Synthesis and Reactivity of a Potential Carcinogenic Metabolite of

cancer (3) and an increase in the incidence of endometrial cancer (4). These troubling side effects demonstrate the need to further explore the potent...
0 downloads 0 Views 188KB Size
Download PDF
Chem. Res. Toxicol. 2000, 13, 53-62

53

Synthesis and Reactivity of a Potential Carcinogenic Metabolite of Tamoxifen: 3,4-Dihydroxytamoxifen-o-quinone Fagen Zhang, Peter W. Fan, Xuemei Liu, Lixin Shen, Richard B. van Breemen, 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 Received August 3, 1999

Although tamoxifen is approved for the treatment of hormone-dependent breast cancer as well as for the prevention of breast cancer in high-risk women, several studies in animal models have shown that tamoxifen is heptocarcinogenic, and in humans, tamoxifen has been associated with an increased risk of endometrial cancer. One potential mechanism of tamoxifen carcinogenesis could involve metabolism of tamoxifen to 3,4-dihydroxytamoxifen followed by oxidation to a highly reactive o-quinone which has the potential to alkylate and/or oxidize cellular macromolecules in vivo. In the study presented here, we synthesized the 3,4dihydroxytamoxifen, prepared its o-quinone chemically and enzymatically, and studied the reactivity of the o-quinone with GSH and deoxynucleosides. The E (trans) and Z (cis) isomers of 3,4-dihydroxytamoxifen were synthesized using a concise synthetic pathway (four steps). This approach is based on the McMurry reaction between the key 4-(2-chloroethoxy)-3,4methylenedioxybenzophenone and propiophenone, followed by selective removal of the methylenedioxy ring of (E,Z)-1-[4-[2-(N,N-dimethylamino)ethoxy]phenyl]-1-(3,4-methylenedioxyphenyl)2-phenyl-1-butene with BCl3. Oxidation of 3,4-dihydroxytamoxifen by activated silver oxide or tyrosinase gave 3,4-dihydroxytamoxifen-o-quinone as a mixture of E and Z isomers. The resulting o-quinone has a half-life of approximately 80 min under physiological conditions. Reaction of the o-quinone with GSH gave two di-GSH conjugates and three mono GSH conjugates. Incubation of 3,4-dihydroxytamoxifen with GSH in the presence of microsomal P450 gave the same GSH conjugates which were also detected in incubations with human breast cancer cells (MCF-7). Reaction of 3,4-dihydroxytamoxifen-o-quinone with deoxynucleosides gave only thymidine and deoxyguanosine adducts; neither deoxyadenosine nor deoxycytosine adducts were detected. Preliminary studies conducted with human breast cancer cell lines showed that 3,4-dihydroxytamoxifen exhibited cytotoxic potency similar to that of 4-hydroxytamoxifen and tamoxifen in an estrogen receptor negative (ER-) cell line (MDAMB-231); however, in the ER+ cell line (MCF-7), the catechol metabolite was about half as toxic as the other two compounds. Finally, in the presence of microsomes and GSH, 4-hydroxytamoxifen gave predominantly quinone methide GSH conjugates as reported in the previous paper in this issue [Fan, P. W., et al. (2000) Chem. Res. Toxicol. 13, XX-XX]. However, in the presence of tyrosinase and GSH, 4-hydroxytamoxifen was primarily converted to o-quinone GSH conjugates. These results suggest that the catechol metabolite of tamoxifen has the potential to cause cytotoxicity in vivo through formation of 3,4-dihydroxytamoxifeno-quinone.

Introduction Tamoxifen is widely used in the treatment of all stages of hormone-dependent breast cancer (1). Recent largescale clinical trials of tamoxifen as a chemopreventive agent in women considered at risk of developing breast cancer demonstrated protection in 45% of the subjects (2). Although the Food and Drug Administration has approved the use of tamoxifen for the prevention of breast cancer, tamoxifen has been associated with uterine cancer (3) and an increase in the incidence of endometrial * To whom correspondence should be addressed: Department of Medicinal Chemistry and Pharmacognosy (M/C 781), College of Pharmacy, University of Illinois at Chicago, 833 S. Wood St., Chicago, IL 60612-7231. Telephone: (312) 996-5280. Fax: (312) 996-7107. E-mail: [email protected].

cancer (4). These troubling side effects demonstrate the need to further explore the potential carcinogenic mechanisms of tamoxifen. Although the carcinogenic effects of tamoxifen have been attributed to hormonal properties (5), there is an interest in metabolites of tamoxifen acting as chemical carcinogens by binding to cellular macromolecules (68). Tamoxifen can be metabolized to at least three different electrophilic metabolites, including a carbocation (9), quinone methide (10), and o-quinone (11, 12; Scheme 1). The o-quinone resulting from oxidation of the catechol metabolite, 3,4-dihydroxytamoxifen (13, 14), can alkylate amino acid residues on proteins (11). In addition, it is possible that this o-quinone has the potential to cause damage to cellular DNA through generation of

10.1021/tx990145n CCC: $19.00 © 2000 American Chemical Society Published on Web 12/21/1999

54

Chem. Res. Toxicol., Vol. 13, No. 1, 2000

Zhang et al.

Scheme 1. Metabolism of Tamoxifen to Electrophilic Metabolites

Scheme 2. Synthesis of 3,4-Dihydroxytamoxifena

tyrosinase converted 4-hydroxytamoxifen into 3,4-dihydroxytamoxifen-o-quinone GSH conjugates.

Materials and Methods

a Reagents and conditions: (a) (CF CO) O, room temperature, 3 2 96 h; (b) propiophenone, TiCl4, Zn, THF, reflux, 4 h; (c) NH(CH3)2 in CH3OH, reflux, 96 h; (d) BCl3 in CH2Cl2, room temperature, 72 h.

reactive oxygen species, and/or alkylation of DNA. In the study presented here, we have developed an efficient synthesis of (E/Z)-3,4-dihydroxytamoxifen (Scheme 2) for investigating this potentially carcinogenic pathway in vitro. 3,4-Dihydroxytamoxifen was oxidized either chemically or enzymatically to the o-quinone which reacted with GSH, giving mono-, di-, and tri-GSH conjugates. These conjugates were also observed in microsomal incubations and in MCF-7 cells. We also found the o-quinone formed adducts with deoxyguanosine and thymidine. In addition, the relative cytotoxicity of 3,4dihydroxytamoxifen was compared with that of tamoxifen and 4-hydroxytamoxifen in human breast cancer cell lines. Finally, we showed that 4-hydroxytamoxifen primarily gave quinone methide GSH conjugates in microsomal incubations as reported in ref 15, whereas

Caution: 3,4-Dihydroxytamoxifen-o-quinone was handled in accordance with the NIH Guidelines for the Laboratory Use of Chemical Carcinogens (16). All chemicals were purchased from Aldrich Chemical Co. (Milwaukee, WI), Fisher Scientific (Itasca, IL), or Sigma (St. Louis, MO) unless stated otherwise. [3H]GSH (glycine-2-3H) was obtained from New Life Science Products, Inc. (Boston, MA) and diluted to a specific activity of 50 nCi/ nmol. 4-Hydroxytamoxifen was synthesized according to the literature procedure (17). 3,4-Dihydroxytamoxifen was synthesized as described below (Scheme 2). HPLC Methodology. Three general methods were used to analyze and separate the various metabolites, conjugates, and deoxynucleoside adducts. All retention times reported in the text were obtained using method A. Method A. Analytical HPLC analysis was performed using a 4.6 mm × 150 mm Ultrasphere C-18 column (Beckman) on a Shimazu LC-10A gradient HPLC system equipped with an SIL10A autoinjector and an SPD-10AV detector set at 280 nm. The mobile phase consisted of 20% methanol in 0.25% perchloric acid/0.25% acetic acid (pH 3.5) at a flow rate of 1.0 mL/min for 5 min, which was increased to 35% CH3OH over the course of 1 min, to 60% over the course of the next 39 min, and then to 90% CH3OH over the course of the the last 5 min. Method B. For isolation of conjugates, a semipreparative method was developed using a 10 mm × 250 mm Ultrasphere C-18 column (Beckman) and a mobile phase consisting of 20% methanol in 0.25% perchloric acid/0.25% acetic acid (pH 3.5) at a flow rate of 3.0 mL/min for 5 min, which was increased to 35% CH3OH over the course of 1 min, to 60% over the course of the next 39 min, and then to 90% CH3OH over the course of the last 5 min. Method C. LC/MS analysis was carried out using a 4.6 mm × 150 mm Ultrasphere C-18 column (Beckman) with a HewlettPackard (Palo Alto, CA) 1050L gradient HPLC system equipped with a photodiode array UV/vis absorbance detector set at 230350 nm and a quadrupole mass spectrometer (described below). The mobile phase consisted of 20% methanol in 0.5% ammonium

