Chemopreventive efficacies of aspirin and sulindac against lung tumorigenesis in A/J mice

Transcription

1 Carcinogenesis vol.18 no.5 pp , 1997 Chemopreventive efficacies of aspirin and sulindac against lung tumorigenesis in A/J mice Caroline Duperron and Andre Castonguay 1 human gastrointestinal tract and metabolized mainly in the liver to sulfide and sulfone intermediates (5,6). Structures of Laboratory of Cancer Etiology and Chemoprevention, School of Pharmacy, Laval University, Quebec City, Canada, G1K 7P4 sulindac and its two metabolites are shown in Figure 1. The sulfide metabolite has analgesic, anti-pyretic and anti- 1 To whom correspondence should be addressed inflammatory properties (5). Renal elimination of sulindac Non-steroidal anti-inflammatory drugs (NSAIDs) are occurs via its sulfone metabolite (5,7). This metabolite exhibits among the most widely prescribed drugs. In this study, no anti-inflammatory property (8). In addition to its antiwe demonstrated the efficacy of aspirin to inhibit lung inflammatory properties, sulindac is a chemopreventive agent tumorigenesis in A/J mice. Lung tumors (9.9 tumors/ that causes regression of large-bowel polyps in patients with mouse) were induced by the tobacco-specific nitrosamine, familial polyposis (9). 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), ad- In parallel with clinical studies, numerous models of animal ministered in drinking water between week 0 and week tumors were used to demonstrate the inhibition of chemically 7. Groups of mice were fed sulindac (123 mg/kg diet), induced cancers by various NSAIDs, such as piroxicam (colon), acetylsalicylic acid (ASA; 294 mg/kg), non-buffered indomethacin (esophagus, pancreas, tongue, mammary gland, Aspirin (294 mg/kg) or buffered Aspirin (294 mg/kg) in uterine cervix), ibuprofen (colon) and sulindac (colon) (10 AIN-76A diet from week 2 to the end of the bioassay 16). Reddy et al. (17) demonstrated that aspirin inhibits colon (week 23). These doses are comparable to the maximal carcinogenesis in rats by 70%. Sakata et al. (18) observed that doses recommended for humans. ASA and non-buffered aspirin co-administered with N-[4-(5-nitro-2-furyl)-2-thiazolyl] Aspirin were the most effective inhibitors and reduced formamide inhibits the incidence of bladder carcinoma in rats lung multiplicities by 60 and 62%, respectively. Sulindac by 80%. Our research group was the first to show that inhibited lung tumor multiplicity by 52%. Inhibition by NSAIDs, and especially sulindac, are effective against lung buffered Aspirin was not statistically significant. We tumorigenesis in laboratory animals (19,20). We reported that evaluated the efficacies of NSAIDs to inhibit NNK activation sulindac, in non-toxic doses, reduced the multiplicity of lung by h1a2 v2 cells expressing human P-450 1A2. Salicylates, tumors in A/J mice by 50% (19,20). Aspirin was not included at doses of 500 µm and 1 mm, had no effect on NNK in our previous studies. activation. Sulindac and its sulfide and sulfone metabolites Tobacco smoke contains over 4000 compounds, 43 of which (1 mm) inhibited NNK metabolism by 90, 92 and 65%, are known to be carcinogens (21). One of these carcinogens respectively. We observed a 76% inhibition with SKF 525A, a P-450 inhibitor. Taken together, these results indicate that salicylates and sulindac could be equally effective as chemopreventive agents, but they could differ in their mode of action. Introduction Aspirin is a non-steroidal anti-inflammatory drug (NSAIDs*) extensively used for its analgesic, anti-pyretic and antirheumatic properties (1). Retrospective and prospective epidemiological studies (reviewed in 2) concluded that regular use of aspirin reduces the risk of colorectal cancer. In a large