Toxicology in Vitro 21 (2007) 558–565www.elsevier.com/locate/toxinvit

Toxic eVects of zearalenone and �-zearalenol on the regulationof steroidogenesis and testosterone production in mouse Leydig cells

Jianying Yang a, Yongfa Zhang b, Yongqiang Wang c, Sheng Cui a,¤

a Department of Animal Physiology, College of Biological Sciences, China Agricultural University, Beijing 100094, PR Chinab College of Life Sciences, Northwest Science and Technology University of Agriculture and Forestry, Yangling Shaanxi 712100, PR China

c CIHR Group in Matrix Dynamics, Faculty of Dentistry, University of Toronto, Toronto, Canada

Received 28 August 2006; accepted 26 October 2006Available online 6 November 2006

Abstract

Zearalenone (ZEA) and its derivative �-zearalenol (�-ZOL) are produced by fungi of the genus Fusarium and, after ingestion via con-taminated cereals, may lead to animal fertility disturbances and other reproductive pathologies. The previous study demonstrated thetoxic eVects of ZEA and �-ZOL through disturbances in male fertility and other reproductive pathologies in mice. In this study, we fur-ther examined the direct biological eVects of ZEA and �-ZOL on steroidogenesis production, primarily in Leydig cells of mice. Maturemouse Leydig cells were puriWed by Percoll gradient centrifugation and the cell purity was determined by 3�-hydroxysteroid dehydroge-nase (3�-HSD) staining. To examine ZEA and �-ZOL-induced biological consequences, we measured testosterone secretion andtranscription level of 3 key steroidogenic enzymes including 3�-HSD-1, P450scc and StAR, in ZEA and �-ZOL/human chorionicgonado-tropin (hCG) co-treated cells. Our results showed that ZEA and �-ZOL (10¡4 M, 10¡6 M and 10¡8 M) signiWcantly suppressed hCG(10 ng/ml)-induced testosterone secretion. The suppressive eVect is correlated with a decrease in the level of transcription of 3�-HSD-1,P450scc, and StAR (P < 0.05).© 2006 Elsevier Ltd. All rights reserved.

Keywords: ZEA; �-ZOL; 3�-HSD-1; p450scc; StAR; Mouse Leydig cells

1. Introduction

Zearalenone (ZEA) and its derivative, �-zearalenol(�-ZOL), are a family of phenolic compounds produced byseveral species of Fusarium (F. graminearum, F. culmorum,F. crookwellense, F. sambucinum and F. equiseti), which caninfect many important crops such as corn, wheat, sorghum,barley, oats, sesame seed, hay and corn silage (Manka et al.,1985; D’Mello, 1997). Several studies carried out in Europeand a number of transcontinental countries have reported ahigh incidence of ZEA in cereals and in animal feed (Botta-lico et al., 1989; Muller et al., 1998; Scudamore et al., 1998;Scudamore and Patel, 2000).

* Corresponding author. Tel./fax: +86 10 62733443.E-mail address: sheng_cui@163.com (S. Cui).

0887-2333/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.tiv.2006.10.013

ZEA binds to estrogen receptors (ERs) in vitro, withsimilar aYnity for both forms: ERa and ER� (Kuiperet al., 1998). Despite having a lower aYnity for estrogenreceptors than 17�-estradiol (100–1000 times less), ZEAand �-ZOL act through ERs (Kiang et al., 1978; Nikovet al., 2000) to activate transcription of estrogen-responsivegenes in vivo (Gray et al., 1985; Katzenellenbogen et al.,1979; Mehmood et al., 2000) and in vitro (Kuiper et al.,1998; Mayr, 1988), as well as promoting undesirable estro-genic eVects.

These fungal toxins have been associated with hypere-strogenism and other reproductive disorders in swine. Insows, a series of reproductive disorders may occur atgreater levels of ZEA in feed (50–100 mg/g feed), includingthe induction of vulvovaginitis, vaginal and rectalprolapses, delayed onset of the Wrst estrus, infertility char-acterized by continuous estrus, pseudo-pregnancy, ovarian

J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565 559

abnormalities and pregnancy loss (Osweiler, 1986). The fer-tility-inhibiting eVect associated with long-term consump-tion of ZEA-infected maize, has been studied in adult malealbino rats. The fertility rate was further reduced by 25–30% if the animals were kept on a contaminated diet up to14 weeks (Ruzsas et al., 1979). There was a tendency forboars, fed 9 ppm dietary ZEA, to produce lower total andgel free volumes of semen, with lower total motile sperm(Young and King, 1986). ZOL may reduce aggressivebehavior, testes growth, and sexual activity in farmed fal-low bucks by ear implants at a dose of 36 mg at 90 d inter-vals (Wilson et al., 2002). However, these reproductivetoxicities of ZEA and �-ZOL for testicular function havemostly relied on the in vivo approach of using animal mod-els. Complications in pharmacokinetic distribution andsecondary eVects attributed to other unidentiWed factorsmay make it diYcult to decipher the direct mechanistictoxicities of ZEA and �-ZOL to the cells. Therefore, it isnecessary to adopt cell models to determine direct biologi-cal eVects of ZEA and �-ZOL to validate the in vivo Wnd-ings.

