J. Comp. Path. 1996 Vol. 114, 407 418

Effects o f In-utero E x p o s u r e to Z e r a n o l or D i e t h y l s t i l b o e s t r o l on M o r p h o l o g i c a l D e v e l o p m e n t

o f the Fetal Tes t i s in Mice

C. P6rez-Martinez, M.J. Garcia-Iglesias, M. C. Ferreras-Estrada, A. M. Bravo-Moral*,

J. Espinosa-Alvarez and A. Escudero-Diez Histology and Pathological Anatomy Section, Department of Animal Pathology: Animal Medicine, Faculty of Veterinary Science, University of Ledn, 24071 Ledn, and *Department of Pathology,

Faculty of Veterinary Science, University of Santiago de Compostela, E-27002-Lugo, Spain

Summary

The morphological development of the fetal mouse testis exposed to ~- zearalanol (zeranol) or diethylstilboestrol (DES) was evaluated as part of an examination of the effects oftransplacental exposure to non-steroid oestrogens on susceptible tissues. On days 9 and 10 of gestation, pregnant NMRI mice were given subcutaneous injections of ethyl oleate alone (0" 1 ml) or zeranol or DES (150 gg/kg body weight) in ethyl oleate. The mice were killed from days 12 to 18 of gestation and the male fetuses were examined. Microscopical examination of the gonads indicated that the onset of testicular differentiation was earlier in the oestrogen-treated fetuses than in controls. Abnormal differentiation of gonocytes and foci of hyperplasia of fetal Leydig cells were observed in the oestrogen-treated mice. Male fetuses from female mice treated with DES showed a delay in testicular descent and progressive decrease in reactivity for cytokeratin (CK) 8 in fetal Sertoli cells. These morphological findings suggest that prenatal exposure to zeranol or DES induces abnormal testicular differentiation in the mouse. �9 1996 W.B. Saunders Company Limited

Introduction

Diethylstilboestrol (DES), a potent non-steroidal synthetic oestrogen, has been widely used for therapeutic and agricultural purposes. Its use has been associated with the appearance of certain abnormalities, including tumours, in the offspring of women exposed to it during gestation (Barter et al., 1986; Newbold, 1993). As a consequence of the adverse effects of DES, other oestrogenic substances, such as the mycoestrogen, ~-zearalanol (zeranol), have been used as alternatives (Baldwin et al., 1983).

Although the effects of DES on fetal development have been the subject of numerous studies (Maydl and Metzler, 1984; Wordinger et al., 1989; Zimm- erman et al., 1991), the action of zeranol is not well known. Zeranol is obtained from a mycotoxin (zearalenone; "ZEN") produced by Fusarium graminearum (Ueno, 1985). Since ZEN and its derivatives, including zeranol, possess a

0021-9975/96/040407 + 12 $12.00/0 �9 1996 W.B. Saunders Company Limited

408 C. P6rez-Martlnez e t al .

potent anabolic property, they have been widely used in agriculture (Ueno, 1985). Their use is forbidden in most countries, but ZEN has been detected in hay, feed, corn, sorghum, dairy rations and barley (Ueno, 1985); thus human beings may be exposed to this environmental oestrogen via meat.

Administration of oestrogenic compounds to laboratory animals at specific times during prenatal development is a useful method for examining their influence on the normal development of offspring (Arai et al., 1983; Pylkk~nen et al., 1991). In most mammals, including human beings and mice, differ- entiation of the reproductive tract is strongly affected by prenatal exposure to DES (Bullock et al., 1988). Some changes in the genital tract may result directly or indirectly in carcinogenesis, as well as sterility in later life (Newbold et al., 1987; Ozawa et al., 1991).

The objective of the study reported here was to examine the effects of prenatal administration of zeranol or DES (positive control) on the morphological development of the testis. The mouse was used because it is particularly sensitive to oestrogens and provides a good model system for understanding the influence of prenatal exposure to oestrogens in man (Walker, 1989).

M a t e r i a l s and M e t h o d s

Animals and Treatment Schedule

NMRI female mice (n= 94), aged 6-8 weeks and weighing 25-30 g, were obtained from Antibi6ticos Laboratories S.A. (Le6n, Spain). They were housed in groups of eight for 2-3 weeks, with food and water ad libitum. A cycle of 14 h light and 10 h dark was maintained, with controlled temperature (22-t-2~ Male mice of proven fertility were housed individually under similar conditions. To obtain timed preg- nancies, individual female mice were placed randomly with individual male mice at 20.00 h and inspected daily at 09.00 h for the appearance of a vaginal plug, which indicated mating. The morning when the vaginal plug was found was designated day 0 of pregnancy. Pregnant mice were housed in individual cages. The numbers of pregnant mice used were 21, 38, and 35 for the control, DES-treated and zeranol- treated groups, respectively.

