Oxo Substituents Markedly Alter the Phase II Metabolism of α

Aug 18, 1997 - The P450-catalyzed hydroxylation of tamoxifen to give α-hydroxytamoxifen [(E)-4-{4-[2-(dimethylamino)ethoxy]phenyl}-3 ...
0 downloads 0 Views 291KB Size
Download PDF
Chem. Res. Toxicol. 1997, 10, 887-894

887

Oxo Substituents Markedly Alter the Phase II Metabolism of r-Hydroxybutenylbenzenes: Models Probing the Bioactivation Mechanisms of Tamoxifen Kornepati V. Ramakrishna,† Peter W. Fan,† C. Scott Boyer,‡ Deepak Dalvie,‡ and Judy L. Bolton*,† Department of Medicinal Chemistry and Pharmacognosy (M/C 781), College of Pharmacy, University of Illinois at Chicago, 833 South Wood Street, Chicago, Illinois 60612-7231, and Drug Metabolism Department, Central Research Division, Pfizer Inc., Groton, Connecticut 06340 Received April 18, 1997X

The P450-catalyzed hydroxylation of tamoxifen to give R-hydroxytamoxifen [(E)-4-{4-[2(dimethylamino)ethoxy]phenyl}-3,4-diphenyl-3-buten-2-ol] and subsequent formation of reactive sulfate esters which alkylate DNA has been proposed to be a potential carcinogenic pathway for tamoxifen. In the present study, the ability of R-hydroxytamoxifen analogs to form GSH and sulfate conjugates was investigated in order to understand the structural features influencing reactivity. The para oxo analogs 1 [1-(4-methoxyphenyl)-3-hydroxy-1-butene], 2 [1-(4-hydroxyphenyl)-3-hydroxy-1-butene], and 4 [1-(4-hydroxyphenyl)-1-phenyl-3-hydroxy-1butene] reacted with GSH instantaneously under strong acidic conditions to yield GSH conjugates in greater than 90% yields. Interestingly, the meta phenolic analogs 3 [1-(3hydroxyphenyl)-3-hydroxy-1-butene] and 5 [1-(3-hydroxyphenyl)-1-phenyl-3-hydroxy-1-butene] did not react with GSH to any significant extent under similar conditions. Characterization of the GSH conjugates with 1H-NMR, electrospray mass spectrometry, and UV showed that all of the conjugates resulted from attack of GSH at the R-position of the substrates with displacement of the hydroxyl group. The formation of a single pair of diastereomeric conjugates strongly supported adduct formation to proceed through a direct SN2 displacement mechanism and not through a quinone methide (4-alkyl-2,5-cyclohexadien-1-one) intermediate. At physiological pH and temperature only the para hydroxy analogs 2 and 4 gave GSH conjugates, a reaction which seems to be catalyzed by isoforms of glutathione S-transferase. Similar substituent effects were observed in the sulfotransferase-mediated formation of R-hydroxy sulfate esters in that only the para hydroxy analogs formed conjugates at the aliphatic hydroxyl group. Finally, the present investigation showed a remarkable difference in the reactivities of para and meta phenolic analogs of R-hydroxybutenylbenzenes toward GSH and sulfate conjugation reactions.

Introduction Tamoxifen remains the endocrine therapy of choice in the treatment of all stages of breast cancer (1). In addition, large-scale clinical trials have recently been initiated to determine the potential of tamoxifen to act as a chemopreventive agent in women considered at high risk for developing breast cancer (2). However, several studies in animal models have raised concern over the safety of chronic treatment with this drug (3). For example, tamoxifen induces hepatocarcinomas (4) and mammary tumors (5) in rats. Also, an increased incidence of endometrial adenocarcinomas has been diagnosed in women treated with tamoxifen (6, 7), and tamoxifen-induced DNA adducts have recently been detected in the endometrial tissue of breast cancer patients (8). These troubling reports demonstrate the need of fully understanding the potential carcinogenic mechanisms of this drug before it is in wide spread use as a prophylactic agent in high-risk but otherwise healthy individuals. * Address correspondence to this author. Fax: (312) 996-7107. E-mail: [email protected]. † University of Illinois at Chicago. ‡ Pfizer Inc. X Abstract published in Advance ACS Abstracts, August 1, 1997.

S0893-228x(97)00060-X CCC: $14.00

Although the carcinogenic effects of estrogens and antiestrogens have been mainly attributed to hormonal properties (9), there is interest in these compounds acting as chemical carcinogens by binding to cellular macromolecules. In the case of tamoxifen, there is evidence of two distinct bioactivation pathways both resulting in electrophilic species (Scheme 1). Group I DNA adducts are produced by aromatic hydroxylation of tamoxifen producing 4-hydroxytamoxifen as the proximate carcinogen (Scheme 1) (10). This phenol can form a highly electrophilic quinone methide (TAM-QM)1 by a P450catalyzed direct two-electron oxidation mechanism. Alternatively, R-hydroxylation giving 4,R-dihydroxytamoxifen followed by dehydration could be the mechanism of 1 Abbreviations: tamoxifen, (Z)-1-{4-[2-(dimethylamino)ethoxy]phenyl]-1,2-diphenyl-1-butene; R-hydroxytamoxifen, (E)-4-{4-[2-(dimethylamino)ethoxy]phenol}-3,4-diphenyl-3-buten-2-ol; P450, cytochrome P450; QM, quinone methide, 4-alkyl-2,5-cyclohexadien-1-one; o-quinone, 3,5-cyclohexadiene-1,2-dione; 1, 1-(4-methoxyphenyl)-3hydroxy-1-butene; 2, 1-(4-hydroxyphenyl)-3-hydroxy-1-butene; 3, 1-(3hydroxyphenyl)-3-hydroxy-1-butene; 4, 1-(4-hydroxyphenyl)-1-phenyl3-hydroxy-1-butene; 5, 1-(3-hydroxyphenyl)-1-phenyl-3-hydroxy-1butene; DCNP, 2,6-dichloro-4-nitrophenol; PCP, pentachlorophenol; DHEA, dehydroepiandrosterone; PCA, perchloric acid; PAP, 3′-phosphoadenosine-5jm’-phosphate; PAPS, 3-phosphoadenosine-5′-phosphosulfate; TBDMS, tert-butyldimethylsilyl; CI-MS, chemical ionization mass spectrometry; FAB-MS, fast atom bombardment mass spectrometry, LC-MS, liquid chromatography-mass spectrometry; electrospray-MS, electrospray mass spectrometry.

