Fluoride-induced IL-8 release in human epithelial lung cells: Relationship to EGF-receptor-, SRC- and MAP-kinase activation

Available online at www.sciencedirect.com

Toxicology and Applied Pharmacology 227 (2008) 56 – 67 www.elsevier.com/locate/ytaap

Fluoride-induced IL-8 release in human epithelial lung cells: Relationship to EGF-receptor-, SRC- and MAP-kinase activation Magne Refsnes ⁎,1 , Tonje Skuland 1 , Per E. Schwarze 2 , Johan Øvrevik 2 , Marit Låg 2 Norwegian Institute of Public Health, Division of Environmental Medicine, PO Box 4404 Nydalen, NO-0403 Oslo, Norway Received 28 June 2007; revised 19 September 2007; accepted 25 September 2007 Available online 3 October 2007

Abstract Exposure of human epithelial lung cells to fluorides is known to induce a marked increase in the release of interleukin (IL)-8, a chemokine involved in neutrophil recruitment. In the present study, the involvement of mitogen-activating protein kinases (MAPKs), the role of upstream activation of Src family kinases (SFKs), epidermal growth factor receptor (EGFR) activation and the interrelationships between these pathways in fluoride-induced IL-8 were examined in a human epithelial lung cell line (A549). Sodium fluoride strongly activated MAPK, in particular JNK1/2 and p38. The ERK1/2-inhibitor PD98059, the p38-inhibitor SB202190 and the JNK1/2-inhibitor SP600125 partially inhibited the fluorideinduced IL-8 response. Combinations of these inhibitors reduced the responses nearly to basal levels. Treatment with siRNA against JNK2 also reduced the IL-8 response to fluoride. Furthermore, fluoride activated SFKs, which was abolished by the SFK-inhibitor PP2. PP2 substantially inhibited the increased levels of IL-8, and partially reduced the fluoride-induced activation of ERK1/2, p38 and JNK1/2. Fluoride exposure also led to a phosphorylation of the EGFR, that was partially inhibited by PP2. AG1478, an EGFR-inhibitor, partially reduced the fluoride-induced IL8 response and the phosphorylation of JNK1/2 and ERK1/2, but less the phosphorylation of p38. The effects of AG1478 were less than that of PP2. In conclusion, our findings suggest that the fluoride-induced IL-8 release involves the combined activation of ERK1/2, JNK1/2 and p38, and that the phosphorylation of these kinases, and in particular JNK1/2 and ERK1/2, partly, is mediated via a SFK-dependent EGFR-linked pathway. SFK-dependent, but EGFR-independent mechanisms seem important, and especially for phosphorylation of p38. © 2007 Elsevier Inc. All rights reserved. Keywords: Fluoride; IL-8; MAPK; src; EGFR; Epithelial lung cells

Introduction Fluoride exposure has been associated with asthmatic symptoms among workers in the aluminum industry (Soyseth and Kongerud, 1992). We have previously shown that hydrogen fluoride (HF) exposure of human volunteers in concentrations that may occur in the pot rooms of aluminum melters, induced acute inflammatory responses in the respiratory tract, with changes in inflammatory cells and cytokines in broncheoalveolar and nasal lavage fluids (Lund et al., 1999, 2002). This may suggest that HF-induced inflammatory changes are important for subsequent development of airway disease. We have pre⁎ Corresponding author. E-mail addresses: [email protected] (M. Refsnes), [email protected] (T. Skuland), [email protected] (P.E. Schwarze), [email protected] (J. Øvrevik), [email protected] (M. Låg). 1 Fax: +47 22042686. 2 Fax: +47 22042386. 0041-008X/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2007.09.022

viously demonstrated that fluoride induced a strong increase in the release of interleukin (IL)-8 from a human epithelial lung cell line (A549) (Refsnes et al., 1999). IL-8 (or CXCL8) belongs to the CXC chemokines and is released from different types of cells. This chemokine is a crucial component in inflammatory processes that involve various cytokines in complex interactions between different cell types. IL-8 is important in recruiting neutrophilic cells to the site of injury, and has a pivotal role in several pathological conditions such as chronic inflammation, fibrosis and cancer (Mukaida, 2003). Increased IL-8 release involves the activation of a range of receptors and intracellular signal transduction pathways (Hoffmann et al., 2002). Among the key enzymes are the mitogen-activated protein kinase (MAPK) family of serine/threonine kinases, to which many other pathways are converging (Puddicombe and Davies, 2000). Both extra-cellular signal-regulated kinase-1 and -2 (ERK1/2), the c-jun-N-terminal kinases (JNKs) and the p38 MAPKs may be involved in regulation of IL-8 synthesis,

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

but the contribution of the different MAPKs varies depending on the cell type and the stimulant (Hashimoto et al., 2000; Refsnes et al., 2001; Furuichi et al., 2002; Hoffmann et al., 2002; Li et al., 2002). The detailed mechanisms underlying the fluoride-induced IL8 release from A549 cells are unclear. Sodium fluoride (NaF) has been reported to activate the MAPKs ERK1/2, p38 and JNK1/2 (Anderson et al., 1991; Susa et al., 1997; Thrane et al., 2001). However, the relative contribution of the different MAPK-pathways in fluoride-induced IL-8 release, the interplay between these pathways and the upstream mechanisms of fluorideinduced MAPK-activation remain to be clarified. The Src family of non-receptor tyrosin kinases (SFKs) and the receptor tyrosin kinase epidermal growth factor (EGFR) are likely candidates as upstream regulators of MAPK activation in fluoride-exposed cells. For some agents, SFKs have been reported to be involved in IL-8 release, via both MAPK-dependent and -independent pathways (Liu et al., 2001; Kitagawa et al., 2002; Ovrevik et al., 2004; Trevino et al., 2005). Furthermore, upstream to the MAPKs, EGFR transactivation may also mediate part of the IL-8 production (Wu et al., 2001; Richter et al., 2002; Adachi et al., 2004). Our previous findings suggest that fluorideinduced IL-8 release from A549 cells involves activation of p38 and protein kinase C (PKC), whereas Ca2+-linked signals and protein kinase A (PKA) seemed less important (Refsnes et al., 1999, 2001). Generally, fluoride is known to mediate its effects both through GTP-binding protein (G protein)-dependent and -independent pathways (Blackmore et al., 1985; Murthy and Makhlouf, 1994). In the present study, we elucidate the role of MAPKs p38, ERK1/2 and JNK1/2, in relation to upstream SFK and EGFR activation, in fluoride-induced IL-8 responses in A549 cells. Materials and methods Reagents. Culture medium, nutrition Mixture F12 HAM Kaighn's modification (F12K) and phenylmethylsulfonyl fluoride (PMSF) were obtained from Sigma-Aldrich, St. Louis, MO, USA. Fetal bovine serum (FBS) was from Euroclone, Pero, Italy. Ampicillin and fungizone were purchased from BristolMyer Squibb, Bromma, Sweden, and penicillin/streptomycin from Bio Whittaker, Walkersville, MD, USA. SB202190 (4-[4-fluorophenyl]-2-[4hydroxyphenyl]-5-[4-pyridyl]1 H-imidazole), PD98059 (2-amino-3methoxyflavone), PP2 (4-amino-5-[4-chlorophenyl]-7-[t-butyl]pyroazolo [3,4-d]pyrimidine), SP600125 (anthrax[1,9-cd]pyrazol-6(2H)-one) and AG1478 (4-[3-chloroanilino]-6,7-dimethoxyquinazoline) were purchased from CalbiochemNovabiochem Corporation, La Jolla, CA, USA. D-JNKI1 (c-jun N-terminal kinase inhibitor 1; D-stereoisomer) was from Alexis Biochemicals San Diego, CA. U0126 (1,4-diamino-2,3-dicyano-1,4-bis[2-aminophenylthio]-butadiene) and cytokine ELISA assays for human IL-8 (Cytoset CHC1304) and rat macrophage inflammatory protein (MIP-2) (Cytoset CRC1024) were supplied by Biosource International, Camarillo, CA, USA. JNK2 smartpool siRNA reagent against JNK2 and non-specific control pool (negative control) were from Upstate, Lake Placid, NY, USA. HiPerfect transfection reagent was from Qiagen, Hilden, Germany. Magnetic beads coated with IgG against rabbits were from Dynal, Oslo, Norway. All other chemicals were purchased from commercial sources at the highest purity available. Antibodies. Specific antibodies against phospho-p38, p38, phospho-JNK1/2, JNK2, phospho-Src family kinases (Tyr416) and phospho-tyrosine (P-Tyr-100) were obtained from Cell Signaling Technology Inc., Beverly, MA, USA. Antibodies against phospho-ERK1/2, ERK2 and EGFR, were obtained from Santa

