Synthesis and DNA Reactivity of α-Hydroxylated Metabolites of

Martin R. Osborne, Warren Davis, Alan J. Hewer, Ian R. Hardcastle, and David H. Phillips. Chemical Research in Toxicology 1999 12 (2), 151-158. Abstra...
4 downloads 0 Views 141KB Size
Download PDF
Chem. Res. Toxicol. 1998, 11, 369-374

369

Synthesis and DNA Reactivity of r-Hydroxylated Metabolites of Nonsteroidal Antiestrogens Ian R. Hardcastle,* Martin N. Horton,‡ Martin R. Osborne,† Alan Hewer,† Michael Jarman, and David H. Phillips† CRC Centre for Cancer Therapeutics and Section of Molecular Carcinogenesis, Institute of Cancer Research, Cotswold Road, Sutton, Surrey SM2 5NG, U.K. Received October 31, 1997

Tamoxifen [(E)-1-(4-(2-(N,N-dimethylamino)ethoxy)phenyl)-1,2-diphenylbut-1-ene], a nonsteroidal antiestrogen, induces liver tumors in rats by a genotoxic mechanism. The mechanism of DNA adduct formation is believed to proceed via the formation of a reactive carbocation at the R-position from the R-hydroxylated metabolite. Molecular mechanics calculations [Kuramochi, H. (1996) J. Med. Chem. 39, 2877-2886] have predicted that 4-substitution will affect the stability of the carbocation and thus will alter its reactivity toward DNA. We have synthesized the putative R-hydroxylated metabolites of 4-hydroxytamoxifen [(E)-1-(4-(2-(N,N-dimethylamino)ethoxy)phenyl)-1-(4-hydroxyphenyl)-3-hydroxy-2-phenylbut-1-ene] and idoxifene [(Z)-1-(4iodophenyl)-3-hydroxy-2-phenyl-1-(4-(2-(N-pyrrolidino)ethoxy)phenyl)but-1-ene] and compared their reactivities with DNA with that of R-hydroxytamoxifen [(E)-1-(4-(2-(N,N-dimethylamino)ethoxy)phenyl)-3-hydroxy-1,2-diphenylbut-1-ene]. As predicted, the bis-hydroxylated compound reacted with DNA in aqueous solution at pH 5 to give 12-fold greater levels of adducts than R-hydroxytamoxifen, whereas R-hydroxyidoxifene gave one-half the number of adducts. The results demonstrate that idoxifene presents a significantly lower genotoxic hazard than tamoxifen for the treatment and prophylaxis of breast cancer.

Introduction The prophylactic use of tamoxifen for breast cancer prevention is currently being investigated in several countries in large-scale clinical trials. However doubts over the long-term safety of tamoxifen in this setting have provoked controversy (1-4). Tamoxifen has been found to be a genotoxic carcinogen in rats, forming high levels of DNA adducts in the liver, but the situation is less clear in humans with evidence of low levels of DNA adduct formation in some studies and not others (5-8). A full understanding of the mechanisms of adduct formation will allow a more complete assessment of the risks of long-term tamoxifen use and will assist in the discovery of possible safer alternatives. Novel selective estrogen receptor modulators (SERMs), such as idoxifene, could be as effective as tamoxifen for breast cancer treatment and prevention and be significantly less genotoxic. The proposed mechanism for adduct formation proceeds via R-hydroxylation followed by protonation of the hydroxyl group or the formation of an activated ester or sulfate which then leaves to give a stabilized R-carbocation (1) which reacts with nucleophilic sites on DNA (Scheme 1) (9). The experimental evidence for this mechanism is compelling: [ethyl-d5]tamoxifen produces lower levels of adducts in vivo, consistent with a kinetic isotope effect at the R-position (10, 11); the R-hydroxylated metabolite of tamoxifen (2) produces abundant adducts in vitro and in rat hepatocytes (12); and the products of the reaction of trans-R-acetoxytamoxifen with DNA have been identified as cis- and trans-R-(N2-deox† ‡

Section of Molecular Carcinogenesis. Deceased.

