Benzo[a]pyrene up-regulates cyclooxygenase-2 ... - Semantic Scholar

carc$$0317

Carcinogenesis vol.18 no.4 pp.795–799, 1997

Benzo[a]pyrene up-regulates cyclooxygenase-2 gene expression in oral epithelial cells

Daniel J.Kelley2, Juan R.Mestre2, Kotha Subbaramaiah1, Peter G.Sacks2, Stimson P.Schantz2, Tadashi Tanabe3, Hiroyasu Inoue3, John T.Ramonetti1 and Andrew J.Dannenberg1,4 1Department

of Medicine, Cornell University Medical College and Strang Cancer Prevention Center, New York, NY 10021, 2Head and Neck Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY 10021, USA and 3Department of Pharmacology, National Cardiovascular Research Institute, Osaka 565, Japan 4To

whom correspondence should be addressed at: Division of Digestive Diseases, Room F-231, The New York Hospital-Cornell Medical Center, New York, NY 10021, USA

Cyclooxygenase may be important in the pathogenesis of smoking-related cancer because it activates carcinogens and catalyzes prostaglandin biosynthesis. We determined the effects of benzo[a]pyrene (B[a]P), a polycyclic aromatic hydrocarbon in tobacco smoke, on cyclooxygenase-2 (Cox2) mRNA, protein and synthesis of prostaglandin E2 (PGE2) in normal and transformed oral epithelial cells. Treatment with B[a]P caused a dose-dependent increase in production of PGE2, with a maximal increase of ~100%. Enhanced synthesis of PGE2 was associated with increased amounts of Cox-2 protein. B[a]P also caused a two-fold increase in Cox-2 mRNA in both normal and transformed cells. Transient transfections with a Cox-2 promoter construct showed that B[a]P-mediated induction of Cox-2 mRNA reflected increased transcription. Levels of Cox-1 were unaffected by B[a]P. B[e]P did not affect the synthesis of PGE2 or amounts of Cox-2. These data are important because B[a]P-mediated induction of Cox-2 may predispose to carcinogenesis by enhancing the production of mutagens and the synthesis of prostaglandins.

Introduction Exposure to polycyclic aromatic hydrocarbons is a recognized and important cause of cancer because these otherwise inert chemicals are metabolized by selected enzymes, including P450s and cyclooxygenase (Cox*, PGH synthase), to carcinogenic metabolites. For example, benzo[a]pyrene (B[a]P), a component of cigarette smoke, undergoes metabolic biotransformation to electrophilic intermediates that react with cellular macromolecules, including DNA and protein. The extent to which a given exposure to carcinogens produces adducts between cellular components and reactive metabolites in vivo depends on the balance between rates of oxidation of the parent compound and rates of detoxification of reactive products via conjugation (1). Thus, induction of mono-oxygenase enzymes *Abbreviations: Cox, cyclooxygenase-2; PGE2, prostaglandin E2; B[a]P, benzo[a]pyrene; B[e]P, benzo[e]pyrene; FBS, fetal bovine serum; LDH, lactate dehydrogenase; NSAID, non-steroidal anti-inflammatory drug; SDS, sodium dodecyl sulfate; SSPE, saline–sodium phosphate–EDTA buffer; XRE, xenobiotic responsive element. © Oxford University Press

