immunotoxic potential.
Human neutrophil granulocytes play a key role in host defenses against invading pathogens and are major effectors of the acute inflammatory reactions. In response to a variety of agents, neutrophils release large quantities of superoxide anion () in a phenomenon known as respiratory burst. Neutrophil production of is dependent on the NADPH oxidase, a multicomponent membrane-bound enzyme that catalyzes NADPH-dependent reduction of oxygen to (Babior, 1999). Superoxide is rapidly converted to hydrogen peroxide (H2O2), either spontaneously or enzymatically by superoxide dismutase. H2O2 is then reduced to water by catalase or converted to hypochlorous acid (HOCl) by myeloperoxidase (MPO). H2O2 can also be converted to hydroxyl radicals in the presence of transition metal ions. Inappropriate activation of respiratory burst is associated with tissue injury and impairment of the ability to clear invading microorganisms (Labro, 2000).
Previously a correlation has been found between wildlife animals' exposure to environmental contaminants, such as polychlorinated biphenyls (PCB) and methyl mercury, and effects on immune parameters. Some of these findings have been attributed to activation of neutrophil granulocytes in vitro (Duffy et al., 2002; Sweet and Zelikoff, 2001; Voie et al., 1998). Because of the high production volume of TBBPA, its presence in biotic samples and the close resemblance to other environmental contaminants, we have examined its effect on human neutrophil granulocytes.
MATERIALS AND METHODS
Chemicals. Tetrabromobisphenol-A (TBBPA, BA-59P, Great Lakes, Lot nr 6L16,C) was all obtained from Promochem (Stockholm, Sweden). Stock solutions were prepared by dissolving the compounds in DMSO. The final DMSO concentration in the samples was always less than 1%. Bisphenol-A, bisindolylmaleimide (BIM), bromphenol blue, ponceau S concentrate, bis-N-Methylacridinium-nitrate (Lucigenin), 2,7-dichlorofluorescein diacetate (DCFH-DA), dimethyl sulfoxide (DMSO), 2-merchaptoethanol, diphenyleneiodonium (DPI), diethyldithio-carbamic acid (DDC), EGTA, verapamil hydrochloride, cyclosporine A (CSA), methanol, phosphate-buffered saline (PBS), phorbol 12-myristate 13-acetate (PMA), SB203580, superoxide dismutase (SOD), and sodium dodecyl sulphate (SDS) were all from Sigma-Aldrich (St. Louis, MO). U0126 was obtained from Promega Corporation (Madison, WI). Hanks Balanced Salt Solution (HBSS) and HEPES buffer were purchased from GibcoBRL (U.K.). Lymphoprep was purchased from Nycomed Pharma (Oslo, Norway). Enhanced chemiluminescence (ECL) reagent was from Amersham Pharmacia Biotech AB (Uppsala, Sweden). Monoclonal mouse anti-phospho-ERK antibody (Tyr204) and polyclonal rabbit anti-ERK2 antibody were from Santa Cruz Biotechnology (Santa Cruz, CA). Horseradish peroxidase (HRP)-conjugated rabbit-anti-mouse antibody and HRP-conjugated goat-anti-rabbit antibody were purchased from DAKO A/S (Glostrup, Denmark). 2,5-Dihydroxymethylcinnamate (Erbstatin-A), FK-506 (Tacrolimus), and Fura-2/AM were from Calbiochem Novabiochem Corp. (San Diego, CA). All other reagents used were analysis grade laboratory chemicals from standard commercial suppliers.
Isolation of human neutrophil granulocytes. Human venous blood was obtained from nonsmoking healthy adult male volunteers in the morning. Neutrophil granulocytes were separated from EDTA blood by dextran sedimentation followed by a standard density-gradient centrifugation as previously described (Boyum et al., 1991). In brief, EDTA blood from individual donors (30 ml) were mixed with 3 ml 6% dextran and left for sedimentation at room temperature for 30 min. The supernatant, containing the granulocytes, was subject to Lymphoprep density gradient centrifugation at 600 x g for 15 min. The pellet was washed in 0.9% NaCl and then resuspended in 7 ml 0.83% NH4Cl in 7 min for lysis of the erythrocytes, and then centrifuged for 7 min (600 x g). This was repeated if not proper lysis was obtained. Cells were then resuspended in HBSS and the number of granulocytes was determined in an AVIDA 60 hematology system. The cells were kept on ice (approximately 4°C) until use.
Lactate dehydrogenase (LDH) assay. Leakage of LDH was assessed as an index of cell injury (Koh and Choi, 1987). The measurements were performed as described elsewhere (Ring and Tanso, manuscript in preparation). In brief, cells (2 x 106/ml) were exposed to BRF for the indicated times (5 or 30 min). Cells were then spun down and supernatant from each sample was transferred to sample tubes and stored at 4°C until measured (usually within 2 h). LDH measurements were performed by transfer of 100 μl aliquots of the supernatant to the wells of a custom made 48 well microplate with glass bottom and the volume was adjusted to 425 μl with 0.1 M KPO2 buffer (pH 7.5). The reactions were started by automated injection of 50 μl of an 846 μM stock solution of NADH (final concentration 84.6 μM) followed by automated injection of 25 μl of a 13.6 mM stock solution of pyruvate (final concentration 0.68 mM). The LDH activity was measured, using a BMG FLUOstar Optima fluorimeter, from the decay rate of NADH fluorescence for 30 min at 28°C. The LDH activity was calculated off line and is given as the rate constant of the decrease in fluorescence emission at 460 nm (excitation wavelength 340 nm). The LDH activity (fluorescence units/s) is not a direct measure of the number of dead cells but it gives a qualitative measure of the relative amount of cell necrosis. 100% cell death was estimated by administration of 0.01% triton and corresponded to a NADH fluorescence decay rate of approximately 95 units/s (control values, 11 units/s). In Table 1 the values are shown as % of triton ± SEM.
Assay for measuring reactive oxygen species by DCF-fluorescence. Formation of ROS was measured with use of the fluorescent probe DCFH-DA. The method is based on the incubation of the granulocytes with DCFH-DA, which diffuses passively through the cellular membrane. Intracellular esterase activity results in the formation of DCFH, which emits fluorescence when oxidized to 2', 7'-dichlorofluorescein (DCF). The fluorescence emitted by DCF reflects the oxidative status of the cell and was determined essentially as described previously (Myhre et al., 2000). Briefly, the cells (final concentration 2 x 106/ml suspension) were incubated with DCFH-DA (5 μM) in HEPES-buffered (20 mM) HBSS (CaCl2 1.26 mM, KCL 5.37 mM, KH2PO4 0.44 mM; MgCl2 0.49 mM, MgSO4 0.41 mM, NaCl 140 mM, NaHCO3 4.17 mM, Na2HPO4 0.34 mM) with glucose (5 mM) at 37°C for 15 min. Following centrifugation, the extracellular buffer with DCFH-DA was exchanged with fresh buffer and the suspension was mixed gently. The cells (2 x 106/ml, 125 μl) were transferred to 250 μl wells (microtiter plate reader, 96 wells) containing 125 μl buffer with the BFR and/or the different inhibitors. Fluorescence was recorded in a Perkin-Elmer LS50B luminescence spectrometer (excitation wavelength 485 nm, emission wavelength 530 nm) at 37°C for 60 min. Results are calculated as area under the curve (AUC) and presented as values relative to control (% of control). PMA (1 x 107 M) was included as a positive control in all experiments (n = 5–8).