yguanosinyl)- (3) and -R-(N6-deoxyadenosinyl)tamoxifen (4) (13, 14), the cis-trans isomerism observed demonstrating that the reaction occurs via the predicted SN1 mechanism. The trans-R-N2-deoxyguanosinyl adduct (3) corresponds to the major adduct observed in vivo. Substitution at the 4-position of tamoxifen is predicted to affect the stability and hence the DNA reactivity of the reactive intermediate. Potter et al. (9) predicted that 4-hydroxytamoxifen (5) following oxidation by a cytochrome P450 enzyme should form an R-carbocationic species (6) capable of resonance stabilization to a neutral species dubbed ‘tam#’ (7), and this could be a major pathway to DNA damage (Scheme 2). This prediction is supported by molecular modeling calculations (15). The carbocation 6 is calculated to be 1.4 kcal/mol more stable than that from tamoxifen itself. Deprotonation of the carbocation to tam# (7) results in a significantly lower energy species which is predicted to be less reactive toward nucleophiles. The same calculations also show that the presence of a 4-iodo substituent lowers the stability of the carbocation by 2.6 kcal/mol predicting a less DNA-reactive species. An alternative pathway to the formation of the DNAreactive quinone methide (7) occurs directly via a twoelectron oxidation of the 4-hydroxytamoxifen (5). This mechanism is reported to be enhanced in mice treated with tamoxifen in conjunction with the sulfotransferase inhibitor pentachlorophenol (16, 17). The quinone methide 7 has been synthesized from 4-hydroxytamoxifen (5) and the major DNA adducts characterized as cis- and trans-R-(deoxyguanosin-N2-yl)-4-hydroxytamoxifen (18). We have prepared the R-hydroxylated derivatives of 4-hydroxytamoxifen and idoxifene (8 and 9) and inves-

S0893-228x(97)00198-7 CCC: $15.00 © 1998 American Chemical Society Published on Web 03/12/1998

370 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

Hardcastle et al.

Scheme 1. Mechanism of Formation of DNA Adducts from Tamoxifen

tigated their reactivity with DNA in vitro. The results are presented in this paper.

Experimental Section Chemical Methods. General Procedures. 1H NMR spectra (internal Me4Si) were obtained with a Bruker AC250 instrument. Melting points were obtained on a Reichert hotstage and are uncorrected. Chromatography refers to flash column chromatography on silica gel (Merck 15111) with the eluant indicated applied at a positive pressure of 0.5 atm. All reactions performed under an inert atmosphere were carried out in oven-dried glassware (110 °C, 24 h). Ether refers to diethyl ether. Petrol refers to the fraction with the boiling range 60-80 °C. Anhydrous tetrahydrofuran (THF) was obtained by distillation from potassium and benzophenone. Purification of E- and Z-geometrical isomers was monitored by 1H NMR spectroscopy and was carried out until none of the undesired isomer could be detected. Tamoxifen and salmon testis DNA were purchased from Sigma Chemical Co. (Poole, U.K.). Idoxifene, R-hydroxytamoxifen (2), and (E)-1-(4-(2-chloroethoxy)phenyl)-1-(4-iodophenyl)-2-phenylbut-1-ene (18) were prepared as previously described (19, 20). All operations with tamoxifen derivatives were carried out in darkness or subdued light where possible. Syntheses. (E,Z)-1-(4-(2-Chloroethoxy)phenyl)-1-(4-hydroxyphenyl)-2-phenylethene (12). n-Butyllithium solution

(2.5 M in hexanes; 7.2 mL, 18 mmol) was added dropwise to a solution of 1-bromo-4-(tetrahydropyranyloxy)benzene (11) (4.63 g, 18.0 mmol) in THF (50 mL) at -78 °C, and stirring continued for 30 min. A solution of 1-(4-(chloroethoxy)phenyl)-2-phenylethanone (10) (4.14 g, 15.0 mmol) in THF was added, the reaction mixture allowed to warm to ambient temperature, and the reaction quenched with ammonium chloride (30 mL). The mixture was diluted with water (100 mL) and extracted with ethyl acetate (2 × 100 mL). The organic extracts were dried (MgSO4) and concentrated. The resulting oil was heated to reflux in a mixture of ethanol (50 mL), concentrated hydrochloric acid (10 mL) for 2 h, allowed to cool, diluted with water (100 mL), and extracted with ethyl acetate (100 mL + 2 × 50 mL). The organic extracts were washed with water (50 mL), dried (MgSO4), and concentrated. Chromatography (hexane-ethyl acetate, 4:1) gave a yellow oil (4.5 g, 86%): 1H NMR δ (250 MHz, CDCl3) 3.88-3.81 (2, m, CH2Cl), 4.28-4.23 (2, m, OCH2), 7.296.77 (14, m, ArH and CHdC); FABHRMS found m/z 350.1063, C22H19ClO2 requires m/z 350.1074. (E,Z)-1-(4-Acetoxyphenyl)-1-(4-(2-chloroethoxy)phenyl)2-phenylethene (13). Acetic anhydride (1.44 mL, 15.3 mmol) was added to a mixture of 12 (3.59 g, 10.2 mmol), 2,6(dimethylamino)pyridine (0.061 g, 0.5 mmol), and dichloromethane (100 mL) and stirred for 1 h. The mixture was washed with sodium hydroxide solution (2 M; 50 mL) and water (50 mL), then dried (MgSO4), and concentrated. Chromatography (hexane-ether, 4:1) gave 13 as a mixture of isomers (3.62 g, 90%): 1H NMR δ (250 MHz, CDCl3) 2.31 (3, s, OCOCH3), 3.80-3.87 (2, m, CH2Cl), 4.22-4.28 (2, m, OCH2), 6.85-7.34 (14, m, ArH and CHdC); HRMS found m/z 392.1174, C24H21ClO3 requires m/z 392.1179. (E,Z)-1-(4-Acetoxyphenyl)-1-(4-(2-chloroethoxy)phenyl)2-bromo-2-phenylethene (14). Pyridinium hydrobromide perbromide (90%, w/w; 3.23 g, 9.1 mmol) was added to a solution of 13 (3.57 g, 9.1 mmol) in dichloromethane (120 mL) and stirred for 30 min. Triethylamine (2.52 mL, 18.2 mmol) was added, and stirring continued for 2 h. The mixture was washed with hydrochloric acid (1 M; 2 × 50 mL) and then water (50 mL), dried (MgSO4), and concentrated. Chromatography (hexane-

