Effects of in utero Exposure to Nonsteroidal Estrogens - NCBI

5 downloads 0 Views 1MB Size Report
of the effects of transplacental exposure to non-steroidal estrogens ..... gen Carcinogenesis 1989; 25: 349-364. 12. REEL JR, LAWTON AD, WOLKOW-. SKI-TYL ...
Effects of in utero Exposure to Nonsteroidal Estrogens on Mouse Testis Claudia Perez-Martinez, Maria C. Ferreras-Estrada, Maria J. Garcia-Iglesias, Ana M. Bravo-Moral, Javier Espinosa-Alvarez, and Alfredo Escudero-Diez

ABSTRACT Male mice exposed in utero to a-zearalanol (zeranol) or diethylstilbestrol (DES) were analyzed postnatally to evaluate the possible changes on their testicular morphology as part of an examination of the effects of transplacental exposure to non-steroidal estrogens on sensitive tissues. Pregnant NMRI mice were injected subcutaneously with ethyl oleate (0.1 mL) alone (negative control) or with 150 pg/kg of body weight of zeranol or DES (positive control) on days 9 and 10 of gestation. Experimental and control male offspring were euthanized at days 45 (n = 47), 90 (n = 44), 180 (n = 40) and 365 (n = 26) after birth and their gonads were examined by light and electron microscopy. The results suggested that prenatal zeranol or DES exposure induced more severe and earlier (at 45 d) testicular abnormalities than in negative control (at 6 mo). These age-related alterations were characterized by regressive changes in the germinal epithelium and Sertoli's cells as well as foci of Leydig's cells around atrophied seminiferous tubules and dysplasia of the rete testis epithelium. On the contrary, the presence of Leydig's cells with immature morphology and their arrangement in sheet could be attributable exclusively to estrogen treatment. The presence of no neoplasm was confirmed.

des souris males exposes au ot-zearalanol (zeranol) ou au diethylstilbestrol (DES) in utero ont ete' examines dans la periode postnatale afin d'e'valuer des changements possibles de morphologie testiculaire. Des souris NMRI gestantes furent injectees par voie sous-cutanee avec 0,1 mL d'oleate d'ethyl (temoin negatif) ou contenant 150 pg/kg de poids corporel de zeranol ou de DES (temoin positif) aux jours 9 et 10 de gestation. Les descendants males des groupes temoin et traite furent euthanasies aux jours 45 (n = 47), 90 (n = 44), 180 (n = 40) et 365 (n = 26) apres la naissance et leurs gonades examinees par microscopie photonique et electronique. Les resultats suggerent qu'une exposition prenatale au zeranol ou au DES induit chez les animaux issus de femelles traitees des anormalites testiculaires plus severes et plus tot (a 45 jours d'age) que chez les temoins negatifs (a 6 mois d'age). Ces alte'rations reliees 'a l'age etaient caracterisees par des changements regressifs de l'epithelium germinal et des cellules de Sertoli de meme que des foyers de cellules de Leydig autour des tubules seminiferes atrophies et de dysplasie de l'epithelium du rete testis. La presence de cellules de Leydig morpholgiquement immatures et leur arrangement en feuillets pouvait etre exclusivement attribuable au traitement aux estrogenes. Aucune forme neoplasique ne fut detectee. (Traduit par docteur Serge Messier)

RESUME Dans le cadre d'une etude des INTRODUCTION effets sur les tissus susceptibles d'une exposition transplacentaire In most mammals, including human a des estrogenes non-steroidiens, beings and mice, differentiation of the

reproductive tract is strongly affected by exposure to estrogens (1). The best known example of an estrogenic compound with adverse effects is DES, which is an established transplacentally active genotoxic agent (2). Its use has been associated with the subsequent development of cervical and vaginal neoplasms in young women that were exposed to DES during gestation (3). Human male offspring of treated mothers with DES also have an increased incidence of genital tract abnormalities such as cryptorchidism, hypoplastic retained testes and epididymal cysts (4). Although these authors did not observe signs of malignancy or tumours, it would be premature to assume such, as the patients were still young at the time of study (5). Some morphological development changes in the male mouse reproductive tract associated with in utero exposure to DES are: (a) testes with substantially reduced spermatogenesis, giant cells contained within the lumina of the seminiferous tubules, (b) testes composed of atrophied seminiferous tubules with hyalanized basement membranes and thickened arterioles, (c) testes composed mostly of necrotic or scar tissue with absence of spermatogenesis, (d) interstitial cell tumors, and (e) multifocal or diffuse changes in the rete testis (6). As a consequence of the adverse effects of DES, zeranol was considered as a possible alternative because it is a weak mycoestrogen with low toxicity (the oral mouse LD50 exceeding 40 g/kg) (7). However, available information on the possible effects of this compound is limited and mainly refers to in vitro (8) and in vivo (9) studies after postnatal exposure. Recently it has been pointed out that administration of estrogenic