Zeranol and DES were supplied by Sigma Chemical Company, St Louis, MO, USA. Pregnant mice were given subcutaneous injections of zeranol or DES (150 gg/ kg body weight) in 0" 1 ml of ethyl oleate (Syva Laboratories, Le6n, Spain) on days 9 and 10 of gestation. Control mice received 0"l-ml injections of ethyl oleate alone. For the collection of fetuses, pregnant mice from each group were killed by cervical dislocation on days 12 (three control, five DES-treated and five zeranol-treated mice), 13 (three control, five DES-treated and five zeranol-treated mice), 14 (three control, five DES-treated and five zeranol-treated mice), 15 (three control, eight DES-treated and five zeranol-treated mice), 16 (three control, five DES-treated and five zeranol- treated mice), 17 (three control, five DES-treated and five zeranol-treated mice), and 18 (three control, five DES-treated and five zeranol-treated mice) of pregnancy. At each age and treatment, at least 10 male fetuses were examined.

Light Microscopy

At each age, 50% of the male fetuses collected were fixed in Bouin's solution, and embedded in paraffin wax. Serial 3-gm sections of each fetus were made and stained with haematoxylin and eosin (HE) and by Masson-Goldner's trichrome stain.

E f f e c t s o f N o n - s t e r o i d O e s t r o g e n s o n T e s t i s

Table 1 P r i m a r y a n t i b o d i e s u s e d in i m r n u n o b i s t o c h e m i c a l s t u d y on fetal t e s t i s

409

Clones Character~Species Specificity Dilution Origin

AntiCK 5* p o / R b C K 5 1 in 500 D. Roop LL001* m o / M C K 14 ud Lane and Leigh T R O M A I * m o / R C K 8 1 in 2 R. Kemler LE61* m o / M C K 18 ud E.B. Lane LP1K* m o / M C K 7 ud E.B. Lane LP2K* m o / M C K 19 ud E.B. Lane Ant iCK 6* p o / R b C K 6 1 in 500 D. Roop L-9393~ p o / R b Laminin 1 in 10 Sigma

mo = Monoclonal, po = polyclonal, M = mouse, R = rat, Rb = rabbit, ud = undiluted, C K = cytokeratin. * Primary antibodies used with the indirect immunofluorescence technique (on frozen tissue). "~ Primary antibody used with the Lab-SA system (on fetuses fixed in Bouin's solution). Primary antibodies kindly supplied by R. Kemler, Max Planck Institute, Freiburg, Germany; E. B. Lane, Medical Science Institute, University of Dundee, Scotland, UK; D. Roop, Baylor College of Medicine, Houston, Texas, USA.

Immunohistochemistry

Serial 7-gm sections of fetuses fixed in Bouin's solution and embedded in paraffin wax were immunolabelled by the labelled-Streptavidin biotin system (Lab-SA; Zymed Laboratories, San Francisco, CA, USA). The remaining 50% of the male fetuses collected were immediately frozen at --20~ in embedding medium (Tissue-Tek II, OCT compound; Miles Laboratories, IN, USA) for examination by the indirect immunofluorescence technique. Frozen sections of 7-gm thickness were cut with a cryostat (Leitz, Switzerland).

Table 1 shows the working dilutions and the characteristics of the primary antibodies used. For Lab-SA validation, a biotinylated goat anti-rabbit IgG (whole molecule) (Sigma Chemical Co., St Louis, MO, USA) was used as secondary antibody, diluted 1 in 200. Diaminobenzidine tetrahydrochloride was used in the substrate-chromogen solution. For the indirect immunofluorescence technique, the secondary antibodies were conjugated to fluorescein isothiocyanate and used at a 1 in 100 dilution. Sections were examined under a microscope equipped for epifluorescence (Leitz, Switzerland). Negative and positive controls for the primary and secondary antibodies were included.

R e s u l t s

Studies on serial sections of control male fetuses revealed that the gonads, initially located close to the mesonephric masses, began their descent on day 15 of gestation (Fig. l a). Similar results were obtained in the zeranol-treated mice. However, in six of 10 fetuses from female mice treated with DES, the gonads were located at the caudal end of the metanephros on day 15 of gestation (Fig. lb). On the following days, testicular descent in these fetuses became similar to that of the control and zeranol-treated fetuses.