Assay for measuring reactive oxygen species by lucigenin-amplified chemiluminescence. Lucigenin chemiluminescence was used to detect in neutrophil granulocytes. The reaction mixture (250 μl) contained 0.1 mM lucigenin, 2 x 105 cells and different concentrations of the compounds. CL was measured by a Labsystem Luminoskan luminometer at 37°C for 60 min. PMA (1 x 10– 7 M) was included as a positive control in all experiments (n = 5–7). The cells and reagents were prepared in HEPES-buffered (20 mM) HBSS with 5 mM glucose. When calcium free buffer was used, 2 mM EGTA was also added. The reaction was started by adding 100 μl of the cell suspension to each well. Results are calculated as AUC and presented as values relative to control (% of control).
Western blotting. The neutrophils were isolated as described above and incubated at 37°C for 30 min before stimulation. TBBPA was added in different concentrations (6–24 μM) and the cells were incubated for different time-spans (2–20 min). After the indicated incubation period the cells were added 500 μl ice-cold PBS and immediately lysed in sample buffer (final concentration 3% SDS, 5% glycerol, 62.5 mM Tris/HCl pH 6.9, 0.1% bromphenol blue, 6% ?-Mercaptoethanol). Total cell samples were heated for 5 min at 95°C and analyzed on a 3% stacking 12% separating SDS-PAGE gel (2 h at 90 V). The separated proteins were then electrophoretically transferred to nitrocellulose membranes (0.45 μm) overnight (30 mA), and stained with Ponceau S to confirm complete transfer. The nitrocellulose blots were incubated in blocking buffer (Tris-buffered saline containing 0.05% Tween 20 [TBST] and 5% low-fat dry milk) for 1 h and probed with monoclonal anti phospho-ERK 1/2 primary antibody (1:200 dilution in blocking buffer) for 1 h. The blots were washed in TBST (6 x 5 min) and then incubated with peroxidase-conjugated rabbit anti-mouse secondary antibody (1:1000 dilution in blocking buffer) for 1 h. After washing in TBST (6 x 5 min), the blots were developed with Amersham's ECL system according to instructions provided by the producer. The signals were visualized on X-OmatBlue XB-1 film (Kodak). The experiments were repeated at least three times. The membranes were then stripped in 100 mM ?-Mercaptoethanol, 2% SDS, and 62.5 mM Tris/HCl (pH 7.6) for 30 min at 50°C and proceeded again with a rabbit polyclonal anti-ERK-2 primary antibody and peroxidase-conjugated goat anti-rabbit secondary antibody to confirm equal amounts of protein in each well.
Measurement of intracellular free calcium in granulocytes. Intracellular free [Ca2+] was measured by using the fluorescent Ca2+-binding probe fura-2/AM by the method previously described (Grynkiewicz et al., 1985). An increase in Ca2+ concentration is indicated by an increase in the fluorescence excitation ratio (I340/I380). Granulocytes (4.5 x 106 cells/ml) in HBSS containing 20 mM HEPES and 5 mM glucose were incubated at 37°C with 5 μM Fura-2/AM for 20 min. The cells were washed and resuspended in HEPES-buffered HBSS with glucose. Measurements of Fura-2 mediated fluorescence was performed on a computerized Shimadzu RF-5301PC Spectrofluorophotometer, using excitation wavelength ranging between 340 and 380 nm and emission wavelength 510 nm. All data are results of 5–9 separate measurements.
Statistical analyses. Differences between controls and treated groups were evaluated using a two-way Student's t-test (paired, two tail distribution), or by one-way ANOVA followed by Dunnett's 2-sided Post Hoc test. The calculations were performed using SPSS 11.5.
RESULTS
The Effect of BFRs on Human Neutrophil Granulocytes
Relatively low concentrations of the brominated flame retardant TBBPA induced a concentration dependent increase in DCF fluorescence in human neutrophil granulocytes (Fig. 1). TBBPA induced formation of ROS in the granulocytes was confirmed by lucigenin (Fig. 2). DMSO, which was used for dilution of the test compounds, reduced ROS formation in unstimulated cells to 84 ± 6% (mean ± SEM) of the control in the DCF assay and to 97 ± 7% (mean ± SEM) of the control in the lucigenin assay. The neutrophils were also exposed to bisphenol-A, which is a non-brominated analog to TBBPA. Bisphenol-A induced a concentration dependent increase in DCF-fluorescence, but had no effect on the lucigenin assay (Fig. 1).
The cytotoxicity assay shown as LDH-leakage showed a small, but significant solvent (DMSO) effect both at 5 and 30 min of exposure. The test compounds had no effect at the concentrations tested (Table 1).
The Involvement of the NADPH Oxidase, Superoxide Dismutase, Protein Kinase C, Tyrosine Kinase, and Calcium in TBBPA Induced ROS Formation
For mechanistic studies we have used inhibitors of different intracellular signaling pathways. The concentration of the inhibitors is based on what is used in the literature. In each assay used in the mechanistic studies we chose the concentration that gave the highest ROS response which for lucigenin was 4 μM TBBPA and for DCF 12 μM TBBPA.
Incubation of the granulocytes with 4 μM of the NADPH oxidase inhibitor DPI (O'Donnell et al., 1993) reduced TBBPA induced ROS formation by 60% with the DCF assay and gave total inhibition with the lucigenin assay. This indicates that activation of the NADPH oxidase complex is involved in the ROS formation induced by TBBPA (Figs. 2 and 4). The superoxide dismutase inhibitor DDC (Misra, 1979) attenuated the TBBPA induced DCF response almost completely, whereas the lucigenin chemiluminescence on the contrary was increased, demonstrating that TBBPA induce superoxide anion radical formation. Addition of superoxide dismutase (50 U/ml) showed an almost total inhibition of TBBPA induced ROS formation in the lucigenin assay and had no effect in the DCF assay. This shows that lucigenin measure extracellular ROS while DCF measure intracellular ROS. Incubation of the granulocytes with 0.25 μM BIM, a specific PKC inhibitor (Bit et al., 1993), reduced the DCF-fluorescence by 69%, and the lucigenin response by 28%. The TBBPA induced response was also strongly inhibited by 25 μM erbstatin-A, an inhibitor of tyrosine kinases (Kawada et al., 1993). The inhibition was 82% relative to the control in the DCF assay and total inhibition in the lucigenin assay. Calcium free buffer and the voltage-dependent calcium channel blocker verapamil (Hille, 1992) reduced TBBPA induced DCF fluorescence by 76 and 61%, respectively and the TBBPA induced lucigenin chemiluminescence by 69 and 9%, respectively. The mitochondrial transition pore blocker CSA (1 μM), used to indicate release of ROS from mitochondria (Baysal et al., 1991), had no inhibitory effect (data not shown).
DISCUSSION
The results presented in this study demonstrate that the BFR, TBBPA, activates respiratory burst in human neutrophil granulocytes as shown with DCF fluorescence and lucigenin-amplified chemiluminescence. This is an important finding, as TBBPA are used extensively for a large variety of applications and can be detected in wildlife and humans (de Wit, 2002). Bisphenol-A, a non-brominated analog of TBBPA, also induced ROS, but with lower potency than TBBPA, both on molar and weight basis.
