Alkylation of 2 '-Deoxynucleosides and DNA by the Premarin

Premarin (Wyeth-Ayerst) is the estrogen replacement treatment of choice and continues to be one of the most widely dispensed prescriptions in the Unit...
0 downloads 0 Views 189KB Size
Download PDF
94

Chem. Res. Toxicol. 1998, 11, 94-101

Articles Alkylation of 2′-Deoxynucleosides and DNA by the Premarin Metabolite 4-Hydroxyequilenin Semiquinone Radical Li Shen, Shengxiang Qiu, Yumei Chen, Fagen Zhang, Richard B. van Breemen, Dejan Nikolic, 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 October 7, 1997

Premarin (Wyeth-Ayerst) is the estrogen replacement treatment of choice and continues to be one of the most widely dispensed prescriptions in the United States. In addition to endogenous estrogens, Premarin contains unsaturated estrogens including equilenin. We synthesized the catechol metabolite of equilenin, 4-hydroxyequilenin (4-OHEN), and found that the semiquinone radical of 4-OHEN reacted with 2′-deoxynucleosides generating very unusual adducts. 2′-Deoxyguanosine (dG), 2′-deoxyadenosine (dA), or 2′-deoxycytosine (dC) all gave four isomers, but no product was observed for thymidine under similar physiological conditions. The structures of these adducts were determined by electrospray mass spectrometry and NMR experiments including 1H, 13C, DQF-COSY, ROESY, HOHAHA, HMQC, and HMBC. The spectral data show that dG forms a cyclic adduct with the 4-OHEN producing 2-N1,3-N2deoxyguanosyl-1,3-dihydroxy-5,7,9(10)-estratriene-4,17-dione. Similarly, reaction with dA produced 1-N6,3-C2-deoxyadenosyl-2,3-dihydroxy-5,7,9(10)-estratriene-4,17-dione, and incubations with dC resulted in 1-N3,3-N4-deoxycytosyl-2,3-dihydroxy-5,7,9(10)-estratriene-4,17-dione. We found that care needed to be taken during the isolation of the dA adducts in particular, as any exposure to acidic environments caused hydrolysis of the sugar moiety leaving alkylated adenine. In mixtures of the deoxynucleosides treated with 4-OHEN, reaction occurred primarily with dG followed by dC and dA. With DNA significant apurinic sites were produced as 4-OHEN-adenine adducts were detected in the ethanol wash prior to hydrolysis. When the DNA was hydrolyzed to deoxynucleosides and analyzed by electrospray mass spectrometry, only one isomer of 4-OHEN-dG and one isomer of 4-OHEN-dC were observed. Our data suggest that several different types of DNA lesions could be expected from 4-OHEN including apurinic sites and bulky stable adducts, in addition to the published oxidized damage to DNA caused by 4-OHEN. The production of these semiquinone radical-derived DNA adducts could play a role in the carcinogenic effects of Premarin estrogens.

Introduction A firm link between female reproductive variables and increased risk of developing cancer in hormone-sensitive tissues, especially the breast, has been established from epidemiological studies (1-3). Excessive exposure to estrogens either through early menarche and late menopause and/or through estrogen replacement therapy (ERT)1 increases the risk of women developing cancer (47). Recent data indicate that as many as 30% of postmenopausal women in the United States are currently receiving ERT (8), and as the population ages an increas* Corresponding author. Fax: (312) 996-7107. E-mail: Judy.Bolton@ UIC.edu. 1 Abbreviations: dN, 2′-deoxynucleoside; dG, 2′-deoxyguanosine; dA, 2′-deoxyadenosine; dC, 2′-deoxycytosine; 8-OHdG, 8-hydroxydeoxyguanosine; 2-OHE, 2-hydroxyestrone, 2,3-dihydroxy-1,3,5(10)-oestratrien17-one; 4-OHE, 4-hydroxyestrone, 3,4-dihydroxy-1,3,5(10)-oestratrien17-one; 4-OHEN, 4-hydroxyequilenin, 3,4-dihydroxy-1,3,5(10),6,8estrapenten-17-one; estrone, 3-hydroxy-1,2,5(10)-oestratrien-17-one;

ing number of women will be faced with the decision of whether to accept ERT. There are many benefits of ERT, including the relief of menopausal symptoms, a substantial reduction in the risk of cardiovascular disease and osteoporosis (7), and a possible correlation between the reduced risk of stroke (9) and Alzheimer’s disease (10, 11). However, the consequences of estrogen deficiency versus the potential risks of ERT therapy, especially the ability of estrogens to initiate and/or promote the carcinogenic process, remain a highly controversial issue. The mechanism(s) of estrogen-induced carcinogenesis could involve modification of critical cellular macromolecules by electrophilic/redox-active quinoids. Aromatic hydroxylation of estrone and 17β-estradiol forming catequilenin, 3-hydroxy-1,3,5(10),6,8-estrapenten-17-one; equilin, 3-hydroxy-1,3,5(10),7-estratetren-17-one; ERT, estrogen replacement therapy; P450, cytochrome P450; QM, quinone methide, 4-alkyl-2,5-cyclohexadien-1-one; o-quinone, 3,5-cyclohexadiene-1,2-dione.

S0893-228x(97)00181-1 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/23/1998

DNA Alkylation by 4-Hydroxyequilenin Quinoids

Chem. Res. Toxicol., Vol. 11, No. 2, 1998 95

Scheme 1. Reaction of 4-OHEN with 2′-Deoxynucleosides

echol metabolites represents the major phase I metabolic pathway for endogenous estrogens (12). Peroxidase/ P450-catalyzed oxidation of these catechols gives oquinones which have previously been implicated as the ultimate carcinogens (13). Redox cycling between the o-quinones and their semiquinone radicals generates superoxide and ultimately reactive hydroxyl radicals which cause oxidation of the purine/pyrimidine residues of DNA (12) as well as single-strand breaks (14). In support of this mechanism, increased levels of 8-hydroxydeoxyguanosine (8-OHdG) have been associated with a predictive significance for breast cancer risk assessment (15). In addition to oxidative damage, model studies with estrogen quinoids (o-quinones, semiquinone radicals, quinone methides) have shown that these quinoids can alkylate purine/pyrimidine bases of nucleosides (16-19) as well as DNA in vitro (20, 21) which may contribute to initiation of the carcinogenic process in vivo. The equine estrogen equilenin [3-hydroxy-1,3,5(10),6,8estrapenten-17-one] or its 17β-hydroxylated analogue make up 15% of the most widely prescribed estrogen replacement formulation, Premarin (Wyeth-Ayerst), and yet there is very little information on the metabolism of these estrogens in animal models (22-24) let alone in women. It is known that treating hamsters for 9 months with either estrone, equilin plus equilenin, or sulfatasetreated Premarin resulted in 100% tumor incidences and abundant tumor foci (22). We previously synthesized the major metabolite of equilenin, 4-hydroxyequilenin (4OHEN; Scheme 1), and examined how aromatization of the B ring affects the formation and reactivity of the equilenin quinoids (25). Unlike the endogenous catechol estrogens, 4-OHEN rapidly autoxidized to 4-OHEN-oquinone which readily entered into a redox couple with the semiquinone radical catalyzed by NAD(P)H, P450

reductase, or quinone reductase (21). Significant oxygen consumption was also detected consistent with in vitro models that have shown that 4-OHEN-o-quinone increases the amount of oxidative damage to DNA by 50% compared to control levels (26). In addition to DNA oxidation, an alternative mechanism for equilenin carcinogenesis may involve alkylation of DNA. We report here the synthesis and characterization of the 2′-deoxynucleoside adducts of 4-OHEN-semiquinone radical. These data facilitated an investigation of the principal sites of DNA alkylation by 4-OHEN-semiquinone radical in vitro.

