A

34aWqitt©

K

1

bmaaomefaciaq

0d

Available online at www.sciencedirect.com

Journal of Steroid Biochemistry & Molecular Biology 109 (2008) 22–30

Reduction of estrogen-induced transformation of mouse mammaryepithelial cells by N-acetylcysteine

Divya Venugopal a,1, Muhammad Zahid a,1, Paula C. Mailander a, Jane L. Meza b,Eleanor G. Rogan a, Ercole L. Cavalieri a, Dhrubajyoti Chakravarti a,∗

a Eppley Institute for Research in Cancer and Allied Diseases, 986805 Nebraska Medical Center, Omaha,NE 68198-6805, United States

b Preventive and Societal Medicine, 984350 Nebraska Medical Center, Omaha,NE 68198-4350, United States

bstract

A growing number of studies indicate that breast cancer initiation is related to abnormal estrogen oxidation to form an excess of estrogen-,4-quinones, which react with DNA to form depurinating adducts and induce mutations. This mechanism is often called estrogen genotoxicity.-Catechol estrogens, precursors of the estrogen-3,4-quinones, were previously shown to account for most of the transforming and tumorigenicctivity. We examined whether estrogen-induced transformation can be reduced by inhibiting the oxidation of a 4-catechol estrogen to its quinone.e demonstrate that E6 cells (a normal mouse epithelial cell line) can be transformed by a single treatment with a catechol estrogen or its

uinone. The transforming activities of 4-hydroxyestradiol and estradiol-3,4-quinone were comparable. N-Acetylcysteine, a common antioxidant,nhibited the oxidation of 4-hydroxyestradiol to the quinone and consequent formation of DNA adducts. It also drastically reduced estrogen-inducedransformation of E6 cells. These results strongly implicate estrogen genotoxicity in mammary cell transformation. Since N-acetylcysteine is wellolerated in clinical studies, it may be a promising candidate for breast cancer prevention.

ion; N

ccgt[

iewtnec

2007 Elsevier Ltd. All rights reserved.

eywords: Catechol estrogen quinone; Mammary epithelial cells; Transformat

. Introduction

In 1896, Beatson reported that removal of the ovary regressedreast cancer in women, suggesting a role of endogenous hor-ones in the disease [1]. Initial studies identified this hormone

s estrogen, and found it to act by a receptor-mediated mech-nism to cause breast cancer [2]. Later studies suggested thatxidative metabolites, formed as a result of abnormal estrogenetabolism in the breast, are also a major cause of this dis-

ase [3,4]. The abnormal metabolism primarily involves theormation of high levels of carcinogenic 4-catecholestrogensnd their quinones (estrogen-3,4-quinones) [4,5]. A signifi-ant number of breast cancer patients appear to suffer from

t. For example, in one study with 77 patients (49 controlsnd 28 breast cancer cases), 4-hydroxyestrogens, estrogen-3,4-uinones and their derivatives were found elevated in 54% of

∗ Corresponding author. Tel.: +1 402 559 2951; fax: +1 402 559 8068.E-mail address: dchakrav@unmc.edu (D. Chakravarti).

1 Equal contributors.

ttc(tbat

960-0760/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.jsbmb.2007.12.003

-acetylcysteine

ases and 10% of controls [5]. This breast cancer phenotypean be the result of polymorphisms in estrogen-metabolizingenes, and/or alterations in the expression of the enzymeshat favor the oxidation of estrogens to form these metabolites4,6,7].

The 4-catecholestrogens and the estrogen-3,4-quinones maynitiate breast cancer by a genotoxic mechanism [4]. Thestrogen-3,4-quinones are the ultimate carcinogens, as they reactith DNA to form depurinating adducts, which are detectable in

he breast tissue [8] and in the urine [4]. Estrogen-DNA depuri-ating adducts show a strong association with breast cancer. Forxample, one study indicated that cancerous human breast tissuean have 30-fold more estrogen-induced depurinating adductshan normal human breast [8], and another study indicated thathe urine of breast cancer patients as well as high risk womenontain ∼2-fold more of these adducts than normal womenGaikwad et al., submitted). The depurinating adducts spon-

aneously dissociate from DNA, forming abasic sites. In thereast, these abasic sites induce the mutations that may initi-te breast cancer. These ideas are supported by the observationhat 4-catecholestrogens and estrogen-3,4-quinones account for

hemis

tc

bocrfvtaioSetttt

capwohtIi[hNdacpas

2

2

ddp(

2

KNm1w

ecmsL

2

59o(miCatom

tpi3cwpct(vbeei

2

2

(t3a(hvi(

2m

D. Venugopal et al. / Journal of Steroid Bioc

he majority of the mutagenic [9–12], transforming [3,13] andarcinogenic activities of estrogens [14–17].

To establish the initiating role of estrogen-3,4-quinones inreast cancer, it is necessary to demonstrate a direct link betweenxidative estrogen metabolism and an early event of breast can-er. In a preliminary study, four antioxidants (N-acetylcysteine,esveratrol, melatonin and reduced lipoic acid) were examinedor their ability to inhibit estrogen-DNA adduct formation initro [18]. These antioxidants, when added in equimolar doseso 4-catecholestrogens, showed 33–77% inhibition of DNAdduct formation [18]. In the present study, we have exam-ned whether N-acetylcysteine can be used to explore the rolef oxidative metabolites of estrogen in breast cancer initiation.pecifically, we examined whether N-acetylcysteine can reducestrogen-DNA adduct formation and inhibit estrogen-inducedransformation of murine breast epithelial cells. A demonstrationhat targeting estrogen metabolism can prevent the transforma-ion of murine breast epithelial cells may lead to new approacheso breast cancer prevention.

N-Aceylcysteine is an aminothiol antioxidant and a pre-ursor of cysteine and glutathione [19]. It was initially useds a topical mucolytic drug [20] and as an antidote foraracetamol poisoning [21]. Later, its antioxidant propertiesere found useful for preventing the progression of chronicbstructive pulmonary disease (COPD) [22]. Other studiesave demonstrated that N-acetylcysteine has very low sys-emic toxicity and can cross the blood–brain barrier [23].ts long-term use has been evaluated at 1–3 × 600 mg/dn clinical trials for the prevention of chronic bronchitis24,25] and renal dysfunction [26]. Thus, N-acetylcysteineas desirable properties for long-term use. The discovery that-acetylcysteine can prevent mutagenesis by oxidative DNAamage [27–29] and decrease the incidence of pre-neoplasticnd neoplastic lesions by various smoke-related chemical car-inogens in rodents [23,30–32] pointed to its chemopreventiveotential. Unfortunately, clinical trials did not indicate N-cetylcysteine to be effective in preventing lung cancer inmokers [33].