Formation of ROS in granulocytes may be induced by different mechanisms, of which activation of the NADPH oxidase is the most important (Fig. 8). The neutrophil NADPH oxidase is a multicomponent membrane-bound enzyme that catalyses NADPH dependent reduction of oxygen to , which may be converted to H2O2, in the presence of peroxidases, OONO–, HOCl, and OH? (Babior, 1999). Superoxide anion () formed by NADPH oxidase activation can be reduced to hydrogen peroxide (H2O2), either spontaneously or by superoxide dismutase. The superoxide dismutase inhibitor DDC reduced the TBBPA induced DCF fluorescent by 92% indicating that is the precursor for the ROS measured by DCF fluorescence after TBBPA stimulation. The DCF assay is an attractive and sensitive method as an overall index for oxidative stress in biological systems. It is reported to detect several types of reactive molecules such as H2O2, in presence of cellular peroxidases, OONO– and OH?, but have no sensitivity towards (Myhre et al., 2003). Lucigenin is a sensitive probe for the detection of the superoxide anion radical, and is frequently used to demonstrate activation of respiratory burst in granulocytes (Halliwell and Gutteridge, 1999). The involvement of was confirmed in the lucigenin assay where DDC on the contrary increased the TBBPA induced ROS production. The DCF assay is primarily an indicator of intracellular formation of ROS, whereas lucigenin assay primarily measures extracellular ROS (Caldefie-Chezet et al., 2002). TBBPA appeared to induce both extracellular and intracellular formation of ROS by its activity towards both assays. The addition of SOD, which generally is nonpermeable to the cell membrane (Halliwell and Whiteman, 2004), strengthened this assumption by a total inhibition of TBBPA induced lucigenin-amplified chemiluminescence and no apparent effect towards TBBPA induced DCF fluorescence. DPI, a potent inhibitor of the activation of the NADPH oxidase complex, reduced ROS formation induced by TBBPA by 60% with use of the DCF assay and was completely abolished in the lucigenin assay. Lucigenin must undergo reduction to lucigenin cation to detect . The primary reducing agent in phagocytes is the NADPH-oxidase system (Halliwell and Gutteridge, 1999). This suggests that the effect of TBBPA is mediated mainly through activation of the NADPH oxidase complex.
Phosphorylation of the cytosolic subunits, p47PHOX, p67PHOX, and p40PHOX are essential elements in activation of the NADPH oxidase complex. p47PHOX is the subunit chiefly responsible for transporting the cytosolic complex from cytosol to the membrane during oxidase activation (Babior, 1999). Several pathways may activate the NADPH oxidase of which mitogen-activated protein kinase (MAPK) and PKC seem to be most important in our experiments (Fig. 8). The MAPKs are major information pathways from the cell surface to the nucleus. The MAPK pathway includes the c-Jun N-terminal kinase (JNK) and p38 MAPK cascade, which function mainly in stress responses like inflammation and apoptosis, as well as the extracellular signal-regulated kinase 1 and 2 (ERK 1/2), which preferentially regulate cell growth and differentiation (Lewis et al., 1998). U0126 is a selective inhibitor for MEK 1/2, the upstream activator of ERK 1/2 type of MAPK (Favata et al., 1998). The ERK 1/2 pathway participate in the phosphorylation of the NADPH oxidase component p47PHOX (Dewas et al., 2000), resulting in activation of the NADPH oxidase complex, and thereby production. We found a large reduction in ROS formation after incubation with U0126 in combination with TBBPA, indicating an involvement of ERK 1/2 in the formation of ROS. In contrast, the p38 inhibitor SB203580 (Cuenda et al., 1995) and the p38/JNK inhibitor FK506 (Matsuda et al., 2000) had no effect on the TBBPA stimulated DCF fluorescence, at the inhibitor concentration tested. Activation of ERK 1/2 was confirmed by Western blotting of proteins from TBBPA stimulated neutrophils with anti-phospho ERK antibodies. This indicates that ERK 1/2 are important activators of TBBPA induced ROS formation in neutrophil granulocytes, and that the p38 and JNK branches of the MAP kinase pathway are not involved in this activation.
PKC is another kinase that phosphorylates p47PHOX, and this phosphorylation plays a major role in activation of the NADPH oxidase complex (Fig. 8) (Nauseef et al., 1991). This was also the case for TBBPA induced ROS formation since the PKC inhibitor BIM reduced the DCF fluorescence by 69% and lucigenin amplified chemiluminescence by 28%. Previously it has been shown that fMLP activates the NADPH oxidase by a co-operation between PKC and the ERK 1/2 pathway (Dewas et al., 2000). Western blot analysis indicated a similar mechanism in TBBPA induced NADPH oxidase activation since the PKC inhibitor BIM also inhibited TBBPA stimulated phosphorylation of ERK 1/2. Fontayne et al. (2002) showed that four out of five known PKC isoforms in neutrophils could induce differential phosphorylation and translocation of p47PHOX, and they suggested that different phosphorylation of p47PHOX by these PKC isoforms could be important in fine-tuning of the NADPH oxidase activity.
Erbstatin-A, a selective and potent inhibitor of tyrosine protein kinases (Kawada et al., 1993), strongly inhibited the ROS formation induced by TBBPA. In previous studies it has been shown that protein tyrosine kinases are involved in the signaling pathways employed by chemotactic factors and hydrocarbons in the stimulation of superoxide production in human neutrophils (Dreiem et al., 2003; Naccache et al., 1990). The phosphorylation of ERK 1 and 2 was, however, not affected by erbstatin-A. Erbstatin-A is an analogue to the tyrphostines, which previously were shown to have anti oxidant properties (Sagara et al., 2002). Erbstatin's effect on TBBPA induced ROS formation may therefore be attributed to a similar effect. However, previously it has been shown that respiratory burst induced by fMLP and PMA is very sensitive to tyrosine kinase inhibition, an effect that has not been attributed to PKC and ERK inhibition (Mocsai et al., 1997). These results indicate that tyrosine kinases might be involved in TBBPA induced ROS formation, but must be acting in parallel to or downstream of the MAP kinase cascade and PKC, rather than upstream to it (Fig. 8).
Oxidative stress is often linked to calcium uptake in cells. However, respiratory burst may be activated in both calcium dependent and independent pathways (Downey et al., 1995). TBBPA increased intracellular calcium measured with Fura-2 AM in a concentration dependent manner. An interesting effect of TBBPA was the observed elevation of intracellular calcium even though extracellular calcium was removed and EGTA added, indicating that TBBPA also elevates cytosolic free calcium through release from intracellular compartments. In the absence of extracellular calcium the ROS formation was reduced. Our findings therefore suggest an involvement of calcium dependent activation of PKC (Fig. 8). The calcium antagonist verapamil, a voltage-dependent calcium channel blocker, also attenuated the ROS formation induced by TBBPA. Verapamil also reduced ERK phosphorylation as shown by Western blot (Fig. 6C). However, voltage-dependent calcium channels have not been reported to exist in neutrophil surface membrane (Schrenzel et al., 1995). It is therefore likely that inhibition of oxidant production by verapamil is attributable to some mechanism independent of its calcium channel-blocking properties. It has been reported that verapamil reduces PMA-stimulated superoxide production in neutrophils by inhibition of protein phosphorylation, probably catalyzed by PKC (Irita et al., 1986). This indicates that TBBPA also activates respiratory burst through a calcium dependent pathway, probably via activation of calcium dependent PKC. The TBBPA induced increase in intracellular calcium may also explain the residual formation of ROS in the DCF-assay after knocking out the NADPH-oxidase with DPI.
