Mutation Research, 238 (1990) 269-276

269

Elsevier MUTREV 02811

Genotoxic effects of estrogens Joachim G. Liehr Department of Pharmacology and Toxicology, University of Texas Medical Branch, Galveston, TX 77550- 2774 (U.S.A.)

(Accepted 12 October 1989) Keywords: Oestrogen-inducedcancer; Catecboloestrogens;Diethylstilboestrol;DNA adducts; Oestrogens, metabolic activation

Summary Estrogens are associated with several cancers in humans and are known to induce tumors in rodents. In this review a mechanism of carcinogenesis by estrogens is discussed which features the following key events: (1) Steroid estrogens are metabolized by estrogen 2- and 4-hydroxylases to catecholestrogens. Target organs of estrogen-induced carcinogenesis, hamster kidney or mouse uterus, contain high levels of estrogen 4-hydroxylase activity. Since the methylation of 4-hydroxyestradiol by catechol-O-methyltransferase is inhibited by 2-hydroxyestradiol, it is proposed that a build up of 4-hydroxyestrogens precedes estrogen-induced cancer. (2) The catecholestrogen or diethylstilbestrol (DES) are oxidized to semiquinones and quinones by the peroxidatic activity of cytochrome P-450. The quinones are proposed to be (the) reactive intermediates of estrogen metabolism. (3) The quinones may be reduced to catecholestrogens and DES and redox cycling may ensue. Redox cycling of estrogens has been shown to generate free radicals which may react to form the organic hydroperoxides needed as cofactors for oxidation to quinones. (4) The quinone metabolites of catechol estrogens and of DES bind covalently to D N A in vitro whereas DNA binding in vivo has only been examined for DES. When DES is administered to hamsters, the resulting D E S - D N A adduct profile in liver, kidney, or other organs closely matches that of DES q u i n o n e - D N A adducts in vitro. In vitro, D E S - D N A adducts are chemically unstable and are generated in incubations with organic hydroperoxide as cofactor. It is proposed that the instability of adducts and the lower sensitivity of previous assay methods contributed to the reported failures to detect adducts. Steroid estrogen-DNA adducts in vivo are currently under investigation. (5) Tumors are postulated to arise in cells rapidly proliferating due to the growth stimulus provided by the estrogenic activity of the primary estrogen or of hormonaily potent metabolites such as 4-hydroxyestradiol. The covalent modification of D N A in these cells is temporary because of the chemical instability of adducts and will result in altered genetic messages in daughter cells, whereas in non-proliferating cells there may be no lasting genetic damage. The sequence of events described above is a plausible mechanism for tumor initiation by estrogens and is partially substantiated by experimental evidence obtained in vitro a n d / o r in vivo.

The studies from this laboratory described in this review were supported by grants from the National Cancer Institute, NIH (CA43232, CA43233 and CA44069).

Abbreviations: CE, catecholestrogen(s);DES, diethylstilbestrol;

DES Q, diethylstilbestrol-4',4"-quinone;2-OH-E2, 2-hydroxy-. estradiol; 4-OH-E2, 4-hydroxyestradioi.

Correspondence: Dr. Joachim G. Liehr, Department of Pharmacologyand Toxicology,Universityof Texas Medical Branch, Galveston, TX 77550-2774 (U.S.A.). 0165-1110/90/$03.50 © 1990 ElsevierSciencePublishers B.V. (BiomedicalDivision)

270

The use of estrogens has increasingly been associated with human cancer (IARC Monogr., 1979). The administration of the synthetic estrogen diethylstilbestrol (DES) to pregnant women with the goal of stabilizing pregnancies has been associated with an increased risk of mammary cancer in these women (Greenberg et al., 1984) and with vaginal adenocarcinomas in their daughters (Herbst et al., 1971). Use of steroidal estrogens has been implicated in the rising incidence of endometrial adenocarcinoma observed in the 1960's (Ziel and Finkle, 1975; Smith et al., 1975; Mack et al., 1976). More recently, indications were found that prolonged use of estrogen-containing contraceptive medication at an early age may be associated with an increased risk of breast cancer [discussed in an Editorial of The Lancet (1989)]. In laboratory animals, estrogens induce kidney tumors in hamsters (Kirkman, 1959), uterine tumors in mice (Walker, 1983), Leydig cell tumors in mice (Huseby, 1972), or pituitary tumors in rats (Clifton and Mayer, 1956). These and additional animal models are reviewed in the IARC Monograph (1979). The mechanism of carcinogenesis by estrogens is discussed in this report.

