Chemico-Biological Interactions 170 (2007) 20–30

Trichloroethylene metabolism in the rat ovaryreduces oocyte fertilizability

Katherine Lily Wu, Trish Berger ∗Department of Animal Science, University of California, Davis One Shields Avenue, Davis, CA 95616, USA

Received 27 February 2007; received in revised form 20 June 2007; accepted 20 June 2007Available online 23 June 2007

Abstract

Exposure to trichloroethylene (TCE, an environmental toxicant) reduced oocyte fertilizability in the rat. In vivo, TCE may bemetabolized by cytochrome P450 dependent oxidation or glutathione conjugation in the liver or kidneys, respectively. CytochromeP450 dependent oxidation is the higher affinity pathway. The primary isoform of cytochrome P450 to metabolize TCE in the liver,cytochrome P450 2E1, is present in the rodent ovary. Ovarian metabolism of TCE by the oxidative pathway and the production ofreactive oxygen species (ROS) may occur given the presence of the metabolizing enzyme. The objectives of this study were to definethe sensitive interval of oocyte growth to TCE exposure, and to determine if TCE exposure resulted in the formation of ovarianprotein carbonyls, an indicator of oxidative damage. Rats were exposed to TCE in drinking water (0.45% TCE (v/v) in 3% Tween) or3% Tween (vehicle control) during three 4–5 day intervals of oocyte development preceding ovulation. Oocytes from TCE-exposedfemales were less fertilizable compared with vehicle-control oocytes. Immunohistochemical labeling of ovaries and Western blottingof ovarian proteins demonstrated TCE treatment induced a greater incidence of protein carbonyls compared with vehicle controls.

Protein carbonyl formation in the ovary is consistent with TCE metabolism by the cytochrome P450 pathway. Oxidative damagefollowing ovarian TCE metabolism or the presence of TCE metabolites may contribute to reduced oocyte fertilizability. In summary,these results indicate maturing oocytes are susceptible to very short in vivo exposures to TCE.© 2007 Elsevier Ireland Ltd. All rights reserved.

etabolis

Keywords: Ovary; In vitro fertilization; Oocyte; Trichloroethylene m

1. Introduction

Trichloroethylene (TCE), a chlorinated hydrocarbon,is primarily used as a solvent for degreasing metal parts;approximately 85% of the TCE produced in the United

States is used for cleaning metal [1]. Although TCE isnot naturally found in the environment, it is an environ-mental toxicant due to production, use, and disposal.

∗ Corresponding author. Tel.: +1 530 752 1267;fax: +1 530 752 0175.

E-mail address: tberger@ucdavis.edu (T. Berger).

0009-2797/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reservdoi:10.1016/j.cbi.2007.06.038

m

Trichloroethylene is detectable in underground watersources, surface water, and the air. Degradation rates forTCE vary in the environment based on the physical stateof TCE, and the half-life of TCE in groundwater rangesfrom 10.7 months to 4.5 years depending upon theconcentration of TCE [2]. Volatilization of TCE fromsurface water produces TCE vapor in the air, which hasa half-life of approximately 7 days [3]. Some removalof TCE from the air also occurs with precipitation

because of the moderate solubility of TCE in water,1.1 g/l [4]. Trichloroethylene has been detected in the airthroughout the United States with levels of TCE beingabout three times higher in urban areas compared with

ed.

ologica

rflw

hotutadswtol

agtooswrIrd0wwTncmhotd[smieta

poIo

K.L. Wu, T. Berger / Chemico-Bi

ural areas. Analysis of measurements taken in 1998rom 115 monitors in 14 different states indicated TCEevels in the air ranged between 0.01 and 3.90 �g/m3

ith a mean of 0.88 �g/m3 [4].Trichloroethylene is most frequently detected at

igh levels in persons exposed to TCE throughccupational degreasing operations, silk screening,axidermy, and electronic cleaning. The general pop-lation is likely to have low levels of exposureo TCE via the environment through inhalation ofmbient air, ingestion of drinking water, and/or trans-ermal absorption. Commercial products such as woodtains, varnishes, finishes, lubricants, adhesives, type-riter correction fluid, paint removers, and cleaners

hat contain TCE may also contribute to exposuref the general population at a low to moderateevel.

Studies in rats and mice have substantiated thedverse effects of TCE on male reproduction at theamete level [5,6]. There is some conflict as to whetherhe effects are specific to reproduction or a resultf general systemic toxicity due to the high levelsf TCE typically used in toxicology studies [7]. Intudies where no significant decreases in body weightere observed, it has been suggested that an effect on

eproduction is not confounded by systemic toxicity [7].n a study published in 2003 by DuTeaux et al., male ratseceived 0.205–0.293 g TCE/kg body weight/day viarinking water in the 0.2% TCE-treatment group, and.410–0.585 g TCE/kg body weight/day via drinkingater in the 0.4% TCE-treatment group [5,8]. Thereas no effect on final body weights as a result ofCE-treatment [5], suggesting systemic toxicity didot confound reproductive parameters. Histologicalhanges observed in the efferent ductule epithelium ofale rats exposed to TCE, oxidized proteins found in the

ead and midpiece of sperm, and the decreased abilityf sperm to fertilize untreated oocytes were attributedo TCE metabolites formed by the cytochrome P450ependent oxidative pathway within the efferent ducts5]. In mice, TCE exposure leads to impairment ofperm fertilizing ability [6]. Trichloroethylene-treatedice and control mice had no significant differences

n body weight, which suggests a lack of confoundingffects on reproduction from systemic toxicity. Thus,here is some evidence suggesting TCE exposure mayffect male gamete quality.

Decreased oocyte fertilizability, apparently inde-

endent of hormones, ovulatory concerns and weight,ccurred in female rats after in vivo exposure to TCE [9].n vitro exposure of mouse oocytes to three metabolitesf TCE, trichloroacetic acid, dichloroacetic acid, and

l Interactions 170 (2007) 20–30 21

trichloroethanol, but not to TCE itself also reduced fer-tilization rates [10]. Taken together, these observationssuggest metabolites of TCE produced in the ovary orelsewhere are responsible for the effects on oocytes.

