The Role of Ovarian Metabolism in 4-Vinylcyclohexene Metabolites and 7,12-

Dimethylbenz[a]anthracene Induced Ovotoxicity in Mice

Item Type text; Electronic Dissertation

Authors Rajapaksa, Kathila Seuwandhi

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 06/05/2018 22:39:54

Link to Item http://hdl.handle.net/10150/194403

THE ROLE OF OVARIAN METABOLISM IN 4-VINYLCYCLOHEXENE

METABOLITES AND 7,12-DIMETHYLBENZ[A]ANTHRACENE-INDUCED

OVOTOXICITY IN MICE

by

Kathila Seuwandhi Rajapaksa

________________________

A Dissertation Submitted to the Faculty of the

GRADUATE INTERDISCIPLINARY PROGRAM IN PHYSIOLOGICAL SCIENCES

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2007

2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Kathila Seuwandhi Rajapaksa entitled The Role of Ovarian Metabolism in 4-Vinylcyclohexene Metabolites and 7,12-Dimethylbenz[a]anthracene-Induced Ovotoxicity in Mice. and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy

_______________________________________________________________________ Date: 01/11/2007Dr. Patricia B. Hoyer _______________________________________________________________________ Date: 01/11/2007Dr. I. Glenn Sipes _______________________________________________________________________ Date: 01/11/2007Dr. Nathan J. Cherrington _______________________________________________________________________ Date: 01/11/2007Dr. Qin Chen _______________________________________________________________________ Date: 01/11/2007Dr. Stephen H. Wright

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College. I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

________________________________________________ Date: 01/11/2007Dissertation Director: Dr. Patricia B. Hoyer

3

STATEMENT BY THE AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Request for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: _Kathila S. Rajapaksa___

4

ACKNOWLEDGMENTS

I would like to acknowledge everyone that has made graduate school a memorable experience. First and foremost I would like to thank my mentor, Dr. Patricia B. Hoyer, for taking me into her lab and training me in the field of Reproductive Physiology. Thank you for all the time you spend mentoring me to become the independent scientist I am today. The freedom you have given me in the lab has allowed me to design experiments and trouble shoot them by consulting literature. You have not only taught me science and lab skills, but also helped me develop critical thinking and presentation skills. Thanks to your mentoring, I am able to present my data with confidence to other scientists. You have made graduate school a truly wonderful learning experience. I would like to equally thank Dr. I. Glenn Sipes for taking the time to help me with my project. Even though you had your own students, you were always available for my questions and to go over my data. Your insight and criticism has helped me learn and develop my knowledge of drug metabolism and toxicology. In addition, thank you for taking the time to talk to me about my career and my future as a scientist. I would also like to thank the rest of my dissertation committee, Dr. Qin Chen, Dr. Stephen Wright, and Dr. Nathan Cherrington for their time spent with suggestions and critiques. I would like acknowledge couple of other researchers that has influenced this dissertation work and my career. I would like to thank Dr. Mark P. Grillo for giving me the opportunity to intern in his lab. You have taught me a great deal about drug metabolism, drug discovery process, and the pharmaceutical industry. I appreciate the time you spent working with me on my dissertation, even though you had deadlines of your own. I would also like to thank my undergraduate mentor, Dr. Nancy Staub, for opening my eyes to the world of research. Thank you for taking me into your lab as a freshman who has never touched a pipettor, and laying the foundation for my scientific career. I like to thank all of my friends during graduate school for their support. Special thanks to Khameeka for the countless drives to the airport, to Nicola for the all the laughs at UMC, to Ji-eun for all the movie breaks, to Ellen for taking the time to help me with this project even after you have left the lab, and Michelle for working on my project after I have left SF. I would like to acknowledge everyone in the Hoyer lab. Especially Patty for talking the time to help collect ovaries, Jess for all the midnight trips to Starbucks, Aileen for the lunch and coffee breaks, and Shannon for listening to me practice my talks. I would like to thank all the other past and present members of the Hoyer lab, especially Kary, Zeli, and Sam. Finally I would like to thank my parents and my brother for their support. Thank you for all your sacrifices through my 20 years of school. Your sacrifices and unconditional love have made this work possible.

5

DEDICATION

This dissertation is dedicated to my mom and dad. Thank you for your love, support, and

encouragement throughout my education.

6

TABLE OF CONTENTS PAGE

LIST OF FIGURES .............................................................................................................9 LIST OF TABLES.............................................................................................................11 LIST OF ABBREVIATIONS............................................................................................12 ABSTRACT.......................................................................................................................15 CHAPTER 1 INTRODUCTION .......................................................................................17

Ovarian Follicular Development and Ovarian Physiology............................................17

Ovarian Toxicants and Consequences on Reproduction ...............................................22

Metabolism of Xenobiotics............................................................................................25

Ovarian Metabolism.......................................................................................................27

Role of Metabolism in 4-Vinylcyclohexene-Induced Ovotoxicity................................29

VCD-Induced Ovotoxicity.............................................................................................33

Role of Metabolism in 7,12-Dimethylbenz[a]anthracene-Induced Ovotoxicity ...........35

CHAPTER 2 GENERAL OVERVIEW AND METHODS ..............................................40

Statement of Problem.....................................................................................................40

Research Objectives.......................................................................................................41

Materials and Methods...................................................................................................43 Reagents.....................................................................................................................43 Animals ......................................................................................................................44 In Vitro Ovarian Cultures...........................................................................................45 In Vivo Animal Dosing ..............................................................................................47 Histological Evaluation..............................................................................................47 Toluidine Blue Staining .............................................................................................48 RNA Isolation ............................................................................................................49 First Strand cDNA Synthesis and Real Time Polymerase Chain Reaction (PCR)....49 Confocal Microscopy.................................................................................................50

7

TABLE OF CONTENTS – Continued PAGE

Reactivity of VCD with GSH ....................................................................................51 Synthesis and Purification of VCD-Mono and DiGSH Adducts...............................51 Extraction of VCD-GSH Adduct from Liver and Ovarian Tissue ............................54 Detection of VCD-GSH Adduct from Ovarian Culture Media .................................54 Statistical Analysis.....................................................................................................55

CHAPTER 3 OVARIAN BIOACTIVATION OF 4-VINYLCYCLOHEXENE METABOLITES TO THE OVARIAN TOXICANT 4-VINYLCYCLOHEXENE DIEPOXIDE: THE INVOLVEMENT OF CYP 2E1 ENZYME ......................................56

Abstract ..........................................................................................................................56

Introduction....................................................................................................................57

Experimental Methods ...................................................................................................58

Results............................................................................................................................59

Discussion ......................................................................................................................68 CHAPTER 4 OVARIAN DETOXIFICATION OF OVOTOXICANT 4-VINYLCYCLOHEXENE DIEPOXIDE: THE INVOLVEMENT OF GLUTATHIONE CONJUGATION ...............................................................................................................74

Abstract ..........................................................................................................................74

Introduction....................................................................................................................75

Experimental Methods ...................................................................................................78

Results............................................................................................................................79

Discussion ......................................................................................................................87

8

TABLE OF CONTENTS – Continued PAGE

CHAPTER 5 OVARIAN BIOACTIVATION OF 7,12-DIMETHYLBENZ[A]ANTHRACENE IN B6C3F1 MICE: THE INVOLVEMENT OF MICROSOMAL EPOXIDE HYDROLASE (mEH) .........................................................92

Abstract ..........................................................................................................................92

Introduction....................................................................................................................93

Experimental Methods ...................................................................................................94

Results............................................................................................................................95

Discussion ....................................................................................................................105 CHAPTER 6 COMPARISON OF 4-VINYLCYCLOHEXENE DIEPOXIDE AND 7,12- DIMETHYLBENZ[A]ANTHRACENE-INDUCED OVOTOXICITY IN THE CYP 2E1 WILD-TYPE AND NULL OVARIES IN CULTURE ...................................................110

Abstract ........................................................................................................................110

Introduction..................................................................................................................111

Experimental Methods .................................................................................................113

Results..........................................................................................................................114

Discussion ....................................................................................................................118 CHAPTER 7 SUMMARY: THE ROLE OF OVARIAN METABOLISM IN 4- VINYLCYCLOHEXENE METABOLITES AND 7,12-DIMETHYLBENZ[A]ANTHRACENE-INDUCED OVOTOXICITY IN MICE .........122 REFERENCES ................................................................................................................127

9

LIST OF FIGURES

PAGE

Figure 1.1: Structure of the Mammalian Ovary..........................................................21 Figure 1.2: Proposed VCD Metabolism Scheme........................................................32 Figure 1.3: DMBA Metabolic Pathway......................................................................39 Figure 2.1: In Vitro Neonatal Whole Ovary Culture System......................................46 Figure 3.1: VCD-Induced Ovotoxicity in Various Strains of Mouse Ovaries in Culture.......................................................................................................61 Figure 3.2: 1,2-VCM-Induced Ovotoxicity in B6C3F1 Ovarian Cultures..................62 Figure 3.3: Involvement of CYP 2E1 in 1,2-VCM-Induced Ovotoxicity in Ovarian Cultures .....................................................................................................64 Figure 3.4: Effect of In Vitro 1,2-VCM and VCD Exposure on Ovarian Morphology...............................................................................................65 Figure 3.5: Involvement of CYP 2E1 in VCH-Induced Ovotoxicity .........................67 Figure 4.1: Proposed Reaction of VCD with GSH.....................................................77 Figure 4.2: Reactivity of VCD in the Presence of GSH to Form VCD-Mono and DiGSH Adducts .................................................................................80 Figure 4.3: Synthesis and Purification of VCD-MonoGSH Adduct...........................81 Figure 4.4: Synthesis and Purification of VCD-DiGSH Adduct ................................83 Figure 4.5: Detection of VCD-Mono and DiGSH Adducts in 28d Female B6C3F1

Mice ..........................................................................................................84 Figure 4.6: VCD-Mono and DiGSH Adduct Formation in CYP 2E1 Wild-Type and Null Ovary Culture Media ........................................................................86 Figure 5.1: Effect of Varying Concentrations of DMBA on Follicle Loss ................96

Figure 5.2: Time Course of DMBA-Induced Follicle Loss........................................98

10

LIST OF FIGURES – Continued

PAGE Figure 5.3: Effect of DMBA on Toluidine Blue Staining ..........................................99 Figure 5.4: Effect of DMBA on mEH mRNA Expression and Total Follicle Loss .........................................................................................................101 Figure 5.5: Effect of DMBA on mEH Protein..........................................................103 Figure 5.6: Effect of mEH Inhibitor (CHO) on DMBA-Induced Follicle Loss .......104 Figure 6.1: VCD-Induced Ovotoxicity in CYP 2E1 Wild-Type and Null Ovarian Cultures ...................................................................................................115 Figure 6.2: DMBA-Induced Ovotoxicity in CYP 2E1 Wild-Type and Null Ovarian Cultures ...................................................................................................117

11

LIST OF TABLES

PAGE Table 2.1: Custom Designed Primer Sequences for mEH, GST and β-actin .............50 Table 2.2: HPLC Gradient Used in LC/MS Detection of VCD-GSH Adducts..........53 Table 2.3: HPLC Gradient Used in VCD-GSH Adduct Purification .........................53 Table 2.4: SRM Conditions for VCD-Mono and DiGSH Adduct Detection .............53

12

LIST OF ABBREVIATIONS

AU Arbitrary Units

B[a]P: benzo[a]pyrene

BD: 1,3-butadine

BDE: butadiene diepoxide

2-BP: 2-bromopropane

4,5-BPO: benzo[a]pyrene 4,5-oxide

CHO: cyclohexene oxide

CL: corpus luteum

CPA: cyclophosphamide

CYP 450: cytochrome P450

d: day(s)

DDT: dichloro-diphenyl-trichloroethane

DEHP: di-(2-ethylhexyl) phthalate

DMBA: 7,12-dimethylbenz[a]anthracene

FSH: follicle stimulating hormone

GnRH: gonadodotropin-releasing hormone

GSH: glutathione

GST: glutathione-S-transferase

h: hour(s)

hCG: human chorionic gonadotropin hormone

13

LIST OF ABBREVIATIONS- Continued

HPLC: High-performance liquid chromatography

LC/MS: Liquid Chromatography/Mass Spectrometry

LH: luteinizing hormone

3-MC: 3-methylcholanthrene

mEH: microsomal epoxide hydrolase

min: minute(s)

MS/MS: tandem mass spectrometry

mo: month(s)

m/z: mass to charge ratio

PAH: polycyclic aromatic hydrocarbon

PMSG: pregnant mare serum gonadotropin

PND: postnatal day

4-PC: 4-phenylcyclohexene

TCDD: 2,3,7,8-tetrachlorodibenzo-p-dioxin

tetrol: 4-[1,2-dihydroxy]ethyl-1,2-dihydroxycyclohexane

TPT: triphenyltin

ST: sulfonate

UDP-GT: UDP-glucuronosyltransferase

VCD: 4-vinylcyclohexene diepoxide

VCH: 4-vinylcyclohexene

14

LIST OF ABBREVIATIONS- Continued

1,2-VCM: vinylcyclohexene 1,2-monoepoxide

7,8-VCM: vinylcyclohexene 7,8-monoepoxide

wk: week(s) yr: year(s)

15

ABSTRACT

Ovarian toxicants 4-vinylcychlohexene (VCH) and 7,12-

dimethylbenz[a]anthracene (DMBA) requires bioactivation to induce follicle loss. VCH

is bioactivated to monoepoxides (1,2-VCM and 7,8-VCM), and subsequently to an

ovotoxic diepoxide (VCD) by hepatic CYP 2A and CYP 2B. DMBA is sequentially

bioactivated to the ovotoxicant DMBA-3,4-diol-1,2-epoxide by hepatic CYP 1B1,

microsomal epoxide hydrolase (mEH), and CYP 1A1/1B1 enzymes. Even though the

liver is the primary organ metabolizing VCH and DMBA to reactive intermediates,

several studies suggest that the ovary can also metabolize these two compounds. Studies

were designed to investigate the role of ovarian metabolism in the resulting ovotoxicity

of these two compounds using a novel mouse ovarian culture system. The hypothesis

was that the ovary can participate in bioactivation and detoxification of VCH/VCM and

DMBA and thereby influence the resulting ovotoxicity.

Postnatal day 4 CYP 2E1 wild-type, null and B6C3F1 mouse ovaries were

incubated with 1,2-VCM, VCD or DMBA for various lengths of time. 28 day old female

CYP 2E1 wild-type and null mice were dosed (15d, i.p) with VCH, 1,2-VCM, VCD, or

sesame oil (control). Following incubations and dosing, ovaries were prepared for

histological evaluation of follicle numbers, mEH mRNA level, or mEH protein level.

Medium from cultures were analyzed by LC/MS for VCD-GSH adducts.

DMBA was found to be a potent ovotoxicant compared to VCH/VCM/VCD. In

the ovarian culture system, VCM-induced toxicity required the CYP 2E1 enzyme.

However, in vivo dosing studies indicated that in the presence of hepatic metabolism the

16

ovary plays a minimal role in VCH/VCM-induced toxicity. Studies utilizing LC/MS

showed that once bioactivated to VCD, this ovotoxic metabolite can be detoxified by

glutathione conjugation in the ovary. Follicle loss induced by the ovotoxicant DMBA

was found to involve mEH enzyme in culture.

Collectively, these studies show that the ovary has the capacity to bioactivate and

detoxify ovotoxicants. In the presence of hepatic metabolism ovarian effects might play

only a minimal role in the resulting toxicity. The role of ovarian metabolism in the whole

animal needs to be further investigated, especially for potent toxicants such as DMBA

that can induce ovotoxicity at nanomolar concentrations.

17

CHAPTER 1

INTRODUCTION

Ovarian Follicular Development and Ovarian Physiology

The embryonic ovary contains three cell types that have different origins within

the developing embryo. The germ cells that develop into oocytes arise from the primitive

ectodermal cells, epithelial cells that develop into granulosa cells arise from supporting

cells of the coelomic epithelium, and interstitial cells arise from mesenchymal cells of the

gonadal ridge (La Barbera, 1997).

In humans, at around nine weeks of gestation, bi-potential gonads develop into

ovaries in the absence of testis-determining gene SRY (Porterfield, 1997). Once germ

and somatic cells reach the ovary, they increase in number by undergoing mitosis. Once

germ cell mitotic division stops, meiosis follows. Germ cells will not complete the first

meiotic division, but become arrested in the first meiotic prophase. Once arrested in

meiosis, these oocytes become surrounded by squamous-shaped somatic cells known as

granulosa cells, and a semi-permeable basement membrane. These structures are called

primordial follicles, and once formed these follicles become metabolically quiescent (Fig.

1.1). During meiosis not all oocytes will be surrounded by somatic cells, and as a result

they will undergo apoptosis (atresia in the ovary). This pre-natal atretic process

drastically reduces the number of oocytes in a female at birth. For example, in the rat

ovary, the number of oocytes is reduced from 75,000 to 27,000 during this process.

18

Furthermore, since the remaining oocytes have already undergone the beginning stages of

meiosis, primordial follicles cannot be replenished. Therefore, a female is born with a

finite number of primordial follicles (Flaws and Hirsfield, 1997).

During the reproductive life span of a female, primordial follicles grow into large

follicles (pre-ovulatory) that upon ovulation, release the oocyte for fertilization. At the

start of a new menstrual cycle, a cohort of primordial follicles will be recruited to form

primary follicles. This recruitment drives primordial follicles from a dormant to a

metabolically active state. Primary follicles are characterized by a single layer of

cuboidal-shaped granulosa cells surrounding an enlarged oocyte (due to an increase in

cytoplasmic and nuclear volume), and a glycoprotein matrix that surrounds the oocyte

called the zona pellucida. Granulosa cells develop processes that penetrate the zona

pellucida for communication and nutrient exchange with the oocyte. Primary follicles

then continue to grow and form secondary follicles that contain multiple layers of

granulosa and theca cells. Theca cells that surround granulosa cells are termed theca

interna, and the theca cells that surround theca interna are termed theca externa. Theca

interna cells produce androgen that diffuses into granulosa cells, where it is converted to

17β-estradiol via aromatase. Theca externa is thought to be a protective layer of

fibroblastic cells. Two hypotheses exists that debate the origin of the theca cells: 1) theca

cells arise from connective tissue, or 2) they arise from the same embryonic cells that

give rise to granulosa cells. With an increase in production of 17β-estradiol, an antrum

or an estrogen-filled cavity is formed within the granulosa cell layer. Follicles containing

an antrum are classified as antral follicles. The antrum continues to grow and the

19

follicular fluid contains a collection of many components such as FSH, lutenizing

hormone (LH), prolactin, progesterone, and androgens in addition to 17β-estradiol. The

increased antral space pushes granulosa cells towards the basement membrane, leaving

only a single layer of granulosa cells called the cumulus oophorus surrounding the

oocyte. The cumulus oophorus remains attached to granulosa cells surrounding the

antrum via a bridge-like granulosa cell structure called the stalk. These large follicles

containing enlarged antrums are classified as pre-ovulatory follicles (Fig. 1.1; Flaws and

Hirsfield, 1997).

In women, the ovary undergoes several distinct phases (follicular phase,

ovulation, and luteal phase) collectively know as the menstrual cycle. This cycle

culminates in the release (ovulation) of the oocyte for fertilization. The ovarian cycle is

under the control of hypothalamic and pituitary hormones. The hypothalamic hormone,

gonadotropin-releasing hormone (GnRH), is released into the portal blood vessels, where

it travels to the anterior pituitary, and induces the synthesis and release of FSH and LH in

the gonadotrophs.

The first phase of the menstrual cycle, termed the follicular phase, begins with an

increase in GnRH, FSH, and LH levels. FSH promotes the development of follicles in

the ovary. LH acting on theca interna cells induces synthesis (steroidogenesis) of

androgens from cholesterol. Androgens diffuse into the granulosa cell layer, and are

converted to 17β-estradiol by the enzyme aromatase. Aromatase expression in the

granulosa cells is induced by FSH. Increased production of 17β-estradiol has many

functions during the follicular phase, including follicular growth, and induction of LH

20

receptor expression on the granulosa cells to prepare the follicle for the LH surge.

During this phase 17β-estradiol inhibits the production of LH (negative-feedback) in the

anterior pituitary. However, at the end of this phase, negative feedback induced by 17β-

estradiol on LH switches to a positive feedback loop, where the increase in production of

17β-estradiol increases the release of LH from the anterior pituitary. As a result,

increased 17β-estradiol stimulates a peak of LH release (LH surge) that triggers

ovulation. By this time a primordial follicle has developed into the pre-ovulatory stage.

Ovulation immediately follows the LH surge at the end of the follicular phase.

Following the LH surge, the oocyte of the pre-ovulatory follicle will complete the first

meiotic division, and become arrested in metaphase II. During this time cumulus

oophorus cells loose their connection with the stalk, the follicular wall ruptures, and the

oocyte is released into the oviduct for fertilization.

Release of the oocyte marks the beginning of the luteal phase of the menstrual

cycle. Once the oocyte is released the remaining cells (granulosa and theca) in the

follicle undergo a differentiation process called lutenization, and forms a corpus luteum

(CL). Luteal cells produce progesterone when stimulated by LH. Progesterone keeps

the myometrium in a quiescent stage; therefore, this hormone is crucial for the

maintenance of pregnancy. If the oocyte is fertilized, the CL is maintained by the human

chorionic gonadotropin (hCG) produced from the placenta. Conversely, if there is no

fertilization, production of progesterone will cease at the end of the menstrual cycle, and

the CL will undergo regression (death by apoptosis; Porterfield, 1997).

21

Figure 1.1: Structure of the Mammalian Ovary. Ovary is a heterogeneous organ

containing follicles at different stages of development, corpus lutea (CL), and interstitial

cells. At the start of the ovarian cycle, a primordial follicles will grow into a pre-

ovulatory follicle, and release the oocyte for fertilization. Following ovulation, granulosa

and theca cells will differentiate and form a CL. If the oocyte is not fertilized the CL will

regress and become part of the interstitial cells.

22

During the reproductive life span of a female most follicles (99%) undergo

atresia, and will not be ovulated. A female is born with one to two million oocytes, and

at the time of puberty two to four hundred thousand oocytes remain in the ovary. Only

about four hundred of these follicles will be ovulated for fertilization. The exact

mechanism that triggers atresia is not known. Excess or insufficiency of LH, FSH,

estradiol, androgen, and growth factors are thought to be involved (Flaws and Hirshfield,

1997). Unlike males, the female gonad does not contain stem cells that can replace the

germ cell pool. Because a female is born with a finite number of primordial follicles,

depletion of follicle populations via atresia results in ovarian senescence or menopause in

women. Menopause is associated with risk factors for several diseases including

cardiovascular, osteoporosis, and cancer (La Barbera, 1997; Porterfield, 1997).

Ovarian Toxicants and Consequences on Reproduction

In recent years, epidemiology studies have reported a decrease in fertility among

women. This decrease in fertility is correlated with exposure to numerous xenobiotics

such as environmental (polycyclic aromatic hydrocarbons; Mlynarcikova et al., 2005),

pharmaceutical (cyclophosphamide; Anderson et al., 2006), and industrial chemicals (2-

bromopropane; Kim et al., 1996) that target follicle populations in the ovary.