Mechanism of estrogen-induced carcinogenesis Most mechanistic studies in estrogen-induced carcinogenesis have been carried out using the hamster kidney tumor model described by Kirkman (1959). In this model, the hormonal potency of various estrogens did not correlate with the incidence of tumor induction (Liehr, 1983; Li and Li, 1984). Therefore, an epigenetic mechanism of kidney tumor induction by estrogens based on hormonal imbalance or uncontrolled stimulation of cell proliferation (Furth, 1982) was considered unlikely. Rather, metabolic activation of estrogens was postulated (Metzler and McLachlan, 1978; Liehr et al., 1982, 1987) to play an etiologic role in tumor induction, because it was found to be necessary for the expression of genotoxic activity of DES in several in vitro test systems such as induction of sister chromatid exchange (Rudiger et al., 1979), unscheduled DNA synthesis (Tsutsui et al., 1984), and mutation of several strains of S. cerevisiae (Mehta and von Borstel, 1982). Specifically, the CE have been considered (Metzler and Mc-

Lachlan, 1978; Liehr et al., 1982) to be genotoxic metabolic intermediates of estrogens, because (i) they are hydroquinones which can be readily oxidized to reactive quinones (Rao and Axelrod, 1962); (ii) they are oxidized to intermediates which bind covalently to peptides (Kuss, 1969) and proteins (Marks and Hecker, 1969); (iii) of their efficient detoxification in vivo to the less reactive methyl ether conjugates (reviewed by Ball and Knuppen, 1980). The reactivity of CE and their corresponding quinones support the proposal of a mechanism of carcinogenesis by estrogens based on tumor initiation by covalent DNA damage (Fig. 1) and subsequent hormone-dependent growth stimulation of initiated cells. The following key features of this proposed carcinogenic action of estrogens will be discussed: (1) Conversion of steroid estrogens to CE. A build up of CE because of either elevated rates of CE synthesis or deficient conjugation of catechol metabolites creates favorable conditions for carcinogenesis to occur. (2) Oxidation of CE to quinones. This oxidation is mediated by the organic hydroperoxide-dependent peroxidatic activity of cytochrome P-450. DES is a hydroquinone and may directly be oxidized to its quinone, DES Q, without initial conversion to a catechol. (3) Redox cycling between estrogen quinones, semiquinones and the CE or DES (i.e. the hydroquinones). The redox cycling generates free radicals which may covalently damage DNA. The free radicals may also react to form lipid peroxides which in turn support organic hydroperoxide-dependent CE formation and redox cycling. (4) Covalent binding of estrogen quinones to DNA. This binding may be considered a tumor initiating event, but may be chemically unstable. (5) Estrogen-dependent growth of initiated cells. DNA damage by estrogens may be a necessary but not sufficient event for tumor formation. In this review, key features of this proposed mechanism of estrogen-induced carcinogenesis are examined: (1) Metabolic formation of CE A wide range of rates of CE formation has been reported for microsomes of rat or hamster liver or