Metabolites of Nonsteroidal Antiestrogens

Chem. Res. Toxicol., Vol. 11, No. 4, 1998 371

Scheme 2. Hypothetical Mechanism of Formation of DNA Adducts from 4-Hydroxytamoxifen (5)

dichloromethane, 1:1) gave 14 as a mixture of isomers (3.65 g, 85%): 1H NMR δ (250 MHz, CDCl3) 2.23 (3, s, cis OCOCH3), 2.31 (3, s, trans OCOCH3), 3.75 (2, t, trans CH2Cl), 3.84 (2, t, J ) 5.9 Hz, cis CH2Cl), 4.12 (2, t, J ) 5.8 Hz, trans OCH2), 4.26 (2, t, J ) 5.9 Hz, cis OCH2), 6.61 (2, d, J ) 8.8 Hz, trans ArH), 6.79-7.39 (24, m, cis and trans ArH); HRMS found m/z 493.0186, C24H2079BrClO3Na requires m/z 493.0182. (E,Z)-1-Bromo-2-(4-(2-(N,N-dimethylamino)ethoxy)phenyl)-2-(4-hydroxyphenyl)-1-phenylethene (15). A mixture of 14 (2.47 g, 4.6 mmol) and dimethylamine solution (33%, ethanol; 17 mL) was heated in a sealed vessel at 100 °C for 16 h. The mixture was diluted with water (100 mL) and extracted with ethyl acetate (2 × 100 mL). The organic extracts were washed with water (50 mL) and brine (50 mL), dried (Na2CO3), and concentrated. Chromatography (ethyl acetate-methanoltriethylamine, 20:2:1) gave 15 as a mixture of isomers (1.89 g, 94%): 1H NMR δ (250 MHz, DMSO-d6) 2.15 (6, s, trans N(CH3)2), 2.22 (6, s, cis N(CH3)2), 2.48-2.55 (4, m, OH and trans CH2N), 2.63 (2, t, J ) 5.8 Hz, cis CH2N), 3.91 (2, t, J ) 5.8 Hz, trans OCH2), 4.06 (2, t, J ) 5.8 Hz, cis OCH2), 6.47 (2, d, J ) 8.6 Hz, cis ArH), 6.65 (2, d, J ) 8.8 Hz, trans ArH), 6.70 (2, d, J ) 8.6 Hz, cis ArH), 6.77 (2, d, J ) 8.6 Hz, trans ArH), 6.81 (2, d, J ) 8.7 Hz, trans ArH), 6.94 (2, d, J ) 8.7 Hz, cis ArH), 7.11 (2, d, J ) 8.6 Hz, trans ArH), 7.16-7.27 (12, m, ArH); HRMS found m/z 438.1065, C24H2579BrNO2 requires m/z 438.1069. (E,Z)-1-Bromo-2-(4-((tert-butyldimethylsilyl)oxy)phenyl)2-(4-(2-(N,N-dimethylamino)ethoxy)phenyl)-1-phenylethene (16). A mixture of 15 (0.58 g, 1.3 mmol), tert-butyldimethylsilyl chloride (0.30 g, 1.5 mmol), imidazole (0.18 g, 2.6 mmol), and dimethylformamide (5 mL) was stirred for 18 h. The mixture was diluted with sodium hydroxide solution (2 M; 30 mL) and extracted with ethyl acetate (3 × 20 mL). The combined organic extracts were washed with water (2 × 25 mL) and brine (25 mL), dried (Na2SO4), and concentrated. Chromatography (hexane-ethyl acetate-triethylamine, 25:5:1) gave 16 as an oil (0.51 g, 71%): 1H NMR δ (250 MHz, CDCl3) 0.12 (6, s, cis Si(CH3)2), 0.24 (6, s, trans Si(CH3)2), 0.93 (9, s, cis SiC(CH3)3), 1.00 (9, s, trans SiC(CH3)3), 2.30 (6, s, trans N(CH3)2), 2.36 (6, s, cis N(CH3)2), 2.66 (2, t, J ) 5.5 Hz, trans CH2N), 2.75 (2, t, J ) 5.8 Hz, cis CH2N), 3.95 (2, t, J ) 5.6 Hz, trans OCH2), 4.10 (2, t, J ) 5.9 Hz, cis OCH2), 6.35 (2, d, J ) 8.6 Hz, cis ArH ortho to OSi), 6.61 (2, d, J ) 8.8 Hz, trans ArH ortho to OSi), 6.77-6.95 (8, m, cis and trans ArH), 7.14-7.33 (14, m, cis and trans ArH); HRMS found m/z 552.1937, C30H3979BrNSiO2 requires m/z 552.1933.