The consequences of follicle loss vary depending on the follicle population

targeted by the chemical, and the time of exposure to the chemical during the life span.

Chemical-induced depletion of the primordial follicle population results in irreversible

23

damage, because a female is born with a finite number of primordial follicles. Loss of

this pool of follicles can result in premature ovarian failure or early menopause in women

(Hoyer, 1997). For example, the industrial chemicals 4-vinylcyclohexene (VCH) and 4-

vinylcyclohexene diepoxide (VCD) have been shown to selectively target small pre-

antral (primordial and primary) follicles in the rodent ovary. This depletion of small pre-

antral follicles results in a loss of larger follicles (due to lack of small follicles available

for recruitment), a decrease in ovarian volume, decrease in 17β-estradiol, and an increase

in FSH (Hooser et al., 1994; Mayer et al., 2002). These changes are similar to those

associated with menopause in women. Other ovotoxicants that target small pre-antral

follicles include two polycyclic aromatic hydrocarbons (PAH) found in cigarette smoke,

3-methylcholanthrene (3-MC) and benzo[a]pyrene (B[a]P); and the alkylating agent 1,3-

butadiene (BD; Mattison and Thorgeirsson, 1978; Basler and Rohrborn, 1976; Doerr et

al., 1995).

Conversely, disruption of larger follicles (secondary and antral) by a compound

results in reversible damage, because primordial follicles are not damaged and remain to

replenish the larger follicle pool. This can result in temporary infertility, and once

exposure to the ovotoxicant is removed the female may regain normal cyclicity.

However, if the exposure to the compound persists for longer periods of time, the small

pre-antral pool can be depleted due to increased demands on recruitment and this can

result in ovarian failure (Hoyer, 1997). Some chemicals that target larger follicles

include the chemotherapeutic agent cyclophosphamide (CPA), and the phthalate di-(2-

24

ethylhexyl) phthalate (DEHP; Plowchalk and Mattison, 1992; Shiromizu et al., 1984;

Davis et al., 1994).

Some compounds such as CPA, 7,12-dimethybenz[a]anthracene (DMBA;

component of cigarette smoke) and 2-bromopropane (2-BP; occupational chemical) can

target all follicle populations in the ovary (Mattison et al., 1980; Yu et al., 1999). This

can simultaneously deplete the ovary of all follicles, and can result in abrupt onset of

ovarian failure.

The time at which a female is exposed to ovotoxicants is also important in the

resulting toxicity. In utero exposure to ovotoxicants can result in sterility and impaired

ovarian development, because this is the time when a female acquires the primordial

follicle pool (Hoyer, 1997). For example, ionizing radiation targets rapidly dividing

cells, and can deplete the formation of the primordial pool (Mattison and Schulman,

1980).

Exposure to ovotoxic chemicals during pre-pubertal years can result in sterility,

delayed or early onset of puberty, and ultimately a shortened reproductive life span

(Hoyer, 1997). For example, exposure to the fungicide and antifouling agent triphenyltin

(TPT) during pre-pubertal development accelerated vaginal opening at the lower dose

(2mg/kg/d), and delayed vaginal opening at the higher dose (6mg/kg/d) in rats. An

increase in ovarian weights was also observed in this study at both doses of TPT (Grote et

al., 2006). A single dose (10µg/kg) of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at

29d of age resulted in a delayed onset of puberty, acceleration in the onset of irregular

25

cyclicity, and a decrease in number of days of reproductive lifespan in rats (Franczak et

al., 2006).

In summary, if an adult is exposed to ovotoxic chemicals this can result in

acyclicity and temporary infertility if larger follicles are affected. Conversely, exposure

to compounds that target small pre-antral follicles can result in premature ovarian failure

or early menopause as discussed earlier (Hoyer, 1997).

Most ovotoxic compounds have to be metabolized to a bioactive form to cause

ovarian toxicity. For example, the ovotoxic form of CPA is the metabolite

phosphoramide, the ovotoxic metabolite of BD is butadiene diepoxide (BDE), and the

ovotoxic metabolite of VCH is VCD (Polwachalk and Mattison, 1991; Doerr et al.,

1995).

Metabolism of Xenobiotics

Females are exposed to ovotoxic compounds via three main routes (skin, lung, or

gastrointestinal tract). The physical property that allows these compounds to be readily

absorbed is their lipophilicity. This same chemical property can hinder the excretion of

these compounds from the body. Therefore, xenobiotics undergo a process called

metabolism or biotransformation that decreases this lipophilicity, and increases water

solubility. The primary organ involved in metabolism is the liver. Metabolism or

biotransformation is divided into two phases (phase I and II).

26

Phase I is termed bioactivation, and this phase result in the formation of bioactive

metabolites due to the addition of functional groups. There are three main phase I

reactions: hydrolysis, reduction, and oxidation. Numerous enzymes are involved in

catalyzing these reactions. Some enzymes involved in catalyzing hydrolysis reactions

are: carboxylesterases, peptidases, and epoxide hydrolases (microsomal and soluble).

Enzymes involved in catalyzing reduction reactions include: alcohol dehydrogenase,

carbonyl reductases, cytochrome P450 (CYP 450) isoforms, and NADPH quinone

oxidoreductase. Some enzymes involved in catalyzing oxidation reactions include: CYP

450 isoforms, monoamine oxidases, flavin monooxygenases, and alcohol dehydrogenase.

This phase only slightly increases water solubility of the ovotoxicants.

Phase II is termed detoxification, and during this phase large conjugates are added

to functional groups that are exposed by phase I. Reactions in this phase include

glucuronidation, sulfation, methylation, acetylation, amino acid conjugation, and

glutathione conjugation. Unlike with phase I reactions, addition of these conjugates

greatly increases water solubility. Furthermore, the addition of these large conjugates

increases the size of these xenobiotics, and restricts the ability of these compounds to

readily cross the plasma membrane. Phase II reactions do not necessarily follow phase I

reactions (Parkinson, 2001).

Following metabolism, these compounds can be excreted in urine or bile. Even

though metabolism results mainly in excretion of ovotoxicants, this process can also lead

to the formation of reactive intermediates. This especially can occur following phase I

27

(bioactivation) reactions, where functional groups are added. These functional groups

can react with cellular protein or DNA resulting in toxicity (Gregus and Klaassen, 2001).

Ovarian Metabolism

Even though the liver is the primary organ involved in metabolism or

biotransformation of xenobiotics, extra-hepatic organs such as the gastrointestinal tract,

kidney, lung, testis, placenta, and ovary have the capacity to participate in metabolism

(Krishna and Koltz, 1994). This extra-hepatic metabolism, especially bioactivation, can

pose a problem if the organ metabolizing xenobiotics to their reactive intermediates is

also the target organ.

Studies have shown that the ovary expresses enzymes involved in phase I and II

reactions, such as isoforms of CYP 450 family, microsomal epoxide hydrolase (mEH),

and glutathione-S-transferase (GST). In addition to expression, these enzymes were also

shown to be catalytically active in the ovary (Cannady et al., 2002, 2003; Mukhart et al.,

1978).

Mukhtar et al. (1978) evaluated the activity of several bioactivation enzymes in

the rat ovary during development. Ovarian mitochondrial and microsomal CYP 450

content gradually increased with age, and reached a peak at puberty (40d) at which point

the level remained constant. In this study, the highest level of ovarian mEH activity was

detected between 20 and 40d of age. Following 40d, mEH activity slightly decreased,

and then a plateau was reached. In a study by Eliasson et al. (1997) the highest DMBA

28

bioactivation was observed in porcine ovarian CL mitochondrial and microsomal

fractions compared with different follicle populations, and the highest CYP 450 content

was also observed in the CL. When CYP 450 activity was inhibited using ellipticine

(inhibits the interaction between CYP 450 and NADPH CYP 450 reductase), DMBA

metabolism in the CL significantly decreased. Bengtsson et al. (1983) have detected

metabolites of B[a]P and DMBA following rat ovarian incubations. Collectively, these

studies suggest that the ovary has the capacity to bioactivate xenobiotics.

Studies also show that the ovary expresses enzymes that catalyze phase II

conjugation reactions. Mukhtar et al. (1978) have shown an increase in ovarian GST

activity up to 40d of age, and following 40d of age, GST activity slightly decreased

reaching a plateau. Studies by Becedas and Bengtsson-Ahlberg (1995) have shown that

UDP-glucuronosyltransferase (UDP-GT), sulfotransferase (ST), and GST are

catalytically active in the rat ovary. In this study, the greatest UDP-GT and ST activity

was detected during the luteal phase, while GST activity did not change as a function of

ovarian cycle.

Taken together, these studies suggest that the ovary expresses metabolic enzymes,

and the expression of these enzymes might be regulated by the ovarian cycle. Thus,

ovarian metabolism of ovotoxicants has the potential to result in ovarian toxicity.

29

Role of Metabolism in 4-Vinylcyclohexene-Induced Ovotoxicity

The industrial chemical 4-vinylcyclohexene (VCH) causes ovotoxicity and

premature ovarian failure in mice (Hooser et al., 1994). Human exposure to VCH can

occur through an occupational or industrial setting. VCH is used as an intermediate in

the manufacture of rubber tires, flame retardants, insecticides, plasticizers, and

antioxidants. VCH can also be formed spontaneously in the manufacture of butadiene

(BD). To our knowledge there are no epidemiological reports of human exposure to

VCH to date.

It has been reported that VCH-induced ovotoxicity is due to bioactivation of this

compound to a diepoxide (4-vinylcyclohexne diepoxide; VCD). Studies show that VCH

can undergo biotransformation or metabolism in the liver by CYP 450 enzymes to form

monoepoxides (1,2- VCM and 7,8-VCM). Subsequently, these monoepoxides are

converted to the diepoxide, VCD, which is the ultimate ovarian toxicant (Fontaine et al.,

2001b; Fig 1.2).

Numerous studies suggest that VCH bioactivation to VCD is required for

ovotoxicity. ED50 values for VCH, 1,2-VCM, 7,8-VCM and VCD for small pre-antral

follicle loss in mice were 2.7mmol/kg/d, 0.57mmol/kg/d, 0.77mmol/kg/d, and

0.27mmol/kg/d respectively (Smith et al., 1990b). Furthermore, VCD was shown to be

more chemically reactive compared to the 1,2-VCM metabolite (Doerr et al., 1995). Co-

incubation of VCH with chloramphenicol (CYP 450 inhibitor) reduced the formation of

1,2-VCM in a concentration dependent manner. Follicle loss induced by VCH was also

partially prevented by the treatment of chloramphenicol in mice (Smith et al., 1990b).

30

The VCH structural analogue 4-phenylcyclohexene (4-PC) did not affect small pre-antral

follicles, or plasma serum FSH levels following 30d treatment (Hooser et al., 1993).

Other structural analogues of VCH/VCM that could not form diepoxides following

microsomal incubations, such as ethylcyclohexene, vinylcyclohexane, cyclohexene,

ethylcyclohexene oxide, vinylcyclohexane oxide, and cyclohexene oxide also did not

reduce small pre-antral follicles in mice (Doerr et al., 1995). Collectively these data

suggest that VCD is the bioactive form, and VCH metabolism to VCD is necessary for

the induction of ovotoxicity.

VCH metabolism to the ovotoxicant VCD is well characterized in the liver. VCH

and VCM treatment for 10d induced microsomal CYP 2A, CYP 2B, and CYP 2E1

immunoreactive protein levels. The induction in CYP 2A and CYP 2B protein levels

corresponded to an increase in catalytic activity of these two enzymes (Doerr-Stevens et

al., 1999; Fontaine et al., 2001a). Even though hepatic CYP 2E1 immunoreactive protein

expression was increased, there was no change in catalytic activity. Additionally, there

was no difference in the formation of either VCM or VCD between CYP 2E1 wild-type

and null hepatic microsomes incubated with VCH or VCM (Fontaine et al., 2001b).

These studies suggest that hepatic CYP 2A and CYP 2B are the likely isoforms involved

in bioactivation of VCH and VCM to VCD, and that hepatic CYP 2E1 is not involved.

Species variation in VCH-induced ovotoxicity provides further evidence that

metabolism of VCH to VCD is necessary in the resulting ovotoxicity. Rats are not

susceptible to ovotoxicity induced by VCH (Smith et al., 1990c). This lack of ovarian

toxicity in rats can be explained by reduced bioactivation. Rat hepatic microsomes

31

produced less monoepoxides, the precursors of VCD. Following multiple doses of VCH

elevation in hepatic CYP 2A or CYP 2B immunoreactive protein was not observed.

Similarly, catalytic activity was not elevated, and VCD was not detected in rat

microsomes (Fontaine et al., 2001a).

Following bioactivation, VCD can be detoxified by mEH to form an inactive

tetrol (4-[1,2-dihydroxy]ethyl-1,2-dihydroxycyclohexane) metabolite (Keller et al., 1997;

Fig. 1.2). This tetrol metabolite was detected in mouse and rat plasma and urine (Salyers,

1995). The epoxides of VCD are also substrates for glutathione (GSH) conjugation. A

polar conjugate of VCD was detected in mouse and rat urine. This conjugate was not

hydrolyzed by β-glucuronidase or sulfatase, and was thought to be a GSH adduct

(Salyers, 1995). Additionally, VCM and VCD depleted hepatic GSH level within 2-4

hours following a single dose (Giannarini et al. 1981).

In summary, these studies suggest that in the liver VCH is bioactivated to the ovarian

toxicant, VCD, via CYP 2A and CYP 2B enzymes. Over several days of dosing, the

VCD which is formed then causes loss of small pre-antral (primordial and primary)

follicles leading to ovarian atrophy and ovarian failure (Hooser et al., 1994; Mayer et al.,

2002). Important to the ultimate ovotoxicity caused by VCD (formed from VCH or

administered directly) is its role of detoxification by mEH and probably GSH

conjugation. As rats are more capable of converting VCD to the tetrol than mice, they

are also more resistant to VCH/VCD ovotoxicity than mice.

32

Figure 1.2: Proposed VCD Metabolism Scheme. The parent compound, VCH, is

bioactivated by CYP 450 to form either 1,2 or 7,8-monoepoxide (VCM) metabolites.

These intermediate metabolites undergo further bioactivation by CYP 450 enzymes to

form the ultimate ovotoxic metabolite VCD. The ovotoxicant VCD is detoxified by

hydrolysis of epoxides by the microsomal epoxide hydrolase (mEH), first to 1,2-diol-7,8-

epoxide or 7,8-diol-1,2 -epoxide intermediates, and subsequently to an inactive tetrol (4-

[1,2-dihydroxy]ethyl-1,2-dihydroxycyclohexane) metabolite. Note: conjugation of 7,8-

VCM, 1,2-VCM and VCD with GSH is also likely to occur.

7,8 VCM O

O

OHO

HO

O

OO

HH

H

H

OHOH

HH

VCD

Tetrol

mEH

mEH

O

O

VCH

CYP 450

CYP 450 1,2 VCM

33

VCD-Induced Ovotoxicity

A number of studies have investigated the mechanism(s) by which the bioactive

metabolite, VCD, cause ovotoxicity in mice and rats. Studies by Flaws et al. (1994a)

showed that VCD treatment reduced primordial follicles by 33%, and primary follicles by

38% of control in adult (51d) rats. In immature rats (28d), VCD decreased primordial

and primary follicles by 19% and 45% of control respectively. In adult rats VCD did not

affect growing (secondary and antral) follicles; however, in immature rats VCD

decreased growing follicles by 54% of control. The immature rat ovary had not

developed many growing follicles; therefore, this effect was thought to be due to the

decrease in small pre-antral follicles available for recruitment. Studies by Kao et al.

(1999) have shown that 12d VCD treatment in mice results in a loss of primordial and

primary follicles, and an increase in atretic follicles. In this study, follicle loss observed

in mice was greater (p < 0.05) than that of rats. This difference in sensitivity to VCD

between mice and rats was also observed in other studies (NTP, 1989).

Long term effects of VCD include ovarian atrophy and ovarian failure. Treatment

with VCD for 15 or 24 months caused ovarian atrophy in female mice. Furthermore, in

this study, hyperplasia of the ovarian surface epithelium and granulosa cell tumors were

observed at high doses (5 and 10mg/mouse/d; NTP, 1989). VCD depleted small pre-

antral, secondary and antral follicles at 30, 60 and 120d respectively following 30d of

treatment in mice (Mayer et al., 2002). Large follicle loss was most likely due to the lack

of small pre-antral follicles available for recruitment. The loss of follicles in these mice

corresponded to a decrease in ovarian weight, irregular cyclicity, and an increase in FSH

34

levels at one year (Mayer et al., 2002). Studies have also shown that 30d VCD treatment

also decreases uterine weights in rats (Flaws et al., 1994a). Thus, ovarian failure occurs

as a consequence of VCD-induced small pre-antral follicle loss.

Follicle loss triggered by VCD has been attributed to an acceleration of the

natural atretic process (apoptosis) in the ovary. Rat small pre-antral follicles showed

signs of DNA degradation following 10 and 12d of VCD dosing. Follicles in these

ovaries have lost focal contact between the granulosa cells and oocytes (Springer et al.,

1996a). Bax (pro-apoptotic protein; increases mitochondrial permeability) expression

increased in rat small pre-antral follicle but not in larger follicle populations or liver

following 10d of VCD dosing (Springer et al., 1996b). VCD dosing also (15d) triggered

translocation of the anti-apoptotic protein Bcl-xL (blocks the release of cytochrome c

from mitochondria) from small pre-antral follicle mitochondria to the cytosol. This

resulted in an increase in the Bax/Bcl-xL ratio in the small pre-antral follicle

mitochondria (Hu et al., 2001a). Increase in the Bax/Bcl-xL ratio favors activation of the

pro-apoptotic pathway by inducing the release of cytochrome c. In addition, VCD

induced an increase in cytochrome c level in the cytosol of rat small pre-antral follicles

(Hu et al., 2001a), suggesting cytochrome c was released from the mitochondria.

Cytochrome c is thought to trigger the activation of caspases that degrade protein during

the apoptotic process. Small pre-antral follicle caspase-3 activity increased following

VCD treatment in rats. Also, in this study, both pro-caspase-3 and active caspase-3

protein expression increased with VCD treatment (Hu et al., 2001b). Taken together,

35

these studies suggest that VCD-induced follicle loss is due to activation of apoptotic

cascades in small pre-antral follicles.

In summary, VCH bioactivation to the ovotoxicant VCD by hepatic CYP 2A and

CYP 2B enzymes ultimately results in loss of small pre-antral follicles by apoptosis.

Long term effects of VCD mimic those seen in post-menopausal women, such as loss of

ovarian follicles, ovarian failure, decreases in estrogen, and increases in FSH.

Role of Metabolism in 7,12-Dimethylbenz[a]anthracene-Induced Ovotoxicity

The polycyclic aromatic hydrocarbon (PAH), 7,12-dimethylbenz[a]anthracene

(DMBA), is a widely studied model carcinogen for the induction of mammary (Russo

and Russo, 1996), skin (Diagaradjane et al., 2006) and ovarian (Kanter et al., 2006)

tumors in rodents. In addition to having carcinogenic properties, this compound has also

been shown to cause ovotoxicity in rodents (Mattison, 1979).

Human exposure to DMBA can occur from cigarette smoke, car exhaust, and

burning of organic substances such as coal, oil, wood, and rubbish. Epidemiology studies

in females have shown a correlation with smoking and early menopause. Studies have

shown that female smokers or women exposed to cigarette smoke are likely to reach

menopause earlier (44.3yr) than the average age for natural onset of menopause (49.6yr;

Everson et al., 1986; Daniell, 1978). Kaufman et al. (1980) have shown that the mean

age of menopause decreases with an increase in the numbers of cigarettes smoked per

day. Additionally, women exposed to cigarette smoke during fetal development were

36

less able to conceive (Weinberg et al., 1989; Baird and Wilcox, 1985), suggesting that

cigarette smoke can also decrease fecundity. Thus, these epidemiological studies suggest

that disruption of follicles by components of cigarette smoke, such as DMBA, can lead to

a decrease in fertility in women.

DMBA had been shown to target all follicle populations and CL in the rodent

ovary leading to a decrease in ovarian volume and premature ovarian failure. Studies by

Mattison (1979) have shown that 6d following DMBA treatment, there was a 99.4% loss

of primordial follicles in C57BL/6N mice, and a 94.4% loss in DBA/2N mice. Unlike

with mice, in rats 14d of DMBA treatment resulted in a 49.3% primordial follicle loss

compared to control, suggesting that rats are less sensitive to ovotoxic effects of DMBA.

In addition to small follicles, DMBA decreased large follicles in mice following 45d of

treatment, and CL following 4wk of treatment (Manoharan and Ramesha, 1980). In mice

this decrease in ovarian follicle populations and CL leads to a decrease in ovarian volume

and weight (Weitzman et al., 1992; Manoharan and Ramesha, 1980). The loss of

follicles induced by DMBA was followed by granulosa cell carcinoma development (Jull,

1973).

Like VCD, DMBA-induced follicle loss was shown to be due to activation of the

apoptotic cascade. Work done by Matikainen et al. (2001) has shown an increase in Bax

mRNA and protein expression in mouse ovaries following 24h of continuous DMBA

exposure. Furthermore, in this study, follicle loss was not observed in Bax null ovaries

cultured with DMBA for two days.

37

As with other ovotoxicants, DMBA is the parent form of this compound, and it

undergoes metabolism to form a 3,4-diol-1,2-epoxide metabolite (Fig. 1.3). Carcinogenic

and ovotoxic properties of DMBA have been attributed to this metabolite. Unlike

DMBA, DMBA-3,4-diol-1,2-epoxide is an electrophilic compound, containing an

epoxide that can react with nucleophiles in the cell, such as proteins and DNA.

DMBA undergoes a series of metabolic steps to form the electrophilic metabolite.

DMBA is bioactivated to a 3,4-epodixe by CYP 450 isoform 1B1, which is then

hydrolyzed by mEH to form a 3,4-diol. This intermediate metabolite further undergoes

epoxidation via CYP 450 isforms 1A1 or 1B1 to form the ultimate carcinogenic and

ovotoxic metabolite, DMBA-3,4-diol-1,2-epoxide (Fig. 1.3).

Incubation of DMBA with recombinant (rec) mouse CYP 1B1 (recCYP 1B1m)

protein (along with an excess of mEH and P450 reductase) resulted in the formation of

several diols, in which 3,4-diol formation was favored. Incubation of anti-CYP 1B1

antibody with recCYP 1B1m and DMBA prevented the formation of the 3,4-diol. In

these studies, mEH was shown to be the rate limiting step in the formation of 3,4-diol

(Savas et al., 1997; Shimada and Fujii-Kuriyama, 2004). Therefore, these studies suggest

that DMBA metabolism to DMBA-3,4-diol involves CYP 1B1 and mEH.

Studies indicate that the formation of the epoxide at the 1,2 position following the

formation of the 3,4-diol is necessary for toxicity of DMBA. This epoxide is situated in

the bay region of DMBA, and epoxides in this bay region of PAHs have been shown to

be biologically active (Sawicki et al., 1983; Vigny et al., 1985). Studies by Shimada et

al. (2001) have shown that the Km of CYP 1A1 and CYP 1B1 enzymes for DMBA-3,4-

38

diol are 4.0 and 3.0µM respectively. Incubation of DMBA-3,4-diol with recCYP 1A1

and recCYP 1B1 caused DNA damage, suggesting that both CYP 1A1 and CYP 1B1

were involved in bioactivation of the 3,4-diol to DMBA-3,4-diol-1,2-epoxide (Shimada

et al. 2001).

