e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860

Available online at www.sciencedirect.com

ScienceDirect

j o ur nal ho me pa ge: www.elsev ier .com/ locate /e tap

Role of rat cytochromes P450 in the oxidation of17�-ethinylestradiol

Lucie Borek-Dohalská, Petra Valásková, Vera Cerná, Marie Stiborová ∗

Department of Biochemistry, Faculty of Science, Charles University, Albertov 2030, 128 40 Prague 2, Czech Republic

a r t i c l e i n f o

Article history:

Received 28 June 2014

Received in revised form

4 October 2014

Accepted 10 October 2014

Available online 18 October 2014

Keywords:

17�-Ethinylestradiol

Endocrine disruptor

a b s t r a c t

17�-Ethinylestradiol (EE2) is an endocrine disruptor (ED) used as an ingredient of oral

contraceptives. Rat hepatic microsomes metabolize EE2 to three products; two of them

are hydroxylated EE2 derivatives. Of the hydroxylation reactions, 2-hydroxylation, is the

major reaction. Cytochrome P450 (CYP) plays a major role in EE2 hydroxylation. To resolve

which rat CYPs are responsible for EE2 oxidation, three approaches were used: induc-

tion of specific CYPs, selective inhibition of CYPs, and recombinant rat CYPs. The results

demonstrate that EE2 is hydroxylated by several rat CYPs, among them CYP2C6 and 2C11

are most efficient in 2-hydroxy-EE2 formation, while CYP2A and 3A catalyze EE2 hydrox-

ylation to the second product. EE2 is also an inhibitor of CYP2C- and CYP3A-catalyzed

hydroxylation of endogenous EDs progesterone and testosterone. EE2 acts as a reversible

Oxidative metabolism

Cytochrome P450

inhibitor of CYP3A-mediated progesterone 6�-hydroxylation and inactivates CYP3A- and

CYP2C-catalyzed testosterone 6�-hydroxylation and progesterone 21- or 16�-hydroxylation,

respectively, in a mechanism-based manner.

© 2014 Elsevier B.V. All rights reserved.

16� position of the steroid nucleus (Fig. 1) (Rogers et al., 1987;

1. Introduction

Synthetic estrogen 17�-ethinylestradiol (EE2, Fig. 1A) belongsto the group of chemicals assigned as endocrine disruptivecompounds, which disrupt endocrine system in wild life. Thissynthetic estrogen is widely used as the major componentin oral contraceptives (OCs) (Bolt, 1979). Incorporation of anacetylenic moiety into the estradiol molecule (Fig. 1B) resultedin an increase in the oral availability of the drug. The oral

contraceptives are used in young healthy women in a hith-erto wide range. The general toxicological aspects of OCs andtheir impact on the environment have been considered and

Abbreviations: �-NF, �-naphthoflavone; CO, carbon monoxide;

ethinylestradiol; EtOH, ethanol; HPLC, high performance liquid chropregnenolon 16�-carbonitrile; UPLC, ultra pressure liquid chromatogra

∗ Corresponding author. Tel.: +420 22195121285; fax: +420 221951283.E-mail address: stiborov@natur.cuni.cz (M. Stiborová).

http://dx.doi.org/10.1016/j.etap.2014.10.0041382-6689/© 2014 Elsevier B.V. All rights reserved.

investigated carefully (Cajthaml et al., 2009; Kresinová et al.,2012).

The level of EE2 exposition is crucial for its biologicaleffects. Besides the ingested amounts of EE2, metabolism ofthis compound dictates its effective concentration, therebymodulating also the clinical consequences of exposure. There-fore, characterization of metabolism of EE2 is of generalimportance. The oxidative metabolism of EE2 has been stud-ied extensively. It undergoes hydroxylation at the 2, 4, 6, and

CYP, cytochrome P450; DMSO, dimethyl sulfoxide; EE2, 17�-matography; OCs, oral contraceptives; PB, phenobarbital; PCN,phy.

Stanczyk et al., 2013; Zhang et al., 2007). Of the hydroxyl-ation reactions, 2-hydroxylation is clearly the major oxidativereaction (Ball et al., 1990). Therefore, alterations in the rate

e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860 853

l (A)

oc2vItcgoBEoq1G

ieamito2mlCoCh

um(btharo2nnrsmr

Fig. 1 – Structure of 17�-ethinylestradio

f 2-hydroxylation can have major effects on the pharma-okinetics and effectiveness of EE2 as a contraceptive. A-hydroxy-EE2 derivative can be subsequently methylated inivo to give 2-methoxyethinylestradiol (Rogers et al., 1987).n addition, EE2 oxidation represents only small fraction ofhe total EE2 metabolism, which can occur by way of glu-uronidation at 17 and 3 positions by uridine diphosphatelucuronosyltransferase 1A1 (Ebner et al., 1993), methylation,r sulfation at the 3 position by sulfotransferase 1E1 (Forbes-amforth and Coughtrie, 1994). Other possible pathways ofE2 metabolism are (i) D-homoannulation by rearrangingne ethinyl group carbon into the D ring of EE2 and subse-uent removal of the second carbon as CO2 (Schmid et al.,983), and (ii) removal of the ethinyl group (Helton andoldzieher, 1977).

Because hydroxylation of EE2 represents a key step ints overall metabolism, understanding which hydroxylationnzymes are involved in these reactions is important in thessessment of susceptibility to this endocrine disruptor. Aicrosomal mixed-function-oxidase enzyme system contain-

ng cytochrome P450 (CYP) as a terminal oxidase is well knowno be the major enzymatic system catalyzing C-hydroxylationf a variety of substrates [for a review see (Guengerich, 2007,008)]. The CYP enzymes were clearly found to play also aajor role in hydroxylation of EE2 in human and rat. Neverthe-

ess, the discrepancies in identification of individual humanYP enzymes were found in several former studies. More-ver, there is still not properly resolved whether orthologousYP enzymes of human and animal models catalyze EE2ydroxylation.