Materials and Methods Caution: All quinones used in this study and the catechol estrogen o-quinones in particular were handled in accordance with NIH guidelines for the Laboratory Use of Chemical Carcinogens (27). Preparation and Characterization of Adducts. All chemicals were purchased from Aldrich (Milwaukee, WI), Fisher Scientific (Itasca, IL), or Sigma (St. Louis, MO) unless stated otherwise. H2O18 (97%) was obtained from Isotec Inc. (Miamisburg, OH). 4-OHEN was synthesized by treating equilin with Fremy’s salt as described previously (23, 28) with minor modifications (25). 4-OHEN (20 mg, 0.07 mmol, dissolved in 1 mL of methanol) was incubated with dNs (0.14 mmol in 200 µL of DMSO) in pH 7.4 potassium phosphate buffer (25 mM, 20 mL) at 37 °C for 7 h. Under these reaction conditions, four major adducts were detected with dG and two major and two minor adducts with dA and dC, and no reaction was observed with thymidine. The major adducts were isolated from the aqueous phase on C-18 extraction cartridges (Waters Oasis TM HLB) and eluted with methanol. The eluates were concentrated and subjected to semipreparative HPLC with an Ultrasphere ODS column (10 × 250 mm; Beckman) with a flow rate of 3.5 mL/min on a Shimadzu LC-10A gradient HPLC equipped with

96 Chem. Res. Toxicol., Vol. 11, No. 2, 1998 an SIL-10A autoinjector, SPD-M10AV UV/vis photodiode array detector, and SPD-10AV UV detector set at 280 nm. The HPLC mobile phases were as follows: 4-OHEN-dG adducts, 10% methanol in water for 2 min, increased to 35% in 3 min, 43% in another 25 min, and increased to 90% CH3OH in the last 5 min of the run; 4-OHEN-dA adducts, 10% CH3OH in water for 2 min, increased to 37% CH3OH over 5 min, isocratic for 17 min and increased to 90% CH3OH over the remaining 7 min of the run; 4-OHEN-dC adducts, 5% methanol in water for 2 min, increased to 27% in 3 min, 33% in another 45 min, and increased to 90% CH3OH in the last 5 min of the run. For LC/MS analysis aliquots (25 µL) were analyzed directly by HPLC with a 4.6- × 150-mm Ultrasphere C-18 column (Beckman) on a HewlettPackard (Palo Alto, CA) 1090L gradient HPLC equipped with a 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. For analytical HPLC analyses on the Simadzu system the ammonium acetate buffer was replaced with 0.25% perchloric acid/0.25% acetic acid (pH 3.5). The structures of the adducts are shown in Scheme 1, and the spectral data are as follows. 4-OHEN-dG1: 1H NMR (DMSO-d6) δ 0.04 (s, 3H, CH3), 1.55 (dd, J ) 3, 8 Hz, 1H, H12β), 1.75 (m, 1H, H12R), 1.78 (m, 1H, H15β), 2.05 (m, 1H, 2′R-dG), 2.28 (m, 2H, H16R, 2′β-dG), 2.35 (m, 1H, H15R), 2.51 (m, 1H, H16β), 2.76 (dd, J ) 8, 17.5 Hz, 1H, H11β), 2.83 (dd, J ) 3, 9 Hz, 1H, H11R), 3.03 (dd, J ) 6, 12 Hz, 1H, H14), 3.47 (m, 2H, 2 × 5′-dG), 3.74 (dd, J ) 4.5, 7.5 Hz, 1H, 4′-dG), 4.24 (bs, 1H, 3′-dG), 4.78 (d, J ) 3.0 Hz, 1H, H2), 4.89 (t, J ) 5.5 Hz, 1H, 5′OH-dG, D2O exchangeable), 5.22 (d, J ) 3.5 Hz, 1H, 3′OH-dG, D2O exchangeable), 5.92 (m, 2H, 1′H-dG, H1), 6.28 (d, J ) 3 Hz, 1H, OH1, D2O exchangeable), 7.24 (d, J ) 6.5 Hz, 1H, H7), 7.25 (s, 1H, OH3, D2O exchangeable), 7.52 (d, J ) 7.5 Hz, 1H, H6), 7.94 (s, 1H, H8-dG), 9.29 (s, 1H, N2H-dG, D2O exchangeable); 13C NMR (CD3OD) δ 14.2 (CH3), 22.2 (C15), 23.4 (C11), 29.6 (C12), 37.3 (C16), 41.5 (2′C-dG), 47.9 (C13), 48.3 (C14), 61.5 (C1), 63.3 (5′C-dG), 68.4 (C2), 72.6 (3′C-dG), 85.7 (1′C-dG), 85.9 (C3), 89.3 (4′C-dG), 118.0 (C5-dG), 126.0 (C7), 127.0 (C6), 132.0 (C5), 137.0 (C9), 138.0 (C8-dG), 140.0 (C10), 146.0 (C8), 152.0 (C4-dG), 155.0 (C2-dG), 157.0 (C6-dG), 193.0 (C4), 221.0 (C17); UV (CH3OH) 220, 265 nm; electrospray-MS, positive ion m/z 564 (100) (MH+), negative ion m/z 562 (100) (MH-), retention time 16 min. 4-OHEN-dG2: 1H NMR (DMSO) δ 0.62 (s, 3H, CH3), 1.76 (m, 2H, H12, H15), 1.75 (m, 1H, H12), 2.19 (m, 1H, 2′R-dG), 2.27 (m, 2H, H15, H16), 2.50 (m, 2H, 2′β-dG, H16), 2.80 (m, 1H, H11), 3.06 (m, 2H, H11, H14), 3.45 (m, 2H, 5′-dG), 3.74 (bd, J ) 2.8 Hz, 1H, 4′-dG), 4.24 (bs, 1H, 3′-dG), 4.78 (d, J ) 2.3 Hz, 1H, H2), 4.87 (t, J ) 5.4 Hz, 1H, 5′OH-dG), 5.23 (d, J ) 4.0 Hz, 1H, 3′OH-dG), 5.91 (d, J ) 3 Hz, 1H, H1), 5.98 (t, 1H, 1′H-dG), 6.21 (d, J ) 3 Hz, 1H, OH1), 7.24 (d, J ) 7.3 Hz, 1H, H7), 7.25 (s, 1H, OH3), 7.57 (d, J ) 7.3 Hz, 1H, H6), 7.97 (s, 1H, H8-dG), 9.40 (bs, 1H, N2H-dG); UV (CH3OH) 220, 265 nm; electrospray-MS, positive ion m/z 564 (100) (MH+), negative ion m/z 562 (MH-), retention time 20 min. 4-OHEN-dG3: 1H NMR (DMSO) δ 0.36 (s, 3H, CH3), 1.62 (m, 1H, H12), 1.87 (m, 2H, H12, H15), 2.15 (m, 2H, 2′R-dG, 2′β-dG), 2.34 (m, 1H, H15), 2.50 (m, 1H, 2 × H16), 2.85 (m, 1H, H11), 3.06 (m, 1H, H14), 3.42 (m, 1H, H11), 3.51 (m, 2H, 2 × 5′-dG), 3.76 (bd, 1H, 4′-dG), 4.20 (m, 1H, H2), 4.30 (bs, 1H, 3′-dG), 4.87 (t, J ) 5.4 Hz, 1H, 5′OH-dG), 5.24 (d, J ) 4.0 Hz, 1H, 3′OH-dG), 6.02 (t, 1H, 1′H-dG), 6.37 (d, J ) 4 Hz, 1H, OH1), 6.42 (d, J ) 4 Hz, 1H, H1), 6.69 (s, 1H, OH3), 7.33 (d, J ) 7.9 Hz, 1H, H7), 7.92 (d, J ) 7.9 Hz, 1H, H6), 7.96 (s, 1H, H8-dG), 8.78 (bs, 1H, N2H-dG); UV (CH3OH) 245, 280 nm; electrospray-MS, positive ion m/z 564 (100) (MH+), negative ion m/z 562 (MH-), retention time 26 min. 4-OHEN-dG4: 1H NMR (DMSO) δ 0.54 (s, 3H, CH3), 1.85 (m, 3H, H12 × 2, H15), 2.15 (m, 1H, 2′R-dG), 2.34 (m, 2H, H15, H16), 2.50 (m, 2H, 2′β-dG, H16), 3.06 (m, 2H, H11, H14), 3.24 (m, 1H,