. Materials and methods

.1. Chemicals and reagents

Estradiol (E2) was oxidized to 4-OHE2 and E2-3,4-Q asescribed previously [34]. E2-3,4-Q was reacted with Ade andG to generate the depurinating adduct standards as describedreviously [34]. N-Acetylcysteine was purchased from SigmaSt. Louis, MO).

.2. Cell lines and culture conditions

Mouse mammary epithelial cells (E6) were a gift from Dr..H. Cowan (University of Nebraska Medical Center, Omaha,

E) [35]. These cells were originally isolated from the mam-ary gland of the Brca1fl/fl mouse (loxP sites flanking exon

1 of the Brca1 gene). They were immortalized by infectingith HPV-16E6 (Neo+) retrovirus to inhibit p53. The E6 cells

wd8

try & Molecular Biology 109 (2008) 22–30 23

xpress a full-length Brca1 protein and are considered “normal”ells. E6 cells were cultured in 1:1 Dulbecco’s minimal essentialedium: Ham’s F-12 (DMEM:F-12, Mediatech, Herndon, VA)

upplemented with 10% bovine growth serum (BGS, Hyclone,ogan, UT) in a 5% CO2 incubator at 37 ◦C.

.3. Estrogen cytotoxicity

Exponentially growing cells were seeded at a density of000 cells/well in 96-well plates. After a day (day 0), cells in one6-well plate were counted by the MTT assay (see below), whilether cells were treated with 4-OHE2 (5-100 �M) or E2-3,4-Q5-100 �M) and incubated for 24 h. The estrogen solutions wereade in acetonitrile (final concentration 0.007%). Following this

ncubation, cells were rinsed with PBS (Invitrogen, Carlsbad,A), fresh media was added, and the cells were cultured fornother 3 days. Cell numbers at days 1–3 were determined byhe MTT assay. The effect of N-acetylcysteine on the growthf estrogen-treated cells was similarly determined (used in 1:1olar ratio).In the MTT [3-(4, 5-dimethyl-thiazolyl-2)-2,5-diphenyl-

etrazolium bromide, Sigma] assay, the media in the 96-welllate cultures were replaced with 100 �l fresh medium contain-ng 25 �l of MTT (5 mg/ml in PBS), and incubated for 2 h at7 ◦C to allow the reduction of MTT by metabolically-activeells to form a purple formazan precipitate. The precipitateas then solubilized by adding 20% SDS in 1:1 DMF:H2O,H 4.7 (100 �L) and incubating overnight at 37 ◦C. The purpleolor was read (at 570 nm) in a �Quant microplate spectropho-ometer (Bio-Tek Instruments) and analyzed by the KCjuniorversion 1.41) software. The absorbance values were con-erted into cell numbers using a standard curve constructedy plotting MTT assay absorbance against cell counts. Lin-ar interpolation was used to estimate the IC50. Survivalstimates are presented using means and 95% confidencentervals.

.4. Analysis of estrogen-DNA adducts

.4.1. Cell culture and treatmentE6 cells were cultured for 72 h in estrogen-free medium

phenol red-free DMEM-F12 with charcoal stripped FBS), andreated (20 × 106 cells) either with N-acetylcysteine (30 �M,0 min pre-treatment) and 4-OHE2 (30 �M) or with N-cetylcysteine (50 �M, 30 min pre-treatment) and E2-3,4-Q50 �M) for 24 h. Following the treatments, the media werearvested and supplemented with 2 mM ascorbic acid (to pre-ent possible decomposition of the compounds) and processedmmediately. Media from E6 cells treated with 10 �l acetonitrilesolvent) were used as controls.

.4.2. Sample preparation for analyzing estrogenetabolites and DNA adducts

Varian C8 Certify II cartridges (Varian, Harbor City, CA)

ere equilibrated by sequentially passing 1 mL of methanol,istilled water, and potassium phosphate buffer (100 mM, pH) through them. The harvested media (40 mL) were adjusted

2 hemistry & Molecular Biology 109 (2008) 22–30

tpt8a

2

t(fiEaebati5(got6rpqrt

tA[

2

wwNcisi0www

2

vHtXf

Few

aE

3

3e

sndpdt

sOa7otcbd

4 D. Venugopal et al. / Journal of Steroid Bioc

o pH 8.0 with 1 mL of 1 M potassium phosphate buffer, andassed through these cartridges. The retained analytes in the car-ridges were washed with the above phosphate buffer, eluted with:1:1:0.1 of methanol:acetonitrile:water:trifluoroacetic acid,nd processed as described previously [34].

.4.3. Analysis of estrogen metabolites and adductsThe eluted samples were analyzed in an HPLC appara-

us equipped with a multi-channel electrochemical detectorModel 580 solvent delivery modules, Model 540 auto-samplertted with a 12-channel CoulArray electrochemical detector,nvironmental Sciences Association, Chelmsford, MA). Thenalytes and adducts were separated with solvent gradients gen-rated by solvent A [15:5:10:70 of acetonitrile:methanol:CAAuffer (5.25% citric acid, 3.85% ammonium acetate, 11.5%cetic acid):water] and solvent B [50:20:10:20 of acetoni-rile:methanol:CAA buffer:water]. The samples were injectednto a Phenomenex Luna-2 C-18 column (250 mm × 4.6 mm,�m; Phenomenex, Torrance, CA), and eluted isocratically

90% solvent A:10% solvent B) for 10 min, then by a linearradient (up to 90% solvent B) in the next 35 min, at a flow ratef 1 ml/min. The 12 coulometric electrodes were set at poten-ials of −35, 10, 70, 140, 210, 280, 350, 420, 490, 550, 620 and90 mV. The analyte and adduct peaks were identified by theiretention times and peak height ratios between the dominanteak and the peaks in the two adjacent channels. The data wereuantified by comparing with known amounts of standards. Theesults were compared between groups using a Mann–Whitneyest.

The results were confirmed by a MicroMass QuattroMicroriple stage quadrupole mass spectrometer attached to a Waterscquity UPLC (Waters, Milford, MA) as described previously

34].

.5. Soft agar colony formation assay

Cells (1.17 × 106) were cultured for 24 h, and then treatedith 4-OHE2 (3 or 30 �M) or E2-3,4-Q (5 or 50 �M) with orithout equimolar amounts of N-acetylcysteine for another 24 h.ext, the cells were harvested, counted (Coulter Z-series particle

ount and size analyzer, Beckman, Fullerton, CA) and seededn 6-well plates for the soft agar assay. Briefly, 5000 cells wereuspended in 2 mL of 0.4% Noble agar (Sigma, St Louis, MO)n 10% BGS (Hyclone) and poured over a 2-mL underlayer of.6% Noble agar (in the culture medium) in each well. After 3eeks of incubation at 37 ◦C in a 5% CO2 incubator, coloniesere counted under a phase-contrast microscope. The resultsere compared between groups using a Mann–Whitney test.