57

Cruz Biotechnology Inc., Santa Cruz, CA, USA. The antibody against JNK2 was from Upstate, CA, USA. Beta-actin was obtained from Sigma-Aldrich, St. Louis, MO, USA. Cell cultures and exposure. A549 cells, a human epithelial lung cell line, from the American Type Culture Collection, Rockville, MD, USA was cultured in Ham's medium, supplemented with ampicillin (0.1 mg/ml), penicillin (0.1 mg/ml), streptomycin (0.1 mg/ml), fungizone (0.25 μg/ml) and 10% heatinactivated FBS. The cells (passage number 79–100) were plated in 35 mm sixwell culture dishes (2 × 104 cells/well) and grown to near confluency at 37 °C in a humidified atmosphere of 5% CO2 in air. The cells were exposed to 1.0 to 5.0 mM NaF for 2 h (for immunoblotting) and 20 h for IL-8 analysis. The MAPK inhibitors PD98059, U0126, SB202190, D-JNKI1, SP600125, alone or in combination, the SFK-inhibitor PP2 and the EGFR antagonist AG1478, were added 1 h prior to NaF. The inhibitors were kept in the culture medium during the whole incubation periods. Primary type 2 cells from rat lung were isolated as previously described (Lag et al., 1996). The purity of type 2 cells was approximately 90%, and the cell viability as assayed by trypan blue exclusion exceeded 90%. The type 2 cells were cultured in Williams' medium with 5% inactivated FBS and other supplements as described by Lag et al. (1996) and cultured for 2 days before exposure to NaF (0.1 to 0.5 mM). Transfection with siRNA against JNK2. A549 cells at a density of 400 000/ well were transfected with siRNA against JNK2, using HiPerfect transfection reagent according to the procedure to Quiagen, Germany. Briefly, siRNA against JNK2 was diluted in culture medium without serum to give a final SiRNA concentration of 5 nM, supplemented with HiPerfect transfection reagent, mixed by vortexing and incubated for 5–10 min at room temperature to allow the formation of tranfection complexes. The complexes were added drop-wise onto the cells cultured at low density; the cultures were gently swirled and further incubated for 24 to 96 h. The effectiveness of gene silencing was monitored at 24, 48, 72 and 96 h by measuring the JNK2 levels in relation to β-actin, as analyzed by Western blotting. As a negative control, “Smart pool”, Upstate, USA, was used. Cells transfected with siJNK2 or exposed to the HiPerfectbuffer for 48 h, were exposed to NaF for 20 h, and IL-8 levels were measured by ELISA. Measurements of IL-8 and macrophage inflammatory protein (MIP-2). At the end of exposure, supernatants were removed and centrifuged at 250×g to remove cells. The final supernatants were stored at − 70 °C. IL-8 and MIP-2 levels were determined by enzyme-linked immunosorbant assay (ELISA) according to the manufacturer's guidelines. Absorbance was measured and quantified using a plate reader (TECAN Sunrise, Phoenix Research Products, Hayward, CA, USA) complete with software (Magellan V 1.10), color intensity was converted to nanograms IL-8 and MIP-2 using appropriate standards. Measurements of cell viability. At the end of exposure to NaF, detached cells in the A549 cultures were removed. The remaining attached cells were subsequently attached and combined with the respective detached cells. Plasma membrane damage and apoptosis were determined after staining cells with propidium iodide (PI) (5.0 μg/ml) and Hoechst 333342 (10 μg/ml) for 30 min in the dark. Briefly the cells were centrifuged at 250×g at 4 °C for 10 min and washed twice. Smears on slides made of the pelleted cells, suspended in FBS, were quickly air-dried. Cell morphology was evaluated using a Nicon Eclipse E 4000 microscope. The potential of fluoride to induce plasma membrane damage was determined by the ability to exclude PI. Apoptotic cells were identified by their distinct condensed nuclei and/or nuclei fragmentation. Also cells with these changes but with compromised plasma membranes were included as apoptotic cells. Three to four hundred cells were counted. The concentration– response relationships for fluoride-induced plasma membrane damage and apoptotic response in A549 cells have previously been published (Refsnes et al., 2003). The plasma membrane damage, was only marginally affected by 3.75 mM NaF with an increase from 6.1 ± 1.0% in the controls to 7.9 ± 0.9% in the fluoride-treated cells. At higher concentrations of fluoride (5–10 mM), the membrane damage was strongly enhanced (Refsnes et al., 2003). At 3.75 mM NaF, the percentage of apoptotic cells was only weakly increased (from 1.0± 0.3% in the controls to 5.3 ± 0.7%).