It is well known that Leydig cells play a crucial role insynthesizing testosterone and regulating the process ofspermatogenesis. Alteration of Leydig cell function canlead to adverse eVects on testicular functions. In this study,the aim was to elucidate the eVects of ZEA and �-ZOL onthe process of steroidogenesis and testosterone secretion inLeydig cells of mice. To determine the mechanistic activi-ties, ZEA or �-ZOL-stimulated mRNA of the key steroido-genic enzymes and the level of testosterone secretion weremeasured.

2. Materials and methods

All experimental animal use and experimental design forthis study was approved by the Chinese Association forLaboratory Animal Sciences.

2.1. Cell culture

Leydig cells were isolated from testes of 60- to 90-dayold Kunming mice. The cells were cultured for 2 daysaccording to Biegel et al. (1995). The testes were decapsu-lated, digested in Erlenmyer Xasks within an oscillatingincubator (100 r.p.m and 34 °C) for 15 min by M199medium containing 0.05% collagenase and 1% BSA. Thesuspension cells were transferred to a 50 ml tube and kepton ice for 2 min to allow the tubules to settle. The superna-tant containing Leydig cells was Wltered through a 70 �mnylon cell strainer (BD Biosciences). The cells were centri-fuged at 350g for 20 min at 4 °C. The pellet was re-sus-pended in 10 ml M199 and loaded onto the top of Percollgradient (5%, 30%, 58% and 70%) (Sigma) and centrifugedat 800g for 30 min at 4 °C. The cells in the third layer werecollected, washed with M199 twice, and were re-suspendedin phenol red-free DMEM/F12 (1:1) containing 10%charcoal stripped fetal calf serum (GeminiBio-Products,

Woodland, CA, USA), 50 U/ml penicillin and 50 �g/mlstreptomycin (GIBCO/BRL, Carlsbad, CA, USA). Thecells were plated at a density of 105 cells/cm2 in 24-wellplates (Nunc, Nalge Nunc International, Rochester, NY,USA) at 0.5 ml/well and maintained at 37 ° C with 5% CO2.

2.2. Histochemical staining of 3�-HSD and testosterone induction assay

Following 2 days of incubation, the purity of Leydigcells was examined by histochemical staining for 3�-hydroxy-steroid dehydrogenase, according to the histochemicalmethod of Mendelson et al with some modiWcations (Men-delson et al., 1975). In brief, Leydig cells were incubated in a24-well plate with 0.4 ml/well staining solution containing0.05 M PBS, pH 7.4 supplemented with 0.2 mg/ml nitro-bluetetrazolium (Sigma Chemical Co.), 1 mg/ml NAD and0.12 mg/ml dehydroepiandrosterone (Sigma Chemical Co)for 90 min at 34 °C. The positive cells were stained a darkblue color and the purity of the Leydig cells was observedto be over 90%. Secondly, in the testosterone inductionassay, Leydig cells were exposed to 10 ng/ml human chori-onic gonadotrop (hCG) (Sigma Chemical Co) for 24 h.

2.3. Cell treatment

Two-day cultured Leydig cells grown in phenol red-freeDMEM/F12 medium supplemented with 10% charcoalstripped fetal calf serum and antibiotics (50 U/ml penicillinand 50�g/ml streptomycin) were washed three times in0.05 M PBS pH 7.4. The conditioned media (phenol red-freeDMEM/F12 medium supplemented with 50 U/ml penicillinand 50�g/ml streptomycin and the corresponding leveled-drugs and 10 ng/ml hCG) were added to the 24-well platesat 0.3 ml/well. The Leydig cells were exposed for 24 h toone of the following four treatments: (i) 10 ng/ml hCG(Sigma) + dimethylsulfoxide (DMSO) (Sigma) solvent con-trol; (ii) 10 ng/ml hCG + 10¡4 M ZEA (Sigma, St. Louis,Mo.); (iii) 10 ng/ml hCG + 10¡6 M ZEA; (iv) 10 ng/mlhCG + 10¡8 M ZEA. Administration of �-ZOL was treatedin the same manner. Cell viability was determined by thetrypan blue dye-exclusion test, according to Kellokumpu(1987). The viability of the control and treated cells wasover 90%. At the end of the incubation, the conditionedmedia were collected and stored at ¡20 °C until determina-tion of testosterone concentration. The treated cells wereused for the measurement of mRNA for steroidogenicacute regulatory protein (StAR), cytochrome P450 side-chain cleavage enzyme (P450scc) and 3�-HSD-1.