Sertoli Cells

Gonadal ridges of control fetuses consisted of the gonadal blastema and the surface epithelium on day 12 of gestation. The gonadal blastema was composed mainly of somatic cells, which partly surrounded numerous primodial germinal cells (PGCs) and formed irregular cellular cords. In the immunohistochemical

410 C. P~rez-Martinez e t al .

Fig. 1. Position of the testes of fetuses on day 15 of gestation. (a) Control fetus. Initiation oftransabdominal migration of the gonad. HE. Scale b a r = 1000 gm. (b) Male fetus exposed to DES on days 9 and 10 of gestation. Gonad retained close to the metanephros. Masson-Goldner 's trichrome. Scale b a r = 1000 gm. M =metanephros ; T=tes t i s ; L=liver ; P=pancreas .

evaluation of laminin expression, neither epithelial organization nor any continuous basement membrane was observed (Fig. 2a). At 12 days of gestation, differentiation of primordial Sertoli cells in the fetal gonad was the first

E f f e c t s o f N o n - s t e r o i d O e s t r o g e n s o n T e s t i s 41 1

Fig. 2. Gonadal blastema from fetuses on day 12 of gestation. Immunohistochemical detection of laminin. Lab-SA. (a) Irregular arrangement of the cellular cords of control fetus. No visible basement membrane, x 60. (b) Distinct seminiferous cords (arrowheads) surrounded by a basement membrane of a fetus exposed to zeranol, x 25. S = somatic cell; PGC = primordial germinal cell; P = primordial Sertoli cell.

morphological evidence of testicular differentiation. The progressive ag- gregation of (hese cells determined the formation of seminiferous cords on day 13 of gestation. The primordial Sertoli cells were characterized by an elongated or oval-shaped nucleus, often with one or more prominent nucleoli

412 C. P6rez-Martinez e t al .

and a clear cytoplasm. They could be distinguished readily from PGCs. However, an early aggregation of the primordial Sertoli cells was observed in fetuses from female mice treated with zeranol or DES on day 12 of gestation (Fig. 2b).

From day 15 of gestation onwards, Sertoli cells were polarized, displaying their nucleus and most of their cytoplasm at the periphery of the seminiferous cords, in the three groups analysed. Immunohistochemical examination for eytokeratins (CKs) showed a positive reaction with the anti-CK8 antibody in the primordial Sertoli cells, whereas no significant staining was observed in the PGCs of control or oestrogen-treated fetuses. The reaction was strongly positive in the basal part of the Sertoli cells, but the lateral cytoplasmic processes extending between the PGCs were immunolabelled more faintly. The reaction was lost in the lateral cytoplasmic processes and remained only weakly at the basal pole of these cells on day 18 of gestation in the DES- treated fetuses (Fig. 3a)--unlike the control and zeranol-treated fetuses, in which it was still evident at this age (Fig. 3b). In undifferentiated and differentiated gonads, no significant staining was observed with the other anti- CK antibodies used.

Gonocytes

PGCs or gonocytes were identified by their large, roundish nucleus, containing one or two prominent nucleoli. At 12 days of gestation, PGCs were found throughout the gonadal ridge in control fetuses (Fig. 2a), whereas these cells were uniformly distributed within the seminiferous cords in fetuses from female mice treated with zeranol or DES (Fig. 2b).

In the control and zeranol-treated fetuses, gonocytes proliferated until day 15 of gestation and their nuclear morphology varied, depending on the phase of the cell cycle. At 15 days of gestation, cellular proliferation ceased and gonocytes were centrally located in the sex cords. However, in all DES-treated fetuses these cells continued their division until day 16 or 17 of gestation, when all divisions ceased, to be resumed after birth, as in the control and zeranol-treated male mice. The mitoses, more numerous in the fetuses exposed to the oestrogens than in the control fetuses, were on occasion atypical (abortive mitoses, irregular arrangement of chromosomes at the equatorial plate, or failure of certain chromosomes to follow the movement of others towards the poles of the achromatic apparatus) in the DES-treated group. Degenerating and developing gonocytes were more numerous in the fetuses treated with zeranol or DES (Fig. 4a) than in the control fetuses (Fig. 4b). Multinucleated giant cells were frequently observed on days 17 and 18 of gestation in the oestrogen-treated mice (Fig. 4c).