US cohort study, participants were asked about their intake of aspirin during the month before the initial interview. The results revealed a reduced incidence of lung cancer associated with increased aspirin use (3). In a prospective study, Thun et al. (4) observed no association between aspirin use and the risk of respiratory cancer. The debate on the effectiveness of aspirin in preventing or delaying lung cancer development needs to be explored further. Data from laboratory studies can provide rationales and incentive for investigating the efficacy of aspirin in lung cancer prevention. Sulindac is a prodrug that is absorbed rapidly from the *Abbreviations: NSAIDs, non-steroidal anti-inflammatory drugs; NNK, 4- (methylnitrosamino)-1-(3-pyridyl)-1-butanone; ASA, acetylsalicylic acid; SA, salicylic acid. Fig. 1. Biotransformation of ASA and sulindac to their metabolites. Oxford University Press 1001

2 C.Duperron and A.Castonguay is a nicotine-derived nitrosamine, 4-(methylnitrosamino)-1-(3- included in Table I. Group 1 received the diet without a chemopreventive pyridyl)-1-butanone (NNK). NNK is present in the mainstream agent and were given water ad libitum. Groups 2 6 received NNK in drinking water for 7 weeks. Group 2 was fed AIN-76A without NSAIDs. Groups 3 6 smoke of commercial cigarettes at levels ranging from 6 to were fed NSAIDs 2 weeks before the treatment with NNK and throughout 197 ng in American cigarettes (22,23) and from 6 to 97 ng in the whole assay. Group 3 was fed AIN-76A supplemented with sulindac at a Canadian cigarettes (24). Hecht and Hoffmann suggested a dose of 123 mg/kg of diet, which is below the maximum tolerated dose as causal relationship between NNK and pulmonary carcino- determined previously (20). Groups 4, 5 and 6 were given diets containing ASA, non-buffered Aspirin and buffered Aspirin, respectively. Doses of genesis related to tobacco smoking (25). NNK is a potent lung ASA were 50% of the maximum tolerated dose. carcinogen in laboratory animals, such as A/J mice (26,27). At 16 weeks after the end of the NNK treatment, the mice were fasted Studies with microsomes prepared either from human liver overnight, killed by CO 2 asphyxiation and necropsied. Lungs were fixed in and lung tissues or from transfected human cell lines demonadenomas Tellyesniczky s fixative for 7 days before counting the number of surface 1 mm. Stomachs were fixed in situ with 0.5 ml of 10% formalin, strated that NNK is a pro-carcinogen activated by cytochrome excised and stored in formalin. Papillomas 1 mm were counted. P-450 isoenzymes 1A2, 2A6, 2D6 and 2E1 (28 32). Pathways Cells culture of NNK activation by α-carbon hydroxylation are shown in AHH TK / human lymphoblastoid cell lines (chol and h1a2 v2) were Figure 2. Smith et al. (33) demonstrated that purified human purchased from Gentest Corporation (Woburn, MA). chol, the negative cytochrome P-450 1A2 incorporated into a reconstituted system control, is a cell line containing extrachromosomal plasmid vectors that confer can catalyze the activation of NNK. resistance to L-histidinol. This cell line expresses low levels of P-450 activities. The aims of this study were: (i) to determine the efficacy The h1a2 v2 cell line expresses human CYP1A2 with basal CYP1A1 activity. of aspirin as a chemopreventive agent of lung carcinogenesis; The cells were maintained in culture in RPMI 1640 medium (Gentest) and supplemented with horse serum 9% v/v (Gentest), L-glutamine (292 µg/ml) (ii) to investigate NNK metabolism in cells expressing human and gentamicin (50 µg/ml) (Sigma Chemical Co.). P-450 1A2; (iii) to compare, in these cells, the inhibition of Assay of NNK metabolism NNK activation by aspirin, sulindac and its sulfide and sulfone The cells ( /3 ml) of