Further evidence suggesting that DMBA requires metabolism to induce toxicity

comes from transgenic mouse studies. CYP 1B1 null mice had a higher survival rate

compared to wild-type mice treated with DMBA. In CYP 1B1 null mice, the incidents of

skin and lymphoid tumors following DMBA treatment were less than those observed in

wild-types (Buters et al., 2003). Similarly, treatment of mEH null mice with DMBA

resulted in a decrease in skin papillomas, and number of papillomas per mouse compared

to the wild-types also treated with DMBA. Incubation of DMBA with mEH null mouse

fibroblast cells did not induce cytotoxicity. The DMBA-3,4-diol metabolite was not

detected in these cultures. On the other hand, incubation of DMBA-3,4-diol with mEH

wild-type and null fibroblast cells induced cytotoxicity in both cell types. Additionally,

DMBA- 3,4-diol metabolite was not detected in the plasma of mEH null mice treated

with DMBA (Miyata et al., 1999; 2002).

Taken together, these studies suggest that DMBA-induced toxicity requires

sequential bioactivation steps to the 3,4-diol-1,2-epoxide by CYP 1B1, mEH, and CYP

1A1/1B1 enzymes. In the ovary this metabolite targets all follicle populations via

apoptosis.

39

Figure 1.3: DMBA Metabolic Pathway. The parent compound, DMBA, is bioactivated

by CYP 450 isoform 1B1 to a DMBA-3,4-epoxide intermediate, which is hydrolyzed by

microsomal epoxide hydrolase (mEH) to form DMBA-3,4-diol. This compound further

undergoes bioactivation by either CYP 1B1 or 1A1 to form the ultimate carcinogenic and

ovotoxic metabolite, DMBA-3,4-diol-1,2-epoxide. Adapted from Miyata et al. (1999).

40

CHAPTER 2

GENERAL OVERVIEW AND METHODS

Statement of Problem

Some ovotoxic compounds, such as the industrial chemical 4-vinylcyclohexene

(VCH) and polycyclic aromatic hydrocarbon (PAH) 7,12-dimethylbenz[a]anthracene

(DMBA), require biotransformation in the organism to reactive intermediate metabolites

to induce ovotoxicity. VCH and DMBA are bioactivated by cytochrome P450 (CYP

450) isoforms and microsomal epoxide hydrolase (mEH) to their ovotoxic metabolites. It

has been presumed that liver is the primary organ bioactivating VCH and DMBA to

reactive intermediates. However, studies suggest that extra-hepatic organs, such as the

ovary, are capable of participating in the metabolic process. Mouse ovary expresses

catalytically active CYP 450 isoforms and mEH. Studies also show that the ovary has the

capacity to detoxify ovotoxicants via glutathione conjugation.

Although previous studies indicate that the ovary is capable of metabolizing VCH

and DMBA, the direct role of ovarian metabolism in the resulting ovotoxicity has not

been studied. Therefore, studies were designed to evaluate the role of ovarian

metabolism in VCH/VCM and DMBA-induced ovotoxicity using a novel in vitro mouse

ovarian culture system. This ovarian culture system eliminates the metabolic role of the

liver, and allows evaluation of direct ovarian metabolic capacities.

41

Research Objectives

The working hypothesis of this dissertation is that ovarian metabolism plays a role

in ovarian toxicity induced by ovotoxicants such as VCH/VCM and DMBA. The major

objectives to test this hypothesis are described below.

1) The first objective was to characterize the involvement of CYP 450 enzyme

isoform CYP 2E1 in bioactivation of VCH and VCM to the ovarian toxicant VCD.

CYP 2A and 2B bioactivate VCH in the liver; however, a variety of observations support

a role for CYP 2E1 in ovarian bioactivation of VCH to VCD. In vivo VCH dosing

increased Cyp 2E1 mRNA level in small pre-antral follicles (target population of VCH

and VCD), and CYP 2E1 catalytic activity in the whole ovary. Thus, unlike the liver, the

CYP 2E1 isoform appears to bioactivate VCH and VCM in the ovary. It is hypothesized

that ovarian CYP 2E1 is required for ovarian bioactivation of VCH/VCM as evident by

resultant ovotoxicity. For this study ovaries from PND 4 and 28d female CYP 2E1 wild-

type and CYP 2E1 null mice were exposed to VCH, VCM and VCD for 15d. Following

exposure follicle populations were classified and counted. Loss of follicles was used as

an indication of bioactivation occurring in these ovaries.

2) The second objective was to characterize the ability of the ovary to detoxify the

ovotoxicant VCD via GSH conjugation. The two epoxide moieties on VCD are

possible substrates for GSH conjugation to form inactive metabolites. In vivo studies

indicate that VCD can be detoxified via conjugation with a polar metabolite that was

presumed to be GSH. The ovary synthesizes GSH, and the amount of GSH in the ovary

42

has been shown to be regulated by the ovarian cycle. It is hypothesized that VCD can be

detoxified by GSH conjugation in the mouse ovary. In this study VCD-GSH conjugate

formation in 28d B6C3F1 female mice dosed with VCD, and PND4 CYP 2E1 wild-type

and null ovaries incubated with VCD for 15d was evaluated using liquid

chromatography/mass spectrometry (LC/MS).

3) The third objective was to further characterize the role of ovarian metabolism

leading to ovarian toxicity using an environmentally relevant compound, DMBA.

CYP 450 enzymes and mEH have been implicated in the bioactivation of PAHs such as

DMBA to their ovotoxic forms. DMBA is bioactivated to the ovotoxicant 3,4-diol-1,2-

epoxide by CYPs 1B1 and 1A1 and mEH. mEH is present in the ovary and has been

shown to be catalytically active. Therefore, it is hypothesized that ovarian mEH

bioactivates DMBA to the ovarian toxicant leading to ovarian toxicity. For this study

ovaries from PND4 B6C3F1 mice were cultured with various concentrations of DMBA

for various time points. Following culture, follicle numbers were utilized as an indication

of bioactivation occurring in the ovary. mEH mRNA and protein were also evaluated

following exposure to DMBA via real time PCR and confocal microscopy.

4) The fourth objective was to compare the difference between ovotoxicity

induced by VCD and that by DMBA using CYP 2E1 wild-type and null ovaries in

culture. Preliminary studies show that VCD-induced toxicity differs between CYP 2E1

wild-type and null mouse ovaries, with CYP 2E1 null ovaries being less sensitive to the

43

effects of VCD. VCD is hydrolyzed by mEH, and is not a substrate for CYP 2E1.

Therefore, this difference in toxicity between CYP 2E1 wild-type and null ovaries might

be due to a difference in mEH expression. Since VCD is not a substrate for CYP 2E1, it

is hypothesized that there will be no difference in ovotoxicity induced by VCD or DMBA

between CYP 2E1 wild-type and null ovaries. For this study ovaries from PND4 CYP

2E1 wild-type and null mice were cultured with various concentrations of VCD and

DMBA. Following culture, difference in loss of follicles between CYP 2E1 wild-type

and null ovaries were utilized as an indication of difference in mEH expression.

Materials and Methods

Reagents. VCH (racemic mixture; purity 99%), 1,2-VCM (mixture of isomers; purity

98%), VCD (mixture of isomers; purity 97%), sesame oil, DMBA, cyclohexene oxide

(CHO), L-glutathione reduced (GSH), carbamazepine (CBZ), bovine serum albumin

(BSA), ascorbic acid (Vitamin C), and transferrin were purchased from Sigma-Aldrich

Inc. (St Louis, MO). Dulbecco’s Modified Eagle Medium: nutrient mixture F-12 (Ham)

1X (DMEM/Ham’s F12), Albumax, penicillin/streptomycin (5000U/ml, 5000µg/ml,

respectively), Hanks’ Balanced Salt Solution (without CaCl2, MgCl2, or MgSO4), mEH,

GST and β-actin custom designed primers, and Superscript III were obtained from

Invitrogen Co. (Carlsbad, CA). Millicell-CM filter inserts were purchased from

Millipore (Bedford, MA), and 48 well cell culture plates were obtained from Corning Inc.

44

(Corning, NY). 8% pure LC/MS grade formic acid (FA) was purchased from Fluka

(Switzerland). Acetonitrile, high purity solvent for HPLC analysis (ACN) was purchased

from Honeywell Burdick and Jackson (Morristown, NJ). The mEH antibody (goat anti-

rabbit) was obtained from Detroit R and D (Detroit, MI). Secondary antibody (horse

anti-goat) and Cy-5-streptavidin were obtained from Vector (Burlingame, CA). YOYO-1

was purchased from Molecular Probes (Eugene, OR). RNeasy Mini kit, QIAshredder kit,

RNeasy MinElute kit, and QuantitectTM SYBR Green PCR kit were purchased from

Qiagen Inc. (Valencia, CA). RNAlater was obtained from Ambion Inc. (Austin, TX).

Animals. CYP 2E1 wild-type (+/+; 129S1/SvImJ background strain; 2 females, 2 males)

and null (-/-; 2 females, 2 males) mice were purchased from Jackson Laboratories (Bar

Harbor, ME) and bred at the University of Arizona’s Animal Care Facility. Late

gestation day pregnant mice C57Bl6 (carrying B6C3F1 litters) and 21d B6C3F1 female

mice were purchased from Harlan Laboratories (Indianapolis, IN). All animals were

housed in plastic cages and maintained in a controlled environment (22 ± 2°C; 12h light/

12h dark cycles). The animals were provided standard diet with ad libidum access to

food and water. 129S1/SvImJ mice breeding pairs were housed one male and one female

per cage, and pups were weaned 4 per cage at 21d by sex. Pregnant C57Bl6 females

were housed one per cage, and 21d B6C3F1 females were housed 4 per cage. All animal

experiments were approved by the University of Arizona’s or Amgen Inc.’s Institutional

Animal Care and Use Committee.

45

In Vitro Ovarian Cultures. On postnatal day (PND) 4 female B6C3F1, CYP 2E1 wild-

type (+/+), and CYP 2E1 null (-/-) mice were killed by CO2 inhalation followed by

decapitation. Each ovary was removed, oviduct and excess tissue was trimmed, and the

ovary was placed on a piece of Millicell-CM filter membrane floating on 250µl of

DMEM/Ham’s F12 medium containing 1mg/ml BSA, 1mg/ml albumax, 50µg/ml

ascorbic acid, 5U/ml penicillin/5µg/ml streptomycin, and 27.5µg/ml transferrin in a well

in a 48 well plate previously equilibrated to 37°C. Using fine forceps a drop of medium

was placed to cover the top of the ovary to prevent drying (Fig. 2.1). VCD (5, 10, 15, 20,

25, or 30µM) or 1,2-VCM (125, 250, 500, 750, or 1000µM) diluted in medium was

added to each well. No treatment was added to control wells. For the DMBA study,

ovaries were incubated with 1% DMSO (vehicle control), DMBA (12.5nM - 1µM), 2mM

CHO, or 2mM CHO + 1µM DMBA for 6h-15d as indicated in figure legends. Plates

containing ovaries were cultured at 37°C and 5% CO2 in air. For those cultures lasting

more than 2d, 200µl of media were removed combined with 200µl of 3% FA in

methanol, and stored at -20oC. Remaining media were removed from each well and fresh

media and treatment were added.

46

Figure 2.1: In Vitro Neonatal Whole Ovary Culture System. Postnatal day (PND) 4

female mice were killed by CO2 inhalation followed by decapitation. Each ovary was

removed, oviduct and excess tissue trimmed, and placed on a piece of Millicell-CM

membrane floating on 250µl of DMEM/Ham’s F12 medium containing 1mg/ml BSA,

1mg/ml Albumax, 50µg/ml ascorbic acid, 5U/ml penicillin/5µg/ml streptomycin, and

27.5µg/ml transferrin in a well in a 48 well plate previously equilibrated to 37°C. Using

fine forceps a drop of medium was placed to cover the top of the ovary to prevent drying.

Trim Ovary

Place in Culture

Excise ovaries from mouse

Membrane Ovary

Media

47

In Vivo Animal Dosing. Age day 28 (28d) female offspring of both CYP 2E1 wild-type

(+/+) and CYP 2E1 null (-/-) mice, and B6C3F1 females were dosed daily (15d; i.p.) with

sesame oil (vehicle control) or sesame oil containing VCH (7.4 mmol/kg/d), 1,2-VCM

(2.74mmol/kg/d), or VCD (0.57 mmol/kg/d). Previously it was determined that those

levels of each chemical provided maximal reduction in small pre-antral follicles in

B6C3F1 female mice following 30d of daily i.p. dosing (Smith et al., 1990a). Animals

were killed by CO2 inhalation 4h, 24h, or 4h following 15d of daily dosing.

Histological Evaluation. On 15d of culture in vitro treated ovaries (5 ovaries/treatment)

were fixed in Bouin’s for 1.5h, transferred to 70% ethanol, embedded in paraffin, serially

sectioned (5µm thick), and every 6th section was mounted. 4h following the final in vivo

dose, animals (5/group) were euthanized, ovaries removed, and oviduct and excess fat

trimmed. In vivo treated ovaries (one ovary per animal) were placed in Bouin’s fixative

for 2h, transferred to 70% ethanol, embedded in paraffin, serially sectioned (5µm thick),

and every 20th section was mounted. All ovarian sections were stained with hematoxylin

and eosin (H and E). Healthy follicle populations containing oocytes were classified and

counted in every 12th (in vitro group) and 20th (in vivo group). Previously it was

established that counting of every 12th (in vitro group) and 20th (in vivo group) sections

avoids repetitive counting of the same follicle (Devine et al., 2002a,b; Flaws et al., 1994).

Unhealthy follicles were distinguished from healthy follicles by pyknosis of granulosa

cells and intense eosinophilic staining of oocytes (Devine et al., 2002b). Follicle

48

population classification was according to the procedure of Flaws et al. (1994a) which

was adapted from that described by Pedersen and Peters (1968). Briefly, primordial

follicles contained the oocyte surrounded by a single layer of squamous-shaped granulosa

cells, primary follicles contained the oocyte surrounded by a single layer of cuboidal-

shaped granulosa cells, and secondary follicles contained the oocyte surrounded by

multiple layers of granulosa cells. Total follicle loss for Fig. 5.4 was calculated by

subtracting the total number of follicles (primordial + primary + secondary) remaining

following DMBA treatment at each time point from total number of follicles in vehicle

control treated group at that time point. These values were divided by the follicle

numbers present in control ovaries, and the resulting value was multiplied by 100 to

obtain a percentage; (Control-DMBA)/Controlx100.

Toluidine Blue Staining. This histological method was adapted from Tome et al. (2001).

Briefly, following incubation ovaries were fixed in 3% glutaraldehyde in 0.1M

cacodylate (pH = 7.2) for 1.5h, transferred to 1mM cacodylate, embedded in epoxy resin,

serially sectioned (1µm thick), and every 10th section was mounted. All ovarian sections

were stained with toluidine blue for observation of pyknotic nuclei as a marker for

apoptosis.

49

RNA Isolation. Following 3h, 6h, 24h, and 2d of in vitro culture, ovaries (12/pool)

treated with vehicle control (1% DMSO) or DMBA (1µM) were stored in RNAlater at

–80°C. Total RNA was isolated using RNeasy Mini kit. Briefly, ovaries were lysed and

homogenized using a motor pestle followed by applying the mixture onto a QIAshredder

column. The QIAshredder column containing ovarian tissue sample was then centrifuged

at 14,000rpm for 2min. The resulting supernatant was applied to an RNeasy mini

column, allowing the RNA to bind to the filter cartridge. Following washing, the RNA

was eluted from the filter, and concentrated using RNeasy MinElute kit. Briefly, isolated

RNA was applied to an RNeasy MinElute spin column, and after washing, RNA was

eluted using 14µL of RNase-free water. RNA concentration was determined using a

NanoDrop (λ = 260/280nm; ND 1000).

First Strand cDNA Synthesis and Real Time Polymerase Chain Reaction (PCR). Total

RNA (1µg) was reverse transcribed into cDNA utilizing the Superscript III Reverse

Transcription System. cDNA was diluted (1:10) in RNase-free water. Two microlitters

of diluted cDNA was amplified on a Rotor-Gene 3000 using QuantitectTM SYBR Green

PCR kit and custom designed primers for mEH, GST, and β-actin (Table 2.1). The

regular cycling program consists of a 15min hold at 95°C and 45 cycles of: denaturing at

95°C for 15s, annealing at 58°C for 15s, and extension at 72°C for 20s at which point

data were acquired. Product melt conditions were determined using a temperature

gradient from 72°C to 99°C with a 1°C increase at each step.

50

Table 2.1: Custom Designed Primer Sequences for mEH, GST and β-actin.

Gene Forward Primer 5’ to 3’

Reverse Primer 5’ to 3’

Reference

mEH GGG TCA AAG CCA TCA GGC A

CCT CCA GAA GGA CAC CAC TTT

Cannady et al.,2002

GST CTG CAG CAG GGG TGG A

CTC TCT CCT TCA TGT CCT TCC

NCBI Genbank accession # AK002712

β-actin ACG CAG CTC AGT AAC AGT CC

TCT ATC CTG GCC TCA CTG TC

NCBI Genbank accession # AK151010

Confocal Microscopy. Following 6h of in vitro culture, ovaries (3/ group) treated with

vehicle control or DMBA (1µM) were fixed in 4% buffered formalin for 2h, transferred

to 70% ethanol, embedded in paraffin, serially sectioned, and every 10th section was

mounted. Sections were deparaffinized (approximately 10 sections/ovary) and incubated

with primary antibody directed against mEH (goat anti-rabbit; 1:50 dilution) at 4°C

overnight. Specificity for this antibody was determined in previous studies (Cannady et

al., 2002). Secondary biotinylated antibody (horse anti-goat; 1:75 dilution) was applied

for 1h, followed by CY-5-streptavidin (1h; 1:50 dilution). Sections were treated with

Ribonuclease A (100µg/ml) for 1h, followed by staining with YOYO-1 (10 min; 5nM).

Slides were repeatedly rinsed with phosphate buffer saline (PBS), cover-slipped, and

stored in the dark (4°C) until visualization. Primary antibody was not added to immuno-

negative ovarian sections. Immunofluorescence was visualized on a Zeiss (LSM 510

NLO-Meta) confocal microscope with an argon and helium-neon laser projected through

51

the tissue into a photomultiplier at λ = 488 and 633 nm for YOYO-1 (green) and CY-5

(red), respectively. All images were captured using a 40 X objective lens. Multiple

readings were taken throughout the sections. Analysis was performed at control settings

on the confocal microscope, in which 110 follicles were evaluated/ovary.

Reactivity of VCD with GSH. 1µM VCD was incubated with 1mL of 10mM GSH in

0.1M potassium phosphate buffer (pH 7.4) at 37oC for 0.5-24h. Following incubation,

the reaction was stopped by adding 1ml of 3% FA in 0.1µM CBZ in ACN. 25µL of the

reaction was injected on a Luna 5u C18(2) 100A, 50 X 2mm 5µm column, and analyzed

using liquid chromatography/mass spectrometry (LC/MS; Thermofinagen Quantum

Discovery Tandem MS; Triple Quad) on Positive Ion Scan setting (Table 2.2). Peak area

of VCD, VCD-monoGSH, VCD-diGSH at retention time (RT) of 7, 5, and 4min

respectively was quantified by normalizing to the peak area of the internal standard CBZ.

Synthesis and Purification of VCD-Mono and DiGSH Adducts. VCD-monoGSH adduct

was synthesized by reacting 10mM VCD with 1mM GSH in 0.1M potassium phosphate

buffer (pH 7.4) at 37°C overnight. VCD-diGSH adduct was synthesized by reacting

1mM VCD with 10mM GSH in 0.1M potassium phosphate buffer (pH 7.4) at 37°C

overnight. Both reactions were stopped by rotovapping the excess water for 30min at

40°C followed by extracting the un-reacted VCD using ethyl acetate. Following

52

extraction, high-performance liquid chromatography (HPLC; HP 1050 Series) was used

to purify VCD-GSH adducts from un-reacted GSH. A total of 2mL from each reaction

was injected (100uL injection volume per run) on a Phenomenex Synergi 4u Fusion RP,

size: 250 X 10.00mm 4µm column, and fractions were collected every 30s and monitored

at a wavelength of 214nm. See Table 2.3 for HPLC gradient conditions used. 25µL of

each fraction was then injected into LC/MS (Table 2.2), and analyzed using Positive Ion

Scan. Fractions containing the highest concentration of VCD mono and di-GSH adduct

from each reaction was then re-purified using a Seppac column. These fractions were re-

analyzed using LC/MS via Positive Ion Scan (Table 2.2), and tandem MS (MS/MS) was

performed to detect the most abundant fragments to be used for Selective Reaction

Monitoring (SRM) analysis of biological samples.

53

Table 2.2: HPLC Gradient Used in LC/MS Detection of VCD-GSH Adducts.

Time (min) 100% ACN 0.1% FA in Water Flow (ml/min) 0 5% 95% 0.300 12 100% 0% 0.300 13 100% 0% 0.300

13.01 5% 95% 0.300 17 5% 95% 0.300

17.01 5% 95% 0.300

Table 2.3: HPLC Gradient Used in VCD-GSH Adduct Purification.

Time (min) 100% ACN 0.1% FA in Water

Flow (ml/min) Pressure (Bar)

0 5% 95% 4.000 4002 50% 50% 4.000 40010 95% 5% 4.000 400 11 5% 95% 4.000 400 15 5% 95% 4.000 400

Table 2.4: SRM Conditions for VCD-Mono and DiGSH Adduct Detection.

Compound Parent (m/z) Product (m/z) Collision Energy

CBZ 237 194 25V

VCD-monoGSH 448 176 30V

VCD-diGSH 378 176 27V

VCD-diGSH 755 479 30V

54

Extraction of VCD-GSH Adduct from Liver and Ovarian Tissue. 4 or 24h following the

final in vivo dose, animals (5/group) were euthanized, liver and ovaries removed.

Oviduct and excess fat was trimmed from ovaries. Liver and ovarian tissues were frozen

at -80°C until use. In vivo treated ovaries (10 ovaries per treatment per time point; n= 1

pool of 10 ovaries) and livers (1mg piece of liver per animal from each treatment at each

time point; n= 5 livers) were homogenized on ice in 0.1M potassium phosphate buffer

(pH 7.4). An equal volume of ACN was added for a final concentration of 1mg/ml and

re-homogenized. 3% FA in 0.1µM CBZ in ACN was added to the homogenate (1:1),

centrifuged at 14,000 rpm for 5min, 25µL of the supernatant was injected onto a

Phenomenex, Luna 5u C18(2) 100A, 50 X 2mm 5µm column, and analyzed using

LC/MS (Table 2.2). VCD-GSH adduct content was detected using SRM (see Table 2.4

for product m/z ratio and collision energy used for SRM).

Detection of VCD-GSH Adduct from Ovarian Culture Media. 200µL of medium stored

in 3% FA in methanol was removed. 200µL of 0.1µM CBZ in ACN was added to the

medium. Media were centrifuged at 14,000rpm for 5min, 25µL of supernatant was

injected onto a Phenomenex, Luna 5u C18(2) 100A, 50 X 2mm 5µm column, and

analyzed using LC/MS (Table 2.2). VCD-GSH adduct content was detected using SRM

(Table 2.4).