Synthesis and Reactivity of 3,4-Di-OHTAM-o-quinone acetate (pH 3.5) at a flow rate of 1.0 mL/min for 3 min, which was increased to 90% CH3OH over the course of 53 min. Mass Spectrometry. LC/MS was carried out using a HewlettPackard 5989B mass spectrometer equipped with an Analytica (Branford, CT) electrospray ion source and ion guide. LC/MS/ MS1 spectra were obtained using a Micromass (Manchester, U.K.) Quattro II triple-quadrupole mass spectrometer equipped with an electrospray ionization detector and a Hewlett-Packard 1050 HPLC system. Collision-induced dissociation (CID) was carried out using argon as the collision gas at 2.7 µbar and collision energies between 25 and 70 eV. Additional details are provided in ref 15. Electron impact (EI) and chemical ionization (CI) exact mass measurements were obtained using a Finnigan MAT (Bremen, Germany) MAT90 magnetic sector mass spectrometer at a resolving power of 10 000. Synthesis of 3,4-Dihydroxytamoxifen (Scheme 2). (1) (2Chloroethoxy)benzene 1. (2-Chloroethoxy)benzene was synthesized as described in the literature (18). Briefly, thionyl chloride (50 mL, 81.6 g, 68.4 mmol) was slowly added to a flask containing 2-phenoxyethanol (32.6 g, 0.24 mol). The evolving HCl was absorbed into a NaOH-saturated solution via an inverted funnel. The mixture was refluxed for 8 h, at which time extra thionyl chloride was removed under reduced pressure. Pyridine (0.1 mL) was added, and the mixture was gradually heated to 160 °C to complete the reaction. The final residue was purified by distillation to give 1 (28.91 g, 78%): bp 90 °C at 5 mmHg; 1H NMR (CDCl3) δ 3.77 (t, 2H, J ) 5.9 Hz, CH2Cl), 4.18 (t, 2H, J ) 5.9 Hz, OCH2), 6.93 (m, 2H, ArH), 6.98 (m, 1H, ArH), 7.27 (m, 2H, ArH); 13C NMR (CDCl3) δ 42.09, 68.06, 114.86, 121.55, 129.74, 158.33. (2) 4-(2-Chloroethoxy)-3,4-methylenedioxybenzophenone 3. Piperonylic acid 2 (13.2 g, 80 mmol), (2-chloroethoxy)benzene 1 (13.8 g, 88 mmol), and trifluoroacetic anhydride (3 mL, 84 mmol) were added to a 50 mL flame-dried flask, and the mixture was stirred for 96 h at room temperature under a stream of nitrogen. The solution was poured into a saturated NaHCO3 solution (200 mL) and extracted with ethyl acetate (3 × 200 mL), and the combined organic layers were washed with 10% KOH (2 × 200 mL) and saturated NaCl (2 × 200 mL). The organic layer was dried over anhydrous Na2SO4 and filtered, and the solvent was removed under reduced pressure. The final residue was recrystallized from ethyl acetate to give 3 (14.5 g, 60% yield): 1H NMR (CDCl3) δ 3.84 (t, 2H, J ) 5.7 Hz, CH2Cl), 4.29 (t, 2H, J ) 5.7 Hz, OCH2), 6.04 (s, 2H, OCH2O), 6.84 (d, 1H, J ) 7.8 Hz, ArH), 6.96 (d, 2H, J ) 8.1 Hz, ArH), 7.28 (s, 1H, ArH), 7.33 (d, 1H, J ) 7.8 Hz, ArH), 7.76 (d, 2H, J ) 8.1 Hz, ArH); 13C NMR (CDCl3) δ 41.8, 68.1, 101.9, 107.7, 109.9, 114.1, 126.3, 131.2, 132.3, 132.4, 147.9, 151.2, 161.5, 193.9; EIMS m/z (rel intensity) 304 (100%) [M]+, 306 (33%) [M + 2]+; exact mass (EI) calcd for C16H13ClO4 [M]+ 304.0502, found 304.0505; mp 105-106 °C. (3) (E,Z)-1-[4-(2-Chloroethoxy)phenyl]-1-(3,4-methylenedioxyphenyl)-2-phenyl-1-butene 4. TiCl4 (9.12 g, 48 mmol) was added dropwise to a stirred suspension of zinc powder (6.12 g, 96 mmol) in THF (80 mL) at -10 °C under nitrogen. The resulting mixture was heated at reflux for 1 h. The suspension was cooled to room temperature, and a mixture of 3 (4.84 g, 16 mmol) and propiophenone (2.14 g, 16 mmol) in THF (55 mL) was added dropwise. The mixture was refluxed and stirred for 3.5 h, cooled to room temperature, and combined with 10% aqueous potassium carbonate (100 mL). The resulting mixturewas stirred for 5 min and extracted with ethyl acetate (2 × 200 mL). The combined organic extracts were dried over anhydrous 1 Abbreviations: 3,4-di-OHTAM, 3,4-dihydroxytamoxifen; 4-OHTAM, 4-hydroxytamoxifen; 3,4-di-OHTAM-triSG, triglutathione conjugates of 3,4-di-OHTAM; 3,4-di-OHTAM-diSG, diglutathione conjugates of 3,4-di-OHTAM; 3,4-di-OHTAM-SG, glutathione conjugates of 3,4-diOHTAM; 3,4-di-OHTAM-diCYS, dicysteine conjugates of 3,4-diOHTAM; 3,4-di-OHTAM-CYS, cysteine conjugates of 3,4-di-OHTAM; 3,4-di-OHTAM-T, thymidine adducts of 3,4-di-OHTAM; 3,4-diOHTAM-dG, deoxyguanosine adducts of 3,4-di-OHTAM; P450, cytochrome P450; T, thymidine; dG, deoxyguanosine; electrospray-MS, electrospray mass spectrometry; MS/MS, tandem mass spectrometry.