Bisphenol-A also showed a small but significant increase in intracellular calcium at relatively high concentrations (20 μM). These findings indicate that the bromine substitution seems to play a crucial role in the ability to induce elevation in intracellular calcium and ROS formation. A previous study also shows that bromine substitution on the bisphenol structure plays a crucial role for activity (Meerts et al., 2000).
The present article demonstrates that one of the most frequently used BFR, TBBPA, potently activates the NADPH oxidase in granulocytes primarily through elevation of intracellular calcium, activation of PKC and the MAP kinase pathway. Although one should be careful to apply these results to an in vivo situation; this is of great concern since the levels of BFR are rapidly increasing, both in human and environmental samples. Human exposure to TBBPA has not been extensively investigated and due to its phenolic structure it is not expected to accumulate in the environment to the same degree as the more lipid-soluble environmental toxicant. Plasma levels up to a few ng g–1 lipids (low nM concentrations) have, however, been reported (Thomsen et al., 2001) and TBBPA is also found accumulated in predators eggs (Berger et al., 2004). The concentrations used in this study are higher than what is detected in biota; however, it is important to remember that we are also continuously exposed to other environmental contaminants with similar chemical and toxicological properties such as PCBs and dioxins. Some studies also show that environmental contaminants may act additive or even synergistically when combined (Bemis and Seegal, 1999; Eriksson et al., 2003). Earlier findings in our laboratory show that environmental contaminants have similar effects on human neutrophil granulocytes as shown for TBBPA in this study. Ortho substituted PCBs increase respiratory burst and elevates intracellular calcium in granulocytes at concentrations around 10 μM (Voie and Fonnum, 1998; Voie et al., 1998). A preliminary experiment in this study showed that PCB153, one of the most frequently found compound in environmental samples, and TBBPA in combination induced an additive effect as shown by lucigenin-amplified chemiluminescence. PCB 153 did not induce DCF fluorescence at this concentration indicating a PCB induced extracellular formation of ROS. PCB 153 and TBBPA in combination did not influence the TBBPA induced DCF-fluorescence making it unlikely that PCB153 inhibits SOD as previously shown for PCB47 and the PCB mixture A1242 (Narayanan et al., 1998).
Human neutrophil granulocytes play a key role in host defenses against invading pathogens and are major effectors of the acute inflammatory reactions. Activation of these cells during an immune response leads to formation of reactive oxygen species used to kill microorganisms. A possible threat to the cellular homeostasis arises from these reactive oxygen species, as they are known to be involved in cellular signaling and gene regulation (Allen and Tresini, 2000; Finkel, 1998). In addition to the direct threat to own cells and tissue, Koner et al. (1997) have shown a possible connection between free radical formation and immune suppression in rabbit.
ACKNOWLEDGMENTS
The authors are indebted to Dr. Anne Dreiem for helpful discussions, Dr. Yngvar Gundersen for proofreading the manuscript, and Avi Ring for assistance in the LDH measurements. The authors also acknowledge The Norwegian Defence Research Establishment and Norwegian Research Council, under the PROFO program, for financial support.
REFERENCES
Saint-Louis, R., and Pelletier, E. (2004). LC-ESI-MS-MS method for the analysis of tetrabromobisphenol A in sediment and sewage sludge. Analyst 129, 724–730. Epub 2004 Jul 2006.
Schrenzel, J., Demaurex, N., Lew, D. P., and Krause, K. H. (1995) [Characterization of Ca2+ influx in human neutrophils using the patch clamp technique]. Schweiz Med. Wochenschr. 125, 1174–1178.
Sellstrom, U., and Jansson B. (1995). Analysis of tetrabromobisphenol a in a product and environmental-samples. Chemosphere 31, 3085–3092.
Sweet, L. I., and Zelikoff, J. T. (2001). Toxicology and immunotoxicology of mercury: A comparative review in fish and humans. J. Toxicol. Environ. Health B Crit. Rev. 4, 161–205.
Thomsen, C., Lundanes, E., and Becher, G. (2001). Brominated flame retardants in plasma samples from three different occupational groups in Norway. J. Environ. Monit. 3, 366–370.
Voie, O. A., and Fonnum, F. (1998). Ortho substituted polychlorinated biphenyls elevate intracellular [Ca2+] in human granulocytes. Environ. Toxicol. Pharmacol. 5, 105–112.
Voie, O. A., Wiik, P., and Fonnum, F. (1998). Ortho-substituted polychlorinated biphenyls activate respiratory burst measured as luminol-amplified chemoluminescence in human granulocytes. Toxicol. Appl. Pharmacol. 150, 369–375.
Watanabe, I., Kashimoto, T., and Tatsukawa, R. (1983). Identification of the flame retardant tetrabromobisphenol-A in the river sediment and the mussel collected in Osaka. Bull. Environ. Contam. Toxicol. 31, 48–52.(Trine Reistad*,1, Espen M)
This study investigates the effects of one of the most frequently used brominated flame-retardants (BFR), tetrabromobisphenol-A (TBBPA), on formation of reactive oxygen species (ROS) and calcium levels in human neutrophil granulocytes. TBBPA enhanced ROS production in a concentration-depended manner (1–12 μM), measured as 2,7-dichlorofluorescein diacetate amplified (DCF) fluorescence. The results on ROS production by TBBPA was confirmed by lucigenin-amplified chemiluminescence. The TBBPA induced formation of ROS was due to activation of respiratory burst, as shown by the NADPH oxidase inhibitor DPI (10 μM). TBBPA induced activation of respiratory burst was also inhibited by the MEK 1/2 inhibitor U0126 (10 μM), the PKC inhibitor BIM (0.25 μM), and the tyrosine kinase inhibitor erbstatin-A (25 μM). We also found a small reduction in ROS formation in the absence of extracellular calcium and when verapamil was added. The phosphorylation of ERK 1/2 was confirmed by Western blotting. TBBPA also induced a concentration dependent increase in intracellular free calcium measured with Fura-2/AM. We suggest that exposure of human neutrophil granulocytes to the brominated flame retardant TBBPA leads to an activation of the NADPH oxidase primarily by an ERK 1/2 stimulated pathway. The data also show that PKC, calcium, and tyrosine kinases may be involved in the activation
Key Words: brominated flame-retardants (BFR); tetrabromobisphenol-A; neutrophil granulocytes; reactive oxygen species (ROS); MAP kinase pathway; calcium; extracellular signal-regulated kinase (ERK).
INTRODUCTION
Brominated flame-retardants (BFRs) are a large group of compounds widely used to protect various products, such as plastics, textiles, and electronic equipment from catching fire. Several of the BFRs are persistent and lipophilic compounds. They may bioaccumulate and are thus regarded as a potential environmental health problem (de Wit, 2002). Within this group we find compounds such as polybrominated diphenyl ethers (PBDE), tetrabromobisphenol-A (TBBPA), hexabromocyclododecane (HBCD), and polybrominated biphenyls (PBBs).