the 25 cm 2 Falcon tissue culture flask were metabolites. incubated with 8 µm [5 3 H]NNK (19.2 µci/ml) without NSAIDs or with 1 mm ASA, 1 mm salicylic acid (SA), 1 µm sulindac, 1 µm sulindac sulfide, Materials and methods 1 µm sulindac sulfone, 100 µm SKF 525A (a P-450 inhibitor). After 48 h of Chemicals incubation, the cells and the media were centrifuged at 1500 g and harvested. The medium was filtered with a 0.45 µm Minisart filter (Sartorius, Germany) NNK (98% purity) and [5-3 H]NNK (2.4 Ci/mmol, 97.4% purity) were and frozen at 20 C until its analysis. purchased from Chemsyn Science Lab. (Lenexa, KS). Sulindac and acetylsali- The medium was analyzed for NNK and NNK metabolites on a reversecylic acid (ASA) were purchased from Sigma Chemical Co. (Mississauga, phase HPLC system, using a Spherisorb ODS 2, 5 µm column mm Ont., Canada). Bayer non-buffered Aspirin and Bayer buffered Aspirin (Jones Chromatography Inc., Columbus, OH). Aliquots of 750 µl of the (Sterling Winthrop Inc., Ont., Canada) were purchased in a local market in filtered medium and 7 µl of the NNK metabolites standards were co-injected October and eluted with a ph 6.0 sodium acetate buffer and methanol as described Lung tumor assay in A/J mice previously (34). The eluent was monitored at 254 nm and 1 ml fractions were collected. Aliquots of 5 ml of Scintiverse II (Fisher Scientific, Montreal, Female A/J mice (6 7 weeks old) were obtained from Jackson Laboratories Canada) were added to each fraction and radioactivity was measured. (Bar Harbor, ME). They were housed in groups of five per cage and kept under standard conditions (22 2 C; 45 5% relative humidity; 12:12h Kinetic of ASA hydrolysis light dark cycle). They were sorted out on weight basis into six groups and Stock solution of ASA 10 mm was prepared with 9.01 mg ASA, 0.4 ml of given tap water and powdered AIN-76A diet from Harlan Teklad (Madison, 2% NaHCO 3 and 4.6 ml of culture medium. ASA (1 mm) was added to 3 ml WI) ad libitum. Powdered NSAIDs were mixed with the diet and two feeders of cell culture media (with /ml or without cells). Incubations were were placed in each cage; diet consumption and spillage were monitored three performed at periods ranging from 0 min to 48 h. As described above, cells times before and after carcinogen treatment. Stock solutions of NNK were and media were then harvested. prepared in distilled water (3.49 mg/ml) and diluted in tap water. The initial Levels of ASA and SA in culture media were measured by a reverse-phase concentration of NNK was 87.4 µg/ml and was adjusted thereafter for each HPLC. The HPLC system consisted of a Waters 501 pump, Rheodyne 7125 cage according to water consumption. Water consumption was monitored injector, Waters 441 UV detector, Hewlett Packard 3390A integrator and a twice a week for 7 weeks. C 18 µbondapak column ( mm; Waters, Milford, MA). Aliquots of Details of the treatment with NNK and the chemopreventive agents are 100 µl of the filtrate were injected and eluted with a 0.35 N acetic acid:methanol mobile phase (75:25, v/v) during a 20-min period at a flow rate of 1.2 ml/ min. The eluent was monitored at 280 nm. Retention time was 11 min for ASA and 16 min for SA. Results were calculated from the linear regression curve, which related peak areas and the given concentrations of ASA and SA. Concentrations of salicylates in culture medium were compared by a splitplot statistical analysis. Fig. 2. Metabolic pathways of NNK in h1a2 v2 cells. Structures in brackets are hypothetical intermediates Results Lung adenoma assay The numbers of tumors induced by NNK with or without NSAIDs are shown in Table I. The total dose of NNK was 9.1 mg per mouse. The number of lung tumors in untreated mice was low (0.2 tumor/mouse) and comparable to previous studies (35). Treatment with NNK only induced 9.92 tumors/ mouse, but this number was reduced to 4.72 tumors/mouse (52% inhibition) in mice treated with sulindac. ASA and nonbuffered Aspirin reduced lung tumor multiplicity to 3.93 (60%) and 3.71 tumors/mouse (62%), respectively. In contrast, buffered Aspirin had negligible effects on lung tumorigenesis. Body weight gains of NNK-treated mice were 11 to 21% lower than the untreated mice (Table I). Therefore, giving