.6. Analysis of H-ras mutations

The procedure for mutation analysis has been described pre-iously [9,36,37]. Briefly, the exon 1-2 region of the mouse

-ras gene was PCR amplified, the product cloned in pUC18,

he recombinant plasmids transformed in E. coli, and using the-gal/IPTG color assay, bacterial colonies harvested, cultured

or the extraction of plasmids and the sequence of H-ras DNA

barl

ig. 1. Cytotoxicity of 4-OHE2 and E2-3,4-Q in normal mouse mammarypithelial cells (E6). Cells were treated with the estrogens at day 0. Cell numbersere determined by the MTT assay using a standard curve.

nalyzed. The results were statistically evaluated by Fisher’sxact test.

. Results

.1. 4-OHE2 and E2-3,4-Q toxicity in E6 mammarypithelial cells

Survival plots for examining estrogen cytotoxicity were con-tructed by converting the MTT absorbance values to cellumbers using a standard curve. The ratios of cell numbers atays 0–3 with respect to those in the control wells at day 0 werelotted as a time course (Fig. 1). These experiments were con-ucted under exponential growth conditions, as can be seen fromhe growth of the untreated control cells.

Both 4-OHE2 and E2-3,4-Q showed a remarkably narrowpectrum of cytotoxicity in E6 breast epithelial cells. For 4-HE2, a dose-dependent response was observed between 10

nd 50 �M, and for E2-3,4-Q this spectrum was between 25 and0 �M. The duration of decline in cell numbers in the 4-OHE2-r E2-3,4-Q-treated cells was dose-dependent. For example, athe high doses (≥50 �M for 4-OHE2 and ≥60 �M for E2-3,4-Q),ell numbers declined for all 3 days and there was no regrowtheyond this period (not shown). At the low and intermediateoses (3–40 �M 4-OHE2 and 25–40 �M E2-3,4-Q), cell num-

ers declined initially (compared to control) and then increasedgain (not shown). The short cytotoxic period followed byegrowth suggests that these treatments of the estrogen metabo-ites induce acute cytotoxicity. Although these results suggest

hemistry & Molecular Biology 109 (2008) 22–30 25

4ivf

dET4t

3

thtsirptfet

i4soOaphi

avcm

m4rm0itt[cm

7oEN

Fid

viQq

tccoicw

mtqatiFt

3a

Nm

D. Venugopal et al. / Journal of Steroid Bioc

-OHE2 to be more cytotoxic than E2-3,4-Q, it may be an exper-mental artifact, as the extreme reactivity of the quinone witharious cellular macromolecules may minimize its availabilityor mechanisms that elicit the cytotoxic response.

To determine the 50% killing dose (IC50), we chose estrogenoses within the linear decline periods. For both 4-OHE2 and2-3,4-Q, the cytotoxic doses showed linear declines up to 1 d.herefore, the IC50 values were calculated at 1 d. The IC50 for-OHE2 is estimated to be 34 �M and for E2-3,4-Q it is estimatedo be 48 �M.

.2. N-Acetylcysteine inhibits estrogen cytotoxicity

N-Acetylcysteine is known to rescue cells from the cyto-oxic effects of various chemicals. Several mechanisms of rescueave been indicated, including N-acetylcysteine blocking oxida-ive DNA damage [38–43] by reacting with reactive oxygenpecies [44]. Metabolism of estrogens and similar compoundss known to induce oxidative DNA damage as a byproduct ofedox cycling between semiquinones and quinones [45–47]. Arevious study showed that N-acetylcysteine can inhibit cyto-oxicity by the catechol estrogen 2-OHE2 by minimizing theormation of oxidative DNA damage by 2-OHE2 [48]. How-ver, whether 2-OHE2-induced oxidative DNA damage is thearget of the protective activity of N-acetylcysteine is unclear.

Although ortho-semiquinones are thought to have poor abil-ty to reduce oxygen to superoxide [41,49], both 2-OHE2 and-OHE2 are known to induce oxidative stress [48,50]. They haveimilar redox potentials [51,52], and form comparable levels ofxidative DNA damage [53–55]. Despite these similarities, 4-HE2 is a much stronger carcinogen than 2-OHE2 [17,56,57]

nd it is detected in greater amounts in cancerous breast com-ared to normal human breast [5,58]. These results support theypothesis that 4-OHE2 is the precursor of the quinone thatnitiates breast cancer.

We treated E6 cells with 1:1 mixtures of N-acetylcysteinend either 4-OHE2 or E2-3,4-Q for 24 h and examined cell sur-ival at 48 h (Fig. 2). The results indicate that N-acetylcysteinean protect cells from the cytotoxic effects of these estrogenetabolites.For example, the addition of N-acetylcysteine improved the

ean survival of E6 cells treated with 30 �M 4-OHE2 from 0 to3.9% (i.e., a mean survival of 43.9%, 95% confidence interval,ange 35.6–52.2%). Similarly, N-acetylcysteine improved theean survival of E6 cells treated with 50 �M E2-3,4-Q from

.7 to 64.5% (i.e., a mean survival of 63.8%, 95% confidencenterval, range 41.8–85.8%). This level of protection is similar tohat reported when MCF-10A human breast epithelial cells werereated with 2-OHE2 (10–20 �M) and N-acetylcysteine (10 mM)48]. The similarity of N-acetylcysteine rescue of E6 cells fromytotoxicity by 4-OHE2 or E2-3,4-Q suggests that the quinoneay play a critical role in cytotoxicity.E2-3,4-Q has a relatively short half-life (t1/2 ≈ 45 min at pH

.0) (unpublished results); it reacts rapidly with DNA [59]r self-polymerizes to form inert compounds [60]. Although2-3,4-Q can be reduced by NAD(P)H alone [61] or byAD(P)H-quinone oxidoreductase 1 (NQO1) in vitro [62], pre-

otat

ig. 2. N-Acetylcysteine rescues E6 cells from 4-OHE2 and E2-3,4-Q cytotox-city. Equimolar amounts of N-acetylcysteine were added to the experiments atay 0. Cell numbers were determined by the MTT assay using a standard curve.

ious studies suggest that the reduction of E2-3,4-Q is inefficientn cells [47,63]. Therefore, the cytotoxic effect of E2-3,4-

may be related mainly to the chemical reactions of theuinone.