271 kidney, when CE were measured by different assay procedures (reviewed by Roy et al., 1989a). These discrepancies exist because one of these methods, the catechol-O-methyltransferase-coupied radioenzymatic assay has been shown to underreport 2- and, more markedly, 4-hydroxylation of estrogens (Roy et al., 1989a). When only the results of the more reliable direct product isolation assays are considered, hamster kidney microsomal activity of total estrogen 2- and 4-hydroxylase were 56 pmoles/mg protein/min at saturating substrate concentration (Liehr et al., 1990) with 2-OH-E 2 and 4-OH-E 2 formed in approximately equal amounts. Chronic estrogen treatment decreased the activity of renal estrogen 4-hydroxylase by 25%, that of 2-hydroxylase by 75% of control values making estrogen 4-hydroxylase the dominant activity over 2-hydroxylase. In liver, which is not a target organ of estrogen-induced carcinogenesis, these ratios do not change appreciably as a function of estrogen treatment (Liehr et al., 1990). These data demonstrate that hamster kidney, a target susceptible to estrogen-induced cancer, contains high basal estrogen 4-hydroxylase activity which becomes predominant during the course of chronic estrogen treatment. The channeling of metabolic activity towards formation of 4-OH-E 2 assumes added significance when the low basal activity of catechol-O-methyltransferase in hamster is considered (Li et al., 1989). This enzyme methylates CE and thus detoxifies these reactive metabolites of estrogens. However, the methylation of 4-OH-E z is inhibited by 2-OH-E 2 in vitro (Roy et al., 1989b). If this inhibition occurs in vivo, significant quantities of 4-OH-E 2 may build up in kidneys of estrogentreated hamsters as substrates for quinone formation and redox cycling. (2) Metabolic oxidation of CE or DES to quinones The oxidation of CE or DES to corresponding quinones in vitro is catalyzed by peroxidases and H202 (Liehr et al., 1983) or by the peroxidatic activity of cytochrome P-450 and organic hydroperoxide (Liehr et al., 1986). In hamster kidney, the basal peroxidatic activity (38 pmoles/mg protein/min) is 40% of hepatic activity (94 pmoles/ mg protein/min) and increases upon chronic

estrogen treatment. When normalized for specific content of cytochrome P-450, peroxidatic activity in hamster liver does not change with estrogen treatment for 2 months, whereas that in kidney increases from 317 in controls to 1025 pmoles/ nmole P-450/min (Liehr et al., 1990). Enzyme activity measurements are supported by concentrations of quinone metabolites found. DES Q was found in low albeit detectable quantities in liver, kidney and uterus of pregnant female hamsters treated with 3H-labeled DES (Roy and Liehr, 1989). DES Q was also detected in kidney and liver of fetus in vivo. Moreover, fetal liver homogenate formed DES Q from DES in vitro. These data demonstrate that estrogen quinones are metabolically formed in vitro and in vivo. (3) Redox cycling of estrogens Redox cycling between DES and DES Q or between CE and their corresponding quinones has been demonstrated (Liehr et al., 1986). The metabolic oxidation of estrogen to quinones by the organic hydroperoxide-dependent peroxidatic activity of cytochrome P-450 is described above. The reduction of quinones to their corresponding hydroquinones (CE or DES) is catalyzed by NADPH-dependent cytochrome P-450 reductase (Liehr et al., 1986; Roy and Liehr, 1988). Renal activity of this enzyme increases upon chronic estrogen treatment 2-3-fold when expressed in relation to specific content of cytochrome P-450 (Liehr et al., 1990). This change in renal enzyme activity may not necessarily decrease estrogen quinone concentrations, because resulting hydroquinones (DES or CE, respectively) are substrates for renewed oxidation to quinones by the peroxidatic activity of cytochrome P-450 (Liehr et al., 1986). The result, redox cycling of estrogens, is likely to be enhanced in estrogen-exposed hamster kidney because of the changes in renal enzyme activities outlined above. Formation of free radicals from redox cycling of stilbene estrogen is elevated when catalyzed by kidney microsomes of estrogen-treated hamsters compared to controls (Roy and Liehr, 1988; Liehr et al., 1990). The role of free radicals in estrogen-induced carcinogenesis has not yet been sufficiently elucidated. One of the consequences of free radical formation is lipid peroxidation (Nagata et al., 1982). Lipid hydro-

272

METHYLATED CONJUGATED ESTROGENS

E2

1

)CE

the postulated tumor-initiating activity of estrogen quinones and their DNA adducts.