The phenolic TBBPA is industrially the most important individual BFR used with an annual demand of approximately 120,000 metric ton (de Wit, 2002). TBBPA is a phenolic compound primarily used as a chemically bound flame retardant, which is supposed to limit its spread in the environment and reduce its accumulative properties. However, studies have shown that this compound may leak from treated products (Sellstrom and Jansson, 1995) and several recent reports on TBBPA in human and wildlife samples have shown the presence of this compound. Human TBBPA serum levels were measured by Thomsen et al. (2001), who found TBBPA in blood serum from electronic dismantlers, circuit board producers, and laboratory personnel. TBBPA has also been reported to be a contaminant in sediments and mussels (Saint-Louis and Pelletier, 2004; Watanabe et al., 1983) and in eggs from predatory bird species (Berger et al., 2004).
The majority of studies conducted to date on BFRs, both with regards to toxicology and environmental levels, have focused on the PBDEs (de Wit, 2002). There is little knowledge about the toxicity of TBBPA, which is of concern considering its extensive use and presence in the environment. A few toxicological studies have been carried out and the TBBPA can elicit thyroidogenic and estrogenic-like activity in vitro (Darnerud, 2003; Kitamura et al., 2002; Meerts et al., 2000), and has a neurotoxic potential (Mariussen and Fonnum, 2003; Reistad et al., 2002). A recent work by Fukuda et al. (2004) showed that TBBPA induces polycystic lesions in the kidney of exposed newborn rats. Pullen et al. (2003) showed that TBBPA suppresses the induction of interleukin-2 in murine splenocytes in vitro, indicating an immunotoxic potential.
Human neutrophil granulocytes play a key role in host defenses against invading pathogens and are major effectors of the acute inflammatory reactions. In response to a variety of agents, neutrophils release large quantities of superoxide anion () in a phenomenon known as respiratory burst. Neutrophil production of is dependent on the NADPH oxidase, a multicomponent membrane-bound enzyme that catalyzes NADPH-dependent reduction of oxygen to (Babior, 1999). Superoxide is rapidly converted to hydrogen peroxide (H2O2), either spontaneously or enzymatically by superoxide dismutase. H2O2 is then reduced to water by catalase or converted to hypochlorous acid (HOCl) by myeloperoxidase (MPO). H2O2 can also be converted to hydroxyl radicals in the presence of transition metal ions. Inappropriate activation of respiratory burst is associated with tissue injury and impairment of the ability to clear invading microorganisms (Labro, 2000).
Previously a correlation has been found between wildlife animals' exposure to environmental contaminants, such as polychlorinated biphenyls (PCB) and methyl mercury, and effects on immune parameters. Some of these findings have been attributed to activation of neutrophil granulocytes in vitro (Duffy et al., 2002; Sweet and Zelikoff, 2001; Voie et al., 1998). Because of the high production volume of TBBPA, its presence in biotic samples and the close resemblance to other environmental contaminants, we have examined its effect on human neutrophil granulocytes.
MATERIALS AND METHODS
Chemicals. Tetrabromobisphenol-A (TBBPA, BA-59P, Great Lakes, Lot nr 6L16,C) was all obtained from Promochem (Stockholm, Sweden). Stock solutions were prepared by dissolving the compounds in DMSO. The final DMSO concentration in the samples was always less than 1%. Bisphenol-A, bisindolylmaleimide (BIM), bromphenol blue, ponceau S concentrate, bis-N-Methylacridinium-nitrate (Lucigenin), 2,7-dichlorofluorescein diacetate (DCFH-DA), dimethyl sulfoxide (DMSO), 2-merchaptoethanol, diphenyleneiodonium (DPI), diethyldithio-carbamic acid (DDC), EGTA, verapamil hydrochloride, cyclosporine A (CSA), methanol, phosphate-buffered saline (PBS), phorbol 12-myristate 13-acetate (PMA), SB203580, superoxide dismutase (SOD), and sodium dodecyl sulphate (SDS) were all from Sigma-Aldrich (St. Louis, MO). U0126 was obtained from Promega Corporation (Madison, WI). Hanks Balanced Salt Solution (HBSS) and HEPES buffer were purchased from GibcoBRL (U.K.). Lymphoprep was purchased from Nycomed Pharma (Oslo, Norway). Enhanced chemiluminescence (ECL) reagent was from Amersham Pharmacia Biotech AB (Uppsala, Sweden). Monoclonal mouse anti-phospho-ERK antibody (Tyr204) and polyclonal rabbit anti-ERK2 antibody were from Santa Cruz Biotechnology (Santa Cruz, CA). Horseradish peroxidase (HRP)-conjugated rabbit-anti-mouse antibody and HRP-conjugated goat-anti-rabbit antibody were purchased from DAKO A/S (Glostrup, Denmark). 2,5-Dihydroxymethylcinnamate (Erbstatin-A), FK-506 (Tacrolimus), and Fura-2/AM were from Calbiochem Novabiochem Corp. (San Diego, CA). All other reagents used were analysis grade laboratory chemicals from standard commercial suppliers.
Isolation of human neutrophil granulocytes. Human venous blood was obtained from nonsmoking healthy adult male volunteers in the morning. Neutrophil granulocytes were separated from EDTA blood by dextran sedimentation followed by a standard density-gradient centrifugation as previously described (Boyum et al., 1991). In brief, EDTA blood from individual donors (30 ml) were mixed with 3 ml 6% dextran and left for sedimentation at room temperature for 30 min. The supernatant, containing the granulocytes, was subject to Lymphoprep density gradient centrifugation at 600 x g for 15 min. The pellet was washed in 0.9% NaCl and then resuspended in 7 ml 0.83% NH4Cl in 7 min for lysis of the erythrocytes, and then centrifuged for 7 min (600 x g). This was repeated if not proper lysis was obtained. Cells were then resuspended in HBSS and the number of granulocytes was determined in an AVIDA 60 hematology system. The cells were kept on ice (approximately 4°C) until use.
Lactate dehydrogenase (LDH) assay. Leakage of LDH was assessed as an index of cell injury (Koh and Choi, 1987). The measurements were performed as described elsewhere (Ring and Tanso, manuscript in preparation). In brief, cells (2 x 106/ml) were exposed to BRF for the indicated times (5 or 30 min). Cells were then spun down and supernatant from each sample was transferred to sample tubes and stored at 4°C until measured (usually within 2 h). LDH measurements were performed by transfer of 100 μl aliquots of the supernatant to the wells of a custom made 48 well microplate with glass bottom and the volume was adjusted to 425 μl with 0.1 M KPO2 buffer (pH 7.5). The reactions were started by automated injection of 50 μl of an 846 μM stock solution of NADH (final concentration 84.6 μM) followed by automated injection of 25 μl of a 13.6 mM stock solution of pyruvate (final concentration 0.68 mM). The LDH activity was measured, using a BMG FLUOstar Optima fluorimeter, from the decay rate of NADH fluorescence for 30 min at 28°C. The LDH activity was calculated off line and is given as the rate constant of the decrease in fluorescence emission at 460 nm (excitation wavelength 340 nm). The LDH activity (fluorescence units/s) is not a direct measure of the number of dead cells but it gives a qualitative measure of the relative amount of cell necrosis. 100% cell death was estimated by administration of 0.01% triton and corresponded to a NADH fluorescence decay rate of approximately 95 units/s (control values, 11 units/s). In Table 1 the values are shown as % of triton ± SEM.