3 Efficacies of aspirin and sulindac Table I. Effects of feeding NSAIDs on lung tumorigenesis induced by NNK in A/J mice a Group Treatment No. of surviving Total dose Dose of NSAIDs Body weight Lung tumor Incidence of mice No. with NSAIDs mice/no. of of NNK mg/kg diet (g/mouse) c multiplicity d with tumor initial mice (mg/mouse) b (mg/kg body weight) 1 None 10/10 None None /10 2 None 24/ None * /24 3 Sulindac 25/ (15) * ** 25/25 4 ASA 15/ (35) * ** 14/15 5 Non-buffered Aspirin 14/ (32) * ** 14/14 6 Buffered Aspirin 15/ (32) /15 a Six- to 7-week-old mice were given NNK in drinking water between week 0 and week 7. NSAIDs were given between weeks 2 to 23. b Mean SD (n 14 25). c Mean SD (at week 23). Statistically different from group 1 (*P 0.005, Student s t-test). d Mean SE. Lung tumors larger than 1mm were counted. Statistically different from group 2 (**P 0.05, Student s t-test). Table II. Metabolism of NNK by h1a2 v2 cells cultured with or without salicylates NSAIDs Metabolites (pmol/ml) a No. 10 No. 11 No. 12 No b None c (n 1) None d (n 4) ASA 500 µm (n 2) ASA1mM(n 3) SA1mM(n 3) a Mean SD. b Numbers refer to structures in Figure 2. c chol (human lymphoblastoid cells not transfected). d h1a2 v2 (transfected cells). NNK in drinking water for 7 weeks delays the gain of body Discussion weight. After the end of NNK treatment, body weight gains resumed. Body weight gains of NNK and NNK NSAIDsreached 22% in men and 11% in women (36). Considering The incidence of lung cancer among the Canadian population treated mice were not significantly different. the low 5-year survival rate (8 12%) of lung cancer, prevention Metabolism of NNK becomes a realistic alternative. The use of NSAIDs in clinical In the presence of h1a2 v2 cells, ~50% of NNK were converted trials on the chemoprevention of colorectal cancer has shown to NNAL after 48 h (data not shown). Metabolic activation of promising results, but no information on the effectiveness of NNK is initiated by α-carbon hydroxylation (Figure 2). The these agents against lung cancer is available. Obviously, a four end products of this activation are the metabolite numbers better understanding of the mechanism of the action of NSAIDs 10, 11, 12 and 13. These numbers refer to structures in Figure would help in the design of chemoprevention clinical trials. 2. The sum of the NNK metabolites reflect the extent of the This study is the first one to demonstrate the effectiveness total activation of NNK. The extent of NNK metabolism and of aspirin in inhibiting lung tumorigenesis in laboratory its inhibition by salicylates, sulindac and the two sulindac animals. This observation is particularly significant considering metabolites are shown in Tables II and III, respectively. Results that it has been estimated that thousand tons of aspirin shown in Table II indicate that neither ASA nor SA (500 µm are consumed in the USA every year (1). The doses of NSAIDs or 1 mm) inhibits NNK α-carbon hydroxylation. As shown in used in this study are comparable to the maximal recommended Table III, sulindac and its sulfide and sulfone metabolites at daily doses for humans. In this study, we used non-buffered 1 mm concentrations inhibited NNK metabolism by 90, 92 Aspirin at a dose of 31.7 mg/kg of body weight, while