(Z)-1-(4-((tert-Butyldimethylsilyl)oxy)phenyl)-1-(4-(2(N,N-dimethylamino)ethoxy)phenyl)-3-hydroxy-2-phenylbut-1-ene (17). n-Butyllithium (2.5 M in hexanes; 0.8 mL, 2.0 mmol) was added dropwise to a solution of 16 (0.69 g, 1.3 mmol) in THF (10 mL) at -78 °C and stirred for 15 min. Acetaldehyde (1 mL) was added, and stirring continued for 2 h at -78 °C; then the mixture was allowed to warm to ambient temperature and the reaction quenched with water (10 mL). The mixture was extracted with ethyl acetate (10 mL), and the organic extracts were washed with water (10 mL) and brine (10 mL), then dried (Na2CO3), and concentrated. Chromatography (hexane-ethyl acetate-triethylamine, 8:8:1) gave 17 as the pure Z (trans) isomer (0.28 g, 42%); further elution (ethyl acetatetriethylamine, 16:1) gave the E (cis) isomer (0.19 g, 28%). Trans isomer: 1H NMR δ (250 MHz, CDCl3) 0.20 (6, s, Si(CH3)2), 0.98 (9, 2, SiC(CH3)3), 1.18 (3, d, J ) 6.5 Hz, CH3C(OH)H), 2.22 (6, s, N(CH3)2), 2.63 (2, t, J ) 5.7 Hz, CH2N), 3.91 (2, t, J ) 5.8 Hz, OCH2), 4.87 (1, q, J ) 6.5 Hz,), 6.52 (2, d, J ) 8.9 Hz, ArH ortho to OCH2), 6.78-6.85 (4, m, ArH), 7.12 (2, d, J ) 8.6 Hz, ArH meta to OSi), 7.14-7.25 (5, m, ArH); HRMS found m/z 518.3084, C32H44NSiO3 requires m/z 518.3090. Cis isomer: inter alia 1H NMR δ (250 MHz, CDCl3) 0.007 (6, s, Si(CH3)2), 0.89 (9, 2, SiC(CH3)3), 2.36 (6, s, N(CH3)2), 2.76 (2, t, J ) 5.7 Hz, CH2N), 4.09 (2, t, J ) 5.8 Hz, OCH2), 6.45 (2, d, J ) 8.6 Hz, ArH), 6.72 (2, d, J ) 8.8 Hz, ArH), 6.91 (2, d, J ) 8.7 Hz, ArH), 7.13-7.21 (7, m, ArH). (E)-1-(4-(2-(N,N-Dimethylamino)ethoxy)phenyl)-1-(4-hydroxyphenyl)-3-hydroxy-2-phenylbut-1-ene (8). Tetrabutylammonium fluoride solution (1 M in THF; 0.22 mL, 0.22 mmol) was added dropwise to a solution of 17 (0.19 g, 0.20 mmol) in THF (2.5 mL). The resulting mixture was stirred for 15 min and then concentrated. The concentrate was partitioned between ethyl acetate (10 mL) and water (10 mL), the aqueous layer was extracted with ethyl acetate (2 × 5 mL), and the combined organic extracts were washed with water (10 mL) and brine (10 mL), dried (Na2CO3), and concentrated. Chromatography (ethyl acetate-methanol-triethylamine, 20:1:1) followed by crystallization from methanol gave 8 as white crystals (0.077 g, 95%): mp 151-154 °C; 1H NMR δ (250 MHz, CDCl3) 1.12 (3, d, J ) 6.5 Hz, CH3C(OH)H), 2.28 (6, s, N(CH3)2), 2.67 (2, t, J ) 5.5 Hz, CH2N), 3.30-3.31 (1, m, OH), 3.93 (2, t, J ) 5.4 Hz, OCH2), 4.86-4.91 (1, m, CH3C(OH)H), 6.55 (2, d, J ) 8.9 Hz, ArH ortho to OCH2), 6.77 (2, d, J ) 8.5 Hz, ArH ortho to OH), 6.80 (2, d, J ) 8.9 Hz, ArH meta to OCH2), 7.04 (2, d, J ) 8.5 Hz, ArH meta to OH), 7.11-7.20 (5, m, PhH). Anal. C, H, N.