SEMIQUlNONE

QUINONE

DNA

ADDUCTS Fig. 1. Proposed mechamsm of genoto~city of estrogens. Steroid estrogens are converted to CE by estrogen 2- and/or 4-hydroxylases. (1) Redox cycling between the CE or DES and their respective quinones generates free radicals. Oxidations are catalyzed by the peroxidatic activity of cytochrome P-450. (2) Reductions by cytochrome P-450 reductase. (3) The reduction of quinones by quinone reductase (4) does not proceed via the semiquinone free radical. The quinones bind covalently to DNA and these adducts initiate tumorigenesis. The conversion of CE or DES to methylated a n d / o r conjugated estrogens metabolites prevents the sequence of events outlined,

peroxides in turn fuel the organic hydroperoxidedependent peroxidatic activity of cytochrome P450 and organic hydroperoxide-dependent CE formation. Thus, conditions which generate organic hydroperoxides facilitate additional CE formation from estrogens (Levin et al., 1987) and quinone formation from CE and DES (Liehr et al., 1986). Enzymatic and non-enzymatic processes possibly protect from quinone formation and redox cycling. Quinone reductase (DT-diaphorase) has been shown to reduce DES Q mainly to Z-DES and some E-isomer (Roy and Liehr, 1988). Hamster renal quinone reductase activity temporarily decreases by 80% in response to chronic estrogen treatment for 1 month. Although this activity recovers after 2 or more months of exposure to estrogen, it may be too late to prevent the genotoxic action of estrogen quinones. An efficient reduction of DES Q to DES is also effected by L-ascorbic acid (vitamin C). Indeed, administration of L-ascorbic acid to estrogen-treated hamsters lowers the incidence of kidney tumors by approximately 50% (Liehr and Wheeler, 1983). This decrease in tumor incidence correlates with a 50% decrease in DES Q concentrations in vivo (Roy and Liehr, 1989). These correlations support

(4) Covalent binding of estrogens to DNA The possibility of covalent binding of an activated estrogen metabolite to DNA has been repeatedly examined, since such covalent interaction is accepted as an initiating event in chemical carcinogenesis. Both stilbene and steroid estrogens have been found to bind to DNA following microsomal or chemical activation (Blackburn et al., 1976; Metzler and McLachlan, 1978). In vivo, binding of radioactivity to DNA has also been reported after injection of radioactive estrone, ethinylestradiol (Jaggi et al., 1978) or DES (Lutz et al., 1982) into rats and hamsters. Unfortunately, a covalent nature of the binding could not be demonstrated in any of these studies. During isolation of the nucleic acids used in these experiments, radioactivity bound to the DNA or nucleotides continued to dissociate and was extracted with organic solvents (Liehr et al., 1985). Because of the lack of evidence for covalent binding of estrogens to DNA, estrogens have been proposed to be epigenetic carcinogens (Furth, 1982). Covalent binding in vivo When a more rapid and more sensitive assay for DNA adduct formation such as 32p-postlabeling was used, a unique and defined set of adducts was demonstrated in DNA of liver, kidney or uterus of hamsters injected with a single large dose of DES (Fig. 2) (Gladek and Liehr, 1989). These adducts were observed in addition to background radioactivity present in DNA of treated animals and of controls. The DES-DNA modifications consisted of 1 major and several minor adducts. These data demonstrate that DES has genotoxic activity. Adduct levels range from 0.9 × 108 nucleotides/adduct in liver to 7.8 × 108 nucleotides/adduct in uterus of hamsters having received 200 mg/kg DES. Covalent binding in vitro The formation and structures of DES-DNA adducts were investigated also in vitro to elucidate mechanistic details. When DNA modification was studied in vitro under conditions used previously in incubations of DES, DNA, rat-liver micro-