© 1997 American Chemical Society

888 Chem. Res. Toxicol., Vol. 10, No. 8, 1997

Ramakrishna et al.

Scheme 1. Proposed Mechanisms of Tamoxifen Bioactivation to DNA-Binding Speciesa

a

The bold lines show the model compounds.

QM formation (10). Evidence in support of TAM-QM as the ultimate carcinogen includes (1) 32P-postlabeling experiments of chemical or microsomal oxidation of 4-hydroxytamoxifen in the presence of DNA gave adducts that cochromatographed with in vivo hepatic group I DNA adducts isolated from mice treated with tamoxifen (11). (2) Coadministration of tamoxifen with the sulfotransferase inhibitor pentachlorophenol enhanced group I DNA adduct formation 11-fold while group II adducts were suppressed 6-fold (12). The increase in group I adducts was presumably the result of the greater availability of 4-hydroxytamoxifen which had not been depleted through sulfate conjugation. (3) Droloxifene (3hydroxytamoxifen) does not induce liver tumors in rats, and no DNA adducts were detected from this analog (13). The lack of observed genotoxic effects for droloxifene may be the result of altered metabolism caused by the 3-hydroxy substituent. The mechanism of formation of group II DNA adducts involves hydroxylation at the R-carbon generating R-hydroxytamoxifen, conjugation with sulfate, and alkylation of DNA with displacement of sulfate group (Scheme 1) (14, 15). Evidence for this pathway includes (1) a significant reduction occurs in DNA adducts with [ethyld5]tamoxifen compared to the nondeuterated compound suggesting C-H bond cleavage at the R-position is required for DNA adduct formation; (2) treatment of rat hepatocytes with R-hydroxytamoxifen results in 15-63fold higher level of DNA adducts than with comparable concentrations of tamoxifen (14, 16); (3) toremifene, an analog of tamoxifen with a β-Cl substituent, did not

produce modified DNA bases in rat liver DNA in vivo, suggesting the β-Cl residue inhibited P450-catalyzed R-hydroxylation (17); and (4) as mentioned above, group II DNA adducts are suppressed by sulfotransferase inhibitors supporting formation of a sulfate ester as the proximate carcinogen (12). In the present investigation, we analyzed the effect of changing ring position and type of oxo substituents on the R-hydroxytamoxifen bioactivation mechanism. As shown in Chart 1, analogs of R-tamoxifen (1), R-hydroxydroloxifene (3, 5), and 4,R-dihydroxytamoxifen (2, 4) were synthesized in order to define the structural features influencing reactivity with GSH as well as their ability to form reactive sulfate esters. The results suggest that substituents and pH have a dramatic effect on reaction of these allylic alcohols with biological nucleophiles as well as determining the extent and type of sulfate ester formed.

Materials and Methods Materials. All chemicals were purchased from Aldrich (Milwaukee, WI), Fisher Scientific (Itasca, IL), or Sigma (St. Louis, MO) unless stated otherwise. The compounds shown in Chart 1 were synthesized by a general procedure involving an aldol condensation with acetone and the corresponding benzaldehyde followed by reduction of the resulting ketone with NaBH4. The detailed procedure for synthesis of 2 is described here. trans-1-(4-Hydroxyphenyl)-3-hydroxy-1-butene (2). The ketone 4-(4-hydroxyphenyl)-2-oxo-3-butene was prepared as follows. To a solution of 4-hydroxybenzaldehyde (2.0 g, 16.0

Tamoxifen Bioactivation Model Studies Chart 1. Structures of the Model Compounds Studied

mmol) in acetone (7.5 mL) was added 50% NaOH (2.7 mL) with stirring resulting in a reddish yellow precipitate. Water (5 mL) was added and the mixture heated until the precipitate dissolved. The resulting red solution was allowed to stand at room temperature for 72 h after which time 10 mL of H2O was added, and the solution was acidified with concentrated HCl and extracted with ethyl acetate (3 × 25 mL). The combined organic layers were washed with water (20 mL), dried over anhydrous magnesium sulfate, and concentrated to yield 1-(4-hydroxyphenyl)-3-oxo-1-butene (2.2 g, 84%). To a solution of the ketone (0.77 g, 4.7 mmol) in methanol (5 mL) was added NaBH4 (0.14 g, 3.7 mmol), and the solution stirred for 15 h. The reaction was quenched with saturated ammonium chloride (1 mL). Methanol was removed under reduced pressure, and the residue was redissolved in water (10 mL), neutralized with 0.1 M HCl at 0 °C, and extracted with ethyl acetate (4 × 10 mL). The combined organic layers were washed with water (20 mL), dried over anhydrous magnesium sulfate, and concentrated to yield a brown solid. The residue was redissolved in hexane and purified by flash chromatography on silica gel with 40% ethyl acetate:hexane as eluent: 1H-NMR (CDCl3) δ 1.21 (d, J ) 6.4 Hz, 3H, CH3), 3.80 (2H, bs, OH), 4.25-4.33 (quintet, J ) 6.8 Hz, 1H, CHOH), 5.94 (dd, J ) 15.7, 6.67 Hz, 1H, ArCHdCH), 6.34 (d, J ) 15.70 Hz, 1H, ArCHdCH), 6.65 (d, J ) 8.5 Hz, 2H, ArH2,6), 7.11 (d, J ) 8.5 Hz, 2H, ArH3,5); UV (CH3CN) 260, 296 nm (sh); positive ion CI-MS (methane) m/z 165 (MH+, 30), 147 (MH+ - H2O, 100). Using analogous synthetic methods 3 was prepared from 3-hydroxybenzaldehyde. trans-1-(3-Hydroxyphenyl)-3-hydroxy-1-butene (3): 1H-NMR (CDCl3) δ 1.34 (d, J ) 6.5 Hz, 3H, CH3), 4.42 (quintet, J ) 6.5 Hz, 1H, CHOH), 6.21 (dd, J ) 15.9, 7.59 Hz, 1H, ArCHdCH), 6.48 (d, J ) 15.9 Hz, 1H, ArCHdCH), 6.72 (m, 1H, ArH4), 6.88 (m, 2H, ArH2, ArH6), 7.14 (1H, dd, J ) 8.2, 8.1 Hz, ArH5); UV (CH3CN) 254, 296 nm; positive ion CI-MS (methane) m/z 165 (46) (MH+), 147 (100) (MH+ - H2O). Compound 1 was synthesized by initially methylating the ketone 1-(4-hydroxyphenyl)-3-oxo-1-butene with dimethyl sulfate followed by NaBH4 reduction. 1-(4-Methoxyphenyl)-3hydroxy-1-butene (1): 1H-NMR (CDCl3) δ 1.36 (d, J ) 6.0 Hz, 3H, CH3), 1.72 (bs, 1H, OH), 3.8 (s, 3H, OCH3), 4.48 (m, 1H, CHOH), 6.12 (dd, J ) 6.6, 16 Hz, 1H, ArCHdCH), 6.50 (d, J ) 16 Hz, 1H, ArCHdCH), 6.85 (d, J ) 9 Hz, 2H, ArH), 7.32 (d, J ) 9 Hz, 2H, ArH); UV (CH3CN) 262, 292 nm (sh); positive ion CI-MS (methane) m/z 179 (22) (MH+), 161 (100) (MH+ - H2O). Compound 5 was prepared according to the reaction shown in Scheme 2. The identification of all intermediates was established by 1H-NMR. 3-Hydroxybenzophenone was initially protected with tert-butyldimethylsilyl (TBDMS) chloride by adding TBDMSCl (540 mg, 3.6 mmol) and imidazole (510 mg, 7.5 mmol) to a solution of 3-hydroxybenzophenone (594 mg, 3.0