and 65%, respectively. SKF 525A, a P-450 inhibitor, was used the recommended maximal anti-pyretic and anti-inflammatory as a positive control and inhibited the metabolism of NNK doses are 55.7 and 83.6 mg/kg of body weight, respectively. by 76%. These results lead to the hypothesis that regular use of aspirin Rate of ASA hydrolysis might reduce the risk of lung cancer in smokers. Results of the stability of ASA in the culture media over a Previous studies in our laboratory (19,20) have shown that 48-h period, in the presence and absence of h1a2 v2 cells, NSAIDs other than aspirin inhibit pulmonary carcinogenesis. are shown in Figure 3. Hydrolysis occurred linearly with time. These studies demonstrate that the extent of inhibition varies After 18 h, almost all ASA had been hydrolyzed to SA. The depending on the type of NSAID. While sulindac was extent of hydrolysis of ASA to salicylates was higher with effective against NNK-induced lung tumorigenesis, ibuprofen rather than without h1a2 v2 cells between 0 and 12 h (P and piroxicam were less effective and naproxen had no effect ). In this study, we observed 52% inhibition with the use of 1003

4 C.Duperron and A.Castonguay Table III. Metabolism of NNK in h1a2 v2 cells cultured with or without sulindac, with its sulfide and sulfone metabolites NSAIDs Metabolites (pmol/ml) a No. 10 b,e No. 11 e No. 12 e No. 13 e No e None c (n 1) 1.2 ND None d (n 4) Sulindac 1mM (n 2) ND f * * * Sulindac 1mM (n 2) ND * ND * Sulindac 1mM (n 2) ND * * SKF 525A 100 mm (n 3) * * ND * a Mean SD. b Numbers refer to Figure 2. c chol (control cells not transfected). d h1a2 v2 (transfected cells). e Statistically different from h1a2 v2 transfected cells (*P 0.05, Student s t-test). f ND, not detected ( 1.7 pmol/ml). Fig. 3. Kinetic of ASA hydrolysis with or without h1a2 v2 cells. j, ASA; d, ASA with h1a2 v2 cells; u, SA; s, SA with h1a2 v2 cells. Each point corresponds to the mean SD (n 3). sulindac. This result is comparable to the percentage inhibition previously reported (51 and 53%) (19,20). ASA and non- buffered Aspirin reduce by 60 and 62%, respectively, the number of lung tumors (Table I). Our results demonstrate that aspirin is as effective as sulindac in preventing lung carcinogenesis. One of the major side effects of ASA is its gastric toxicity. Various formulations of aspirin, such as non-buffered, buffered and enteric-coated formulations, have been developed to eliminate this problem. The absorption of ASA can be influenced by many factors: tablet solubility, gastric ph, rate of disintegration and food ingestion. In a study on aspirin pharmacokinetics in rats, Fu et al. (37) compared the bio- availability of buffered and unbuffered ASA. The unbuffered formulation was found to give higher AUC (area under curve) than the buffered formulation. Fu et al. attributed this difference to the shift of the absorption of the aspirin from the stomach to the intestine (where the aspirin may be more hydrolyzed). Our results show that while non-buffered Aspirin and ASA reduce the lung tumor multiplicity, buffered Aspirin has no significant effect (Table I). An increased gastric ph, related to the buffer and food absorption could have reduced or delayed aspirin absorption. Our results demonstrate that the formulation 1004 of NSAIDs has an impact on their chemopreventive efficacies in A/J mice. A previous study has demonstrated that α-carbon hydroxylation of NNK is catalyzed by various P-450s, including human P-450 1A2 (30). Our results confirm these observations. Human lymphoblastoid cells that express human P-450 1A2 catalyze α-carbon hydroxylation (activation pathway, Table II) but not pyridine N-oxidation (detoxification pathway) of NNK (data not shown). We observed no inhibition