L dig CeZls

In the control and experimental gonads, between days 12 and 14 of gestation, most of the cells in the interstitium were undifferentiated interstitial cells, with varying morphology. These cells were characterized by a small amount of

E f f e c t s o f N o n - s t e r o i d O e s t r o g e n s o n T e s t i s 413

Fig. 3. Reactivity of anti-CK8 monoclonal antibody by an indirect immunofluorescence assay. Day 18 of gestation. (a) Male fetus exposed to DES. Negative reaction in the lateral cytoplasmic processes and weak immunolabelling at the basal pole of fetal Sertoli cells, x 154. (b) Control fetus. Positive reaction in the basal part and lateral cytoplasmic processes of fetal Sertoli cells, x 38. TC = testicular cords.

cytoplasm and an oval or elongated nucleus, containing one or two nucleoli. After day 15 of gestation, in the control testes, many interstitial cells had differentiated into fetal Leydig cells with an abundant eosinophilic cytoplasm

414 C. P6rez-Martlnez e t al.

Fig. 4. Testes from fetuses on day 17 of gestation. (a) Male fetus exposed to zeranol. Note greater number of developing and degenerating (arrowheads) gonocytes. HE. x 60. (b) Control testis. No degenerating gonocytes are observed. Note the fetal Leydig cells (L) differentiated in the interstitium. HE. x 60. (c) Male fetus exposed to zeranol. Multinucleated giant cells (arrowheads) are observed. HE. x 154.

a n d a r o u n d e d nucleus centrally or eccentrically located (Fig. 4b), whereas in thd fetuses from mice treated with zeranol or DES, fetal Leydig cells remained less differentiated and there was an increased number of them in the inter- tubular stroma, forming clusters (interstitial cell hyperplasia) (Fig. 5).

Effec t s o f N o n - s t e r o i d O e s t r o g e n s o n T e s t i s 415

Fig. 4. continued.

D i s c u s s i o n

This study indicated that prenatal administration of zeranol to pregnant NMRI mice on days 9 and 10 of gestation altered prespermatogenesis and affected normal differentiation of fetal Leydig cells. Although human and rodent fetuses are bathed in endogenous oestrogens, it is believed that these substances have little effect on the genital tract of the fetuses, since they bind to extracellular carrier proteins and are conjugated and metabolized to inactive forms (McLachlan and Newbold, 1987). However, it seems clear that exposure to exogenous oestrogens during the fetal period has a pronounced effect on sexual differentiation (Newbold et al., 1987). In the present study, prenatal treatment with zeranol or DES induced the appearance of seminiferous cords in certain areas of the gonadal blastema at earlier ages than in controls, a feature that, to the authors' knowledge, has not been reported hitherto.

In agreement with previous studies on control rat fetuses (Fridmacher et al., 1992), our immunohistochemical results showed the presence of CK8 in fetal Sertoli cells in undifferentiated and differentiating gonads, both in control and oestrogen-treated fetuses. The fact that the Sertoli cells in the fetuses treated with DES lost CK8 expression earlier than did those of the other two groups indicated a conformational change in their cytoskeleton, probably related to functional alterations in these cells. However, the positive results obtained for CK18 and CK19 in rat Sertoli cells during the prenatal period (Fridmacher et al., 1992) are not easily reconciled with our negative results for these anti- CK antibodies. Possibly the amount of CK18 and CK19 in the fetal mouse testis was insufficient to permit their detection. The delay in testicular descent and the persistence of the mtillerian duct (data not shown) in the fetuses from

416 C. P6rez-Martinez e t al .

Fig. 5. Testis froln fetus exposed to zeranol; day 17 of gestation. Hyperplastic areas of fetal Leydig cells between the seminiferous cords. HE. x 60.

the DES-treated female mice are findings similar to those recorded by Luthra and Hutson (1989) after prenatal administration of oestradiol benzoate to mice. Those lesions were associated with altered secretion of the "mtillerian inhibiting substance" (MIS) by the Sertoli cells (Luthra and Hutson, 1989).

In the fetuses treated with oestrogens, large numbers of gonocytes and multinucleated cells were observed. These changes may be attributed to a direct effect on gonocytes since, as has been demonstrated already with DES and its metabolites, these substances can establish a covalent binding to their DNA and modify it (Gladek and Lierh, 1991). It has also been pointed out that this synthetic oestrogen can even alter the mitotic spindle of the gonocytes (Pylkkanen et al., 1991).

The increased number of immature Leydig ceils from day 16 of gestation, found in the fetuses treated with zeranol or DES, has also been reported in 18-day-old fetuses treated with ethinyl oestradiol (Yasuda et al., 1986). To explain this phenomenon, Yasuda et al. (1986) concluded that sites affected by ethinyl oestradiol in the immature Leydig cells were located in the nucleus, where functional disturbances of DNA appeared as proliferation of Leydig cells, with reduction in their smooth endoplasmic reticulum (site of enzymes related to the metabolism of steroid hormones); this resulted in reduced utilization of substrate for steroid biosynthesis, with accumulation of great numbers of lipid droplets in the cytoplasm.