372 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

Hardcastle et al.

Scheme 3. Synthesis of r,4-Dihydroxytamoxifen (8)a

a Reagents and conditions: (a) (i) THF, -78 °C, (ii) HCl, EtOH; (b) Ac O, DMAP, CH Cl ; (c) PyBr‚Br, CH Cl , Et N; (d) HN(CH ) , 2 2 2 2 2 3 3 2 EtOH; (e) TBDMSCl, imidazole, DMF; (f) (i) n-BuLi, THF, (ii) CH3CHO; (g) TBAF, THF.

(Z)-3-Acetoxy-1-(4-(2-chloroethoxy)phenyl)-1-(4-iodophenyl)-2-phenylbut-1-ene (19). A suspension of (E)-1-(4-(2chloroethoxy)phenyl)-1-(4-iodophenyl)-2-phenylbut-1-ene (18) (1.01 g, 2.05 mmol), silver(I) acetate (1.36 g, 8.20 mmol), and iodine (1.04 g, 4.10 mmol) in carbon tetrachloride (20 mL) was stirred for 2 h. The mixture was filtered through Celite and the filtrate washed with sodium thiosulfate solution (10%; 50 mL) and water, then dried (MgSO4), and concentrated. Chromatography (hexane-ether, 5:1) gave 19 as white crystals (0.337 g, 30%): mp 130-133 °C; 1H NMR δ (250 MHz, CDCl3) 1.28 (3, d, J ) 6.7 Hz, CH3C(OAc)H), 1.91 (3, s, OCOCH3), 3.70 (2, t, J ) 6.2 Hz, CH2Cl), 4.08 (2, t, J ) 6.2 Hz, OCH2), 5.72 (1, q, J ) 6.5 Hz, CH3C(OAc)H); HRMS found m/z 546.0480, C26H24ClIO3 requires m/z 546.0459. (Z)-1-(4-(2-Chloroethoxy)phenyl)-1-(4-iodophenyl)-3-hydroxy-2-phenylbut-1-ene (20). A solution of 19 (0.55 g, 1 mmol) in THF (20 mL) and sodium hydroxide solution (1 M; 10 mL) was refluxed for 48 h. The mixture was partitioned between ether (30 mL) and water (30 mL). The organic phase was washed with water (20 mL) and brine (20 mL), dried (MgSO4), and concentrated. Chromatography (hexane-ether, 2:1) gave 20 as an oil (0.49 g, 98%): 1H NMR δ (250 MHz, CDCl3) 1.27 (3, d, J ) 6.7 Hz, CH3C(OH)H), 1.73-1.79 (4, m, N(CH2CH2)2), 1.91 (3, s, OCOCH3), 2.53-2.59 (4, m, N(CH2CH2)2), 2.79 (2, t, J ) 6.1 Hz, CH2N), 3.94 (2, t, J ) 6.0 Hz, OCH2), 5.61 (1, q, J ) 6.6 Hz, CH3C(OAc)H), 6.53 (2, d, J ) 8.8 Hz, ArH ortho to OCH2), 6.74 (2, d, J ) 8.7 Hz, ArH meta to OCH2), 7.03 (2, d, J ) 8.2 Hz, ArH meta to I), 7.11-7.21 (5, m, PhH), 7.70 (2, d, J ) 8.2 Hz, ArH ortho to I); HRMS found m/z 527.0256, C24H22ClIO2Na requires m/z 527.0251. (Z)-1-(4-Iodophenyl)-3-hydroxy-2-phenyl-1-(4-(2-(N-pyrrolidino)ethoxy)phenyl)but-1-ene (9). A mixture of 20 (0.47 g, 0.93 mmol), pyrrolidine (0.80 mL, 10 mmol), and ethanol (10 mL) was refluxed for 48 h. The mixture was concentrated and the residue partitioned between sodium hydroxide solution (2 M; 30 mL) and ethyl acetate (30 mL). The aqueous phase was extracted with ethyl acetate (2 × 10 mL), and the combined organic extracts were washed with water (25 mL) and brine (25 mL), dried (Na2SO4), and concentrated. Chromatography (hexane-ether-triethylamine, 10:10:1) gave 9 as white crystals (0.46 g, 92%): mp (ex methanol) 148-150 °C; 1H NMR δ (250 MHz, DMSO-d6) 0.99 (3, d, J ) 6.4 Hz, CH3C(OH)H), 1.591.67 (4, m, N(CH2CH2)2), 2.41-2.47 (4, m, N(CH2CH2)2), 2.67 (2, t, J ) 6.1 Hz, CH2N), 3.88 (2, t, J ) 6.0 Hz, OCH2), 4.504.59 (1, m, CH3C(OH)H), 4.72 (1, d, J ) 3.3 Hz, OH), 6.57 (2, d,