of NNK metabolism even at relatively high concentrations of ASA (500 µm and 1 mm) or SA (1 mm). These results correlate with other studies that show that aspirin does not affect the metabolism of NNK (38,39). Using rat lung microsomes, Smith et al. (38) observed a 0 9% inhibition of the NNK metabolism by aspirin at concentrations varying from 30 to 300 µm. With rat liver microsomes, Guo et al. (39) observed no inhibition of NNK metabolism with a 100 µm aspirin. We conclude that inhibition of tumorigenesis by aspirin could not involve an inhibition of NNK activation by P-450 1A2. Aspirin is known to be an inhibitor of prostaglandin synthesis (40). Bilodeau et al. (41) reported that sulindac reduces plasma levels of PGE 2 by 51%, and that naproxen has no effect. Considering that PGE 2 favors clonal expansion of initiated cells, Bilodeau et al. (41) suggested that lower levels of PGE 2 would delay tumor development. Further studies by our research group are focusing on the role of the inhibition of prostaglandin synthesis, which is a crucial element in lung cancer chemoprevention. In order to study the effect of ASA on NNK metabolism, our research group decided to dissolve ASA in a culture medium at ph 7.6. However, Thiessen (42) has reported that ASA is quickly hydrolyzed to SA in aqueous solutions. As a result, we also had to determine the rate of ASA hydrolysis by h1a2 v2 cells. In the first part of the curve (from 0 to 18 h), we observed a difference between results with cells and results without cells. This difference was probably due to a cellular absorption of ASA. These results show that we still had half of the initial amount of ASA after 12 h. Sulindac is an anti-inflammatory prodrug that has to be reduced to a sulfide to exhibit anti-inflammatory effectiveness. Sulindac sulfone, a second metabolite, is formed by an irreversible oxidation of the sulfoxide function. In this study, we observed that sulindac and its two metabolites, at a concentration of 1 mm, inhibited the metabolism of NNK in

5 Efficacies of aspirin and sulindac h1a2 v2 cells. Results in Table III show that sulindac sulfide (1995) Chemoprevention of colon carcinogenesis by sulindac, a is more effective in inhibiting NNK metabolism than sulindac nonsteroidal antiinflammatory agent. Cancer Res., 55, Reddy,B.S., Rao,C.V., Rivenson,A. and Kelloff,G. (1993) Inhibitory effect sulfone is (92% vs 65%). SKF 525A, a P-450 inhibitor, of aspirin on azoxymethane-induced colon carcinogenesis in F344 rats. inhibited NNK metabolism (see Table III). Because the P-450 Carcinogenesis, 14, A2 is the only P-450 in the h1a2 v2 cells, we can conclude 18. Sakata,T., Hasegawa,R., Johansson,S.L., Zenser,T.V. and Cohen,S.M. that sulindac and its sulfide metabolite act by inhibiting the (1986) Inhibition by aspirin of N-[4-(5-nitro-2-furyl)-2-thiazolyl]- formamide initiation and sodium saccharin promotion of urinary bladder P-450 isoform, which activates NNK. These results are in line carcinogenesis in male F344 rats. Cancer Res., 46, with our previous study showing that the metabolism of NNK 19. Pepin,P., Bouchard,L., Nicole,P. and Castonguay,A. (1992) Effects of by mouse lung tissues was inhibited by sulindac sulfide (43). sulindac and oltipraz on the tumorigenicity of 4-(methylnitrosamino)-1- In summary, the results of this study show that aspirin is a (3-pyridyl)-1-butanone in A/J mouse lung. Carcinogenesis, 13, lung chemopreventive agent that is as effective as sulindac in 20. Jalbert,G. and Castonguay,A. (1992) Effects of NSAIDs on NNK-induced pulmonary and gastric tumorigenesis in A/J mice. Cancer Lett., 66, A/J mice; aspirin does not inhibit NNK activation by the 21. US Department of Health and Human Services (1989) Reducing the health human P-450 