Chem. Res. Toxicol., Vol. 13, No. 1, 2000 55 sodium sulfate and evaporated to give a yellow oil, which was purified by column chromatography (silica gel) using ethyl acetate/hexane (1:9, v/v) as the eluent to give 4 as a mixture of E (trans) and Z (cis) isomers (10:7 Z:E) (4.71 g, 73% yield): 1H NMR (CDCl3) δ 0.89 (m, 6H, cis and trans CH3), 2.47 (m, 4H, cis and trans CH2), 3.71 (t, 2H, J ) 6 Hz, trans CH2Cl), 3.82 (t, 2H, J ) 6 Hz, cis CH2Cl), 4.08 (t, 2H, J ) 6 Hz, trans OCH2), 4.24 (t, 2H, J ) 6 Hz, cis OCH2), 5.79 (s, 2H, cis OCH2O), 5.94 (s, 2H, trans OCH2O), 6.35-6.44 (m, 3H, cis ArH), 6.87-6.90 (m, 7H, trans ArH), 7.06 (d, 2H, cis ArH), 7.09-7.12 (m, 12H, cis and trans ArH); EI-MS m/z (rel intensity) 406 (100%) [M]+, 408 (29%) [M + 2]+; exact mass (EI) calcd for C25H23ClO3 [M]+ 406.1336, found 406.1332. (4) (E,Z)-1-[4-[2-(N,N-Dimethylamino)ethoxy]phenyl]-1(3,4-methylenedioxyphenyl)-2-phenyl-1-butene 5. A mixture of 4 (3.5 g, 8.64 mmol) and dimethylamine (2 M in methanol, 40 mL) was refluxed for 96 h under a stream of nitrogen. The mixture was then concentrated and partitioned between Et2O (300 mL) and KOH (0.1 M, 200 mL). The Et2O solution was concentrated, and the residue was purified by a flash chromatography (silica gel). Elution with hexane/ethyl acetate/triethylamine (3:1:0.4, v/v/v) gave (E)-5A (trans) (100 mg) containing 24% (Z)-5B (cis), a mixture of (E)- and (Z)-5 (1:1 E:Z, 2.9 g), and the (Z)-5B (cis) isomer (0.2 g) containing 10% 5A. The total yield was 89%. For 5A: 1H NMR (CDCl3) δ 0.93 (m, 3H, CH3), 2.31 (s, 6H, NCH3), 2.47 (m, 2H, CH2), 2.69 (t, 2H, J ) 6 Hz, CH2N), 3.95 (t, 2H, J ) 6 Hz, OCH2), 5.97 (s, 2H, OCH2O), 6.59 (d, 2H, ArH), 6.70-6.81 (m, 5H, ArH), 7.10-7.17 (m, 5H, ArH); CI-MS (positive ion, methane) m/z (rel intensity) 416 (100%) [M + H]+; exact mass (EI) calcd for C27H29NO3 [M]+ 415.2147, found 415.2146. For 5B: 1H NMR (CDCl3) δ 0.93 (m, 3H, CH3), 2.45 (m, 8H, NCH3, CH2), 2.87 (t, 2H, J ) 6 Hz, CH2N), 4.16 (t, 2H, J ) 6 Hz, OCH2), 5.80 (s, 2H, OCH2O), 6.346.35 (m, 2H, ArH), 6.70-6.83 (m, 5H, ArH), 7.10-7.17 (m, 5H, ArH); CI-MS (positive ion, methane) m/z (rel intensity) 416 (100%) [M + H]+; exact mass (EI) calcd for C27H29NO3 [M]+ 415.2147, found 415.2146. (5) (E/Z)-3,4-Dihydroxytamoxifen. To a flame-dried flask were added compound 5 (1.35 g, 2.94 mmol), a 2 M solution of boron trichloride in methylene chloride (11 mL, 22 mmol), and 1,2-dichloroethane (60 mL) under a stream of nitrogen (19). The solution was stirred for 72 h at room temperature under a nitrogen atmosphere. MeOH (70 mL) was gradually added to the solution, and the mixture was stirred for 40 min. The solvent was removed under reduced pressure, and saturated NaHCO3 (80 mL) and MeOH (10 mL) were added to the residue. The pH of the solution was adjusted to 8.5-9.0 with 1 M NaOH solution, and the resulting solution was extracted with ethyl acetate (2 × 200 mL). The combined organic layers were washed with saturated NaHCO3 (80 mL) and saturated NaCl (80 mL) and dried over anhydrous sodium sulfate. After filtration, the solvent was removed under reduced pressure, and the final residue was purified by flash chromatography (silica gel) using CHCl3/CH3OH (7:3, v/v) as the eluent to give 3,4-dihydroxytamoxifen (500 mg, 39% yield) as a mixture of E (trans) and Z (cis) isomers (1:1 E:Z ratio): 1H NMR (CD3OD) δ 0.75 (m, 6H, cis and trans CH3), 2.14 (s, 6H, trans NCH3), 2.21 (s, 6H, cis NCH3), 2.37 (m, 4H, cis and trans CH2), 2.54 (t, 2H, J ) 5.4 Hz, trans CH2N), 2.64 (t, 2H, J ) 5.4 Hz, cis CH2N), 3.78 (s, 2H, J ) 5.4 Hz, trans OCH2), 3.95 (t, 2H, J ) 5.4 Hz, cis OCH2), 6.03-6.27 (m, 3H, cis ArH), 6.39-6.65 (m, 7H, trans ArH), 6.77 (d, 2H, trans ArH), 6.91-7.00 (m, 12H, cis and trans ArH); CI-MS (positive ion, methane) m/z (rel intensity) 404 (100%) [M + H]+; exact mass (EI) calcd for C26H29NO3 [M]+ 403.2147, found 403.2136. LC/MS analysis (method C) gave the following additional spectroscopic information. For 3,4-dihydroxytamoxifen isomer 1: UV (CH3OH/H2O) 242, 290 nm; positive ion electrospray MS m/z (rel intensity) 404 (100%) [MH]+; retention time of 37 min. For 3,4-dihydroxytamoxifen isomer 2: UV (CH3OH/H2O) 239, 289 nm; positive ion electrospray MS m/z (rel intensity) 404 (100%) [M + H]+; retention time of 39 min.

56

Chem. Res. Toxicol., Vol. 13, No. 1, 2000

Oxidation of 3,4-Dihydroxytamoxifen by Silver Oxide. To a 10 mL flask were added 3,4-dihydroxytamoxifen (4 mg), acetonitrile (4 mL), and activated silver oxide (400 mg; 20). The mixture was stirred for 15 min at 60 °C. After filtration, the solution was concentrated to a final volume of 1 mL, and aliquots (25 µL) of this solution was analyzed directly by LC/MS (method C). For 3,4-dihydroxytamoxifen-o-quinone: UV (CH3OH/H2O) 232, 375, 475 nm; positive ion electrospray MS m/z (rel intensity) 402 (100%) [M + H]+; retention time of 36 min. Reaction of 3,4-Dihydroxytamoxifen-o-quinone with Cysteine or GSH. A solution of 3,4-dihydroxytamoxifen-oquinone in acetonitrile was concentrated to a final volume of 4.5 mL (∼0.01 M) and combined with cysteine or GSH (0.1 M) in incubation buffer (4 mL, 50 mM phosphate buffer, pH 7.4). After the mixture had been stirred at room temperature for 5 min, 1 mL of PCA was added, and the solution was concentrated to remove the remaining acetonitrile under a stream of N2. The final solution was extracted with 2 × 10 mL of ether and centrifuged for 6 min at 13 000 rpm. The conjugates were purified by semipreparative HPLC (method B). For 3,4-diOHTAM-diCYS: positive ion electrospray MS m/z (rel intensity) 642 [M + H]+ (100%); MS/MS with CID of m/z 642 (rel intensity) m/z 554 (20%) [MH - Cys]+; retention time of 16 min. For 3,4-di-OHTAM-CYS 1: 1H NMR (CD3OD) δ 0.90 (t, 3H, CH3), 2.5 (m, 2H, CH2), 2.83 (m, 2H, CH2S), 2.93 (s, 6H, NCH3), 3.19 (3.51, 1H, CH), 3.43 (t, 2H, J ) 4.5 Hz, NCH2), 4.20 (t, 2H, J ) 4.5 Hz, OCH2), 6.35 (s, 1H, ArH), 6.40 (s, 1H, ArH), 6.72 (d, 2H, J ) 9 Hz, ArH), 7.16 (m, 7H, ArH); positive ion electrospray MS m/z (rel intensity) 523 (100%) [M + H]+; MS/MS of m/z 523 (rel intensity) m/z 435 (20%) [MH - Cys]+; retention time of 17 min. For 3,4-di-OHTAM-CYS 2: 1H NMR (CD3OD) δ 0.93 (t, 3H, CH3), 2.51 (m, 2H, CH2), 2.91 (s, 6H, NCH3), 2.94 (m, 2H, CH2S), 3.34 (t, 2H, J ) 4.5 Hz, NCH2), 3.51 (m, 1H, CH), 4.18 (t, 2H, J ) 4.5 Hz, OCH2), 6.65 (d, 2H, J ) 8.7 Hz, ArH), 6.72 (s, 1H, ArH), 6.77 (s, 1H, ArH), 6.82 (d, 2H, J ) 8.7 Hz, ArH), 7.11 (m, 5H, ArH); positive ion electrospray MS m/z (rel intensity) 523 (100%) [M + H]+; MS/MS with CID of m/z 523 (rel intensity) m/z 435 (20%) [MH - Cys]+; retention time of 20 min. For 3,4-di-OHTAM-CYS 3: positive ion electrospray MS m/z (rel intensity) 523 (100%) [M + H]+; retention time of 25 min. Structures of the GSH conjugates are shown in Scheme 2. For 3,4-di-OHTAM-diSG 1: UV (CH3OH/H2O) 248, 279, 316 nm; positive ion electrospray MS m/z (rel intensity) 1014 (100%) [M + H]+; MS/MS with CID of m/z 1014 (rel intensity) m/z 884 (20%) [MH - Glu]+; retention time of 29 min. For 3,4-diOHTAM-diSG 2: UV (CH3OH/H2O) 239, 274 nm; positive ion electrospray MS m/z (rel intensity) 1014 (100%) [M + H]+; MS/ MS with CID of m/z 1014 (rel intensity) m/z 884 (20%) [MH Glu]+; retention time of 31 min. For 3,4-di-OHTAM-SG 1: UV (CH3OH/H2O) 239, 274 nm; 1H NMR (CD3OD) δ 0.90 (t, 3H, CH3), 2.08 (m, 2H, Gluβ), 2.45 (m, 4H, Gluγ, CH2), 2.80 (m, 1H, Cysβ), 2.95 (s, 6H, NCH3), 3.15 (m, 1H, Cysβ), 3.53 (t, 2H, J ) 4.5 Hz, CH2N), 3.58 (m, 1H, GluR), 3.69 (d, 2H, GlyR), 4.21 (m, 1H, CysR), 4.36 (t, 2H, J ) 4.5 Hz, OCH2), 6.29 (s, 1H, ArH), 6.99 (s, 1H, ArH), 7.01 (d, 2H, J ) 8.4 Hz, ArH), 7.18 (m, 7H, ArH); positive ion electrospray MS m/z (rel intensity) 709 (100%) [M + H]+; MS/MS with CID of m/z 709 (rel intensity) m/z 580 (100%) [MH - Glu]+, 691 (80%) [MH - H2O]+; retention time of 34 min. For 3,4-di-OHTAM-SG 2: UV (CH3OH/H2O) 248, 284, 336 nm; 1H NMR (CD3OD) δ 0.92 (t, 3H, CH3), 2.08 (m, 2H, Gluβ), 2.43 (m, 2H, Gluγ), 2.54 (m, 2H, CH2), 2.88 (s, 6H, NCH3), 3.01 (m, 2H, Cysβ), 3.44 (t, 2H, J ) 4.5 Hz, CH2N), 3.62 (m, 1H, GluR), 3.68 (d, 2H, GlyR), 4.20 (t, 2H, J ) 4.5 Hz, OCH2), 4.27 (m, 1H, CysR), 6.57 (s, 1H, ArH), 6.65 (d, 2H, J ) 8.6 Hz, ArH), 6.78 (s, 1H, ArH), 6.82 (d, 2H, J ) 8.6 Hz, ArH), 7.12 (m, 5H, ArH); positive ion electrospray MS m/z (rel intensity) 709 (100%) [M + H]+; MS/MS m/z (rel intensity) 691 (85%) [MH H2O]+, 580 (80%) [MH - Glu]+; retention time of 37 min. For 3,4-di-OHTAM-SG 3: UV (CH3OH/H2O) 248, 285 nm; positive ion electrospray MS m/z (rel intensity) 709 (100%) [M + H]+, 691 (41%) [MH - H2O]+; MS/MS with CID of m/z 709 (rel intensity) m/z 580 (50%) [MH - Glu]+; retention time of 41 min.