Chem. Res. Toxicol., Vol. 10, No. 8, 1997 889 mmol) in DMF (1.5 mL) and stirring at room temperature under nitrogen. After 16 h water (10 mL) was added to the reaction, and the resulting solution was extracted with hexane (10 mL × 5). The combined organic layers were washed with water (10 mL × 2) and brine (10 mL), dried over anhydrous magnesium sulfate, and concentrated to a yellow oil. The conversion of the TBDMS-protected 3-hydroxybenzophenone, I, to 1-[3-(tert-butyldimethylsiloxy)phenyl]-1-phenyl-2-propenol, II (reaction a, Scheme 2), was carried out by adding vinylmagnesium bromide (4.15 mmol) to a solution of the ketone (650 mg, 2.08 mmol) in 5 mL of dry THF. After the reaction mixture was stirred at 0 °C for 30 min, 5 mL of aqueous saturated ammonium chloride was added to quench the reaction. The aqueous phase was extracted with ethyl acetate (3 × 2.5 mL). The combined organic layers were washed with aqueous saturated NaCl and evaporated to give 506 mg of II. This was converted without purification to (E/Z)-1-[3-(tertbutyldimethylsiloxy)phenyl]-1-phenylpropen-3-ol, III (reaction b), by dissolving II in 5 mL of dioxane and treating it with 2-3 drops of sulfuric acid. After the reaction mixture was stirred at room temperature for 30 min, 5 mL of aqueous saturated sodium bicarbonate was added to quench the reaction. The aqueous phase was extracted with ether (3 × 3 mL), and the combined organic layers were washed with aqueous saturated NaCl. After the solvent was removed in vacuo, the crude product was purified by flash chromatography on silica gel with 10% ethyl acetate:hexane as eluent. Compound III was oxidized to (E/Z)-1-[3-(tert-butyldimethylsiloxy)phenyl]-1-phenylpropen-3-al, IV (reaction c), by dissolving it in 4 mL of dry THF and treating it with 1 g of manganese dioxide. After the reaction mixture was stirred at room temperature for 15 h, the reaction suspension was filtered through a bed of silica gel. After the solvent was removed in vacuo, the crude product was purified by flash chromatography on silica gel with 5% ethyl acetate:hexane as eluent. IV (283 mg, 0.84 mmol) was converted to (E/Z)-1-[3-(tertbutyldimethylsiloxy)phenyl]-1-phenylbuten-3-ol, V (reaction d), by dissolving it in 5 mL of dry THF and treating it with methylmagnesium bromide (1.3 mmol). After the reaction mixture was stirred at 0 °C for 30 min, 5 mL of aqueous saturated ammonium chloride was added to quench the reaction. The aqueous phase was extracted with ether (3 × 2.5 mL), and the combined organic layers were washed with aqueous saturated NaCl and evaporated to give 241 mg of V. This was converted without purification to the final 5 (reaction e) by dissolving V in 3 mL of THF and treating it with tetrabutylammonium fluoride (1.0 mmol). After the reaction mixture was stirred for 30 min, the solvent was evaporated in vacuo and the crude product was purified by flash chromatography on silica gel with 20% ethyl acetate:hexane as eluent. (E/ Z)-1-(3-Hydroxyphenyl)-1-phenylbuten-3-ol (5): E:Z ratio ) 1:1; 1H-NMR (CDCl3) δ 1.31 (d, J ) 6.3 Hz, 3H, CH3, E and Z isomers), 4.35-4.46 (m, 2H, CHOH, E and Z isomers), 5.01 (d, 1H, J ) 9 Hz, CdCH, E or Z isomer), 6.04 (d, 1H, J ) 9 Hz, CdCH, E or Z isomer), 6.68-6.84 (m, 6H, ArH, E and Z isomers), 7.08-7.35 (m, 12H, ArH, E and Z isomers); UV (CH3CN) 214, 250 nm; positive ion CI-MS (methane) m/z 241 (17) (MH+), 223 (100) (MH+ - H2O). Using analogous synthetic methods, 4 was prepared from 4-(tert-butyldimethylsiloxy)benzophenone. (E/Z)-1-(4-Hydroxyphenyl)-1-phenylbuten-3-ol (4): E : Z ratio ) 3:2; 1H-NMR (CDCl3) δ 1.32 (d, J ) 6.3 Hz, 3H, CH3, E or Z isomer), 1.36 (d, 3H, CH3, E or Z isomer), 4.32-4.52 (m, 2H, CHOH, E and Z isomers), 5.97 (d, J ) 9 Hz, 1H, CdCH, E or Z isomer), 6.01 (d, J ) 9 Hz, 1H, CdCH, E or Z isomer), 6.73 (d, J ) 8.7 Hz, 2H, ArH, E or Z isomer), 6.83 (d, J ) 9.9 Hz, 2H, ArH, E or Z isomer), 7.02-7.40 (m, 16H, ArH, E and Z isomers); UV (CH3OH) 224, 262 nm; positive ion CI-MS (methane) m/z 241 (17) (MH+), 223 (100) (MH+ - H2O). GSH Conjugates of r-Hydroxybutenylbenzenes. The GSH conjugates were prepared by combining the substrate (0.5 mM) with 1.0 mM GSH in 5 mL of H2O containing 250 µL of PCA, 37 °C. Under these conditions 2 and 1 were almost

890 Chem. Res. Toxicol., Vol. 10, No. 8, 1997

Ramakrishna et al.