Shen et al. H14), 3.46 (m, 2H, 2 × 5′-dG), 3.74 (bd, 1H, 4′-dG), 4.18 (m, 1H, H2), 4.30 (bs, 1H, 3′-dG), 4.85 (t, J ) 5.4 Hz, 1H, 5′OH-dG), 5.25 (d, J ) 4.0 Hz, 1H, 3′OH-dG), 6.02 (t, 1H, 1′H-dG), 6.37 (d, J ) 4 Hz, 1H, OH1), 6.42 (d, J ) 4 Hz, 1H, H1), 6.68 (s, 1H, OH3), 7.32 (d, J ) 7.9 Hz, 1H, H7), 7.92 (d, J ) 7.9 Hz, 1H, H6), 7.97 (s, 1H, H8-dG), 8.81 (bs, 1H, N2H-dG); UV (CH3OH) 245, 280 nm; electrospray-MS, positive ion m/z 564 (100) (MH+), negative ion m/z 562 (MH-), retention time 30 min. 4-OHEN-dA3: 1H NMR (DMSO-d6) δ 0.51 (s, 3H, CH3), 1.87 (m, 3H, 2 × H12, H15), 2.25 (m, 1H, 2′-dA), 2.35 (m, 1H, H16), 2.42 (m, 1H, H15), 2.50 (m, 1H, 2′-dA), 2.60 (m, 1H, H16), 3.13 (m, 2H, H11, H14), 3.20 (m, 1H, H11), 3.50 (m, 2H, 2 × 5′-dA), 3.80 (m, 1H, 4′-dA), 4.18 (bs, 1H, H2), 4.35 (bs, 1H, 3′-dA), 4.87 (t, J ) 5.4 Hz, 1H, 5′OH-dA), 5.26 (d, J ) 4.0 Hz, 1H, 3′OHdA), 5.82 (d, J ) 4.4 Hz, 1H, H1), 6.05 (bs, 1H, OH2), 6.12 (s, 1H, OH3), 6.18 (t, J ) 6.9 Hz, 1H, 1′H-dA), 7.32 (d, J ) 8.0 Hz, 1H, H7), 7.91 (d, J ) 8.0 Hz, 1H, H6), 8.12 (s, 1H, H8-dA), 8.33 (s, 1H, N6H-dA); 13C NMR (DMSO-d6) δ 14.1 (CH3), 20.9 (C15), 23.9 (C11), 28.4 (C12), 36.1 (C16), 39.6 (2′C-dA), 46.1 (C13), 46.2 (C14), 55.9 (C1), 61.6 (5′C-dA), 67.3 (C2), 70.6 (3′C-dA), 83.3 (1′CdA), 85.2 (C3), 87.9 (4′C-dA), 123.2 (C5-dA), 125.1 (C6), 126.1 (C7), 130.2 (C5), 135.4 (C10), 136.0 (C9), 138.0 (C8-dA), 140.4 (C6-dA), 142.0 (C4-dA), 145.8 (C8), 146.2 (C2-dA), 192.6 (C4), 218.3 (C17); UV (CH3OH) 220, 260 nm; electrospray-MS, positive ion m/z 548 (100) (MH+), retention time 18 min. 4-OHEN-dA4: 1H NMR (DMSO-d6) δ 0.52 (s, 3H, CH3), 1.76 (m, 1H, H15), 1.97 (m, 2H, 2 × H12), 2.20 (m, 1H, 2′-dA), 2.35 (m, 2H, H15, H16), 2.50 (m, 1H, 2′-dA), 2.60 (m, 1H, H16), 2.85 (m, 2H, 2 × H11), 3.10 (m, 1H, H14), 3.50 (m, 2H, 2 × 5′-dA), 3.81 (m, 1H, 4′-dA), 4.15 (bs, 1H, H2), 4.32 (bs, 1H, 3′-dA), 4.90 (bs, 1H, 5′OH-dA), 5.27 (bs, 1H, 3′OH-dA), 5.80 (d, J ) 4.3 Hz, 1H, H1), 6.06 (bs, 1H, OH2), 6.13 (s, 1H, OH3), 6.17 (t, J ) 6.9, 1H, 1′H-dA), 7.30 (d, J ) 7.6 Hz, 1H, H7), 7.90 (d, J ) 7.6 Hz, 1H, H6), 8.12 (s, 1H, H8-dA), 8.36 (s, 1H, N6H-dA); 13C NMR (DMSO-d6) δ 15.1 (CH3), 20.9 (C15), 23.7 (C11), 28.3 (C12), 36.1 (C16), 39.6 (2′C-dA), 46.0 (C13), 46.7 (C14), 55.9 (C1), 61.6 (5′CdA), 67.3 (C2), 70.6 (3′C-dA), 83.2 (1′C-dA), 85.2 (C3), 87.9 (4′CdA), 124.0 (C5-dA), 124.9 (C6), 125.8 (C7), 130.2 (C5), 135.4 (C10), 137.0 (C9), 137.8 (C8-dA), 140.5 (C6-dA), 142.0 (C4-dA), 146.3 (C8, C2-dA), 192.6 (C4), 218.3 (C17); UV (CH3OH) 220, 260 nm; electrospray-MS, positive ion m/z 548 (100) (MH+), retention time 21 min. The other two isomers gave similar UV and electrospray mass spectrometry data. The retention times were 4-OHEN-A1, 13 min, and 4-OHEN-A2, 15 min. 4-OHEN-dC3: 1H NMR (DMSO-d6) δ 0.43 (s, 3H, CH3), 1.62 (m, 1H, H12), 1.90 (m, 2H, H12, H15), 2.06 (m, 2H, 2 × H2′-dC), 2.38 (m, 2H, H15, H16), 2.57 (m, 1H, H16), 2.70 (m, 1H, H11), 3.07 (m, 1H, H14), 3.49 (m, 2 H, 2 × H5′-dC), 3.73 (m, 2H, H11, H4′dC), 4.04 (d, J ) 4.6 Hz, 1H, H2), 4.20 (m, 1H, H3′-dC), 4.97 (bs, 1H, 5′-OH-dC, D2O exchangeable), 5.25 (bs, 1H, 3′-OH-dC, D2O exchangeable), 5.60 (d, J ) 8.1 Hz, 1H, H5-dC), 5.98 (d, J ) 4.4 Hz, 1H, H1), 6.13 (t, J ) 6.9 Hz, 1H, H1′-dC), 7.28 (d, J ) 8.0 Hz, 1H, H7), 7.38 (d, J ) 8.2 Hz, 1H, H6-dC), 7.88 (d, J ) 7.7 Hz, 1H, H6); 13C NMR (DMSO-d6) δ 15.5 (CH3), 20.6 (C15), 23.4 (C11), 28.5 (C12), 36.2 (C16), 39.4 (2′C-dC), 46.2 (C14), 46.8 (C13), 50.0 (C1), 61.3 (5′C-dC), 67.2 (C2), 70.4 (3′C-dC), 85.1 (1′C-dC), 85.2 (C3), 87.3 (4′C-dC), 103.0 (C5-dC), 125.1 (C6), 125.3 (C7), 130.0 (C5), 133.0 (C6-dC), 137.0 (C10), 138.0 (C9), 145.0 (C4-dC), 146.0 (C8), 149.0 (C2-dC), 193.0 (C4), 219.0 (C17); UV (CH3OH) 222, 265 nm; electrospray-MS, positive ion m/z 546 (6) [M + Na]+, 524 (100) [M + H]+, 408 (10) [M + H - ribose]+, retention time 49 min. 4-OHEN-dC4: 1H NMR (DMSO-d6, exchangeable protons were not visible) δ 0.52 (s, 3H, CH3), 1.75 (m, 1H, H12), 1.90 (m, 2H, H12, H15), 2.05 (m, 2H, 2 × H2′-dC), 2.35 (m, 2H, H15, H16), 2.55 (m, 1H, H16), 3.10 (m, 1H, H11), 3.20 (m, 2H, H14, H11), 3.50 (m, 2H, 2 × H5′-dC), 3.75 (m, 1H, H4′-dC), 4.02 (d, J ) 4.5 Hz, 1H, H2), 4.19 (m, 1H, H3′-dC), 5.60 (d, J ) 8.1 Hz, 1H, H5-dC), 6.01 (d, J ) 4.5 Hz, 1H, H1), 6.16 (t, J ) 6.9 Hz, 1H, H1′-dC), 7.27 (d, J ) 7.9 Hz, 1H, H7), 7.37 (d, J ) 8.2 Hz, 1H, H6-dC), 7.86 (d, J ) 7.9 Hz, 1H, H6); 13C NMR (DMSO-d6) δ 14.7 (CH3), 20.9 (C15), 23.6 (C11), 28.3 (C12), 36.1 (C16), 39.4 (2′C-dC), 46.1