Trichloroethylene metabolism occurs primarily in theliver and kidneys, although it may occur in other tis-sues. The two known pathways responsible for TCEmetabolism are oxidation by cytochrome P450 and con-jugation with glutathione [11]. The oxidative pathwayis the higher affinity pathway [12], and metabolizesmost of the TCE when pathway elements are presentor until the pathway is saturated. Enzymes involved inboth metabolic pathways are present in the ovary [13–15]which is consistent with the possibility that the ovary par-ticipates in TCE metabolism and production of bioactiveTCE metabolites. The purpose of this study was to deter-mine if there is a specific interval of sensitivity to TCEduring the 2 weeks of oocyte growth preceding ovula-tion and to determine if protein oxidation was a potentialmechanism for altered oocyte fertilizability.

2. Materials and methods

2.1. Animals

Simonson albino rats, a Sprague-Dawley derivedstrain, were used in this study. Female rats between 28and 45 days old, and male rats approximately 100 daysold came from the Department of Animal Science breed-ing colony at the University of California, Davis. Allanimals were housed under a 14-h light:10-h dark lightcycle in a temperature (70 ± 2 F) and humidity (40–70%)controlled facility. Rats had ad libitum access to PurinaFormulab 5008 rat chow (St. Louis, MO, USA). Betweenweaning (21 days of age) and treatment with TCE, ratshad ad libitum access to deionized water. The Universityof California, Davis Animal Use and Care Administra-tive Advisory Committee approved all animal use.

2.2. Chemicals and antibodies

Trichloroethylene (ACS reagent ≥99.5%), 2,4-dinitrophenylhydrazine ((DNPH), reagent grade 97%),and rabbit anti-dinitrophenyl (DNP) antibody were pur-chased from Sigma–Aldrich (St. Louis, MO, USA).Media components for in vitro fertilization (IVF)were cell-culture tested or molecular biology gradeand obtained from Fisher Scientific (Pittsburgh, PA,

USA) or Sigma–Aldrich. Water was obtained from aMilli-Q Synthesis system (Millipore, Bedford, MA,USA). Chorionic gonadotropin was obtained fromIntervet, Inc. (Millsboro, DE, USA). Pregnant mare

ologica

22 K.L. Wu, T. Berger / Chemico-Bi

serum gonadotropin was obtained from Sioux Bio-chemical, Inc. (Sioux Center, IA, USA). Trifluoroaceticacid and a bicinchoninic acid (BCA) protein assaywas purchased from Pierce Biotechnology, Inc. (Rock-ford, IL, USA). Citrisolv was acquired from FisherScientific. Antigen Unmasking Solution, Vector BlueAlkaline Phosphatase Substrate Kit III, and VectastainABC-AP kit were obtained from Vector Laborato-ries (Burlingame, CA, USA). Cover Safe mountingmedium was acquired from American Master Tech Sci-entific, Inc. (Lodi, CA, USA). The secondary antibody,peroxidase-conjugated donkey anti-rabbit IgG was pur-chased from Jackson ImmunoResearch Laboratories,Inc. (West Grove, PA, USA). Western Lightning West-ern Blot Chemiluminescence Reagent Plus was obtainedfrom Perkin-Elmer Life Sciences, Inc. (Wellesley, MA,USA).

2.3. Chemical treatment

Drinking water was made with water from a Milli-Q Synthesis system and contained 0.45% TCE (v/v)in 3% Tween for TCE-treated rats, and 3% Tween forvehicle-control rats [9]. Tween was used as a vehi-cle to dissolve TCE in drinking water. Each female inthe TCE-exposure group received an estimated 0.66 gTCE/kg body weight/day based on the average waterconsumption of 10 ml/100 g body weight [8]. A com-puter simulation model suggests this is a slightly highereffective concentration of TCE than the previouslyreported highest industrial concentration [16,17] and 2-week exposure to this level of TCE is known to reduceoocyte fertilizability [9]. Female rats were exposed toTCE or to vehicle for 4, 5, or 14 days for IVF andimmunohistochemistry trials. Rats received TCE-wateron days 1–5, 6–10, or 11–14 of the 2-week interval pre-ceding ovulation (oocytes collected the morning of day15). During the 2-week period, when females were notreceiving drinking water containing TCE (days 6–14,1–5 and 11–14, and 1–10), they received water con-taining Tween (vehicle). Females providing tissue forWestern blot analysis received 0.45% TCE (v/v) in3% Tween or vehicle alone for 5 days similar to theday 11–14 treatment without the additional exposure toTween. Ovaries recovered from females exposed to TCEfor the 2 or 3 previous days provided tissue for micro-some isolation. Drinking water was freshly prepared foreach treatment group, stored in tightly capped bottles,

and protected from light exposure. Glass water bottles,fit with rubber and stainless steel stoppers, were refilledevery 24–48 h to minimize headspace and volatilizationof TCE.

l Interactions 170 (2007) 20–30

2.4. Gamete retrieval and in vitro fertilization

Gamete retrieval and IVF were performed as previ-ously described [9,18]. Briefly, one male rat, 100 daysor older, was used as a semen donor for all femalesin a replicate of IVF. The male was sacrificed byCO2 asphyxiation followed by cervical dislocation ina warm room (30–32 ◦C) on the morning the oocyteswere collected from females. Cauda epididymides andthe attached vas deferens were dissected using sterilesurgical instruments. Each epididymis and vas defer-ens was rinsed in warm fertilization media (94.19 mMNaCl, 4.75 mM KCl, 1.71 mM CaCl2, 1.17 mMKH2PO4, 1.17 mM MgSO4, 25.1 mM NaHCO3, 26 mMNa lactate, 0.5 mM Na pyruvate, 5.5 mM glucose,50 �g gentamycin sulfate/ml, 4 mg bovine serumalbumin (BSA)/ml, with 20 mM [4-(2-hydroxyethyl)-1-piperazine]ethanesulfonic acid (HEPES)) to remove fatand hair. Sperm were flushed retrogradely through a ver-tical incision in the epididymis with a 23 gauge needleinserted into the vas deferens. Collected spermatozoawere evaluated with phase contrast optics. One millilitreof 7 × 106 sperm/ml was prepared in fertilization mediaand allowed to capacitate in a 37 ◦C, 5% CO2 incubatorfor 3 h.