58

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

Immunoblotting. Cells were harvested in ice-cold PBS containing PMSF (1 mM) and resuspended in lysis buffer (20 mM Tris–HCl, pH = 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 2.4 mM Na-pyrophosphate, 1.0 mM orthovanadate, 1 mM NaF, 21 μM leupeptin, 1.5 μM aprotinin, 15 μM Pepstatin A and 1% Triton-X) prior to protein determination by the BioRad DC Protein Assay (BioRad Life Science, CA, USA). Proteins (12.5 μg/well) from whole-cell lysates were separated by 10% SDS-PAGE and blotted onto nitrocellulose membranes. To ensure that the protein levels of each well were equal, Ponceau-staining was used for loading control. The membranes were then probed with antibodies for the respective phosphorylated kinases (pERK, p-JNK, p-p38, p-SFK) prior to incubation with horseradish peroxidaseconjugated secondary antibodies. The blots were developed using the SuperSignal® West Dura chemoluminiscence system (Pierce, Perbio Science, Sweden) according to the manufacturer's instructions. Finally the membranes were stripped by incubation for 15 min at room temperature with Mild antibody stripping solution® from Chemicon International, Temecula, CA, USA, and re-

probed for the total amount of the respective kinases (ERK2, JNK1/2, p38) and/or β-actin. Immunoprecipitation. To explore the effect of fluoride exposure on EGFR activity in A549 cell, the cells were incubated with fluoride for 2 h and with EGF (as positive control) for 15 min. The cells were washed twice in ice-cold PBS and once in ice-cold immunoprecipitation (IP)-buffer (50 mM Tris–HCl [pH 7.4], 280 mM NaCl, 0.2 mM EDTA, 50 mM Na4P2O7, 2 mM EGTA, 1 mM Na3VO4, 1 mM PMSF, 10% glycerol, complete protease inhibitor tablet [1 tablet/50 ml] and distilled water). The cells were then lysed in IP-buffer with 1% NP-40. Cell lysates were centrifuged at 2500×g for 10 min and incubated with antibodies against EGFR. Lysates (with antibodies) were incubated overnight under constant rotation with magnetic beads coated with IgG against rabbit. The complexes were washed once in IP-buffer containing 0.5% NP-40 and the immunoprecipitates were released from the beads by boiling in 2× Laemmli buffer for 5 min. The precipitated proteins were separated by SDS-PAGE prior to Western

Fig. 1. Fluoride-induced IL-8 response and phosphorylation of MAPKs in lung epithelial cells. (A) A549 cells were exposed to NaF for 20 h, and release of IL-8 release was measured by ELISA. The data represent the mean ± SEM of IL-8 responses from 5 experiments. The IL-8 response is presented as fold increase. The basal IL8 level was 1250 ± 290 pg/ml. (B) Primary type 2 cells were exposed to NaF for 20 h, and release of MIP-2 was measured by ELISA. The data represent the mean ± SEM of MIP-2 responses from 3 experiments. The data are presented as fold increase. The basal levels were 350 ± 250 pg/ml. (C) A549 cells were cultured for 2 h, with and without NaF (3.75 mM). Phosphorylated ERK1/2 and total ERK2, phosphorylated p38 and total p38, phosphorylated JNK1/2 and total JNK1/2 were detected by Western blotting. Typical blots and optical quantification (mean ± SEM) of the protein bands from 4 to 5 independent experiments are shown. Values are expressed as relative phosphorylation of the ERK2, p38, JNK1 and JNK2 compared to the total levels of the respective proteins, and are presented as fold increase. ⁎Significant increase in IL-8 release by Student's t-test (p ≤ 0.05).

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

59

Fig. 2. Involvement of MAPKs in fluoride-induced IL-8 release from A549 cells. A549 cells were pre-treated with different concentrations of the ERK1/2 inhibitor PD98059 (A), the p38 inhibitor SB202190 (B), the JNK-inhibitors D-JNKI1(C) and SP600125 (D) for 1 h prior to addition of 3.75 mM NaF, and further incubated for 20 h. IL-8 release was measured by ELISA. Each value represents mean ± SEM of 4–5 experiments, and is presented in percentage of the fluoride response in absence of inhibitors. The basal IL-8 levels (mean ± SEM) in A, B, C and D are 439 ± 138 pg/ml, 791 ± 81 pg/ml, 640 ± 180 pg/ml and 680 ± 130 pg/ml, respectively. ⁎Significant reduction ( p ≤ 0.05) by one-way ANOVA using Dunnet's post-test.

blotting, as described above, and incubated with an antibody against phosphotyrosine to detect phosphorylated EGFR. Statistical analysis. Statistical calculations were performed by Student's t-test or analysis of variance (ANOVA), with post-tests for multiple comparisons, as indicated in the figure legends. Significance was assigned to a p-value ≤ 0.05.

Results Involvement of MAPK on fluoride-induced IL-8 release As illustrated in Fig. 1A, fluoride induced a concentrationdependent increase in IL-8 release from A549 cells, with a 10-fold increase after 20 h exposure to 3.75 mM NaF, a concentration that only slightly affects cell viability. A significant increase was observed from 2.5 mM. As shown in Fig. 1B, primary type 2 cells from rats showed a higher sensitivity to fluoride, as shown by measuring of MIP-2, a chemokine analogous to IL-8 in humans. Thus, fluoride induced a 2-fold increase in the type 2 cells, with a maximal response at approximately 0.25–0.5 mM NaF. After 2 h of exposure of A549 cells to 3.75 mM fluoride, phosphorylation of MAPKs ERK1/2, JNK1/2 and p38 was markedly increased and most pronounced for JNK1/2 and p38, as shown by phosphospecific antibodies and Western analysis (Fig. 1C). MAPK p38 has previously been shown to be involved in fluoride-induced

IL-8 release in A549 cells (Refsnes et al., 2001). Using different MAPK inhibitors, alone and in combination, the role of MAPKs in fluoride-induced IL-8 responses was further examined. Fig. 2 shows concentration-effect curves for PD98059 that inhibits MEK1/2 upstream of ERK1/2 (A), the p38-inhibitor SB202190 (B), the JNK-inhibitors (D-JNKI1) (C) and SP600125 (D). Increasing concentrations of PD98059 progressively inhibited the fluoride-induced IL-8 release, with a maximal reduction at 25–50 μM. With SB202190 the maximal reduction was obtained at 3–10 μM. D-JNKI1, that is known to block the interaction between JNK and c-jun, and thereby inhibiting the signaling events downstream of JNK, did not reduce the fluoride-induced IL-8 at a concentration (1.0 μM) reported to be effective in other cell types (Borsello and Bonny, 2004). In fact, D-JNKI1 significantly increased the cytokine response. In contrast, SP600125 that is known to inhibit JNK1, JNK2 and JNK3 with about equal potency (Bennett et al., 2001) significantly reduced the IL-8 response at 10–40 μM. The basal IL-8 release was not affected by SB202190, D-JNKI1 and SP600125, and only slightly by PD98059. Another MEK-1 inhibitor U0126, also reduced the fluoride-induced IL-8 release, but affected the basal IL-8 release (data not shown). Since SP600125 reduced the fluoride-induced IL-8 response, whereas D-JNKI1 increased the response, we examined how

60

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

Fig. 3. Effect of siRNA against JNK-2 on protein levels (A) and on fluoride-induced IL-8 release (B) in A549 cells. A549 cells at a density of approximately 40,000 cells/cm2 were transfected with 5 nM siRNA against JNK-2 or non-specific siRNA in Hi Perfect-buffer. (A) Left panel: The cells were incubated for 24–72 h. The levels of JNK2 were examined by Western analysis, and related to β-actin levels. A typical blot and quantification of 3–4 experiments (mean ± SEM) are presented. Right panel: Cells were transfected with siRNA against JNK2 for 72 h and compared with cells transfected with non-specific siRNA for 72 h (negative control). (B) The cells were transfected with siRNA for 48 h against JNK2 and compared with the cells incubated with and without Hi Perfect-buffer. The cells were subsequently exposed to 3.75 mM NaF for 24 h and analyzed for IL-8 release by ELISA. The values represent mean ± SEM of 3 experiments. ⁎Significant reduction by Student's t-test ( p ≤ 0.05).