(2) or inhibition of conjugating enzymes (1,3) enhance the covalent binding of B[a]P to DNA. The cytochrome P-450 family of enzymes is widely recognized for catalyzing oxidative reactions and activating xenobiotics to reactive electrophiles that are carcinogenic. The peroxidative oxidation of chemicals that occurs during arachidonic acid metabolism catalyzed by Cox also bioactivates chemical carcinogens. Cox catalyzes the oxidation of B[a]P7,8-dihydrodiol to B[a]P-diolepoxide, which is a highly reactive and strongly mutagenic carcinogen (4–8). The Cox enzyme system is especially important in bioactivating chemicals in extrahepatic tissues, where the P-450 mono-oxygenase system has low activity. Recently, an inducible isoform of Cox designated Cox-2, was identified (9,10). It was shown subsequently that the synthesis of Cox-2 is stimulated by phorbol esters, cytokines and growth factors (11,12). In contrast, the constitutive isoform, Cox-1, is unaffected by these factors. By analogy with the inducing effect of B[a]P on selected P-450s, we wondered whether B[a]P could also induce Cox-2. Thus, we describe in the present report the effects of B[a]P on levels of Cox-2 and prostaglandin synthesis in normal and transformed oral epithelial cells. Our results indicate that B[a]P up-regulates Cox-2 and the synthesis of prostaglandins. Materials and methods Materials Keratinocyte growth media (KGM) was obtained from Clonetics Corp. (San Diego, CA). DMEM/F-12, Amniomax-C100 and FBS were from Life Technologies, Inc. (Grand Island, NY). Enzyme immunoassay reagents for assays of PGE2 were from Cayman Co. (Ann Arbor, MI). Random-priming kits were from Boehringer–Mannheim Biochemicals (Indianapolis, IN). Nitrocellulose membranes were from Schleicher and Schuell (Keene, NH). pFx-3, the cationic lipid used for transfections, was from Invitrogen (San Diego, CA). Reagents for the luciferase assay were from Analytical Luminescence (San Diego, CA). B[a]P, benzo[e]pyrene (B[e]P) and 7,12-dimethylbenz[a]anthracene, lactate dehydrogenase diagnostic kits and trypan blue were from Sigma Chemical Co. (St Louis, MO). Normal oral epithelial cells were grown in Primaria culture dishes from Falcon, Beckton, Dickinson (Lincoln Park, NJ). Tissue culture Squamous carcinoma cells (1483) have been described previously (13). Cells were maintained in a 1:1 mixture of DMEM/F12 supplemented with 10% FBS and 50 µg/ml gentamicin. Cells were grown to 70–80% confluency, trypsinized with 0.05% trypsin–2 mM EDTA, and plated for experimental use as described below. Normal oral epithelial cells were cultured using explant outgrowth techniques (14). Briefly, mucosal epithelium was dissected from specimens obtained from patients undergoing tonsillectomy for benign disease. Mucosa was cut into 1–3 mm sized pieces and explanted into 60-mm Petri dishes in Amniomax-C100 medium supplemented with 50 µg/ml gentamicin. Epithelial cells were allowed to outgrow for 10–12 days in a 5% CO2–water saturated incubator at 37°C. Following outgrowth, epithelial cells were dissociated with 0.05% trypsin–2 mM EDTA and plated into 100-mm Primaria dishes and Primaria 6-well plates in serum-free KGM. All experiments were performed on first passage cells. Cells were allowed to attach for 24 h prior to experiments with 100-mm dishes or 6-well plates, which were run in parallel under identical conditions.

795

D.J.Kelley et al.

Fig. 4. Benzo[a]pyrene does not affect amounts of Cox-1. 1483 cells were treated with vehicle (0.01% DMSO, lane 1) or B[a]P (1000 ng/ml, lane 2) for 4.5 h. Cellular lysate protein (100 µg/lane) was loaded onto a 10% sodium dodecyl sulfate-polyacrylamide gel, electrophoresed and subsequently transferred onto nitrocellulose. The immunoblot was probed with antibody specific for Cox-1. Densitometry was performed and the results expressed in arbitrary units (a.u.): lane 1, 533 a.u.; lane 2, 491 a.u.

Fig. 1. Benzo[a]pyrene induces the synthesis of PGE2 by oral epithelial cells. 1483 cells were treated with vehicle (0.01% DMSO) or B[a]P (125– 1000 ng/ml) for 6 h. Culture medium was removed and then assayed for PGE2 by enzyme immunoassay. Values are means 6 SD; n 5 6. **P , 0.01.

Fig. 2. Benzo[a]pyrene induces Cox-2 in oral epithelial cells. 1483 cells were treated with vehicle (0.01% DMSO, lane 2) or B[a]P (125, 250, 500, 1000 ng/ml; lanes 3–6) for 4.5 h. Lane 1 represents ovine Cox-2 which was used as a standard. Cellular lysate protein (25 µg/lane) was loaded onto a 10% sodium dodecyl sulfate-polyacrylamide gel, electrophoresed and subsequently transferred onto nitrocellulose. The nitrocellulose membrane was probed with antibody specific for Cox-2. Results of densitometry in arbitrary units were as follows: lane 2, 23 6 5; lane 3, 41 6 17; lane 4, 57 6 13a; lane 5, 58 6 7b; lane 6, 71 6 6c. Values are means 6 SD; n 5 3. aP 5 0.01 compared with control, bP , 0.01 compared with control, cP , 0.001 compared with control.