273

D

E

F

O

1

~4

m

Fig. 2. 32p-Postlabeling analysis of DNA of DES-treated hamsters (220 mg/kg) and of untreated controls. The upper panel shows adduct maps of liver (A), kidney (B), uterus (C) of untreated controls and liver (D), kidney (E), and uterus (F) of DES-treated animals. The major adduct (Spot 4) was seen in all three DNAs of treated animals (D-F). Kidney and liver DNA (D,E) also showed minor DES-DNA adducts in the upper fight quadrant of the adduct map. somes and N A D P H (Blackburn et al., 1976), only minor amounts of D E S - D N A adduction occurred (Gladek and Liehr, unpublished) (Fig. 3). In contrast, adduct formation was increased 10-fold with an organic hydroperoxide as cofactor replacing N A D P H . The adduct pattern generated with hydroperoxide as cofactor in vitro closely matched that obtained in vivo. Since DES Q is formed under these conditions, pure DES Q, and D N A without any protein were also incubated and resuited in almost identical adduct patterns (Gladek and Liehr, 1989). A comparable adduct pattern was generated by the reaction of DES Q and dGMP. Thus, adduct formation in vitro or in vivo occurs by the interaction of DES Q with the guanine base. d A M P reacted with DES Q to a minor extent whereas the pyrimidine nucleotides

did not react at all. These reactions demonstrated that DES Q is the major reactive genotoxic intermediate of DES in D N A adduction in vitro and in vivo. Moreover, the data show that D N A damage by DES is critically dependent on the availability of organic hydroperoxide as cofactor necessary for the formation of reactive quinone metabolite. When incubations of DES Q and D N A were carried out for various lengths of time, the adduct yields were found to decrease. Since there was no protein in the incubation mixtures, the decreases in adduct yield could not have been caused by repair processes. Rather, a chemical instability of the adducts was documented to be the cause for the decreasing adduct yield (A. Gladek and J.G. Liehr, unpublished). The half-life of the major adduct spot was 4 - 5 days at 37°C. In vivo, repair

274

Fig. 3. 32p-Postlabeling analysis of products of the incubation of purified DNA, hamster liver microsomes and DES in absence of cofactor (A), or in the presence of NADPH (B) or cumene hydroperoxide (C). Panel C shows the same adduct pattern observed when DES Q was reacted with dGMP, i.e. a major adduct (spot 4) and 3 minor adducts. Only traces of spot 4 are observed when NADPH was used as cofactor (B), whereas panel A showed only the background radioactivity of hamster-liver DNA.

processes acted synergistically resulting in a biological half-life of this adduct of approximately 14 h. The chemical instability of the D E S - D N A adducts in addition to inadequate detection limits of the previously used assay systems are very likely among the causes for the previously experienced failures to detect D N A adduction in vitro. Quinones generated from catechol estrogens have also been found to be D N A reactive in vitro when investigated by 32p-postlabeling (A. Gladek and J.G. Liehr, unpublished). The existence of C E - D N A adducts in vivo and its role in tumor initiation by steroid estrogens is currently under investigation. These data demonstrate that quinone intermediates formed by metabolic activation of DES or CE are genotoxic agents in estrogen metabolism. The role of estrogens in D N A damage and tumor initiation is tightly linked to the metabolic oxidation of estrogens to reactive quinones.

Indirect DN.4 adduction In addition to estrogen q u i n o n e - D N A adducts described above, another class of estrogen-induced covalent D N A adducts have been observed specifically in kidney of hamsters, a target of

estrogen-induced cancer (Liehr et al., 1986a). Identical sets of these adducts were induced by chronic treatment of hamsters with structurally diverse estrogens. Therefore, it was concluded that estrogens were not part of the D N A adduct, but induced endogenous substances of as yet unknown structure to bind covalently to DNA. Structures and biological significance of these modified nucleotides are currently under investigation. Because of their chromatographic mobilities these adducts were postulated (Liehr et al., 1986a) to have been formed by binding of lipids (lipid peroxides?).