Zhang et al. Kinetic Studies. The rate of disappearance of the o-quinone was determined by monitoring the apparent disappearance of the corresponding o-quinone GSH conjugates by HPLC as described previously (21, 22). 3,4-Dihydroxytamoxifen-o-quinone (0.5 mM) was incubated in 5 mL of potassium phosphate buffer (pH 7.4, 37 °C). Aliquots were removed at various times (250 µL) and the reactions quenched with GSH (5.0 mM) followed by perchloric acid (50 µL/mL). Aliquots of the supernatant (100 µL) were analyzed directly by HPLC using method A. The rate of o-quinone disappearance was determined from the apparent first-order decrease in the peak area ratio of the corresponding o-quinone GSH conjugates. Oxidation of 3,4-Dihydroxytamoxifen and 4-Hydroxytamoxifen by Tyrosinase. 3,4-Dihydroxytamoxifen or 4-hydroxytamoxifen was added as a solution in DMSO, and GSH was added in phosphate buffer to achieve a final concentration 0.5 or 5.0 mM, respectively. Reaction solutions containing tyrosinase (mushroom tyrosinase, 0.1 mg/mL) were held for 30 min at 37 °C in 50 mM phosphate buffer (pH 7.4, 1.0 mL total volume). Reactions were terminated by chilling in an ice bath followed by the addition of perchloric acid (50 µL/mL). For control incubations, GSH or tyrosinase was omitted. The reaction mixtures were centrifuged at 13 000 rpm for 6 min, and the solution was extracted using solid-phase extraction cartridges (Oasis, HLB, Waters Corp., Milford, MA), eluted with methanol, and concentrated to a final volume of 100 µL. Aliquots (25 µL) were analyzed directly by LC/MS using method C. In addition to the mono- and di-GSH conjugates described above, tri-GSH conjugates were also obtained (Scheme 2). For 3,4-diOHTAM-triSG 1: UV (CH3OH/H2O) 244, 270 nm; positive ion electrospray MS m/z (rel intensity) 1319 (100%) [M + H]+; retention time of 14 min. For 3,4-di-OHTAM-triSG 2: UV (CH3OH/H2O) 240, 280 nm; positive ion electrospray MS m/z (rel intensity) 1319 (100%) [M + H]+; retention time of 14.5 min. For 3,4-di-OHTAM-triSG 3: UV (CH3OH/H2O) 239, 294 nm; positive ion electrospray MS m/z (rel intensity) 1319 (100%) [M + H]+; retention time of 14.8 min. When 4-hydroxytamoxifen was used as a substrate, 3,4-di-OHTAM-SG 1-3 were detected. Microsomal Incubations and Incubations in MCF-7 Cells. Microsomal incubations and incubations in MCF-7 cells were conducted by following the procedures described in ref 15. Cytotoxicity Studies in MCF-7 and MDA-231 Cells. Cell viability was determined by trypan blue exclusion (23, 24). Briefly, MCF-7 and MDA-MB-231 cells were maintained in MEME supplemented with 1% penicillin, 10 mg/L streptomycin, 10 mg/L fungizome, 1% nonessential amino acids, and 10% bovine serum. The medium was changed 24 h before beginning cytotoxicity assays to maintain logarithmic growth. The cells were treated with 3,4-dihydroxytamoxifen, 4-hydroxytamoxifen, or tamoxifen for 18 h. The test samples were assayed in triplicate, and final concentrations ranged from 1.6 to 100 µM. Each assay included negative controls (cells treated with DMSO only) that were used to define 100% cell viability. After treatment, floating cells were collected by centrifugation at 3000 rpm for 5 min, and attached cells were first trypsinized and then harvested by centrifugation. Floating and attached cells were combined, washed with PBS, and then stained with 0.4% trypan blue. A drop of cell suspension was placed on a hemocytometer, and the number of cells was determined under a microscope. The dead cells were stained blue, while viable cells remained unstained. The LC50 values were obtained by regression and linear estimation analysis. The data represent the means ( SD of triplicate determinations. Reaction of 3,4-Dihydroxytamoxifen-o-quinone with Deoxynucleosides. A solution of 3,4-dihydroxytamoxifen-oquinone (0.25 mL, 10 mM) in acetonitrile prepared as described above was combined with deoxynucleosides (0.025 mM) in 4.5 mL of phosphate buffer (50 mM, pH 7.4), and the solution was incubated for 5 h at 37 °C. The solution was extracted using solid-phase extraction cartridges (Oasis; Waters Corp.), eluted

Synthesis and Reactivity of 3,4-Di-OHTAM-o-quinone with methanol, and concentrated to a final volume of 100 µL. Aliquots (25 µL) were analyzed directly by LC/MS (method C). Four adducts were obtained with both thymidine and deoxyguanosine (Figure 3); however, no adducts were detected with deoxyadenosine or deoxycytosine. Since tandem mass spectrometry of the thymidine and deoxyguanosine adducts showed a similar pattern for each isomer, MS/MS data are only reported for the first eluting isomer in each case. For 3,4-di-OHTAM-T 1: UV (CH3OH/H2O) 247, 284 nm; positive ion electrospray MS m/z (rel intensity) 644 (100%) [M + H]+; MS/MS with CID of m/z 644 (rel intensity) m/z 528 (100%), 363 (10%), 166 (5%), 117 (3%), 72 (40%); retention time of 22 min. For 3,4-di-OHTAM-T 2: UV (CH3OH/H2O) 246, 285 nm; positive ion electrospray MS m/z (rel intensity) 644 (100%) [M + H]+; retention time of 24 min. For 3,4-di-OHTAM-T 3: UV (CH3OH/H2O) 248, 289 nm; positive ion electrospray MS m/z (rel intensity) 644 (100%) [M + H]+; retention time of 24.3 min. For 3,4-di-OHTAM-T 4: UV (CH3OH/H2O) 247, 290 nm; positive ion electrospray MS m/z (rel intensity) 644 (100%) [M + H]+; retention time of 26 min. For 3,4-di-OHTAM-dG 1: positive ion electrospray MS m/z (rel intensity) 669 (100%) [M + H]+; MS/MS with CID of m/z 669 (rel intensity) m/z 553 (100%), 117 (3%), 72 (5%); retention time of 21 min. For 3,4-di-OHTAM-dG 2: positive ion electrospray MS m/z (rel intensity) 669 (100%) [M + H]+; retention time of 23 min. For 3,4-di-OHTAM-dG 3: positive ion electrospray MS m/z (rel intensity) 669 (100%) [M + H]+; retention time of 24 min. For 3,4-di-OHTAM-dG 4: positive ion electrospray MS m/z (rel intensity) 669 (100%) [M + H]+; retention time of 25 min.