Histology and Pathological Anatomy Section, Department of Animal Pathology, Animal Medicine, Faculty of Veterinary Science, University of Le6n, 24071 Le6n, Spain. Received October 2, 1995.

94

Can J Vet Res 1997; 61: 94-98

Table I. Effects of prenatal exposure to zeranol or DES on various parameters in NMRI male mice at different ages postpartum Testicular weight (g)

Testicular density (g/mL)

Diameter seminiferous tubules (pum)

Group n Body weight (g) 45 days postnatal Control 20 34.72 + 0.77 0.10 1 0.01 1.34 ± 0.10 207.34 ± 2.61 DES 14 34.36 + 1.17 0.09 ± 0.01 1.08 ± 0.06 207.78 ± 2.66 Zeranol 13 31.53 ± 1.29 0.09 10.01 1.12 ± 0.12 203.58 ± 2.93 90 days postnatal Control 21 39.64 + 1.26 0.10 + 0.01 1.08 ± 0.06 220.08 + 2.33 DES 11 39.76 1 1.42 0.10 1 0.01 1.42 ± 0.34 226.05 1 2.38 Zeranol 12 42.49 1 1.48 0.10 1 0.01 1.05 ± 0.04 221.90 1 2.93 180 days postnatal Control 15 47.49 ± 1.69a 0.10 + 0.01 1.19 + 0.06 225.81 + 2.46 DES 11 48.96 ± 1.40a 0.11 ± 0.01 1.22 ± 0.03 233.78 + 2.38 Zeranol 14 41.37 + 1.54 0.10 ± 0.01 1.23 ± 0.07 219.36 + 2.80b 365 days postnatal Control 7 44.76 + 1.40 0.11 + 0.01 1.19 + 0.08 204.55 1 3.30c DES 10 48.25 1 1.43 0.15 ± 0.04 1.13 10.11 228.69 ± 2.38 Zeranol 9 46.39 1.51 0.10 1 0.01 1.38 10.11 212.08 1 3.09c n = Number of male mice studied Values are means 1 standard error of the mean (SEM): a Significantly different from zeranoltreated group (P < 0.01); bSignificantly different from DES-treated group (P < 0.01); cSignificantly different from DES-treated group (P < 0.001).

compounds at an early stage of testic- MATING AND TREATMENT SCHEDULE ular development (prenatal period) To achieve timed pregnancies, makes a more sensitive toxicity test individual female mice were ranthan many in vitro tests (10). Admin- domly placed with individual male istration of these substances to mice (20:00 h) and inspected daily rodents at specific times during prena- (09:00 h) for the appearance of a vagital development has been documented nal plug, which was considered an as a useful method for examining indication of mating. The morning their potential influence on normal when the vaginal plug was found was development of the gonad in human designated day 0 of gestation. Pregmales (11). Thus, the main objectives nant mice were housed in individual of this study were to evaluate the cages. effects of prenatal exposure to zeranol On days 9 and 10 of gestation, on the subsequent morphological pregnant mice were injected subcutadevelopment and occurrence of neo- neously with zeranol (Sigma Chemiplasms in the gonad of male mice off- cal Company, St-Louis, Missouri, spring, using DES as a positive USA) or DES (Sigma Chemical Comcontrol (12). pany) at a dose of 150 ,ug/kg of body weight in 0.1 mL ethyl oleate (Syva MATERIALS AND METHODS Laboratorie, Le6n, Spain). Negative control mice received 0.1 mL injections of ethyl oleate alone. ANIMALS Pregnant mice were allowed to give Six- to 8-week old NMRI mice, birth. When 28 d of age offspring weighing 25 to 30 g, were obtained were weaned, from Antibioticos Laboratories S.A. and housed in segregated by gender groups of 5 per cage. (Leon, Spain). Female mice were housed in groups of 8 for 2 to 3 wk, TISSUE RECOVERY with food and water provided ad libiAt 45, 90, 180 and 365 d of age, tum. A cycle of 14 h light and 10 h dark was maintained, with a con- 63 control, 46 DES-treated and trolled temperature (22 ± 2°C). Male 48 zeranol-treated male offspring mice of proven fertility were housed were euthanized by cervical dislocaindividually under similar conditions. tion. The abdominal cavity was All experiments were performed fol- quickly opened and gross structural lowing the guidelines of the Guide to abnormalities in the reproductive tract the Care and Use of Experimental including location of testes were noted. The body and testicular Animals (13).