In early studies with human CYP enzymatic systems,tilizing antibody inhibition, CYP3A4 was reported as theajor enzyme involved in oxidation of EE2 in human liver

Guengerich, 1988). More recently, EE2 has been shown toe oxidized also by other CYPs. Wang et al. (2004) describedhat the CYP enzymes predominantly contributing to the 2-ydroxylation of EE2 in human liver microsomes are CYP2C9nd 3A4, whereas CYP2C8, 2C19, and 1A2 contribute to thiseaction to a lesser extent. In addition, the involvement ofther CYP enzymes tested by these authors, CYP2A6, 2B6,D6 and 2E1, was also detectable, but their contribution wasegligible. These authors also showed that human recombi-ant CYP1A1 exhibited higher intrinsic catalytic activity than

ecombinant CYP3A4 and/or 2C9. Furthermore, a very recenttudy of Winkler et al. (2014) showed that CYP3A4 plays ainor role in EE2 metabolism in human in vivo. Using the

at hepatic enzymatic system Ball et al. (1990), who used

and endogenous hormone estradiol (B).

immunoinhibitory CYP antibodies in their experiments, foundthat CYP2C and 2E could be involved in the 2-hydroxylationreaction of EE2. Moreover, in contrast to results found by Wanget al. (2004) who showed a role of human CYP1A1 in EE2metabolism, Ball et al. (1990) postulated that rat CYP1A1 doesnot play an important role in this process.

The inhibitors of several CYP enzymes metabolizing EE2and several other drugs were found to elevate EE2 plasmaconcentrations and increase risks of vascular disease andhypertension (Zhang et al., 2007). Other studies in vitrohave shown that EE2 itself act also as an inhibitor of sev-eral CYP enzymes. Because EE2 influences the activities ofthese enzymes, it can contribute considerably to the drug-drug interactions. EE2 was found to be a mechanism-basedinhibitor of CYP3A4, 2B1 and 2B6 enzymes (Guengerich, 1988;Kent et al., 2002, 2008; Lin et al., 2002). Whereas the pri-mary loss in activity of CYP3A4 was found to be caused byheme modification, CYP2B1 and 2B6 are inactivated due tothe formation of a covalent adduct to the apo-protein. TheEE2-mediated inactivation of both rat and human liver micro-somal CYPs was postulated to occur during oxidation of theacetylenic moiety of EE2 (Guengerich, 1990). Indeed, mecha-nistic studies suggest that oxidation of the acetylenic moietyof EE2 by CYP enzymes leads to the formation of an unsta-ble and reactive acetylene-oxide, a strong nucleophile thatcan covalently bind to CYP. The EE2-mediated inactivationof liver microsomal CYP was found in both rat and human(Guengerich, 1990). However, the toxicological relevance ofthe inactivation of CYP enzymes has not been establishedas yet. It is only known that apoprotein modification of cel-lular macromolecules by reactive intermediates can lead todifferent disease states ranging from autoimmune diseasesto neurotoxicity (LoPachin and DeCaprio, 2005). In addition tomechanism-based inhibition, EE2 has been found as a weakreversible inhibitor of CYP2C19 (Rodrigues and Lu, 2004) andCYP2C8 (Walsky et al., 2005).

Since the discrepancies in data identifying CYP enzymesthat participate in oxidation of EE2 exist, the aim of the presentstudy was to extend our knowledge on this issue. The rat wasused as a model in the study. Even though this animal was con-sidered as a suitable model to mimic the fate of estradiol andits derivatives in human, information on EE2 metabolism andCYP enzymes participating in this process is scarce (Ball et al.,

1990; Klinger et al., 2002; Laurenzana et al., 2002). The objec-tive of the present study was to determine a role of individualrat hepatic CYP enzymes in EE2 oxidation. Because there areseveral reports suggesting EE2-containing oral contraceptive

d p h

854 e n v i r o n m e n t a l t o x i c o l o g y a n

formulations as perpetrators of pharmacokinetic drug interac-tions (Back and Orme, 1990; Dickinson et al., 2001; Zhang et al.,2007), here we also examined whether EE2 and its metabolitesmight influence metabolism of two endogenous endocrinedisruptors, hormones progesterone and testosterone.

2. Materials and methods

2.1. Chemicals and enzymes

17-Ethinylestradiol, glucose-6-phosphate, ellipticine, NADP+,NADPH, �-naphthoflavone (-NF), phenacetine and dithiothre-itol were obtained from Sigma Chemical Co. (St. Louis, MO,USA); Sudan I from BDH (Poole, UK). Progesterone, testos-terone, 16-hydroxyprogesterone and 6-hydroxytestosteronewere purchased from Merck (Darmstadt, Germany). Glucose-6-phosphate dehydrogenase was from Serva (Heidelberg,Germany). Bicinchoninic acid was from Pierce (Rockford,IL, USA). All chemicals were of a reagent grade or better.SupersomesTM, microsomal samples isolated from insect cellstransfected with Baculovirus constructs, containing cDNAof CYPs, NADPH:CYP reductase with or without cytochromeb5, expressing all these enzymes were from Gentest corp.(Woburn, MA).