A previous study also implicated E2-3,4-Q in 4-OHE2 cyto-oxicity [41]. It was found that hypoxic or aerobic cultureonditions did not alter 4-OHE2 cytotoxicity in human breastarcinoma cells (MCF-7), but the addition of ascorbic acidr cysteine protected the cells, whereas a nitroxide (Tempol)ncreased the cytotoxicity. It was proposed that ascorbic acid andysteine act by minimizing oxidation of 4-OHE2 to E2-3,4-Q,hereas Tempol acts by favoring quinone formation [41].N-Acetylcysteine may act as a quencher of E2-3,4-Q to

inimize estrogen cytotoxicity. It conjugates with E2-3,4-Qo form 4-OHE2-2-NAcCys (Fig. 3A), thereby decreasing theuinone pool for the DNA adduct-forming reaction [18,64]. Inddition, N-acetylcysteine, through the mercapturic acid biosyn-hesis pathway, can be transformed to cysteine, which is involvedn the formation of glutathione that can also quench E2-3,4-Q.urthermore, the intracellular glutathione level itself is related

o cell survival [41].

.3. N-Acetylcysteine inhibits estrogen oxidation and DNAdduct formation in E6 cells

To address the above issues, we examined the effects of-acetylcysteine on estrogen metabolism and DNA adduct for-ation (Fig. 3B and C). Treatment of the cells with 4-OHE2

r E2-3,4-Q produces three major conjugates (methoxy, glu-athione and N-acetylcysteine conjugates) and two major DNAdducts (adenine- and guanine-depurinating adducts). Underhese conditions, relatively little of the 4-OHE2 remains free,

26 D. Venugopal et al. / Journal of Steroid Biochemistry & Molecular Biology 109 (2008) 22–30

F E2 ana A ad

am(bfbC(opNaaawc

QaOkOcc

sit

ig. 3. N-Acetylcysteine inhibits estrogen-DNA adduct formation in E6 cells.dducts from E1 and E2 together. The quinone produced less conjugates and DN

major portion is converted by the abundant catechol-O-ethyltransferase (COMT) to form the methoxy conjugates

4-OCH3E1/2) [5,6,65–67] and the remaining 4-OHE2 appears toe oxidized to form the E2-3,4-Q. The quinones have three majorates. Some are reduced by NAD(P)H:quinone oxidoreductasesack to the catechols [7,62], which are then methylated byOMT. Second, the quinones react with endogenous glutathione

GSH) to form glutathione conjugates (4-OHE1/2-2-SG), a partf which are catabolized by the mercapturic acid biosynthesisathway, forming N-acetylcysteine conjugates (4-OHE1/2-2-AcCys) [68]. Third, the quinones react with DNA and formdducts. E2-3,4-Q forms primarily (99.99%) depurinating DNA

dducts (roughly equal amounts of N3Ade and N7Gua adducts)nd 0.001% stable DNA adducts [59] Treatment of E6 cellsith 4-OHE2 or E2-3,4-Q for 24 h showed the expected three

onjugates and the two DNA adducts (Fig. 3B and C).

4ti6

d E1 are interconverted in the cell; therefore, we combined the conjugates andducts than the catechol (note the difference in the Y-axes of B and C).

We used moderately cytotoxic doses of 4-OHE2 and E2-3,4-for these experiments for efficient analysis of the estrogen

nalytes. As can be seen from Fig. 1, the 30 �M dose of 4-HE2 kills ∼22% of the cells and the 50 �M dose of E2-3,4-Qills ∼59% of the cells at 24 h. Under these conditions, the 4-HE2 -treated cells generated greater quantities of the three

onjugates and the two DNA adducts than the E2-3,4-Q-treatedells (Fig. 3B and C, note the scales on the Y-axes).

The addition of equimolar amounts of N-acetylcysteine mea-urably altered the levels of the estrogen conjugates. Specifically,n the 4-OHE2-treated cells (Fig. 3B), N-acetylcysteine alteredhe methoxy conjugate levels from 19,192 ± 952 (S.E.M.) to

5,600 ± 3308 (S.E.M.) pmol (138% increase, p = 0.05), the glu-athione conjugate levels from 36.7 ± 4.1 to 79.3 ± 6.4 (116%ncrease, p = 0.05) and N-acetylcysteine-conjugate levels from6.6 ± 6.4 to 112.7 ± 4.4 pmol/20 million cells (69% increase,

hemistry & Molecular Biology 109 (2008) 22–30 27

pa1cnfpteectccnitta

NmtOgo(t(rfuj

afettf

Nca5fp

etlNoc

3t

h

T

T

E

C

4

E

E

D. Venugopal et al. / Journal of Steroid Bioc

= 0.05). In E2-3,4-Q-treated cells (Fig. 3C), N-acetylcysteineltered the methoxy conjugate levels from 3079.7 ± 132.9 to260.7 ± 42.4 pmol (59% decrease, p = 0.05), the glutathioneonjugate levels from 36.7 ± 0.7 to 32.7 ± 3.3 pmol (no sig-ificant change, p = 0.2), and N-acetylcysteine-conjugate levelsrom 36 ± 3.1 to 78 ± 3.1/20 million cells (117% increase,= 0.05). N-Acetylcysteine can react directly with E2-3,4-Q

o form the 4-OHE2-2-NAcCys conjugate [64] (Fig. 3A). Asxpected, the addition of N-acetylcysteine increased the lev-ls of these conjugates in both 4-OHE2 and E2-3,4-Q-treatedultures. Similarly, N-acetylcysteine is a precursor of glu-athione [19], which can react with E2-3,4-Q to produce aonjugate (Fig. 3A). N-Acetylcysteine increased glutathioneonjugate levels in 4-OHE2-treated cultures. However, this wasot observed in the E2-3,4-Q-treated cultures. This differencen result could be due to rapid depletion of E2-3,4-Q by reac-ion with various nucleophilic groups present in the cell andhe medium, before sufficient glutathione is produced from N-cetylcysteine.

The most remarkable result was that in 4-OHE2-treated cells,-acetylcysteine caused a great increase in the levels of the 4-ethoxy conjugates (Fig. 3B). This result may be explained with

he observation that the semiquinone produced by oxidation of 4-HE2 in these cells, can be reduced back to 4-OHE2 by the thiolroup of cysteine [41]. This process can increase the pool sizef 4-OHE2 that is methylated by catechol-O-methyltransferaseFig. 3A). In contrast, N-acetylcysteine caused a decrease inhe levels of 4-methoxy conjugates in the E2-3,4-Q-treated cellsFig. 3C). In this case, the 4-OHE2 can only be obtained byeduction of the quinone, and N-acetylcysteine apparently inter-eres with this process. We are conducting further studies tonderstand the contributions of N-acetylcysteine towards con-ugate formation.