DNA Alkylation by 4-Hydroxyequilenin Quinoids (C14), 46.6 (C13), 49.9 (C1), 61.4 (5′C-dC), 67.5 (C2), 70.4 (3′CdC), 85.2 (1′C-dC), 85.3 (C3), 87.4 (4′C-dC), 102.6 (C5-dC), 124.9 (C6), 125.6 (C7), 130.3 (C5), 133.7 (C6-dC), 136.3 (C10), 137.6 (C9), 145.0 (C4-dC), 146.0 (C8), 149.0 (C2-dC), 192.8 (C4), 218.8 (C17); UV (CH3OH) 222, 265 nm; electrospray-MS, positive ion m/z 524 (100) [M + H]+, retention time 52 min. The other two isomers gave similar UV and electrospray mass spectrometry data. The retention times were 4-OHEN-dC1, 23 min, and 4OHEN-dC2, 24 min. Acetylation of 4-OHEN-dG4, 4-OHEN-dA3, and 4-OHENdC3. Each adduct (2 mg) was dissolved in 2 mL of pyridine and 2 mL of acidic anhydride and left at room temperature for 24 h. The organic solvents were removed in vacuo, and the derivatives were characterized by electrospray mass spectrometry or 1H NMR. All three adducts showed that four acetate groups were incorporated. 4-OHEN-dG4-tetraacetate: 1H NMR (DMSO-d6) δ 1.93 (s, 3H, acetate CH3), 1.97 (s, 3H, acetate CH3), 2.06 (s, 3H, acetate CH3), 2.14 (s, 3H, acetate CH3) in addition to dG peaks; electrospray-MS, positive ion m/z 732 (100) [M + H]+, 532 (10) [M - sugar with two acetate groups]+. 4-OHEN-dA3-tetraacetate: electrospray-MS, positive ion m/z 716 (100) [M + H]+, 738 (20) [M + Na]+. 4-OHEN-dC3-tetraacetate: electrospray-MS, positive ion m/z 692 (40) [M + H]+, 650 (50) [M - COCH3 + H]+. Reaction of 4-OHEN with DNA. Calf thymus DNA (Sigma) (10 mg) was dissolved in 5 mL of 50 mM phosphate buffer containing 1 mM EDTA and repeatedly precipitated with 70% ethanol/water to remove all impurities. A solution of DNA (2 mg/mL) was incubated with 4-OHEN (2.0 mM) and ampicillin (100 µg/mL) for 4 h at 37 °C, pH 7.4 phosphate buffer. At the conclusion of the incubation the DNA was precipitated with 70% ethanol/water and washed with three 5-mL portions of 70% ethanol. The ethanol washings were pooled, the solvent was removed in vacuo, and the residue was redissolved in 1.0 mL of methanol for LC/MS analysis of alkylated deoxynucleosides. The precipitated DNA was hydrolyzed to deoxynucleosides using the following literature procedures with minor modifications (29, 30). Briefly, the DNA was redissolved in 1 mL of DNase buffer (50 mM sodium acetate, 2 mM CaCl2, and 10 mM MgCl2, pH 6.5) and incubated with DNase I (24 units/mg DNA) for 20 h at 37 °C. Then 250 µL of 0.5 M Tris buffer (pH 9.0) and 0.013 unit/mg DNA of snake venom phosphodiesterase were added, and the mixture was incubated at 37 °C for 4 h. This treatment resulted in complete hydrolysis to the deoxynucleoside level. Adducts were isolated as described above and analyzed by LC/ MS with electrospray detection. 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. 1H NMR spectra were obtained with a Varian XL300 spectrometer at 300 MHz, and CI/EI and FAB mass spectra were obtained with a Finnigan MAT 90 magnetic sector mass spectrometer. 2D NMR experiments were performed with either a GE Omega 500 or a Bruker Avance DPX500 instrument at 500 MHz. Positive ion 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 abovementioned 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 250 °C and a flow rate of 10 L/min was used for evaporation of solvent from the electrospray. The range m/z 200-900 was scanned every 2 s during LC/MS. LC/MS-MS experiments were performed on a Micromass quattro II electrospray triple quadrupole mass spectrometer. Samples were infused using a syringe pump in 50% methanol/water containing 1% acetic acid.