Results and Discussion Synthesis of 3,4-Dihydroxytamoxifen. To the best of our knowledge, the synthesis of 3,4-dihydroxytamoxifen has not been reported in the literature. As a result, we report here an efficient synthetic method for preparation of 3,4-dihydroxytamoxifen as E and Z isomers (Scheme 2). This approach is based on the McMurry reaction between the key benzophenone derivative 3 and propiophenone, followed by selective removal of the methylenedioxy ring of 5. (2-Chloroethoxy)benzene 1 was prepared as reported by McCague et al. (18). FriedelCrafts acylation of 1 by piperonylic acid 2 in the presence of trifluoroacetic anhydride produced 4-(2-chloroethoxy)3,4-methylenedioxybenzophenone 3. The McMurry reaction between 3 and propiophenone in the presence of TiCl4/Zn gave E and Z isomers of 1-[4-(2-chloroethoxy)phenyl]-1-(3,4-methylenedioxyphenyl)-2-phenyl-1-butene 4. Compound 4 was converted into (E,Z)-1-[4-[2(N,N-dimethylamino)ethoxy]phenyl]-1-(3,4-methylenedioxyphenyl)-2-phenyl-1-butene 5, by refluxing 4 with dimethylamine in methanol. By selective removal of the methylenedioxy ring in 5 using BCl3 in CH2Cl2 at room temperature (17), 3,4-dihydroxytamoxifen was obtained as a mixture of E and Z (1:1 E:Z ratio) isomers. Attempts were made to decrease the reaction time for removal of the methylenedioxy ring in compound 5 by using other demethylation agents, including BBr3‚SMe2 (25) or aluminum chloride/ethanethiol (26); however, both methods gave undesired byproducts resulting from cleavage of the (N,N-dimethylamino)ethoxy group even at low temperatures. This could be due to the fact these agents were too reactive to selectively remove the methylenedioxy ring without effecting the (N,N-dimethylamino)ethoxy group in compound 5. For separation of E and Z isomers of 3,4-dihydroxytamoxifen, a series of chromatographic conditions were employed, including preparative TLC using silica gel or C-18 reverse phase, and flash chromatography (silica gel) with mixtures of different

Chem. Res. Toxicol., Vol. 13, No. 1, 2000 57

solvent systems (methanol, chloroform, benzene, and triethylamine); however, we were unsuccessful at separating the geometric isomers. The mixture is still very useful for biological studies because there are many reports in the literature which state that most metabolites of tamoxifen are formed as E and Z isomers (2729). These products were readily assigned a trans or cis configuration by proton NMR spectroscopy based on literature reports which have shown that the OCH2 protons on the side chain of the trans isomer have a chemical shift of less than 4.0 ppm because of the through-space shielding influence of the vicinal 2-phenyl substituent (30, 31). GSH Adducts of 3,4-Dihydroxytamoxifen-o-quinone. Previous work has shown that catechol estrogens can readily be oxidized by a variety of chemical and enzymatic agents, including manganese dioxide (32) and tyrosinase (33). These model oxidizing agents are more efficient than liver microsomes or purified P450s in generating the sufficient quantities of o-quinones for spectroscopic characterization. In the study presented here, we generated the 3,4-dihydroxytamoxifen-o-quinone in 90% yield using activated silver oxide in acetonitrile. Reaction of the o-quinone with GSH gave five major GSH conjugates. An HPLC method was developed to separate and characterize the GSH conjugates by UV, electrospray MS, and LC/MS/MS. Two of the conjugates were di-GSH conjugates (3,4-di-OHTAM-diSG 1 and 3,4-di-OHTAMdiSG 2, Scheme 3). These di-GSH conjugates were not stable and could not be purified for NMR characterization. One reason for the instability of the di-GSH conjugates could be the additional electron-donating sulfur substituent facilitates the oxidation of the aromatic ring, producing initially the o-quinone di-GSH conjugate which can further react with other nucleophiles in solution or undergo dimerization reactions (34). The mono-GSH conjugates were sufficiently stable for HPLC purification, and their proton NMR spectra were obtained. On the basis of the 1H NMR, COSY-NMR, and literature precedent (35), we predict that substitution has occurred at the C6 position, giving 3,4-di-OHTAM-SG 1 and 3,4-di-OHTAM-SG 2 as the major products (Scheme 3). Our previous work has shown that the C6 position is the predominant site of attack for sulfur nucleophiles on o-quinones since this position is less sterically hindered and the resulting anion enjoys greater resonance stabilization (22, 36). Their configurations were assigned on the basis of the same principle described previously, since the OCH2 protons on the side chain of the trans isomer are shifted upfield from the cis isomer because of the through-space shielding influence of the vicinal 2-phenyl substituent (30, 31). The structures of these GSH conjugates were further confirmed when cysteine was used in place of GSH. Like the diGSH conjugates, the di-CYS conjugates were not stable, and only two mono-CYS conjugates were isolated as stable compounds. Tyrosinase-catalyzed oxidation of 3,4-dihydroxytamoxifen in the presence of GSH gave three tri-GSH conjugates (3,4-di-OHTAM-triSG 1-3), two di-GSH conjugates (3,4-di-OHTAM-diSG 1 and 3,4-di-OHTAM-diSG 2), and three mono-GSH conjugates (3,4-di-OHTAM-SG 1-3, Scheme 3 and Figure 1). Like the di-GSH conjugates, these tri-GSH conjugates were also unstable, and only their MS and UV data were obtained. The possible reason for the formation of tri-GSH conjugates is that

58

Chem. Res. Toxicol., Vol. 13, No. 1, 2000

Zhang et al.

Scheme 3. Structures of the GSH Conjugates of 3,4-Dihydroxytamoxifen-o-quinone

Table 1. Conversion of 3,4-Dihydroxytamoxifen to o-Quinone-Derived GSH Conjugates by Rat Liver Microsomesa conjugate 3,4-di-OHTAM-diSG 1 3,4-di-OHTAM-diSG 2 3,4-di-OHTAM-SG 1 3,4-di-OHTAM-SG 2 3,4-di-OHTAM-SG 3

Figure 1. HPLC analysis (method A) of incubations of 3,4dihydroxytamoxifen with tyrosinase: (a) 3,4-di-OHTAM-triSG 1, (b) 3,4-di-OHTAM-triSG 2, (c) 3,4-di-OHTAM-triSG 3, (d) 3,4-di-OHTAM-diSG 1, (e) 3,4-di-OHTAM-diSG 2, (f) 3,4-diOHTAM-SG1, (g) 3,4-di-OHTAM-SG 2, and (h) 3,4-di-OHTAMSG 3.

di- and mono-GSH conjugates are more easily oxidized into the corresponding o-quinone thiol ethers than the parent catechol in the presence of tyrosinase (34), resulting in subsequent GSH Michael addition reactions. Michael type nucleophilic reaction of 3,4-dihydroxytamoxifen-o-quinone with GSH and cysteine has the following implications. The o-quinone could produce toxicity by depletion of cellular GSH, or once GSH is depleted, the o-quinone could alkylate cysteine residues on cellular proteins, resulting in toxic effects (11, 12). Reactivity of 3,4-Dihydroxytamoxifen-o-quinone. To study the reactivity of this o-quinone, we took advantage of the fact that GSH reacts very rapidly with o-quinones (37). As a result, the lifetime of the o-quinone under physiological conditions could be determined by removing aliquots at various times and “quenching” the remaining o-quinone with GSH. Analysis of the apparent decrease in the peak area ratio of the o-quinone GSH conjugates at different times by HPLC gave a good estimate of the lifetime of the o-quinone. From these studies, we determined that the half-life of the 3,4dihydroxytamoxifen-o-quinone was about 80 min. Interestingly, no isomerization of the o-quinone to a quinone

retention rate of formation [nmol time (min) (nmol of P450)-1 (10 min)-1] 29 31 34 37 41