Thus, N-acetylcysteine can interfere with estrogen oxidationnd minimize quinone levels. Since the quinones are neededor reaction with DNA to form adducts, N-acetylcysteine can be

xpected to reduce DNA adduct formation. A drastic reduction inhe depurinating DNA adduct levels was observed. In 4-OHE2-reated cells, N-acetylcysteine reduced the N3Ade adductsrom 52 ± 2.0 to 5.3 ± 0.7 pmol (90% reduction, p = 0.04) and

OemF

able 1

reatments

strogen metabolite (�M) N-Acetylcysteine (�M)

ontrol – –– 30– 50

-OHE2 3 –30 –30 30

2-3,4-Q 5 –50 –50 50

2-2,3-Q 10 –

a 5000 cells/well.b p-values are adjusted for multiple comparisons with the control using the method

Fig. 4. Colony morphology of estrogen-transformed E6 cells.

7Gua adducts from 43.3 ± 1.8 to 7.3 ± 0.7 pmol/20 millionells (83% reduction, p = 0.05). In E2-3,4-Q-treated cells, N-cetylcysteine reduced the N3Ade adducts from 18 ± 1.2 to.3 ± 0.7 pmol (71% reduction, p = 0.05) and N7Gua adductsrom 18 ± 1.2 to 6.7 ± 0.7 pmol/20 million cells (63% reduction,= 0.05).

Thus, N-acetylcysteine can drastically reduce depurinatingstrogen-DNA adduct formation in breast epithelial cells. Sincehe depurinating adducts are considered to be the precursoresions that induce the mutations leading to breast cancer [9–12],-acetylcysteine may be useful for probing the link betweenxidative metabolism of estrogens and the initiation of breastancer.

.4. N-Acetylcysteine inhibits estrogen-inducedransformation of E6 cells

Previous studies showed that estrogens can transform normaluman breast epithelial cells [3]. We examined whether 4-

HE2 or E2-3,4-Q can similarly transform E6 cells and whether

quimolar amounts of N-acetylcysteine can block the transfor-ation. A typical estrogen-transformed E6 colony is shown inig. 4.

Transformation total colonies/totalwells (% cells transformeda)

p-valueb (comparedwith control)

70/88 (0.016%) –17/24 (0.015%) 0.9918/24 (0.016%) 0.99

58/19 (0.060%) <0.00152/16 (0.067%) <0.00121/15 (0.028%) 0.50

53/20 (0.053%) <0.00160/20 (0.058%) <0.00142/20 (0.042%) 0.002

27/24 (0.025%) 0.99

of Bonferroni.

28 D. Venugopal et al. / Journal of Steroid Biochemistry & Molecular Biology 109 (2008) 22–30

F (Gens below

tn0t

(qarotent

Ebppatr[

a5(tf(0Aapt

coo

tsc1pt

pt(cnmfsmTow

4

hMbl[bcsbt3

oditiOcsa

cpsThe ability of N-acetylcysteine in inhibiting estrogen-induced

ig. 5. Spectrum of pre-existing H-ras mutations in E6 cells. Exon 1-2 regionhown over the top line of the cartoon, and the wild-type nucleotides are shown

E6 cells showed a background frequency of transforma-ion of 0.016 % (Table 1). Addition of N-acetylcysteine didot significantly alter transformation: 0.015% (p = 0.99) and.016% (p = 0.99) with 30 or 50 �M N-acetylcysteine, respec-ively.

Treatment of E6 cells with 4-OHE2 (3 or 30 �M) or E2-3,4-Q5 or 50 �M) increased the number of transformed colonies [fre-uency = 0.06% (p < 0.001) or 0.067% (p < 0.001) for 4-OHE2,nd 0.053% (p < 0.001) or 0.058% (p < 0.001) for E2-3,4-Q,espectively]. The chosen estrogen doses [4-OHE2 (3 or 30 �M)r E2-3,4-Q (5 or 50 �M)] did not show a dose-response inransformation. The lack of a dose-response suggests that thextent of cytotoxicity induced by these doses of estrogen doesot allow significant changes in the frequency of transforma-ion.

In contrast to 4-OHE2 and E2-3,4-Q, treatment with 10 �M2-2,3-Q showed a minimal increase in transformation overackground (0.025%, p = 0.99, compared with control). Theoor transforming activity of E2-2,3-Q was expected, as itsrecursor, 2-OHE2, is known to have weaker transformingctivity compared to 4-OHE2 [3]. It has been suggestedhat the weak carcinogenic activity of 2-OHE2/E2-2,3-Q iselated to their poor ability to form depurinating adducts59].

Next, we examined whether the addition of equimolarmounts of N-acetylcysteine (30 �M to 30 �M 4-OHE2 and0 �M to 50 �M E2-3,4-Q) would reduce transformationTable 1). In both cases, N-acetylcysteine drastically reducedransformation of the E6 cells. N-Acetylcysteine reduced trans-ormation by 4-OHE2, from 0.067 to 0.028% for 30 �M 4-OHE258% reduction, p = 0.01), and by E2-3,4-Q from 0.058 to.042% for 50 �M E2-3,4-Q (28% reduction, p = 0.09). N-cetylcysteine reduction of transformation was more effective

gainst 4-OHE2 (residual transformation was similar to control,= 0.50), than against E2-3,4-Q (residual transformation higher

han control, p = 0.002).These effects of N-acetylcysteine provide a tool to directly

orrelate oxidative metabolism of estrogens with transformationf breast epithelial cells, which is expected to be an early eventf breast cancer.

It is possible that the background frequency of transforma-ion could be related to intrinsic properties of the E6 cells,uch as pre-existing mutations. To examine whether E6 cells

ontained pre-existing mutations, we PCR amplified the exon–2 region of the H-ras gene (500 bp), cloned the product inUC18, transformed into E. coli, isolated single colonies usinghe IPTG-X-gal colony color assay, extracted the recombinant

tHpp

Bank accession No. U89950) was analyzed for mutations. The mutations are.

lasmids and sequenced the inserts (Fig. 5). This PCR pro-ocol for mutation analysis generates low levels of mutationsone mutation in ∼40 plasmids, mutation frequency = 5 × 10−5,alculated by dividing the number of mutations with the totalumber of bases sequenced) [9,12,37]. Analysis of 31 plas-id clones from the E6 cells showed 8 mutations (mutation

requency = 51.6 × 10−5, p = 0.01, Fisher’s Exact test compari-on of PCR-induced vs. pre-existing mutations). Seven of theseutations were G.C to T.A and the eighth mutation was A.T to.A. Since the E6 cells contained pre-existing mutations, a rolef these mutations in spontaneous transformation of E6 cellsarrants consideration.

. Discussion

The E6 are normal mouse breast epithelial cells. Similaruman cells exist (MCF-10A and MCF-10F). In particular, theCF-10F cells are considered to be a good model of the human

reast [13]. Unfortunately, both MCF-10A and 10F cells haveow CYP1B1, and thus show poor ability to oxidize estrogens69]. Estrogen oxidation is more efficient in normal humanreast [5]. Estrogens can transform both MCF-10A and 10Fells, but four treatments are required. Results presented herehow that the E6 cells are efficient estrogen oxidizers, and cane transformed with a single estrogen treatment. This study ishe first to demonstrate the transforming activity of estrogen-,4-quinones.