Chem. Res. Toxicol., Vol. 11, No. 2, 1998 97

Results Characterization of 4-OHEN-deoxyadenosine Adducts. Previously we showed that deoxyguanosine reacts with the 4-OHEN-semiquinone radical to give four major isomers (31). In the present study, we have demonstrated that similar products are produced with two of the three additional deoxynucleosides. Deoxyadenosine and dC gave two major adducts and two minor products; however, no reaction was observed with thymidine. The major structural difference between thymidine and the other DNA bases is the lack of an exocyclic amino group in the latter which suggests that coupling between 4-OHEN and the other deoxynucleosides occurs at this site. We found that care needed to be taken during the isolation of the dA adducts in particular, as any exposure to acidic environments caused hydrolysis of the sugar moiety leaving alkylated adenine. As observed with the dG adducts, the electrospray mass spectra of 4-OHEN-dA allowed the assignment of a molecular weight (MH+ ) 548) consistent with the addition of deoxyadenosine and oxygen to 4-OHEN. Positive ion electrospray MS-MS experiments were performed on 4-OHEN-dA3 to obtain molecular structure information (Supporting Information). Upon collisioninduced dissociation, the ion at m/z 548 [M + H]+ decomposed to form major fragment ions of m/z 432 [M + H - deoxyribose, 20]+, 414 [M + H - deoxyribose H2O, 15]+, 386 [M + H - deoxyribose - H2O - CO, 28]+, 297 [MH+ - dA, 5]+, 269 [MH+ - dA - CO, 23]+, 241 [MH+ - dA - 2CO, 5]+, 136 [adenine, 100]+, 117 [deoxyribose, 3]+. All of the fragments are consistent with the structure shown in Scheme 1. Finally, when 4-OHEN was incubated with dA in 50% H218O, the positive ion electrospray mass spectrum gave apparent molecular ions at 548 [M + H]+, 550 [M + H + 2]+, and 552 [M + H + 4]+ in a ratio of 1:2:1 which was consistent with the addition of two water molecules to the 4-OHENdA adduct, although the molecular formula showed only one additional oxygen. This suggests that like 4-OHENdG (31) formation of 4-OHEN-dA involved the addition of two molecules of water and the loss of one water molecule presumably from one of the catechol hydroxy groups. The structure of 4-OHEN-dA3 was further elucidated by a combination of 1D 1H NMR and 13C NMR and 2D NMR experiments including DQF-COSY, ROESY, HOHAHA, HMQC, and HMBC. The data are consistent with the adduct structure shown in Scheme 1. The B, C, and D rings of 4-OHEN remained intact as evident from the interpretation of the data derived from NMR experiments which suggested dA had alkylated the A ring of 4-OHEN similar to 4-OHEN-dG1 (31). The proton NMR spectrum (Figure 1) showed exchangeable protons at δ 4.90, 5.27, 6.06, 6.13, and 8.36 which were assigned to 5′OH-dA, 3′OH-dA, OH2, OH3, and N6H-dA, respectively. As shown in the DQF-COSY and HOHAHA spectra, the two hydroxyl groups were assigned to the C2 and C3 positions of 4-OHEN instead of C1 and C3 as in the structure of 4-OHEN-dG1 because the proton H2 at δ 4.18 was split by the vicinal hydroxyl proton (δ 6.13, assigned to OH2) and further coupled with the adjacent H1 proton at δ 5.82 which showed ROESY cross-peaks with two methylene protons at C11. The exchangeable N6H proton at δ 8.36 integrates to one proton suggesting reaction had occurred at the exocyclic amino group. In

98 Chem. Res. Toxicol., Vol. 11, No. 2, 1998

Figure 1. Partial proton NMR (500 MHz) of 4-OHEN-dA3 in DMSO-d6.

addition, there were two informative cross-correlations in the HMBC spectrum between the N6H of adenine and the C1 of 4-OHEN, as well as the H1 of 4-OHEN and the C6 of adenine, suggesting that the site of attachment between adenine and the steroid was between the exocyclic amino group and C1. This attachment was further confirmed by the significant cross-peaks between the proton pairs of N6H-dA/H1 and N6H-dA/(2 × H11) in the ROESY spectrum. The assignment of OH3 at δ 6.13 was based on the two-bond and three-bond correlations of C2 and C3 with OH3 in the HMBC spectrum. In the 13C NMR spectrum, a typical conjugated carbonyl signal was observed at δ 192.6, allowing it to be assigned at the C4 position which was further supported by a cross-correlation between the C4 carbonyl and H6 in the HMBC spectrum. Interestingly, only one aromatic resonance was detected in the proton NMR for dA at 8.12 ppm suggesting a second site of attachment between C8-dA or C2-dA and C3 on 4-OHEN. The resonance at δ 8.12 was assigned to H8-dA from cross-correlations in the HMBC spectrum between C4-dA, C5-dA and H8-dA, C8dA and 1′-H-dA. This allowed assignment of the cyclic adduct to be joined between C2-dA and C3 on 4-OHEN as well as the above-mentioned N6-dA/C1 attachment. Finally, acetylation of 4-OHEN-dA3 with acetic anhydride in the presence of pyridine produced a tetraacetate as determined by electrospray mass spectrometry. This is consistent with the two hydroxyl groups on 4-OHEN as well as the two hydroxyl groups on the sugar moiety of dA. The other 4-OHEN-dA adducts are likely diastereoisomers of 4-OHEN-dA3; however, the possibility of structural isomers cannot be ruled out at the present time. Characterization of 4-OHEN-deoxycytidine Adducts. Analysis of the spectral data for 4-OHEN-dC3 produced a similar structural pattern as 4-OHEN-dA3. The MH+ ions (524) in the electrospray mass spectra were again consistent with the addition of dC and oxygen to 4-OHEN. Positive ion electrospray MS-MS experiments were performed to obtain molecular structure information (Figure 2). Upon collision-induced dissociation, the ion at m/z 524 [M + H]+ decomposed to form major fragment ions of m/z 408 [M + H - deoxyribose, 100]+, 390 [M + H - deoxyribose - H2O, 75]+, 362 [M + H - H2O fragment shown in Figure 2, 18]+, 321 [MH+ - dR fragment shown in Figure 2, 5]+, 297 [MH+ - dC, 3]+, 117 [deoxyribose, 8]+, 112 [cytosine, 18]+. All of the fragment ions are consistent with the structure shown in Scheme 1. Finally, incubation of 4-OHEN with dC in

Shen et al.