J ) 8.7 Hz, ArH ortho to OCH2), 6.75 (2, d, J ) 8.7 Hz, ArH meta to OCH2), 7.05 (2, d, J ) 8.2 Hz, ArH meta to I), 7.107.23 (5, m, PhH), 7.74 (2, d, J ) 8.2 Hz, ArH ortho to I); FABMS m/z 562 (65, M+ + Na), 540 (100, M+). Anal. C, H, N, I. In Vitro DNA Reactions. Reaction of 2, 8, and 9 with DNA. Salmon testis DNA (0.5 mg) was treated with 2, 8, or 9 (0.13 mmol) in water-ethanol (0.5 mL; 9:1) containing an acidic buffer (0.1 M NaOAc-HOAc, pH 5, or mixtures of 0.1 M citric acid and 0.2 M Na2HPO4 in various proportions to give a range of pH from 3 to 7) for 18 or 66 h at 37 °C. The tamoxifen derivative was removed by ether extraction (7 × 0.7 mL). The DNA was precipitated with ethanol, washed, and dried. A portion (4 µg) of the DNA was analyzed for adducts by the 32Ppostlabeling technique, using the standard labeling procedure (13). Briefly, the DNA was digested to mononucleotides with micrococcal nuclease (0.14 unit) and spleen phosphodiesterase (2.4 munits) at 37 °C overnight and then incubated with 75 µCi of [γ-32P]ATP and T4 polynucleotide kinase (6 units) for 30 min to produce radioactive nucleoside bisphosphates. These were resolved on poly(ethylenimine)-cellulose TLC as described previously using the following solvents: D1, 2.3 M sodium phosphate, pH 5.8; D2, 2.275 M lithium formate, 5.525 M urea, pH 3.5; D3, 0.25 M LiCl, 0.325 M Tris-HCl, 5.525 M urea, pH 8. Autoradiography was carried out on a Canberra-Packard Instantimager.

Results and Discussion Synthesis: 1. r-4-Dihydroxytamoxifen (8). The synthesis of 8 is outlined in Scheme 3. Ketone 10 is reacted with 1-lithio-4-(tetrahydropyranyloxy)benzene (11), formed in situ from the reaction of 1-bromo-4(tetrahydropyranyloxy)benzene and n-butyllithium, in THF at -78 °C to give the carbinol which is dehydrated by treatment with hydrochloric acid in refluxing ethanol to give the deprotected product (12) as a mixture of cis and trans isomers in 68% yield. Bromination, with pyridinium bromide perbromide, of the unprotected phenol 12 resulted in a mixture of ring-brominated products along with the desired product in low yield. However, protection of the phenol with the electronwithdrawing acetyl group to give 13 allowed the bromination to proceed smoothly affording 14 in 85% yield.

Metabolites of Nonsteroidal Antiestrogens

Chem. Res. Toxicol., Vol. 11, No. 4, 1998 373

Scheme 4. Synthesis of r-Hydroxyidoxifene (9)a

a

Reagents and conditions: (a) AgOAc, I2, CCl4; (b) NaOH, H2O, THF; (c) pyrrolidine, EtOH, reflux.