Scheme 2. Synthesis of 5a

a

Reactions a-e are described in the text.

quantitatively converted to two diastereomeric GSH conjugates. The percent yields of GSH conjugates were 99% (2) and 100% (1). Similarly, 4 gave two sets of diastereomeric GSH conjugates (from E and Z isomers) in 100% yield. In contrast, only 4% of 3 formed conjugates under similar conditions, and only one set of diastereomeric GSH conjugates was formed from 5 in 6% yield. After centrifugation for 15 min at 3000 rpm, the conjugates were isolated from the aqueous phase on C-18 extraction cartridges (J. T. Baker) and eluted with methanol. The methanol was concentrated to 500 µL, and aliquots (25 µL) were analyzed directly by HPLC with a 4.6 × 150 mm Ultrasphere C-18 column (Beckman) on a Hewlett-Packard (Palo Alto, CA) 1090L gradient HPLC equipped with an photodiode array UV/ vis absorbance detector set at 230-350 nm and a 5989B MS engine quadrupole mass spectrometer. The mobile phase consisted of 5% methanol in 0.5% ammonium acetate (pH 3.5) at 1.0 mL/min for 5 min, increased to 40% CH3OH over 45 min, isocratic for 5 min, and increased to 90% CH3OH over the last 20 min. GSH conjugates of 2: 1H-NMR (CD3OD) δ 1.37 (d, J ) 6.8 Hz, 3H, CH3), 2.14 (m, 2H, Glu-β), 2.52 (m, 2H, Glu-γ), 2.8 (dt, H, Cys-β′), 2.95 (dt, 1H, Cys-β), 3.62 (m, 1H, Glu-R), 3.75 (q, J ) 7.6 Hz, 1H, CHdCH-CHSG-), 3.91 (s, 2H, Gly-R), 4.65 (m, 1H, Cys-R), 5.9 (m, 1H, CHdCH-CH-SG-), 6.45 (dd, J ) 1.5, 16.5 Hz, 1H, CHdCH-CH-SG-), 6.81 (d, J ) 8.3 Hz, 2H, ArH), 7.31 (d, J ) 8.7 Hz, 2H, ArH); UV (CH3OH) 270, 298 nm; electrospray-MS m/z 454 (80) (MH+); retention time 44.6, 45.3 min. GSH conjugates of 1: 1H-NMR (CD3OD) δ 1.39 (d, J ) 6.8 Hz, 3H, CH3), 2.15 (m, 2H, Glu-β), 2.53 (m, 2H, Glu-γ), 2.85 (dt, H, Cys-β′), 2.93 (dt, 1H, Cys-β), 3.65 (m, 1H, Glu-R), 3.74 (q, 1H, CHdCH-CHSG-), 3.83 (s, 3H, OCH3), 3.90 (s, 2H, GlyR), 4.56 (m, 1H, Cys-R), 5.95 (m, 1H, CHdCH-CH-SG-), 6.45 (dd, J ) 2.1, 15.6 Hz, 1H, CHdCH-CH-SG-), 6.94 (d, J ) 8 Hz, 2H, ArH), 7.40 (d, J ) 8 Hz, 2H, ArH); UV (CH3OH) 268 nm; FAB-MS (positive ion, glycerol) m/z 468 (30) (MH+), 161 (MH+ - GSH); retention time 62, 62.5 min. GSH conjugates of 4: UV (CH3OH) 267 (67 min), 267 (68 min), 240, 260 nm (70 min); electrospray-MS m/z 530 (100) (MH+); retention time 67, 68, 70 min. GSH conjugates of 5: electrospray-MS m/z 530 (100) (MH+); retention time 68, 69 min. Kinetic Experiments. The effects of pH, reaction time, and GSH concentration were determined by monitoring GSH adduct formation by HPLC. In general, GSH was combined with the substrate (0.5 mM) in 1 mL of buffer (50 mM K2HPO4, pH 7-8, 50 mM Na2CO3, pH 8-10, NaOH, pH >10) and incubated at 37 °C for 30 min. The reaction was quenched by addition of 5 mL of cold ether and mixed vigorously to separate the substrate from GSH and the GSH conjugates. Aliquots of the aqueous layer (100 µL) were analyzed directly by HPLC with a 4.6 × 150 mm Ultrasphere C-18 column (Beckman) on a Shimadzu LC-10A gradient HPLC equipped with an SIL-10A autoinjector and SPD-10AV UV detector set at 280 nm. The mobile phase consisted of 5% methanol in 0.25% perchloric acid/0.25% acetic

acid (pH 3.5) at 1.0 mL/min for 5 min, increased to 40% CH3OH over 45 min, isocratic for 5 min, and increased to 90% CH3OH over the last 20 min. Incubations. Male Sprague-Dawley rats (160-180 g) were obtained from Sasco Inc. (Omaha, NE). Cytosol was prepared from rat liver (18), and protein concentrations were determined as described previously (19). Incubations containing cytosolic protein (3.0 mg/mL) were conducted for 30 min at 37 °C in 50 mM phosphate buffer (pH 7.4, 500 µL total volume). Substrates were added as solutions in dimethyl sulfoxide, and GSH was added in phosphate buffer to achieve final concentrations of 0.5 and 1.0 mM, respectively. PAPS (0.5 mM) was used to initiate the reaction, and for control incubations PAPS was omitted. Experiments with inhibitors contained a DMSO solution of DCNP, PCP, or DHEA and phosphate-buffered solutions of PAP of 0.5 mM final volume. The incubations were quenched by extracting the substrate with ethyl ether (2 × 5.0 mL). The aqueous layer was separated, and 100 µL of this solution was analyzed on HPLC. Sulfate conjugates of 2: UV (CH3OH) 295 (30 min), 252 nm (33 min); electrospray-MS (negative ion) m/z 243 (100) (M-); retention time 30, 33 min. Sulfate conjugates of 3: UV (CH3OH) 247 nm; electrosprayMS (negative ion) m/z 243 (100) (M-); retention time 33 min. Instrumentation. HPLC experiments were performed on the above-mentioned Shimadzu HPLC system. Peaks were integrated with Shimadzu EZ-Chrom software and a 486-33 computer. UV spectra were measured with a Hewlett-Packard Model 8452 diode array UV/vis spectrophotometer, and 1H-NMR spectra were obtained with a Varian XL-300 spectrometer at 300 MHz. Electrospray mass spectra were obtained using a Hewlett-Packard 5989B MS engine quadrupole mass spectrometer equipped with a ChemStation data system and high-flow pneumatic nebulizer-assisted electrospray LC-MS interface. The mass spectrometer was interfaced to the above-mentioned Hewlett-Packard gradient HPLC system. The quadrupole analyzer was maintained at 120 °C, and unit resolution was used for all measurements. Nitrogen at a pressure of 80 psi was used for nebulization of the HPLC effluent, and nitrogen bath gas at 350 °C and a flow rate of 50 mL/min where used for evaporation of solvent from the electrospray. The range m/z 200-900 was scanned over approximately 2 s during LC-MS.