1:1 H218O:H216O gave a similar apparent molecular ion pattern (524 [M + H]+, 526 [M + H + 2]+, and 528 [M + H + 4]+ in a ratio of 1:2:1) as described above for 4-OHEN-dA suggesting that the 4-OHEN-dC adducts also incorporate two water molecules and lose one during their formation. As with the 4-OHEN-dA3 adduct, high-field NMR experiments were performed on 4-OHEN-dC3 (Supporting Information). Unfortunately the sample was quite hydroscopic, and exchangeable protons were not visible. In the 13C NMR spectrum, a typical conjugated carbonyl signal was observed at δ 192.6, allowing it to be assigned at the C4 position. The C4 assignment position was further supported by a three-bond cross-correlation between the C4 carbonyl and H6 as well as with H2 in the HMBC spectrum. The C2-dC had informative crosscorrelations in the HMBC spectrum between the H1 proton on 4-OHEN as well as with H6-dC and 1′C-dC. Similarly, the C4-dC carbon had a three-bond correlation with the H1 proton as well as expected correlations with H6-dC and H5-dC. Derivatization with acetic anhydride also showed the addition of 4-acetate groups to 4-OHENdC3. On the basis of the molecular weight information, MS-MS data, NMR experiments, and derivatization experiment, the structure of 4-OHEN-dC3 was assigned as shown in Scheme 1. As mentioned above for the 4-OHEN-dA adducts, we believe the other dC adducts are diastereoisomers of 4-OHEN-dC3 although extensive 2D NMR experiments will be necessary in order to discount the possibility of structural isomers. Mechanistic Studies on the Rate of Formation of 4-OHEN-dG Adducts. Based on the unusual structures of the 4-OHEN-dN adducts, it seemed likely that the semiquinone radical, and not the o-quinone or the quinone methides, is involved in adduct formation. Accordingly, we determined the effects of anaerobic conditions, acidic and basic pH, and reducing agents, as well as scavengers of reactive oxygen species, on the rate of formation of the 4-OHEN-dG adducts (Table 1). Significant increases in the rate of 4-OHEN-dG formation could be observed in basic solution and in the presence of NADPH or ascorbate. In contrast, adduct formation could be completely abolished under anaerobic conditions, in acidic solution, in methanol, and in the presence of a high concentration of chelator, which strongly suggests that the deoxynucleosides react with the semiquinone radical of 4-OHEN. Superoxide dismutase, mannitol, and desferal had no effect on adduct formation indicating that superoxide, hydroxyl radical, and iron, respectively, are not involved, or at least not limiting, under these reaction conditions. In contrast, addition of hydrogen peroxide increased the rate of adduct formation, whereas catalase completely blocked the reaction which implicates hydrogen peroxide as an intermediate in the process. Competitive Reaction of Deoxynucleosides and DNA with 4-OHEN. Incubation of all four deoxynucleosides with 4-OHEN followed by HPLC analysis gave the chromatogram shown in Figure 3. All four isomers for each adduct were observed with the exception of thymidine which, as stated previously, does not react with 4-OHEN. On the basis of qualitative peak area comparisons, the rate of addition of 4-OHEN to deoxynucleosides decreases in the following order: dG > dC > dA . thymidine (no adducts detected). We then incubated solutions of calf thymus DNA with 4-OHEN and isolated, washed, and hydrolyzed the DNA. Figure 4A shows the

DNA Alkylation by 4-Hydroxyequilenin Quinoids

Chem. Res. Toxicol., Vol. 11, No. 2, 1998 99

Figure 2. Collision-induced LC/MS-MS analysis of [M + H]+ ion at m/z 524 from 4-OHEN-dC3. Table 1. Effect of Incubation Conditions on the Rate Of 4-OHEN-dG Formation conditiona

effect on 4-OHEN-dG

anaerobic pH 4.0c pH 10 CH3OHc

NDb NDb 70d NDb

Mg2+ (0.5 mM)c NADPH (5.0 mM) ascorbate (1.0 mM) H2O2 (10 mM) catalase (844 units/mL) SOD (320 units/mL)

NDb 20d 3d 6d NDb

mannitol (10 mM) desferal (5.0 mM)

1d 1d 1d

explanation semiquinone radical not formed quinone stable in acid increase in autoxidation rate redox cycling does not occur in organic solvents semiquinone radical chelated increase in rate of redox cycling increase in rate of redox cycling critical role for H2O2 critical role for H2O2 lack of requirement for superoxide lack of requirement for HO‚ lack of requirement for iron

a

Reactions contained 4-OHEN (0.4 mM), 4.0 mM dG, and pH 7.4 phosphate buffer, 37 °C, 30 min. b ND ) none detected. c 24 h. d Fold increase in peak area ratio of 4-OHEN-dG adducts as compared to 4-OHEN (0.4 mM), 4.0 mM dG, and pH 7.4 phosphate buffer, 37 °C, 30 min.

LC/MS (electrospray) analysis of the adducts obtained from the ethanol wash, and the adducts obtained after the DNA was hydrolyzed to deoxynucleosides are shown in Figure 4B. As mentioned previously, the 4-OHENdA adducts are very unstable and readily lose the sugar leaving apurinic sites on the DNA. As a result we detected these adducts in the ethanol wash before the DNA was hydrolyzed to deoxynucleosides. Adducts consistent with the MH+ of 4-OHEN-guanine adducts were also observed, although synthesis and spectral characterization will be necessary to confirm this assignment. In contrast, of the standards synthesized, only 4-OHEN-dC1 and 4-OHEN-dG1 were observed in the hydrolyzed DNA sample suggesting that there is regiospecificity in the reaction of DNA with the 4-OHENsemiquinone radical. There are additional 4-OHENdependent peaks in the full scan LC/MS chromatogram of the DNA hydrolysate which we have not identified as yet. They do not fit the retention times or mass spectra of any of our standards, and future experiments will be required to fully characterize these unknown adducts.

Figure 3. HPLC chromatogram of reaction of 4-OHEN with deoxynucleosides. 4-OHEN (1.5 mM) was incubated at pH 7.4, 37 °C, with 5.0 mM dG, dA, dC, and thymidine. After 5 h, an aliquot (60 µL) was analyzed by HPLC using a 4.6- × 150-mm Ultrasphere C-18 column (Beckman) with a flow rate of 1.0 mL/ min. The mobile phase consisted of 5% methanol in 0.25% perchloric acid/0.25% acetic acid (pH 3.5) for 2 min, increased to 27% methanol in 18 min, isocratic for 25 min, and increased to 60% methanol in 15 min.

Discussion The formation of covalent DNA adducts in vivo is usually regarded as the initiation event in the carcinogenic process (32). Unless the modified bases are promptly repaired, miscoding may result during DNA replication leading to mutations (33). With estrogens it has been shown by 32P-postlabeling methods that DNA adducts can be detected in susceptible target organs such as the Syrian hamster kidney (34, 35) or the rat prostrate (21). In addition, in vitro studies with catechol estrogens, horseradish peroxidase, and calf thymus DNA also gave DNA adducts as detected by 32P-postlabeling (20). Recent model studies with catechol estrogen o-quinones and deoxynucleosides or bases showed that different types of adducts are obtained depending on the reaction conditions and the reactivity of the o-quinone. For example, when 4-OHE-o-quinone was reacted with adenine in dimethylformamide under cathodic reduction, an adduct

100 Chem. Res. Toxicol., Vol. 11, No. 2, 1998

Figure 4. LC/MS (electrospray, positive ion) analysis of the reaction of 4-OHEN with DNA. The HPLC mobile phase was as described for LC/MS analysis in Materials in Methods except that the starting conditions were changed to 5% CH3OH for 15 min instead of 5 min to prevent the initial eluting salts from entering the mass spectrometer. (A) Analysis of ethanol washings before hydrolysis of DNA. Traces represent extracted ions at m/z 432 (MH+ of 4-OHEN-adenine) and 448 (MH+ of 4-OHEN-guanine). (B) DNA was hydrolyzed to deoxynucleosides as described in Materials in Methods. Traces represent extracted ions at m/z 524 (MH+ of 4-OHEN-dC1) and 564 (MH+ of 4-OHEN-dG1). The identity of the adducts was verified by coinjection of synthetic standards off-line using the Shimadzu HPLC system.