1A2. consequences of smoking, 25 years of progress. Office on Smoking and Health. DHHS Publication No. CDC , pp Adams,J.D., Lee,S.J., Vinchkoski,N., Castonguay,A. and Hoffmann,D. Acknowledgements (1983) On the formation of the tobacco-specific carcinogen 4- We gratefully acknowledge Dr Marlène Bouillon for her precious advice. We (methylnitrosamino)-1-(3-pyridyl)-1-butanone during smoking. Cancer also thank Dr D.Lévesque and R.De La Durantaye for their technical assistance Lett., 17, in the tumor inhibition study. This study was supported by a grant from the 23. Fischer,S., Spiegelhalder,B., Eisenbarth,J. and Preussmann,R. (1990) Cancer Research Society Inc. (Canada). Investigations on the origin of tobacco-specific nitrosamines in mainstream smoke of cigarettes. Carcinogenesis, 11, References 24. Fischer,S., Castonguay,A., Kaiserman,M., Spiegelhalder,B. and Preussmann,R. (1990) Tobacco-specific nitrosamines in Canadian cigarettes. J. Cancer Res. Clin. Oncol., 116, Goodman Gilman,A., Goodman,L.S. and Gilman,A. (1990) Analgesic- 25. Hecht,S.S. and Hoffmann,D. (1989) The relevance of tobacco-specific antipyretics, anti-inflammatory agents: the salicylates. In Goodman,L.S. nitrosamines to human cancer. Cancer Surv., 8, and Gilman,A. (eds) The Pharmacological Basis of Therapeutics, 8th edn. 26. Hecht,S.S. and Hoffmann,D. (1988) Tobacco-specific nitrosamines, an MacMillan Publishing Co., New York, pp important group of carcinogens in tobacco and tobacco smoke. 2. Trujillo,M.A., Garewald,H.S. and Sampliner,R.E. (1994) Nonsteroidal Carcinogenesis, 9, antiinflammatory agents in chemoprevention of colorectal cancer: at what 27. Belinsky,S.A., Stefanski,S.A. and Anderson,M.W. (1993) The A/J mouse cost? Dig. Dis. Sci., 39, lung as a model for developing new chemointervention strategies. Cancer 3. Schreinemachers,D.M. and Everson,R.B. (1994) Aspirin use and lung, Res., 53, colon, and breast cancer incidence in a prospective study. Epidemiology, 28. Gonzalez,F.J., Crespi,C.L. and Gelboin,H.V. (1991) cdna-expressed 5, human cytochrome P-450s: a new age of molecular toxicology and human 4. Thun,M.J., Namboodiri,M.M., Calle,E.E., Flanders,W.D. and Heath,C.W.Jr risk assessment. Mutat. Res., 247, (1993) Aspirin use and risk of fatal cancer. Cancer Res., 53, Yamazaki,H., Inui,Y., Yun,C.H., Guengerich,F.P. and Shimada,T. (1992) 5. Brogden,R.N., Heel,R.C., Speight,T.M. and Avery,G.S. (1978) Sulindac: a Cytochrome P-450 2E1 and 2A6 enzymes as major catalysts for metabolic review of its pharmacological properties and therapeutic efficacy in activation of N-nitrosodialkylamines and tobacco-related nitrosamines in rheumatic diseases. Drugs, 16, human liver microsomes. Carcinogenesis, 13, Duggan,D.E. (1981) Sulindac: therapeutic implications of the prodrug/ 30. Smith,T.J., Guo,Z., Gonzalez,F.J., Guengerich,F.P., Stoner,G.D. and pharmacophore equilibrium. Drug Metab. Rev., 12, Yang,C.S. (1992) Metabolism of 4-(methylnitrosamino)-1-(3-pyridyl)-1-7. Strong,H.A., Warner,N.J., Renwick,A.G. and George,C.F. (1985) Sulindac butanone in human lung and liver microsomes and cytochromes P-450 metabolism: The importance of an intact colon. Clin. Pharmacol. Ther., expressed in hepatoma cells. Cancer Res., 52, , Penman,B.W., Reece,J., Smith,T., Yang,C.S., Gelboin,H.V., Gonzalez,F.J. 8. Van Arman,C.G., Risley,E.A., Nuss,G.W., Hucker,H.B. and Duggan,D.E. and Crespi,C.L. (1993) Characterization of a human cell line expressing (1976) Pharmacology of sulindac. In Huskisson,E.C. and Franchimont,P. high levels of cdna-derived CYP2D6. Pharmacogenetics, 3, (eds) Clinoril in the Treatment of Rheumatic Disorders: A New Nonsteroidal Antiinflammatory/Analgesic Agent. Proceedings of a Symposium VIII 32. Smith,T.J., Stoner,G.D. and Yang,C.S. (1995) Activation of 4- European Rheumatology Congress, Helsinski 1 7 June Raven Press, (methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in human lung New York, p. 9. microsomes by cytochromes P-450, lipoxygenase, and hydroperoxides. 