Assay for measuring reactive oxygen species by DCF-fluorescence. Formation of ROS was measured with use of the fluorescent probe DCFH-DA. The method is based on the incubation of the granulocytes with DCFH-DA, which diffuses passively through the cellular membrane. Intracellular esterase activity results in the formation of DCFH, which emits fluorescence when oxidized to 2', 7'-dichlorofluorescein (DCF). The fluorescence emitted by DCF reflects the oxidative status of the cell and was determined essentially as described previously (Myhre et al., 2000). Briefly, the cells (final concentration 2 x 106/ml suspension) were incubated with DCFH-DA (5 μM) in HEPES-buffered (20 mM) HBSS (CaCl2 1.26 mM, KCL 5.37 mM, KH2PO4 0.44 mM; MgCl2 0.49 mM, MgSO4 0.41 mM, NaCl 140 mM, NaHCO3 4.17 mM, Na2HPO4 0.34 mM) with glucose (5 mM) at 37°C for 15 min. Following centrifugation, the extracellular buffer with DCFH-DA was exchanged with fresh buffer and the suspension was mixed gently. The cells (2 x 106/ml, 125 μl) were transferred to 250 μl wells (microtiter plate reader, 96 wells) containing 125 μl buffer with the BFR and/or the different inhibitors. Fluorescence was recorded in a Perkin-Elmer LS50B luminescence spectrometer (excitation wavelength 485 nm, emission wavelength 530 nm) at 37°C for 60 min. Results are calculated as area under the curve (AUC) and presented as values relative to control (% of control). PMA (1 x 107 M) was included as a positive control in all experiments (n = 5–8).
Assay for measuring reactive oxygen species by lucigenin-amplified chemiluminescence. Lucigenin chemiluminescence was used to detect in neutrophil granulocytes. The reaction mixture (250 μl) contained 0.1 mM lucigenin, 2 x 105 cells and different concentrations of the compounds. CL was measured by a Labsystem Luminoskan luminometer at 37°C for 60 min. PMA (1 x 10– 7 M) was included as a positive control in all experiments (n = 5–7). The cells and reagents were prepared in HEPES-buffered (20 mM) HBSS with 5 mM glucose. When calcium free buffer was used, 2 mM EGTA was also added. The reaction was started by adding 100 μl of the cell suspension to each well. Results are calculated as AUC and presented as values relative to control (% of control).
Western blotting. The neutrophils were isolated as described above and incubated at 37°C for 30 min before stimulation. TBBPA was added in different concentrations (6–24 μM) and the cells were incubated for different time-spans (2–20 min). After the indicated incubation period the cells were added 500 μl ice-cold PBS and immediately lysed in sample buffer (final concentration 3% SDS, 5% glycerol, 62.5 mM Tris/HCl pH 6.9, 0.1% bromphenol blue, 6% ?-Mercaptoethanol). Total cell samples were heated for 5 min at 95°C and analyzed on a 3% stacking 12% separating SDS-PAGE gel (2 h at 90 V). The separated proteins were then electrophoretically transferred to nitrocellulose membranes (0.45 μm) overnight (30 mA), and stained with Ponceau S to confirm complete transfer. The nitrocellulose blots were incubated in blocking buffer (Tris-buffered saline containing 0.05% Tween 20 [TBST] and 5% low-fat dry milk) for 1 h and probed with monoclonal anti phospho-ERK 1/2 primary antibody (1:200 dilution in blocking buffer) for 1 h. The blots were washed in TBST (6 x 5 min) and then incubated with peroxidase-conjugated rabbit anti-mouse secondary antibody (1:1000 dilution in blocking buffer) for 1 h. After washing in TBST (6 x 5 min), the blots were developed with Amersham's ECL system according to instructions provided by the producer. The signals were visualized on X-OmatBlue XB-1 film (Kodak). The experiments were repeated at least three times. The membranes were then stripped in 100 mM ?-Mercaptoethanol, 2% SDS, and 62.5 mM Tris/HCl (pH 7.6) for 30 min at 50°C and proceeded again with a rabbit polyclonal anti-ERK-2 primary antibody and peroxidase-conjugated goat anti-rabbit secondary antibody to confirm equal amounts of protein in each well.
Measurement of intracellular free calcium in granulocytes. Intracellular free [Ca2+] was measured by using the fluorescent Ca2+-binding probe fura-2/AM by the method previously described (Grynkiewicz et al., 1985). An increase in Ca2+ concentration is indicated by an increase in the fluorescence excitation ratio (I340/I380). Granulocytes (4.5 x 106 cells/ml) in HBSS containing 20 mM HEPES and 5 mM glucose were incubated at 37°C with 5 μM Fura-2/AM for 20 min. The cells were washed and resuspended in HEPES-buffered HBSS with glucose. Measurements of Fura-2 mediated fluorescence was performed on a computerized Shimadzu RF-5301PC Spectrofluorophotometer, using excitation wavelength ranging between 340 and 380 nm and emission wavelength 510 nm. All data are results of 5–9 separate measurements.
Statistical analyses. Differences between controls and treated groups were evaluated using a two-way Student's t-test (paired, two tail distribution), or by one-way ANOVA followed by Dunnett's 2-sided Post Hoc test. The calculations were performed using SPSS 11.5.
RESULTS
The Effect of BFRs on Human Neutrophil Granulocytes
Relatively low concentrations of the brominated flame retardant TBBPA induced a concentration dependent increase in DCF fluorescence in human neutrophil granulocytes (Fig. 1). TBBPA induced formation of ROS in the granulocytes was confirmed by lucigenin (Fig. 2). DMSO, which was used for dilution of the test compounds, reduced ROS formation in unstimulated cells to 84 ± 6% (mean ± SEM) of the control in the DCF assay and to 97 ± 7% (mean ± SEM) of the control in the lucigenin assay. The neutrophils were also exposed to bisphenol-A, which is a non-brominated analog to TBBPA. Bisphenol-A induced a concentration dependent increase in DCF-fluorescence, but had no effect on the lucigenin assay (Fig. 1).
The cytotoxicity assay shown as LDH-leakage showed a small, but significant solvent (DMSO) effect both at 5 and 30 min of exposure. The test compounds had no effect at the concentrations tested (Table 1).
The Involvement of the NADPH Oxidase, Superoxide Dismutase, Protein Kinase C, Tyrosine Kinase, and Calcium in TBBPA Induced ROS Formation
For mechanistic studies we have used inhibitors of different intracellular signaling pathways. The concentration of the inhibitors is based on what is used in the literature. In each assay used in the mechanistic studies we chose the concentration that gave the highest ROS response which for lucigenin was 4 μM TBBPA and for DCF 12 μM TBBPA.