2.4. Total RNA extraction and semi-quantitative reverse transcription-polymerase chain reaction (PCR)

Total RNA was extracted from diVerent treatmentgroups using Trizol reagent (Invitrogen, Carlsbad, CA,USA) followed by deoxyribonuclease I (Life Technolo-gies, Inc.) treatment to remove DNA contamination. One

560 J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565

microgram of total RNA isolated from Leydig cells or theovary was reverse transcribed using 200 U of Superscript IIRnase H-Reverse Transcriptase (Gibco BRL, Bethesda,MD) in a 50 ml reaction volume in the presence of 25 g/mlOligo (dT), Wrst strand buVer (50 mM Tris–HCl, 75 mMKCl, 3 mM MgCl2), 0.01 M dithiothreitol, and 10 mM ofeach dATP, dGTP, dCTP,and dTTP. The RNA and OligodT mix were heated at 65 °C for 10 min and then cooled to4 °C. The other reagents were added and the reverse-tran-scription (RT) was performed at 42 °C for 1 h.

The PCR of 3�-HSD-1, P450scc, StAR and HPRT(housekeeping gene) were carried out utilizing primer-pairs as described by Akingbemi et al. (2003) and Jin et al.(2000) (Table 1). 3�-HSD-1, P450scc and StAR cDNAswere ampliWed by PCR for 35 cycles (94 °C for 40 s, 55 °Cfor 30 s, and 72 °C for 1 min); HPRT cDNAs were ampli-Wed by PCR for 28 cycles (94 °C for 30 s, 50 °C for 30 s,and 72 °C for 40 s) .The sizes of the PCR products (3�-HSD-1, P450scc, StAR and HPRT) were determined bycomparison with a gene marker (100 bp DNA, Promega)run in a parallel fashion with RT-PCR products in 1.2%agarose gels containing ethidium bromide. The relativeband intensity was quantiWed using a computer-assistedimage analysis system (Visage 2000, BioImage, AnnArbor, MI). The integrated optical density (IOD) valuesfor 3�-HSD-1, P450scc, StAR and HPRT in each bandwere normalized with the corresponding HPRT expres-sion.

2.5. Determination of testosterone production

The conditioned media was assayed for testosteronecontent by radioimmunoassay (RIA) using a testosteroneRIA Kit (HY-020, Beijing SINO-UK Institute Of Biologi-cal Technology, Beijing) in a single batch as described pre-viously by O’Shaughnessy PJ in 1990. Testosterone levelswere determined in duplicate tubes with a within-assaycoeYcient of 6.5% variance.

2.6. Statistical analysis

Drug treatments were performed in triplicate in thesame experiments and individual experiments wererepeated at least three times. All data are represented asmeans§ SE. Statistical signiWcance was tested by Stu-dent’s t-test. Groups were considered signiWcantly diVerentif P < 0.05.

3. Results

3.1. Highly puriWed Leydig cells can be prepared by Percoll gradient centrifugation

The four-step Percoll gradient centrifugation yielded cellsof three major populations. The top layer (layer-1, �»1.035g/ml) consisted of small round cells and cell debris. Layer-2(�»1.076g/ml) was composed of spermatozoa and a smallfraction of Leydig cells. The cells in the third layer (�»1.085g/ml) consisted of over 90% Leydig cells, as they were posi-tively stained by the 3�-HSD method (Fig. 1). According tothe 3�-HSD staining, the layer-3 cells were conWrmed asLeydig cells and were used for the latter part of this study.

3.2. Inhibitory eVects of ZEA and �-ZOL on mRNA levelsof steroidogenic key enzymes

RT-PCR was used to investigate whether 3�-HSD-1,p450scc and StAR mRNA expression was altered in themouse Leydig cells exposed to hCG + ZEA. The results areshown in Fig. 2A–C. PCR products around the predictedsizes of 565 bp, 370 bp and 310 bp had 100% homology inmouse Leydig cells with those of mouse ovary 3�-HSD-1,p450scc and StAR cDNA, respectively. In addition, com-paring 3�-HSD-1, p450scc and StAR mRNA expressions

Fig. 1. Histochemical staining of mouse Leydig cells. PuriWed adult Leydigcells were cultured for 2 days. After the medium was changed, Leydig cellswere stained with 3�-HSD. The positive cells were stained a dark bluecolor. (For interpretation of the references to colour in this Wgure legend,the reader is referred to the web version of this article.)