treatment of the cells with siRNA against JNK2 affected the fluoride-induced IL-8 response. The left panel in Fig. 3A shows that the JNK2 levels were substantially reduced 48, 72 and 96 h after transfection with the siRNA against JNK2, whereas no effect was observed at 24 h. Addition of fluoride did not affect the transfection (data not shown). The right panel in Fig. 3A shows that a non-specific siRNA duplex (Smart-pool from Upstate, USA) was without effect on the JNK2 levels. Fig. 3B shows that the IL-8 responses to fluoride were partially reduced in A549 cells transfected with siRNA against JNK2, in comparison to cells treated with transfection buffer. The reduction was only partial, as observed with the chemical JNKinhibitor SP600125, and was statistically significant in the exposure period from 48 to 72 h. The basal levels were not affected. The chemical inhibitors of the respective MAPKs only partially reduced the fluoride-induced IL-8 response. There-

fore, an additional set of experiments with inhibitors alone and in combination was performed (Fig. 4). SB202190 (10 μM), SP600125 (20 μM) and PD98059 (25 μM) reduced the fluoride-induced IL-8 release by 40%, 59% and 44%, respectively. The combination experiments with either SP600125/ SB202190, SB202190/PD98059 or SP600125/PD98059, seemed almost to abolish the fluoride-induced IL-8 responses. During these experiments the cell viability was carefully monitored, and no changes were observed in the presence of the inhibitors (data not shown). Involvement of SFKs in fluoride-induced IL-8 release Having established that inhibition of ERK, p38 and JNK pathways reduced the fluoride-induced IL-8 release, we investigated the involvement of signaling events potentially located upstream of MAPK activation. Since SFKs have been

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

61

Fig. 4. Effect of MAPK inhibitors, alone and in combination, on fluoride-induced IL-8 release in A549 cells. A549 cells were pre-treated for 1 h with the ERK1/2 inhibitor PD98059, the p38 inhibitor SB202190, the JNK-inhibitor SP600125, alone or in combination, prior to addition of 3.75 mM NaF, and further incubated for 20 h. IL-8 release was measured by ELISA. Each value represents mean ± SEM of 4–12 experiments, and is presented in percentage of the fluoride response in absence of inhibitors. The basal level (mean ± SEM) in absence of inhibitors is 680 ± 110 pg/ml. The data were analyzed by one-way ANOVA, using Bonferroni's post-test. ⁎Significant reduction compared to NaF alone (p ≤ 0.05). +Significant reduction compared to NaF + SB202190 and NaF + SP600125 ( p ≤ 0.05). ¤Significant reduction compared to NaF + SB202190 and NaF + PD9805 (p b 0.05). §Significant reduction compared to NaF + SP600125, and NaF + PD98059 ( p ≤ 0.05).

reported to be important in cytokine release (Kitagawa et al., 2002; Liu et al., 2001; Ovrevik et al., 2004; Trevino et al., 2005), we examined whether SFKs were involved in fluorideinduced IL-8 release from the A549 cells. The SFK-inhibitor PP2 at 10 and 20 μM inhibited the IL-8 release induced by 2.5 mM NaF at 20 h by 50% and 58%, respectively. The basal IL-8 levels were less affected by the inhibitor treatment (Fig. 5). The cell viability in controls and the fluoride-treated cells was not diminished by the inhibitor (not shown), suggesting that the reduction in IL-8 release cannot be attributed to viability changes. The ability of fluoride to stimulate SFKs in A549 cells was examined using a phospho-specific antibody (tyr416) which cross-reacts with several SFK-members (c-Src, Lyn, Lck, Yes

and Hck). Fluoride exposure for 2 and 4 h induced significant phosphorylation of a protein estimated to approximately 61 kDa. The antibody also revealed induction of additional phosphoproteins, mainly a band of approximately 57 kDa protein. A band of approximately 53 kDa occurred, but was only visible in some experiments (data not shown) (Fig. 6A). The phosphorylation levels of 57 and 53 kDa were considerable weaker than the 61 kDa band and were not quantified. The phosphorylation of the SFK bands was strongly reduced by 20 μM PP2 (Fig. 6B). Since the EGFR may interact with SFKs, we examined whether the EGFR antagonist AG1478 affected the phosphorylation of SFK. Pre-treatment of the cells with 10 μM AG1478 seemed to enhance the SFK phosphorylation at 2 h (Fig. 6B). Involvement of SFKs in fluoride-induced MAPK activation SFKs have been shown to be located upstream to MAPK activation in response to various stimuli (Kitagawa et al., 2002; Liu et al., 2001; Ovrevik et al., 2004; Trevino et al., 2005). We therefore examined how activation of fluoride-induced SFKs was related to MAPK activation in the signaling pathway leading to IL-8 release in the A549 cells. PP2 partially reduced the phosphorylation of ERK2, p38 and JNK1/2 in these experiments (Fig. 7). The inhibitor did not affect the basal phosphorylations of the MAPKs (data not shown). Importance of the EGFR phosphorylation in fluoride-induced IL-8 release

Fig. 5. Involvement of SFKs in fluoride-induced IL-8 release in A549 cells. A549 cells were pre-treated with the SFK inhibitor PP2 (10 and 20 μM) for 1 h prior to exposure to NaF (2.5 mM) for 20 h. IL-8 release was measured by ELISA. Each value represents mean ± SEM of 5–7 experiments, and is presented in percentage of the fluoride response in the absence of inhibitors. The basal level (mean ± SEM) in absence of inhibitors was 1057 ± 157 pg/ml. ⁎Significant reduction estimated by one-way ANOVA analysis, using Dunnet's post-test ( p ≤ 0.05).

Phosphorylation/activation of EGFR has been shown to initiate cytokine responses, including IL-8, after exposure to EGF (Richter et al., 2002), but also in response to various other external stimuli via trans-activating mechanisms (Wu et al., 2001; Richter et al., 2002; Adachi et al., 2004). To examine the potential contribution of the EGFR pathway in fluoride-induced IL-8 release, the A549 cells were exposed to the EGFR anta-

62

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

Involvement of EGFR phosphorylation in fluoride-induced MAPK activation Activation of the EGFR is known to be linked to MAPK signal transduction (Chen et al., 2001; Eguchi et al., 2001; Kitagawa et al., 2002; Onan et al., 2004; Purdom and Chen, 2005). To explore the potential role of MAPK signaling subsequent to fluoride-induced EGFR phosphorylation, we examined how AG1478 (10 μM) interfered with phosphorylations of the different MAPKs. The effect of the inhibitor was thus studied on phosphorylation of ERK1/2, p38 and JNK1/2 in the A549 cells after induction by 3.75 mM NaF. The experiments indicated that AG1478 reduced fluoride-induced phosphorylation of ERK and JNK, but less than observed in the presence of PP2. The phosphorylation of p38 was not affected (Fig. 10). AG1478 alone only slightly affected phosphorylation of ERK (data not shown). Discussion