The purpose of this study was to examine whether the anti-xidant N-acetylcysteine can be used to test the hypothesis thatepurinating DNA adduct formation by estrogen-3,4-quinoness the initiating event in breast cancer. A transformation protocolhat requires four treatments can model the combined effects ofnitiation and promotion from sustained carcinogen exposure.n the other hand, transformation by a single treatment of a

arcinogen can report its initiating activity. Therefore, for thistudy, we think that the E6 are more suitable than the MCF-10And 10F cells.

N-Acetylcysteine prevented estrogen-DNA damage and asso-iated death of E6 cells (Figs. 2 and 3). In addition toreventing estrogen-DNA adduct formation, N-acetylcysteinetrongly inhibited estrogen-induced transformation (Table 1).

ransformation should be validated in appropriate human cells.owever, we think that the present results provide a proof-of-rinciple and warrant the exploration of N-acetylcysteine in therevention of breast cancer.

hemis

A

GD

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

D. Venugopal et al. / Journal of Steroid Bioc

cknowledgements

We are grateful for the technical assistance from Drs. N.W.aikwad and M. Saeed in analyzing estrogen metabolites andNA adducts.

eferences

[1] G.T. Beatson, On the treatment of inoperable cases of carcinoma of themamma: suggestions for a new treatment with illustrative cases, Lancet 2(1896) 104–107.

[2] B.E. Henderson, H.S. Feigelson, Hormonal carcinogenesis, Carcinogenesis21 (2000) 427–433.

[3] J. Russo, M. Hasan Lareef, G. Balogh, S. Guo, I.H. Russo, Estrogen andits metabolites are carcinogenic agents in human breast epithelial cells, J.Steroid Biochem. Mol. Biol. 87 (2003) 1–25.

[4] E. Cavalieri, D. Chakravarti, J. Guttenplan, E. Hart, J. Ingle, R. Jankowiak,P. Muti, E. Rogan, J. Russo, R. Santen, T. Sutter, Catechol estrogen quinonesas initiators of breast and other human cancers: implications for biomark-ers of susceptibility and cancer prevention, Biochim. Biophys. Acta 1766(2006) 63–78.

[5] E.G. Rogan, A.F. Badawi, P.D. Devanesan, J.L. Meza, J.A. Edney, W.W.West, S.M. Higginbotham, E.L. Cavalieri, Relative imbalances in estrogenmetabolism and conjugation in breast tissue of women with carcinoma:potential biomarkers of susceptibility to cancer, Carcinogenesis 24 (2003)697–702.

[6] S. Singh, D. Chakravarti, J.A. Edney, R.R. Hollins, P.J. Johnson, W.W.West, S.M. Higginbotham, E.L. Cavalieri, E.G. Rogan, Relative imbalancesin the expression of estrogen-metabolizing enzymes in the breast tissue ofwomen with breast carcinoma, Oncol. Rep. 14 (2005) 1091–1096.

[7] S. Singh, M. Zahid, N. Gaikwad, D. Chakravarti, E. Cavalieri, E. Rogan,The effect of NAD(P)H:quinone oxidoreductase 1 polymorphisms on estro-gen metabolism in relation to initiation of breast cancer, Proc. Am. Assoc.Cancer Res. 47 (2006) 835.

[8] Y. Markushin, W. Zhong, E.L. Cavalieri, E.G. Rogan, G.J. Small, E.S.Yeung, R. Jankowiak, Spectral characterization of catechol estrogenquinone (CEQ)-derived DNA adducts and their identification in humanbreast tissue extract, Chem. Res. Toxicol. 16 (2003) 1107–1117.

[9] D. Chakravarti, P.C. Mailander, K.-M. Li, S. Higginbotham, H.L. Zhang,M.L. Gross, J.L. Meza, E.L. Cavalieri, E.G. Rogan, Evidence that a burstof DNA depurination in SENCAR mouse skin induces error-prone repairand forms mutations in the H-ras gene, Oncogene 20 (2001) 7945–7953.

10] Z. Zhao, W. Kosinska, M. Khmelnitsky, E.L. Cavalieri, E.G. Rogan,D. Chakravarti, P.G. Sacks, J.B. Guttenplan, Mutagenic activity of 4-hydroxyestradiol, but not 2-hydroxyestradiol, in BB rat2 embryonic cells,and the mutational spectrum of 4-hydroxyestradiol, Chem. Res. Toxicol.19 (2006) 475–479.

11] S.V. Fernandez, I.H. Russo, J. Russo, Estradiol and its metabolites 4-hydroxyestradiol and 2-hydroxyestradiol induce mutations in human breastepithelial cells, Int. J. Cancer 118 (2006) 1862–1868.

12] P.C. Mailander, J.L. Meza, S. Higginbotham, D. Chakravarti, Induction ofA.T to G.C mutations by erroneous repair of depurinated DNA followingestrogen treatment of the mammary gland of ACI rats, J. Steroid Biochem.Mol. Biol. 101 (2006) 204–215.

13] J. Russo, Q. Tahin, M.H. Lareef, Y.F. Hu, I.H. Russo, Neoplastic trans-formation of human breast epithelial cells by estrogens and chemicalcarcinogens, Environ. Mol. Mutagen. 39 (2002) 254–263.

14] J.J. Li, S.A. Li, Estrogen carcinogenesis in hamster tissues: a critical review,Endocr. Rev. 11 (1990) 524–531.

15] J.D. Yager, Endogenous estrogens as carcinogens through metabolic acti-

vation, J. Natl. Cancer Inst Monogr. 27 (2000) 67–73.

16] J.G. Liehr, Is estradiol a genotoxic mutagenic carcinogen? Endocr. Rev. 21(2000) 40–54.

17] R.R. Newbold, J.G. Liehr, Induction of uterine adenocarcinoma in CD-1mice by catechol estrogens, Cancer Res. 60 (2000) 235–237.

[

try & Molecular Biology 109 (2008) 22–30 29

18] M. Zahid, N.W. Gaikwad, E.G. Rogan, E.L. Cavalieri, Inhibition of depuri-nating estrogen-DNA adduct formation by natural compounds, Chem. Res.Toxicol. 20 (2007) 1947–1953.

19] L. Bonanomi, A. Gazzaniga, Toxicological, pharmacokinetic and metabolicstudies on acetylcysteine, Eur. J. Respir. Dis. Suppl. 111 (1980) 45–51.

20] W.R. Webb, New Mucolytic Agents for Sputum Liquefaction, Postgrad.Med. 36 (1964) 449–453.