Fig. 3. Benzo[e]pyrene does not induce Cox-2 in oral epithelial cells. 1483 cells were treated with vehicle (0.01% DMSO, lane 2), B[a]P (1000 ng/ml, lane 3) or B[e]P (1000 ng/ml, lane 4) for 4.5 h. Lane 1 represents ovine Cox-2, which was used as a standard. A 25 µg aliquot of cellular lysate protein was used in each lane. The nitrocellulose membrane was probed with anti-Cox-2 antiserum. Results of densitometry in arbitrary units (a.u.) were as follows: lane 2, 44 a.u.; lane 3, 102 a.u.; lane 4, 46 a.u.

Cellular cytotoxicity was assessed by release of LDH and trypan blue exclusion. Levels of LDH were measured in the supernatants used for PGE2 analyses. LDH assays were performed according to the manufacturer’s instructions. For trypan blue analysis, following treatment with B[a]P for 3– 6 h, cells were trypsinized and combined 1:1 with 0.4% trypan blue and examined for dye exclusion. PGE2 Production Cells were plated at a density of 2.53105 cells/well in 6-well plates and allowed to attach for 24 h prior to experiments. Medium was obtained from each well and centrifuged to sediment cell debris; the supernatants were frozen at –80°C until assay. Supernatants were assayed for spontaneously released PGE2 by enzyme immunoassay (15). To normalize the PGE2 data, cell number was measured with an electronic particle counter (Coulter, Hialeah, FL).

796

Western blotting Cells were plated in 100-mm Petri dishes at 1.5–2.03106 cells/dish and allowed to attach for 24 h prior to experiments. Cell lysates were prepared by treating cells with lysis buffer consisting of 150 mM NaCl, 100 mM Trisbuffered saline, 1% Tween-20, 50 mM diethyldithiocarbamate, 1 mM EDTA and 1 mM phenylmethylsulfonyl fluoride. Lysates were sonicated twice for 20 s and centrifuged at 10 000 g for 10 min to sediment the particulate material. Protein concentration was measured by the method of Lowry et al. (16) and the supernatant stored at –80°C until used. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis was performed under reducing conditions on 10% polyacrylamide gels as described by Laemmli (17). The resolved proteins were transferred onto nitrocellulose sheets as detailed by Towbin et al. (18). The nitrocellulose membrane was then incubated with a rabbit polyclonal antibody raised against the unique 18-amino acid sequence from the carboxy-terminal portion of Cox-2 that does not react with Cox-1 (19). Nitrocellulose membranes were also probed with a monoclonal antiCox-1 antibody (19). The membrane was subsequently probed with a goat anti-rabbit or a goat anti-mouse antibody conjugated to alkaline phosphatase as described previously (20). A computer densitometer (Molecular Dynamics, Sunnyvale, CA) was used to determine the density of the bands. Northern blotting Cells were plated in 100-mm Petri dishes at a density of 1.5–2.03106 cells/ dish and allowed to attach for 24 h prior to experiments. To prepare total cellular RNA, cell monolayers were washed and then directly lysed in 4 mol/ l guanidinium isothiocyanate solution. RNA was then isolated by phenol– chloroform extraction according to Chomczynski and Sacchi (21). For Northern blots, 15 µg of total cellular RNA per lane was electrophoresed in formaldehyde-containing 1.2% agarose gels and transferred to nylon-supported membranes. After baking, membranes were prehybridized for 3 h and then hybridized in a solution containing 50% formamide, 53 saline–sodium phosphate–EDTA buffer (SSPE), 53 Denhardt’s solution, 0.1% SDS, and 100 µg/ml single-stranded salmon sperm DNA. Hybridization was carried out for 16 h at 42°C with a human Cox-2 cDNA probe. To verify equivalency of RNA loading in the different lanes, the blot was stripped of radioactivity and rehybridized to determine levels of 18S rRNA. Cox-2 and 18S rRNA probes were labeled with [32P]dCTP by random priming. After hybridization, membranes were washed for 20 min at room temperature in 23 SSPE-0.1% SDS, twice for 20 min in the same solution at 55°C, and twice for 20 min in 0.13 SSPE-0.1% SDS at 55°C. Washed membranes were then subjected to autoradiography. The signal level of the bands was quantified by densitometry. Transient transfection assays 1483 cells were seeded at a density of 83105 cells/well in 6-well dishes and grown to 30–40% confluence in DMEM/F-12 containing 10% FBS. For each well, 2 µg of plasmid DNA containing 1432 bases of 59-flanking region of Cox-2 ligated to luciferase, was transfected (22). Transfections were done with pFx-3 at a 1:12 ratio of DNA to lipid in Opti-MEM for 12 h as per the manufacturer’s instructions. Subsequently, cells were treated with medium containing different concentrations of B[a]P (0–1000 ng/ml). Luciferase activity was measured in cellular extract 24 h later. Luciferase activity was measured as follows. Each well was washed twice with PBS. 200 µl of 13 lysis buffer (Analytical Luminescence Laboratories, San Diego, CA) was added to each well for 30 min. Lysate was centrifuged for 5 min at 4°C. The supernatant was used to assay luciferase activity and the protein concentration. Luciferase activity was measured using a Monolight 2010 Luminometer (Analytical Luminescence Laboratories, San Diego, CA) according to the manufacturer’s instructions. Luciferase activities were normalized to protein concentrations. Statistics Comparisons between groups were made by the Student’s t-test. A difference between groups of P , 0.05 was considered significant.