Incubation of the granulocytes with 4 μM of the NADPH oxidase inhibitor DPI (O'Donnell et al., 1993) reduced TBBPA induced ROS formation by 60% with the DCF assay and gave total inhibition with the lucigenin assay. This indicates that activation of the NADPH oxidase complex is involved in the ROS formation induced by TBBPA (Figs. 2 and 4). The superoxide dismutase inhibitor DDC (Misra, 1979) attenuated the TBBPA induced DCF response almost completely, whereas the lucigenin chemiluminescence on the contrary was increased, demonstrating that TBBPA induce superoxide anion radical formation. Addition of superoxide dismutase (50 U/ml) showed an almost total inhibition of TBBPA induced ROS formation in the lucigenin assay and had no effect in the DCF assay. This shows that lucigenin measure extracellular ROS while DCF measure intracellular ROS. Incubation of the granulocytes with 0.25 μM BIM, a specific PKC inhibitor (Bit et al., 1993), reduced the DCF-fluorescence by 69%, and the lucigenin response by 28%. The TBBPA induced response was also strongly inhibited by 25 μM erbstatin-A, an inhibitor of tyrosine kinases (Kawada et al., 1993). The inhibition was 82% relative to the control in the DCF assay and total inhibition in the lucigenin assay. Calcium free buffer and the voltage-dependent calcium channel blocker verapamil (Hille, 1992) reduced TBBPA induced DCF fluorescence by 76 and 61%, respectively and the TBBPA induced lucigenin chemiluminescence by 69 and 9%, respectively. The mitochondrial transition pore blocker CSA (1 μM), used to indicate release of ROS from mitochondria (Baysal et al., 1991), had no inhibitory effect (data not shown).
DISCUSSION
The results presented in this study demonstrate that the BFR, TBBPA, activates respiratory burst in human neutrophil granulocytes as shown with DCF fluorescence and lucigenin-amplified chemiluminescence. This is an important finding, as TBBPA are used extensively for a large variety of applications and can be detected in wildlife and humans (de Wit, 2002). Bisphenol-A, a non-brominated analog of TBBPA, also induced ROS, but with lower potency than TBBPA, both on molar and weight basis.
Formation of ROS in granulocytes may be induced by different mechanisms, of which activation of the NADPH oxidase is the most important (Fig. 8). The neutrophil NADPH oxidase is a multicomponent membrane-bound enzyme that catalyses NADPH dependent reduction of oxygen to , which may be converted to H2O2, in the presence of peroxidases, OONO–, HOCl, and OH? (Babior, 1999). Superoxide anion () formed by NADPH oxidase activation can be reduced to hydrogen peroxide (H2O2), either spontaneously or by superoxide dismutase. The superoxide dismutase inhibitor DDC reduced the TBBPA induced DCF fluorescent by 92% indicating that is the precursor for the ROS measured by DCF fluorescence after TBBPA stimulation. The DCF assay is an attractive and sensitive method as an overall index for oxidative stress in biological systems. It is reported to detect several types of reactive molecules such as H2O2, in presence of cellular peroxidases, OONO– and OH?, but have no sensitivity towards (Myhre et al., 2003). Lucigenin is a sensitive probe for the detection of the superoxide anion radical, and is frequently used to demonstrate activation of respiratory burst in granulocytes (Halliwell and Gutteridge, 1999). The involvement of was confirmed in the lucigenin assay where DDC on the contrary increased the TBBPA induced ROS production. The DCF assay is primarily an indicator of intracellular formation of ROS, whereas lucigenin assay primarily measures extracellular ROS (Caldefie-Chezet et al., 2002). TBBPA appeared to induce both extracellular and intracellular formation of ROS by its activity towards both assays. The addition of SOD, which generally is nonpermeable to the cell membrane (Halliwell and Whiteman, 2004), strengthened this assumption by a total inhibition of TBBPA induced lucigenin-amplified chemiluminescence and no apparent effect towards TBBPA induced DCF fluorescence. DPI, a potent inhibitor of the activation of the NADPH oxidase complex, reduced ROS formation induced by TBBPA by 60% with use of the DCF assay and was completely abolished in the lucigenin assay. Lucigenin must undergo reduction to lucigenin cation to detect . The primary reducing agent in phagocytes is the NADPH-oxidase system (Halliwell and Gutteridge, 1999). This suggests that the effect of TBBPA is mediated mainly through activation of the NADPH oxidase complex.
Phosphorylation of the cytosolic subunits, p47PHOX, p67PHOX, and p40PHOX are essential elements in activation of the NADPH oxidase complex. p47PHOX is the subunit chiefly responsible for transporting the cytosolic complex from cytosol to the membrane during oxidase activation (Babior, 1999). Several pathways may activate the NADPH oxidase of which mitogen-activated protein kinase (MAPK) and PKC seem to be most important in our experiments (Fig. 8). The MAPKs are major information pathways from the cell surface to the nucleus. The MAPK pathway includes the c-Jun N-terminal kinase (JNK) and p38 MAPK cascade, which function mainly in stress responses like inflammation and apoptosis, as well as the extracellular signal-regulated kinase 1 and 2 (ERK 1/2), which preferentially regulate cell growth and differentiation (Lewis et al., 1998). U0126 is a selective inhibitor for MEK 1/2, the upstream activator of ERK 1/2 type of MAPK (Favata et al., 1998). The ERK 1/2 pathway participate in the phosphorylation of the NADPH oxidase component p47PHOX (Dewas et al., 2000), resulting in activation of the NADPH oxidase complex, and thereby production. We found a large reduction in ROS formation after incubation with U0126 in combination with TBBPA, indicating an involvement of ERK 1/2 in the formation of ROS. In contrast, the p38 inhibitor SB203580 (Cuenda et al., 1995) and the p38/JNK inhibitor FK506 (Matsuda et al., 2000) had no effect on the TBBPA stimulated DCF fluorescence, at the inhibitor concentration tested. Activation of ERK 1/2 was confirmed by Western blotting of proteins from TBBPA stimulated neutrophils with anti-phospho ERK antibodies. This indicates that ERK 1/2 are important activators of TBBPA induced ROS formation in neutrophil granulocytes, and that the p38 and JNK branches of the MAP kinase pathway are not involved in this activation.
PKC is another kinase that phosphorylates p47PHOX, and this phosphorylation plays a major role in activation of the NADPH oxidase complex (Fig. 8) (Nauseef et al., 1991). This was also the case for TBBPA induced ROS formation since the PKC inhibitor BIM reduced the DCF fluorescence by 69% and lucigenin amplified chemiluminescence by 28%. Previously it has been shown that fMLP activates the NADPH oxidase by a co-operation between PKC and the ERK 1/2 pathway (Dewas et al., 2000). Western blot analysis indicated a similar mechanism in TBBPA induced NADPH oxidase activation since the PKC inhibitor BIM also inhibited TBBPA stimulated phosphorylation of ERK 1/2. Fontayne et al. (2002) showed that four out of five known PKC isoforms in neutrophils could induce differential phosphorylation and translocation of p47PHOX, and they suggested that different phosphorylation of p47PHOX by these PKC isoforms could be important in fine-tuning of the NADPH oxidase activity.