21] R.J. Flanagan, The role of acetylcysteine in clinical toxicology, Med. Tox-icol. 2 (1987) 93–104.

22] C.J. Doelman, A. Bast, Oxygen radicals in lung pathology, Free Radic.Biol. Med. 9 (1990) 381–400.

23] S. De Flora, C.F. Cesarone, R.M. Balansky, A. Albini, F. D’Agostini,C. Bennicelli, M. Bagnasco, A. Camoirano, L. Scatolini, A. Rovida, etal., Chemopreventive properties and mechanisms of N-acetylcysteine. Theexperimental background, J. Cell Biochem. Suppl. 22 (1995) 33–41.

24] A.B. Tattersall, K.M. Bridgman, A. Huitson, Acetylcysteine (Fabrol) inchronic bronchitis–a study in general practice, J. Int. Med. Res. 11 (1983)279–284.

25] E.M. Grandjean, P. Berthet, R. Ruffmann, P. Leuenberger, Efficacy of orallong-term N-acetylcysteine in chronic bronchopulmonary disease: a meta-analysis of published double-blind, placebo-controlled clinical trials, Clin.Ther. 22 (2000) 209–221.

26] I. Goldenberg, M. Shechter, S. Matetzky, M. Jonas, M. Adam, H. Pres,D. Elian, O. Agranat, E. Schwammenthal, V. Guetta, Oral acetylcysteineas an adjunct to saline hydration for the prevention of contrast-inducednephropathy following coronary angiography. A randomized controlledtrial and review of the current literature, Eur. Heart J. 25 (2004) 212–218.

27] S. De Flora, C. Bennicelli, P. Zanacchi, A. Camoirano, A. Morelli, A. DeFlora, In vitro effects of N-acetylcysteine on the mutagenicity of direct-acting compounds and procarcinogens, Carcinogenesis 5 (1984) 505–510.

28] S. De Flora, C. Bennicelli, P. Zanacchi, F. D’Agostini, A. Camoirano, Muta-genicity of active oxygen species in bacteria and its enzymatic or chemicalinhibition, Mutat. Res. 214 (1989) 153–158.

29] S. De Flora, A. Izzotti, F. D’Agostini, C.F. Cesarone, Antioxidant activityand other mechanisms of thiols involved in chemoprevention of mutationand cancer, Am. J. Med. 91 (1991) 122S–130S.

30] G.J. Kelloff, J.A. Crowell, C.W. Boone, V.E. Steele, R.A. Lubet, P. Green-wald, D.S. Alberts, J.M. Covey, L.A. Doody, G.G. Knapp, et al., Clinicaldevelopment plan: N-Acetyl-l-cysteine, J. Cell Biochem. Suppl. 20 (1994)63–73.

31] S. De Flora, A. Izzotti, F. D’Agostini, R. Balansky, Mechanisms ofN-acetylcysteine in the prevention of DNA damage and cancer, with spe-cial reference to smoking-related end-points, Carcinogenesis 22 (2001)999–1013.

32] R.M. Balansky, G. Ganchev, F. D’Agostini, S. De Flora, Effects of N-acetylcysteine in an esophageal carcinogenesis model in rats treated withdiethylnitrosamine and diethyldithiocarbamate, Int. J. Cancer 98 (2002)493–497.

33] S.M. Lippman, M.R. Spitz, Lung cancer chemoprevention: an integratedapproach, J. Clin. Oncol. 19 (2001) 74S–82S.

34] D. Chakravarti, M. Zahid, M. Backora, E.M. Myers, D.D. Weisenburger,E.L. Cavalieri, E.G. Rogan, S.S. Joshi, Ortho-quinones of benzene andestrogens induce hyperproliferation of human peripheral blood mononu-clear cells, Leuk. Lymphoma 47 (2006) 2635–2644.

35] M.K. Sgagias, K.U. Wagner, B. Hamik, S. Stoeger, R. Spieker, L.J. Huber,L.A. Chodosh, K.H. Cowan, Brca1-deficient murine mammary epithelialcells have increased sensitivity to CDDP and MMS, Cell Cycle 3 (2004)1451–1456.

36] D. Chakravarti, P. Mailander, J. Franzen, S. Higginbotham, E.L. Cava-lieri, E.G. Rogan, Detection of dibenzo[a,l]pyrene-induced H-ras codon61 mutant genes in preneoplastic SENCAR mouse skin using a new PCR-RFLP method, Oncogene 16 (1998) 3203–3210.

37] D. Chakravarti, P.C. Mailander, E.L. Cavalieri, E.G. Rogan, Evidence thaterror-prone DNA repair converts dibenzo[a,l]pyrene-induced depurinatinglesions into mutations: formation, clonal proliferation and regression ofinitiated cells carrying H-ras oncogene mutations in early preneoplasia,Mutat. Res. 456 (2000) 17–32.

3 hemis

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[69] F. Lu, M. Zahid, M. Saeed, E.L. Cavalieri, E.G. Rogan, Estrogen

0 D. Venugopal et al. / Journal of Steroid Bioc

38] E. Hoffer, Y. Baum, A. Tabak, U. Taitelman, N-Acetylcysteine increases theglutathione content and protects rat alveolar type II cells against paraquat-induced cytotoxicity, Toxicol. Lett. 84 (1996) 7–12.

39] S. Kanno, M. Ishikawa, M. Takayanagi, Y. Takayanagi, K. Sasaki, Exposureto hydrogen peroxide induces cell death via apoptosis in primary culturedmouse hepatocytes, Biol. Pharm. Bull. 22 (1999) 1296–1300.

40] X. Zhou, A. Zhao, G. Goping, P. Hirszel, Gliotoxin-induced cytotoxicityproceeds via apoptosis and is mediated by caspases and reactive oxygenspecies in LLC-PK1 cells, Toxicol. Sci. 54 (2000) 194–202.

41] A.M. Samuni, E.Y. Chuang, M.C. Krishna, W. Stein, W. DeGraff, A. Russo,J.B. Mitchell, Semiquinone radical intermediate in catecholic estrogen-mediated cytotoxicity and mutagenesis: chemoprevention strategies withantioxidants, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 5390–5395.

42] H. Terasaka, Y. Kadoma, H. Sakagami, S. Fujisawa, Cytotoxicity andapoptosis-inducing activity of bisphenol A and hydroquinone in HL-60cells, Anticancer Res. 25 (2005) 2241–2247.

43] S.Y. Chan, A.L. Hilchie, M.G. Brown, R. Anderson, D.W. Hoskin, Apop-tosis induced by intracellular ceramide accumulation in MDA-MB-435breast carcinoma cells is dependent on the generation of reactive oxygenspecies, Exp. Mol. Pathol. 82 (2007) 1–11.