B[a]P upregulation of cyclo-oxygenase gene expression

Fig. 5. Benzo[a]pyrene induces amounts of Cox-2 mRNA in oral epithelial cells. In panel (A) normal oral epithelial cells were treated with vehicle (0.01% DMSO) or B[a]P (125, 250, 500, 1000 ng/ml) for 3 h. In panel (B) 1483 cells were treated with vehicle (0.01% DMSO, lane 1) or B[a]P (1000 ng/ml, lane 2) for 3 h. Total cellular RNA was isolated; 15 µg of RNA was added to each lane. Northern blots were probed sequentially with probes that recognized Cox2 mRNA and 18S rRNA, respectively. Results of densitometry in arbitrary units were as follows. Panel (A): 0 ng/ml, 177 6 30; 125 ng/ml, 297 6 40a; 250 ng/ml, 364 6 62b; 500 ng/ml, 371 6 50b; 1000 ng/ml, 477 6 82b. Panel (B): lane 1, 599 6 69; lane 2, 1842 6 639c. Values are means 6 SD; n 5 3. aP 5 0.01 compared with control; bP , 0.01 compared with control; cP , 0.05 compared with control.

Table I. Benzo[e]pyrene does not enhance production of PGE2 by oral epithelial cells Treatment group

Production of PGE2 (pg/cell)

Control Benzo[a]pyrene Benzo[e]pyrene 7,12-Dimethylbenz[a]anthracene

0.019 0.036 0.025 0.023

6 6 6 6

0.003 0.005* 0.003 0.005

1483 cells were treated with vehicle (0.01% DMSO) or 1000 ng/ml B[a]P, B[e]P or 7,12-dimethylbenz[a]anthracene for 6 h. Culture medium was removed and assayed for PGE2 by enzyme immunoassay. Values are means 6 SD; n 5 6. *P , 0.01.

Fig. 6. Benzo[a]pyrene enhances Cox-2 promoter activity. 1483 cells were transfected with 2 µg of a Cox-2 promoter construct containing 1432 bases 59 of the transcription start site of Cox-2 ligated to luciferase. Cells were then treated with B[a]P at doses ranging from 0–1000 ng/ml. Six wells were used for each transfection condition. Columns, mean; bars, SD. *P , 0.01; **P 5 0.01.

Results Effect of B[a]P on PGE2 synthesis Induction of Cox by B[a]P can be assessed in a variety of ways including changes in the synthesis of PGE2. Using this assay, B[a]P caused a dose-dependent increase in synthesis of PGE2 in transformed oral epithelial cells (Figure 1). Treatment with B[a]P at a dose of 1000 ng/ml resulted in about an 80% increase in production of PGE2; higher doses of B[a]P did not cause further increases in prostaglandin synthesis. Similar effects of B[a]P were observed after treatment of cells for 6 or 12 h. Normal oral epithelial cells also were used to investigate the effect of B[a]P on synthesis of PGE2. Treatment with B[a]P (1000 ng/ml) led to approximately a 100% increase in PGE2 synthesis in these cells; higher concentrations of B[a]P did not cause further increases in synthesis of PGE2 (data not shown). In contrast to the inducing effects of B[a]P, B[e]P and 7,12-dimethylbenz[a]anthracene did not cause a significant increase in the production of PGE2 (Table I). Measurements of prostaglandin production were done when cells were equally confluent because differences in cell proliferation affect the release of arachidonic acid (23). In these