9. Waddell,W.R. and Loughry,R.W. (1983) Sulindac for polyposis of the Cancer Res., 55, colon. J. Surg. Oncol., 24, Smith,T.J., Guo,Z.Y., Guengerich,F.P. and Yang,C.S. (1995) Metabolism 10. Reddy,B.S., Maruyama,H. and Kelloff,G. (1987) Dose-related inhibition of of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) by human colon carcinogenesis by dietary piroxicam, a nonsteroidal antiinflammatory cytochrome P-450 1A2 and inhibition by phenethyl isothiocyanate drug, during different stages of rat colon tumor development. Cancer Res., (PEITC). Proc. Am. Assoc. Cancer Res., 36, , Jorquera,R., Castonguay,A. and Schuller,H.M. (1992) Effects of pregnancy 11. Rubio,C.A. (1984) Antitumoral activity of indomethacin on experimental and ethanol treatment on the metabolism of 4-(methylnitrosamino)-1-(3- esophageal tumors. J. Natl Cancer Inst., 72, pyridyl)-1-butanone by hamster liver and lung microsomes. Drug Metab. 12. Takahashi,M., Furukawa,F., Toyoda,K., Sato,H., Hasegawa,R., Imaida,K. Dispos., 20, and Hayashi,Y. (1990) Effects of various prostaglandin synthesis inhibitors 35. Shimkin,M.B. and Stoner,G.D. (1975) Lung tumors in mice: application on pancreatic carcinogenesis in hamsters after initiation with N- to carcinogenesis bioassay. Adv. Cancer Res., 21, nitrosobis(2-oxopropyl)amine. Carcinogenesis, 11, National Cancer Institute of Canada (1992) Canadian Cancer Statistics 13. Tanaka,T., Nishikawa,A., Mori,Y., Morishita,Y. and Mori,H. (1989) Toronto, Canada, p. 14. Inhibitory effects of nonsteroidal antiinflammatory drugs, piroxicam and 37. Fu,C.J., Melethil,S. and Mason,W.D. (1991) The pharmacokinetics of indomethacin on 4-nitroquinoline 1-oxide-induced tongue carcinogenesis aspirin in rats and the effect of buffer. J. Pharmacokin. Biopharm., 19, in male ACI/N rats. Cancer Lett., 48, Carte,C.A., Milholland,R.J., Shea,W. and Ip,M.M. (1983) Effect of the 38. Smith,T.J., Guo,Z., Hong,J.Y., Ning,S.M., Thomas,P.E. and Yang,C.S. prostaglandin synthetase inhibitor indomethacin on 7,12- (1992) Kinetics and enzyme involvement in the metabolism of 4- dimethylbenz[a]anthracene-induced mammary tumorigenesis in rats fed (methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in microsomes of different levels of fat. Cancer Res., 43, rat lung and nasal mucosa. Carcinogenesis, 13, Rao,A.R. and Hussain,S.P. (1988) Modulation of methylcholanthrene- 39. Guo,Z., Smith,T.J., Thomas,P.E. and Yang,C.S. (1992) Metabolism of 4- induced carcinogenesis in the uterine cervix of mouse by indomethacin. (methylnitrosamino)-1-(3-pyridyl)-1-butanone by inducible and Cancer Lett., 43, constitutive cytochrome P-450 enzymes in rats. Arch. Biochem. Biophys., 16. Rao,C.V., Rivenson,A., Simi,B., Zang,E., Kelloff,V.S. and Reddy,B.S. 298,

6 C.Duperron and A.Castonguay 40. Marnett,L.J. (1992) Aspirin and the potential role of prostaglandins in colon cancer. Cancer Res., 52, Bilodeau,J.F., Wang,M., Chung,F.L. and Castonguay,A. (1995) Effects of nonsteroidal antiinflammatory drugs on oxidative pathways in A/J mice. Free Rad. Biol. Med., 18, Thiessen,J.J. (1982) Pharmacokinetics of salicylates. In Barnett,H.J.M., Hirsh,J. and Mustard,J.F. (eds) Acetylsalicylic Acid: New Uses for an Old Drug. Raven Press, New York, pp Bouchard,L. and Castonguay,A. (1993) Inhibitory effects of nonsteroidal antiinflammatory drugs (NSAIDs) on the metabolism of 4- (methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in mouse lung explants. Drug Metab. Disp., 21, Received on June 12, 1996; revised on September 30, 1996; accepted on December 16,