Estrogen-dependent growth of initiated cells D E S - D N A adducts after administration of DES to a hamster have been observed in m a n y organs, target and non-target cells for estrogeninduced cancer (Gladek and Liehr, 1989). Therefore, estrogen quinone metabolite and q u i n o n e D N A adduct formation m a y be a necessary but not sufficient event for tumors to arise. Since most estrogen-associated cancer is estrogen-dependent or -responsive, at least in its early stages, estrogen-dependent cell transformation a n d / o r growth m a y also be necessary for the development

275 of cancer. T h e h o r m o n a l p o t e n c y of the p r i m a r y estrogens or the high h o r m o n a l activity of 4 - O H - E 2 (Martucci a n d F i s h m a n , 1979) or other m e t a b o lites m a y s u p p o r t this e s t r o g e n - d e p e n d e n t growth. H o r m o n e - d e p e n d e n t growth is b e y o n d the scope of this discussion, b u t is reviewed elsewhere (Lel a n d et al., 1981). Conclusion T h e data presented here d e m o n s t r a t e that D E S is genotoxic in vitro a n d i n vivo. C E are genotoxic in vitro, a role of this process i n steroid-estrogen toxicity a n d carcinogenesis in vivo r e m a i n s to be s h o w n a n d is c u r r e n t l y u n d e r investigation. Metabolic o x i d a t i o n of the stilbene estrogen to a q u i n o n e , the D N A - r e a c t i v e metabolite, is the key feature of the p a t h w a y of genotoxicity of DES. This D N A a d d u c t i o n has n o t b e e n detected previously, because a d d u c t nucleotides are chemically u n s t a b l e a n d because less sensitive D N A adduction assays have previously b e e n used. A central role of q u i n o n e s i n steroid-estrogen toxicity a n d genotoxicity in vivo r e m a i n s to be ascertained. Q u i n o n e f o r m a t i o n a n d a d d u c t f o r m a t i o n are critically d e p e n d e n t o n lipid hydroperoxides as cofactors for c y t o c h r o m e P-450 metabolism. A likely source of organic hydroperoxides is the f o r m a t i o n b y free radicals generated b y redox cycling of C E or D E S a n d s u b s e q u e n t reaction of these radicals with lipids.

References Ball, P., and R. Knuppen (1980) Catecholestrogens: Chemistry, biogenesis, metabolism, occurrence and physiological significance, Acta Endocrinol., Suppl. 232, 2-127. Blackburn, G.M., M.H. Thompson and H.W.S. King (1976) Binding of diethylstilbestrol to DNA by rat liver microsomal fractions in vitro and in mouse fetal cells in culture, Biochem. J., 158, 643-646. Clifton, K.H., and R.K. Meyer (1956) Mechanism of anterior pituitary tumor induction by estrogen, Anat. Record, 125, 65-81. Editorial (1989) Cancer risks of oral contraception, The Lancet, No. 8628, 21-22. Furth, J. (1982) Hormones as etiological agents in neoplasia, in: F.F. Becker (Ed.), Cancer, a Comprehensive Treatise, Etiology: Chemical and Physical Carcinogenesis, 2nd edn. Plenum, New York, pp. 89-134. Gladek, A., and J.G. Liehr (1989) Mechanism of genotoxicity of diethylstilbestrol in vivo, J. Biol. Chem., 264, 1684716852.