experiments, cytotoxicity was assessed by cell number, LDH release and trypan blue exclusion. No evidence of cellular toxicity was detected (data not shown). B[a]P induces Cox-2 in oral epithelial cells The possibility that the above differences in production of PGE2 reflected at least, in part, differences in levels of Cox was evaluated by Western blotting of cell lysate protein. As shown in Figure 2, B[a]P caused a dose-dependent increase in amounts of Cox-2. In contrast to the inducing effects of B[a]P, B[e]P had no effect on levels of Cox-2 (Figure 3), consistent with the inability of this compound to enhance the production of PGE2. Levels of Cox-1 were unaffected by B[a]P (Figure 4). Since B[a]P-mediated induction of Cox-2 enzyme could reflect either increased protein synthesis or decreased degradation, Northern blotting of Cox-2 mRNA was carried out. In representative experiments shown in Figure 5, higher levels of Cox-2 mRNA were detected in normal and transformed cells after treatment with B[a]P. In fact, B[a]P caused a dosedependent increase in levels of Cox-2 mRNA. B[a]P at a dose of 1000 ng/ml led to about a two-fold increase in amounts of Cox-2 mRNA in both types of cells. Differences in levels of mRNA could reflect altered rates of transcription or mRNA stability. To distinguish between these two possibilities, transient transfections were conducted using a human Cox-2 luciferase reporter construct. As shown in Figure 6, B[a]P caused a doubling of luciferase activity consistent with increased transcription. Discussion In this study, we showed that B[a]P induced Cox-2 in oral epithelial cells. This observation is important for a variety of 797

D.J.Kelley et al.

reasons. To begin with, in extrahepatic tissues such as those of the head and neck in which P450 content is low (24), Cox is likely to be important for carcinogen metabolism. Cox converts a broad array of carcinogens to reactive metabolites, which can form DNA adducts (4–8). Thus, several classes of chemical carcinogens, e.g. aromatic amines, heterocyclic amines and dihydrodiol derivatives of polycyclic aromatic hydrocarbons, are activated to mutagenic derivatives by Cox. It is possible, therefore, that B[a]P-mediated induction of Cox2 will amplify the effect of a given dose of B[a]P on tumor initiation. The potential importance of this idea is underscored by the recent report that B[a]P-diolepoxide, a mutagen formed by Cox, causes adducts along exons of the p53 gene that correspond to p53 mutational hot spots in human lung cancer (25). Another interesting possibility is that up-regulation of Cox-2 will enhance the activation of tobacco-specific nitrosamines such as NNN or NNK (26). Aside from potentially increasing the activation of carcinogens, B[a]P-mediated induction of Cox-2 could predispose to carcinogenesis via other mechanisms. The oxidative function of Cox produces prostaglandins that mediate inflammation, that down-regulate immune surveillance of mutant cells (27–29), and that appear to modulate apoptosis (30,31). For example, PGE2 frequently acts as a feedback inhibitor for cellular immune processes, such as T-cell proliferation, lymphokine production and cytotoxicity. Overexpression of Cox-2 in epithelial cells inhibits apoptosis (30). Possibly then, B[a]P-mediated up-regulation of Cox-2 will inhibit apoptosis. Although this study was performed in oral epithelial cells, it is important to recognize that induction of Cox-2 by B[a]P could contribute to tobacco-related cancer in other tissues such as the lung. Prior to the present study, Levine reported that B[a]P caused a 13.6-fold increase in prostaglandin production in canine kidney cells (32,33). Increased synthesis of prostaglandins was attributed at least, in part, to induction of Cox activity (33). These investigators also showed that B[a]P may activate phospholipase A2 in addition to Cox (33), an effect that could contribute to increased production of prostaglandins. We have extended upon these initial observations by showing that B[a]P increases Cox activity by enhancing the expression of the Cox2 gene. Other aryl hydrocarbons such as 7,12-dimethylbenz[a]anthracene, were weaker inducers of prostaglandin production in canine kidney cells. Interestingly, induction depended on cell type; human fibroblasts, rabbit endothelial cells and mouse fibroblasts were unresponsive to B[a]P (32). It seems likely that the lack of effect of 7,12-dimethylbenz[a]anthracene on prostaglandin production in our work reflects the cell type used. In the present work, B[a]P but not B[e]P induced Cox-2 and prostaglandin synthesis, which could be important for understanding why B[a]P is a more potent carcinogen than B[e]P (34). B[a]P is metabolized more extensively than B[e]P in most types of cells (35). Hence, upregulation of Cox-2 by B[a]P but not B[e]P could reflect differences in the metabolism of these aryl hydrocarbons. Cox has both cyclooxygenase and peroxidase activities. Aside from being important for the synthesis of prostaglandins, the peroxidase function contributes to the activation of procarcinogens. Drugs such as non-steroidal anti-inflammatory drugs inhibit the cyclooxygenase but not the peroxidase function of Cox, which potentially limits their effectiveness as anticancer agents. Although aspirin and indomethacin delay the formation of oral tumors in experimental animals (36,37), it is possible that this approach to cancer prevention can be improved upon. 798