The results of this study of prenatal exposure to DES are similar to those described by other authors, although the treatment schedule was different.

Effects of Non-steroid Oestrogens on Testis 417

However, the effects produced by the prenatal administration of zeranol have not been recorded previously. The main conclusion is that zeranol, although a less potent oestrogen than DES, resembles it in causing abnormalities in the testicular differentiation of the fetal mouse. Such abnormalities might result in decreased fertility.

A c k n o w l e d g m e n t s

We thank R. Kemler, E. B. Lane, I. M. Leigh and D. Roop for providing the primary antibodies used in this work.

R e f e r e n c e s

Arai, Y., Mori, T., Suzuki, Y. and Bern, H.A. (1983). Long-term effects of perinatal exposure to sex steroids and diethylstilbestrol on the reproductive system of male mammals. International Review of Cytology, 84, 235-268.

Baldwin, R. S., Williams, R.D. and Terry, M.K. (1983). Zeranol: a review of the metabolism, toxicology, and analytic methods for detection of tissue residues. Regulatory Toxicology and Pharmacology, 3, 9-25.

Barter,J. F., Austin,J. M. and Shingleton, H. M. (1986). Endometrial adenocarcinoma after in utero DES exposure. Obstetrics and Gynecology, 67, 845-851.

Bullock, B.C., Newbold, R.R. and McLachlan, J.A. (1988). Lesions of testis and epididymis associated with prenatal diethylstilbestrol exposure. EnvironmentalHealth Perspective, 77, 29-31.

Fridmacher, V., Locquet, O. and Magre, S. (1992). Differential expression of acidic cytokeratins 18 and 19 during sexual differentiation of the rat gonad. Development, 115, 503-517.

Gladek, A. and Liehr, J. G. (1991). Transplacental genotoxicity of diethylstilbestrol. Carcinogenesis, 12, 773 776.

Luthra, M. and Hutson, J .M. (1989). Late-gestation exogenous oestrogen inhibits testicular descent in fetal mice despite Mallerian duct regression. Pediatric Surgery International, 4, 260-264.

Maydl, R. and Metzler, M. (1984). Oxidative metabolites of DES in the fetal Syrian golden hamster. Toxicology, 30, 351-357.

McLachlan,J. A. and Newbold, R. R. (1987). Estrogen and development. Environmental Health Perspective, 75, 25 27.

Newbold, R.R. (1993). Gender-related behavior in women exposed prenatally to diethylstilbestrol. Environmental Health Perspective, 101,208-213.

Newbold, R.R., Bullock, B.C. and McLachlan, J.A. (1987). Testicular tumors in mice exposed in utero to diethylstilbestrol. Journal of Urology, 138, 1446 1450.

Ozawa, S., Iguchi, T., Sawada, K., Ohta, Y., Takasugi, N. and Bern, H.A. (1991). Postnatal vaginal nodules induced by prenatal diethylstilbestrol treatment cor- relate with later development of ovary-independent vaginal and uterine changes in mice. Cancer Letters, 58, 167-175.

Pylkk~tnen, L., Jahnukainen, K., Parvinen, M. and Santti, R. (1991). Testicular toxicity and mutagenicity of steroidal and non-steroidal estrogens in the male mouse. Mutation Research, 261, 181 191.

Ueno, Y. (1985). The toxicology of mycotoxins. CRC Critical Reviews in Toxicology, 14, 445 448.

Walker, B. E. (1989). Animal models of prenatal exposure to diethylstilbestrol. Perinatal and Multigeneration Carcinogenesis, 25, 349 364.

Wordinger, R.J., Brown, D., Atkins, E. and Jackson, F.L. (1989). Superovulation and early embryo development in the adult mouse after prenatal exposure to diethylstilbestrol. Journal of Reproduction and Fertility, 85, 383 388.

418 G. P6rez-Martinez e t al.

Yasuda, Y., Konishi, H. and Tanimura, T. (1986). Leydig cell hyperplasia in fetal mice treated transplacentally with ethinyl-estradiol. Teratology, 33, 281-288.

Zimmerman, S.A., Clevenger, W.R., Brimhall, B.B. and Bradshaw, W. S. (1991). Diethylstilbestrol induced perinatal lethality in the rat. II. Perturbation of par- turition. Biology of Reproduction, 44, 583-589.

Received, September 15th, 19951 Accepted, January 15th, 1996 J