Table 1Primer sequences

Sense (5�-3�) Antisense (5�-3�) Product size Reference

3�-HSD-1 ACTGCAGGAGGTCAGAGCT GCCAGTAACACACAGAATACC 565 bp Akingbemi et al. (2003)P450scc AGGTGTAGCTCAGGACTTCA AGGAGGCTATAAAGGACACC 370 bp Akingbemi et al. (2003)StAR TGTCAAGGAGATCAAGGTCCTG CGATAGGACCTGGTTGATGAT 310 bp Akingbemi et al. (2003)HPRT CTTGCTCGAGATGTCATGAAG GTTTGCATTGTTTTACCAGTG 290 bp Jin et al. (2000)

J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565 561

with diVerent ZEA-treated groups by semi-quantitativePCR was examined. The 3�-HSD-1 mRNA level relative tothat of HPRT decreased signiWcantly at a dose of 10 ng/mlhCG + 10¡4 M ZEA (lane 6, Fig. 2A and D, P < 0.05). Therewere signiWcant decreases in the p450scc mRNA level rela-tive to that of HPRT in Leydig cells exposed to ZEA at

doses of 10 ng/ml hCG + 10¡6 M and 10 ng/ml hCG +10¡4 M (lane 5–6, Fig. 2B and E, P < 0.05). SigniWcantdecreases in the StAR mRNA level relative to that ofHPRT were observed at doses of 10 ng/ml hCG + 10¡8 M,10 ng/ml hCG + 10¡6 M, and 10 ng/ml hCG + 10¡4 M ZEA(lane 4–6, Fig. 2C and F, P < 0.05).

Fig. 2. RT-PCR analysis of 3�-HSD-1 (A), p450scc (B), and StAR (C) expression in mouse ovary (lane 2, positive controls) and mouse Leydig cells (lane 1,negative control; lanes 3–6) after 24 h incubation with hCG (10 ng/ml) and hCG (10 ng/ml) + ZEA (10¡4 M, 10¡6 M, or 10¡8 M). (A)–(C) show 565 bp,370 bp and 310 bp DNA fragments for 3�-HSD-1; P450scc, and StAR respectively. Lanes 3–6: 10 ng/ml hCG, 10 ng/ml hCG + 10¡8 M ZEA, 10 ng/mlhCG + 10¡6 M ZEA, and 10 ng/ml hCG + 10¡4 M ZEA respectively. HPRT (290 bp) was used as an internal control. The densitometry analysis shows theratio (%) of 3�-HSD-1, p450scc, and StAR : HPRT mRNA expression post-treatment in (D)–(F), respectively. Values are expressed as the mean § SE ofthree separate experiments performed in triplicates in each treatment; Histograms within experiments with diVerent letters denote signiWcant diVerentratios (P < 0.05).

562 J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565

We also explored whether 3�-HSD-1, p450scc, andStAR mRNA expression is altered in the mouse Leydigcells exposed to hCG + �-ZOL using RT-PCR (Fig. 3A–C).PCR products around the predicted sizes of 565 bp, 370 bpand 310 bp had 100% homology in mouse Leydig cellswith those of mouse ovary 3�-HSD-1, p450scc and StARcDNA, respectively. As well, comparisons of 3�-HSD-1,p450scc, and StAR mRNA expression with diVerent �-ZOL-treated groups by semi-quantitative PCR were studied.

Exposure to hCG + �-ZOL caused signiWcant decreases inthe 3�-HSD-1 mRNA level relative to that of HPRT, at adose of 10 ng/ml hCG + 10¡4 M �-ZOL (lane 6, Fig. 3A andD, P < 0.05). Lanes 4–6 in Fig. 3B and E show signiWcantreductions in the p450scc mRNA level relative to that ofHPRT in Leydig cells exposed to hCG + �-ZOL, at doses of10 ng/ml hCG + 10¡8 M, 10 ng/ml hCG + 10¡6 M, and 10 ng/ml hCG + 10¡4 M �-ZOL. SigniWcant decreases in the StARmRNA level relative to that of HPRT were observed at

Fig. 3. RT-PCR analysis of 3�-HSD-1 (A), p450scc (B) and StAR (C) expression in mouse ovary (lane 2, positive controls) and mouse Leydig cells (lane 1,negative control; lanes 3–6) after 24 h incubation with hCG (10 ng/ml) and hCG (10 ng/ml) + �-ZOL(10-4 M, 10-6 M or 10-8 M). (A)–(C) show 565 bp,370 bp and 310 bp DNA fragments for 3�-HSD-1; P450scc and StAR, respectively. Lanes 3–6: 10 ng/ml hCG, 10 ng/ml hCG + 10¡8 M �-ZOL, 10 ng/mlhCG + 10¡6 M �-ZOL, and 10 ng/ml hCG + 10¡4 M �-ZOL respectively. HPRT (290 bp) was used as an internal control. The densitometry analysis showsthe ratio (%) of 3�-HSD-1, p450scc and StAR : HPRT mRNA expression post-treatment in (D)–(F), respectively. Values are the mean § SE of three sep-arate experiments performed in triplicates in each treatment; Histograms within experiments with diVerent letters denote signiWcant diVerent ratios(P < 0.05).