Results Acid-Catalyzed Reaction of r-Hydroxytamoxifen Analogs with GSH. In order to further investigate the bioactivation mechanism(s) of tamoxifen, analogs of R-hydroxytamoxifen (1), R-hydroxydroloxifene (3, 5), and 4,R-dihydroxytamoxifen (2, 4) were synthesized. Interestingly, under strongly acidic reaction conditions the para oxo analogs formed GSH conjugates instantaneously in very high yields, while the meta phenolic substrates failed to react with GSH to any significant extent (Figure 1). The HPLC analysis of the reaction products of 1 and

Tamoxifen Bioactivation Model Studies

Chem. Res. Toxicol., Vol. 10, No. 8, 1997 891

Figure 2. Influence of pH on reaction of 2 with glutathione, correlation with the titration curve for 2: (b) amount of adduct formed and (×) titration curve of 2. Reactions contained 0.5 mM substrate, 1.0 mM GSH, and buffered solutions at various pH values in a total volume of 500 µL at 37 °C. The titration curve was generated spectrophotometrically by monitoring the increase in UV absorbance at 260 nm as a function of increasing pH.

Figure 1. HPLC analysis of the reaction mixture of acidcatalyzed reaction of GSH with (A) 2, (B) 3, (C) 4, and (D) 5. Reactions contained 0.5 mM substrate, 1.0 mM GSH, and 25 µL of PCA, in a total volume of 500 µL. *Peaks are decomposition products of 5.

2 with GSH in the presence of PCA displayed two adduct peaks. The LC-MS analysis of the adduct peaks showed that each pair of conjugates had the same molecular weight confirming them to be a pair of diastereomeric products resulting from attack of GSH at the prochiral R-carbon of the substrates. The attack by GSH at the R-position was confirmed from the 1H-NMR spectra of the conjugates. Particularly characteristic was the R-hydrogen at 3.75 ppm shifted upfield by 0.55 ppm relative to that of 2 as expected for the effect of a sulfur atom compared to oxygen. Loss of water forming the corresponding quinone methide and reaction with GSH does not occur as four conjugates would be expected resulting from both 1,6- and 1,8-addition (20, 21). In the case of diaryl substrates, the para phenolic analog 4 formed GSH conjugates quantitatively like the monoaryl analogs (Figure 1C). In this case two sets of diastereomeric products were produced for each E and Z isomer, although the second set coeluted as one peak and could not be resolved using a variety of HPLC conditions. The diaryl meta phenolic analog 5 did give some GSH conjugates as minor products (Figure 1D); however, most of the substrate formed other products likely involving dehydration and cyclization reactions. Reaction of Tamoxifen Analogs with GSH at Physiological pH and Temperature. At pH 7.4, 2 reacted very slowly with GSH at a rate of approximately 4.0 µM conjugate formation/h (data not shown). Similarly, small amounts of GSH conjugates were observed from reaction of 4 with GSH at physiological pH and temperature. No evidence of GSH adduct formation from

the other analogs was detected under these conditions suggesting that the phenolic hydroxy group in the para position has considerable influence on the reaction. For 2, conjugate formation depends on GSH concentration, saturating at a 5-fold excess of GSH over substrate concentration (data not shown). Changing pH also influenced the extent of adduct formation from 2 (Figure 2). No conjugates were detected below pH 7.0, whereas from pH 7-11 adduct formation increased with increasing pH correlating with the pKa of the phenolic hydroxy group of 2 (pKa ) 9.0). Similar experiments with 1 showed no adduct formation from pH 6 to 11 confirming the importance of deprotonation of the para phenolic group on reaction of R-hydroxybutenylbenzenes with GSH. Formation of GSH Conjugates in Rat Liver Cytosol. As the above experiments suggest conjugate formation through a direct SN2 displacement mechanism at physiological pH and temperature, the ability of glutathione S-transferase (GST) to catalyze the reaction was explored (Table 1). Experiments with commercial rat liver GST and 2 showed an increase in GSH adduct formation of 2-3-fold. Similarly, incubations with rat liver cytosol gave 4-fold more product relative to heatinactivated cytosol or in the absence of cytosol (Table 1). GSH conjugate formation from 2 depended on the concentration of cytosolic protein suggesting GST isozyme(s) could be involved (Figure 3). In support of this, an inhibitor of GSTs, pentachlorophenol (PCP) (22), and a competitive substrate for GSTs, dichloronitrophenol (DCNP) (23), reduced GSH conjugate levels by 60% (Table 1). No GSH conjugates were detected during corresponding cytosolic incubations with the methoxy (1) and meta hydroxy (3) analogs. Sulfotransferase-Mediated Conjugation of Tamoxifen Analogs. Since one bioactivation pathway for tamoxifen is believed to involve R-hydroxylation, conju-

892 Chem. Res. Toxicol., Vol. 10, No. 8, 1997

Ramakrishna et al.

Table 1. Formation of GSH Conjugates of 2 in Rat Liver Cytosol incubation conditionsa

GSH conjugates (µM)b

cytosol without cytosol boiled cytosol -PAPS (0.5 mM) +PCP (0.5 mM) +DCNP (0.5 mM) commercial GSTc (50 µg/mL) (250 µg/mL) without GSTc

3.8 ( 0.4 0.95 0.85 3.8 1.6 ( 0.02 1.4 ( 0.01 6.2 9.4 2.7

a Incubations were conducted for 30 min at 37 °C in pH 7.4 phosphate buffer with 0.5 mM substrate, 1.0 mM GSH, untreated rat liver cytosol (3.0 mg/mL), amd 0.5 mM PAPS. b Determined from the peak area ratio of the adducts formed in the acidcatalyzed reaction which gives quantitative formation of GSH conjugates from 2. c GSH concentration was 5.0 mM.