2.2. Preparation of microsomes

All animal experiments were conducted in accordance withthe Regulations for the Care and Use of Laboratory Animals(311/1997, Ministry of Agriculture, Czech Republic), which is incompliance with the Declaration of Helsinki. Male Wistar rats(150 g, AnLab, Czech Republic) were housed in groups of 3 inwire cages at 22 ◦C with a 12 h light/dark period and ad libitumdiet (ST-1 diet from Velaz, Czech Republic) and water access.Microsomes were isolated from livers of untreated rats andthose of pre-treated with Sudan I, phenobarbital (PB), ethanol(EtOH) and pregnenolon 16-carbonitrile (PCN) by proceduresas described (Stiborová et al., 2002). Total CYP content wasmeasured based on a complex of reduced CYP with carbonmonoxide (CO) (Omura and Sato, 1964).

2.3. 17-Ethinylestradiol oxidation

Incubation mixtures contained in a final volume of 0.5 ml:0.1 M potassium phosphate buffer, pH 7.4, 50 �M EE2 [2.5 �lof stock dimethyl sulfoxide (DMSO) solution per incuba-tion], 50 �l of NADPH-generating system (10 mM MgCl2, 10 mMd-glucose 6-phosphate, 1 mM NADP+, 1 unit/ml d-glucose6-phosphate dehydrogenase) and 0.5 �M microsomal CYPof untreated rats and/or those of pre-treated with above-mentioned inducers. The mixtures were incubated for 15 minat 37 ◦C in a shaking incubator. In some cases, SupersomesTM

(see above) containing rat recombinant CYPs (50 pmol),NADPH:CYP reductase (400.5 nmol cytochrome c reduced permin) were also used instead of microsomes. Incubation mix-

tures contained in a final volume of 0.25 ml: 0.1 M potassiumphosphate buffer, pH 7.4, 50 �M EE2 (2.5 �l of stock DMSO solu-tion per incubation), 1 mM NADPH and 50 pmol recombinantCYP. The mixtures were incubated for 30 min at 37 ◦C in a

a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860

shaking incubator (EE2 oxidation was linear up to 30 min ofincubation). The reactions were terminated by addition of 1 mlof ethyl acetate. Then, phenacetin (5 �l of 1 mM stock solu-tion) was added as an internal standard. The metabolites wereextracted and the extracts were evaporated to dryness. Theresidues were dissolved in the mobile phase for high perfor-mance liquid chromatography (HPLC) (see below).

In the case of inhibition studies, 50 �M specific inhibitorof CYP was pre-incubated 10 min with microsomes in thepresence or absence of NADPH before addition of EE2. Thefollowing procedure was the same as mentioned above.

2.4. Progesterone hydroxylation and testosterone6ˇ-hydroxylation

The incubation mixtures for measuring the progesteroneand/or testosterone metabolism contained in a final volumeof 0.5 ml: 0.1 M potassium phosphate buffer, pH 7.4, 50 �M pro-gesterone and/or testosterone (2 �l of stock methanol solutionper incubation), 10 mM MgCl2, 10 mM d-glucose 6-phosphate,1 mM NADP+, 1 unit/ml d-glucose 6-phosphate dehydrogenaseand 0.5 �M CYP (for a detail see, Borek-Dohalská et al., 2001).

The mixtures were incubated for 15 min, at 37 ◦C in a shak-ing incubator. The reaction was terminated by addition of0.1 ml of 1 M aqueous Na2CO3 containing 2 M NaCl. Then,phenacetin (5 �l of 1 mM stock solution) was added as aninternal standard. The metabolites were extracted with 2 mlof CH2Cl2 and the extracts were evaporated to dryness. Theresidues were dissolved in the mobile phase for HPLC (seebelow). In order to determine the effect of EE2 on CYP3A sub-strate specific activity, rat microsomes of untreated rats and/orthose of pre-treated with above mentioned inducers werepreincubated (10 min at 37 ◦C) with 50 �M EE2 in the presenceor absence of NADPH.

2.5. HPLC and (ultra pressure liquid chromatography)UPLC conditions and mass spectrometry

EE2 and its metabolites were separated on a column of Nucle-osil (C18) (4.6 mm × 250 mm, 5 �m, Macherey-Nagel, Germany)with optimized gradient elution. The mobile phase consistedof solvent A (20% acetonitrile) and solvent B (80% acetoni-trile). A linear gradient was constructed and the flow rateswere as follows: 10–50% B, 0–25 min and 0.8 ml/min; 50–100% B,25–30 min and 1 ml/min; 100–10% B, 30–37 min and 0.8 ml/min,10% B, 37–40 min and 0.8 ml/min. Detection wavelength andtemperature was 280 nm and 35 ◦C, respectively.

Separation on UPLC (Waters Acquity UPLC system) withmass spectroscopy (Waters LCT Premier XE, Manchester UK)was used for identification of EE2 metabolites. Metaboliteswere separated on the Acquity UPLC BEH (C18) column(50 mm × 2.1 mm, 1.7 �m) with gradient elution [0.4 ml/min−1;(A) HCOOH:H2O 0.1:99.9 (v/v), (B) ACN: 0 min 95% (A), 12 min50% (A), 15 min 30% (A) and 17 min 0% (A)]. The flow rate was0.4 ml/min. The analytes were detected at positive and neg-ative ion modes. The voltage at the input cone was 40 V; the

voltage at the spray capillary was set at ±2500 V. The sourceblock was maintained at 120 ◦C and temperature of desolva-tion gas–nitrogen at 350 ◦C. The other parameters were asfollows: desolvation gas flow, 800 L h−1; gas flow of the inlet

e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r

Phenacetine

X-OH E E2

2-OH EE2

EE2

-2.0

5.0

10.0

15.0

20.0

25.0

29.8

0.1 5.0 10.0 15.0 20.0 27.8

mAU

min

WVL:280 nm

Fig. 2 – HPLC chromatogram for 17�-ethinylestradiolmetabolism by rat microsomes. HPLC conditions are givenin Section 2. The peaks with retention time of 9.1 and22.7 min belong to internal standard phenacetine andparental compound EE2, respectively. Three metaboliteswere formed with retention times of 10.3, 18.5 and21.8 min, which were assigned as the X-hydroxy-EE2 (X-OHEE2), 2-hydroxy-EE2 (2-OH EE2) and dehydrogenatedmetabolite of EE2. The peak with retention time of 13 minbelongs to background peak presents also in the controls

csiim

aMaam

3

3

MirdtOpitml

rat liver microsomes. Among them, the inhibitor of CYP3A,

amples.