J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565 563

doses of 10 ng/ml hCG + 10¡8 M and 10 ng/ml hCG+10¡6 M (lane 5–6, Fig. 3C and F, P < 0.05).

3.3. Inhibitory eVects of ZEA and �-ZOL on testosterone secretion of hCG-treated Leydig cells

Exposure to hCG at a dose of 10 ng/ml resulted in a sig-niWcant increase in the levels of testosterone secretion inmouse Leydig cells. Furthermore, a signiWcant reduction intestosterone secretions were detected in Leydig cells co-treated with 10 ng/ml hCG and 10¡4 M, 10¡6 M, or 10¡8 MZEA (Fig. 4A) or �-ZOL (Fig. 4B).

4. Discussion

Previous research has demonstrated the negative eVectsof ZEA and �-ZOL on male fertility and other reproductivepathologies in male animals (Ruzsas et al., 1979; Young andKing, 1986; Wilson et al., 2002). These studies suggest thatZEA and �-ZOL exposure can interfere with the process ofspermatogenesis. As a result, these Wndings prompted thepresent investigation to examine possible eVects of ZEA and�-ZOL on Leydig cell functions, particularly for the synthe-sis and secretion of testosterone. In these experiments, basaltestosterone synthesis was very low and no signiWcantinhibitory eVects of ZEA and �-ZOL were found in unstim-ulated cells.

Steroid hormones are synthesized from cholesterol in thegonads in response to pituitary hormones, such as hCG viathe classical Wrst messenger/second messenger pathway.Conversion of cholesterol to biologically active steroids is amulti-step enzymatic process. Along with some importantenzymes, like cholesterol side-chain cleavage enzyme(P450scc) and 3�-hydroxysteroid dehydrogenase/isomer-ase (3�-HSD), several proteins play key roles in steroido-

genesis. Among these, the role of steroidogenic acuteregulatory (StAR) protein appears to transfer cholesterolfrom cellular stores to the inner mitochondrial membrane,where cholesterol is enzymatically converted to pregneno-lone by P450scc (Clark et al., 1994; Lin et al., 1995; Stoccoand Clark, 1996; Stocco, 1999; Strauss et al., 1999; Boseet al., 2002). The conversion of pregnenolone to progester-one is a step metabolized by 3�-HSD. The 3�-HSD-Ienzyme complex plays a crucial role in the conversion of�5-3�-hydroxysteroids to �4-3-oxosteroids, which is anessential step in the production of all active steroid hor-mones (Payne and Hales, 2004). Considering the impor-tance of 3�-HSD-1, P450scc, and StAR in regulatingsteroidogenesis and to delineate the underlying mechanismof hCG-mediated induction of testosterone synthesis, thechanges of 3�-HSD-1, P450scc and StAR expression inmouse Leydig cells were measured. The results indicate thathCG-stimulation of testosterone synthesis in primary cul-ture of Leydig cells was in agreement with the resultsreported by Caprio et al. (1999) and was further extendedby the enhanced expression of 3�-HSD-1, P450scc andStAR transcripts.

In the adult, testosterone supports spermatogenesis,sperm maturation, and sexual function (Ewing and Keeney,1993). Therefore, disruption of testosterone biosynthesis inLeydig cells can adversely aVect male fertility. The presentstudy was designed to evaluate the ability of ZEA and �-ZOL to alter (i) Leydig cell steroidogenic function, and (ii)the levels of testosterone in the co-treatment studies usingdoses of 10¡4 M, 10¡6 M and 10¡8 M ZEA or �-ZOL and10 ng/ml hCG. The results suggest that ZEA and �-ZOL atconcentrations of 10¡4 M, 10¡6 M and 10¡8 M (Fig. 4) sig-niWcantly inhibited the 10 ng/ml hCG-stimulated testoster-one synthesis. Therefore, the results demonstrate that ZEAand �-ZOL are potent inhibitors of testosterone production

Fig. 4. EVects of increasing concentrations of ZEA (A) or �-ZOL (B) on T production by mouse Leydig cells in culture in the absence or presence of hCG(10 ng/ml) after 24 h incubation. Data (means§ SE) are from three separate experiments performed in triplicates in each treatment. Histograms withinexperiments with diVerent letters denote signiWcant diVerent ratios (P < 0.05).

564 J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565

in mouse Leydig cells. Similar Wndings were observed ininterstitial cells of adult gerbils that were incubated with369� M ZEA, in addition, the literature also mentions thatZEA acts directly on testicular tissue, presumably by inhib-iting early steps of the steroidogenic pathway (Fenske andFink-Gremmels, 1990).