Female rats in each replicate were induced toovulate by intraperitoneal (i.p.) injection of pregnantmare serum gonadotropin (PMSG, 15 IU) on day 12,followed 48 h later by an i.p. injection of humanchorionic gonadotropin (hCG, 15 IU) [9,19]. Femalerats were sacrificed by CO2 asphyxiation followedby cervical dislocation 16–18 h after receiving hCG.Oviducts were removed and cumulus masses werereleased into warm saline-BSA (0.9% NaCl, 1 mgBSA/ml). Cumulus masses and zonae pellucidae wereremoved with hyaluronidase (1 mg/ml saline-BSA) andacid Tyrode’s [20], respectively, and oocytes wererinsed with saline-BSA. Oocytes were then groupedin clusters of 10 and placed in 100 �l drops of pre-equilibrated fertilization media under water-saturateddimethylpolysiloxane (DMPS; Sigma–Aldrich). Eachgroup of oocytes was incubated with 10 �l of 7 × 106

capacitated rat sperm/ml.After 20 h of incubation at 37 ◦C in 5% CO2 in air,

oocytes were rinsed through three drops of fertiliza-tion media to remove loosely bound sperm. Oocytesand associated sperm were then stained with Hoescht33342 (Sigma–Aldrich, 0.08 mg/ml fertilization media)

for 10 min. Oocytes were rinsed with fertilization media,placed on slides, compressed with a coverslip rest-ing on vaseline-paraffin, and sealed with nail polish.Oocytes were examined by fluorescence microscopy

ologica

(fat

2

TdvfisdpUws

2i

rhAiocDwiicomgFitwsPmetiuQ

Wf

K.L. Wu, T. Berger / Chemico-Bi

400× magnification), and the percentage of oocytesertilized (decondensed sperm heads) and the percent-ge of oocytes binding sperm were calculated for eachreatment in each replicate.

.5. Tissue preparation

Ovaries were collected from females that receivedCE on days 1–5, 6–10, and 11–14 on the morning ofay 15 along with ovaries from females that receivedehicle water and TCE water for 2 weeks. Tissue wasxed with 4% paraformaldehyde in phosphate bufferedaline (PBS). Ovaries were rinsed in PBS then dehy-rated with 30, 50, and 70% ethanol solutions prior toaraffin embedding (Tissue-Tek VIP5, Sakura FinetekSA, Torrance, CA). Paraffin-embedded ovarian tissueas cut in 5 �m sections and allowed to dry on glass

lides.

.6. Immunochemical detection of oxidized proteinn the rat ovary

Ovarian tissue sections were deparaffinized with Cit-isolv and rehydrated in graded ethanols. Subsequently,eat-induced antigen retrieval was performed usingntigen Unmasking Solution. The tissue was rehydrated

n PBS (pH 7.4) and underwent an endogenous per-xidase quench with 0.3% H2O2 in methanol. Proteinarbonyls were derivatized using a solution of 5 mM 2,4-NPH in PBS (pH 6.3) for 45 min [5]. Ovarian sectionsere then rinsed four times with PBS, treated with block-

ng solution (Vectastain ABC-AP kit) for 20 min, andncubated with rabbit anti-DNP antibody (1:800 in PBSontaining 2% normal goat serum). Additional sectionsf tissue from the same ovaries were incubated with nor-al rabbit serum (1:800 in PBS containing 2% normal

oat serum) to serve as controls for the primary antibody.ollowing four rinses with PBS, ovarian sections were

ncubated with biotinylated secondary antibody (Vec-astain ABC-AP kit) for 30 min. Sections were rinsedith PBS, then incubated with alkaline phosphatase sub-

trate solution (Vectastain ABC-AP kit), and rinsed withBS again. All slides were coverslipped with Cover Safeounting medium. Ovarian sections were examined for

vidence of protein carbonyls using an Olympus BH-2rinocular phase contrast microscope (Olympus Amer-ca, Inc., Melville, NY, USA) and images were takensing a QImaging MicroPublisher 3.3 RTV camera and

Capture Pro software (Burnaby, BC, Canada).Protein oxidation was quantitatively assessed using

estern blots. Ovaries from rats treated with TCEor 5 days were collected, weighed, and immediately

l Interactions 170 (2007) 20–30 23

immersed in 1 ml of 0.1 M phosphate buffer (pH 7.0)on ice. Tissue was homogenized using a glass tube andTeflon pestle for approximately 2 min until the tissuewas completely ground. The suspension was centrifugedat 10,000 × g in a Fisher Scientific Micro-CentrifugeModel 59A for 20 min. The supernate was collected andstored at −20 ◦C until use. Protein concentration wasassessed using the BCA protein assay. Protein carbonylswere derivatized by incubating 25 �g of ovarian proteinwith 10 mM 2,4-DNPH in 10% trifluoroacetic acid for15 min [21]. Non-derivatized aliquots from each samplewere incubated with 10% trifluoroacetic acid to serveas negative controls. After 15 min, all solutions wereneutralized with 2 M Tris in 30% glycerol [21]. Deriva-tized and non-derivatized ovarian protein samples weresolubilized with 0.9% SDS (w/v), separated on a 12%SDS-polyacrylamide gel, and electroblotted to 0.45 �mpolyvinylidene fluoride (PVDF) membrane with a wettransfer tank (Hoefer, Inc., San Francisco, CA, USA).Membranes were blocked with 1% BSA, 1% normaldonkey serum in PBS-T (PBS containing 0.5% Tween-20, pH 7.4). Subsequently, membranes were incubatedwith rabbit anti-DNP antibody (1:24,000 in PBS-T con-taining 1% normal donkey serum). After rinsing withPBS-T, membranes were incubated with peroxidase-conjugated donkey anti-rabbit IgG (1:30,000 in PBS-Tcontaining 1% normal donkey serum). Immunoreactivebands were visualized using chemiluminescence reagentand X-ray film.

2.7. Cytochrome P4502E1 activity

Microsomes were prepared from rat ovaries previ-ously frozen in 1 ml 0.1 M phosphate buffer (pH 7.0) asdescribed by Forkert et al. [22] except isolation buffercontained 0.1 mM AEBSF protease inhibitor. Activityof p-nitrophenol hydroxylase was used as an assay ofcytochrome P4502E1 activity [22,23]. In vitro inhibitionwas assessed following a 60-min incubation with 5 mMTCE or buffer alone [24]. Ovarian microsomes were pre-pared from ovaries from rats previously treated with TCEfor 2 or 3 days and their respective vehicle controls andanalyzed for p-nitrophenol hydroxylase activity.