0.0032 ( 0.0002 0.0014 ( 0.0004 0.061 ( 0.003 0.032 ( 0.003 0.028 ( 0.001 total: 0.13 ( 0.01

a Incubations were conducted for 30 min with 0.5 mM substrate and rat liver microsomes (1.0 nmol of P450/mL) in the presence of an NADPH-generating system and [3H]GSH at 37 °C. Radioactivity eluting from the HPLC column was measured in fractions collected at 18 s intervals. Results are the averages ( SD of three incubations.

methide was observed (HPLC showed no new GSHdependent peaks, data not shown), which was different from other o-quinones, such as 4-hydroxyequilin-oquinone (38) or 4-hydroxyestrone-o-quinone (21, 39). It is likely that the extended π conjugation stabilizes the 3,4-dihydroxytamoxifen-o-quinone, making isomerization unfavorable. In addition, although previous work has shown that the o-quinone of 4-allylcatechol (hydroxychavicol) readily isomerizes to the corresponding quinone methide (22), 4-(1-propenyl)catechol-o-quinone, which could be considered a structural isomer of 3,4-dihydroxytamoxifen-o-quinone, does not isomerize presumably because the R-hydrogens are not sufficiently acidic (40). Oxidation of 3,4-Dihydroxytamoxifen by Rat Liver Microsomes. We examined the oxidation of 3,4-dihydroxytamoxifen to quinoid metabolites in rat river microsomes by trapping these reactive species with [3H]GSH (Table 1). The trapping reaction should be very efficient due to the high concentration of GSH in the medium, and the fast rate of addition of thiols to quinoids relative to amino or hydroxy groups (34, 41). Nevertheless, a small amount of binding to microsomal protein is

Synthesis and Reactivity of 3,4-Di-OHTAM-o-quinone

Chem. Res. Toxicol., Vol. 13, No. 1, 2000 59

Table 2. Cytotoxicity of Tamoxifen and Its Metabolites in Breast Cancer Cell Linesa LC50 (µM) substrate

MDA-MB-231

MCF-7

3,4-di-OHTAM 4-OHTAM tamoxifen

30 ( 0.5 30 ( 0.6 34 ( 3

30 ( 0.9 17 ( 0.6 17 ( 1

a Cell viability was assessed by trypan blue exclusion in incubations containing 105 cells/mL for 18 h as described in Materials and Methods. Values are expressed as means ( SD of at least three determinations.

Figure 3. HPLC analysis (method A) of the reaction of 3,4dihydroxytamoxifen-o-quinone with thymidine and dG. (A) Reaction with thymidine: (a) 3,4-di-OHTAM-T 1, (b) 3,4-diOHTAM-T 2, (c) 3,4-di-OHTAM-T 3, and (d) 3,4-di-OHTAM-T 4. (B) Reaction with deoxyguanosine: (e) 3.4-di-OHTAM-dG 1, (f) 3,4-di-OHTAM-dG 2, (g) 3,4-di-OHTAM-dG 3, and (h) 3,4-di-OHTAM-dG 4.

Figure 2. HPLC analysis (method A) of GSH conjugates produced by 3,4-dihydroxytamoxifen in MCF-7 cells. 3,4-Dihydroxytamoxifen (20 µM) was incubated in 10 mL of medium (106 cells/mL) for 30 min at 37 °C as described in Materials and Methods. (A) DMSO. (B) 3,4-Dihydroxytamoxifen: (a) 3,4-diOHTAM-SG 1, (b) 3,4-di-OHTAM-SG 2, and (c) 3,4-diOHTAM-SG 3.

possible, so the conjugate formation shown in Table 1 is a lower limit for the generation of quinoids. The radiochromatograms gave peaks with retention times identical to those derived from addition of GSH to the o-quinone. The total amount of GSH conjugates produced was approximately 6-fold higher than that observed for P450catalyzed conversion of 4-hydroxytamoxifen to quinone methide GSH conjugates (15); however, the rate of formation of the GSH conjugates is considerably slower than that observed with catechol estrogen-o-quinone GSH conjugate formation (21). Incubation of 3,4-Dihydroxytamoxifen with MCF-7 Cells. Incubation of 3,4-dihydroxytamoxifen in MCF-7 cells showed that these cells are capable of oxidizing this catechol to the o-quinone which then reacts with cellular GSH to form conjugates. Isolation of the cell supernatant and HPLC separation showed the formation of three GSH conjugates (3,4-di-OHTAM-SG 1-3, Figure 2). The identity of these conjugates was confirmed by co-injections with the authentic standards. These preliminary results may explain the cytotoxicity of 3,4-dihydroxytamoxifen (see below), which occurs via depletion of cellular GSH through conjugate formation. In support of this, we have recently determined that the concentration of GSH in MCF-7 cells is 45 ( 9.4 µM (M. Chang and J. L. Bolton, unpublished results) compared to millimolar concentrations in the liver. Cytotoxicity Studies in MCF-7 and MDA-MB-231 Cells. Preliminary studies conducted with human breast cancer cell lines which are either ER+ (MCF-7) or ER(MDA-MB-231) showed that tamoxifen, 4-hydroxytamoxifen, and 3,4-dihydroxytamoxifen are all cytotoxic in these

cell lines. Interestingly, tamoxifen and its metabolites had equivalent cytotoxic potency in the ER- cell line (MDA-MB-231), whereas 3,4-dihydroxytamoxifen was about half as toxic in the MCF-7 cell line which contains the estrogen receptor. These data may suggest that the estrogen receptor is involved in the mechanism of cell death for tamoxifen and 4-hydroxytamoxifen but not for 3,4-dihydroxytamoxifen. Future studies with stable transfectants will be necessary to confirm this hypothesis. Reaction of 3,4-Dihydroxytamoxifen-o-quinone with Deoxynucleosides. 3,4-Dihydroxytamoxifen-oquinone was incubated with the four deoxynucleosides, and the solutions were analyzed using LC/MS and LC/ MS/MS. Results showed that only thymidine and deoxyguanosine can form adducts with the o-quinone under physiological conditions. Four adducts were detected in each case, although the yields were very low (Figure 3). Because of the low yields, it was very difficult to obtain sufficient quantities of these adducts for NMR characterization. As a result, their structures could not be determined unambiguously. However, on the basis of previous work with o-quinones (35), position 5 (Scheme 1) is the most reactive toward amino nucleophiles, although reaction could also occur at position 6 which is the preferred site for thiol nucleophiles. In addition, it has previously been shown that 4-hydroxyestrone-oquinone forms N-3 adducts with thymidine and exocyclic amino adducts with deoxyguanosine (42). Other adducts have been reported (42); however, all result in loss of the sugar moiety which was not observed in the study presented here. As a result, we predict that 3,4-dihydroxytamoxifen-o-quinone forms a covalent bond between the N-3 position of thymidine and the C5 position of the o-quinone (Figure 4). Similarly, the dG adducts could result from reaction between the C5 position on the o-quinone and the exocyclic amino group (Figure 5). The four adducts obtained in each case may result from E and Z isomers, or it is possible that other sites on the o-quinone ring (i.e., C6) have been modified. MS/MS analyses of the thymidine and deoxyguanosine adducts are consistent with the structures proposed

60

Chem. Res. Toxicol., Vol. 13, No. 1, 2000

Zhang et al.

Figure 6. HPLC analysis of incubations of 4-hydroxytamoxifen with tyrosinase in the presence of GSH: (a) 3,4-di-OHTAMSG 1, (b) 3,4-di-OHTAM-SG 2, and (c) 3,4-di-OHTAM-SG 3.

Figure 4. Positive ion electrospray MS/MS with CID of the protonated molecule of 3,4-di-OHTAM-T 1 at m/z 644.