55

Statistical Analysis. Comparisons were made using one-way analysis of variance

(ANOVA). When significant differences were detected, individual groups were

compared with the Fisher’s protected least significant difference (PLSD) multiple range

test. The assigned level of significance for all tests was p < 0.05.

56

CHAPTER 3

OVARIAN BIOACTIVATION OF 4-VINYLCYCLOHEXENE METABOLITES TO

THE OVARIAN TOXICANT 4-VINYLCYCLOHEXENE DIEPOXIDE: THE

INVOLVEMENT OF CYP 2E1 ENZYME

Abstract

4-vinylcyclohexene (VCH) is bioactivated by hepatic CYP 2A and 2B to a

monoepoxide (VCM), and subsequently to an ovotoxic diepoxide metabolite (VCD).

Studies suggest that the ovary can directly bioactivate VCH via CYP 2E1. The current

study was designed to evaluate the role of ovarian CYP 2E1 in VCM-induced

ovotoxicity. Postnatal day 4 B6C3F1, and CYP 2E1 wild-type (+/+) and null (-/-) mouse

ovaries were cultured (15d) with VCD (30µM), 1,2-VCM (125-1000µM), or vehicle.

28d female CYP 2E1 +/+ and -/-mice were dosed daily (15d; ip) with VCH, 1,2-VCM,

VCD or vehicle. Following culture or in vivo dosing, ovaries were prepared for

histological evaluation of follicles. In culture, VCD decreased (p < 0.05) primordial and

primary follicles in all three groups of mouse ovaries compared to controls. 1,2-VCM

decreased (p < 0.05) primordial follicles in B6C3F1 and CYP 2E1 +/+ ovaries, but not in

CYP 2E1 -/- ovaries in culture. 1,2-VCM did not affect primary follicles from any group

of mouse ovaries. Following in vivo dosing, VCD, VCM, and VCH reduced (p < 0.05)

primordial and primary follicles in CYP 2E1 +/+ mice. In CYP 2E1 -/- mice VCD and

VCM reduced primordial and primary follicles, whereas, the parent compound VCH only

reduced (p < 0.05) primary follicles. Collectively, these data demonstrate that in vitro

57

ovarian bioactivation of VCM requires CYP 2E1 enzyme; however, in vivo CYP 2E1

plays a minimal role. This study also demonstrates the use of a novel ovarian culture

system to evaluate ovary-specific metabolism of xenobiotics.

Introduction

Exposure to the occupational chemical, 4-vinylcyclohexene (VCH), results in a

loss of small pre-antral (primordial and primary) follicles in the mouse ovary (NTP,

1986; Smith et al., 1990b). VCH-induced ovarian toxicity has been attributed to the

bioactivation of this compound to a diepoxide metabolite (VCD). VCH is bioactivated in

the liver to either a 1,2 or 7,8-monoepoxide (VCM) and subsequently to VCD via

cytochrome P450 (CYP 450) isoforms 2A and 2B (Fig. 1.2). Structure activity studies

suggest that bioactivation to the diepoxide is necessary for VCH-induced ovarian toxicity

(Doerr et al., 1995). Findings support that liver is likely to be the major site of

bioactivation of VCH to VCD, for subsequent circulation to the ovary for follicle

destruction (Doerr-Stevens et al., 1999; Smith et al., 1990b).

Even though the liver might be the major bioactivation site for VCH, studies by

Cannady et al. (2003) suggest that the mouse ovary also has the capacity to bioactivate

VCH. In vivo VCH dosing increased levels of mRNA encoding Cyp 2B in ovarian small

pre-antral follicles (those targeted by VCD). Additionally, in vivo VCD dosing increased

mRNA encoding Cyp 2A and Cyp 2E1 in those same follicles. CYP 2A, CYP2B and

CYP 2E1 protein expression was detected in all ovarian compartments. CYP 2B and

58

CYP 2E1 enzymes were shown to be catalytically active in the B6C3F1 mouse ovary.

However, unlike liver, following in vivo exposure to VCH, CYP 2E1 catalytic activity in

the ovary was increased, while CYP 2B activity did not change. Collectively, these

studies suggest a potential role for ovarian CYP 2E1 in metabolism of VCH and VCM to

the ovotoxic metabolite VCD.

This study was designed to more directly evaluate the role of ovarian CYP 2E1

enzyme in the ovotoxicity that might result from bioactivation of VCH and VCM using a

neonatal mouse whole ovary culture system, where the metabolic role of the liver was not

involved (Devine et al., 2002a, 2004). The hypothesis is that ovarian CYP 2E1 enzyme

can participate in bioactivation of VCH and VCM to the ovotoxicant VCD.

Experimental Methods

On postnatal day (PND) 4, ovaries from B6C3F1, CYP 2E1 wild-type (+/+), and

CYP 2E1 null (-/-) mice were cultured with control medium (no treatment), VCD or

VCM for 15d. Age 28d female CYP 2E1 wild-type (+/+) and CYP 2E1 null (-/-) mice

were dosed daily (15d; i.p.) with sesame oil (vehicle control) or sesame oil containing

VCH, 1,2-VCM, or VCD. Following in vitro cultures and in vivo dosing, ovaries were

processed for histological evaluation of follicle numbers. All methods are described in

detail in Chapter 2.

59

Results

Strain dependence on VCD-induced follicle loss

To demonstrate the ovotoxic effects of VCD in a neonatal mouse whole ovary

culture system, follicle loss was evaluated in PND4 B6C3F1, CYP 2E1 wild-type (+/+),

and null (-/-) mouse ovaries cultured with 30µM VCD (Fig. 3.1). Incubation with VCD

for 15d essentially depleted (p < 0.05) primordial and primary follicles compared to

medium control. Unlike primordial follicles, 30µM VCD did not deplete all healthy

primary follicles following 15d of culture. When comparing the three groups of mouse

ovaries used, primary follicles in B6C3F1 and CYP 2E1 null (-/-) ovaries were less

sensitive to the effects of VCD compared to that of CYP 2E1 wild-type (+/+) ovaries

(Fig. 3.1B).

Follicle populations in the PND4 ovary are all primordial and primary, secondary

follicles are rarely observed, and antral follicles are not seen. Thus, the potential effect of

VCD on larger follicle populations could not be evaluated.

60

VCM-induced follicle loss in B6C3F1 ovaries

To investigate the potential for ovarian bioactivation of 1,2-VCM in the absence

of hepatic tissue, follicle loss was evaluated in PND4 B6C3F1 mouse ovaries cultured

with increasing concentrations of 1,2-VCM (Fig. 3.2). Relative to control, primordial

follicles were reduced (p < 0.05) in ovaries incubated with 750 and 1000µM 1,2-VCM

(Fig. 3.2A). Unlike VCD, incubation with 1,2-VCM at any concentration did not affect

primary follicles in B6C3F1 ovary cultures (Fig. 3.2B).

VCM-induced follicle loss in CYP 2E1 wild-type and null ovaries

To investigate the potential for CYP 2E1 enzyme in ovarian bioactivation of 1,2-

VCM to the ovotoxicant VCD, follicle loss was evaluated in PND4 CYP 2E1 wild-type

(+/+) and null (-/-) mouse ovaries cultured with increasing concentrations of 1,2-VCM

for 15d (Fig. 3.3). Relative to control, in CYP 2E1 wild-type (+/+) ovaries incubated

with ≥ 500µM 1,2-VCM primordial follicle loss (p < 0.05) was seen. Conversely,

primordial follicle numbers in CYP 2E1 null (-/-) mouse ovaries were not affected by

1,2-VCM incubation at any concentration (Fig. 3.3A). There was no effect of 1,2-VCM

on primary follicles in ovaries from CYP 2E1 wild-type (+/+) or null (-/-) mice (Fig.

3.3B).

61

Figure 3.1: VCD-Induced Ovotoxicity in Various Strains of Mouse Ovaries in

Culture. Ovaries from PND4 B6C3F1, CYP 2E1 wild-type (+/+), or CYP 2E1 null (-/-)

mice were cultured with medium control or 30µM VCD for 15d. Following incubation,

ovaries were collected, and processed for histological evaluation as described in materials

and methods. Healthy (A) primordial, and (B) primary follicles were classified and

counted. Values are mean ± SE total follicles counted/ovary, n=5; different letters differ

(p<0.05) from one another within each follicle type.

62

Figure 3.2: 1,2-VCM–Induced Ovotoxicity in B6C3F1 Ovarian Cultures. Ovaries

from PND4 B6C3F1 mice were collected and cultured for 15d with control medium, or

1,2-VCM (125, 250, 500, 750, or 1000 µM). Following incubation, ovaries were

collected, and processed for histological evaluation as described in materials and

methods. Healthy (A) primordial, and (B) primary follicles were classified and counted.

Values are mean ± SE total follicles counted/ovary, n=5; different letters differ (p<0.05)

from one another within each follicle type.

63

Effect of VCM and VCD on ovarian morphology in CYP 2E1 wild-type and null ovaries

Fig. 3.4 shows the effect of VCD (30µM) and 1,2-VCM (1000µM) on

morphology of ovaries from CYP 2E1 wild-type (+/+) and null (-/-) mice. Following

incubation with VCD, all primordial and numerous primary follicles appeared unhealthy

as characterized by intense eosin staining in oocytes. This follicle damage was

equivalent between CYP 2E1 wild-type (+/+) and null (-/-) incubations (Fig. 3.4E and F)

compared to that of vehicle control (Fig. 3.4A and B). Furthermore, in those ovaries

incubated with VCD, follicles have shrunk and contain predominantly follicular debris

compared to those ovaries in control incubations.

Compared to VCD, following incubation with 1000µM 1,2-VCM there were

fewer unhealthy primordial and primary follicles in the CYP 2E1 wild-type (+/+) ovary

(Fig. 3.4C). Compared to control incubations, there was no difference in the

morphological appearance of the CYP 2E1 null (-/-) ovary treated with 1000µM 1,2-

VCM (Fig. 3.4B vs. Fig. 3.4D).

64

Figure 3.3: Involvement of CYP 2E1 in 1,2-VCM-Induced Ovotoxicity in Ovarian

Cultures. Ovaries from PND4 CYP 2E1 wild-type (+/+) or null (-/-) mice were cultured

with medium control or 1,2-VCM (125, 250, 500, 750, or 1000 µM) for 15d. Following

incubation, ovaries were collected, and processed for histological evaluation as described

in materials and methods. Healthy (A) primordial, and (B) primary follicles were

classified and counted. Values are mean ± SE total follicles counted/ovary, n=5;

different letters differ (p<0.05) from one another within each follicle type.

65

Figure 3.4: Effect of In Vitro 1,2-VCM and VCD Exposure on Ovarian Morphology.

Ovaries from PND4 CYP 2E1 wild-type (+/+; A,C and E) and null (-/-; B,D and F) mice

were cultured with control medium (A and B), 1,2-VCM (1000 µM; C and D), or 30µM

VCD (E and F) for 15d. Following incubation, ovaries were collected, and processed for

histological evaluation as described in materials and methods.

66

Effect of VCH, 1,2-VCM, VCD dosing on follicle loss in CYP 2E1 wild-type and null mice

To evaluate the significance of ovarian bioactivation in the whole animal, ovarian

follicle loss was evaluated in CYP 2E1 wild-type (+/+) and null (-/-) mice following

repeated in vivo exposure to VCH, 1,2-VCM, VCD, or vehicle control (15 daily doses;

Fig. 3.5). Compared to vehicle control, primordial follicles (VCD target population)

were significantly reduced (p < 0.05) in ovaries from CYP 2E1 wild-type (+/+) and null

(-/-) mice following VCD exposure. As compared with VCD, 1,2-VCM exposure also

resulted in a significant loss (p < 0.05) of primordial follicles in CYP 2E1 wild-type (+/+)

and null (-/-) mouse ovaries. VCH exposure resulted in a significant loss (p < 0.05) of

primordial follicles in CYP 2E1 wild-type (+/+) mice; however, VCH exposure did not

decrease (p = 0.09) primordial follicles in CYP 2E1 null (-/-) mice (Fig. 3.5A).

Primary follicles (VCD target population) were reduced (p < 0.05) in CYP 2E1

wild-type (+/+) and null (-/-) mouse ovaries with in vivo VCD, VCM, and VCH exposure

compared to vehicle controls (Fig. 3.5B).

Secondary and antral follicle populations (VCD non-target populations) were not

affected by VCD, VCM, and VCH exposure compared to vehicle controls in either CYP

2E1 wild-type (+/+) or null (-/-) mice (Fig. 3.5C and D).

67

Figure 3.5: Involvement of CYP 2E1 in VCH–Induced Ovotoxicity. Female CYP

2E1 wild-type (+/+) and null (-/-) mice were treated with repeated daily doses (i.p.) of

sesame oil, VCH, 1,2-VCM, or VCD for 15d. Four hours following the final dose

ovaries were collected, and processed for histological evaluation as described in materials

and methods. Healthy (A) primordial, (B) primary, (C) secondary, and (D) antral

follicles were classified and counted. Values are mean ± SE total follicles counted/ovary,

n=5; different letters differ (p<0.05) from one another within each follicle type.

68

Discussion

Previous studies indicate that VCH metabolism to VCM and to the subsequent

ovotoxic metabolite VCD is primarily hepatic. Hepatic CYP 2B and 2A isoforms

catalyze both bioactivation steps, while CYP 2E1 enzyme in liver appears to not be

required (Doerr-Stevens et al., 1999; Fontaine et al., 2001a, b). However, in the mouse

ovary CYP 2E1 may play a role in VCH/VCM bioactivation (Cannady et al., 2003).

CYP 2A, 2B and 2E1 expression was detected throughout the ovary. In vivo VCH dosing

induced CYP 2E1 catalytic activity, while CYP 2B activity did not change. Specific

activity for CYP 2A was not detected in the mouse ovary (Cannady et al., 2003).

The current study was designed to further investigate a potential role of ovarian

CYP 2E1 in bioactivation of VCM utilizing a novel ovarian culture system. Using this

culture system the role of ovarian metabolism in the resulting ovotoxicity can be

evaluated independent of hepatic contributions. In a previous report, VCD caused loss of

primordial and primary follicles in neonatal rat ovaries cultured using this system

(Devine et al., 2002a, 2004). However, ovotoxicity in the culture system using ovaries

from neonatal mice has not been studied. Thus, the bioactive form, VCD, was used to

validate the use of this culture system. In neonatal rat ovarian cultures unhealthy follicles

were observed, characterized by intense eosinophilic staining of oocytes and pyknosis of

granulosa cells (Devine et al., 2002b). Unlike the ovary in vivo, in which circulating

blood can remove cellular debris, cellular remnants remain in ovaries incubated in vitro.

Therefore, in this study only healthy follicles were classified and counted. Incubation

with 30µM VCD decreased healthy small pre-antral follicles (primordial and primary) in

69

cultured ovaries from neonatal B6C3F1 mice agreeing with previous in vivo data (NTP,

1989; Chhabra et al., 1990; Kao et al., 1999; Smith et al., 1990a). Additionally, 30µM

VCD also depleted small pre-antral follicles in ovaries from neonatal CYP 2E1 wild-type

(+/+) and null (-/-) mice. These data are in agreement with previous incubations of

ovaries from PND4 rats with VCD, in which a reduction (p < 0.05) in small pre-antral

follicles was seen (Devine et al., 2004). Interestingly, unlike the three groups of mouse

ovaries utilized in this study, in which healthy primordial follicles were depleted

following 15d of incubation with VCD, not all primordial follicles were damaged in rat

ovaries. This reduced sensitivity of rats to VCD was also observed in vivo, where the

ED50 for small follicle loss was 2-3 times higher in rats compared to mice (Smith et al.,

1990b). In the current in vitro study with mice as well as in the in vitro study with rats

(Devine et al., 2004), VCD had less effect on primary follicles than on primordial

follicles. These observations suggest that primary follicles are less sensitive to VCD-

induced toxicity. The difference in susceptibility to VCD between species and follicle

populations are consistent with findings in mice and rats following in vivo dosing (Kao et

al., 1999).

To determine whether the ovarian culture system can be used to study the role of

ovarian bioactivation in the absence of hepatic tissue, ovotoxicity induced by the

proximate-ovotoxicant 1,2-VCM in B6C3F1 ovaries was evaluated. B6C3F1 mouse

ovaries were utilized, because previous studies characterizing VCH/VCM-induced

ovotoxicity were conducted in this strain (Smith et al., 1990a,b). Relative to control, 1,2-

VCM induced primordial follicle loss in ovaries from neonatal B6C3F1 mice at

70

concentrations ≥ 750µM. Unlike VCD, 1,2-VCM did not affect primary follicles (also

shown to be targeted in vitro by the ovotoxicant). These data support indirectly that in

the absence of hepatic tissue the ovary can bioactivate 1,2-VCM to the ovotoxicant,

VCD. However, bioactivation of 1,2-VCM to VCD in the ovary might not be sufficient

to induce loss of primary follicles which are less sensitive than primordial follicles to the

effects of VCD. The reduced sensitivity of ovaries to 1,2-VCM compared to that of VCD

might also be due to the capacity of ovarian microsomal epoxide hydrolase (mEH) to

metabolize the proximate-ovotoxicant 1,2-VCM to a VCH-1,2-diol (Keller et al. 1997).

Cannady et al. (2002) showed that the B6C3F1 mouse ovary expresses catalytically active

mEH, which can be induced following in vivo VCH and VCD dosing. In addition, the

epoxide of 1,2-VCM can be conjugated with glutathione. Preliminary results

demonstrate that a 1,2-VCM glutathione conjugate is formed in ovarian cultures and that

its formation decreases over 15 days of incubation with 1,2-VCM (Rajapaksa et al.,

unpublished data). Thus, insufficient ovarian bioactivation and/or detoxification of 1,2-

VCM by mEH and/or glutathione conjugation could account for the need to use 1,2-VCM

at relatively high concentrations to see ovotoxicity.

The role of ovarian CYP 2E1 in ovarian bioactivation of 1,2-VCM to VCD was

also investigated in this culture system. Following 15d incubation of CYP 2E1 wild-type

(+/+) and null (-/-) ovaries with 1,2-VCM, primordial follicle loss was detected in wild-

type (+/+) ovaries at concentrations ≥ 500µM. However, 1,2-VCM did not affect

primordial follicles in null (-/-) ovaries at any concentration. As with B6C3F1 ovaries,

1,2-VCM did not cause primary follicle loss in either CYP 2E1 wild-type (+/+) or null (-

71

/-) ovaries. These data suggest that 1,2-VCM was bioactivated to VCD in the CYP 2E1

wild-type (+/+) ovary to induce follicle loss, while the CYP 2E1 null (-/-) ovary was not

capable of this bioactivation. Thus, it appears that in this in vitro culture system CYP

2E1 enzyme is necessary for 1,2-VCM-induced ovarian toxicity. Due to the chemical

property of VCH (vapor pressure = 10.2mmHg @ 25°C), the appropriate capacity to

study VCH in culture was not available.

LC/MS analysis detected the VCD metabolite from culture media from CYP 2E1

wild-type (+/+) ovaries incubated with 1,2-VCM (125, 250, 750, and 1000µM; but not

500µM). VCD was not detected in control treated ovary and CYP 2E1 null (-/-) ovary

culture media (data not shown). The detection of VCD was not consistent through out

the samples analyzed (n=3 for each concentration). This might be due to the instability of

VCD in aqueous media. These studies were conducted in culture media stored at -80°C;

therefore, VCD might have degraded during the freeze thaw process. Regardless, this

study provides further evidence that the ovary can bioactivate 1,2-VCM to VCD.

When comparing the morphology of CYP 2E1 wild-type (+/+) ovaries incubated

with VCD and 1,2-VCM, those incubated with 30µM VCD mostly contained unhealthy

follicular debris. However, in ovaries incubated with 1000µM 1,2-VCM, numerous

healthy follicles remained. Thus, the amount of VCD formed from 1,2-VCM in CYP

2E1 wild-type (+/+) ovaries incubated with the highest concentration of 1,2-VCM did not

induce similar damage to that of VCD at a much lower concentration.

CYP 2E1 wild-type (+/+) and null (-/-) mice were dosed daily for 15d to evaluate

the role of CYP 2E1 enzyme in VCH/VCM-induced ovotoxicity in vivo. In vivo VCD

72

dosing caused ovarian toxicity in target follicle groups (primordial and primary) in both

CYP 2E1 wild-type (+/+) and null (-/-) mice, agreeing with previous studies in other

strains of mice (NTP, 1989; Chhabra et al., 1990; Kao et al., 1990; Smith et al., 1990b).

1,2-VCM also decreased (in vivo) follicles in both CYP 2E1 wild-type (+/+) and null (-/-)

mice, suggesting that 1,2-VCM had been bioactivated to the ovotoxic form VCD, in the

absence of CYP 2E1 enzyme. Because CYP 2E1 was required for bioactivation in

ovarian cultures, it appears that in vivo bioactivation of 1,2-VCM by hepatic CYP 2A and

CYP 2B is sufficient to induce ovarian toxicity.

VCH decreased primordial and primary follicles in CYP 2E1 wild-type (+/+)

mice following in vivo dosing. In CYP 2E1 null (-/-) mice VCH decreased primary but

not primordial follicles. Even though VCH did not affect primordial follicles in CYP

2E1 null (-/-) mice, these mice behave similarly to wild-type (+/+) and B6C3F1 mice

following in vivo dosing with all three compounds. In ovarian culture CYP 2E1

contributes to VCM-induced ovotoxicity; however, in the whole animal its role appears

minimal.

Although previous studies suggest that liver is the primary organ involved in

metabolism of VCH and VCM to the ovotoxic form, VCD, the current study suggests that

CYP 2E1 enzyme activity in the ovary can metabolize VCM to VCD. However, in vivo

ovarian bioactivation of VCH/VCM likely plays little if any role in the resulting toxicity.

Furthermore, preliminary data suggest that ovarian detoxification of VCH/VCM via mEH

and/or glutathione conjugation may be more important in the resulting follicle loss.

Therefore, the capacity of the ovary to detoxify xenobiotics needs to be further

73

investigated. The results presented here also demonstrate the potential usefulness of the

neonatal mouse whole ovary culture system for studying ovarian metabolism of

xenobiotic chemicals.

74

CHAPTER 4

OVARIAN DETOXIFICATION OF OVOTOXICANT 4-VINYLCYCLOHEXENE

DIEPOXIDE: THE INVOLVEMENT OF GLUTATHIONE CONJUGATION

Abstract

The occupational chemical, 4-vinylcyclohexene diepoxide (VCD) causes loss of

ovarian small pre-antral follicles in mice. VCD is the ultimate ovotoxicant, and can be

detoxified by microsomal epoxide hydrolase (mEH) or phase II conjugation. This

conjugate has been presumed to be a glutathione (GSH) adduct. The mouse ovary

expresses mEH and synthesizes GSH, and thus, has the capacity to detoxify VCD. This

study was designed to evaluate the role of ovarian GSH in detoxification of VCD in mice.

28d old B6C3F1 mice were dosed with sesame oil (control) or VCD (0.57 mmol/kg/d).