one, 50 L h−1; W mode; scan time, 0.1 s; time delay betweencans, 0.01 s. For metabolite identification, full scan analysisn the mass range of 80–800 amu was performed and producton scan analyses were used to investigate EE2 metabolites [for

ore details, see (Kresinová et al., 2012)].Progesterone, testosterone and their metabolites were sep-

rated on a column of Nucleosil (C18) (4.6 mm × 25 mm, 5 �m,acherey-Nagel, Germany). The flow rates, mobile phases

nd detection wavelengths for progesterone and testosteronessays were 0.7 ml/min and 0.5 ml/min, respectively, 75:25ethanol/H2O (v/v), and 254 nm (Borek-Dohalská et al., 2001).

. Results and discussion

.1. Metabolism of EE2 by rat hepatic microsomes

etabolism of EE2 by rat microsomal CYP enzymes wasnvestigated in ex-vivo incubations containing this compound,at hepatic microsomes and NADPH, a cofactor for CYP-ependent enzymatic system. Under the used conditions,hree EE2 metabolite products were separated by HPLC (Fig. 2).n the basis of mass spectrometry and their chromatographicroperties, the structures of the metabolites were character-

zed. In the positive-ion electrospray mass-spectrum, one of

hree metabolites eluting at 10.3 min showed the protonated

olecule at m/z 295.17 [M−H2O+H]+, indicating the molecu-ar mass of hydroxylated derivatives of EE2 (X-hydroxy-EE2).

m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860 855

Even though the structure of this hydroxylated EE2 metabo-lite has not been identified as yet, one can speculate fromits chromatographic properties on HPLC that it might be 4-hydroxy-EE2 (Wang et al., 2004). The second metabolite elutingat 18.5 min that was formed as the major EE2 metabolite exhib-ited the same protonated molecule at m/z 295.17 [M−H2O+H]+

and m/z 313.4 [M+H]+. Based on this finding and the resultsfrom the tandem liquid chromatography/mass spectrome-try (Kresinová et al., 2012), this metabolite was identified as2-hydroxyethinylestradiol (2-hydroxy-EE2). The amount of 2-hydroxy-EE2 formed in these microsomes was more than2.4-fold higher than that of X-hydroxy-EE2 (Figs. 2 and 3). Thenext EE2 metabolite eluting at 21.8 min, which was formedat low amounts, had m/z 277.16 [M−H2O+H]+ indicating tobe a dehydrogenated metabolite of EE2. Because of its lowamounts, its formation was not investigated in further exper-iments.

3.2. Involvement of rat CYP enzymes in oxidation ofEE2

In order to resolve which rat CYP enzymes are responsiblefor oxidation of EE2, three experimental approaches wereemployed: (i) induction of specific CYPs, (ii) selective inhibitionof CYPs, and (iii) heterologous expression systems containingrecombinant rat CYPs (SupersomesTM).

Besides the hepatic microsomes of control (untreated) rats,liver microsomes of rats pretreated with Sudan I (an inducer ofCYP1A) (Stiborová et al., 2002), phenobarbital (PB, an inducer ofCYP2B and 2C, Rice et al., 1994), ethanol (an inducer of CYP2E1,Wu and Cederbaum, 1993) and pregnenolone-16�-carbonitrile(PCN, an inducer of CYP3A, Surry et al., 2000) also oxidizedEE2. An increase in formation of X-hydroxy-EE2 was found inmicrosomes, in which CYP2B/2C and 3A were induced by phe-nobarbital and PCN, respectively (Fig. 3). Other CYP inducerswere found to decrease formation of this metabolite. Forma-tion of 2-hydroxy-EE2 was increased by pretreatment of ratswith PB and ethanol indicating that CYP2B/2C and 2E1 mightbe involved in 2-hydroxylation of EE2 in rat liver. These resultssuggest that CYP2B, 2C and 3A can participate in EE2 oxidativemetabolism in rat liver microsomes.

The inhibition effects of specific CYP inhibitors were stud-ied using two different protocols. In the first one, the reactionmixtures containing microsomes and a specific inhibitorwere incubated 10 min in the presence of NADPH before theaddition of EE2. In the second protocol, inhibitors were pre-incubated with microsomes for the same time interval, butwithout NADPH. Such experimental approaches are suitableto resolve whether the used inhibitors themselves or theirmetabolites act as inhibitors.