Figure 5. Positive ion electrospray MS/MS with CID of the protonated molecule of 3,4-di-OHTAM-dG 1 at m/z 669.

above (Figures 4 and 5). The ion at m/z 644 (10%) was consistent with a protonated adduct of 3,4-dihydroxytamoxifen and thymidine (Figure 4). The product ion spectrum of the ion at m/z 644 gave prominent fragments at m/z 528 [MH - 116]+ (loss of the deoxyribosyl fragment), m/z 363 [MH - 281]+ [loss of the deoxyribosyl fragment and 4-(dimethylamino)ethoxyphenyl fragment], m/z 166 [4-(dimethylamino)ethoxyphenyl], and m/z 72 [dimethylaziridinium, CH2CH2N(CH3)2+]. Interestingly, all four thymidine adducts had the same MS/MS patterns as described above, and the earliest eluting isomer is shown in Figure 4. The MS/MS spectrum of the 3,4dihydroxytamoxifen-dG adduct is shown in Figure 5. The ion at m/z 669 was consistent with a protonated adduct of 3,4-dihydroxytamoxifen and dG. The product ion spectra of m/z 669 gave abundant fragment ions at m/z 553 [MH - deoxyribose]+ and m/z 72 (dimethylaziridinium). Similar to what was observed with the thymidine adducts, all four dG adducts produced indistinguishable tandem mass spectra. It is known that tamoxifen forms several DNA adducts both in animal models (43-45) and in humans (46, 47). The majority of these adducts appear to result from the R-hydroxylation pathway, ultimately producing the tamoxifen carbocation (9, 42). However, many of the minor adducts have not been identified, and it is possible that

the tamoxifen-o-quinone contributes to the DNA adduct profile. Quinone methide formation has been invoked to explain adducts not resulting from the tamoxifen carbocation (10, 48, 49); however, data presented in ref 15 as well as literature reports (50, 51) suggest that very little of the quinone methide is produced from 4-hydroxytamoxifen, and if formed, it might be too stable to cause significant DNA adduct formation in vivo. In contrast, we have shown that the tamoxifen-o-quinone is capable of forming adducts with thymidine and deoxyguanosine, and work is in progress to determine if similar adducts are produced with DNA. Finally, quinones are highly redox active compounds which could cause oxidation of the phosphate-sugar backbone, leading to single-strand cleavage and/or oxidation of the DNA bases (52). It has been reported that tamoxifen can cause oxidative DNA damage (53), and it is possible that formation of the tamoxifen-o-quinone is responsible for this effect. Oxidation of 4-Hydroxytamoxifen by Tyrosinase. It has been shown that tamoxifen and 4-hydroxytamoxifen can be metabolized to 3,4-dihydroxytamoxifen (1114). This reaction is probably catalyzed by P450 in most tissues but could be catalyzed by tyrosinase in melanocytes or in the cornea. Microsomal incubations with 4-hydroxytamoxifen showed that the major GSH conjugates result from two-electron oxidation to the quinone methide which reacts nonenzymatically with GSH. However, small amounts of 3,4-dihydroxytamoxifen-o-quinone GSH conjugates were detected as described in ref 15. In contrast, incubation of 4-hydroxytamoxifen with tyrosinase in the presence of GSH gave primarily 3,4-dihydroxytamoxifen-o-quinone GSH conjugates (Figure 6). This change in metabolism could be used to explain the ocular toxicity reported in some patients on high doses of tamoxifen which may be caused by tyrosinase-mediated o-quinone formation in the cornea (54). In conclusion, data have been presented describing the formation of 3,4-dihydroxytamoxifen-o-quinone GSH conjugates in microsomes and breast cancer cell lines, the reactivity of the o-quinone with GSH and deoxynucleosides, and the relative cytotoxicity of tamoxifen and metabolites in breast cancer cell lines. We have found that 3,4-dihydroxytamoxifen can be oxidized to the oquinone chemically or enzymatically which then reacts with GSH or deoxynucleosides forming the corresponding adducts under physiological conditions. These data suggest that this o-quinone could contribute to the cytotoxicity and/or genotoxicity of tamoxifen through depletion of GSH and alkylation of proteins and DNA. Interestingly, we have also found that 4-hydroxytamoxifen is primarily converted to 3,4-dihydroxytamoxifen-o-quinone in the presence of tyrosinase. The implications of these

Synthesis and Reactivity of 3,4-Di-OHTAM-o-quinone

quinoids with respect to the biological effects of tamoxifen are not known; however, given the direct link between tamoxifen therapy and the enhanced risk of endometrial cancer, the potential for formation of redox-active and/ or electrophilic metabolites from tamoxifen needs to be explored.

Acknowledgment. This research was supported by NIH Grant CA-79870 (J.L.B.) and CA-83124 (R.B.v.B.). We thank Mr. Richard Dvorak for performing highresolution mass spectrometry measurements.

Chem. Res. Toxicol., Vol. 13, No. 1, 2000 61

(19) (20)

(21)

(22)

References (1) Jordan, V. C. (1993) A current view of tamoxifen for the treatment and prevention of breast cancer. Br. J. Pharmacol. 110, 507517. (2) Fisher, B., Costantino, J. P., Wickerham, D. L., Redmond, C. K., Kavanah, M., Cronin, W. M., Vogel, V., Robidoux, A., Dimitrov, N., Atkins, J., Daly, M., Wieand, S., Tan-Chiu, E., Ford, L., and Wolmark, N. (1998) Tamoxifen for prevention of breast cancer: Report of National Surgical Adjuvant Breast and Bowel Project 1 Study. J. Natl. Cancer Inst. 90, 1371-1388. (3) Smith, L. L., and White, I. N. (1998) Antiestrogen therapy: uncertainties and risk assessment. Oncology 12, 14-22. (4) Killackey, M. A., Hakes, T. B., and Price, V. K. (1985) Endometrial adenocarcinoma in breast cancer patients receiving antiestrogen. Cancer Treat. Rep. 69, 237-238. (5) 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. (6) Hellmann-Blumberg, U., Cartner, M. G., Wurz, G. T., and DeGregorio, M. W. (1998) Intrinsic reactivity of tamoxifen and toremifene metabolites with DNA. Breast Cancer Res. Treat. 50, 135-141. (7) Hemminki, K., Rajaniemi, H., Lindahl, B., and Moberg, B. (1996) Tamoxifen-induced DNA adducts in endometrial samples from breast cancer patients. Cancer Res. 56, 4374-4377. (8) Hemminki, K., Rajaniemi, H., Koskinen, M., and Hansson, J. (1997) Tamoxifen-induced DNA adducts in leukocytes of breast cancer patients. Carcinogenesis 18, 9-13. (9) 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. (10) 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. (11) Dehal, S. S., and Kupfer, D. (1995) Evidence that catechol 3,4dihydroxytamoxifen is a proximate intermediate to the reactive species binding covalently to protein. Cancer Res. 56, 1283-1290. (12) Dehal, S. S., and Kupfer, D. (1999) Cytochrome P-450 3A and 2D6 catalyze ortho hydroxylation of 4-hydroxytamoxifen and 3-hydroxytamoxifen (Droloxifene) yielding tamoxifen catechol: involvement of catechols in covalent binding to hepatic proteins. Drug Metab. Dispos. 27, 681-688. (13) Lim, C. K., Lamb, J. H., Yuan, Z. X., and Smith, L. L. (1994) Identification of epoxide metabolites of tamoxifen by on-line liquid chromatography-electrospray ionization mass spectrometry. Biochem. Soc. Trans. 22, S165. (14) Lim, C. K., Yuan, Z. X., Lamb, J. H., White, I. N. H., De Matteis, F., and Smith, L. L. (1994) A comparative study of tamoxifen metabolism in female rat, mouse and human liver microsomes. Carcinogenesis 15, 589-593. (15) Fan, P. W., Zhang, F., and Bolton, J. L. (2000) 4-Hydroxylated metabolites of the antiestrogens tamoxifen and toremifene are metabolized to unusually stable quinone methides. Chem. Res. Toxicol. 13, 45-52. (16) NIH Guidelines for the Laboratory Use of Chemical Carcinogens, NIH Publication 81-2385 (1981) U.S. Government Printing Office, Washington, DC. (17) Gauthier, S., Mailhot, J., and Labrie, F. (1996) New highly stereoselective synthesis of (Z)-4-hydroxytamoxifen and (Z)-4hydroxytoremifene via McMurry reaction. J. Org. Chem. 61, 3890-3893. (18) McCague, R., Potter, G. A., and Jarman, M. (1994) An efficient, large-scale synthesis of Idoxifene {(E)-1-[4-[2-(N-pyrrolidino)-

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32) (33)

(34)

(35)

(36)

(37)

(38)