2.8. Experimental design and data analysis

Effects on oocyte fertilizability were assessed by IVFwith each interval of exposure evaluated in three sepa-

rate replicates. Each replicate consisted of six or sevenfemale rats and one untreated male rat. Two females weretreated with vehicle water for 2 weeks to serve as neg-ative controls, two or three females were treated with

ological Interactions 170 (2007) 20–30

Fig. 1. (A) Decreased fertilizability of oocytes from TCE-exposed(0.45% TCE (v/v) in 3% Tween) rats compared with oocytes fromvehicle-control (3% Tween) rats. Fertilization was assessed by the pres-ence of decondensed sperm heads. Values represent the least squaresmeans from three replicates. *P < 0.05 compared with the vehicle con-trol. (B) Oocytes with bound sperm. Bars represent the percentage of

24 K.L. Wu, T. Berger / Chemico-Bi

TCE water for 4 or 5 days (days 1–5, 6–10, or 11–14)to examine potential critical intervals of exposure, andtwo females were treated with TCE water for 2 weeks toserve as positive controls. In each replicate, oocytes fromfemales that received the same treatment were pooled.The number of females that ovulated was recorded foreach treatment in each replicate. Immunohistochemicallocalization of protein carbonyls within the ovary wasevaluated in three replicates. Each replicate included onefemale per treatment group, for a total of five females perreplicate; treatments included TCE water on days 1–5,TCE water on days 6–10, TCE water on days 11–14,plus the negative control (vehicle water for 2 weeks)and the positive control (TCE water for 2 weeks). Pro-tein carbonyl formation following TCE exposure wasquantitatively assessed on Western blots. Ovaries wererecovered and weighed prior to homogenization. Eightfemales were used in four replicates that consisted ofovaries from one rat treated with TCE for 5 days andone vehicle-control littermate.

In vitro fertilization data, ovarian weight data, andp-nitrophenyl hydroxylase activity were subjected toanalysis of variance (ANOVA) using SAS (SAS Statis-tical Programs, Cary, NC, USA). Treatment was a fixedfactor, pooled females within a treatment and replicatewere random. The percentage of oocytes fertilized andpercentage of oocytes with bound sperm were weightedby the number of oocytes evaluated. The percentageof females that ovulated was analyzed before and aftertransformation (arc sine square root) to improve normal-ity. P-values associated with the transformed data arepresented.

3. Results

3.1. Effect of female exposure to trichloroethyleneon oocyte fertilizability

In vivo TCE exposure of females reduced the fertil-izability of zona-free oocytes regardless of the intervalof TCE exposure. Oocytes from rats exposed to TCE for2 weeks were consistently less fertilizable than oocytesfrom vehicle-control females (23% versus 66%, respec-tively, P < 0.05, Fig. 1A). Females that were treated forshorter intervals within the 2-week period had fertiliza-tion rates similar to oocytes from females that receivedTCE for 2 weeks; 21.7, 23.3, and 29.3% for oocytes fromfemales exposed to TCE on days 1–5, 6–10, and 11–14,

respectively. These oocytes were significantly less fer-tilizable than oocytes from vehicle-control females, butnot different from oocytes from females exposed to TCEfor 2 weeks. Oocytes from females exposed to TCE on

total oocytes (fertilized and unfertilized) that bound sperm. Oocytesfrom females treated with TCE on days 6–10, days 11–14 and for 2weeks bound fewer sperm compared with vehicle controls. *P < 0.05compared with the vehicle control.

days 6–10, days 11–14, and for 2 weeks were also lessable to bind sperm compared with oocytes from vehiclecontrols (P < 0.05, Fig. 1B).

3.2. TCE exposure, ovulation, and the ovary

TCE water and vehicle water removal from the drink-ing water bottles (an estimate of consumption) wereconsistent with reported average consumption for ratsof this size [8]. Liver weights did not differ significantlyafter 3 days of exposure (P > 0.10); 5.68 g versus 5.18 gfor TCE treatment and vehicle controls, respectively. Thepercentage of females responding to hormonal induc-tion of ovulation was not altered by exposure to TCE(P > 0.10, S.E.M. = 15, Table 1). On the rare occasion,females that did not ovulate showed evidence of previ-ous ovulation (corpora lutea), which may account for the

lack of cumulus masses recovered from the oviducts ofthose females. The mean ovarian weights were similarfor TCE-treated females and vehicle controls (0.062 gcompared with 0.071 g, P > 0.25, S.E.M. = 0.008). Thus,

K.L. Wu, T. Berger / Chemico-Biological Interactions 170 (2007) 20–30 25

Table 1Reproductive parameters of female rats after exposure to trichloroethylene

Vehicle TCE

2 weeksa Days 1–5 Days 6–10 Days 11–14 2 weeksa

Total number of females 18 8 8 8 18P 87.5b b b b

nine tott signifi

T

3t

tfbwfw(

FfsdDb

ercentage of total females ovulating 78

a A positive and negative control were used for each IVF replicate (b Percentage of females ovulating in each TCE-treatment group are no

CE-exposure did not affect ovarian development.

.3. Evidence of TCE metabolism in the ovaries ofreated females

The ovarian tissue from TCE-exposed and vehicle-reated rats did not exhibit any gross histological changesollowing treatment. However, increased protein car-onyls (an indication of oxidatively modified proteins)

ere detected immunohistochemically in the ovaries of

emales exposed to TCE water for 2 weeks comparedith females that received vehicle water for 2 weeks

Fig. 2). Granulosa cells from females that received

ig. 2. Immunohistochemical labeling of protein carbonyls following TCE exrom a female given drinking water containing 0.45% TCE (v/v) in 3% Twubstrate indicating the presence of protein carbonyls. Oocytes are not labeledrinking water for 2 weeks. Granulosa cells and oocytes show limited and no) incubated with normal rabbit serum in place of the primary antibody (rabackground labeling. Pictures are representative of three replicates.

100 100 78

al replicates).cantly different compared with vehicle control (P > 0.10, S.E.M. = 15).