For example, a detailed understanding of the mechanism(s) by which B[a]P up-regulates Cox-2 could provide important insight into possible approaches to blocking the expression of Cox-2. Chemopreventive strategies that interfere with the signaling mechanism(s) responsible for the induction of Cox2 should inhibit carcinogenesis. The human Cox-2 promoter contains a motif consistent with a xenobiotic responsive element (XRE), two putative NF-κB sites (38,39) and a CRE site (40). Our data are compatible with the idea that B[a]P induces Cox-2 via an Ah receptordependent reaction, as is the case for other xenobiotic metabolizing enzymes such as CYP1A1 (41). However, a recent report that 2,3,7,8-tetrachlorodibenzo-p-dioxin stimulated transcription of the Cox-2 gene via an XRE-independent mechanism suggests that the Ah receptor may not be involved in mediating the effects of B[a]P (42). B[a]P-induced oxidative stress could activate NF-κB or AP-1 (43) and thereby up-regulate Cox-2. This idea is attractive because it fits with the observation that antioxidants protect against B[a]P-induced cellular transformation (44). Further studies are needed to elucidate the molecular mechanisms responsible for B[a]P-mediated up-regulation of Cox-2. Acknowledgements The authors wish to thank Dr David Zakim for suggestions that were helpful in preparing this manuscript. This work was supported by grants CA68136, CA57166 and 1T32 CA09685-02 from the National Cancer Institute. J.R.M. was the recipient of a fellowship award from the Cancer Research Foundation of America.

References 1. Hu,Z. and Wells,P.G. (1992) In vitro and in vivo biotransformation and covalent binding of benzo[a]pyrene in Gunn and RHA rats with a genetic deficiency in bilirubin uridine diphosphate UDP-glucuronosyltransferase. J. Pharmacol. Exp. Ther., 263, 334–342. 2. Burke,D.M., Vadi,H., Jernstrom,B. and Orrenius,S. (1977) Metabolism of benzo[a]pyrene with isolated hepatocytes and the formation and degradation of DNA-binding derivatives. J. Biol. Chem., 252, 6424–6431. 3. Bock,K.W., Bock-Hennig,K.S., Lilienblum,W. and Volp,R.F. (1981) Release of mutagenic metabolites of benzo[a]pyrene from the perfused rat liver after inhibition of glucuronidation and sulfation by salicylamide. Chem.-Biol. Interactions, 36, 167–177. 4. Marnett,L.J., Reed,G.A. and Dennison,D.J. (1978) Prostaglandin synthetase dependent activation of 7,8-dihydro-7,8-dihydroxy-benzo[a]pyrene to mutagenic derivatives. Biochem. Biophys. Res. Commun., 82, 210–216. 5. Guthrie,J., Robertson,I.G.C., Zeiger,E., Boyd,J.A. and Eling,T. (1982) Selective activation of some dihydrodiols of several polycyclic aromatic hydrocarbons to mutagenic products by prostaglandin synthetase. Cancer Res., 42, 1620–1623. 6. Dix,T.A. and Marnett,L.J. (1983) Metabolism of polycyclic aromatic hydrocarbon derivatives to ultimate carcinogens during lipid peroxidation. Science, 221, 77–79. 7. Eling,T.W., Thompson,D.C., Foureman,G.L., Curtis,J.F. and Hughes,M.F. (1990) Prostaglandin H synthase and xenobiotic oxidation. Annu. Rev. Pharmacol. Toxicol., 30, 1–45. 8. Eling,T.E. and Curtis,J.F. (1992) Xenobiotic metabolism by prostaglandin H synthase. Pharmac. Ther., 3, 261–273. 9. Kujubu,D.A., Fletcher,B.S., Varnum,B.C., Lim,R.W. and Herschman,H.R. (1991) TIS 10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cells, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J. Biol. Chem., 266, 12866–12872. 10. O’Banion,M.K., Sadowsky,H.B., Winn,V. and Young,D.A. (1991) A serum and glucocorticoid-regulated 4-kilobase mRNA encodes a cyclooxygenase-related protein. J. Biol. Chem., 266, 23261–23267. 11. DuBois,R.N., Awad,J., Morrow,J., Roberts,L.J. and Bishop,P.R. (1994) Regulation of eicosanoid production and mitogenesis in rat intestinal epithelial cells by transforming growth factor- and phorbol ester. J. Clin. Invest., 93, 493–498.