(39)

phenyl]]-1-(4-iodophenyl)-2-phenyl-1-butene}. Org. Prep. Proced. Int. 26, 343-346. Teitel, S., and O’Brien, J. P. (1976) Selective removal of an aromatic methylenedioxy group. J. Org. Chem. 41, 1657-1658. Curtin, D. Y., and Harder, R. J. (1960) Stereochemistry and reactions with hydroxy ion and with silver oxide of 2-bromo-4phenyl cyclohexanol and 1-methyl-2-bromo-4-phenyl cyclohexanols. J. Am. Chem. Soc. 82, 2357. Iverson, S. L., Shen, L., Anlar, N., and Bolton, J. L. (1996) Bioactivation of estrone and its catechol metabolites to quinoidglutathione conjugates in rat liver microsomes. Chem. Res. Toxicol. 9, 492-499. Bolton, J. L., Acay, N. L., and Vulkomanovic, V. (1994) Evidence that 4-allyl-o-quinones spontaneously rearrange to their more electrophilic quinone methides: Potential bioactivation mechanism for the hepatocarcinogen safrole. Chem. Res. Toxicol. 7, 443450. Andrews, D. L., Williams, G. S., Mahoney, J. C., and West, J. R. (1999) DNA fragmentation during exposure of rat cerebella to ethanol under hypoxia imposed in vitro. J. Neurobiol. 38, 8292. Sandoval, M., Zhang, X. J., Liu, X., Mannick, E. E., Clark, D. A., and Miller, M. J. (1997) Peroxynitrite-induced apoptosis in T84 and RAW 264.7 cells: Attenuation by L-ascorbic acid. Free Radical Biol. Med. 22, 489-495. Williard, P. G., and Fryhle, C. B. (1980) Boron trihalide-methyl sulfide complexs as convenient reagents for dealkylation of aryl ethers. Tetrahedron Lett. 21, 3731-3734. Node, M., Nishide, K., Fuji, K., and Fujita, E. (1980) Hard acid and soft nucleophile system. 2. Demethylation of methyl ethers of alcohol and phenol with an aluminium halide-thiol system. J. Org. Chem. 45, 4275-4277. Osborne, C. K., Wiebe, V. J., McGuire, W. L., Ciocca, D. R., and DeGregorio, M. W. (1992) Tamoxifen and the isomers of 4-hydroxytamoxifen in tamoxifen-resistant tumors from breast cancer patients. J. Clin. Oncol. 10, 304-310. Brewster, M. E., Huang, M. J., Pop, E., and Bodor, N. (1995) Isomeric interconversions in tamoxifen and related compoundsan AM1 study. Int. J. Quantum Chem. 55, 343-352. Katzenellenbogen, J. A., Carlson, K. E., and Katzenellenbogen, B. S. (1985) Facile geometric isomerization of phenolic nonsteroidal estrogens and antiestrogens: limitations to the interpretation of experiments characterizing the activity of individual isomers. J. Steroid Biochem. 22, 589-596. McCage, R., Leclercq, G., Legros, N., Goodman, J., Blackburn, M., Jarman, M., and Foster, A. B. (1989) Derivatives of tamoxifen. Dependence of antiestrogenicity on the 4-substituent. J. Med. Chem. 32, 2527-2533. Collins, D. J., Hobbs, J. J., and Emmens, C. W. (1971) Antiestrogenic and antifertility compounds. 4. 1,1,2-Triarylalkan-1-ols and 1,1,2-triarylalkenes containing basic ether group. J. Med. Chem. 14, 952. Abul-Hajj, Y. J., and Cisek, P. (1988) Catechol estrogen adducts. J. Steroid Biochem. 31, 107-110. Jacobsohn, M. K., Byler, D. M., and Jacobsohn, G. M. (1991) Isolation of estradiol-2,3-quinone and its intermediary role in melanin formation. Biochim. Biophys. Acta 1073, 1-10. Monks, T. J., and Lau, S. S. (1997) Biological reactivity of polyphenolic-glutathione conjugates. Chem. Res. Toxicol. 10, 1296-1313. Liberato, D. J., Byers, V. S., Dennick, R. G., and Castagnoli, N. (1981) Regiospecific attack of nitrogen and sulfur nucleophiles on quinones derived from poison oak/ivy catechols (urushiols) and analogues as models for urushiol-protein conjugate formation. J. Med. Chem. 24, 28-33. Iverson, S. L., Hu, L. Q., Vukomanovic, V., and Bolton, J. L. (1995) The influence of the para-alkyl substituent on the isomerization of o-quinones to p-quinone methides: Potential bioactivation mechanism for catechols. Chem. Res. Toxicol. 8, 537-544. Kalyanaraman, B., Premovic, P. I., and Sealy, R. C. (1987) Semiquinone anion radicals from addition of amino acids, peptides, and protein to quinones derived from oxidation of catechols and catecholamines: An ESR stabilization study. J. Biol. Chem. 262, 11080-11087. Zhang, F., Chen, Y., Pisha, E., Shen, L., Xiong, Y., van Breemen, R. B., and Bolton, J. L. (1999) The major metabolite of equilin, 4-hydroxyequilin, autoxidized to an o-quinone which isomerizes to the potent cytotoxin 4-hydroxyequilenin-o-quinone. Chem. Res. Toxicol. 12, 204-213. Bolton, J. L., and Shen, L. (1996) p-Quinone methides are the major decomposition products of catechol estrogen o-quinones. Carcinogenesis 17, 925-929.

62

Chem. Res. Toxicol., Vol. 13, No. 1, 2000

(40) Bolton, J. L., Pisha, E., Shen, L., Krol, E. S., Iverson, S. L., Huang, Z., van Breemen, R. B., and Pezzuto, J. M. (1997) The reactivity of o-quinones which do not isomerize to quinone methides correlates with alkylcatechol-induced toxicity in human melanoma cells. Chem.-Biol. Interact. 106, 133-148. (41) Bolton, J. L., Valerio, L. G. J., and Thompson, J. A. (1992) The enzymatic formation and chemical reactivity of quinone methides correlate with alkylphenol-induced toxicity in rat hepatocytes. Chem. Res. Toxicol. 5, 816-822. (42) Akanni, A., and Abul-Hajj, Y. J. (1997) Estrogen-nucleic acid addcts: Reaction of 3,4-estrone-o-quinone radical anion with deoxyribonucleosides. Chem. Res. Toxicol. 10, 760-766. (43) 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. (44) 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. (45) 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. (46) Terashima, I., Suzuki, N., and Shibutani, S. (1999) Mutagenic potential of R-(N2-deoxyguanosinyl)tamoxifen lesion, the majorDNA adducts detected in endometrial tissues of patients treated with tamoxifen. Cancer Res. 59, 2091-2095.

Zhang et al. (47) Shibutani, S., Suzuki, N., Terashima, I., Sugarman, S. M., Grollman, A. P., and Pearl, M. L. (1999) Tamoxifen-DNA adducts detected in the endometrium of women treated with tamoxifen. Chem. Res. Toxicol. 12, 646-653. (48) 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. (49) Marques, M. M., and Beland, F. A. (1997) Identification of tamoxifen-DNA adducts by 4-hydroxytamoxifen quinone methide. Carcinogenesis 18, 1949-1954. (50) Beland, F. A., McDaniel, L. P., and Marques, M. M. (1999) Comparison of the DNA adducts formed by tamoxifen and 4-hydroxytamoxifen in vivo. Carcinogenesis 20, 471-477. (51) Osborne, M. R., Davis, W., Hewer, A. J., Hardcastle, I. R., and Phillips, D. H. (1999) 4-Hydroxytamoxifen gives DNA adducts by chemical activation, but not in rat liver cells. Chem. Res. Toxicol. 12, 151-158. (52) Chen, Y., Shen, L., Zhang, F., Lau, S. S., van Breemen, R. B., Nikolic, D., and Bolton, J. L. (1998) The equine estrogen metabolite 4-hydroxyequilenin causes DNA single-strand breaks and oxidation of DNA bases in vitro. Chem. Res. Toxicol. 11, 11051111. (53) Okubo, T., Nagai, F., Ushiyama, K., Yokoyama, Y., Ozawa, S., Kano, K., Tomita, S., Kubo, H., and Kano, I. (1998) DNA cleavage and 8-hydroxyguanosine formation caused by tamoxifen derivatives in vitro. Cancer Lett. 122, 15. (54) Nayfield, S. G., and Gorin, M. B. (1996) Tamoxifen-associated eye disease. A review. J. Clin. Oncol. 14, 1018-1026.

TX990145N