Ovarian and liver tissues were collected 4 or 24h following dosing, and protein extract

was analyzed for VCD-GSH adducts via LC/MS. Postnatal day (PND) 4 CYP 2E1 wild-

type (+/+) and null (-/-) ovaries were incubated with 30µM VCD or medium only

(control) for 15d. Every 2d media were collected and analyzed for VCD-GSH conjugates

via LC/MS. The reaction of VCD with GSH resulted in the formation of mono and di

conjugates. Relative to the diGSH conjugate, the mono GSH conjugate was formed at a

much greater rate in the absence of biological tissue. However, in liver and ovarian

tissues, mono and diGSH conjugates formed at equal amounts. VCD-mono and diGSH

conjugates were detected in liver and ovarian homogenates 4h, but not 24h following in

vivo dosing with VCD. In the in vitro system, VCD-GSH adducts were also detected in

75

culture media from CYP 2E1 +/+ and -/- ovaries at 2 and 4d, in the absence of hepatic

tissue. Here, mono adduct formation was greater than di adduct formation. This study

shows that VCD can be conjugated with GSH, and that the ovary is capable of

detoxifying VCD via GSH conjugation.

Introduction

The ovarian toxicant, 4-vinylcyclohexene diepoxide (VCD) can be detoxified to

an inactive tetrol (4-[1,2-dihydroxy]ethyl-1,2-dihydroxycycloheane) metabolite by

microsomal epoxide hydrolase (mEH) enzyme (Keller et al., 1997). Following in vivo

dosing the tetrol metabolite was detected in mouse and rat urine (Salyers, 1995).

Furthermore, the two electrophilic epoxides on VCD are good substrates for conjugation

with the cellular nucleophile glutathione (GSH). VCD can be conjugated with GSH to

form VCD-monoGSH and VCD-diGSH adducts (Fig. 4.1).

Studies by Giannarini et al. (1981) have shown that at 500mg/kg VCD depletes

liver GSH to 3.7% that of control in Swiss albino mice. Following 24h the GSH level

was restored to normal levels. The Km and Vmax values for the VCD GSH reaction

catalyzed by glutathione-S-transferase (GST) were determined in these studies to be

7.3nM and 66 nmol/mg protein/min respectively in mice. Studies done by Hayakawa et

al. (1975) have show that there are species differences in GST activity towards VCD,

where in sheep liver supernatant GST activity was not detected following incubation with

76

VCD. However, the structural analogue styrene oxide was shown to be a substrate for

GST in sheep liver.

Species differences in the detoxification route of VCD were also observed.

Studies by Salyers (1995) showed that rats can detoxify VCD by forming both the tetrol

metabolite and a conjugate, while the mouse primarily uses conjugation to detoxify VCD.

In these studies both the tetrol and VCD-conjugate were detected in rat urine following a

single dose, whereas, in mouse urine, the tetrol metabolite was shown to be a minor

component. This conjugate was presumed to contain GSH, because treatment with β-

glucuronidase or sulfatase did not change the HPLC profile of this adduct. Collectively,

these studies suggest species differences in the routes used for detoxification of the

ovarian toxicant VCD.

Although the liver is the major organ involved in detoxification of ovarian

toxicants, the ovary has been shown to have the capacity to protect itself by detoxifying

VCD (Cannady et al., 2002; Flaws et al., 1994b). mEH protein was expressed in all

ovarian compartments with the highest expression in the interstitial compartment. mEH

mRNA levels and catalytic activity were induced following VCD dosing in the pre-antral

follicle population (VCD target follicle population; Cannady et al., 2002). Flaws et al.

(1994) showed that rat pre-antral follicles in vitro can metabolize VCD to the inactive

tetrol.

Previous studies have not evaluated ovarian GSH as a possible route for

detoxification of VCD in mice, where GSH might be the main route of VCD elimination.

77

Figure 4.1: Proposed Reaction of VCD with GSH. (A) GSH reaction with VCD to

form VCD-GSH adducts. Arrows indicate possible carbon positions that can be

conjugated with GSH to form either a monoGSH or a diGSH adducts. mw = molecular

weight of each compound, MH+ = molecular weight of the compound ionized (parent

ion), m/z = mass to charge ratio. VCD-diGSH conjugate can be doubly protonated to

give a m/z value of 378. (B) Example of a VCD-monoGSH adduct that could be formed

during this reaction.

78

It has been shown that the rat and mouse ovary synthesize GSH, and rat ovarian GSH

production is regulated during the ovarian cycle (Luderer et al., 2001). Therefore, this

study was designed to evaluate a possible role of ovarian GSH in ovarian metabolism of

VCD in mice. It is hypothesized that VCD is detoxified in the mouse ovary via GSH

conjugation.

Experimental Methods

VCD-mono and diGSH conjugates were synthesized by reacting VCD with GSH

and purified using high-performance liquid chromatography (HPLC). These standards

were then used to detect the formation of VCD-GSH adducts in liver and ovarian tissue

from mice dosed with sesame oil (control) or VCD via liquid chromatography/mass

spectrometry (LC/MS). Ovaries from postnatal day (PND) 4 CYP 2E1 wild-type (+/+)

and null (-/-) mice were cultured for 15d with media (control) or VCD. Media were

collected every 2d and also analyzed for VCD-GSH adducts using LC/MS. All methods

are described in detail in Chapter 2.

79

Results

Reactivity of VCD with GSH to form mono and diGSH adducts

1µM VCD was incubated with 10mM GSH in 0.1M potassium phosphate buffer

(pH 7.4) for 24h at 37oC, the product was analyzed using LC/MS, and peak areas of

VCD, VCD-monoGSH, and VCD-diGSH were quantified. VCD continued to decrease

over the 24h incubation time (Fig. 4.2A). VCD-monoGSH adduct formed was detected

following 30min of incubation. Formation of the mono adduct reached a plateau

following 4h of incubation (Fig. 4.2B). Formation of the di adduct was detected to a

much lower level following 3h of incubation, and did not reach a plateau during the total

incubation time of 24h (Fig. 4.2C).

Synthesis and purification of VCD-monoGSH conjugate

VCD-monoGSH adduct was synthesized and purified to be used as a standard in

the detection of this adduct from biological samples. LC/MS of VCD-monoGSH adduct

at m/z of 448 is shown in Fig. 4.3A. This standard contains a 1% impurity of un-reacted

GSH. Tandem MS (MS/MS) of the adduct is shown in Fig. 4.3B. Common GSH

fragments, such as loss of gamma-glutamic acid with water at m/z 300 and loss of glycine

at m/z 373, can be seen.

80

Figure 4.2: Reactivity of VCD in the Presence of GSH to Form VCD-Mono and

DiGSH Adducts. 1µM VCD in ACN was reacted with 10mM GSH in 0.1M potassium

phosphate buffer (pH 7.4) at 37oC for increasing lengths of time, and analyzed using

LC/MS in the Positive Ion Scan mode. (A)VCD, (B)VCD-monoGSH, and (C)VCD-

diGSH adduct present was quantified using peak area at retention time of 7, 5, and 4 min,

respectively.

Time-Dependent Decrease of Vinylcyclohexene Diepoxide (VCD, 1 µM) withGlutathione (10 mM) in Potassium Phosphate Buffer (0.1M, pH 7.4, 37 oC)

to Form VCD-GSH Adducts

0 4 8 12 16 20 240

1000000

2000000

3000000

4000000

Time (h)

VCD

-Pea

kA

rea

(RT

=7m

in)

A

Time-Dependent Reaction of Vinylcyclohexene Diepoxide (VCD, 1 µM) withGlutathione (10 mM) in Potassium Phosphate Buffer (0.1 M, pH 7.4, 37 oC)

to Form VCD-GSH Di-Adduct (RT = 4 min)

0 4 8 12 16 20 240

100000

200000

300000

400000

Time (h)

VCD

-GSH

Di-A

dduc

tPe

akA

rea

(RT

=4

min

)

Time-Dependent Reaction of Vinylcyclohexene Diepoxide (VCD, 1 µM) withGlutathione (10 mM) in Potassium Phosphate Buffer (0.1 M, pH 7.4, 37 oC)

to Form VCD-GSH Mono-Adduct (RT = 5 min)

0 4 8 12 16 20 240

50000

100000

150000

200000

250000

Time (h)

VCD

-GSH

Mon

o-A

dduc

tPe

akA

rea

(RT

=5

min

)

C

B

81

Figure 4.3: Synthesis and Purification of VCD-MonoGSH Adduct. VCD was reacted

with GSH in 0.1M potassium phosphate buffer (pH 7.4). Product was rotovapped,

extracted using ethyl acetate, and purified using HPLC at a wavelength of 214 nm. (A)

LC/MS and (B) MS/MS of purified VCD-monoGSH adduct at m/z 448. Arrows indicate

common GSH fragments.

R T : 0 .0 0 - 1 7 .0 0 S M : 7 G

0 2 4 6 8 1 0 1 2 1 4 1 6T im e ( m in )

0

5

1 0

1 5

2 0

2 5

3 0

3 5

4 0

4 5

5 0

5 5

6 0

6 5

7 0

7 5

8 0

8 5

9 0

9 5

1 0 0

Rel

ativ

eAb

unda

nce

N L : 7 .3 4 E 6T IC F : + c s id = - 1 0 .0 0 F u ll m s 2 4 4 8 .1 0 @ - 3 0 .0 0 [ 5 0 .0 0 - 5 0 0 .0 0 ] M S 0 8 _ 1 5 _ 2 0 0 6 _ V C D _ G S H_M o n o _ D i_ M S M S 0 1

08_23_2006_VCM_VCDMono_G SH_Purified_MSMS03 #354 -389 RT : 4 .52 -4 .97 AV: 36 NL : 3 .44E4F : + c s id=-10 .00 F u ll ms2 448 .10@ -30 .00 [ 50 .00 -500 .00 ]

50 100 150 200 250 300 350 400 450 500m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

eAb

unda

nce

1 76 .83

144 .97104 .86

122 .96 197 .91373 .10230 .69

99.7992 .96 300 .69

234 .25 354 .91 390 .26 412 .19 447 .51 497 .47

A

B

NH

HN

OH

SHO

O

O

NH

2

O

HO

m/z 373 Loss of glycine

m/z 301 Loss of gamma- glutamic acid + water

O

O

82

Synthesis and purification of VCD-diGSH conjugate

VCD-diGSH adduct was synthesized and purified to be used as a standard in the

detection of this adduct from biological samples. LC/MS of VCD-diGSH adduct at m/z

of 755 and 378 (di-protonated) is shown in Fig 4.4A and C. This standard also contains a

1% impurity of un-reacted GSH. Tandem MS of the adduct at m/z 755 and 378 is shown

in Fig. 4.4B and D. Loss of gamma-glutamic acid with water at m/z 626 and 231, and

loss of glycine at m/z 680 and 303 are detected for the parent m/z of 755 and 378

respectively. The synthetic adduct detected had a peak area of 4.14x107AU at m/z 378

and 1.96x107AU at m/z 755, thus, the detection of this adduct at m/z 378 was more

sensitive. Therefore VCD-diGSH adduct in all biological samples was detected using the

m/z of 378.

Identification of VCD-GSH adduct formation in B6C3F1 mouse liver and ovary following

in vivo dosing

VCD-GSH adduct formation was evaluated in B6C3F1 mouse liver and ovarian

tissue 4 or 24h following in vivo dosing with VCD. LC/MS analysis using Selective

Reaction Monitoring (SRM) setting detected both VCD-mono and diGSH adducts in the

liver and ovary 4h following dosing (Fig. 4.5). In the liver, the concentration of the

monoGSH adduct formed (1.5X103AU) was slightly higher than that of the di adduct

83

Figure 4.4: Synthesis and Purification of VCD-DiGSH Adduct. VCD was reacted

with GSH in 0.1M potassium phosphate buffer (pH 7.4). Product was rotovapped,

extracted using ethyl acetate, and purified using HPLC. LC/MS of purified VCD-diGSH

adduct at m/z 755 (A) and m/z 378 (C), and MS/MS of fraction collected at m/z 755 (B)

and m/z 378 (D) from HPLC purification. Arrows indicate common GSH fragments.

RT: 0.00 - 16.99 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

eAb

unda

nce

NL: 4.14E7TIC F: + c sid=-10.00 Full ms2 378.10@-30.00 [ 50.00-500.00] MS 08_14_2006_VCD_GSH_MSMS01

08_15_2006_VCD_GSH_Mono_Di_MSMS02 #323-368 RT: 4.13-4.70 AV: 46 NL: 2.96E6F: + c sid=-10.00 Full ms2 378.10@-30.00 [ 50.00-500.00]

50 100 150 200 250 300 350 400 450 500m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

eAb

unda

nce

83.93 129.90

176.88

104.92161.86 282.95

290.87

178.91

230.8775.93

181.00 232.89302.99

202.87 280.93307.89256.97 357.02

354.97 372.94 431.11 469.79

x10

RT: 0.00 - 16.99 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

eAb

unda

nce

NL: 1.96E7TIC F: + c sid=-10.00 Full ms2 755.10@-30.00 [ 50.00-800.00] MS 08_14_2006_VCD_GSH_MSMS02

08_15_2006_VCD_GSH_Mono_Di_MSMS03 #320-361 RT: 4.15-4.68 AV: 42 NL: 1.69E6F: + c sid=-10.00 Full ms2 755.10@-30.00 [ 50.00-800.00]

100 200 300 400 500 600 700 800m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

eAb

unda

nce

478.98

626.01

755.06

496.97431.97

302.88551.00

737.06

176.82282.90

144.84 230.84

305.80 680.07375.85104.83

99.51 762.16

A

B

C

D

84

Figure 4.5: Detection of VCD-Mono and DiGSH Adducts in 28d Female B6C3F1

Mice. 28d female B6C3F1 mice were administered one (i.p.) dose of either sesame oil

(vehicle control) or VCD (0.57 mmol/kg/d). Animals were killed by CO2 inhalation 4h

following the single dose, liver and ovary tissue was homogenized, VCD-GSH adduct

was extracted using ACN, and analyzed using SRM on LC/MS. 4h (A) control and (B)

VCD treated liver homogenate, and 4h (C) control and (D) VCD treated ovary

homogenate.

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 1.00E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver03

NL: 3.62E2TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver03

NL: 1.10E2TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver03

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 9.07E6TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver13

NL: 1.50E3TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver13

NL: 9.54E2TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_18_2006_B6C3F1_Mouse_Dosing_Liver13

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 1.30E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples01

NL: 1.85E3TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples01

NL: 1.57E2TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples01

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100R

elat

ive

Abun

danc

eNL: 1.30E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples04

NL: 1.20E3TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples04

NL: 1.10E3TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_22_2006_B6C3F1_Ovary_VCD_Samples04

A

B

C

DCBZ internal std

monoGSH

diGSH

CBZ internal std

CBZ internal std

CBZ internal std

monoGSH

diGSH

m/z 448

m/z 378

m/z 448

m/z 378

m/z 378 m/z 378

m/z 448 m/z 448

m/z 237 m/z 237

m/z 237 m/z

85

formed (0.954X103AU). In the ovary nearly equal amounts of mono (1.2X103AU) and di

(1.1X103AU) adduct were detected. At 24h following dosing neither of the VCD-GSH

adducts were detected in the liver or ovary (data not shown).

Identification of ovarian VCD-GSH adducts formation in CYP 2E1 wild-type and null

mice following in vitro ovarian culture

Ovarian VCD-GSH adduct formation was evaluated in CYP 2E1 wild-type (+/+)

and null (-/-) ovarian culture media to eliminate the metabolic role of the liver. LC/MS

analysis using SRM setting detected VCD-mono and diGSH adducts in culture media

from both CYP 2E1 wild-type (+/+) and null (-/-) ovaries incubated with VCD for 2 and

4d (Fig. 4.6). The concentration of VCD-monoGSH adduct formed was greater than

VCD-diGSH adduct at both time points in CYP 2E1 wild-type (+/+) and null (-/-) ovary

culture media. For example, at 2d mono and di adduct concentrations in CYP 2E1 wild-

type (+/+) culture media were 6.10X103AU and 1.4X103AU, and in CYP 2E1 null (-/-)

culture media were 1.7X104AU and 1.2X103AU respectively. Following 6d of culture,

VCD-mono and diGSH adducts detected were close to background (data not shown).

86

Figure 4.6: VCD-Mono and DiGSH Adduct Formation in CYP 2E1 Wild-Type and

Null Ovary Culture Media. Ovaries from PND4 CYP 2E1 wild-type (+/+) and null

(-/-) mice were incubated with VCD (30 µM) or medium control for 15d. Culture media

was analyzed for VCD-GSH adducts using SRM on LC/MS at 2d. Culture media from

CYP 2E1 +/+ ovary incubated with (A) control and (B) 30µM VCD for 2d. Culture

media from CYP 2E1 -/- ovary incubated with (C) control and (D) 30µM VCD 2d.

RT: 0.00 - 17.04 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 1.50E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_23_2006_CYP2E1WT_VCD_Media03

NL: 6.10E3TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_23_2006_CYP2E1WT_VCD_Media03

NL: 1.40E3TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_23_2006_CYP2E1WT_VCD_Media03

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 1.50E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_23_2006_CYP2E1WT_VCD_Media205

NL: 6.10E3TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_23_2006_CYP2E1WT_VCD_Media205

NL: 1.40E3TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_23_2006_CYP2E1WT_VCD_Media205

RT: 0.00 - 17.03 SM: 7G

0 2 4 6 8 10 12 14 16Time (min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

NL: 1.70E7TIC F: + c sid=-10.00 SRM ms2 237.00@-25.00 [ 193.75-194.25] MS 08_24_2006_CYP2E1KO_VCD_Media226

NL: 1.70E4TIC F: + c sid=-10.00 SRM ms2 448.10@-30.00 [ 176.65-177.15] MS 08_24_2006_CYP2E1KO_VCD_Media226

NL: 1.20E3TIC F: + c sid=-10.00 SRM ms2 378.10@-27.00 [ 129.75-177.15] MS 08_24_2006_CYP2E1KO_VCD_Media226

A

B D

C

m/z 237

m/z 448

m/z 378 m/z 378

m/z 378

m/z 448

m/z 448

m/z 237

CBZ internal std

CBZ internal std

CBZ internal std

monoGSH monoGSH

monoGSH

diGSH diGSH

diGSH

RT: 0.00 - 17.03 SM: 11G

0 2 4 6 8 10 12 14 16Time(min)

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

0

20

40

60

80

100

Rel

ativ

eAb

unda

nce

MA: 313511127

MA: 1436

NL: 1.70E7TICF: +csid=-10.00 SRM ms2 237.00@-25.00 [193.75-194.25] MS09_20_2006_gsh_VCD01_060921095112

NL: 1.70E4TICF: +csid=-10.00 SRM ms2 448.10@-30.00 [176.65-177.15] MS09_20_2006_gsh_VCD01_060921095112

NL: 1.20E3TICF: +csid=-10.00 SRM ms2 378.10@-27.00 [129.75-177.15] MS09_20_2006_gsh_VCD01_060921095112

m/z 237

monoGSH m/z 448

diGSH m/z 378

87

Discussion

Previous studies have shown that the mouse ovary has the capacity to detoxify

VCD to a tetrol metabolite (Cannady et al., 2002; Flaws et al., 1994b). Studies indicate

that the primary route of detoxification of VCD in the mouse is via conjugation, while in

the rat equal amounts of tetrol and conjugate are formed (Salyers, 1995). The polar

conjugate was thought to be GSH, because it was not hydrolyzed following β-

glucuronidase or sulfatase treatment. The ovary synthesizes GSH, and GSH content in

the ovary has been shown to be regulated by the estrous cycle (Luderer et al., 2001).

However, studies thus far have not looked at the role of ovarian GSH as a possible route

for VCD detoxification in mice. Therefore, the current study was designed to evaluate

whether the ovary is capable of detoxifying VCD via GSH conjugation.

GSH adducts of VCD were synthesized (mono and di) and purified for use as

standards to qualitatively detect VCD-GSH adduct formation by LC/MS/MS in liver and

ovarian samples. Both GSH adducts contained a 1% un-reacted GSH impurity.

Furthermore, LC/MS analysis of the VCD-diGSH conjugate revealed that this compound

is protonated at two sites to give a mass to charge ratio of 378. This m/z 378 detection

via LC/MS was more sensitive compared to that of the m/z 755. MS/MS of the two

adducts revealed a common fragment with an m/z of 177. A fragment with m/z of 177

was also detected in the 1,2-VCM GSH conjugate synthesized (data not shown). This

fragment could not be identified. Other fragments were detected that are common to

GSH. These fragments, such as the loss of gamma-glutamic acid and glycine, were used

to confirm the formation of VCD-GSH adducts in biological samples.

88

VCD-mono and diGSH reaction rates were examined by reacting VCD with GSH

in potassium phosphate buffer at 37oC. The reaction of VCD with GSH to form the

monoGSH adduct occurred rapidly, where by 30min the VCD-monoGSH adduct was

detected. The VCD-monoGSH adduct formation reaction reached a plateau by 4h of

incubation. Unlike the VCD-monoGSH adduct, the reaction rate of VCD with GSH to

form the di adduct was slow; where the first adduct formation was detected at 3h. This

reaction however did not reach a plateau over the 24h incubation time. These data

suggest spontaneous formation (ex vivo) of VCD-monoGSH is the primary reaction, and

in the presence of excess GSH both epoxides can be conjugated. Slow reactivity of GSH

with VCD-monoGSH to form the di adduct could result from stearic hindrance by GSH

conjugated to one of the epoxides on VCD.

Analysis of VCD-GSH conjugates in the biological samples showed that VCD

can be conjugated with GSH in both liver and ovary. VCD-GSH mono and di adducts

were detected in B6C3F1 liver and ovarian homogenates 4h following a single i.p. dose of

VCD. Unlike spontaneous (ex vivo) adduct formation, the amount of the monoGSH

adduct formed in the liver and ovary was only slightly higher than that of the di adduct

formed (mono to di adduct ratio was 1.5 and 1.1 for liver and ovary respectively). Thus,

in biological samples the mono adduct reaction still may be the primary reaction, but di

adduct formation appears to be more efficient than that in ex vivo. This is likely due to an

enzymatic catalysis of this reaction.

VCD-GSH adducts were not detected 24h following dosing (data not shown).

Even though, t1/2 data for VCD-GSH adducts are not known, it can be presumed that by

89

24h VCD-GSH adducts were excreted in these mice. Salyers (1995) has detected a

conjugate of VCD (thought to be GSH) in rat and mouse urine following dosing with

VCD, suggesting this adduct is excreted following conjugation. In a study by Devine et

al. (2001), a decrease in rat hepatic GSH content was detected at 2h following a single

injection of VCD (0.57mmol/kg). Six and 26h following the single dose of VCD, GSH

levels in rats were comparable to controls. Studies by Salyers (1995) have shown that a

single dose of VCD (0.57mmol/kg) reduced mouse ovarian GSH content to 55% of

control 6h following dosing, and by 24h the ovarian GSH level was restored. Taken

together, these studies suggest that in mice 24h following administration of a single dose

of VCD, VCD-GSH adducts might be excreted and GSH levels restored in liver and

ovary.