Fig. 6. Phosphorylation of SFKs induced by fluoride in A549 cells: Inhibition with PP2 and AG1478. (A) The cells were exposed to 3.75 mM fluoride for 2 and 4 h. (B) The cells were pre-treated for 1 h with 20 μM PP2 or 10 μM AG1478, and further incubated for 2 h with fluoride. The phosphorylated SFKs compared to the levels of β-actin (i.e. phospho-SFK/β-actin) were detected by Western blotting, and are shown by a typical blot and optical quantification (mean ± SEM) of protein bands from 3–4 experiments. The data in A are presented as fold increase compared to the band intensity (ratio pSFK/β-actin) in the control at 2 h. In B the data are presented relative to phosphorylation levels in cells exposed to NaF alone.

gonist AG1478. AG1478 (10 μM) significantly reduced the IL-8 release at a low fluoride concentration (Fig. 8). Fig. 9 shows how fluoride exposure substantially increased the phosphorylation of the EGFR (∼ 3-fold at 2 h of exposure). In comparison to EGF, phosphorylation of the EGFR induced by fluoride, was much weaker (data not shown). Involvement of SFKs in fluoride-induced EGFR phosphorylation SFKs have been reported to be localized upstream to the EGFR (Belsches et al., 1997; Jorissen et al., 2003). PP2 (20 μM) substantially reduced the fluoride-induced phosphorylation of the EGFR. As expected, the EGFR antagonist, AG1478, nearly abolished the EGFR phosphorylation (Fig. 9).

The present study suggests that fluoride mediates its effect on IL-8 release in A549 cells via mechanisms involving a SFK- and EGFR-dependent pathway, and subsequent activation of MAPK, and in particular JNK and ERK. A SFK-dependent, EGFRindependent pathway seems to be important, and mostly for the phosphorylation of p38 that also is involved in IL-8 release (see proposed model Fig. 11). Experiments with specific inhibitors against different MAPKs indicated a role for phosphorylation of p38, ERK1/2 and JNK1/2 in fluoride-induced IL-8 release (Figs. 2 and 3). The reduction in fluoride-induced IL-8 response obtained by the MAPK inhibition was only partial, suggesting that neither p38, JNK1/2 nor the ERK1/2 pathways mediated the whole IL-8 response. In contrast, the combined inhibition of p38 and JNK1/ 2, p38 and ERK1/2, and JNK1/2 and ERK1/2 seemed all to abolish the fluoride-induced IL-8 release (Fig. 4). This suggests that a combined activation of several MAPK pathways is required for maximal fluoride-induced IL-8 release from the A549 cells. The role of the MAPK activation in IL-8 regulation seems to be highly cell type- and stimulus-specific. Thus, it has been reported that IL-8 may be regulated by p38, ERK1/2 or JNK alone, but also that maximal IL-8 production may require the activation of all the three major MAPK pathways (Li et al., 2002; Hoffmann et al., 2002). Maximal IL-8 amounts have been suggested to be generated by three different mechanisms: first, derepression of the promoter region of the gene; second, transcriptional activation of the promoter by nuclear factor-κB (NFκB) and the activator protein (AP)-1-activating protein JNK; and third, stabilization of the mRNA for IL-8 by the p38 pathway. Also the ERK pathway seems to contribute to IL-8 expression by activating AP-1 elements in the IL-8 promoter region (Hoffmann et al., 2002). In accordance with Hoffmann and coworkers, our study suggests a role both for p38/JNK and p38/ERK in fluoride-induced IL-8 release, showing the requirement for combined activation of two separate pathways for maximal response. The combined effect of ERK- and JNKactivations indicated by the use of PD98059 and SP600125, is more difficult to explain.

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

63

Fig. 7. Phosphorylation of MAPKs induced by fluoride in A549 cells: Inhibition with PP2. The cells were pre-incubated for 1 h in absence and presence of PP2 (20 μM) and further exposed to 3.75 mM NaF for 2 h. The phosphorylation of ERK1/2 (A), p38 (B) and JNK1/2 (C) was detected by Western blotting. A typical blot and optical quantification of the protein bands from 4 to 7 experiments are also presented. Values represent phosphorylation of ERK2, p38, JNK1 and JNK2 compared to the total levels of the respective proteins, and are presented in percentage (mean ± SEM) of fluoride-induced responses.

In the present experiments, the JNK-inhibitor, D-JNKI1, was used in addition to SP600125. D-JNKI1 did not reduce the fluoride-induced IL-8 release as observed for SP600125, but augmented the response (Fig. 2). Presently, there is no explanation for this discrepancy. The two inhibitors are known to act via different mechanisms, but the different JNK-isoforms seem to be equally affected (Bennett et al., 2001; Borsello and Bonny, 2004). The uptake of D-JNKI1 is critical, and could possibly be of importance. In any case, since SP600125 and siRNA against JNK2 both reduced the fluoride-induced IL-8 release (Fig. 3), our data suggest a positive role for JNK1/2 in the cytokine response.

The NFκB pathway is regarded as indispensable for IL-8 regulation. The effect of NFκB activation alone on IL-8 response is, however, normally marginal, but in combination with JNK, ERK and/or p38 activation, NFκB activation permits a strong IL-8 response (Hoffmann et al., 2002). Thus, it is explainable that the combined inhibition of p38/JNK, p38/ERK or ERK/JNK seemed sufficient to reduce the fluoride-induced increase in IL-8 release down to control levels, even if the NFκB pathway also is required and acts in concert with the different MAPKs. We have ongoing studies to address this question. Fluoride exposure induced phosphorylation of SFKs and EGFR in A549 cells, and both SFK- and EGFR inhibitors

64

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

Fig. 8. Involvement of EGFR activation in fluoride-induced IL-8 release from A549 cells. The cells were pre-treated with 10 μM AG1478 for 1 h, and further exposed to 0–3.0 mM NaF for 20 h. IL-8 release was measured by ELISA. Each value represents mean ± SEM of 6 experiments, and is presented as fold increase compared to the basal levels in absence of AG1478. The basal levels (mean ± SEM) were 1030 ± 180 pg/ml. ⁎Significant reduction estimated by 2-ways ANOVA, using Bonferroni's post-test ( p ≤ 0.05).