Figure 4. HPLC chromatogram of the sulfate and GSH conjugates observed after incubating 2 (0.5 mM) with rat liver cytosol (3.0 mg/mL), PAPS (0.5 mM), and GSH (1.0 mM) in pH 7.4 phosphate buffer for 30 min at 37 °C. Table 2. Effect of Sulfotransferase Inhibitors on Sulfate Ester Formation from 2 incubation conditionsa

Figure 3. Effect of changing cytosolic protein concentration on GSH and sulfate conjugate formation from 2: (b) sum of the sulfate conjugates and (×) GSH conjugates - the amount formed in the absence of cytosol. Incubation conditions were 0.5 mM 2, various concentrations of rat liver cytosol, 5.0 mM GSH, and PAPS (0.5 mM), pH 7.4, 37 °C, 30 min. Reactions were terminated by ether extraction as described in Materials and Methods.

gation with sulfate, and loss of the sulfate ester generating a carbocation, the relative ability of the R-hydroxybutenylbenzenes to form sulfate esters was explored in rat liver cytosol. When the substrates were incubated with rat liver cytosol in the presence of PAPS, sulfated metabolites were observed for most of the analogs (i.e., Figure 4 for 2) as confirmed by electrospray-MS. As observed with the GSH conjugates the amount of sulfated metabolites produced directly depended on the concentration of cytosolic protein (Figure 3), and omission of PAPS from the incubation mixture resulted in no metabolite formation (Table 2). A significant difference was observed in the extent and type of sulfated metabolites formed with the model compounds. In the case of para phenolic mono- and diaryl substrates, two sulfates were observed. Since the molecular weights of the two sulfate conjugates formed from incubations with 2 or 4 with rat liver cytosol and PAPS were same, this suggests that sulfation occurred at the para phenolic and R-allylic alcohol positions. The aliphatic sulfates were relatively stable as confirmed by repeated injections of the same HPLC sample, and they were not affected by the presence of GSH in the incubation. In addition, no increase in the GSH conjugates was detected which suggests that loss of the sulfate group and reaction with GSH do not occur under these conditions. The possibility that GSH is too soft a nucleophile to trap the carbocation was considered; however, similar experiments with adenine as the trapping agent also showed no formation of adenine adducts

cytosol -GSH -PAPS +DHEA +PAP +PCP +DCNP without cytosol boiled cytosol

sulfate conjugates (% of control) 30 min 32 min 100 100 NDb 91 ( 9 25 ( 2 ND ND ND ND

100 100 NDb 102 ( 10 41 ( 2 8(1 ND ND ND

a Incubations were conducted for 30 min at 37 °C in pH 7.4 phosphate buffer with 0.5 mM substrate, 1.0 mM GSH, untreated rat liver cytosol (3.0 mg/mL), and 0.5 mM PAPS, in the presence and absence of sulfotransferase inhibitors (0.5 mM). b None detected.

(data not shown). In the case of the corresponding meta analogs 3 and 5, only one sulfate conjugate was formed likely at the meta phenolic group. In support of this, no sulfate metabolites were detected during cytosolic incubations with the ether analog 1. Inhibition of Sulfotransferase-Mediated Conjugation of 2 and 3. The sulfation of 2 by rat liver cytosol and PAPS was affected by sulfotransferase inhibitors. As shown in Table 2, incubations in the presence of the specific inhibitors for phenol sulfotransferases DCNP and PCP resulted in virtually 100% inhibition of sulfate formation. In contrast, DHEA, a typical inhibitor for hydroxysteroid sulfotransferases, did not affect the formation of sulfates, while PAP, a nonspecific inhibitor, led to 60-75% inhibition. In the case of 3 a similar trend was observed (data not shown). DCNP and PCP very strongly inhibited the formation of phenolic sulfate while DHEA did not. These experiments suggest that phenolic sulfotransferases are involved in the sulfation of both the phenolic and R-allylic hydroxy groups.

Discussion There is considerable evidence that the reactive metabolite of tamoxifen responsible for DNA adduct formation in rodents is the result of oxidative metabolism by hepatic cytochrome P450 enzymes (24, 25). Among the various metabolites of tamoxifen, R-hydroxytamoxifen was shown to possess exceptionally high DNA-binding

Tamoxifen Bioactivation Model Studies

activation in rat hepatocytes (15) and was further found to react with DNA in the absence of metabolizing enzymes. The key step in the mechanism of DNA adduct formation was proposed to involve formation of a reactive sulfate, phosphate, or glucuronide conjugate of R-hydroxytamoxifen followed by loss of the conjugate generating a highly reactive carbocation (Scheme 1). In support of this, incubation of R-acetoxytamoxifen with DNA results in a dramatic increase in DNA adducts (1 in 50 bases) compared to R-hydroxytamoxifen (1 in 105 bases) (26). In contrast, droloxifene which contains a 3-hydroxy substituent in the B ring is not carcinogenic in rats suggesting there could be dramatic substituent effects on the bioactivation mechanism (13). In order to probe these substituent differences, we synthesized R-hydroxy analogs of tamoxifen, 4-hydroxytamoxifen and droloxifene, and explored the phase II metabolism of these compounds. Our first surprise was the substituent effect on reaction of these analogs with GSH. At pH 7.4, 37 °C, only compounds with para hydroxy substituents (2, 4) reacted with GSH to give diastereomeric products. The mechanism of the reaction did not involve dehydration to a quinone methide and reaction with GSH since product studies only showed conjugates resulting from direct SN2 displacement of the hydroxy group and not 1,6- and 1,8-Michael addition conjugates expected from GSH trapping of a quinone methide (20, 21). A para methoxy substituent is much less effective in stabilizing the partial positive charge at the R-carbon [σ+(OH) ) -0.92, σ+(OCH3) ) -0.78 (27)], which likely explains why no reaction with GSH was observed with 1 at pH 7.4. Similarly, no resonance stabilization can occur with a meta hydroxy group [σmeta(OH) ) 0.12], and even under strongly acidic conditions very little reaction with GSH was detected for 3. The additional aryl substituent in 4 and 5 does show a modest increase in the acid-catalyzed rate of reaction with GSH as expected from the σ value [σ+(C6H5) ) -0.17]; however, the effect is not sufficient to activate 5 under physiological conditions. In support of these model studies, calculations have shown that the R-carbocation of droloxifene is significantly less stable (∼2 kcal/mol) than the R-carbocation of tamoxifen (28). There are very few reports in the literature of reaction of benzylic or allylic alcohols with GSH. Dehydroretronecine, a carcinogenic metabolite of the pyrrolizidine alkaloid monocrotaline, reacts with either GSH or cysteine at the benzylic alcohol carbon losing water and generating 7-glutathionyl dehydroretronecine (29). Similarly, microsomal incubations with the benzyl alcohol metabolite of the synthetic antioxidant butylated hydroxytoluene (BHT) gave the same GSH conjugate as would be formed from GSH trapping of the BHT quinone methide (J. L. Bolton, and J. A. Thompson, unpublished results). Formation of the adduct did not depend on NADPH and was only observed if the reactions were “quenched” with PCA since quenching with organic solvents gave no GSH conjugate (30). These data illustrate the importance of avoiding acid workup procedures in metabolism experiments when para benzyl alcohols of phenols or aromatic ethers are potential metabolites. Incubations with rat liver cytosol and GSH gave 4-fold more GSH conjugate from 2 as compared to heatinactivated cytosol or no cytosol. Similarly, increases in GSH adduct formation were observed during experiments with commercial rat liver GST. Finally, the studies with GST substrates and inhibitors in cytosolic incubations