Greenberg, E.R., A.B. Barnes, Z. Resseguie, J.A. Barrett, S. Burnside, L.L. Lanza, R.K. Neff, M. Stevens, R.H. Young and T. Colton (1984) Breast cancer in mothers given diethylstilbestrol in pregnancy, N. Engl. J. Med., 311, 1393-1398. Herbst, A.L., H. Ulfelder and D.C. Poskanzer (1971) Adenocarcinoma of the vagina: Association of maternal stilbestrol therapy with tumor appearance in young women, N. Engl. J. Med., 284, 878-881. Huseby, R.A. (1972) Hormonal factors in relation to cancer, in: (M.D. Anderson Hospital Symposium) Environment and Cancer, Williams and Wilkins, Baltimore, pp. 372-388. IARC (1979) Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Sex Hormones II, Vol. 21, IARC, Lyon. Jaggi, W., W.K. Lutz and C. Schlatter (1978) Covalent binding of ethinylestradiol and estrone to rat liver DNA in vitro, Chem.-Biol. Interact., 23, 13-18. Kirkman, H. (1959) Estrogen-induced tumors of the kidney in the Syrian hamster. III. Growth characteristics in the Syrian hamster, Natl. Cancer Inst. Monogr. No. 1, 1-59. Kuss, E. (1969) Wasserl~SslicheMetabolite des Ostradiols-17fl, III. Trennung und Identifizierung der 1- und 4Glutathionthio~ither von 2,3-Dihydroxy-rstratrienen, Hoppe-Seyler's Z. Physiol. Chem., 350, 95-97. Leland, E.F., M. Iio and D.A. Sirbasku (1981) Hormone-dependent cell lines, in: G.H. Sato (Ed.), Functionally Differentiated Cell Lines, Liss, New York, pp. 1-46. Levin, M., J. Weisz, Q.D. Bui and R.J. Santen (1987) Peroxidatic catecholestrogen production by human breast cancer tissue in vitro, J. Steroid Biochem., 28, 513-520. Li, J.J., and S.A. Li (1984) Estrogen-induced tumorigenesis in hamsters: Roles for hormonal and carcinogenic activities, Arch. Toxicol., 55, 110-118. Li, S.A., R.H. Purdy and J.J. Li (1989) Variations in catecholO-methyltransferase activity in rodent tissues: possible role in estrogen carcinogenicity, Carcinogenesis, 10, 63-67. Liehr, J.G. (1983) 2-Fluoroestradiol: Separation of estrogenicity from carcinogenicity, Mol. Pharmacol., 23, 278-281. Liehr, J.G., and W.J. Wheeler (1983) Inhibition of estrogen-induced renal carcinoma in Syrian hamsters by vitamin C, Cancer Res., 43, 4638-4642. Liehr, J.G., B.B. DaGue, A.M. Ballatore and D.A. Sirbasku (1982) Multiple roles of estrogen in estrogen-dependent renal clear cell carcinoma of Syrian hamster, in: G.H. Sato, A.B. Pardee and D.A. Sirbasku (Eds.), Growth of Cells in Hormonally Defined Media, Cold Spring Harbor Conferences on Cell Proliferation Vol. 9, book A, Cold Spring Harbor Press, Cold Spring Harbor, New York, pp. 445-458. Liehr, J.G., B.B. DaGue, A.M. Ballatore and J. Henkin (1983) Diethylstilbestrol (DES) quinone: a reactive intermediate in DES metabolism, Biochem. Pharmacol., 16, 3711-3718. Liehr, J.G., B.B. DaGue and A.M. Ballatore (1985) Reactivity of 4',4"-diethylstilbestrol quinone, a metabolic intermediate of diethylstilbestrol, Carcinogenesis, 6, 829-836. Liehr, J.G., A.A. Ulubelen and H.W. Strobel (1986a) Cytochrome P-450-mediated redox cycling of estrogens, J. Biol. Chem., 261, 16865-16870. Liehr, J.G., T.A. Avitts, E. Randerath and K. Randerath (1986b) Estrogen-induced endogenous DNA adduction:

276 Possible mechanism of hormonal cancer, Proc. Natl. Acad. Sci. (U.S.A.), 83, 5301-5305. Liehr, J.G., R.H. Purdy, J.S. Baran, E.F. Nutting, F. Colton, E. Randerath and K. Randerath (1987) Correlation of aromatic hydroxylation of lift-substituted estrogens with morphological transformation in vitro but not with in vivo tumor induction by these hormones, Cancer Res., 2583-2588. Liehr, J.G., D. Roy, A.A. Ulubelen, Q.P. Bui, J. Weisz and H.W. Strobel (1990) Effect of chronic treatment of Syrian hamsters on microsomal enzymes mediating formation of catechol estrogens and their redox cycling: Implications for carcinogenesis, J. Steroid Biochem., in press. Lutz, W.K., W. Jaggi and C.H. Schlatter (1982) Covalent binding of diethylstilbestrol to DNA in rat and hamster liver and kidney, Chem.-Biol. Interact., 42, 251-257. Mack, T.M., M.C. Pike, B.E. Henderson, R.I. Pfeffer, V.R. Gerkins, M. Arthur and S.E. Brown (1976) Estrogens and endometrial carcinoma in a retirement community, N. Engl. J. Med., 294, 1262-1267. Marks, F., and E. Hecker (1969) Metabolism and mechanism of action of oestrogens, XII. Structure and mechanism of formation of water-soluble and protein-bound metabolites of oestrone in rat liver microsomes in vitro and in vivo, Biochim. Biophys. Acta, 187, 250-265. Martucci, C.P., and J. Fishman (1979) Impact of continuously administered catechol estrogens on uterine growth and luteinizing hormone secretion, Endocrinology, 105, 12881292. Mehta, R.D., and R.C. yon Borstel (1982) Genetic activity of diethylstilbestrol in Saccharomycescerevisiae, Enhancement of mutagenicity by oxidizing agents, Mutation Res., 92, 49-61. Metzler, M., and J.A. McLachlan (1978) Oxidative metabolism of diethylstilbestrol and steroidal estrogens as a potential factor in their fetotoxicity, in: D. Neubert, H.J. Merker, H. Nau and J. Langman (Eds.), Role of Pharmacokinetics in Prenatal and Perinatal Pharmacology, Georg Thieme, Stuttgart, pp. 157-163. Nagata, C., M. Kodama, Y. Ioki and T. Kimura (1982) Free radicals produced from chemical carcinogens and their significance in carcinogenesis, in: R.A. Floyd (Ed.), Free Radicals and Cancer, Marcel Dekker, New York, pp. 1-62.

Rao, P.N., and L.R. Axelrod (1962) Studies on the preparation of 2,3-o-quinones from 2-hydroxyestrogens, Biochim. Biophys. Acta, 60, 404-406. Roy, D., and J.G. Liehr (1988) Temporary decrease in renal quinone reductase activity induced by chronic administration of estradiol to male Syrian hamsters, Increased superoxide formation by redox cycling of estrogens, J. Biol. Chem., 263, 3646-3651. Roy, D., and J.G. Liehr (1989) Metabolic oxidation of diethylstilbestrol to diethylstilbestrol-4',4"-quinone in Syrian hamsters, Carcinogenesis, 10, 1241-1245. Roy, D., Q.D. Bui, J. Weisz and J.G. Liehr (1989a) Comparison of assays for catechol estrogen synthase activity: Product isolation vs. radioenzymatic catechol-O-methyltransferase-coupled procedures, J. Steroid Biochem., 33, 243249. Roy, D., J.G. Liehr and J. Weisz (1989b) The O-methylation of 4-hydroxyestradiol is inhibited by 2-hydroxyestradiol: implications for carcinogenesis, Proc. Am. Assoc. Cancer Res., 30, 156. Riidiger, H.W., F. Haenisch, M. Metzler, F. Oesch and H.R. Glatt (1979) Metabolites of diethylstilbestrol induce sister chromatid exchange in human cultured fibroblasts, Nature (London), 281, 392-394. Smith, D.C., R. Prentice, D.J. Thompson and W.L. Hermann (1975) Association of exogenous estrogen and endometrial carcinoma, N. Engl. J. Med., 293, 1164-1167. Tsutsui, T., G.H. Degen, D. Schiffmann, A. Wong, H. Maizumi, J.A. McLachlan and J.C. Barrett (1984) Dependence on exogenous metabolic activation for induction of unscheduled DNA synthesis in Syrian hamster embryo cells by diethylstilbestrol and related compounds, Cancer Res., 44, 184-189. Walker, B. (1983) Uterine tumors in old female mice exposed prenatally to diethylstilbestrol, J. Natl. Cancer Inst., 70, 477-484. Ziel, H.K., and W.D. Finkle (1975) Increased risk of endometrial carcinoma among users of conjugated estrogens, N. Engl. J. Med., 293, 1167-1170.