Erbstatin-A, a selective and potent inhibitor of tyrosine protein kinases (Kawada et al., 1993), strongly inhibited the ROS formation induced by TBBPA. In previous studies it has been shown that protein tyrosine kinases are involved in the signaling pathways employed by chemotactic factors and hydrocarbons in the stimulation of superoxide production in human neutrophils (Dreiem et al., 2003; Naccache et al., 1990). The phosphorylation of ERK 1 and 2 was, however, not affected by erbstatin-A. Erbstatin-A is an analogue to the tyrphostines, which previously were shown to have anti oxidant properties (Sagara et al., 2002). Erbstatin's effect on TBBPA induced ROS formation may therefore be attributed to a similar effect. However, previously it has been shown that respiratory burst induced by fMLP and PMA is very sensitive to tyrosine kinase inhibition, an effect that has not been attributed to PKC and ERK inhibition (Mocsai et al., 1997). These results indicate that tyrosine kinases might be involved in TBBPA induced ROS formation, but must be acting in parallel to or downstream of the MAP kinase cascade and PKC, rather than upstream to it (Fig. 8).
Oxidative stress is often linked to calcium uptake in cells. However, respiratory burst may be activated in both calcium dependent and independent pathways (Downey et al., 1995). TBBPA increased intracellular calcium measured with Fura-2 AM in a concentration dependent manner. An interesting effect of TBBPA was the observed elevation of intracellular calcium even though extracellular calcium was removed and EGTA added, indicating that TBBPA also elevates cytosolic free calcium through release from intracellular compartments. In the absence of extracellular calcium the ROS formation was reduced. Our findings therefore suggest an involvement of calcium dependent activation of PKC (Fig. 8). The calcium antagonist verapamil, a voltage-dependent calcium channel blocker, also attenuated the ROS formation induced by TBBPA. Verapamil also reduced ERK phosphorylation as shown by Western blot (Fig. 6C). However, voltage-dependent calcium channels have not been reported to exist in neutrophil surface membrane (Schrenzel et al., 1995). It is therefore likely that inhibition of oxidant production by verapamil is attributable to some mechanism independent of its calcium channel-blocking properties. It has been reported that verapamil reduces PMA-stimulated superoxide production in neutrophils by inhibition of protein phosphorylation, probably catalyzed by PKC (Irita et al., 1986). This indicates that TBBPA also activates respiratory burst through a calcium dependent pathway, probably via activation of calcium dependent PKC. The TBBPA induced increase in intracellular calcium may also explain the residual formation of ROS in the DCF-assay after knocking out the NADPH-oxidase with DPI.
Bisphenol-A also showed a small but significant increase in intracellular calcium at relatively high concentrations (20 μM). These findings indicate that the bromine substitution seems to play a crucial role in the ability to induce elevation in intracellular calcium and ROS formation. A previous study also shows that bromine substitution on the bisphenol structure plays a crucial role for activity (Meerts et al., 2000).
The present article demonstrates that one of the most frequently used BFR, TBBPA, potently activates the NADPH oxidase in granulocytes primarily through elevation of intracellular calcium, activation of PKC and the MAP kinase pathway. Although one should be careful to apply these results to an in vivo situation; this is of great concern since the levels of BFR are rapidly increasing, both in human and environmental samples. Human exposure to TBBPA has not been extensively investigated and due to its phenolic structure it is not expected to accumulate in the environment to the same degree as the more lipid-soluble environmental toxicant. Plasma levels up to a few ng g–1 lipids (low nM concentrations) have, however, been reported (Thomsen et al., 2001) and TBBPA is also found accumulated in predators eggs (Berger et al., 2004). The concentrations used in this study are higher than what is detected in biota; however, it is important to remember that we are also continuously exposed to other environmental contaminants with similar chemical and toxicological properties such as PCBs and dioxins. Some studies also show that environmental contaminants may act additive or even synergistically when combined (Bemis and Seegal, 1999; Eriksson et al., 2003). Earlier findings in our laboratory show that environmental contaminants have similar effects on human neutrophil granulocytes as shown for TBBPA in this study. Ortho substituted PCBs increase respiratory burst and elevates intracellular calcium in granulocytes at concentrations around 10 μM (Voie and Fonnum, 1998; Voie et al., 1998). A preliminary experiment in this study showed that PCB153, one of the most frequently found compound in environmental samples, and TBBPA in combination induced an additive effect as shown by lucigenin-amplified chemiluminescence. PCB 153 did not induce DCF fluorescence at this concentration indicating a PCB induced extracellular formation of ROS. PCB 153 and TBBPA in combination did not influence the TBBPA induced DCF-fluorescence making it unlikely that PCB153 inhibits SOD as previously shown for PCB47 and the PCB mixture A1242 (Narayanan et al., 1998).
Human neutrophil granulocytes play a key role in host defenses against invading pathogens and are major effectors of the acute inflammatory reactions. Activation of these cells during an immune response leads to formation of reactive oxygen species used to kill microorganisms. A possible threat to the cellular homeostasis arises from these reactive oxygen species, as they are known to be involved in cellular signaling and gene regulation (Allen and Tresini, 2000; Finkel, 1998). In addition to the direct threat to own cells and tissue, Koner et al. (1997) have shown a possible connection between free radical formation and immune suppression in rabbit.
ACKNOWLEDGMENTS
The authors are indebted to Dr. Anne Dreiem for helpful discussions, Dr. Yngvar Gundersen for proofreading the manuscript, and Avi Ring for assistance in the LDH measurements. The authors also acknowledge The Norwegian Defence Research Establishment and Norwegian Research Council, under the PROFO program, for financial support.
REFERENCES
Saint-Louis, R., and Pelletier, E. (2004). LC-ESI-MS-MS method for the analysis of tetrabromobisphenol A in sediment and sewage sludge. Analyst 129, 724–730. Epub 2004 Jul 2006.
Schrenzel, J., Demaurex, N., Lew, D. P., and Krause, K. H. (1995) [Characterization of Ca2+ influx in human neutrophils using the patch clamp technique]. Schweiz Med. Wochenschr. 125, 1174–1178.
Sellstrom, U., and Jansson B. (1995). Analysis of tetrabromobisphenol a in a product and environmental-samples. Chemosphere 31, 3085–3092.
Sweet, L. I., and Zelikoff, J. T. (2001). Toxicology and immunotoxicology of mercury: A comparative review in fish and humans. J. Toxicol. Environ. Health B Crit. Rev. 4, 161–205.
Thomsen, C., Lundanes, E., and Becher, G. (2001). Brominated flame retardants in plasma samples from three different occupational groups in Norway. J. Environ. Monit. 3, 366–370.
Voie, O. A., and Fonnum, F. (1998). Ortho substituted polychlorinated biphenyls elevate intracellular [Ca2+] in human granulocytes. Environ. Toxicol. Pharmacol. 5, 105–112.
Voie, O. A., Wiik, P., and Fonnum, F. (1998). Ortho-substituted polychlorinated biphenyls activate respiratory burst measured as luminol-amplified chemoluminescence in human granulocytes. Toxicol. Appl. Pharmacol. 150, 369–375.
Watanabe, I., Kashimoto, T., and Tatsukawa, R. (1983). Identification of the flame retardant tetrabromobisphenol-A in the river sediment and the mussel collected in Osaka. Bull. Environ. Contam. Toxicol. 31, 48–52.(Trine Reistad*,1, Espen M)