B[a]P upregulation of cyclo-oxygenase gene expression 12. Rimarachin,J.A., Jacobson,J.A., Szabo,P., Maclouf,J., Creminon,C. and Weksler,B.B. (1994) Regulation of cyclooxygenase-2 expression in aortic smooth muscle cells. Arterioscl. Thromb., 14, 1021–1031. 13. Sacks,P.G., Parnes,S.M., Gallick,G.E., Mansouri,Z., Lichtner,R., SatyaPrakash,K.L., Pathak,S. and Parsons,D.F. (1988) Establishment and characterization of two new squamous cell carcinoma cell lines derived from tumors of the head and neck. Cancer Res., 48, 2858–2866. 14. Xu,L., Schantz,S.P., Edelstein,D. and Sacks,P.G. (1996) A simplified method for the routine culture of normal oral epithelial (NOE) cells from upper aerodigestive tract mucosa. Methods Cell Sci., 18, 1–9. 15. Weksler,B.B. (1990) Heparin and acidic fibroblast growth factor interact to decrease prostacyclin synthesis in human endothelial cells by affecting both prostaglandin H synthase and prostacyclin synthase. J. Cell. Physiol., 142, 514–522. 16. Lowry,O.H., Rosebrough,N.J., Farr,A.L. and Randell,R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193, 265–275. 17. Laemmli,U.K. (1970) Cleavage of structural proteins during the assembly of bacteriophage T4. Nature (Lond.), 227, 680–685. 18. Towbin,H., Staehelin,T. and Gordon,J. (1979) Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc. Natl Acad. Sci., 76, 4350–4354. 19. Habib,A., Creminon,C., Frobert,Y., Grassi,J., Pradelles,P. and Maclouf,J. (1993) Demonstration of an inducible cyclo-oxygenase in human endothelial cells using antibodies raised against the carboxyl-terminal region of cyclooxygenase-2. J. Biol. Chem., 268, 23448–23454. 20. Yang,E.K., Radominska,A., Winder,B.S. and Dannenberg,A.J. (1993) Dietary lipids coinduce xenobiotic metabolizing enzymes in rat liver. Biochim. Biophys. Acta, 1168, 52–58. 21. Chomczynski,P. and Sacchi,N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem., 162, 156–159. 22. Inoue,H., Yokoyama,C., Hara,S., Tone,Y. and Tanabe,T. (1995) Transcriptional regulation of human prostaglandin endoperoxide synthase2 gene by lipopolysaccharide and phorbol ester in vascular endothelial cells. J. Biol. Chem., 270, 24965–24971. 23. Whatley,R.E., Satoh,K., Zimmerman,G.A., McIntyre,T.M. and Prescott,S.M. (1994) Proliferation-dependent changes in release of arachidonic acid from endothelial cells. J. Clin. Invest., 94, 1889–1900. 24. Janot,F., Massaad,L., Ribrag,V., de Waziers,I., Beaune,P.H., Luboinski,B., Parise,O., Gouyette,A. and Chabot,G.G. (1993) Principal xenobioticmetabolizing enzyme systems in human head and neck squamous cell carcinoma. Carcinogenesis, 14, 1279–1283. 25. Denissenko,M.F., Pao,A., Tang,M-S. and Pfeifer,G.P. (1996) Preferential formation of benzo[a]pyrene adducts at lung cancer mutational hotspots in p53. Science, 274, 430–432. 26. Smith,T.J., Stoner,G.D. and Yang,C.S. (1995) Activation of 4(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in human lung microsomes by cytochromes P450, lipoxygenase and hydroperoxides. Cancer Res., 55, 5566–5573. 27. Taffet,S.M. and Russell,S.W. (1980) Macrophage-mediated tumor cell killing: regulation of expression of cytolytic activity by prostaglandin E. J. Immunol., 126, 424–427. 28. Brunda,M.J., Heberman,R.B. and Holden,H.T. (1980) Inhibition of murine natural killer cell activity by prostaglandins. J. Immunol., 124, 2682–2687. 29. Goodwin,J.S. and Ceuppens,J. (1983) Regulation of the immune response by prostaglandins. J. Clin. Immunol., 3, 295–314. 30. Tsujii,M. and DuBois,R.N. (1995) Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell, 83, 493–501. 31. Lu,X., Xie,W., Reed,D., Bradshaw,W.S. and Simmons,D.L. (1995) Nonsteroidal antiinflammatory drugs cause apoptosis and induce cyclooxygenase in chicken embryo fibroblasts. Proc. Natl Acad. Sci. USA, 92, 7961–7965. 32. Levine,L. (1977) Chemical carcinogens stimulate canine kidney (MDCK) cells to produce prostaglandins. Nature, 268, 447–448. 33. Hassad,A. and Levine,L. (1977) Induction of fatty acid cyclo-oxygenase activity in canine kidney cells (MDCK) by benzo[a]pyrene. J. Biol. Chem., 252, 6591–6593. 34. World Health Organization (1983) IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. International Agency for Research on Cancer Research Publications, Lyon, France, pp. 211–237. 35. Macleod,M.C., Mansfield,B.K. and Selkirk,J.K. (1982) Time course of metabolism of benzo[e]pyrene by hamster embryo cells and the effect of chemical modifiers. Chem.-Biol. Interactions, 40, 275–285. 36. Perkins,T.M. and Shklar,G. (1982) Delay in hamster buccal pouch carcinogenesis by aspirin and indomethacin. Oral Surg., 53, 170–178.