attenuated fluoride-induced IL-8 release (Figs. 5 and 8), suggesting that the IL-8 response involved SFK- and EGFRdependent pathways. Fluoride both phosphorylated a band of approximately 61 kDa band, likely corresponding to c-src (60 kDa), and two bands with lower molecular weight (approximately 57 and 53 kDa) (Fig. 6), presumably representing Lyn known to have isoforms of 53 and 57 kDa. A549 cells have been shown to abundantly express, c-Src and Lyn, but not Lck and Fyn (Huang et al., 2003). The role of SFKs and EGFR phosphorylation/activation in fluorideinduced IL-8 response is in accordance with studies in A549 cells as well as other cell types, using other toxicants and infectious agents (Liu et al., 2001; Wu et al., 2002; Richter et al., 2002; Ovrevik et al., 2004). Notably, the inhibition of fluoride-induced IL-8 release by PP2 and AG1478 was only partial, suggesting that part of the response may also be mediated by SFK- and EGFR-independent pathways. Our present results imply that fluoride-induced SFK activation is involved in phosphorylation of ERK1/2, p38 and JNK1/2, whereas fluoride-induced EGFR activation is involved in phosphorylation of ERK1/2 and JNK1/2, and to less extent p38 (Figs. 7 and 10). SFK activation by other agents has been observed to mediate its effect on IL-8 release at least partly via the MAPK ERK (Liu et al., 2001; Ovrevik et al., 2004; Trevino et al., 2005), but SFKs have also been reported to be localized upstream to p38 (Trevino et al., 2005). In addition to its role in IL-8 release, SFKs have been localized upstream to ERK (Kitagawa et al., 2002; Gardner et al., 2003; Zhuang and Schnellmann, 2004; Purdom and Chen, 2005), but also to JNK (Chen et al., 2001; Kitagawa et al., 2002; Purdom and Chen, 2005) and p38 (Purdom and Chen, 2005). Most studies of the role of EGFR in MAPK activation have focused on a down-stream activation of ERK and JNK (Chen et al., 2001; Eguchi et al., 2001; Kitagawa et al., 2002; Gardner et al., 2003; Onan et al., 2004; Zhuang and Schnellmann, 2004; Purdom and Chen, 2005). Although p38 is generally not

considered to be regulated by the EGFR, some studies implicate EGFR activation also in the p38 pathway (Eguchi et al., 2001; Onan et al., 2004). Notably, the magnitude of the effects of EGFR inhibition by AG1478 on the fluoride-induced activation of the different MAPKs seems less than the effects of the SFK-inhibitor PP2, suggesting that SFK-dependent, but EGFR-independent pathway may be important. At least a part of the SFKs effect on fluoride-induced IL-8 release appeared to be mediated by the EGFR. A lot of evidence suggests the involvement of SFK proteins, and in particular c-Src, in EGFR signaling (Belsches et al., 1997; Jorissen et al., 2003). Interestingly, c-Src may be both an activator of the EGFR and a transducer of the signals down-stream of the receptor (Jorissen et al., 2003). In accordance with Jorissen and coworkers (2003), our data show a critical role for fluorideinduced SFK activation as an upstream event in EGFR activation. However, as SFK phosphorylation was not reduced by AG1478 pre-treatment, no transducing role was observed for SFKs down-stream to the EGF-receptor, as also observed after exposure to several other stress-inducing agents (Kitagawa et al., 2002; Zhuang and Schnellmann, 2004). SFKs are activated following engagement of several classes of cell surface receptors, including immunoreceptors, integrins and adhesion molecules, cytokine receptors, G protein-coupled receptors (GPCRs) and receptor tyrosin kinases (Thomas and Brugge, 1997). In particular, the GPCRs have been studied as upstream activators of the SFKs (Luttrell et al., 1997), and are considered as important in transactivation of EGFRs, allowing communication between different signaling systems that enables cells to integrate a multitude of signals from its environment (Gschwind et al., 2001). Several studies of others suggest that fluoride exerts its cellular effects via GPCR-dependent pathways. Fluoride-

Fig. 9. Fluoride-induced EGFR phosphorylation and involvement of SFKs. The cells were pre-treated with 20 μM PP2 or 10 μM AG1478 for 1 h, and further incubated with and without 3.75 mM NaF prior to immunoprecipitation. The precipitated proteins were separated by SDS-PAGE and Western blots were incubated with an antibody against phosphotyrosine to investigate EGFR phosphorylation. The figure displays a typical blot and optical quantification (mean ± SEM) of the protein bands from 7 to 9 independent experiments, and is presented relative to the maximal EGFR phosphorylation with NaF.

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

induced G protein activation is mediated via formation of a fluoroaluminate (AlF4−) complex, strongly inhibiting G proteinlinked GTP-ases, inducing persistent stimulation of G proteins (Sternweis and Gilman, 1982). Fluoroaluminate-induced Gprotein activation has been reported to activate SFKs (Jeschke et al., 1998) and PKA, PKC and Ca2+-dependent protein kinases (Blackmore et al., 1985). Alternatively, fluoride may also inhibit tyrosine protein phosphatases (Wergedal and Lau, 1992), leading to increased phosphorylation of tyrosine kinases and activation of down-stream signaling. Thus, GPCRs may potentially be an initial/early step in the fluoride-induced activation of SFKs and EGFRs, and in MAPK-activation and IL-8 release; however, this needs to be studied. A crucial question is how the present cytokine responses in A549 cells, using high fluoride concentrations, are related to fluoride-induced inflammatory responses in the respiratory tract

65

in vivo. An important explanation to the high concentrations is that the A549 cell line, commonly used as epithelial lung cell model, is known for its relative insensitivity to many agents. However, when using primary type 2 pneumocytes, fluoride induced a cytokine response (MIP-2) at 5- to 10-fold lower concentrations than A549 cells (Fig. 1). Furthermore, in the respiratory tract several, cell types are known to interact in a complex network, increasing the sensitivity to different agents as reflected by an increased sensitivity in co-cultures (Wottrich et al., 2004). In a co-culture system of epithelial cells and macrophages, the sensitivity to fluoride would presumably be further increased compared to epithelial cells alone. The remaining question is whether these concentrations still are high compared to concentrations that occur in the in vivo situation. We have previously shown that exposure of volunteers to realistic concentrations of HF induces inflammatory reactions in

Fig. 10. Involvement of EGFR phosphorylation in fluoride-induced MAPK activation. The cells were pre-treated with 10 μM AG1478 for 1 h, and further incubated for 2 h in the presence of 3.75 mM NaF. The phosphorylation of ERK1/2, p38, JNK1/2 and their respective total proteins was analyzed by Western blotting. The figure displays a typical blot and optical quantification (mean ± SEM) of the protein bands from 4 to 7 independent experiments. Values are presented as relative phosphorylation of the ERK2, p38, JNK1 and JNK2 compared to the total levels of their respective proteins, and are presented in percentage of the NaF-induced responses without AG1478.