TCE had noticeably more labeling for protein carbonyls(Fig. 2A) compared with granulosa cells in ovaries fromvehicle-control rats (Fig. 2B). The immunolabeling neg-ative controls, sections of ovary from TCE-exposed ratsand vehicle-control rats that were incubated with nor-mal rabbit serum instead of rabbit anti-DNP antibody,were comparable with no detectable protein carbonyls(Fig. 2C and D).

Protein carbonyls were present in all ovarian tis-

sues exposed to TCE for 4–5 days within the 2-weektreatment interval. The granulosa cells were positive forimmunolabeling of protein carbonyls and the oocytesappeared unlabeled. Although immunohistochemistry is

posure in rat ovarian tissue. Bar represents 50 �m. (A) Ovarian tissueeen for 2 weeks. Granulosa cells are prominently labeled with blue. (B) Ovarian tissue from a female that received 3% Tween vehicle inlabeling, respectively. Ovarian tissues from the same females (C and

bit anti-DNP). Neither tissue has positive signal indicating negligible

26 K.L. Wu, T. Berger / Chemico-Biological Interactions 170 (2007) 20–30

Fig. 3. Immunohistochemical detection of protein carbonyls following TCE exposure on days 1–5, 6–10, and 11–14 of a 2-week treatment interval.Bar represents 50 �m. (A) Ovarian tissue from a female treated with 0.45% TCE (v/v) in 3% Tween on days 1–5. Granulosa cells are slightly labeledindicating the presence of protein carbonyls. (B) Ovarian tissue from a female exposed to TCE on days 6–10. Granulosa cells are more intensely

arbonyls. (C) Ovarian tissue from a female exposed to TCE on days 11–14.), indicating increased presence of protein carbonyls. (D–F) Ovarian tissues

ith normal rabbit serum in place of the primary antibody (rabbit anti-DNP).und labeling. Pictures representative of three replicates.

Fig. 4. Western blot analysis of protein carbonyls in the ovaries fromrats receiving TCE in drinking water for 5 days or vehicle water. Lane 1:non-derivatized ovarian protein from TCE-treated rat, lane 2: deriva-tized ovarian protein from TCE-treated rat, lane 3: non-derivatizedovarian protein from vehicle-control rat, and lane 4: derivatized ovar-ian protein from vehicle-control rat. Protein carbonyls are present inderivatized ovarian protein from TCE-treated rats (lane 2). Vehicle-

labeled compared with (A), indicating increased presence of protein cGranulosa cells are more intensely labeled compared with (A and Bfrom the same females represented in (A–C), respectively, incubated wNone of the tissues have positive signal indicating negligible backgro

not quantitative, comparatively small amounts of oxi-dation product were visible in the granulosa cells onday 15 following exposure of females to TCE on days1–5 of the 2-week treatment interval (Fig. 3A). Moreintense labeling of protein carbonyls was visible on day15 following TCE exposure on days 6–10 (Fig. 3B), andthe most oxidation product was present following treat-ment with TCE on days 11–14 (Fig. 3C). Hence, lessintense labeling of protein carbonyls were observed intissues given a recovery period between treatment andhistological examination. Ovarian tissue from femalesexposed to TCE water on days 1–5 and 6–10 were lessintensely labeled for protein carbonyls compared withovarian tissue from females that received TCE water for2 weeks (Figs. 2A and 3A and B). Ovarian tissue fromfemales exposed to TCE water on days 11–14 had sim-ilar labeling of protein carbonyls to ovarian tissue fromfemales that received TCE water for 2 weeks (Figs. 2Aand 3C). Sections of ovary from the TCE-treated andvehicle-control rats that were incubated with normal rab-bit serum, instead of rabbit anti-DNP antibody, served asnegative controls (Fig. 3D–F).

3.4. Quantitative assessment of oxidatively modified

ovarian protein from treated females

The increased presence of oxidized protein in ovar-ian tissue following TCE exposure was confirmed using

control rats had much fainter labeling indicating less protein carbonyls(lane 4). The antibody to the derivatization agent for protein carbonylsexhibited no background binding, as it did not label the non-derivatizedovarian protein (lanes 1 and 3, respectively). Blot representative of fourreplicates.

ologica

WbIdlrtTtin

3

hmif5pS32cS

4

motavotOittmufafvg

wgi

K.L. Wu, T. Berger / Chemico-Bi

estern blotting with the same rabbit anti-DNP anti-ody that was used for immunohistochemistry (Fig. 4).ncreased protein carbonyl formation was visible in theerivatized samples from TCE-treated females (Fig. 4,ane 2) compared with derivatized samples from femaleats treated with vehicle (Fig. 4, lane 4). No immunoreac-ive oxidized proteins were detected in non-derivatizedCE and vehicle samples (Fig. 4, lanes 1 and 3, respec-

ively). Duplicate blots stained with Coomassie bluendicated equal loading of protein in each lane (resultsot shown).

.5. Cytochrome P4502E1 activity

In vitro incubation with TCE reduced p-nitrophenylydroxylase activity (cytochrome P4502E1) in threeicrosomal preparations to 54% of control activ-

ty (P < 0.05). The p-nitrophenol hydroxylase activityollowing 2 days of in vivo exposure averaged30 pmoles/ng protein compared with 299 pmoles/ngrotein in ovaries from vehicle controls (P < 0.20,.E.M. = 136). Values were very similar following

days of in vivo exposure (225 pmoles/ng versus34 pmoles/ng protein for microsomes from vehicle-ontrol and treated animals, respectively, P > 0.25,.E.M. = 60).

. Discussion

The goals of this study were two-fold: (1) to deter-ine if a short interval of in vivo TCE-exposure during

ocyte growth decreased oocyte fertilizability, and (2)o determine if protein oxidation was a potential mech-nism contributing to reduced oocyte fertilizability. Initro fertilization results demonstrated a short intervalf 4 or 5 days of exposure to TCE is sufficient to reducehe fertilizability of oocytes from TCE-treated females.ur laboratory has previously suggested that IVF is an

mportant bioassay which can be used to evaluate in vivooxicant effects on female gamete quality [18]. Otherechniques of assessing oocyte fertility, such as in vivo

ating trials, are insensitive because the normal rat ejac-late has approximately 10-fold more sperm than neededor maximum fertilization [25,26]. Studies by Smith etl. have suggested a more fertile male may compensateor a less fertile female [27,28]. The same holds true initro since increasing concentrations of sperm lead toreater percentages of fertilized oocytes [18].