Reaction of 14 with dimethylamine in ethanol gave the dimethylamino compound (15), with concomitant deprotection, in 55% yield. Protection under standard conditions gave the silyl-protected compound (16) in good yield. Treatment of 16 with n-butyllithium at -78 °C resulted in halogen-metal exchange to give the vinyllithium species which was quenched with acetaldehyde to give the R-hydroxylated product (17) in 42% yield. Subsequent deprotection with fluoride gave 8 in 98% yield. 2. r-Hydroxyidoxifene (9). The synthesis of 9 is outlined in Scheme 4. Allylic acetoxylation, with silver acetate and iodine, of trans-(chloroethoxy)idoxifene (18) gave the trans-R-acetoxy product (19) in 30% yield along with the cis isomer (37% yield). Reaction of the acetoxy compound (19) with pyrrolidine was anticipated to lead directly to the desired product (9); however, an intractable mixture resulted. Deprotection of 19 with aqueous sodium hydroxide in THF proceeded cleanly to give 20. Subsequent reaction of 20 with pyrrolidine gave 9 in excellent yield. Comparison of in Vitro Reactivity of Metabolites with DNA. R-OH-tamoxifen (2), R,4-dihydroxytamoxifen (8), and R-OH-idoxifene (9) were incubated with DNA in order to compare their reactivities. Thus, a solution of 2, 8, or 9 in water-ethanol or water-DMSO containing sodium acetate buffer was incubated with salmon testis DNA for 18 or 66 h at 37 °C. Unreacted 2, 8, or 9 was removed by ether extraction and the DNA precipitated, washed, and dried. A portion of the DNA was analyzed for adducts by the 32P-postlabeling technique. The TLC patterns of adducts are shown in Figure 1. The levels of adducts formed (Figure 2) varied depending on the time of incubation and pH used; however, in all cases 9 produced lower levels of adducts and was on average 0.47 times as reactive as 2. In contrast, 8 reacted strongly with DNA giving an average of 12 times as many adducts as 2. The levels of adducts observed are consistent with the molecular modeling calculations (15). The 4-iodo-substituted carbocation formed from 9 was calculated to be less stable by 2.6 kcal/mol than that from 2, and therefore 9 is less reactive toward nucleophiles. In contrast, the 4-hydroxy-substituted carbocation 6 formed from 8 is calculated to be 1.4 kcal/mol more stable and is likely to exist in equilibrium with the deprotonated species 7 which is significantly more stable and so forms more abundant DNA adducts. Effect of pH on Extent of Adduct Formation by 8. Salmon testis DNA was treated with 8 in waterethanol containing an acidic buffer at pH 7-3. Unreacted 8 was removed by ether extraction and the DNA

Figure 1. Autoradiographs of poly(ethylenimine)-cellulose thin-layer plates after chromatography of 32P-labeled nucleoside bisphosphates. The origins are at the bottom left corner of each plate and the directions of development: D1 downward, D2 upward, and D3 to the right. The relative intensities are given in Figure 2; they are not apparent from this figure, which is rendered on a logarithmic scale in order that both major and minor spots are visible. The adducts were derived from A, R-hydroxytamoxifen (2); B, R,4-dihydroxytamoxifen (8); C, R-hydroxyidoxifene (9). DNA (0.5 mg) was treated with 0.13 mmol of compound in 0.5 mL of aqueous ethanol, pH 5, for 18 h at 37 °C, followed by isolation, degradation, and 32P labeling as described in the Experimental Section.

Figure 2. Extent of reaction of tamoxifen and idoxifene derivatives with DNA at pH 5: R-hydroxytamoxifen (2), R,4dihydroxytamoxifen (8), R-hydroxyidoxifene (9).

precipitated with ethanol, washed, and dried. A portion of the DNA was analyzed for adducts by the 32Ppostlabeling technique. The results (Figure 3) show that the reaction of 8 with DNA is acid-catalyzed. This is consistent with the rate-limiting step for the adduct formation being the protonation of the R-hydroxyl group prior to its leaving to form the carbocation. Kinetics of Reaction of 8 with DNA. The rate of reaction of 8 at pH 5 was determined. Figure 4 shows the extent of reaction at various times of incubation at 37 °C. The data were analyzed with a published program (21) and shown to fit a first-order curve, with a correlation coefficient of 0.96 and a half-life of 27 h. Thus, the bis-hydroxylated compound 8 is more reactive

374 Chem. Res. Toxicol., Vol. 11, No. 4, 1998

Hardcastle et al.

References

Figure 3. Extent of reaction of R,4-dihydroxytamoxifen (8) with DNA at various pH, in water-ethanol containing citratephosphate buffer, for 18 h at 37 °C.

Figure 4. Rate of reaction of R,4-dihydroxytamoxifen (8) with DNA at pH 5, in water-ethanol containing acetate buffer, at 37 °C.

than 2 which has a half-life, under the same conditions, of 43 h (13).