was isolated consistent with coupling between the C1 position of 4-OHE and the C8 position of adenine (16). The mechanism of formation implied that the semiquinone radical was the alkylating species and not the o-quinone or the quinone methide. Similarly, when sodium dithionite was used to generate the semiquinone radical of 4-OHE in the presence of DNA bases or deoxynucleosides, adducts were isolated and characterized consistent with radical coupling reactions, not with Michael addition products expected from reaction with the 4-OHE-o-quinone or quinone methide (18, 19). In contrast, incubating 4-OHE-o-quinone with deoxyguanosine in an acidic environment gave an adduct which had lost the ribose moiety resulting from reaction of the N7 position of dG with the 4-OHEN-o-quinone at C1 (17). In the same study, no adducts were detected with dA. N7 adducts are extremely unstable and readily depurinate leading to mutations (36). In contrast, 2-OHE-oquinone initially isomerizes rapidly to a quinone methide which then reacts at the C6 position with either deoxyadenosine or deoxyguanosine at the exocyclic amino group. The stable adducts could be repaired much more efficiently than apurinic sites which may explain why only 4-OHE was found to be carcinogenic in the male Syrian hamster kidney whereas no carcinogenic effects were detected with 2-OHE treatment (37). Mechanistic studies on the rate of formation of 4-OHENdG adducts strongly implicate the semiquinone radical as the alkylating species. Conditions which enhance the rate of autoxidation (i.e., base, reducing agents) lead to a corresponding increase in the rate of adduct formation.

Shen et al.

In contrast, inhibitors of autoxidation (acid, methanol, metal chelators) prevent formation of 4-OHEN-dG adducts. Finally, preliminary studies with scavengers of reactive oxygen species suggest that hydrogen peroxide and not superoxide or free hydroxyl radical is required for adduct formation. In studies investigating the mechanism of 4-hydroxyestrone-mediated oxidative damage to DNA, similar results were obtained in that hydrogen peroxide appears to be the reactive oxygen species involved (38). The exact mechanism of adduct formation is unknown at the moment; however, it likely involves autoxidation of 4-OHEN generating the semiquinone radical which abstracts a hydrogen atom from the exocyclic amino group of dNs generating dN-NH‚. Alternatively, dN-NH‚ could be formed by abstraction of a hydrogen atom by a hydroxyl radical or a superoxide anion radical. Under physiological conditions it has been shown that radical generation at the exocyclic amino group is favored over carbon-centered radicals (39). This radical combines with another 4-OHEN-semiquinone radical at the C1 position in the case of adenine and the C3 position for guanine and cytosine. Future work will involve probing this mechanism further in order to unequivocally establish the nature of deoxynucleoside adduct formation with the 4-OHEN-semiquinone radical. Characterization of the 4-OHEN-deoxynucleoside adducts enabled determinations of adduct formation in calf thymus DNA. Significant amounts of apurinic sites were produced especially through reaction with adenine as 4-OHEN-adenine adducts were detected in the ethanol wash prior to hydrolysis of the DNA (Figure 4A). As apurinic sites are known to be highly mutagenic (36) in vivo, formation of these sites by the 4-OHEN-semiquinone radical may contribute to the carcinogenic mechanism(s) of Premarin estrogens. After hydrolysis, only 4-OHEN-dG1 and 4-OHEN-dC1 were observed (Figure 4B). The competition experiment with the deoxynucleosides does predict that guanine would be the most reactive base followed by cytosine and adenine; however, as only one isomer for each dC and dG adduct was observed in the DNA hydrolysate, these data show that intact DNA impacts regio- and stereospecificity on 4OHEN-semiquinone radical alkylation. In conclusion, we have shown that 4-OHEN-semiquinone radical forms highly unusual adducts with deoxynucleosides and DNA. Our data suggest that several different types of DNA lesions could be expected including apurinic sites and bulky stable adducts, in addition to the published oxidized damage to DNA caused by 4-OHEN (26). If similar adducts are formed in vivo which are not repaired efficiently, mutations could result leading to initiation of the carcinogenic process in the endometrium or breast. Finally, it should be noted that equilenin or 17β-dehydroequilenin are the major urinary (40) and biliary (41) metabolites of equilin, and it is quite possible that 4-OHEN-semiquinone radicals are formed from these estrogens as well. The implication of these adducts to the biological effects of Premarin is not known; however, given the direct link between long-term estrogen replacement therapy and the enhanced risk of breast cancer, the potential for formation of redox-active/electrophilic metabolites from all of the estrogens in estrogen replacement formulations needs to be explored.

Acknowledgment. This research was supported by NIH Grant CA73638-01. We thank Ms. Sue Blake,

DNA Alkylation by 4-Hydroxyequilenin Quinoids

Department of Chemistry, Queen’s University, for obtaining the 500-MHz 2D NMR spectra (Bruker, Avance DRX 500) of dA and dC adducts. Supporting Information Available: Characterization details of the 4-OHEN-dC and 4-OHEN-dA adducts (20 pages). Ordering information is given on any current masthead page.