In these studies, females were dosed via drinkingater since earlier studies noted adverse effects of oralavage [18]. The Tween vehicle is effective in increas-ng solubility of TCE in the drinking water, but does

l Interactions 170 (2007) 20–30 27

not affect reproductive parameters or weight in thisstrain of rats [9,18]. Negligible differences in liverweights between TCE-treated rats and vehicle-controlrats suggest systemic toxicity is not a confounding factoraffecting oocyte fertilizability.

The TCE-treatment interval was shortened from 2weeks to 4 or 5 days in the present study to investi-gate which stage of oocyte growth and maturation (early,middle, or late) was perturbed by exposure to TCE.These intervals would be expected to vary in relativeimportance of RNA synthesis, protein synthesis, andredistribution of cellular organelles. Originally, a 2-weekTCE treatment was chosen because it covers the intervalof massive oocyte growth in the rodent [29]. The exten-sive cytoplasmic growth and nuclear changes that occurduring oocyte maturation [29–31] suggested it might bea period of time when oocytes are particularly sensi-tive to some toxicants. The age of treatment was chosenbecause a majority of females have not yet cycled, butthe females are readily responsive to the induced ovula-tion regimen of PMSG and hCG [9]. As expected, thevast majority of females (>90%) had not cycled prior toinduced ovulation at 42–45 days.

Granulosa cells including the cumulus cells (thegranulosa cells immediately surrounding the oocyte)exhibited a greater incidence of protein carbonyls inthe ovaries of TCE-treated rats compared with the samecells from vehicle-control rats. Current results along withcytochrome P4502E1 localization in the granulosa cells[13] are consistent with ovarian TCE metabolism bythe cytochrome P450 dependent oxidative pathway [11].Protein carbonyl content is the most commonly usedindicator of oxidative modification of proteins [32–36].Metabolism of some environmental pollutants, such asTCE, may result in oxidative stress which could lead tothe formation of reactive oxygen species (ROS) withinthe body. Covalent modification of a protein inducedeither directly by ROS or indirectly by reaction withsecondary by-products of oxidative stress can result inprotein oxidation [35]. Reactive oxygen species, suchas TCE-epoxide, can oxidize amino acid residues onproteins, which leads to the formation of protein car-bonyls [32]. Such oxidation is not specific to a singleprotein. Protein carbonyls are stable products which canbe assessed in the laboratory and assays for the detectionof protein carbonyls involve derivatization of the car-bonyl group with 2,4-DNPH. Derivatization results in theformation of a hydrazone product which can be labeled

for visualization by immunoassays, spectrophotometricassays or by radioactive labeling [32,33,37,38].

The granulosa cells of ovaries 10 days after TCE-exposure had subtle labeling for protein carbonyls, which

ologica

28 K.L. Wu, T. Berger / Chemico-Bi

is comparable with labeling of the granulosa cells ofovaries from vehicle-control rats. The amount of pro-tein carbonyls in these tissues may represent a baselineamount of oxidative damage. Normal metabolism canproduce reactive oxygen and reactive nitrogen species,which can cause oxidative damage [32]. The less intensestaining for protein carbonyls in ovaries from ratsexposed to TCE on days 1–5 or 6–10 compared withdays 11–14 may be due to tissue repair as these femalesreceived vehicle for several days prior to collection ofovaries on the morning of day 15. The time betweenTCE-exposure and tissue collection could have allowedthe oxidized proteins to be removed from the tissueby proteolysis [35]. Alternatively, the ovary in the finalstages of hormone-induced maturation may be the mostsensitive to oxidative damage. As a parent compound,TCE does not appear to have adverse effects on oocytefertilizability following in vitro exposure [10]. In vitrofertilization data suggests the TCE metabolites formedvia the oxidative pathway elicit toxic effects on oocytes[10]. These metabolites could bind to proteins formingprotein adducts, thus altering the original protein andfunction.

Oxidative modification of proteins can inhibit nor-mal protein function [35] by loss of catalytic or structuralintegrity thereby interrupting normal pathways [39]. Theinhibition of p-nitrophenyl hydroxylase activity follow-ing in vitro exposure is similar to that reported by Lillyet al. Reductions in p-nitrophenol hydroxylase activitywere reported 4 h following in vivo exposure in lungmicrosomes [22] and increased p-nitrophenyl hydrox-ylase activity was observed in hepatic microsomes 3 hfollowing in vivo exposure [40]. Elevated activity con-tinued to be observed for at least 24 h at the highest dosein this later study but a lower dose that exhibited elevatedactivity at 3, 6, and 12 h no longer exhibited elevatedactivity at 16 or 24 h. Hence the arithmetically increasedactivities we observed during continuous treatment for 2days and the similar activities observed after 3 days onovarian microsomal preparations are not surprising.

Given that oxidative modification of protein altersfunction, granulosa cells may be less able to aid inthe growth and maturation of oocytes after exposureto TCE. This may partially explain why oocytes fromTCE-exposed females are less fertilizable. Earlier stud-ies reported females exposed to TCE had decreasedreproductive performance, and increased incidences ofpoor pregnancy outcomes [41,42]; however, it is unclear

in these studies whether the TCE metabolites thatcaused the adverse effects were formed in the reproduc-tive tract. A study conducted by Manson et al. notedthat the cytochrome P450 oxidized TCE metabolites,

l Interactions 170 (2007) 20–30

trichloroacetic acid and trichloroethanol, were present inthe ovaries and uteri of rats, but they did not determinewhere these metabolites were formed [42].