In the case of X-hydroxy-EE2, most tested inhibitors[�-naphthoflavone (�-NF), an inhibitor of CYP1A1/2, adaman-tane, an inhibitor of CYP2B (Hodek et al., 2005), diethyldithio-carbamate, (DDTC), an inhibitor of CYP2E1 and 2A (Guengerichet al., 1991) and ketoconazole, an inhibitor of CYP3A(Borek-Dohalská et al., 2001)] decreased its formation in

ketoconazole, was most effective (Fig. 4). In experimentswhere inhibition of EE2 oxidation by sulfaphenazole (aCYP2C inhibitor, Guengerich et al., 1991) was tested, the

856 e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860

0

25

50

75

100

125

150

175

200

225

PCN MSEtOH MSPB MSSudan MSControl MS

% o

f EE2

met

abol

ite fo

rmat

ion

X-OH EE2

2-OH EE2

*

** *

Fig. 3 – Metabolism of EE2 by hepatic microsomes (MS) of rats pretreated with Sudan I (an inducer of CYP1A1),phenobarbital (PB) (an inducer of CYP2B and 2C), ethanol (EtOH) (an inducer of CYP2E1) and pregnenolone-16�-carbonitrile(PCN) (an inducer of CYP3A). Values are given as means ± SD (n = 3). Area of X-hydroxy-EE2 (X-OH EE2)/area of internalstandard of control microsomes was 0.172 and area of 2-hydroxy-EE2 (2-OH EE2)/area of internal standard of control

tude

microsomes was 0.425. *P < 0.01 different from control MS (S

formation of X-hydroxy-EE2 could not be quantified becausesulfaphenazole metabolites interfered with X-hydroxy-EE2eluting during HPLC analysis. Essentially no changes ininhibitory capacity of �-NF and ketoconazole to decrease for-mation of this metabolite (X-hydroxy-EE2) were detectablebetween the two used experimental protocols. This indi-cated inhibition of the CYP1A- and 3A-catalyzed formationof X-hydroxy-EE2 exclusively by these two compounds. Incontrast, DDTC and adamantane exhibited a different modeof their action; the inhibitory potency of these inhibitorsincreased after their pre-incubation with NADPH and micro-somes (Fig. 4). These findings suggest that metabolites of thesetwo inhibitors, rather than the parent compounds, decreasethe CYP2B-, 2A- and 2E1-mediated formation of X-hydroxy-EE2.

The formation of the second EE2 metabolite, 2-hydroxy-EE2, was inhibited by adamantane and ketoconazole, evenunder the conditions where the reaction mixtures were

Fig. 4 – Inhibition of EE2 oxidation to X-hydroxy-EE2 (X-OH EE2)

presence or absence of NADPH. Values are given as means ± SD

0.172. #Not quantified because sulfaphenazole metabolites interfrom control MS (Student’s t-test). N.D. – not detectable.

nt’s t-test).

incubated in the absence of NADPH before the addition ofEE2 (Fig. 4). On the contrary, under these conditions 1.4-,1.3- and 1.5-fold increase in amounts of this EE2 metabo-lite (2-hydroxy-EE2) was caused by �-NF, a compound knownto stimulate CYP3A activities (Borek-Dohalská et al., 2001;Stiborová et al., 2006), sulfaphenazole and DDTC, respec-tively (Fig. 4). Employing the inhibition protocol, where theinhibitors were pre-incubated with microsomes and NADPHbefore adding EE2, oxidation of this compound to 2-hydroxy-EE2 was also inhibited by adamantane, sulfaphenazole, DDTC,and ketoconazole (Fig. 5). This finding suggests that reactivemetabolites of these compounds, rather than the parentalmolecules, act as inhibitors of 2-hydroxylation of EE2.

All results found in the studies utilizing inducers andinhibitors of CYPs indicate that rat CYP1A, 2A, 2B, 2E1, and

3A can catalyze hydroxylation of EE2 to its X-hydroxylatedmetabolite, while CYP2C, 3A and to a low extent alsoCYP2B and 2E1 mediate 2-hydroxylation of EE2 in rat liver

in rat hepatic microsomes by CYP inhibitors (50 �M) in the(n = 3). Area of X-OH EE2/area of internal standard wasfered with X-OH EE2 eluting during HPLC. *P < 0.01 different

e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860 857

Fig. 5 – Effect of CYP inhibitors (50 �M) on EE2 oxidation to 2-hydroxy-EE2 (2-OH EE2) by microsomes in the presence orabsence of NADPH. Values are given as means ± SD (n = 3). Area of 2-OH EE2/area of internal standard was 0.425. *P < 0.01d ble.

mtbib

3

TytnTooowmCCyii

Fa

ifferent from control MS (Student’s t-test). N.D. – not detecta

icrosomes. Most of these findings are in agreement withhe results found by other authors (Ball et al., 1990). It shoulde noted, however, that the interpretation of the results of

nhibitors is sometimes difficult, because one inhibitor maye more effective with one substrate than another.

.3. Oxidation of EE2 by rat recombinant CYPs

o identify and prove the role of individual rat CYPs in hydrox-lation of EE2 further, we utilized microsomes of Baculovirusransfected insect cells (SupersomesTM) containing recombi-antly expressed rat CYPs and NADPH:CYP reductase (Fig. 6).he recombinant rat CYPs used in the experiments efficientlyxidized their typical substrates (results not shown). Exceptf CYP1A1, all tested recombinant CYPs catalyzed oxidationf EE2 to X-hydroxy-EE2. Rat CYP2A2 followed by CYP3A2ere most efficient to metabolize EE2 to its X-hydroxylatedetabolite. In the case of EE2 oxidation to 2-hydroxy-EE2, only

YP2C6/11, 2D1 and 3A1/2 catalyzed this reaction, of them the

YP2C6 enzyme was most effective. The degree of EE2 hydrox-lation to this metabolite catalyzed by CYP2C6 overexpressedn SupersomesTM was almost 2-times higher than that foundn hepatic microsomes (Fig. 6).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

CYP1A1

CYP1A2

CYP2A1

CYP2A2

CYP2B1

CYP2C6

CYP2C11

CYP2D1

rela

tive

area

of E

E2

met

abol

ites

.D.N.D.N.D.N.D.N.D.N

ig. 6 – Oxidation of EE2 by rat recombinant CYPs. 50 pmol of ratll experiments. Values are given as means (n = 2). Standard devi

3.4. The effect of EE2 on oxidation of progesterone andtestosterone

Since EE2 affects the activities of CYP enzymes metabolizingseveral endogenous endocrine disruptors, it might contributeto the drug-drug interactions in organisms (Choi et al., 2013;Haslemo et al., 2011; Winkler et al., 2014). Therefore, weevaluated the impact of EE2 on metabolism of two steroidhormones, progesterone and testosterone that are endoge-nous endocrine disruptors. Progesterone and testosterone6�-hydroxylations are specific markers of CYP3A activities(Brian et al., 1990), while 21- and 16�-hydroxylation of proges-terone is specifically catalyzed by rat CYPs of the 2C subfamily(Yamazaki and Shimada, 1997).