66

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67

Fig. 11. Model of pathways involved in IL-8 release. Proposed model based on the present results, describing the relationships between fluoride-induced phosphorylation of SFKs, EGFR and MAPKs, and the involvement in IL-8 production in the A549 cells.

the respiratory tract (Lund et al., 1999, 2002). The concentrations of fluoride in the blood were up to five-fold increased during this exposure (Lund et al., 1997). However, it is conceivable that fluoride concentrations will be more elevated in limited areas of the lung such as bifurcations, generating concentrations in the epithelial cell lining fluid that are transiently higher than observed in blood plasma. Furthermore, in the pot rooms of aluminum melters, fluoride attached to particles may favor even higher fluoride concentrations at the localized sites for particle deposition in the respiratory tract. The arguments above indicate that the concentrations used might give important information relevant to the in vivo situation. Furthermore, since fluoride exposure is capable of inducing inflammation and cytokine responses in volunteers, mechanistic studies of fluoride-induced cytokine release in lung cells might give useful knowledge. Even if the regulation of IL-8 is known to be cellspecific, the present data in A549 cells presumably are of relevance for the more sensitive primary type 2 cells. In support of this, our studies with exposure of A549 cells to quartz particles have shown similarities to the regulation pattern for quartz-induced cytokine synthesis in rat primary type 2 cells. In both cell types, MAP-kinases and SFKs were shown to be involved in the cytokine release (Ovrevik et al., 2004). In summary, this study extends our previous findings that MAP-kinase activation is involved in fluoride-induced IL-8 release in A549 cells, showing that a combined activation of MAPK ERK1/2, p38 and JNK1/2 is essential. Furthermore, the study shows that fluoride activates SFKs in the A549 cells, as previously reported in other cell types. Novel findings are that fluoride induces EGF-receptor phosphorylation, and a role of both SFKs and EGF-receptors in fluoride-induced IL-8 release. The present study also shows that fluoride-induced IL-8 re-

sponse at least partly is mediated via SFK-/EGFR-dependent activation of ERK1/2 and JNK1/2, and to a less extent p38. In concert with this, a SFK-dependent/EGFR-independent mechanism seems to be important for fluoride-induced IL-8 release, mainly via a p38-mediated mechanism. The relative contribution of these different signaling pathways in fluoride-induced IL-8 release has, however, not been quantified. We present a model for the main signaling pathways in fluoride-induced IL-8 response in the A549 cells, based on the findings in the present study, as suggested in Fig. 11. The most important aspect with the present study is an increased understanding of initial events involved in fluoride-induced IL-8 release in human epithelial lung cells. References Adachi, T., Cui, C.H., Kanda, A., Kayaba, H., Ohta, K., Chihara, J., 2004. Activation of epidermal growth factor receptor via CCR3 in bronchial epithelial cells. Biochem. Biophys. Res. Commun. 320, 292–296. Anderson, N.G., Kilgour, E., Sturgill, T.W., 1991. Activation of mitogen-activated protein kinase in BC3H1 myocytes by fluoroaluminate. J. Biol. Chem. 266, 10131–10135. Belsches, A.P., Haskell, M.D., Parsons, S.J., 1997. Role of c-Src tyrosine kinase in EGF-induced mitogenesis. Front. Biosci. 2, d501–d518. Bennett, B.L., Sasaki, D.T., Murray, B.W., O'Leary, E.C., Sakata, S.T., Xu, W., Leisten, J.C., Motiwala, A., Pierce, S., Satoh, Y., Bhagwat, S.S., Manning, A.M., Anderson, D.W., 2001. SP600125, an anthrapyrazolone inhibitor of Jun N-terminal kinase. Proc. Natl. Acad. Sci. U. S. A. 98, 13681–13686. Blackmore, P.F., Bocckino, S.B., Waynick, L.E., Exton, J.H., 1985. Role of a guanine nucleotide-binding regulatory protein in the hydrolysis of hepatocyte phosphatidylinositol 4,5-bisphosphate by calcium-mobilizing hormones and the control of cell calcium. Studies utilizing aluminum fluoride. J. Biol. Chem. 260, 14477–14483. Borsello, T., Bonny, C., 2004. Use of cell-permeable peptides to prevent neuronal degeneration. Trends Mol. Med. 10, 239–244.

M. Refsnes et al. / Toxicology and Applied Pharmacology 227 (2008) 56–67 Chen, K., Vita, J.A., Berk, B.C., Keaney Jr., J.F., 2001. c-Jun N-terminal kinase activation by hydrogen peroxide in endothelial cells involves SRC-dependent epidermal growth factor receptor transactivation. J. Biol. Chem. 276, 16045–16050. Eguchi, S., Dempsey, P.J., Frank, G.D., Motley, E.D., Inagami, T., 2001. Activation of MAPKs by angiotensin II in vascular smooth muscle cells. Metalloprotease-dependent EGF receptor activation is required for activation of ERK and p38 MAPK but not for JNK. J. Biol. Chem. 276, 7957–7962. Furuichi, S., Hashimoto, S., Gon, Y., Matsumoto, K., Horie, T., 2002. p38 mitogen-activated protein kinase and c-Jun-NH2-terminal kinase regulate interleukin-8 and RANTES production in hyperosmolarity stimulated human bronchial epithelial cells. Respirology 7, 193–200. Gardner, O.S., Dewar, B.J., Earp, H.S., Samet, J.M., Graves, L.M., 2003. Dependence of peroxisome proliferator-activated receptor ligand-induced mitogen-activated protein kinase signaling on epidermal growth factor receptor transactivation. J. Biol. Chem. 278, 46261–46269. Gschwind, A., Zwick, E., Prenzel, N., Leserer, M., Ullrich, A., 2001. Cell communication networks: epidermal growth factor receptor transactivation as the paradigm for interreceptor signal transmission. Oncogene 20, 1594–1600. Hashimoto, S., Gon, Y., Takeshita, I., Matsumoto, K., Jibiki, I., Takizawa, H., Kudoh, S., Horie, T., 2000. Diesel exhaust particles activate p38 MAP kinase to produce interleukin 8 and RANTES by human bronchial epithelial cells and N-acetylcysteine attenuates p38 MAP kinase activation. Am. J. Respir. Crit. Care Med. 161, 280–285. Hoffmann, E., Dittrich-Breiholz, O., Holtmann, H., Kracht, M., 2002. Multiple control of interleukin-8 gene expression. J. Leukoc. Biol. 72, 847–855. Huang, W.C., Chen, J.J., Chen, C.C., 2003. c-Src-dependent tyrosine phosphorylation of IKKbeta is involved in tumor necrosis factor-alpha-induced intercellular adhesion molecule-1 expression. J. Biol. Chem. 278, 9944–9952. Jeschke, M., Standke, G.J., Scaronuscarona, M., 1998. Fluoroaluminate induces activation and association of Src and Pyk2 tyrosine kinases in osteoblastic MC3T3-E1 cells. J. Biol. Chem. 273, 11354–11361. Jorissen, R.N., Walker, F., Pouliot, N., Garrett, T.P., Ward, C.W., Burgess, A.W., 2003. Epidermal growth factor receptor: mechanisms of activation and signalling. Exp. Cell Res. 284, 31–53. Kitagawa, D., Tanemura, S., Ohata, S., Shimizu, N., Seo, J., Nishitai, G., Watanabe, T., Nakagawa, K., Kishimoto, H., Wada, T., Tezuka, T., Yamamoto, T., Nishina, H., Katada, T., 2002. Activation of extracellular signal-regulated kinase by ultraviolet is mediated through Src-dependent epidermal growth factor receptor phosphorylation. Its implication in an anti-apoptotic function. J. Biol. Chem. 277, 366–371. Lag, M., Becher, R., Samuelsen, J.T., Wiger, R., Refsnes, M., Huitfeldt, H.S., Schwarze, P.E., 1996. Expression of CYP2B1 in freshly isolated and proliferating cultures of epithelial rat lung cells. Exp. Lung Res. 22, 627–649. Li, J., Kartha, S., Iasvovskaia, S., Tan, A., Bhat, R.K., Manaligod, J.M., Page, K., Brasier, A.R., Hershenson, M.B., 2002. Regulation of human airway epithelial cell IL-8 expression by MAP kinases. Am. J. Physiol., Lung Cell. Mol. Physiol. 283, L690–L699. Liu, R., Aupperle, K., Terkeltaub, R., 2001. Src family protein tyrosine kinase signaling mediates monosodium urate crystal-induced IL-8 expression by monocytic THP-1 cells. J. Leukoc. Biol. 70, 961–968. Lund, K., Ekstrand, J., Boe, J., Sostrand, P., Kongerud, J., 1997. Exposure to hydrogen fluoride: an experimental study in humans of concentrations of fluoride in plasma, symptoms, and lung function. Occup. Environ. Med. 54, 32–37. Lund, K., Refsnes, M., Sandstrom, T., Sostrand, P., Schwarze, P., Boe, J., Kongerud, J., 1999. Increased CD3 positive cells in bronchoalveolar lavage fluid after hydrogen fluoride inhalation. Scand. J. Work Environ. Health 25, 326–334. Lund, K., Refsnes, M., Ramis, I., Dunster, C., Boe, J., Schwarze, P.E., Skovlund, E., Kelly, F.J., Kongerud, J., 2002. Human exposure to hydrogen fluoride induces acute neutrophilic, eicosanoid, and antioxidant changes in nasal lavage fluid. Inhal. Toxicol. 14, 119–132. Luttrell, L.M., Della Rocca, G.J., van Biesen, T., Luttrell, D.K., Lefkowitz, R.J., 1997. Gbetagamma subunits mediate Src-dependent phosphorylation of the epidermal growth factor receptor. A scaffold for G protein-coupled receptormediated Ras activation. J. Biol. Chem. 272, 4637–4644.