References (1) 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. (2) Feigelson, H. S., and Henderson, B. E. (1996) Estrogens and breast cancer. Carcinogenesis 17, 2279-2284. (3) Liehr, J. G. (1990) Genotoxic effects of estrogens. Mutat. Res. 238, 269-276. (4) Vogel, V. G., Yeomeans, A., and Higginbotham, E. (1993) Clinical management of women at increased risk for breast cancer. Breast Cancer Res. Treat. 28, 195-210. (5) Colditz, G. A., Hankinson, S. E., Hunter, D. J., Willett, W. C., Manson, J. E., Stampfer, M. J., Hennekens, C., Rosner, B., and Speizer, F. E. (1995) The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. N. Engl. J. Med. 332, 1589-1593. (6) Harris, R. B., Laws, A., Reddy, F. M., King, A., and Haskell, W. L. (1990) Are women using postmenopausal estrogens? A community survey. Am. J. Public Health 80, 1266-1268. (7) Grodstein, F., Stampfer, M. J., Colditz, G. A., Willett, W. C., Manson, J. E., Joffe, M., Rosner, B., Fuchs, C., Hankinson, S. E., Hunter, D. J., Hennekens, C. H., and Speizer, F. E. (1997) Postmenopausal hormone therapy and mortality. N. Engl. J. Med. 336, 1769-1775. (8) Lobo, R. A. (1995) Benefits and risks of estrogen replacement therapy. Am. J. Obstet. Gynecol. 173, 982-989. (9) Paganinihill, A. (1995) Estrogen replacement therapy and stroke. Prog. Cardiovasc. Diseases 38, 223-242. (10) Wickelgren, I. (1997) Estrogen: A new weapon against Alzheimer’s. Science 276, 676-677. (11) Henderson, V. W. (1997) The epidemiology of estrogen replacement therapy and Alzheimer’s disease. Neurology 48, S27-S35. (12) Martucci, C. P., and Fishman, J. (1993) P450 enzymes of estrogen metabolism. Pharmacol. Ther. 57, 237-257. (13) Yager, J. D., and Liehr, J. G. (1996) Molecular mechanisms of estrogen carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 36, 203232. (14) Nutter, L. M., Ngo, E. O., and Abul-Hajj, Y. J. (1991) Characterization of DNA damage induced by 3,4-estrone-o-quinone in human cells. J. Biol. Chem. 266, 16380-16386. (15) Musarrat, J., Arezina-Wilson, J., and Wani, A. A. (1996) Prognostic and aetiological relevance of 8-hydroxyguanosine in human breast carcinogenesis. Eur. J. Cancer 32A, 1209-1214. (16) Abul-Hajj, Y. J., Tabakovic, K., and Tabakovic, I. (1995) An estrogen-nucleic acid adduct. Electroreductive intermolecular coupling of 3,4-estrone-o-quinone and adenine. J. Am. Chem. Soc. 117, 6144-6145. (17) Stack, D. E., Byun, J., Gross, M. L., Rogan, E. G., and Cavalieri, E. L. (1996) Molecular characteristics of catechol estrogen quinones in reactions with deoxyribonucleosides. Chem. Res. Toxicol. 9, 851-859. (18) Akanni, A., Tabakovic, K., and Abul-Hajj, Y. J. (1997) Estrogennucleic acid adducts: Reaction of 3,4-estrone-o-quinone with nucleic acid bases. Chem. Res. Toxicol. 10, 477-481. (19) Akanni, A., and Abul-Hajj, Y. J. (1997) Estrogen-nucleic acid adducts: Reaction of 3,4-estrone-o-quinone radical anion with deoxynucleosides. Chem. Res. Toxicol. 10, 760-766. (20) Dwivedy, I., Devanesan, P., Cremonesi, P., Rogan, E., and Cavalieri, E. (1992) Synthesis and characterization of estrogen 2,3- and 3,4-quinones. Comparison of DNA adducts formed by the quinones versus horseradish peroxidase-activated catechol estrogens. Chem. Res. Toxicol. 5, 828-833. (21) Han, X., Liehr, J. G., and Bosland, M. C. (1995) Induction of DNA adduct detectable by 32P-postlabeling in the dorsolateral prostrate of NBL/Cr rats treated with estradiol-17β and testosterone. Carcinogenesis 16, 951-954. (22) Purdy, R. H., Moore, P. H., Williams, M. C., Goldzheher, H. W., and Paul, S. M. (1982) Relative rates of 2- and 4-hydroxyestrogen synthesis are dependent on both substrate and tissue. FEBS Lett. 138, 40-44. (23) Li, J. J., Li, S. A., Oberley, T. D., and Parsons, J. A. (1995) Carcinogenic activities of various steroidal and nonsteroidal

Chem. Res. Toxicol., Vol. 11, No. 2, 1998 101 estrogens in the hamster kidney: Relation to hormonal activity and cell proliferation. Cancer Res. 55, 4347-4351. (24) Sarrabia, S. F., Zhu, B. T., Kurosawa, T., Tohma, M., and Liehr, J. G. (1997) Mechanism of cytochrome P450-catalyzed aromatic hydroxylation of estrogens. Chem. Res. Toxicol. 10, 767-771. (25) Shen, L., Pisha, E., Huang, Z., Pezzuto, J. M., Krol, E., Alam, Z., van Breemen, R. B., and Bolton, J. L. (1997) Bioreductive activation of catechol estrogen-ortho-quinones: Aromatization of the B ring in 4-hydroxyequilenin markedly alters quinoid formation and reactivity. Carcinogenesis 18, 1093-1101. (26) Han, X., and Liehr, J. G. (1995) Microsome-mediated 8-hydroxylation of guanine bases of DNA by steroid estrogens: Correlation of DNA damage by free radicals with metabolic activation to quinones. Carcinogenesis 16, 2571-2574. (27) NIH. (1981) NIH Guidelines for the Laboratory Use of Chemical Carcinogens, NIH Publication No. 81-2385, U.S. Government Printing Office, Washington, DC. (28) Teuber, H. J. (1953) Reaktionen mit nitrosodisulfonat (III). Equilenin-chinon. Chem. Ber. 86, 1495-1499. (29) Gehrke, C. W., McCune, R. A., and Kenneth, C. K. (1984) Quantitative reversed-phase high-performance liquid chromatography of major and modified nucleosides in DNA. J. Chromatogr. 301, 199-219. (30) Lewis, M. A., Yoerg-Graff, D., Bolton, J. L., and Thompson, J. A. (1996) Alkylation of 2′-deoxynucleosides and DNA by quinone methides derived from 2,6-di-tert-butyl-4-methylphenol. Chem. Res. Toxicol. 9, 1368-1374. (31) Shen, L., Qiu, S., van Breemen, R. B., Zhang, F., Chen, Y., and Bolton, J. L. (1997) Reaction of the Premarin metabolite 4-hydroxyequilenin semiquinone radical with 2′-deoxyguanosine: Formation of unusual cyclic adducts. J. Am. Chem. Soc. 119, 1112611127. (32) Klaassen, C. D. (1996) Casarett & Doull’s Toxicology: The Basic Science of Poisons, McGraw Hill, New York, NY. (33) Singer, B., and Hang, B. (1997) What structual features determine repair enzyme specificity and mechanism in chemically modified DNA? Chem. Res. Toxicol. 10, 713-732. (34) Liehr, J. G., Avitts, T. A., Randerath, E., and Randerath, K. (1986) Estrogen-induced endogenous DNA adduction: Possible mechanism of hormonal cancer. Proc. Natl. Acad. Sci. U.S.A. 83, 53015305. (35) Liehr, J. G., Hall, E. R., Avitts, T. A., Randerath, E., and Randerath, K. (1987) Localization of estrogen-induced DNA adducts and cytochrome P450 activity at the site of renal carcinogenesis in the hamster kidney. Cancer Res. 47, 2156-2159. (36) Loeb, L. A., and Preston, B. D. (1986) Mutagenesis by apurinic/ apyrimidinic sites. Annu. Rev. Genet. 20, 201-203. (37) Liehr, J. G., Fang, W. R., Sirbasku, D. A., and Ari-Ulubelen, A. (1986) Carcinogenicity of catechol estrogens in Syrian hamsters. J. Steroid Biochem. 24, 353-356. (38) Li, Y., Trush, M. A., and Yager, J. D. (1994) DNA damage caused by reactive oxygen species originating from a copper-dependent oxidation of the 2-hydroxy catechol of estradiol. Carcinogenesis 15, 1421-1427. (39) Zady, M. F., and Wong, J. L. (1980) Unusual competition between nitrogen and carbon methylation of nucleosides by methyl radical in various aqueous media. J. Am. Chem. Soc. 45, 2373-2377. (40) Bhavnani, B. R., Cecutti, A., and Wallace, D. (1994) Metabolism of [3H] 17β-dihydroequilin and [3H] 17β-dihydroequilin sulfate in normal postmenopausal women. Steroids 59, 389-394. (41) Ikegawa, S., Itoh, M., and Goto, J. (1994) Separatory determination of biliary metabolites of equilin in rat by high-performance liquid chromatography. J. Liq. Chromatogr. 17, 223-239.

TX970181R