To rule out the possibility that ovarian adducts resulted from disposition of

adducts from other tissue, such as the liver, VCD-GSH adduct formation in the absence

of hepatic tissue was investigated using the in vitro ovarian culture system. Because

previous studies evaluating VCH/VCM metabolism to VCD utilized CYP 2E1 wild-type

(+/+) and null (-/-) mouse ovaries, for culture, ovaries from the same strain of mice were

used. VCD-mono and diGSH adducts were detected in the media from both CYP 2E1

wild-type (+/+) and null (-/-) ovaries cultured with VCD. This study shows that the

ovary has the capacity to detoxify VCD in the absence of hepatic tissue. The

concentration of VCD-monoGSH adduct formed was also greater than VCD-diGSH

adduct in both CYP 2E1 wild-type (+/+) and null (-/-) culture media, similar to but not as

90

great as the VCD-GSH reaction in non-biological samples. Further, di adduct formation

was not as great as that formed during in vivo analysis of ovary and liver homogenates.

GSH conjugation can occur spontaneously or it can be catalyzed by the GST

enzyme (Parkinson, 2001). Previous studies have shown Km and Vmax values for GST

catalyzed reaction of VCD with GSH to be 7.3nM and 66nmol/mg protein/min

respectively in the mouse liver. Doerr-Stevens et al. (1999) showed an increase in

hepatic GST levels following 5, 10 and 15d of dosing with the parent compound of VCD,

4-vinylcyclohexene (VCH). When comparing the ratio of mono to di adduct formed at

the 4h time point the ratio was greater in the absence of biological tissue (8) compared to

in the presence of biological tissue (1.5 and 1.1 for liver and ovary, respectively).

Because the ratio of the two adducts in the biological sample was nearly one, this

suggests that in the biological tissue formation of the two adducts is enzymatically

catalyzed, for example by GST, and not a spontaneous reaction. In the ovarian culture at

2d (earliest time point analyzed) the ratio of mono to di adduct formation was 4.0 and 2.8

in CYP 2E1 wild-type (+/+) and null (-/-) culture media, respectively. Thus, in the

culture media, as with ovary and liver homogenate, formation of the di adduct is still

greater than the non-biological reaction at 4h, suggesting a role for enzyme activity. GST

mRNA expression was detected in CYP 2E1 wild-type (+/+) and null (-/-) ovaries,

providing further evidence for the role of GST in the formation of VCD-GSH adducts

(data not shown). The relative degree of di adduct formation was in vivo (liver and

ovary) > in vitro (ovaries in culture) > ex vivo (spontaneous reaction). The reason for

greater formation in vivo as compared with in vitro could be due to reduced ability of the

91

ovary in culture to maintain GSH production. In the adult rat ovary GSH synthesis was

shown to be regulated by the hypothalamic-pituitary-gonadal (HPG) axis (Luderer et al.,

2001). Without the HPG axis in place in the culture system production of GSH might

decrease over time.

Furthermore, the amount of VCD-GSH adducts formed decreased as a function of

time in both CYP 2E1 wild-type (+/+) and null (-/-) ovarian culture media. After 6d the

adducts detected were close to background (data not shown), suggesting GSH content has

been depleted. Follicle loss in the in vitro system requires constant exposure to VCD in

media (Devine et al., 2002a). Additionally, VCD also has to be administered by a

repeated daily exposure (15d; ip) to induce ovarian toxicity (Borman et al., 1999). This

requirement for constant exposure may relate to efficient detoxification capacity. Thus,

ovarian toxicity may not be induced by VCD until ovarian detoxification capacity

becomes exhausted.

In summary, these studies have shown that VCD can be conjugated via GSH on

both epoxides; however, mono adduct formation is favored substantially in non-

biological reactions and to a lesser degree in biological tissues, suggesting a role for

enzymatic catalysis of this reaction. The ovary like the liver is capable of conjugating

VCD with GSH for detoxification. Thus, ovarian toxicants that are locally bioactivated

and those reaching the ovary following hepatic metabolism can be detoxified in the

ovary. The question remains as to the biological role of the ovary in detoxifying these

epoxides, particularly when presented repeatedly as a result of multiple doses.

92

CHAPTER 5

OVARIAN BIOACTIVATION OF 7,12-DIMETHYLBENZ[A]ANTHRACENE IN

B6C3F1 MICE: THE INVOLVEMENT OF MICROSOMAL EPOXIDE HYDROLASE

(mEH)

Abstract

Ovarian follicle disruption in mice caused by 7,12-dimethylbenz[a]anthracene

(DMBA) is attributed to its sequential bioactivation to the DMBA-3,4-diol-1,2-epoxide

by CYP 1B1, microsomal epoxide hydrolase (mEH), and CYP 1A1/1B1. Studies suggest

that the mouse ovary expresses these enzymes, and thus, may be capable of bioactivating

DMBA to its ovotoxic metabolite. The present study was designed to evaluate the role of

ovarian mEH in DMBA-induced ovotoxicity using a neonatal mouse ovarian culture

system. Ovaries from postnatal day (PND) 4 B6C3F1 mice were incubated with DMBA

(12.5nM -1µM) for various lengths of time. Following incubation, ovaries were

histologically evaluated, or assessed for mEH protein or mRNA. Following 15d

incubation, DMBA reduced (p < 0.05) healthy follicles at concentrations ≥ 12.5nM. At

1µM DMBA, follicle loss and increased mEH protein were measured (p < 0.05) by 6h.

mRNA encoding mEH markedly increased after 2d incubation, and this increase

preceded accelerated follicle loss at 4d. Furthermore, follicle loss induced by DMBA

was prevented when cyclohexene oxide (CHO; 2mM), an mEH inhibitor, was added to

DMBA incubations. These studies suggest that the PND4 mouse ovary is capable of

93

bioactivating DMBA to its ovotoxic form, and that ovarian mEH enzyme activity is likely

involved. Furthermore, these observations support the use of a novel ovarian culture

system to study ovary-specific metabolism of xenobiotic chemicals.

Introduction

The polycyclic aromatic hydrocarbon (PAH), 7,12-dimethylbenz[a]anthracene

(DMBA) causes ovarian follicle disruption and ovarian failure in mice (Mattison, 1980;

Weitzman et al., 1992). Ovarian toxicity of this compound is attributed to bioactivation

of DMBA to a 3,4-diol-1,2-epoxide metabolite. DMBA is sequentially bioactivated to

the ovotoxicant by CYP 1B1, microsomal epoxide hydrolase (mEH), and CYP 1A1/1B1

(Fig. 1.3; Savas et al., 1997; Shimada et al., 2001).

Studies have shown that the mouse ovary expresses CYP1A1, CYP1B1 and mEH,

and mRNA encoding these enzymes is inducible following dosing with substrates for

these enzymes (Cannady et al., 2002; Shimada et al., 2003). Inhibition of rat ovarian

CYP 1B1 activity using an anti-P4501B1 IgG markedly reduced DMBA metabolism

(Otto et al., 1992). mEH enzyme activity in the mouse ovary was induced following in

vivo dosing with the industrial chemical 4-vinylcyclohexene diepoxide, VCD, which is

detoxified by mEH (Cannady et al., 2002). mEH protein expression was detected in

human granulosa, theca interna and luteal cells, and catalytic activity was detected in

granulosa cells and CL microsomal fractions (Hattori et al., 2000). Unilateral intra-

ovarian injection of mice with DMBA destroyed follicles only in the treated ovary

94

(Shiromizu and Mattison, 1985). Studies have shown that DMBA can be metabolized by

ovarian and placental microsomes (Bengtsson et al., 1983; Miyata et al., 2002).

Furthermore, studies by Becedas et al. (1993) suggest that rat granulosa cells in culture

can metabolize DMBA to the carcinogen, 3,4-diol-1,2-epoxide. Collectively, these

findings suggest that the ovary is capable of bioactivating DMBA to the 3,4-diol-1,2-

epoxide.

Even though studies indicate that the ovary is capable of bioactivating DMBA to

the ovotoxicant, the role of ovarian enzymes, such as mEH, in the resulting toxicity of

DMBA has not been assessed. Therefore, the present study was designed to evaluate the

role of ovarian mEH in DMBA-induced ovotoxicity utilizing the novel in vitro ovarian

culture system. Using this system, hepatic contribution to bioactivation of DMBA is

removed, and ovary-specific capabilities can be assessed. The hypothesis is that ovarian

mEH is involved in bioactivation of DMBA to the 3,4-diol-1,2-epoxide as evidenced by

ovarian toxicity.

Experimental Methods

Postnatal day (PND) 4 female B6C3F1 ovaries were incubated with DMSO

(vehicle control), DMBA, CHO or DMBA + CHO at times and concentrations indicated

in figure legends. Following incubation, ovaries were processed for histological

evaluation, and mEH mRNA or protein level. All methods are described in detail in

Chapter 2.

95

Results

Effect of concentration on DMBA-induced follicle loss

Follicle loss was evaluated in PND4 mouse ovaries following 15d of incubation

with various concentrations (12.5nM - 1µM) of DMBA (Fig. 5.1). Compared to vehicle

control, DMBA reduced (p < 0.05) healthy primordial follicles at concentrations ≥

12.5nM (Fig. 5.1A). Healthy primary follicles were reduced (p < 0.05) at concentrations

≥ 25nM DMBA (Fig. 5.1B). DMBA markedly reduced (p < 0.05) healthy secondary

follicles at all concentrations (Fig. 5.1C). All healthy follicle populations were depleted

by DMBA at concentrations ≥ 250nM.

Time course of DMBA-induced follicle loss

Follicle loss was evaluated in PND4 mouse ovaries incubated with 1µM DMBA

for various time points in culture (Fig. 5.2). Relative to vehicle control, ovaries incubated

with DMBA for 6h, showed loss of healthy primordial and primary follicles (p < 0.05).

All healthy primordial and primary follicles were depleted in these ovaries by 8d of

culture (Fig. 5.2A and B). Secondary follicles in PND4 ovaries do not develop until 4d.

No healthy secondary follicles were observed in DMBA-treated ovaries at any time point

96

Figure 5.1: Effect of Varying Concentrations of DMBA on Follicle Loss. Ovaries

from PND4 B6C3F1 mice were cultured with vehicle control or DMBA (12.5nM - 1µM)

for 15d. Following incubation, ovaries were collected, and processed for histological

evaluation as described in materials and methods. Healthy (A) primordial, (B) primary,

and (C) secondary follicles were classified and counted. Values are mean ± SE total

follicles counted/ovary, n=5; * = different from each follicle type control, p<0.05.

97

(Fig. 5.2C). However, some unhealthy secondary follicles in DMBA-treated ovaries

were observed between 4 – 15d of culture (data not shown).

DMBA-induced cell death

The type of cell death occurring in cultured ovaries following incubation with

DMBA (4d) was evaluated using toluidine blue staining (Fig. 5.3). Ovaries treated with

DMBA contained pyknotic bodies in both the oocyte and granulosa cell nuclei. Some

follicles in these ovaries also contained vacuoles along with pyknotic bodies (Fig. 5.3C

and D). These morphological changes are characteristic of apoptosis (Kerr et al., 1972;

Tome et al., 2001).

Effect of DMBA on ovarian expression of mEH mRNA

To investigate the effect of DMBA on ovarian mEH enzyme expression, the level

of mEH mRNA in ovaries collected from PND4 mice following DMBA treatment was

quantified using real-time PCR (Fig. 5.4A). mEH mRNA was detected in RNA isolated

from ovaries in all treatment groups at all time points. Following incubation with 1µM

DMBA for 3, 6, or 24h, the level of mEH mRNA did not change compared to that of the

98

Figure 5.2: Time Course of DMBA-Induced Follicle Loss. Ovaries from PND4

B6C3F1 mice were cultured with vehicle control or 1µM DMBA for 6h - 15d. Following

incubation, ovaries were collected, and processed for histological evaluation as described

in materials and methods. Healthy (A) primordial, (B) primary, and (C) secondary

follicles were classified and counted. Values are mean ± SE total follicles counted/ovary,

n=5; * = different from each follicle type control at each time point, p<0.05.

99

Figure 5.3: Effect of DMBA on Toluidine Blue Staining. PND4 B6C3F1 ovaries were

cultured with vehicle control or 1µM DMBA for 4d. Following incubation, ovaries were

fixed in 3% glutaraldehyde, sectioned, and stained with toluidine blue as described in

materials and methods. Images of ovaries treated with (A) vehicle control and (C)

DMBA captured with a 40X objective lens. Images of ovaries treated with (B) vehicle

control and (D) DMBA captured with a 100X objective lens. GC = granulosa cell.

100

vehicle control treated group. At 2d there was a 5.2 fold increase (p < 0.05) in mEH

mRNA compared to vehicle control (Fig. 5.4A).

Fig. 5.4B shows the total follicle loss expressed as a percentage of control

throughout the 15d DMBA incubation period. The rate of DMBA-induced follicle loss

from 6h to 2d did not differ significantly. At 4d the rate of DMBA-induced follicle loss

increased (p < 0.05), at which point it plateaued until 15d of culture (Fig. 5.4B).

Effect of DMBA on ovarian expression of mEH protein

To further evaluate the ovarian distribution of mEH, protein expression was

visualized using confocal microscopy in ovaries collected from PND4 mice (Fig. 5.5).

mEH protein was detected in all follicle populations present in PND4 ovaries (primordial

and primary). Staining intensity for mEH in these follicles was highly localized to the

oocyte cytoplasm. mEH protein was not detected in granulosa cells of either primordial

or primary follicles (Fig. 5.5A and B). Following incubation with 1µM DMBA for 6h,

mEH staining in primary oocytes increased (p < 0.05) compared to that of vehicle control

ovary (Fig. 5.5 D and F). Interestingly, some diffuse mEH staining was detected in the

primary oocyte nucleus in these DMBA treated ovaries. There was no difference in

staining intensity for mEH expression in primordial follicles between DMBA-treated and

vehicle control ovaries (Fig. 5.5F). No Cy-5 staining was seen in immuno-negative

sections at λ = 647nm (Fig. 5.5E).

101

Figure 5.4: Effect of DMBA on mEH mRNA Expression and Total Follicle Loss.

Ovaries from PND4 B6C3F1 mice were cultured with vehicle control or 1µM DMBA for

varying times. Following incubation, ovaries were collected, and processed for (A)

measurement of mRNA encoding mEH by real-time PCR, or (B) follicle counts as

described in material and methods. Different letters differ from one another within each

group; p<0.05.

102

Effect of mEH inhibitor on DMBA-induced follicle loss

Follicle loss was evaluated in PND4 mouse ovaries following 6h incubation with

an mEH enzyme inhibitor, cyclohexene oxide (CHO; Fig. 5.6). CHO acts as an

alternative substrate for mEH, and competitively inhibits mEH activity. Thus,

metabolism of DMBA to the 3,4-diol-1,2-epoxide (active metabolite) is decreased

(Oesch, 1973). CHO (2mM) alone did not affect primordial and primary follicle

populations following 6h incubation. DMBA significantly (p < 0.05) decreased

primordial and primary follicles following 6h in culture. Loss of primordial and primary

follicles was prevented by the addition of 2mM CHO to DMBA incubations (Fig. 5.6A

and B).

103

Figure 5.5: Effect of DMBA on mEH Protein. PND4 ovaries were cultured with

vehicle control or 1µM DMBA for 6h, and processed for confocal microscopy. mEH

(red stain) staining in a section of a (A) control or (C) DMBA treated ovary at 40X.

Overlay of mEH and DNA (green stain) staining in the same (B) control or (D) DMBA

treated ovarian sections at 40X. Arrows indicate primary follicles. (E) Immuno-negative

ovarian section. (F) Fluorescence intensity of 110 follicles counted/11sections/3 ovaries.

Values are mean ± SE fluorescence intensity/follicle, n=3; * = different from each follicle

type control, p<0.05.

104

Figure 5.6: Effect of mEH Inhibitor (CHO) on DMBA-Induced Follicle Loss.

Ovaries from PND4 B6C3F1 mice were cultured with vehicle control, 2mM CHO, 1µM

DMBA, or 1µM DMBA + 2mM CHO for 6h. Following incubation, ovaries were

collected, and processed for histological evaluation as described in materials and

methods. Healthy (A) primordial and (B) primary follicles were classified and counted.

Values are mean ± SE total follicles counted/ovary, n=5; *, p<0.05.

105

Discussion

DMBA targets all follicle populations in the ovary following in vivo dosing. This

loss leads to ovarian failure (Mattison, 1980; Weitzman et al., 1992). Previous studies

indicate that DMBA metabolism to the 3,4-diol-1,2-epoxide is required for ovarian toxic

effects of this compound (Sawicki et al., 1983; Vigny et al., 1985; Shiromizu and

Mattison, 1985). Metabolic activation of DMBA to 3,4-diol-1,2-epoxide requires three

bioactivation steps mediated by CYP1B1, mEH, and CYP1A1/1B1 (Savas et al., 1997;

Shimada et al., 2001).

Although the liver is the primary organ participating in bioactivation of

xenobiotics, extra-hepatic organs such as the ovary have also been shown to be capable

of bioactivation of xenobiotic chemicals. Both CYP1A1 and CYP1B1 are expressed in

the ovary and mRNA for these CYP 450s was induced in the ovary following a single

dose of DMBA (Shimada et al., 2003). mEH enzyme is expressed in the ovary, and

following dosing with the ovarian toxicant 4-vinylcyclohexene diepoxide, VCD, mEH

activity was induced in VCD-targeted follicle populations (Cannady et al., 2002). The

study reported here provides evidence for the role of mEH in chemical-induced ovarian

toxicity. Therefore, the current study was designed to evaluate a possible role of ovarian

mEH in metabolism of DMBA using the novel ovarian culture system. By using this

approach, the role of ovarian metabolism of DMBA and DMBA-induced ovotoxicity can

be evaluated, independent of hepatic contributions.

DMBA reduced healthy primordial and secondary follicles at concentrations ≥

12.5nM following 15d incubations. Unlike primordial and secondary follicles, primary

106

follicle loss was observed at concentrations ≥ 25nM. This observation is interesting

because VCD also had less of an effect on primary follicles compared to that of

primordial follicles in PND4 rat ovarian cultures (Devine et al., 2004). DMBA ≥ 250nM

depleted all healthy ovarian follicles in culture following 15d of incubation. At that time,

follicles in ovaries incubated with DMBA had lost their shape, burst, and become part of

the interstitial space (data not shown). A similar degree of ovotoxicity was seen in PND4

B6C3F1 ovaries incubated with 30µM VCD (15d; Fig. 3.1). Therefore, DMBA (250nM)

induces ovotoxicity in the culture system at a much lower concentration compared to

VCD (30µM). This greater potency of DMBA compared with VCD was also observed in

an in vivo 15d dosing study, where an equivalent degree of follicle loss (ED50) in mice

was seen with 0.02mg/kg DMBA compared with 80mg/kg for VCD (Borman et al.,

2000).

The highest concentration of DMBA (1µM) was then utilized to determine the

shortest time point for follicle loss, to be used in studies for evaluating the role of mEH.

Ovaries were incubated with 1µM DMBA or vehicle. Following 6h in culture, DMBA

decreased healthy primordial and primary follicles. Secondary follicles in PND4 ovaries

begin to form after 4d in culture. No healthy secondary follicles were observed in ovaries

cultured with DMBA, at any time point. However, unhealthy secondary follicles were

observed in DMBA-treated ovaries cultured ≥ 4d. Therefore, DMBA targeted secondary

follicles directly, rather than preventing recruitment from the primary to secondary stage.

By 8d all healthy ovarian follicle populations were depleted in ovaries incubated with

DMBA.

107

Ovarian toxicants such as VCD have been shown to induce follicle loss in rats via

acceleration of the natural process of atresia (apoptosis; Springer et al., 1996a, b; Hu et

al., 2001a, b). Therefore, to further examine the type of cell death caused by DMBA,

ovaries were stained with toluidine blue following 4d in culture. In ovaries treated with

DMBA (1µM) pyknotic bodies were detected in granulosa cells and oocytes of ovarian

follicles, while other follicles contained pyknotic bodies along with vacuoles. These

morphological changes are characteristic of apoptosis (Kerr et al., 1971; Tome et al.,

2001). Thus, it can be hypothesized that DMBA induces ovarian follicle death via

apoptosis. Previous studies by Matikainen et al. (2001) have shown an increase

expression of Bax (pro-apoptotic member of the Bcl-2 family of proto-oncogenes)

protein in primordial and primary follicles in ovaries incubated with DMBA in culture.

Furthermore, follicle loss was not observed in Bax null ovaries cultured with DMBA

compared to wild-type ovaries. Collectively the results of these studies suggest that, as

with other ovotoxicants, DMBA-induced follicle loss is via an apoptotic dependent

mechanism.

An evaluation was made as to whether follicle loss induced by DMBA involves

ovarian mEH. Previous studies have shown that the adult B6C3F1 mouse ovary

expresses catalytically active mEH, and this activity can be induced in the ovary

following administration of VCD (Cannady et al., 2002). Therefore, expression of mEH

protein in the PND4 ovary was evaluated by confocal microscopy. mEH was seen to be

expressed in all follicle populations of the PND4 ovary. Expression was highly

concentrated in oocyte cytoplasm. Unlike the adult B6C3F1 mouse ovary, mEH

108

expression was not observed in PND4 ovarian granulosa cells (Cannady et al., 2002).

Following DMBA treatment for 6h, mEH expression increased in oocytes of primary but

not primordial follicles. This is inconsistent with the observation that primordial follicles

were more sensitive to DMBA-induced toxicity than primary follicles. Thus, it must be

assumed that even though DMBA is more highly bioactivated in primary follicles,

DMBA or DMBA metabolites can diffuse into other ovarian compartments such as the

primordial follicle pool. Conversely, the basal level of mEH expressed in primordial

follicles may be sufficient to effectively bioactivate DMBA, to cause localized

ovotoxicity.

Even though ovotoxicity and increased mEH protein expression were observed by

6h following DMBA treatment, mRNA encoding mEH was not markedly increased until

2d in culture. Previous studies have shown an increase in hepatic mEH protein level that

could not be explained by an increase in gene transcription, and thus, mEH is thought to

be also regulated post-transcriptionally (Kim and Kim, 1992; Simmons et al., 1987).

Interestingly, the rate of follicle loss markedly increased between 2 and 4d. Therefore,

the induction of expression of mEH mRNA directly precedes a significant increase in the

rate of follicle loss at 4d. The lag in time between the marked increases in mRNA (2d)

for mEH and follicle loss (4d) likely reflects the time between mRNA translation, mEH-

stimulated bioactivation of DMBA, and onset of follicle loss.

CHO, an mEH inhibitor, was used to block DMBA bioactivation mediated by

mEH. Incubation of ovaries with CHO alone for 6h did not affect follicle populations.

However, at that time DMBA induced follicle loss (primordial and primary; p < 0.05).

109

Loss of follicles induced by DMBA was inhibited by co-incubation of ovaries with CHO.

This observation provides functional support that ovarian mEH plays a role in

bioactivation of DMBA.

In summary, data presented here suggest that DMBA is a highly potent ovarian

toxicant. In these ovarian cultures, and most likely in vivo, ovarian mEH plays a key role

in the bioactivation of DMBA. Ovarian mEH can be induced by DMBA at the

transcriptional and translational levels. Thus, DMBA-induced expression of mEH

appears to contribute to the high level of potency of DMBA. These findings support an

extra-hepatic role for target organ metabolism of xenobiotic agents that could amplify

potential hepatic effects. Additionally, this study demonstrates that the in vitro PND4

whole ovary culture system will be useful for investigating ovarian capabilities for

metabolism of xenobiotic chemicals.