Conclusions These results provide further evidence for the mechanism of DNA adduct formation by metabolites of tamoxifen. The 4-hydroxylated tamoxifen metabolite 5 may be activated in vivo to 8, a potent DNA-reactive species, by further hydroxylation at the R-position. Adducts formed from 8 may account for some of the minor adducts detected in tamoxifen-treated rats. In contrast, the presence of a 4-iodo substituent in idoxifene renders the R-hydroxylated metabolite 9 less reactive to DNA, and correspondingly fewer adducts are seen. Almost undetectable levels of DNA adducts were observed in rats treated with idoxifene, whereas tamoxifen produces abundant adducts (22). The lower reactivity of 9 toward DNA may explain the lack of adduct formation. The R-hydroxylated metabolite 9 is detected in very low levels in the plasma of idoxifene-treated patients (23). The low DNA reactivity, coupled with the low circulating levels of 9, suggests that idoxifene represents a safer alternative to tamoxifen for the treatment and prophylaxis of breast cancer.

Acknowledgment. This work was supported by grants from the Cancer Research Campaign and a Cancer Research Campaign studentship (M.N.H.). Mass spectrometry was performed by the ULICS at the London School of Pharmacy.

(1) Williams, G. M., Iatropoulous, M. J., and Hard, G. C. (1992) Longterm prophylactic use of tamoxifen: is it safe? Eur. J. Cancer Prev. 1, 386-387. (2) Fugh-Bergman, A., and Epstein, S. (1992) Tamoxifen: disease prevention or disease substitution. Lancet 340, 1143-1145. (3) Phillips, D. H., and Venitt, S. (1993) Safety of prophylactic tamoxifen. Lancet 1993, 1485-1486. (4) King, C. M. (1995) Tamoxifen and the induction of cancer. Carcinogenesis 16, 1449-1454. (5) Carmichael, P. L., Ugwumadu, A. H. N., Neven, P., Hewer, A. J., Poon, G. K., and Phillips, D. H. (1996) Lack of genotoxicity of tamoxifen in human endometrium. Cancer Res. 56, 1475-1479. (6) Phillips, D. H., Hewer, A., Grover, P. L., Poon, G. K., and Carmichael, P. L. (1996) Tamoxifen does not form detectable DNA adducts in white blood cells of breast cancer patients. Carcinogenesis 17, 1149-1152. (7) 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. (8) Hemminki, K., Rajaniemi, H., Koskinen, M., and Hansson, J. (1997) Tamoxifen-induced DNA adducts in leucocytes of breast cancer patients. Carcinogenesis 18, 9-13. (9) Potter, G. A., McCague, R., and Jarman, M. (1994) A mechanistic hypothesis for DNA adduct formation by tamoxifen following hepatic oxidative metabolism. Carcinogenesis 15, 439-442. (10) Jarman, M., Poon, G. K., Rowlands, M. G., Grimshaw, R. M., Horton, M. N., Potter, G. A., and McCague, R. (1995) The deuterium isotope effect for the R-hydroxylation of tamoxifen by rat liver microsomes accounts for the reduced genotoxicity of [D5ethyl]tamoxifen. Carcinogenesis 16, 683-688. (11) Phillips, D., 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. (12) 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. (13) 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. (14) Osborne, M. R., Hardcastle, I. R., and Phillips, D. H. (1997) Minor products of reaction of DNA with R-acetoxytamoxifen. Carcinogenesis 18, 539-543. (15) 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. (16) 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. (17) 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. (18) Marques, M. M., and Beland, F. A. (1997) Identification of tamoxifen-DNA adducts formed by 4-hydroxytamoxifen quinone methide. Carcinogenesis 18, 1949-1954. (19) Foster, A. B., Jarman, M., Leung, O.-T., McCague, R., Leclercq, G., and Devleeschouwer, N. (1985) Hydroxy derivatives of tamoxifen. J. Med. Chem. 28, 1491-1497. (20) McCague, R., LeClercq, G., Legros, N., Goodman, J., Blackburn, G. M., Jarman, M., and Foster, A. B. (1989) Derivatives of tamoxifen: dependence of antiestrogenicity on the 4-substituent. J. Med. Chem. 32, 2527-2533. (21) Osborne, M. R. (1982) A pocket calculator program to fit firstorder kinetic curves. Comput. Biol. Med. 12, 159-162. (22) Pace, P., Jarman, M., Phillips, D., Hewer, A., Bliss, J., and Coombes, R. C. (1997) Idoxifene is equipotent to tamoxifen in inhibiting mammary carcinogenesis but forms lower levels of hepatic DNA adducts. Br. J. Cancer 76, 700-704. (23) Haynes, B. P., Griggs, L. J., Grimshaw, R. M., Quigley, M., Nutley, B. P., Poon, G. K., and Coombes, R. C. (1996) Pharmacokinetics of idoxifene and identification of metabolites in patients during a loading dose study. Proc. Am. Assoc. Cancer Res. 37, 177.

TX970198+