Female gametogenesis is difficult to examine giventhe numerous factors that can perturb its success. Sev-eral toxicants affect male gametogenesis independentlyof the endocrine system, but female gametogenesis andoocyte quality are much more difficult to study partiallybecause mammalian female gametes are maintainedinternally rather than being expelled. The current studyadds to previous work by demonstrating that short in vivoexposure to TCE significantly reduces fertility, whereasthe females were exposed to TCE for the duration ofoocyte growth and maturation in the previous study.Interestingly, oocytes were sensitive to TCE exposureduring each third of postnatal oocyte growth and matu-ration. Although, oocytes from females exposed to TCEon days 6–10 and 11–14 were slightly less able to bindsperm compared with vehicle controls, oocytes fromfemales exposed to TCE on days 1–5 were as effec-tive at binding sperm as vehicle-control oocytes. Theinability of oocytes from females exposed to TCE ondays 6–10 and 11–14 to bind sperm suggests oocyteplasma membrane proteins have been altered. Thisresult supports earlier work suggesting fertilizability ofoocytes from females exposed to TCE was hamperedby changes to the oocyte plasma membrane (decreasedability to bind sperm plasma membrane proteins) [9].Combined with the consistent reduction in fertilizabil-ity in oocytes from all exposed females, this suggeststhat the ability of oocytes to fuse with sperm is hindered.Trichloroethylene metabolism in the ovary may decreasethe functionality of oocyte plasma membrane proteinsindirectly by adversely affecting cumulus-oocyte inter-actions which take place during oocyte growth andmaturation. The disruption of cumulus-oocyte interac-tions could alter the ability of cumulus cells to supportoocyte cytoplasmic maturation, which includes synthe-sis of plasma membrane receptors for sperm [43]. Thus,the effect of TCE and its metabolites on the oocyte couldbe indirect.

The presence of increased ovarian protein carbonylsfollowing TCE-exposure as demonstrated by immuno-histochemistry and Western blotting suggests the ovarymay participate in the metabolism of TCE. Increasedprotein carbonyls, previously reported presence ofcytochrome P4502E1 [13], presence of TCE metabolites[42] and the p-nitrophenol hydroxylase activity are con-

sistent with metabolism of some TCE by the cytochromeP450 pathway in the ovary. Evidence from this studysuggests the rat ovary, in addition to the liver, is capableof metabolizing TCE via the cytochrome P450 depen-

ologica

doTpTiennTmmfiaat

A

Ad2tCiHss

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

K.L. Wu, T. Berger / Chemico-Bi

ent oxidative pathway. The results demonstrate thatocytes are less fertilizable as a result of short-termCE-exposure, and there is an increased incidence ofrotein carbonyls in granulosa cells of the ovaries fromCE-treated rats. Given these results, oocyte fertilizabil-

ty in vivo may be temporarily hindered after a female isxposed to high levels of TCE. Although litter size mightot be reduced in rats following this in vivo exposure andatural mating, female gamete quality is compromised.his is an important finding when the rat is used as aodel for other species where excess sperm numbersay not be as available to compensate for less fertile

emales [25,26]. Bioactivation of the environmental tox-cant, TCE, within the rat ovary by metabolism may playn important role in reducing rat oocyte fertilizabilityfter in vivo TCE-exposure and be a model for effects ofoxicants on female gametes.

cknowledgements

Portions of this paper were presented at the 38thnnual Meeting of the Society for the Study of Repro-uction, Quebec City, Quebec, Canada, July 24–27,005. Funding for KLW was provided in part byhe National Institutes of Health, National Institute ofhild Health and Human Development (NICHD) train-

ng grant in Fertilization and Early Development (T32D071131), and the University of California Toxic Sub-

tances Research and Teaching Program (UC TSR&TP)tudent fellowship.

eferences

[1] EPA, Sources, Emission and Exposure for Trichloroethylene(TCE) and Related Chemicals, National Center for EnvironmentalAssessment—Washington Office, Office of Research and Devel-opment, U.S. Environmental Protection Agency, 2001, pp. 1–125.

[2] P.H. Howard, R.S. Boethling, W.F. Jarvis, M.W.M.E.M.Michalenko, Handbook of Environmental Degradation Rates,Lewis Publishers, Chelsea, MI, 1991.

[3] ATSDR, ToxFAQs for Trichloroethylene, U.S. Department ofHealth and Human Services, Public Health Service, Agency forToxic Substances and Disease Registry, 2003.

[4] C. Wu, J. Schaum, Exposure assessment of trichloroethylene,Environ. Health Perspect. 108 (2000) 359–363.

[5] S.B. DuTeaux, T. Berger, R.A. Hess, B.L. Sartini, M.G. Miller,Male reproductive toxicity of trichloroethylene: sperm proteinoxidation and decreased fertilizing ability, Biol. Reprod. 70(2004) 1518–1526.

[6] H. Xu, N. Tanphaichitr, P.G. Forkert, A. Anupriwan, W. Weer-

achatyanukul, R. Vincent, A. Leader, M.G. Wade, Exposure totrichloroethylene and its metabolites causes impairment of spermfertilizing ability in mice, Toxicol. Sci. 82 (2004) 590–597.

[7] J.C. Lamb, K.L. Hentz, Toxicological review of male reproductiveeffects and trichloroethylene exposure: assessing the relevance

[

l Interactions 170 (2007) 20–30 29

to human male reproductive health, Reprod. Toxicol. 22 (2006)557–563.

[8] L.F.M. van Zutphen, V. Baumans, A.C. Beynen, Principles ofLaboratory Animal Science: A Contribution to the Humane Useand Care of Animals and to the Quality of Experimental Results,Elsevier, New York, 2001.

[9] T. Berger, C.M. Horner, In vivo exposure of female rats to tox-icants may affect oocyte quality, Reprod. Toxicol. 17 (2003)273–281.

10] N.C. Cosby, W.R. Dukelow, Toxicology of maternally ingestedtrichloroethylene (TCE) on embryonal and fetal development inmice and of TCE metabolites on in vitro fertilization, Fundam.Appl. Toxicol. 19 (1992) 268–274.

11] L.H. Lash, J.W. Fisher, J.C. Lipscomb, J.C. Parker, Metabolism oftrichloroethylene, Environ. Health Perspect. 108 (2000) 177–200.

12] I.W. Davidson, R.P. Beliles, Consideration of the target organtoxicity of trichloroethylene in terms of metabolite toxicity andpharmacokinetics, Drug Metab. Rev. 23 (1991) 493–599.

13] E. Cannady, C.A. Dyer, P.J. Christian, I.G. Sipes, P.B. Hoyer,Expression and activity of cytochromes P450 2E1, 2A, and 2B inthe mouse ovary: the effect of 4-vinylcyclohexene and its diepox-ide metabolite, Toxicol. Sci. 73 (2003) 423–430.

14] D. Singh, R.S. Pandey, Glutathione-S-transferase in rat ovary:its changes during estrous cycle and increase in its activityby estradiol-17 beta, Indian J. Exp. Biol. 34 (1996) 1158–1160.