44] O.I. Aruoma, B. Halliwell, B.M. Hoey, J. Butler, The antioxidant actionof N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radi-cal, superoxide, and hypochlorous acid, Free Radic. Biol. Med. 6 (1989)593–597.

45] Y. Hiraku, S. Kawanishi, Oxidative DNA damage and apoptosis inducedby benzene metabolites, Cancer Res. 56 (1996) 5172–5178.

46] E.L. Cavalieri, E.G. Rogan, D. Chakravarti, Initiation of cancer and otherdiseases by catechol ortho-quinones: a unifying mechanism, Cell Mol. LifeSci. 59 (2002) 665–681.

47] L.M. Nutter, B. Zhou, E.E. Sierra, Y.Y. Wu, M.M. Rummel, P. Gutierrez, Y.Abul-Hajj, Cellular biochemical determinants modulating the metabolismof estrone 3,4-quinone, Chem. Res. Toxicol. 7 (1994) 609–613.

48] Y.J. Hurh, Z.H. Chen, H.K. Na, S.Y. Han, Y.J. Surh, 2-Hydroxyestradiolinduces oxidative DNA damage and apoptosis in human mammary epithe-lial cells, J. Toxicol. Environ. Health A 67 (2004) 1939–1953.

49] B. Kalyanaraman, W. Korytowski, B. Pilas, T. Sarna, E.J. Land, T.G.Truscott, Reaction between ortho-semiquinones and oxygen: pulse radiol-ysis, electron spin resonance, and oxygen uptake studies, Arch. Biochem.Biophys. 266 (1988) 277–284.

50] Z.H. Chen, H.K. Na, Y.J. Hurh, Y.J. Surh, 4-Hydroxyestradiol inducesoxidative stress and apoptosis in human mammary epithelial cells: possibleprotection by NF-kappaB and ERK/MAPK, Toxicol. Appl. Pharmacol. 208(2005) 46–56.

51] E. Cavalieri, Mini symposium on endogenous carcinogens: the catecholestrogen pathway. An introduction, Polycyclic Aromatic Compd. 6 (1994)223–228.

52] J.A. Mobley, A.S. Bhat, R.W. Brueggemeier, Measurement of oxidativeDNA damage by catechol estrogens and analogues in vitro, Chem. Res.Toxicol. 12 (1999) 270–277.

53] X. Han, J.G. Liehr, 8-Hydroxylation of guanine bases in kidney and liver

DNA of hamsters treated with estradiol: role of free radicals in estrogen-induced carcinogenesis, Cancer Res. 54 (1994) 5515–5517.

54] P.H. Lin, J. Nakamura, S. Yamaguchi, S. Asakura, J.A. Swenberg, Alde-hydic DNA lesions induced by catechol estrogens in calf thymus DNA,Carcinogenesis 24 (2003) 1133–1141.

try & Molecular Biology 109 (2008) 22–30

55] Y. Hiraku, N. Yamashita, M. Nishiguchi, S. Kawanishi, Catechol estrogensinduce oxidative DNA damage and estradiol enhances cell proliferation,Int. J. Cancer 92 (2001) 333–337.

56] J.G. Liehr, W.F. Fang, D.A. Sirbasku, A. Ari-Ulubelen, Carcinogenicityof catechol estrogens in Syrian hamsters, J. Steroid Biochem. 24 (1986)353–356.

57] J.J. Li, S.A. Li, Estrogen carcinogenesis in Syrian hamster tissue: role ofmetabolism, Fed. Proc. 46 (1987) 1858–1863.

58] J.G. Liehr, M.J. Ricci, 4-Hydroxylation of estrogens as marker of humanmammary tumors, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 3294–3296.

59] M. Zahid, E. Kohli, M. Saeed, E. Rogan, E. Cavalieri, The greater reactivityof estradiol-3,4-quinone vs estradiol-2,3-quinone with DNA in the for-mation of depurinating adducts: implications for tumor-initiating activity,Chem. Res. Toxicol. 19 (2006) 164–172.

60] K. Tabakovic, W.B. Gleason, W.H. Ojala, Y.J. Abul-Hajj, Oxidative trans-formation of 2-hydroxyestrone. Stability and reactivity of 2,3-estronequinone and its relationship to estrogen carcinogenicity, Chem. Res. Toxi-col. 9 (1996) 860–865.

61] L.M. Nutter, Y.Y. Wu, E.O. Ngo, E.E. Sierra, P.L. Gutierrez, Y.J. Abul-Hajj, An o-quinone form of estrogen produces free radicals in human breastcancer cells: correlation with DNA damage, Chem. Res. Toxicol. 7 (1994)23–28.

62] N.W. Gaikwad, E. Cavalieri, E. Rogan, NQO1-catalyzed reduction ofestradiol-3,4-quinone. Implications for tumor initiation by estrogens, Proc.Am. Assoc. Cancer Res. 47 (2006) 445.

63] L.M. Nutter, E.O. Ngo, Y.J. Abul-Hajj, Characterization of DNA damageinduced by 3,4-estrone-o-quinone in human cells, J. Biol. Chem. 266 (1991)16380–16386.

64] K. Cao, D.E. Stack, R. Ramanathan, M.L. Gross, E.G. Rogan, E.L. Cava-lieri, Synthesis and structure elucidation of estrogen quinones conjugatedwith cysteine, N-acetylcysteine, and glutathione, Chem. Res. Toxicol. 11(1998) 909–916.

65] M. Nakagomi, E. Suzuki, Quantitation of catechol estrogens and their N-acetylcysteine conjugates in urine of rats and hamsters, Chem. Res. Toxicol.13 (2000) 1208–1213.

66] P. Devanesan, R.J. Santen, W.P. Bocchinfuso, K.S. Korach, E.G. Rogan,E. Cavalieri, Catechol estrogen metabolites and conjugates in mammarytumors and hyperplastic tissue from estrogen receptor-alpha knock-out(ERKO)/Wnt-1 mice: implications for initiation of mammary tumors, Car-cinogenesis 22 (2001) 1573–1576.

67] R. Todorovic, P. Devanesan, S. Higginbotham, J. Zhao, M.L.Gross, E.G. Rogan, E.L. Cavalieri, Analysis of potential biomarkersof estrogen-initiated cancer in the urine of Syrian golden ham-sters treated with 4-hydroxyestradiol, Carcinogenesis 22 (2001) 905–911.

68] E. Boyland, L.F. Chasseaud, The role of glutathione and glutathione S-transferases in mercapturic acid biosynthesis, Adv. Enzymol. Relat AreasMol. Biol. 32 (1969) 173–219.

metabolism and formation of estrogen-DNA adducts in estradiol-treatedMCF-10F cells. The effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin induc-tion and catechol-O-methyltransferase inhibition, J. Steroid Biochem. Mol.Biol. 105 (2007) 150–158.