110

CHAPTER 6

COMPARISON OF 4-VINYLCYCLOHEXENE DIEPOXIDE AND 7,12-

DIMETHYLBENZ[A]ANTHRACENE-INDUCED OVOTOXICITY IN THE CYP 2E1

WILD-TYPE AND NULL OVARIES IN CULTURE

Abstract

A mouse whole ovary culture system has been used to investigate follicle loss

caused by several known ovotoxicants including, 4-vinylcyclohexene diepoxide (VCD)

and the polycyclic aromatic hydrocarbon (PAH), 7,12-dimethylbenz[a]anthracene

(DMBA). VCD is the ultimate ovotoxicant, and can be detoxified by microsomal

epoxide hydrolase (mEH) to a tetrol metabolite. Unlike VCD, DMBA disruption of

follicles is attributed to bioactivation to a 3,4-diol-1,2-epoxide by CYP 1A1 and 1B1,

and mEH. Previous studies (Fig. 3.1) have shown differences in toxicity induced by

VCD between CYP 2E1 wild-type (+/+) and null (-/-) ovaries in culture. This study was

designed to further evaluate VCD and DMBA-induced ovotoxicity in CYP 2E1 +/+ and

-/- mouse ovaries using the whole ovary culture system. Ovaries from postnatal day

(PND) 4 CYP 2E1 +/+ and -/- mice were incubated in media containing VCD (5, 10, 15,

20, and 25µM), or DMBA (0.5, 0.625, 0.75, 1µM) for 15d. Following incubation,

ovaries were prepared for histological evaluation of follicle numbers. VCD reduced

primordial follicle loss at concentrations ≥ 5µM in CYP 2E1 +/+ and -/- ovaries. 5 and

10µM VCD-induced primordial follicle loss was greater (p<0.05) in CYP 2E1 +/+

111

ovaries compared to -/- ovaries. VCD reduced primary follicles at concentrations ≥ 5µM

in CYP 2E1 +/+ and at ≥ 15µM in -/- ovaries. DMBA-induced primordial follicle loss at

concentrations ≥ 0.5µM in CYP 2E1 +/+ and -/- ovaries. DMBA-induced primordial

follicle loss was greater (p<0.05) in CYP 2E1 -/- ovaries at concentrations ≥ 0.625µM.

DMBA reduced primary follicles at concentrations ≥ 0.625µM in CYP 2E1 -/- ovaries.

DMBA did not affect primary follicles in CYP 2E1 +/+ ovaries. The enhanced

ovotoxicity for DMBA and reduced ovotoxicity observed for VCD in CYP 2E1 -/- mouse

ovaries suggest a role for CYP 2E1 in regulating ovarian mEH activity. These findings

support that CYP 2E1 may decrease expression/activity of mEH and therein reduce

detoxification of VCD, and increase bioactivation of DMBA.

Introduction

The industrial chemical, 4-vinylcyclohexne diepoxide (VCD) causes ovotoxicity

in mice resulting in premature ovarian failure (Mayer et al., 2002). VCD accelerates the

natural process of atresia (apoptosis) in small pre-antral (primordial and primary) follicles

leading to a loss of ovarian follicles (Hu et al., 2001a, b; Springer et al., 1996a, b). VCD

is the ultimate ovotoxic metabolite, and the diepoxides of VCD can be detoxified by

microsomal epoxide hydrolase (mEH) enzyme to an inactive tetrol (4-[1,2-

dihydroxy]ethyl-1,2-dihydroxycyclohexane) metabolite (Fig. 1.2; Keller et al., 1997;

Salyers, 1995).

112

As with VCD, the polycyclic aromatic hydrocarbon (PAH), 7,12-

dimethylbenz[a]anthracene (DMBA), has also been shown to cause follicle loss in the

ovary resulting in a decrease in ovarian volume (Mattison, 1980; Weitzman et al., 1992).

Similar to VCD, DMBA depletes follicles in an apoptotic dependent manner (Matikainen

et al., 2001). However, unlike the ovotoxicant VCD, DMBA requires bioactivation to a

3,4-diol-1,2-epoxide to induce follicle loss (Shiromizu and Mattison, 1985).

Bioactivation of DMBA involves several phase I enzymes including mEH (Fig. 1.3;

Miyata et al., 1999, 2002; Buters et al., 2003). Thus, unlike with VCD, mEH is involved

in the bioactivation of DMBA to the ovotoxic form.

Studies have shown that the ovary has the capacity to metabolize VCD and

DMBA via mEH. Immunofluorescence microscopy demonstrates that mEH protein is

expressed in all ovarian compartments with the highest expression in the interstitial

compartment in B6C3F1 mice. mEH mRNA level and catalytic activity in small pre-

antral follicles (VCD target) increased following dosing of mice with VCD (Cannady et

al., 2002). Studies conducted in Chapter 5 suggest that ovarian mEH is involved in

DMBA-induced ovotoxicty. DMBA substantially increased mEH mRNA levels by 2d,

and this was followed by a marked increase in the rate of DMBA-induced follicle loss at

4d. The mEH inhibitor, cyclohexene oxide (CHO), prevented DMBA-induced follicle

loss in the B6C3F1 ovarian cultures. Taken together, these studies suggest that ovarian

mEH plays a role in the detoxification of VCD to the inactive metabolite, and

bioactivation of DMBA to the ovotoxicant.

113

Studies in Chapter 3 (Fig. 3.1) using the whole ovary culture system show that

CYP 2E1 wild-type (+/+) ovaries were more sensitive to the ovotoxic effects of VCD

(30µM) compared to that of null (-/-) ovaries. VCD depleted primordial and primary

follicles in CYP 2E1 wild-type (+/+) ovaries following 15d of culture, while some

primary follicles still remained in the CYP 2E1 null (-/-) ovaries. Because VCD contains

two epoxides that are not substrates for CYP 2E1 enzyme, it has been presumed that CYP

2E1 is not involved in the metabolism of VCD. As stated previously, VCD is thought to

be metabolized by mEH enzyme to an inactive tetrol. Thus, this difference in toxicity

between the two groups of mice might be due to a difference in mEH expression between

CYP 2E1 wild-type (+/+) and null (-/-) ovaries.

This study was designed to further evaluate the difference in ovotoxicity between

CYP 2E1 wild-type (+/+) and null (-/-) ovaries in culture using two substrates of mEH

(VCD and DMBA) that are not substrates for CYP 2E1. Since both of these compounds

are not thought to be metabolized by CYP 2E1, the hypothesis is that there will be no

difference in ovotoxicity induced by VCD or DMBA between CYP 2E1 wild-type (+/+)

and null (-/-) ovaries.

Experimental Methods

Ovaries from postnatal day (PND) 4 CYP 2E1 +/+ (129S1/SvImJ) and -/- mice

were incubated in control media (no treatment; vehicle control for VCD), media

containing 1% DMSO (vehicle control for DMBA), VCD, or DMBA for 15d. Following

114

incubation, ovaries were prepared for histological evaluation of follicle numbers. All

methods are described in detail in Chapter 2.

Results

VCD-induced follicle loss in CYP 2E1 wild-type and null mouse ovaries

Follicle loss was evaluated in PND4 CYP 2E1 wild-type (+/+) and null (-/-)

mouse ovaries cultured with VCD and medium control (no treatment) for 15d (Fig. 6.1).

Loss (p < 0.05) of primordial follicles was observed at VCD concentrations ≥ 5µM in

both CYP 2E1 wild-type (+/+) and null (-/-) ovaries compared to medium controls. The

loss of primordial follicles in CYP 2E1 wild-type (+/+) ovaries at 5 (86% loss) and 10µM

(96% loss) concentrations was significantly greater (p < 0.05) than in CYP 2E1 null (-/-)

ovaries (12% loss at 5µM; 55% loss at 10µM; Fig. 6.1A). Significant loss (p < 0.05) of

primary follicles was observed at concentrations ≥ 5µM in CYP 2E1 wild-type (+/+)

ovaries compared to medium controls. In CYP 2E1 null (-/-) ovaries, primary follicle

loss (p < 0.05) was detected at concentrations ≥ 15µM compared to medium control (Fig.

6.1B). In the 129S1/SvImJ PND4 ovary, secondary follicles are rarely observed, and

antral follicles are not seen.

115

Figure 6.1: VCD - Induced Ovotoxicity in CYP 2E1 Wild-Type and Null Ovarian

Cultures. Ovaries from PND4 CYP 2E1 wild-type (+/+) and null (-/-) mice were

cultured with medium control or VCD (5, 10, 15, 20, or 25µM) for 15d. Following

incubation, ovaries were collected, and processed for histological evaluation as described

in materials and methods. Healthy (A) primordial and (B) primary follicles were

classified and counted. Values are mean ± SE total follicles counted/ovary, n=5;

different letters differ from one another within each follicle type, p<0.05.

116

DMBA-induced follicle loss in CYP 2E1 wild-type and null mouse ovaries

Follicle loss was evaluated in PND4 CYP 2E1 wild-type (+/+) and null (-/-)

mouse ovaries cultured with DMBA and vehicle control (1% DMSO) for 15d (Fig. 6.2).

DMBA reduced (p < 0.05) primordial follicles in CYP 2E1 wild-type (+/+) and null (-/-)

ovaries at all concentrations used. Primordial follicles were similarly decreased (p <

0.05) in ovaries from CYP 2E1 wild-type (+/+) and null (-/-) mice at 0.5µM DMBA.

However, at ≥ 0.625µM there was a greater (p < 0.05) decrease in primordial follicles in

null (-/-; 99% loss) ovaries as compared with wild-type (+/+; 23% loss) ovaries (Fig.

6.2A). Primary follicles in ovaries from wild-type (+/+) mice were unaffected by DMBA

at any concentration; however, they were reduced (≥ 93%; p < 0.05) in null (-/-) ovaries

at concentrations ≥ 0.625µM (Fig. 6.2B).

117

Figure 6.2: DMBA - Induced Ovotoxicity in CYP 2E1 Wild-Type and Null Ovarian

Cultures. Ovaries from PND4 CYP 2E1 wild-type (+/+) and null (-/-) mice were

cultured with vehicle control (1% DMSO) or DMBA (0.5, 0.625, 0.75, or 1 µM) for 15d.

Following incubation, ovaries were collected, and processed for histological evaluation as

described in materials and methods. Healthy (A) primordial and (B) primary follicles

were classified and counted. Values are mean ± SE total follicles counted/ovary, n=5;

different letters differ from one another within each follicle type, p<0.05.

118

Discussion

VCD is the ultimate ovarian toxicant and is metabolized to an inactive tetrol (4-

[1,2-dihydroxy]ethyl-1,2- dihydroxycyclohexane) by mEH (Keller et al., 1997; Salyers,

1995). The ovarian toxicant DMBA is also a substrate for mEH. However, unlike VCD,

DMBA is the parent form and is bioactivated to the ovarian toxicant by mEH (Savas et

al., 1997; Shimada et al., 2001). Thus, mEH plays an interesting role in the fate of these

two ovotoxicants; where VCD is detoxified by mEH to an inactive metabolite, and

DMBA is bioactivated by mEH to the ovotoxicant.

Previous studies (Chapter III, Fig. 3.1) have shown a difference in toxicity

induced by VCD (30µM) between CYP 2E1 wild-type (+/+) and null (-/-) ovaries

following 15d of culture. In this study, CYP 2E1 null (-/-) ovaries were less sensitive to

toxicity induced by VCD compared to that of CYP 2E1 wild-type (+/+) ovaries. VCD

contains two epoxides that are hydrolyzed by mEH, and thus, not a substrate for CYP 450

related action. Additionally, DMBA is a large molecule (mw = 256); therefore, not

considered to be a substrate for CYP 2E1 enzyme. CYP 2E1 is thought to only be

involved in the epoxidation of small molecules (mw < 200; Guengerich et al., 1991).

Since CYP 2E1 is not involved in the metabolism of VCD, this difference in

toxicity is interesting, and could be due to a possible difference in mEH level between

CYP 2E1 wild-type (+/+) and null (-/-) ovaries. Therefore, this study was designed to

further evaluate the difference in toxicity between CYP 2E1 wild-type (+/+) and null (-/-)

ovaries, that could be attributed to a difference in mEH level, using both VCD and

DMBA.

119

CYP 2E1 wild-type (+/+) and null (-/-) ovaries were cultured with various

concentrations of VCD and DMBA for 15d. VCD induced primordial follicle loss (p <

0.05) in both wild-type (+/+) and null (-/-) ovaries at concentrations ≥ 5µM. However,

the loss of primordial follicles induced by VCD in CYP 2E1 wild-type (+/+) ovaries was

greater (p < 0.05) at 5 and 10µM concentrations. Even though, at concentrations ≥ 15µM

VCD-induced follicle loss was similar between the two groups, this greater sensitivity of

wild-type (+/+) ovaries was observed at all concentrations. VCD-induced primary

follicle loss (p < 0.05) was seen at concentrations ≥ 5µM in CYP 2E1 wild-type (+/+)

ovaries. In CYP 2E1 null (-/-) ovaries VCD-induced follicle loss was seen at

concentrations ≥ 15µM. Thus, as with primordial follicles, primary follicles in CYP 2E1

wild-type (+/+) ovaries were more sensitive to VCD-induced toxicity.

Unlike VCD, CYP 2E1 wild-type (+/+) ovaries were less sensitive to toxicity

induced by DMBA. DMBA-induced greater (p < 0.05) primordial follicle loss in CYP

2E1 null (-/-) ovaries at concentrations ≥ 0.625µM compared to that of wild-type (+/+)

ovaries in culture. Primary follicle loss was also greater (p < 0.05) in the wild-type (+/+)

ovaries compared to that of the null (-/-) ovaries in culture at DMBA concentrations ≥

0.625µM.

Because mEH is involved in bioactivation of DMBA rather than detoxification as

with VCD, it can be hypothesized that perhaps a common basis for these differences

might be differences in mEH enzyme expression between CYP 2E1 wild-type (+/+) and

null (-/-) ovaries, where CYP 2E1 null (-/-) ovarian expression of mEH is greater. Thus,

more mEH in the CYP 2E1 null (-/-) ovaries would be better able to detoxify VCD,

120

leading to less toxicity in null (-/-) ovaries. Conversely, higher expression of mEH in the

null (-/-) ovary would lead to greater DMBA bioactivation, leading to more toxicity in

CYP 2E1 null (-/-) ovaries.

Although not significant, studies by Hadri et al. (2004) have shown that female

CYP 2E1 wild-type (+/+) mice express more basal mEH protein in the liver compared to

that of null (-/-) mice. mEH protein level in 28d old CYP 2E1 wild-type (+/+) and null

(-/-) mouse ovaries was evaluated to determine whether there is a difference between

these two groups. Basal expression of mEH was higher in the CYP 2E1 wild-type (+/+)

as compared with CYP 2E1 null (-/-) ovaries (data not shown). Thus, both of these

studies are inconsistent with the observation that ovotoxicity was greater with VCD and

less with DMBA in CYP 2E1 wild-type (+/+) ovaries compared to that of the null (-/-)

ovaries. However, those studies evaluated basal mEH expression, VCD and DMBA can

both induce mEH in the ovary. Studies by Cannady et al. (2002) and studies conducted

in Chapter 5 have shown that mEH is inducible at the transcriptional and translational

level following in vivo VCD dosing and in vitro DMBA incubations. Preliminary data

show that VCD (5µM; 15d) treatment induces mEH mRNA in CYP 2E1 null

(-/-) ovaries to a greater extent than in CYP 2E1 wild-type (+/+) ovaries (Keating et al.,

2007). Therefore, it is possible that CYP 2E1 enzyme suppresses the induction of mEH

following exposure to xenobiotics in CYP 2E1 wild-type (+/+) mice. Further studies are

needed to evaluate the differences between CYP 2E1 wild-type (+/+) and null (-/-)

ovarian mEH induction following VCD and DMBA treatment in culture.

121

In summary, the enhanced ovotoxicity for DMBA and reduced ovotoxicity

observed for VCD in CYP 2E1 null (-/-) mouse ovaries suggest a role for CYP 2E1 in

regulating ovarian mEH activity.

122

CHAPTER 7

SUMMARY: THE ROLE OF OVARIAN METABOLISM IN 4-

VINYLCYCLOHEXENE METABOLITES AND 7,12-

DIMETHYLBENZ[A]ANTHRACENE-INDUCED OVOTOXICITY IN MICE

Chemicals that cause loss of ovarian follicles are of a toxicological importance,

because a female is born with a finite number of primordial follicles. Premature loss of

this follicle pool can accelerate the time to ovarian senescence or menopause in women,

ultimately reducing the reproductive life span. Women are postponing the start of a

family for later reproductive years; therefore, this decrease in time to menopause and

shortened reproductive life span can reduce child bearing years (Hoyer, 1997). Recent

studies suggest a correlation between exposure to xenobiotics, and reduced fertility

among females world wide (Mlynarcikova et al., 2005; Anderson et al., 2006; Kim et al.,

1996). Furthermore, in recent years there has been an increase in the use of in vitro

fertilization, which can be costly. Thus, the effects of xenobiotics on the female ovary is

both a health and an economic concern (Hoyer, 1997). Work conducted in this

dissertation evaluated the potential role of ovarian metabolism (bioactivation and

detoxification) in the resulting ovotoxicity of several environmental chemicals (VCM,

VCD and DMBA) using a novel neonatal mouse ovary culture system.

To address ovarian bioactivation, two phase I enzyme systems (CYP 450 isoform

and mEH) in the ovary were evaluated using two different environmental chemicals

123

(VCM and DMBA). Findings from studies with culture of ovaries from neonatal CYP

2E1 wild-type (+/+) or null (-/-) mice showed that ovarian CYP 2E1 is required for

VCM-induced ovotoxicity. Because structure activity studies suggest (Doerr et al., 1995)

that VCD is the ovotoxic metabolite, it can be presumed that this ovotoxicity is due to

ovarian metabolism of VCM to VCD via CYP 2E1. LC/MS analysis provided further

support for this hypothesis. The ovotoxic metabolite, VCD, was detected in culture

media from CYP 2E1 wild-type (+/+) ovaries incubated with VCM, but not CYP 2E1

null (-/-) ovary culture media.

A study was conducted to estimate hepatic contributions to VCM-induced

ovotoxicty. In vivo dosing with VCH/VCM/VCD resulted in a similar degree of

ovotoxicity between CYP 2E1 wild-type (+/+) and null (-/-) mice, which was also

comparable with ovotoxicity seen in B6C3F1 mice dosed with these three compounds

(Smith et al., 1990a). Therefore, this finding suggests that in the presence of hepatic

bioactivation, ovarian contributions to ovotoxicity are minimal.

Human exposure to VCH/VCM/VCD is limited. Threshold exposure limits have

been set by the National Institute for Occupational Safety and Health (NIOSH) for

workers in manufacturing plants that utilize these compounds. However, human

exposure to other ovotoxicants that require bioactivation to induce ovotoxicity such as

DMBA (found in cigarette smoke and car exhaust) is not well regulated. Therefore, the

potential role of ovarian mEH in bioactivation of DMBA was evaluated by assessing

DMBA-induced ovotoxicity in the culture system.

124

Substantial increases in mEH mRNA levels were observed 2d following

incubation of ovaries with DMBA. This corresponded to a large increase in the rate of

follicle loss at 4d. The mEH inhibitor, CHO, prevented DMBA-induced follicle loss

following in vitro culture. Since DMBA metabolism to the ovotoxicant involves mEH

activity, these studies suggest that ovarian mEH participates in this process. Studies

conducted here and studies by Borman et al. (2000) have shown that DMBA is a more

potent ovotoxicant when compared with VCD. A similar degree of toxicity was observed

with 250nM DMBA compared to 30µM VCD following 15d of culture. Thus,

disposition of small amounts of DMBA to the ovary, and subsequent ovarian

bioactivation has the potential to markedly reduce follicle numbers compared to other

less potent ovotoxicants.

Those studies showed that the ovary expresses phase I enzymes and has the capacity

to bioactivate ovotoxicants. However, it is difficult to determine the relevance of this

ovarian bioactivation in vivo, in the presence of hepatic tissue due to the large metabolic

capacity of the liver. For VCM ovarian bioactivation was shown to not likely play a

major role. In the case of more potent ovotoxicants, such as DMBA, ovarian metabolism

might be of more importance.

The potential role of ovarian detoxification was also addressed in this dissertation.

Previous studies have focused on the role of mEH in detoxification of VCD; however,

studies conducted in this dissertation showed a novel ovarian VCD detoxification

pathway via GSH conjugation. VCD-mono and diGSH adducts were detected in mouse

liver and ovarian homogenates following in vivo VCD dosing. Both mono and diGSH

125

adducts were also detected in culture media from ovaries incubated with VCD at 2-4d.

However, following longer time points VCD-GSH adducts were not detected, suggesting

ovarian detoxification capacity was exhausted. These studies suggest that the ovary has

the capacity to detoxify locally formed and circulating VCD via GSH conjugation;

however, following repeated exposure this capacity might be overwhelmed. Further

studies need to be conducted to determine if the VCD-GSH adducts are in fact inactive

metabolites. These studies along with past studies suggest that VCD can be detoxified in

the ovary by mEH and GSH conjugation. However, because the appropriate capacity to

detect the tetrol metabolite in ovary culture media was not available, studies conducted

here did not distinguish which detoxification pathway is favored in the mouse ovary.

Collectively, work conducted in this dissertation show that the ovary has the

capacity to both bioactivate and detoxify ovotoxicants. Work conducted here utilized an

ovarian culture system to evaluate ovary-specific capabilities; therefore, future studies

should focus on the significance of ovarian metabolism in vivo leading to ovarian

toxicity. Even though this might be a difficult task due to the large metabolic capacity of

the liver, novel methodology such as conditional knock out mouse models are available

for answering this question. The capacity of the ovary to detoxify ovarian toxicants, such

as VCD, can further complicate this issue. At least in the case of VCD, it can be

presumed that ovotoxicity occurs due to exhaustion of the detoxification capacity. Even

with this detoxification capacity in place, present and previous studies suggest that

ovarian follicles can be targeted by: 1) compounds in circulation due to hepatic

metabolism and 2) locally formed ovotoxicants due to ovarian metabolism. Thus,

126

understanding ovarian toxicity induced by chemicals needs to consider many factors such

as hepatic bioactivation and detoxification, ovarian bioactivation and detoxification, and

potency (t1/2) of different ovotoxicants.

The work presented here has provided support for an interesting direction that

could be taken to answer several biologically relevant questions: Does the ovary directly

influence effects from exposure to xenobiotics by bioactivating or detoxifying chemicals?

If so, in some cases can it amplify, and in others attenuate effects from a toxic insult?

Does the ovary have some say in determining its own fate by exposure to environmental

chemicals.