15] B.S. Cummings, J.C. Parker, L.H. Lash, Role of cytochrome P450and glutathione S-transferase alpha in the metabolism and cyto-toxicity of trichloroethylene in rat kidney, Biochem. Pharmacol.59 (2000) 531–543.

16] National Research Council (NRC), Drinking Water and Health:Selected Issues in Risk Assessment, Washington, DC, NationalAcademies Press, Washington, DC, 1989.

17] International Agency for Research on Cancer (IARC), Mono-graphs on the Evaluation of Carcinogenic Risks to Humans: DryCleaning, Some Chlorinated Solvents and other Industrial Chem-icals, World Health Organization, 1995.

18] T. Berger, M.G. Miller, C.M. Horner, In vitro fertilization after invivo treatment of rats with three reproductive toxicants, Reprod.Toxicol. 14 (2000) 45–53.

19] G. Evans, D.T. Armstrong, Reduction in fertilization rate in vitroof oocytes from immature rats induced to superovulate, J. Reprod.Fertil. 70 (1984) 131–135.

20] G.L. Nicolson, R. Yanagimachi, H. Yanagimachi, Ultrastructurallocalization of lectin-binding sites on the zonae pellucidae andplasma membranes of mammalian eggs, J. Cell Biol. 66 (1975)263–274.

21] E. Schacter, J.A. Williams, M. Lim, R.L. Levine, Differentialsusceptibility of plasma proteins to oxidative modification: exam-ination by Western blot immunoassay, Free Radic. Biol. Med. 17(1994) 429–437.

22] P.G. Forkert, B. Millen, L.H. Lash, D.A. Putt, B.I. Ghanayem, Pul-monary bronchiolar cytotoxicity and formation of dichloroacetyllysine protein adducts in mice treated with trichloroethylene, J.Pharmacol. Exp. Ther. 316 (2006) 520–529.

23] D.R. Koop, Inhibition of ethanol-inducible cytochrome P450IIE1by 3-amino-1,2,4-triazole, Chem. Res. Toxicol. 3 (1990)

377–383.

24] P.D. Lilly, J.R. Thornton-Manning, M.L. Gargas, H.J. Clewell,M.E. Andersen, Kinetic characterization of CYP2E1 inhibitionin vivo and in vitro by the chloroethylenes, Arch. Toxicol. 72(1998) 609–621.

ologica

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

30 K.L. Wu, T. Berger / Chemico-Bi

25] J.H. Aafjes, J.M. Vels, E. Schenck, Fertility of rats with artificialoligozoospermia, J. Reprod. Fertil. 58 (1980) 345–351.

26] P.K. Working, Male reproductive toxicology: comparison of thehuman to animal models, Environ. Health Perspect. 77 (1988)37–44.

27] K. Smith, L.J. Rodriguez-Rigau, E. Steinberger, Relation betweenindices of semen analysis and pregnancy rate in infertile couples,Fertil. Steril. 28 (1977) 1314–1319.

28] E. Steinberger, K.D. Smith, R.K. Tcholakian, L.J. Rodriguez-Rigau, Testosterone levels in female partners of infertile couples.Relationship between androgen levels in the woman, the malefactor, and the incidence of pregnancy, Am. J. Obst. Gynecol.133 (1979) 133–138.

29] R. Bachvarova, Gene expression during oogenesis and oocytedevelopment in mammals, Dev. Biol. (N Y 1985) 1 (1985)453–524.

30] R. Bachvarova, B.V. Paynton, Gene expression during growth andmeiotic maturation of mouse oocytes, Prog. Clin. Biol. Res. 267(1988) 67–85.

31] E.A. McGee, A.J. Hsueh, Initial and cyclic recruitment of ovarianfollicles, Endocr. Rev. 21 (2000) 200–214.

32] M. Chevion, E. Berenshtein, E.R. Stadtman, Human studiesrelated to protein oxidation: protein carbonyl content as a marker

of damage, Free Radic. Res. 33 (2000) S99–S108.

33] R.L. Levine, Carbonyl modified proteins in cellular regulation,aging, and disease, Free Radic. Biol. Med. 32 (2002) 790–796.

34] B.S. Berlett, E.R. Stadtman, Protein oxidation in aging, disease,and oxidative stress, J. Biol. Chem. 272 (1997) 20313–20316.

[

l Interactions 170 (2007) 20–30

35] E. Shacter, Quantification and significance of protein oxidationin biological samples, Drug Metab. Rev. 32 (2000) 307–326.

36] M.F. Beal, Oxidatively modified proteins in aging and disease,Free Radic. Biol. Med. 32 (2002) 797–803.

37] R.L. Levine, D. Garland, C.N. Oliver, I.A. Amic, I. Climent,A.G. Lenz, B.W. Ahn, S. Shaltiel, E.R. Stadtman, Determinationof carbonyl content in oxidatively modified proteins, MethodsEnzymol. 186 (1990) 464–478.

38] I. Dalle-Donne, R. Rossi, D. Giustarini, A. Milzani, R. Colombo,Protein carbonyl groups as biomarkers of oxidative stress, Clin.Chim. Acta 329 (2003) 23–38.

39] E.R. Stadtman, R.L. Levine, Protein oxidation, Ann. N.Y. Acad.Sci. 899 (2000) 191–208.

40] K.M. Lee, S. Muralidhara, C.A. White, J.V. Bruckner, Mecha-nisms of the dose-dependent kinetics of trichloroethylene: oralbolus dosing of rats, Toxicol. Appl. Pharmacol. 164 (2000) 55–64.

41] S. Coberly, D.J. Oudiz, J.W. Overstreet, L.M. Wiley, Effects ofmaternal exposure to trichloroethylene (TCE) on cell proliferationin the mouse preimplantation embryo, Reprod. Toxicol. 6 (1992)241–245.

42] J.M. Manson, M. Murphy, N. Richdale, M.K. Smith, Effects oforal exposure to trichloroethylene on female reproductive func-

tion, Toxicology 32 (1984) 229–242.

43] M. Zuccotti, R. Yanagimachi, H. Yanagimachi, The ability ofhamster oolemma to fuse with spermatozoa: its acquisition duringoogenesis and loss after fertilization, Development (Cambridge,England) 112 (1991) 143–152.