The inhibition of progesterone and testosterone oxidationby EE2 was again analyzed using the two experimental pro-tocols, utilizing the pre-incubation of microsomes with EE2for 10 min in the presence or absence of NADPH before theaddition of progesterone or testosterone. These approaches

might resolve whether EE2 acts as a reversible inhibitor or asa mechanism-based inhibitor of CYP (Kent et al., 2002, 2008).

As shown in Fig. 7, reversible inhibition of progesterone 6�-hydroxylation was caused by EE2; essentially no changes in

CYP2D2

CYP2E1

CYP3A1

CYP3A2

micros

omes

X-OH EE 22-OH EE 2

.D.N.D.N

recombinant CYP/incubation and 50 �M EE2 were used inations were <10%. N.D. – not detectable.

858 e n v i r o n m e n t a l t o x i c o l o g y a n d p h

0

0.2

0.4

0.6

0.8

1

1.2

control preinc. EE 2+NADPH preinc. EE 2-NADPH

rela

�ve

met

abol

ite a

rea

16α-OH Prg

6β-OH Prg

21-OH Prg* * *

*

Fig. 7 – Effect of 50 �M EE2 on progesterone hydroxylationin control rat microsomes. Values are given as means ± SD

(n = 3). *P < 0.01 different from control MS (Student’s t-test).

inhibitory capacity of EE2 with respect to the progesterone6�-hydroxylation were detectable using both experimentalprotocols. However, whereas 21- and 16�-hydroxylation ofprogesterone were not affected by pre-incubation of EE2 withmicrosomes in the absence of NADPH, the pre-incubation ofthis compound in the presence of NADPH decreased the for-mation of both metabolites. These findings indicate that EE2itself inhibits CYP3A activity (measured as 6�-hydroxylationof progesterone), while its metabolites could modulate activ-ity of rat CYP2C (measured as 21- and 16�-hydroxylation ofprogesterone). These results are in agreement with the resultsobserved by Laurenzana et al. (2002). These authors postulatedthat in EE2-treated rats, this compound may covalently bind toCYP2C11, resulting in degradation of the enzyme, rather thanregulating of its expression.

In the case of the effect of EE2 on testosterone 6�-hydroxylations, a marker of CYP3A-mediated catalysis,inhibition was caused by this endocrine disruptor, but onlyafter pre-incubation of EE2 with microsomes in the presence of

NADPH (Fig. 8). This indicates that in the presence of this sub-strate (testosterone), the CYP3A-mediated metabolites of EE2inhibit the CYP3A-catalyzed 6�-hydroxylation of testosterone.

0

0.5

1

1.5

2

con trol preinc. EE2+NADPH preinc. EE2-NADPH

rela

tive

met

abol

ite a

rea

*

Fig. 8 – Effect of 50 �M EE2 on testosterone6�-hydroxylations in control rat microsomes. Values aregiven as means ± SD (n = 3). *P < 0.01 different from controlMS (Student’s t-test).

r

a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860

It corresponds to results found by Lin et al. (2002), who foundthat EE2 inactivated testosterone 6�-hydroxylation catalyzedby a human CYP3A orthologue, CYP3A4, in a mechanism-based manner. All these results suggest that complex effectsof EE2 on activities of CYP3A enzymes, depending on thesubstrates that are oxidized, should be considered. EE2 caninhibit the CYP3A-catalyzed progesterone 6�-hydroxylation asa reversible inhibitor, but a reactive intermediate of EE2 causesmechanism-based inactivation of CYP3A-mediated testos-terone 6�-hydroxylations.

4. Conclusions

The results found in this study demonstrate that EE2 is asubstrate of various rat hepatic CYPs. Of them the CYP2C6and 2C11 are most efficient in 2-hydroxylation of EE2, form-ing the major EE2 metabolite 2-hydroxy-EE2, whereas ratCYP2A and 3A catalyze EE2 hydroxylation predominantly toits minor hydroxylation metabolite. Its structure remains tobe identified. EE2 is not only the substrate of rat CYP2Cand 3A enzymes, but it is also an inhibitor of their enzymeactivities, the hydroxylation reactions of two endogenousendocrine disruptors, hormones progesterone and testos-terone. Depending on these substrates, EE2 acts either asa reversible inhibitor of CYP3A-mediated progesterone 6�-hydroxylation or inactivates CYP3A- and CYP2C-catalyzedtestosterone 6�-hydroxylation and progesterone 21- or 16�-hydroxylation, respectively, in a mechanism-based manner.Because of these complex interactions of EE2 in its metabolismby CYP2C and 3A and inhibition effects on these enzymes, thiscompound might participate in drug-drug interactions thatmay result in problems of their clinical significance.

Conflict of interest

The authors declare that there are no conflicts of interest.

Transparency document

The Transparency document associated with this article canbe found in the online version.