Chem. Res. Toxicol., Vol. 10, No. 8, 1997 893

suggest a role for this enzyme in catalyzing the nucleophilic substitution reaction. We have found no examples in the literature of this type of phase II reaction, and our data suggest it may play a role in the metabolism of other 4-hydroxybenzyl alcohols. Dramatic substituent effects were also observed with the PAPS-mediated sulfotransferase pathway. In incubations with rat liver cytosol and PAPS, only the para hydroxy analogs 2 and 4 were metabolized to sulfates at the R-allylic alcohol position in addition to sulfation of the phenolic hydroxyl group. Inhibitors of phenol sulfotransferase (PCP, DCNP) essentially abolished sulfate ester formation, whereas the hydroxysteroid sulfotransferase inhibitor DHEA had little effect on either aliphatic or phenolic sulfation. The meta hydroxy analogs 3 and 5 only gave one sulfated metabolite presumably resulting from phenol sulfation, and no sulfated metabolites were detected with the R-hydroxytamoxifen analog 1. In terms of the bioactivation pathway for tamoxifen, these data might suggest that sulfation of R-hydroxytamoxifen followed by carbocation formation may represent a minor metabolic pathway. It is quite possible, however, that other reactive esters of R-hydroxytamoxifen could be formed that could account for the group II DNA adducts. Finally, only R,4-dihydroxytamoxifen appears to be a likely candidate for sulfation of the R-hydroxy group in addition to formation of the phenolic sulfate at the 4-position. These data support the premise that the observed increases in group I DNA adducts in rats treated with the sulfotransferase inhibitor PCP and tamoxifen (12) are due to inhibition of sulfation of the proximate carcinogen 4-hydroxytamoxifen. Our data do not explain why the group II DNA adducts are depleted with PCP treatment. Inhibition of sulfotransferasemediated bioactivation of R-hydroxytamoxifen has been one explanation; however, our data suggest aliphatic sulfation of R-hydroxytamoxifen represents a minor metabolic pathway. An alternative explanation is inhibition of acetylation or glucuronidation of R-hydroxytamoxifen since PCP is also known to be a potent inhibitor of acetyltransferase and UDP glucuronosyltransferase activity (31, 32). However, it should be noted that our data are based on small model compounds, and results with tamoxifen and its metabolites remain to be determined. Finally, to the best of our knowledge, this is the first structure-activity study on sulfotransferasemediated sulfation of phenols and alcohol function groups where both substituents are present within the same molecule. In conclusion, we have shown that under strongly acidic conditions para-oxo-substituted R-hydroxybutenylbenzenes react with GSH to form conjugates in virtually 100% yield, whereas meta phenolic substrates do not react with GSH to any significant extent. Under physiological conditions, only the para hydroxy analogs gave GSH conjugates, a reaction which seems to be catalyzed by isoforms of glutathione S-transferase. Structure determination of GSH conjugates and kinetic studies clearly supported the adduct formation to proceed through a nucleophilic substitution reaction and not the quinone methide pathway. Finally, similar substituent effects were observed in the sulfotransferase-mediated formation of R-hydroxy sulfate esters in that only the para hydroxy analogs formed sulfate conjugates at the R-hydroxy position. Future work will explore whether these substituent effects extend to R-hydroxytamoxifen and other triaryl analogs in order to develop structure-reactivity

894 Chem. Res. Toxicol., Vol. 10, No. 8, 1997

relationships which would more clearly define the role of R-hydroxytamoxifen in the carcinogenicity of tamoxifen.

Acknowledgment. This research was supported by a grant from Pfizer, Inc. The electrospray-MS expertise provided by Dr. Richard B. van Breemen (Liquid Chromatography-Mass Spectrometry Laboratory, University of Illinois at Chicago) is gratefully appreciated.