127

REFERENCES

Anderson, R.A., Themmen, A.P.N., Al-Qahtani, A., Groome, N.P., and Cameron, D.A. (2006). The effects of chemotherapy and long-term gonadotrophin suppression on the ovarian reserve in premenopausal women with breast cancer. Hum. Reprod. 10, 2583-2592. Baird, D.D. and Wilcox, A.J. (1985). Cigarette smoking associated with delayed conception. JAMA. 20, 2979-2983. Basler, A., and Rohrborn, G. (1976). Chromosome aberration in oocytes of NMR1 mice and bone marrow cells of Chinese hamster induced with 3,4-benzpyrene. Mutat. Res. 38,327-332. Becedas, L., Romert, L., Toft, E., Jenssen, D., DePierre, J.W., and Ahlberg-Bengtsson, M. (1993). Metabolism of polycyclic aromatic hydrocarbons to mutagenic species by rat and porcine ovarian granulosa cells: detection by cocultivation with V79 Chinese hamster cells. Reprod. Toxicol. 7, 219-224. Bengtsson, M., Montelius, J., Mankowitz, L., and Rydstrom, J. (1983). Metabolism of polycyclic aromatic hydrocarbons in the rat ovary. Biochem. Pharacol. 32, 129-136. Borman., S.M., VanDePol, B.J., Kao, S., Thompson, K.E., Sipes, I.G., and Hoyer, P.B. (1999). A single dose of the ovotoxicant 4-vinylcyclohexene diepoxide is protective in rat primary ovarian follicles. Toxicol. Appl. Pharmacol. 158, 244-252. Borman, S.M., Christian, P.J., Sipes, I.G., and Hoyer, P.B. (2000). Ovotoxicity in female Fisher rats and B6 mice induced by low-dose exposure to three polycyclic aromatic hydrocarbons: comparison through calculation of an ovotoxic index. Toxicol. Appl. Pharmacol. 167, 191-198. Buters, J., Quintanilla-Martinez, L., Schober, W., Soballa, V.J., Hintermair, J., Wolff, T., Gonzalez, F.J., and Greeim, H. (2003). CYP1B1 determines susceptibility to low doses of 7,12-dimethylbenz[a]anthracene-induced ovarian cancers in mice: correlation of CYP1B1-mediated DNA adducts with carcinogenicity. Carcinogenesis. 24, 327-334. Cannady, E.A., Dyer, C.A., Christian, P. J., Sipes, I.G., and Hoyer, P.B. (2002). Expression and activity of microsomal expoxide hydrolase in follicles isolated from mouse ovaries. Toxicol. Sci. 68, 24-31. Cannady, E.A., Dyer, C.A., Christian, P. J., Sipes, I.G., and Hoyer, P.B. (2003). Expression and activity of cytochrome P450 2E1, 2A, and 2B in the mouse ovary: The effect of 4-vinylcyclohexene and its diepoxide metabolite. Toxicol. Sci. 73, 423-430.

128

Chhabra, R.S., Elwell, M.R., and Peters, A. (1990). Toxicity of 4-vinyl-1-cyclohexene diepoxide after 13 weeks of dermal and oral exposure in rats and mice. Fundam. Appl. Toxicol. 14, 745-751. Daniell, H.W. (1978). Smoking, obesity, and the menopause, Lancett. 2, 373. Davis, B.J., Maronpot, R.R., and Heindel, J.J. (1994). Di-(2-ethylhexyl)phthalate suppress estradiol and ovulation in cycling rats. Toxicol Appl. Pharmacol. 128, 216-223. Devine, P.J., Sipes, I.G., and Hoyer, P.B. (2001). Effect of 4-vinylcyclohexene diepoxide dosing in rats on GSH levels in liver and ovaries. Toxicol. Sci. 62, 315-320. Devine, P.J., Sipes, I.G., Skinner, M.K., and Hoyer, P.B. (2002a). Characterization of a rat in vitro ovarian culture system to study the ovarian toxicant 4-vinylcyclohexene diepoxide. Toxicol. Appl. Pharmacol. 184, 107-115. Devine, P.J., Rajapaksa, K.S., and Hoyer, P.B. (2002b). In vitro ovarian tissue and organ culture: a review. Front. Biosci. 1, d1979-1989. Devine, P.J., Sipes, I.G., and Hoyer, P.B. (2004). Initiation of delayed ovotoxicity by in vitro and in vivo exposure of rat ovaries to 4-vinylcyclohexene diepoxide. Reprod. Toxicol. 19, 71-77. Diagaradjane, P., Yaseen, M.A., Yu, J., Wong, M.S., and Anvari, B. (2006). Synchronous fluorescence spectroscopic characterization of DMBA-TPA-induced squamous cell carcinoma in mice. J. Biomed. Opt. 014012. Doerr, J.K., Hooser, S.B., Smith, B.J., and Sipes, I.G. (1995). Ovarian toxicity of 4- vinylcyclohexene and related olefins in B6C3F1 mice: role of diepoxides. Chem. Res. Toxicol. 8, 963-969. Doerr-Stevens, J.K., Liu, J., Stevens, G.J., Kraner, J.C., Fontaine, S. M., Halpert, J.R., and Sipes, I.G. (1999). Induction of cytochrome P450 enzymes after repeated exposure to 4-vinylcyclohexene in B6C3F1 mice. Drug Metab. Dispos. 27, 281-287. Eliasson, M., Brock, S., and Ahlberg-Bengtsson, M. (1997). Evidence for mitochondrial metabolism of 7,12-dimethylbenz[a]anthracene in porcine ovaries: comparison with microsomal metabolism. Toxicology. 122, 11-21. Everson, R.B., Sandler, D.P., Wilcox, A.J., Schreinemachers, D., Shore, D.L., and Weinberg, C. (1986). Effect of passive exposure to smoking on age at natural menopause. Br. Med. J. (Clin. Res. Ed). 27, 792.

129

Flaws, J.A., Doerr, J.K., Sipes, I.G., and Hoyer, P.B. (1994a). Destruction of preantral follicles in adult rats by 4-vinyl-1-cyclohexene diepoxide. Reprod. Toxicol. 8, 509-514. Flaws, J.A., Slayer, K.L., Sipes, I.G., and Hoyer, P.B. (1994b). Reduced ability of rat preantral ovarian follicles to metabolize 4-vinyl-1-cyclohexene diepoxide in vitro. Toxicol. Appl. Pharmacol. 126, 286-294. Flaws, J. A., and Hirshfield, A.N. (1997). Reproductive, developmental, and endocrine toxicology, in Sipes, I.G., McQueen, C.A., and Gandolfi, A.J. (eds): Comprehensive Toxicology: Vol 10, Oxford, England: Elsevier, 283-291. Fontaine, S.M., Hoyer, P.B., Halpert, J.R., and Sipes, I.G. (2001a). Role of induction of specific hepatic cytochrome P450 isoforms in epoxidation of 4-vinylcyclohexene. Drug Metab. Dispos. 29, 1236-1242. Fontaine, S.M., Hoyer, P.B., and Sipes, I.G. (2001b). Evaluation of hepatic cytochrome P4502E1 in the species-dependent bioactivation of 4-vinylcyclohexene. Life Sci. 69, 923-934. Franczak, A., Nynca, A., Valdez, K., Mizinga, K.M., and Petroff, B.K. (2006). Effects of acute and chronic exposure to the aryl hydrocarbon receptor agonist 2,3,7,8-tetrachlorodibenzo-p-dioxin on the transition to reproductive senescence in female Sprague-Dawley rats. Biol. Reprod. 74, 125-130. Giannarini, C., Citti, L., Gervasi, P.G., and Turchi, G. (1981). Effect of 4-vinylcyclehexene and main oxirane metabolite on mouse hepatic microsomal enzyme and glutathione levels. Toxicol. Lett. 8, 115-121. Gregus, Z., and Klaassen, C.D. (2001). Mechanisms of toxicology, in Klaassen, C.D. (ed): Casarett and Doull’s toxicology: the basic science of poisons 6th edition, McGraw-Hill Medical Publishing Division, 133-224. Grote, K., Andrade, A.J.M., Grande, S.W., Kuriyama, S.N., Talsness, C.E., Appel, K.E., and Chahoud, I. (2006). Effects of peripubertal exposure to triphenyltin on female sexual development of the rat. Toxicology. 222, 17-24. Guengerich, F.P., Kim, D.H., and Iwasaki, M. (1991). Role of human cyctochrome P-450 IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168-179. Hadri, L.E., Chanas, B., and Ghanayem, B.I. (2005). Comparative metabolism of methacrylonitrile and acrylonitrile to cyanide using cytochrome P4502E1 and microsomal epoxide hydrolase-null mice. Toxicol. Appl. Pharmacol. 205, 116-125.

130

Hattori, N., Fujiwara, H., Maeda, M., Fujii, S., and Ueda, M. (2000). Epoxide hydrolase affects estrogen production in the human ovary. Endocrinology. 141, 3353-3365. Hayakawa, T., and Udenfriend, S. (1975). Substrate and inhibitors of hepatic glutathione-S-epoxide transferase. Arch. Biochem. Biophys. 170, 438-451. Hooser, S.B., Parola, L.R., Van Ert, M.D., and Sipes, I.G. (1993). Differential ovotoxicity of 4-vinylcyclohexene and its analogue, 4-phenolcyclohexene. Toxicol. Appl. Pharmacol. 119, 302-305. Hooser, S.B., Douds, D.P., DeMerell, D.G., Hoyer, P.B., and Sipes, I.G. (1994). Long- term ovarian and gonadotropin changes in mice exposed to 4-vinylcyclohexene. Reprod. Toxicol. 8, 315-323. Hoyer, P.B. (1997). Female reproductive toxicology: introduction and overview, in Sipes, I.G., McQueen, C.A., and Gandolfi, A.J. (eds): Comprehensive Toxicology: Vol 10, Oxford, England: Elsevier, 249-254. Hu, X., Christian, P., Sipes, I.G., and Hoyer, P.B. (2001a). Expression and redistribution of cellular Bad, Bax, and Bcl-X(L) protein is associated with VCD-induced ovotoxicity in rats. Biol Reprod. 65, 1489-1495. Hu, X., Christian, P., Thompson, K.E., Sipes, I.G., and Hoyer, P.B. (2001b). Apoptosis induced in rats by 4-vinylcyclohexene diepoxide is associated with activation of the caspase cascades. Biol. Reprod. 65, 87-93. Jull, J.W. (1973). Ovarian tumorigenesis. Methods Cancer Res. 7, 131-186. Kang, K.W., Ryu, J.H., and Kim, S.G. (2001). Activation of phosphatidylinositol 3-kinase by oxidative stress leads to induction of microsomal epoxide hydrolase in H4IIE cells. Toxicol. Lett. 121, 191-197. Kanter, E.M., Walker, R.M., Marion, S.L., Brewer M., Hoyer, P.B., and Barton, J.K. (2006). Dual modality imaging of a novel rat model of ovarian carcinogenesis. J. Biomed. Opt. 4, 041123. Kao, S.W., Sipes, I.G., and Hoyer, P.B. (1999). Early effects of ovotoxicity induced by 4-vinylcyclohexnene diepoxide in rats and mice. Reprod. Toxicol. 13, 67-75. Kaufman, D.W., Slone, D., Rosenberg, L., Miettinen, O.S., and Shapiro, S. (1980). Cigarette smoking and age at natural menopause. Am. J. Public Health. 70, 420-422. Keating, A.F., Rajapaksa, K.S., Sipes, I.G., and Hoyer, P.B. (2007). Interactions between Cytochrome P450 2E1 and microsomal epoxide hydrolase in effects of 4-

131

Vinylcyclohexene diepoxide in the mouse ovary. Submitted to the SOT Annual Meeting. Charlotte, NC. Keller, D.A., Carpenter, S.C., Cagen, S.Z., and Reitman, F.A. (1997). In vitro metabolism of 4-vinylcyclohexne in rat and mouse liver, lung, and ovary. Toxicol. Appl. Pharmacol. 144, 36-44. Kerr, J.F.R, Wyllie, A.H., and Currie, A.R. (1972). Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer. 26, 239-257. Kim, Y., Jung, K., Hwang, T., Jung, G., Kim, H., Park, J., Kim, J., Park, J., Park, D., Park, S., Choi, K., and Moon. Y. (1996). Hematopoietic and reproductive hazards of Korean electronic workers exposed to solvents containing 2-bromopropane. Scand. J. Work. Environ. Health. 22, 387-391. Kim, S.G., and Kim, Y. (1992). Gender-related expression of rat microsomal epoxide hydrolase during maturation: post-transcriptional regulation. Mol. Pharmacol. 42, 75-81. Krishna, D.R., and Klotz, U. (1994). Extrahepatic metabolism of drugs in humans. Clin. Pharmacokinet. 26, 144-160. La Barbera, A. (1997). Differentiation and function of the female reproductive system, in Sipes, I.G., McQueen, C.A., and Gandolfi, A.J. (eds): Comprehensive Toxicology, Vol 10, Oxford, England: Elsevier, 255-281. Luderer, U., Kavanagh, T.J., White, C.C., and Faustman, E.M. (2001). Gonadotropin regulation of glutathione sysnthesis in the rat ovary. Reprod. Toxicol. 15, 495-504. Manoharn, K., and Ramesha Rao, A., (1980). Influence of exogenous estrogens on oocyte depletion induced by 7,12-dimethylbenz[a]antharacene in mice. Cancer Lett. 10,359-363. Matikainen, T., Perez, G.I., Jurisicova, A., Pru, J.K., Schlezinger, J.J., Ryu, H., Laine, J., Sakai, T., Korsmeyer, S., Casper, R., Sherr D.H., and Tilly, J.L. (2001). Aromatic hydrocarbon receptor-driven Bax gene expression is required for premature ovarian failure caused by biohazard environmental chemicals. Nature Genetics. 28, 355-360. Mattison, D.R. (1979). Diffrences in sensitivity of rat and mouse primordial oocytes to destruction by polcyclic aromatic hydrocarbons. Chem. Biol. Interact. 28, 133-137. Mattison, D.R. (1980). Morphology of oocyte and follicle destruction by polycyclic aromatic hydrocarbons in mice. Toxicol. Appl. Pharmacol. 53, 249-259.

132

Mattison, D.R., and Thorgeirsson, S.S. (1978). Smoking and industrial pollution, and their effect on menopause and ovarian cancer. Lancet. 1,187-188. Mattison, D.R., and Schulman, J.D. (1980). How xenobiotic compounds can destroy oocytes. Contemp. Ob. Gyn. 15,157-169. Mayer, L.P., Pearsall, N.A., Christian, P.J., Devine, P.J., Payne, C.M., McCuskey, M.K., Marion, S.L., Sipes, I.G., and Hoyer, P.B. (2002). Long-term effects of ovarian follicular depletion in rats by 4-vinylcyclohexene diepoxide. Reprod. Toxicol. 16, 775-781. Miyata, M., Kudo, G., Lee, Y., Yang, T.J., Gelboni, H.V., Fernandez-Salguero, P., Kimura, S., and Gonzalez, F.J. (1999). Target disruption of the microsomal epoxide hydrolase gene. The J. of Biolog. Chem. 274, 23963-23968. Miyata, M., Motoki, K., Tamura, E., Furukawa, M., Gonzalez, F.J., and Yamazoe, Y. (2002). Relative importance of maternal and embryonic microsomal epoxide hydrolase in 7,12-dimethylbenz[a]antharcene-induced developmental toxicity. Biochem. Pharmacol. 63, 1077-1084. Mlynarcikova, A., Fickova, M., Scsukova, S. (2005). Ovarian intrafollicular processes as a target for cigarette smoke components and selected environmental reproductive disruptors. Endocr. Regul. 39, 20-31. Mukhtar, H., Philpot, R.M., and Bend, J. R. (1978). The postnatal development of microsomal epoxide hydrolase, cytosolic glutathione S-transferase, and mitochondrial and microsomal cytochrome P450 in adrenals and ovaries of female rats. Drug Meatb. Dispos. 6, 577-583. National Toxicology Program (NTP). (1986). Toxicology and carcinogenesis studies of 4-vinylcyclohexhene in F344/N rats and B6C3F1 mice. U.S. Department of Health and Human Services, Public Health Services, National Institutes of Health, Research Triangle Park, NC, Technical Report 303. National Toxicology Program (NTP). (1989). Toxicology and carcinogenesis studies of 4-vinyl-1-cyclohexhene diepoxide in F344/N rats and B6C3F1 mice. U.S. Department of Health and Human Services, Public Health Services, National Institutes of Health, National Toxicology Program, Research Triangle Park, NC, Technical Report 362. Oesch, F. (1973). Mammalian epoxide hydrases: inducible enzymes catalyzing the inactivation of carcinogenic and cytotoxic metabolites derived from aromatic and olefinic compounds. Xenobiotica. 3, 305-340.

133

Otto, S., Bhattacharyya, K.K., and Jefcoate, C.R. (1992). Polycyclic aromatic hydrocarbon metabolism in rat adrenal, ovary, and testis microsomes is catalyzed by the same novel cytochrome P450 (P450RAP)*. Endocrinology. 131, 3067-3076. Parkinson, A. (2001). Biotransformation of xenobiotics, in Klaassen, C.D. (ed): Casarett and Doull’s toxicology: the basic science of poisons 6th edition, McGraw-Hill Medical Publishing Division, 133-224. Pedersen, T., and Peters, H. (1968). Proposal for a classification of oocytes and follicles in the mouse ovary. J. Reprod. Fertil. 17, 555-557. Plowchalk, D.R., and Mattison, D.R. (1991). Phosphoramide mustard is responsible for the ovarian toxicity of cyclophosphamide. Toxicol Appl. Pharmacol. 107, 472-481. Plowchalk, D.R., and Mattison, D.R. (1992). Reproductive toxicity of cyclophosphamide in C57BL/6N mouse. 1. Effects on ovarian structure and function. Reprod. Toxicol. 6,411-421. Porterfield, S.P. (1997). Female reproductive system: Endocrine Physiology, Mosby-Year Book Inc., 173-196. Russo, J., and Russo, I.H. (1996). Experimentally induced mammary tumors in rats. Breast Cancer Res. Treat. 39, 7-20. Salyers, K.L. (1995). Dissertation. Disposition and metabolism of 4-vinyl-1-cyclohexene diepoxide in female F-344 rats and B6C3F1 mice. The University of Arizona, Tucson, AZ. Savas, U., Carsten, C.P., and Jefcoate, C.R. (1997). Biological oxidation and P450 reactions: recombinant mouse CYP 1B1 expressed in Eschericie coli exhibits selective binding by polycyclic hrdrocarbons and metabolism which paralles C3H10T1/2 cell microsomes, but differs from human recombinant CYP 1B1. Arch Biochem Biophys. 15,181-192. Sawicki, J.T., Moschel, R.C., and Dipple, A. (1983). Involvement of both syn- and anti-dihydrodiol-epoxides in the binding of 7,12-dimethylbenz[a]anthracene to DNA in mouse embryo cell cultures. Cancer Res. 43, 3212-3218. Shimada, T., Oda, Y., Gillam, E.M., Guengerich, F.P., and Inoue, K. (2001). Meatabolic activation of polycyclic aromatic hydrocarbons and other procarcinogens by cytochrome P450 1A1 and P450 1B1 allelic variations and other human cytochrome P450 in salmonella typhimurium NM2009. Drug Meatb. Dispos. 29, 1176-1182.

134

Shimada, T., Sugie, A., Shindo, M., Nakajima, T., Azuma, E., Hashimoto, M., and Inoue, K. (2003). Tissue-specific induction of cytochrome P450 1A1 and 1B1 by polcyclic aromatic hydrocarbons and polychlorinated biphenyls in engineered C57BL/6J mice of arylhydrocarbon receptor gene. Toxicol. Appl. Pharmacol. 187, 1-10. Shimada, T., and Fujii-Kuriyama, Y. (2004). Metabolic activation of polycyclic aromatic hydrocarbons to carcinogens by cytochrome P450 1A1 and 1B1. Cancer Sci. 95, 1-6. Shiromizu, K., and Mattison, D.R. (1985). Murine oocyte destruction following intraovarian treatment with 3-methylcholanthrene or 7,12-dimethylbenz[a]anthracene: protection by alpha-napthoflavone. Teratog. Carcinog. Mutagen. 5, 463-472. Shiromizu, K., Thorgeirsson, S.S., and Mattison, D.R. (1984). Effect of cyclophosphamide on oocyte and follicle number in Sprague Dawley rats, C57BL/6N and DBA/2N mice. Pediatr. Pharmacol. 4, 213-221. Simmons, D.L., McQuiddy, P., and Kasper, C.B. (1987). Induction of the hepatic mixed-function oxidase system by synthetic glucocorticoids. J. Biol. Chem. 262, 326-332. Smith, B.J., Mattison, D.R., and Sipes, I.G. (1990a). The role of epoxidation in 4- vinylcyclohexene-induced ovarian toxicity. Toxicol. Appl. Pharmacol. 105, 372-381. Smith, B.J., Carter, D.E., and Sipes, I.G. (1990b). Comparison of the disposition and in vitro metabolism of 4-vinylcyclohexene in the female mouse and rat. Toxicol. Appl. Pharmacol. 105, 364-371. Smith, B.J., Sipes, I.G., Stevens, J.C., and Halpert, J.R. (1990c). The biochemical basis for the species difference in hepatic microsomal 4-vinylcyclohexene epoxidation between female mice and rats. Carciniogenesis. 11, 1951-1957. Springer, L.N., McAsey, M.E., Flaws, J. A., Tilly, J.L., Sipes, I.G., and Hoyer, P.B. (1996a). Involvement of apoptosis in 4-vinylcyclehexene diepoxide induced ovotoxicity in rats. Toxicol. Appl. Pharmacol. 139, 394-401. Springer, L.N., Tilly, J, L., Sipes, I.G., and Hoyer, P.B. (1996b). Enhanced expression of bax in small preantral follicles during 4-vinylcyclohexne diepoxide-induced ovotoxicity in the rat. Toxicol. Appl. Pharmacol. 139, 402-410. Tome, M.E., Baker, A. F., Powis, G., Payne, C. M., and Briehl, M. M. (2001). Catalase-overexpressing thymocytes are resistant to glucocorticoid-induced apoptosis and exhibit increase net tumor growth. Cancer Res. 61, 2766-2773.

135

Vigny, P., Brunissen, A, Phillips, D.H., Cooper, C.S., Hewer, A., Grover, P.L., and Sims, P. (1985). Metabolic activation of 7,12-dimethylbenz[a]anthracene in rat mammary tissue: fluorescence spectral characteristics of hydrocarbon-DNA adducts. Cancer Lett. 26, 51-59. Watabe, T., and Sawahata, T. (1976). Metabolism of the carcinogen bifunctional olefin oxides, 4vinyl-1-cyclohexene dioxide, by hepatic micrososmes. Biochem. Pharmac. 25,601-602. Weinberg, C.R., Wilcox, A.J., and Baird, D.D. (1989). Reduced fecundity in women with prenatal exposure to cigarette smoke. Am. J, Epidemiol. 129, 1072-1078. Weitzman, G. A., Miller, M.M., London, S.N., and Mattison, D. R. (1992). Morphometric assessment of the murine ovarian toxicity of 7,12-dimethylbenz[a]anthracene. Reprod. Toxicol. 6, 137-141. Yu, X., Kamijima, M., Ichihara, G., Li, W., Kitoh, J., Xie, Z., Shibata, E., Hisanaga, N., and Takeuchi, Y. (1999). 2-Bromopropane causes ovarian dysfunction by damaging primordial follicle and their oocytes in female rats. Toxicol. Appl. Pharmacol. 159, 185-193.