Acknowledgment

Financial support from GACR (grant 14-18344S in panel P301)and Charles University in Prague (UNCE 204025/2012) is highlyacknowledged.

e f e r e n c e s

Back, D.J., Orme, M.L., 1990. Pharmacokinetic drug interactionwith oral contraceptives. Clin. Pharmocokinet. 18 (6), 472–484.

Ball, S.E., Forrester, L.M., Wolf, C.R., Back, D.J., 1990. Differences in

the cytochrome P-450 isoenzymes involved in the2-hydroxylation of oestradiol and 17 alpha-ethinyloestradiol.Relative activities of rat and human liver enzymes. Biochem. J.267 (1), 221–226.

p h a r

B

B

B

C

C

D

E

F

G

G

G

G

G

H

H

H

K

K

K

e n v i r o n m e n t a l t o x i c o l o g y a n d

olt, H.M., 1979. Metabolism of estrogens-natural and synthetic.Pharmacol. Ther. 4 (1), 155–181.

orek-Dohalská, L., Hodek, P., Sulc, M., Stiborová, M., 2001.alpha-Naphthoflavone acts as activator and reversible orirreversible inhibitor of rabbit microsomal CYP3A6. Chem.Biol. Interact. 138 (1), 85–106.

rian, W.R., Sari, M.A., Iwasaki, M., Shimada, T., Kaminsky, L.S.,Guengerich, F.P., 1990. Catalytic activities of human livercytochrome P-450 IIIA4 expressed in Saccharomyces cerevisiae.Biochemistry 29 (51), 11280–11292.

ajthaml, T., Kresinová, Z., Svobodová, K., Sigler, K., Rezanka, T.,2009. Microbial transformation of synthetic estrogen17�-ethinylestradiol. Environ. Pollut. 157, 3325–3335.

hoi, Y.H., Lee, Y.K., Lee, M.G., 2013. Effects of17�-ethynylestradiol-induced cholestasis on thepharmacokinetics of doxorubicin in rats: reduced biliaryexcretion and hepatic metabolism of doxorubicin.Xenobiotica 43 (10), 901–907.

ickinson, B.D., Altman, R.D., Nielsen, N.H., 2001. Druginteraction between oral contraceptives and antibiotics.Obstet. Gynecol. 98 (5), 853–860.

bner, T., Remmel, R.P., Burchell, B., 1993. Human bilirubinUDP-glucuronosyltransferase catalyzes the glucuronidation ofethinylestradiol. Mol. Pharmacol. 43 (4), 649–654.

orbes-Bamforth, K.J., Coughtrie, M.W., 1994. Identification of anew adult human liver sulfotransferase with specificity forendogenous and xenobiotic estrogens. Biochem. Biophys. Res.Commun. 198 (2), 707–711.

uengerich, F.P., 1988. Oxidation of 17 alpha-ethynylestradiol byhuman liver cytochrome P-450. Mol. Pharmacol. 33 (5),500–508.

uengerich, F.P., 1990. Inhibition of oral contraceptivesteroid-metabolizing enzymes by steroids and drugs. Am. J.Obstet. Gynecol. 163 (6 (Pt 2)), 2159–2163.

uengerich, F.P., 2007. Mechanisms of cytochrome P450 substrateoxidation: mini review. J. Biochem. Mol. Toxicol. 21 (4),163–168.

uengerich, F.P., 2008. Cytochrome P450 and chemical toxicology.Chem. Res. Toxicol. 21 (1), 70–83.

uengerich, F.P., Kim, D.H., Iwasaki, M., 1991. Role of humancytochrome P-450 IIE1 in the oxidation of many low molecularweight cancer suspects. Chem. Res. Toxicol. 4 (2), 168–179.

aslemo, T., Refsum, H., Molden, E., 2011. The effect ofethinylestradiol-containing contraceptives on the serumconcentration of olanzapine and N-desmethyl olanzapine. Br.J. Clin. Pharmacol. 71 (4), 611–615.

elton, E.D., Goldzieher, J.W., 1977. Metabolism of ethynylestrogens. J. Toxicol. Environ. Health. 3 (1–2), 231–241.

odek, P., Borek-Dohalská, L., Sopko, B., Sulc, M., Smrcek, S.,Hudecek, J., Janku, J., Stiborová, M., 2005. Structuralrequirements for inhibitors of cytochromes P450 2B:assessment of the enzyme interaction with diamondoids. J.Enzyme. Inhib. Med. Chem. 20 (1), 25–33.

ent, U.M., Mills, D.E., Rajnarayanan, R.V., Alworth, W.L.,Hollenberg, P.F., 2002. Effect of 17-alpha-ethynylestradiol onactivities of cytochrome P450 2B (P450 2B) enzymes:characterization of inactivation of P450s 2B1 and 2B6 andidentification of metabolites. J. Pharmacol. Exp. Ther. 300 (2),549–558.

ent, U.M., Sridar, C., Spahlinger, G., Hollenberg, P.F., 2008.Modification of serine 360 by a reactive intermediate of17-alpha-ethynylestradiol results in mechanism-basedinactivation of cytochrome P450s 2B1 and 2B6. Chem. Res.Toxicol. 21 (10), 1956–1963.

linger, W., Lupp, A., Karge, E., Baumbach, H., Eichhorn, F., Feix,

A., Füldner, F., Gernhardt, S., Knels, L., Kost, B., Mertens, G.,Werner, F., Oettel, M., Römer, W., Schwarz, S., Elger, W.,Schneider, B., 2002. Estradiol, testosterone,

m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860 859

dehydroepiandrosterone and androstenedione: novelderivatives and enantiomers. Interactions with rat livermicrosomal cytochrome P450 and antioxidant/radicalscavenger activities in vitro. Toxicol. Lett. 128 (1–3),129–144.