Disruption of androgen biosynthesis in Leydig cells hasbeen associated with factors that cause decreases in steroi-dogenic enzyme gene transcriptional activity, especially ofthe cytochrome P450 enzymes (Majdic et al., 1996). Theresults of the present study showed that ZEA and �-ZOLinhibition of hCG-stimulated mouse Leydig cells was dueto down-regulation of P450scc, the enzyme that catalyzesthe Wrst reaction in the testosterone biosynthetic pathway.In addition, the inhibitory eVect was attributed to down-regulation of 3�-HSD-1 and StAR (Figs. 2 and 3). There-fore, these results support the hypothesis that ZEA and�-ZOL inhibit testosterone production in hCG-stimulatedLeydig cells by inhibiting the expression of 3�-HSD-1,P450scc and StAR at the mRNA levels. However, theinhibitory eVect of HPTE on androgen biosynthesis is notrelated to the gene expression and enzyme activity of StARin rats (Akingbemi et al., 2000). The diVerent study resultsascribe possibly the structures of xenoestrogens, the species,reproductive status of the animal, length of the exposure,the way of treatment administration and treatment level.

Both ZEA and �-ZOL have estrogenic activities, how-ever �-ZOL has been shown to bind with greater aYnity toestrogenic receptors (Olsen, 1989). ZEA and �-ZOL inducehyper-estrogenism due to the competition of 17�-estradiolin the binding to cytosolic estrogen receptors with subse-quent activation of gene expression, increases in RNApolymerase I and II activity in nuclei, and synthesis of newproteins (Ueno, 1991). Some suspected endocrine disrup-tors have been shown to interact with the ER to interferewith steroid hormone synthesis or metabolism (LeBlancet al., 1997). Some xenoestrogens are known to preferen-tially bind ER� in vitro (e.g. bisphenol A and phytoestro-gens), and liganded ER� may have inhibitory eVects onER� signaling (Kuiper et al., 1998). EVects of xenoestro-gens on many functions of primary cells or cell lines aremediated via estrogen receptor signaling, which was vali-dated by using the ER antagonist. For example, the inXu-ence of ZEA on the gene expression of Wnt-7a in anendometrial adenocarcinoma cell line (Wagner and Leh-mann, 2006); the inhibitive eVects of ZEA on the TCDD-induced CYP1A1 activity and gene expression in MCF-7cells (Yu et al., 2004); A metabolite of methoxychlor, 2,2-bis(p-hydroxyphenyl)-1,1,1- trichloroethane, reduces testos-terone biosynthesis in rat Leydig cells (Akingbemi et al.,2000). According to the above literatures, we speculate thatZEA and �-ZOL have the ability to activate the ER andthereby regulate cellular function. Presumably, endocrinedisrupters may also cause these eVects by more than onemechanism. Therefore, further studies are required to ascer-tain the speciWc mechanisms of actions of ZEA and �-ZOLon the functions of Leydig cells.

In summary, this study demonstrates that ZEA and �-ZOL impair the abundance of essential rate-limitingenzymes, including 3�-HSD-1, P450scc and StAR transcrip-tion, and also inhibit the testosterone synthesis in mousehCG-stimulated Leydig cells in vitro. The results indicatethat the regulatory pathways are targets of ZEA and �-ZOLactions. Therefore, ZEA and �-ZOL could inXuence geneexpression at the transcription level. Whether or not ZEAand �-ZOL may also perturb other mechanisms in testos-terone synthesis, like HMG-Co reductase, remains to befurther investigated.

Acknowledgements

This work was supported by Grants of the Natural Sci-ence Foundation for Outstanding Young Scientists ofChina (30325034) and the Natural Science Foundation ofChina (30471264). We thank Prof. Yaoxing Chen (Facultyof Veterinary Medicine, China Agricultural University) forhis helpful suggestions. We are grateful for critical com-ments on this manuscript by Drs. Maryam Fathimani andCarol Forster (at University of Toronto).

References

Akingbemi, B.T., Ge, R.S., Klinefelter, G.R., Gunsalus, G.L., Hardy, M.P.,2000. A metabolite of methoxychlor, 2,2-bis(p-hydroxyphenyl)-1,1,1-trichloroethane, reduces testosterone biosynthesis in rat Leydig cellsthrough suppression of steady-state messenger ribonucleic acid levelsof the cholesterol side-chain cleavage enzyme. Biology of Reproduc-tion 62 (3), 571–578.

Akingbemi, B.T., Ge, R., Rosenfeld, C.S., Newton, L.G., Hardy, D.O., Cat-terall, J.F., Lubahn, D.B., Korach, K.S., Hardy, M.P., 2003. Estrogenreceptor-alpha gene deWciency enhances androgen biosynthesis in themouse Leydig cell. Endocrinology 144 (1), 84–93.