67

Mukaida, N., 2003. Pathophysiological roles of interleukin-8/CXCL8 in pulmonary diseases. Am. J. Physiol., Lung Cell. Mol. Physiol. 284, L566–L577. Murthy, K.S., Makhlouf, G.M., 1994. Fluoride activates G protein-dependent and -independent pathways in dispersed intestinal smooth muscle cells. Biochem. Biophys. Res. Commun. 202, 1681–1687. Onan, D., Pipolo, L., Yang, E., Hannan, R.D., Thomas, W.G., 2004. Urotensin II promotes hypertrophy of cardiac myocytes via mitogen-activated protein kinases. Mol. Endocrinol. 18, 2344–2354. Ovrevik, J., Lag, M., Schwarze, P., Refsnes, M., 2004. p38 and Src-ERK1/2 pathways regulate crystalline silica-induced chemokine release in pulmonary epithelial cells. Toxicol. Sci. 81, 480–490. Puddicombe, S.M., Davies, D.E., 2000. The role of MAP kinases in intracellular signal transduction in bronchial epithelium. Clin. Exp. Allergy 30, 7–11. Purdom, S., Chen, Q.M., 2005. Epidermal growth factor receptor-dependent and -independent pathways in hydrogen peroxide-induced mitogen-activated protein kinase activation in cardiomyocytes and heart fibroblasts. J. Pharmacol. Exp. Ther. 312, 1179–1186. Refsnes, M., Becher, R., Lag, M., Skuland, T., Schwarze, P.E., 1999. Fluorideinduced interleukin-6 and interleukin-8 synthesis in human epithelial lung cells. Hum. Exp. Toxicol. 18, 645–652. Refsnes, M., Thrane, E.V., Lag, M., Thoresen, G.H., Schwarze, P.E., 2001. Mechanisms in fluoride-induced interleukin-8 synthesis in human lung epithelial cells. Toxicology 167, 145–158. Refsnes, M., Schwarze, P.E., Holme, J.A., Lag, M., 2003. Fluoride-induced apoptosis in human epithelial lung cells (A549 cells): role of different G protein-linked signal systems. Hum. Exp. Toxicol. 22, 111–123. Richter, A., O'Donnell, R.A., Powell, R.M., Sanders, M.W., Holgate, S.T., Djukanovic, R., Davies, D.E., 2002. Autocrine ligands for the epidermal growth factor receptor mediate interleukin-8 release from bronchial epithelial cells in response to cigarette smoke. Am. J. Respir. Cell Mol. Biol. 27, 85–90. Soyseth, V., Kongerud, J., 1992. Prevalence of respiratory disorders among aluminium potroom workers in relation to exposure to fluoride. Br. J. Ind. Med. 49, 125–130. Sternweis, P.C., Gilman, A.G., 1982. Aluminum: a requirement for activation of the regulatory component of adenylate cyclase by fluoride. Proc. Natl. Acad. Sci. U. S. A. 79, 4888–4891. Susa, M., Standke, G.J., Jeschke, M., Rohner, D., 1997. Fluoroaluminate induces pertussis toxin-sensitive protein phosphorylation: differences in MC3T3-E1 osteoblastic and NIH3T3 fibroblastic cells. Biochem. Biophys. Res. Commun. 235, 680–684. Thomas, S.M., Brugge, J.S., 1997. Cellular functions regulated by Src family kinases. Annu. Rev. Cell Dev. Biol. 13, 513–609. Thrane, E.V., Refsnes, M., Thoresen, G.H., Lag, M., Schwarze, P.E., 2001. Fluoride-induced apoptosis in epithelial lung cells involves activation of MAP kinases p38 and possibly JNK. Toxicol. Sci. 61, 83–91. Trevino, J.G., Summy, J.M., Gray, M.J., Nilsson, M.B., Lesslie, D.P., Baker, C.H., Gallick, G.E., 2005. Expression and activity of SRC regulate interleukin-8 expression in pancreatic adenocarcinoma cells: implications for angiogenesis. Cancer Res. 65, 7214–7222. Wergedal, J.E., Lau, K.H., 1992. Human bone cells contain a fluoride sensitive acid phosphatase: evidence that this enzyme functions at neutral pH as a phosphotyrosyl protein phosphatase. Clin. Biochem. 25, 47–53. Wottrich, R., Diabate, S., Krug, H.F., 2004. Biological effects of ultrafine model particles in human macrophages and epithelial cells in mono- and co-culture. Int. J. Hyg. Environ. Health 207, 353–361. Wu, W., Samet, J.M., Ghio, A.J., Devlin, R.B., 2001. Activation of the EGF receptor signaling pathway in airway epithelial cells exposed to Utah Valley PM. Am. J. Physiol., Lung Cell. Mol. Physiol. 281, L483–L489. Wu, W., Graves, L.M., Gill, G.N., Parsons, S.J., Samet, J.M., 2002. Srcdependent phosphorylation of the epidermal growth factor receptor on tyrosine 845 is required for zinc-induced Ras activation. J. Biol. Chem. 277, 24252–24257. Zhuang, S., Schnellmann, R.G., 2004. H2O2-induced transactivation of EGF receptor requires Src and mediates ERK1/2, but not Akt, activation in renal cells. Am. J. Physiol. Renal Physiol. 286, F858–F865.