37. Cornwall,H., Odukoya,O. and Shklar,G. (1983) Oral mucosal tumor inhibition by ibuprofen. Oral Maxillofac. Surg., 41, 795–800. 38. Kosaka,T., Miyata,A., Ihara,H., Hara,S., Sugimoto,T., Takeda,O., Takahashi,E. and Tanabe,T. (1994) Characterization of human gene (PTGS2) encoding prostaglandin-endoperoxide synthase 2. Eur. J. Biochem., 221, 889–897. 39. Tazawa,R., Xu,X.-M., Wu,K.K. and Wong,L.-H. (1994) Characterization of the genomic structure, chromosomal location and promoter of human prostaglandin H synthase 2 gene. Biochem. Biophys. Res. Commun., 203, 190–199. 40. Xie,W. and Herschman,H. (1995) v-src induces prostaglandin synthase 2 gene expression by activation of the c-Jun N-terminal kinase and the cJun transcription factor. J. Biol. Chem., 270, 27622–27628. 41. Whitlock,J.P.Jr (1986) The regulation of cytochrome P450 gene expression. Annu. Rev. Pharmacol. Toxicol., 26, 333–369. 42. Kraemer,S.A., Arthur,K.A., Denison,M.S., Smith,W.L. and DeWitt,D.L. (1996) Regulation of prostaglandin endoperoxide H synthase-2 expression by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch. Biochem. Biophys., 330, 319–328. 43. Janknecht,R., Cahill,M.A. and Nordheim,A. (1995) Signal integration at the c-fos promoter. Carcinogenesis, 16, 443–450. 44. Arnold,J.T., Wilkinson,B.P., Sharma,S. and Steele,V.E. (1995) Evaluation of chemopreventive agents in different mechanistic classes using a rat tracheal epithelial cell culture transformation assay. Cancer Res., 55, 537–543. Received on September 19, 1996; accepted on December 6, 1996

799