Biegel, L.B., Liu, R.C., Hurtt, M.E., Cook, J.C., 1995. EVects of ammoniumperXuorooctanoate on Leydig cell function: in vitro, in vivo, andex vivo studies. Toxicology and Applied Pharmacology 134 (1), 18–25.

Bose, H.S., Lingappa, V.R., Miller, W.L., 2002. The steroidogenic acuteregulatory protein, StAR, works only at the outer mitochondrial mem-brane. Endocrine Research 28 (4), 295–308.

Bottalico, A., Logrieco, A., Visconti, A., 1989. Fusarium species and theirmycotoxins in infected corn in Italy. Mycopathologia 107 (2–3), 85–92.

Caprio, M., Isidori, A.M., Carta, A.R., Moretti, C., Dufau, M.L., Fabbri,A., 1999. Expression of functional leptin receptors in rodent Leydigcells. Endocrinology 140 (11), 4939–4947.

Clark, B.J., Wells, J., King, S.R., Stocco, D.M., 1994. The puriWcation, clon-ing, and expression of a novel luteinizing hormone-induced mitochon-drial protein in MA-10 mouse Leydig tumor cells. Characterization ofthe steroidogenic acute regulatory protein (StAR). Journal of Biologi-cal Chemistry 269 (45), 28314–28322.

D’Mello, F.J.P., 1997. Handbook of Plant and Fungal Toxicants, Wrst ed.CRC Press, Boca Raton, New York. pp. 287–301.

Ewing, L.L., Keeney, D.S., 1993. Leydig cells: structure and function. In:Desjardins, C., Ewing, L.L. (Eds.), Cell and Molecular Biology of theTestis. Oxford University Press, New York, pp. 137–165.

Fenske, M., Fink-Gremmels, J., 1990. EVects of fungal metabolites on tes-tosterone secretion in vitro. Archives of Toxicology 64 (1), 72–75.

Gray Jr., L.E., Ferrell, J.M., Ostby, J.S., 1985. Alteration of behavioral sexdiVerentiation by exposure to estrogenic compounds during a criticalneonatal period: eVects of zearalenone, methoxychlor, and estradiol inhamsters. Toxicology and Applied Pharmacology 80 (1), 127–136.

J. Yang et al. / Toxicology in Vitro 21 (2007) 558–565 565

Jin, L., Zhang, S., Burguera, B.G., Couce, M.E., Osamura, R.Y., Kulig, E.,Lloyd, R.V., 2000. Leptin and leptin receptor expression in rat andmouse pituitary cells. Endocrinology 141 (1), 333–339.

Katzenellenbogen, B.S., Katzenellenbogen, J.A., Mordecai, D., 1979. Zea-ralenones: characterization of the estrogenic potencies and receptorinteractions of a series of fungal beta-resorcylic acid lactones. Endocri-nology 105 (1), 33–40.

Kellokumpu, S., 1987. DiVerent processing of LH/hCG receptors in cul-tured rat luteal cells and murine Leydig tumour cells (MLTC-1).Experimental Cell Research 168 (2), 299–308.

Kiang, D.T., Kennedy, B.J., Pathre, S.V., Mirocha, C.J., 1978. Bindingcharacteristics of zearalenone analogs to estrogen receptors. CancerResearch 38 (11 Pt 1), 3611–3615.

Kuiper, G.G., Lemmen, J.G., Carlsson, B., Corton, J.C., Safe, S.H., van derSaag, P.T., van der Burg, B., Gustafsson, J.A., 1998. Interaction ofestrogenic chemicals and phytoestrogens with estrogen receptor beta.Endocrinology 139 (10), 4252–4563.

LeBlanc, G.A., Bain, L.J., Wilson, V.S., 1997. Pesticides: multiple mecha-nisms of demasculinization. Molecular and Cellular Endocrinology126, 1–5.

Lin, D., Sugawara, T., Strauss 3rd, J.F 3rd., Clark, B.J., Stocco, D.M.,Saenger, P., Rogol, A., Miller, W.L., 1995. Role of steroidogenic acuteregulatory protein in adrenal and gonadal steroidogenesis. Science 267(5205), 1828–1831.

Majdic, G., Sharpe, R.M., O’Shaughnessy, P.J., Saunders, P.T.K., 1996.Expression of cytochrome P450 17a-hydroxylase/17–20 lyase in thefetal rat testisis reduced by maternal exposure to exogenous estrogens.Endocrinology 137, 1063–1070.

Manka, M., Visconti, A., Chelkowski, J., Bottalico, A., 1985. Pathogenicityof Fusarium isolates from wheat, rye and triticale toward seedling andtheir ability to produce trichothecenes and zearalenone. Phytopatho-logische Zeitscrift 113, 24–29.