References (1) Furr, B. J. A., and Jordan, V. C. (1984) The pharmacology and clinical uses of tamoxifen. Pharmacol. Ther. 25, 127-205. (2) Fisher, B., and Redmond, C. (1991) New perspective on cancer of the contralateral breast: a marker for assessing tamoxifen as a preventive agent. J. Natl. Cancer Inst. 83, 1278-1280. (3) King, C. M. (1994) Tamoxifen and induction of cancer. Carcinogenesis 16, 1449-1454. (4) Williams, G. M., Iatropoulos, M. J., Djordjevic, M. W., and Kaltenberg, O. P. (1993) The triphenylethylene drug tamoxifen is a strong carcinogen in the rat. Carcinogenesis 14, 315-317. (5) Fendel, K. C., and Zimniski, S. J. (1992) Role of tamoxifen in the induction of hormone-independent rat mammary tumors. Cancer Res. 52, 235-237. (6) Killackey, M. A., Hakes, T. B., and Price, V. K. (1985) Endometrial adenocarcinoma in breast cancer patients receiving antiestrogens. Cancer Treat. Rep. 69, 237-238. (7) Fornander, T., Cedermark, B., Matteson, A., Skoog, L., Theve, T., Askergren, J., Rutqvist, I. E., Glas, U., Silljverswaard, C., Somell, A., Wilking, N., and Hjolmar, M. J. (1989) Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 1, 117-119. (8) Hemminki, K., Rajaniemi, H., Lindahl, B., and Moberger, B. (1996) Tamoxifen-induced DNA adducts in endometrial samples from breast cancer patients. Cancer Res. 56, 4374-4377. (9) Henderson, B. E., Ross, R., and Bernstein, L. (1988) Estrogens as a cause of human cancer: The Richard and Hinda Rosenthal Foundation Award Lecture. Cancer Res. 48, 246-253. (10) Potter, G. A., McCague, R., and Jarman, M. (1994) A mechanistic hypothesis for DNA adduct formation by tamoxifen following hepatic oxidative metabolism. Carcinogenesis 5, 439-442. (11) Moorthy, B., Sriram, P., Pathak, D. N., Bodell, W. J., and Randerath, K. (1996) Tamoxifen metabolic activation: Comparison of DNA adducts formed by microsomal and chemical activation of tamoxifen and 4-hydroxytamoxifen with DNA adducts formed in vivo. Cancer Res. 56, 53-57. (12) Randerath, K., Moorthy, B., Mabon, N., and Sriram, P. (1994) Tamoxifen: evidence by 32P-postlabeling and use of metabolic inhibitors for two distinct pathways leading to mouse hepatic DNA adduct formation and identification of 4-hydroxytamoxifen as a proximate metabolite. Carcinogenesis 15, 2087-2094. (13) White, I. N. H., de Matteis, F., Davies, A., Smith, L. L., CroftonSleigh, C., Venitt, S., Hewer, A., and Phillips, D. H. (1992) Genotoxic potential of tamoxifen and analogues in female Fisher F344/n rats, DBA/2 and C57BL/6 and in human MCL-5 cells. Carcinogenesis 13, 2197-2203. (14) Phillips, D. H., Potter, G. A., Horton, M. N., Hewer, A., CroftonSleigh, C., Jarman, M., and Venitt, S. (1994) Reduced genotoxicity of [D5-ethyl]-tamoxifen implicates R-hydroxylation of the ethyl group as a major pathway of tamoxifen activation to a liver carcinogen. Carcinogenesis 15, 1487-1492. (15) Phillips, D. H., Carmichael, P. L., Hewer, A., Cole, K. J., and Poon, G. K. (1994) R-Hydroxytamoxifen, a metabolite of tamoxifen with exceptionally high DNA-binding activity in rat hepatocytes. Cancer Res. 54, 5518-5522. (16) Phillips, D. H., Carmichael, P. L., Hewer, A., Cole, K. J., Hardcastle, I. R., Poon, G. K., Keogh, A., and Strain, A. J. (1996)

Ramakrishna et al.

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27) (28)

(29)

(30)

(31)

(32)

Activation of tamoxifen and its metabolite R-hydroxytamoxifen to DNA-binding products: comparisons between human, rat and mouse hepatocytes. Carcinogenesis 17, 89-94. Hard, G. C., Iatropoulos, M. J., Jordan, K., Radi, L., Kaltenberg, O. P., Imondi, A. R., and Williams, G. M. (1993) Major Difference in the Hepatocarcinogenicity and DNA Adduct Forming Ability between Toremifene and Tamoxifen in Female Crl:CD(BR) Rats. Cancer Res. 53, 4534-4541. Surh, Y., Kwon, H., and Tannenbaum, S. R. (1993) Sulfotransferase-mediated activation of 4-hydroxy- and 3,4-dihydroxy-3,4dihydrocyclopenta[c,d]pyrene, major metabolites of cyclopenta[c,d]pyrene. Cancer Res. 53, 1017-1022. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the folin-phenol reagent. J. Biol. Chem. 193, 265-275. Bolton, J. L., Comeau, E., and Vukomanovic, V. (1995) The influence of 4-alkyl substituents on the formation and reactivity of 2-methoxy-quinone methides: Evidence that extended π-conjugation dramatically stabilizes the quinone methide formed from eugenol. Chem.-Biol. Interact. 95, 279-290. Bolton, J. L., Acay, N. M., and Vukomanovic, V. (1994) Evidence that 4-allyl-ortho-quinones spontaneously rearrange to their more electrophilic quinone methides: Potential bioactivation mechanism for the hepatocarcinogen safrole. Chem. Res. Toxicol. 7, 443450. Moorthy, B., and Randerath, K. (1996) Pentachlorophenol enhances 9-hydroxybenzo[a]pyrene-induced hepatic DNA adduct formation in vivo and inhibits microsomal epoxide hydrolase and glutathione S-transferase activities in vitro: likely inhibition of epoxide detoxication by pentachlorophenol. Arch. Toxicol. 70, 696-703. Chasseaud, L. F. (1979) The role of glutathione and glutathione S-transferases in the metabolism of chemical carcinogens and other electrophilic agents. Adv. Cancer Res. 29, 175-274. Pathak, D. N., and Bodell, W. J. (1994) DNA adduct formation by tamoxifen with rat and human liver microsomal activation systems. Carcinogenesis 15, 529-532. Pathak, D. N., Pongracz, K., and Bodell, W. J. (1995) Microsomal and peroxidase activation of 4-hydroxytamoxifen to form DNA adducts: Comparison with DNA adducts formed in SpragueDawley rats treated with tamoxifen. Cancinogenesis 16, 11-15. Osborne, M. R., Hewer, A., Hardcastle, I. R., Carmichael, P. L., and Phillips, D. H. (1996) Identification of the major tamoxifendeoxyguanosine adduct formed in the liver DNA of rats treated with tamoxifen. Cancer Res. 56, 66-71. Lowry, T. H., and Richardson, K. S. (1987) Mechanism and theory in organic chemistry, 3rd ed., p 144, Harper & Row, New York. Kuramochi, H. (1996) Conformational studies and electronic structures of tamoxifen and toremifene and their allylic carbocations proposed as reactive intermediates leading to DNA adduct formation. J. Med. Chem. 39, 2877-2886. Robertson, C. A., Seymour, J. L., Hsia, M. T., and Allen, J. R. (1977) Covalent interaction of dehydroretronecine, a carcinogenic metabolites of the pyrrolizidine alkaloid monocrotaline, with cysteine and glutathione. Cancer Res. 37, 3141-3144. Tajima, K., Yamamoto, K., and Mizutani, T. (1985) Formation of a glutathione conjugate from butylated hydroxytoluene by rat liver microsomes. Biochem. Pharmacol. 34, 2109-2114. Saito, K., Shinohara, A., Kamataki, T., and Kato, R. (1985) Metabolic activation of mutagenic N-hydroxyarylamines by Oacetyltransferase in Salmonella tryphimurium TA98. Arch. Biochem. Biophys. 239, 286-295. Malfatti, M. A., Connors, M. S., Mauthe, R. J., and Felton, J. S. (1996) The capability of rat colon tissue slices to metabolize the cooked-food carcinogen 2-amino-1-methyl-6-phenylimidazol[4,5b] pyridine. Cancer Res. 56, 2550-2555.

TX970060R