Kresinová, Z., Moeder, M., Ezechiás, M., Svobodová, K., Cajthaml,T., 2012. Mechanistic study of 17�-ethinylestradiolbiodegradation by Pleurotus ostreatus: tracking of extracelullarand intracelullar degradation mechanisms. Environ. Sci.Technol. 46 (24), 13377–13385.

Laurenzana, E.M., Weis, C.C., Bryant, C.W., Newbold, R., Delclos,K.B., 2002. Effect of dietary administration of genistein,nonylphenol or ethinyl estradiol on hepatic testosteronemetabolism, cytochrome P-450 enzymes, and estrogenreceptor alpha expression. Food. Chem. Toxicol. 40 (1),53–63.

Lin, H.L., Kent, U.M., Hollenberg, P.F., 2002. Mechanism-basedinactivation of cytochrome P450 3A4 by 17alpha-ethynylestradiol: evidence for heme destruction andcovalent binding to protein. J. Pharmacol. Exp. Ther. 301 (1),160–167.

LoPachin, R.M., DeCaprio, A.P., 2005. Protein adduct formation asa molecular mechanism in neurotoxicity. Toxicol. Sci. 86 (2),214–225.

Omura, T., Sato, R., 1964. The carbon monooxide-bindingpigment of liver microsomes. II. solubilization, purificationand properties. J. Biol. Chem. 239, 2379–2385.

Rice, J.M., Diwan, B.A., Hu, H., Ward, J.M., Nims, R.W., Lubet, R.A.,1994. Enhancement of hepatocarcinogenesis and induction ofspecific cytochrome P450-dependent monooxygenaseactivities by the barbiturates allobarbital, aprobarbital,pentobarbital, secobarbital and 5-phenyl- and5-ethylbarbituric acids. Carcinogenesis 15 (2), 395–402.

Rodrigues, A.D., Lu, P., 2004. Is 17 alpha-ethinyl estradiol aninhibitor of cytochrome P450 2C19? Drug. Metab. Dispos. 32(3), 364–365.

Rogers, S.M., Back, D.J., Orme, M.L., 1987. Intestinal metabolism ofethinyloestradiol and paracetamol in vitro: studies usingUssing chambers. Br. J. Clin. Pharmacol. 23 (6), 727–734.

Schmid, S.E., Au, W.Y., Hill, D.E., Kadlubar, F.F., Slikker Jr., W., 1983.Cytochrome P-450-dependent oxidation of the 17alpha-ethynyl group of synthetic steroids. D-homoannulationor enzyme inactivation. Drug. Metab. Dispos. 11 (6), 531–536.

Stanczyk, F.Z., Archer, D.F., Bhavnani, B.R., 2013. Ethinyl estradioland 17�-estradiol in combined oral contraceptives:pharmacokinetics, pharmacodynamics and risk assessment.Contraception 87 (6), 706–727.

Stiborová, M., Martínek, V., Rydlová, H., Hodek, P., Frei, E., 2002.Sudan I is a potential carcinogen for humans: evidence for itsmetabolic activation and detoxication by human recombinantcytochrome P450 1A1 and liver microsomes. Cancer Res. 62(20), 5678–5684.

Stiborová, M., Borek-Dohalská, L., Aimová, D., Kotrbová, V.,Kukacková, K., Janouchová, K., Rupertová, M., Ryslavá, H.,Hudecek, J., Frei, E., 2006. Oxidation pattern of the anticancerdrug ellipticine by hepatic microsomes – similarity betweenhuman and rat systems. Gen. Physiol. Biophys. 25 (3), 245–261.

Surry, D.D., Meneses-Lorente, G., Heavens, R., Jack, A., Evans, D.C.,2000. Rapid determination of rat hepatocyte mRNA inductionpotential using oligonucleotide probes for CYP1A1, 1A2, 3Aand 4A1. Xenobiotica 30 (5), 441–456.

Wang, B., Sanchez, R.I., Franklin, R.B., Evans, D.C., Huskey, S.E.,2004. The involvement of CYP3A4 and CYP2C9 in themetabolism of 17 alpha-ethinylestradiol. Drug. Metab. Dispos.32 (11), 1209–1212.

Walsky, R.L., Gaman, E.A., Obach, R.S., 2005. Examination of 209drugs for inhibition of cytochrome P450 2C8. J. Clin.Pharmacol. 45 (1), 68–78.

d p h

Zhang, H., Cui, D., Wang, B., Han, Y.H., Balimane, P., Yang, Z., Sinz,

860 e n v i r o n m e n t a l t o x i c o l o g y a n

Winkler, J., Goldammer, M., Ludwig, M., Rohde, B., Zurth, C., 2014.Pharmacokinetic drug-drug interaction between ethinylestradiol and gestodene, administered as a transdermalfertility control patch, and two CYP3A4 inhibitors and aCYP3A4 substrate. Eur. J. Drug Metab. Pharmacokinet.,

http://dx.doi.org/10.1007/s13318-014-0215-8 (in press).

Wu, D., Cederbaum, A.I., 1993. Ethanol consumption by thenursing mother induces cytochrome P-4502E1 in neonatal ratliver. J. Pharmacol. Exp. Ther. 267 (1), 560–566.

a r m a c o l o g y 3 8 ( 2 0 1 4 ) 852–860

Yamazaki, H., Shimada, T., 1997. Progesterone and testosteronehydroxylation by cytochromes P450 2C19, 2C9, and 3A4 inhuman liver microsomes. Arch. Biochem. Biophys. 346 (1),161–169.

M., Rodrigues, A.D., 2007. Pharmacokinetic drug interactionsinvolving 17alpha-ethinylestradiol: a new look at an old drug.Clin. Pharmacokinet. 46 (2), 133–157.