Mayr, U.E., 1988. Estrogen-controlled gene expression in tissue culturecells by zearalenone. Febs Letters 239 (2), 223–226.

Mehmood, Z., Smith, A.G., Tucker, M.J., Chuzel, F., Carmichael, N.G.,2000. The development of methods for assessing the in vivo oestrogen-like eVects of xenobiotics in CD-1 mice. Food and Chemical Toxicol-ogy 38 (6), 493–501.

Mendelson, C., Dufau, M., Catt, K., 1975. Gonadotropin binding andstimulation of cyclic adenosine 3�:5�-monophosphate and testosteroneproduction in isolated Leydig cells. The Journal of Biological Chemis-try 250 (22), 8818–8823.

Muller, H.M., Reimann, J., Schumacher, U., Schwadorf, K., 1998. Naturaloccurrence of Fusarium toxins in oats harvested during Wve years in anarea of southwest Germany. Food Additives and Contaminants 15 (7),801–806.

Nikov, G.N., Hopkins, N.E., Boue, S., Alworth, W.L., 2000. Interactions ofdietary estrogens with human estrogen receptors and the eVect on

estrogen receptor-estrogen response element complex formation. Envi-ronmental Health Perspectives 108 (9), 867–872.

Olsen, M., 1989. Metabolism of zearalenone in farm animals. In: Chelkow-ski, J. (Ed.), Fusarium: Mycotoxins, Taxonomy and Pathogenicity.Elsevier, Amsterdam, pp. 167–177.

Osweiler, G.D., 1986. Occurrence and clinical manifestations of trichothe-cene toxicoses and zearalenone toxicoses. In: Richard, J.L., Thurston,J.R. (Eds.), Diagnosis of Mycotoxicoses. Martinus NijhoV, Dordrecht,pp. 31–50.

Payne, A.H., Hales, D.B., 2004. Overview of steroidogenic enzymes in thepathway from cholesterol to active steroid hormones. EndocrineReviews 25 (6), 947–970.

Ruzsas, C., Biro-Gosztonyi, M., Woller, L., Mess, B., 1979. EVect of thefungal toxin (zearalenone) on the reproductive system and fertility ofmale and female rats. Acta Biologica Academiae Scientiarum Hungari-cae 30 (4), 335–345.

Scudamore, K.A., Patel, S., 2000. Survey for aXatoxins, ochratoxin A, zea-ralenone and fumonisins in maize imported into the United Kingdom.Food Additives and Contaminants 17 (5), 407–416.

Scudamore, K.A., Nawaz, S., Hetmanski, M.T., 1998. Mycotoxins in ingre-dients of animal feeding stuVs: II. Determination of mycotoxins in maizeand maize products. Food Additives and Contaminants 15 (1), 30–55.

Stocco, D.M., 1999. Steroidogenic acute regulatory protein. Vitamins andHormones – Advances in Research 55, 399–441.

Stocco, D.M., Clark, B.J., 1996. Role of the steroidogenic acute regulatoryprotein (StAR) in steroidogenesis. Biochemical Pharmacology 51 (3),197–205.

Strauss 3rd., J.F., Kallen, C.B., Christenson, L.K., Watari, H., Devoto, L.,Arakane, F., Kiriakidou, M., Sugawara, T., 1999. The steroidogenicacute regulatory protein (StAR): a window into the complexities ofintracellular cholesterol traYcking. Recent Progress in HormoneResearch 54, 369–394. discussion 394–395.

Ueno, Y., 1991. Biochemical mode of action of mycotoxins. In: Smith, J.E.,Henderson, R.S. (Eds.), Mycotoxins and Animal Foods. CRS Press,Boca Raton, Boston, pp. 437–453.

Wagner, J., Lehmann, L., 2006. Estrogens modulate the gene expression ofWnt-7a in cultured endometrial adenocarcinoma cells. MolecularNutrition & Food Research 50 (4–5), 368–372.

Wilson, T.W., NeuendorV, D.A., Lewis, A.W., Randel, R.D., 2002. EVect ofzeranol or melengestrol acetate (MGA) on testicular and antler devel-opment and aggression in farmed fallow bucks. Journal of Animal Sci-ence 80 (6), 1433–1441.

Young, L.G., King, G.J., 1986. Low concentrations of zearalenone in dietsof boars for a prolonged period of time. Journal of Animal Science 63(4), 1197–1200.

Yu, Z., Hu, D., Li, Y., 2004. EVects of zearalenone on mRNA expressionand activity of cytochrome P450 1A1 and 1B1 in MCF-7 cells. Ecotoxi-cology and Environmental Safety 58 (2), 187–193.