*.l,k'-4>j+7t}S9Wsia?;v X

,

,1;

(

,,

A

;

-

Figure 1. 45-day'old negative control mouse. Active spermatogenesis is observed in the seminiferous tubules. H&E. (Bar ..225 'm)

Figure 2. 365-day-old negative control mouse. Few atrophied seminiferous tubules surrounded by Leydig's cell focal accumulations. Silver methenamine stain. (Bart 125 t(m) ,

-

.

itr. is

wlt*,,.w;4fla.

,-

;

s:

38

A

>

'

i*

'

*e

"

1*

,, 9

4 e. %

^ t .vfi

. 1

_

.6.

Sf

M 4s M'sst tX

:-N

Figure 3. 45-day-old prenatally DES-treated mouse. Various giant multinucleated cells (arrowheads) have sloughed into the lumen and Sertoli's cells are present with vacuoles (arrows) in their cytoplasm. Toluidine blue. (Bar = 75 ,um)

weights were recorded using an electronic analytical and precision balance (BA210S, d = 0.0001 g) (Sartorius GA, Goettingen, Germany) and the testicular size (length and width) was recorded by use of a sliding guage (d = 0.1 mm) (Casa Alvarez, Madrid, Spain). The gonadal volume was measured using the formula V = 4'rAB2/3 where A is half of the longest diameter and B is half of the shortest diameter. Thereafter testicular

95

f7

exposedmouse. Elec

...mi.a

j X2 exoe

mose

Elcto

Folded nucleus (N); (b

mirorah .of

a,lte nra

e

niol a

specializations (JS) between them (N), laminn propria (L). (Bars = 1 p.m) STATISTICAL ANALYSIS

All the data concerning testicular density was obtained using the weight and density, diameter of semifollowing formula: d = mass/volume. niferous tubules as well as body All testicular measurements were weight of the male offspring were recorded on fresh tissue. The diameter analyzed by use of ANOVA. Signifiof 10 round-shaped seminiferous cant differences between groups were tubules was measured in 3 cross- identified by use of Newman-Keuls sections per testis with an ocular sequential range tests (14). For the micrometer (PK 12.5xm) (C. Reichert, study of the testicular parameters, Austria) calibrated with a stage data from right and left gonads were compared by Student's t-test. Data are micrometer.

Figure 5. 45-day-old negative control mouse. Electron micrograph of Sertoli's cells showing normal Sertoli-to-Sertoli cell junctional specializations (JS). Nucleus of Sertoli's cell (N) (Bar = 0.5 p.m) no significant differfound in testicular weight and density depending on factor "time" effect and "time/treatment" interaction effect.

0.01). However,

ence was

HISTOLOGICAL FEATURES

In the negative control mice, active

spermatogenesis was observed at 45 d of age (Fig. 1). At 6 mo, testicular abnormalities appeared in these animals which were consistent with usual

age-related changes. Initially, these histological changes consisted of

a

mild vacuolization in the seminiferous epithelium with damaging of gerexpressed as mean ± standard error of minal and Sertoli's cells (in about 3% LIGHT MICROSCOPY the mean (SEM). Differences were of the testes). These alterations were At each age, half of each gonad considered statistically significant at developed with age appearing atrofrom all the mice were fixed in P < 0.05. phied tubules in about 10% of the Bouin's solution and embedded in testes in 1 year-old mice (Fig. 2). of 1m wax. 3 sections Serial Likewise Leydig' s cell focal accumuparaffin RESULTS each testis were made and stained lations around atrophied tubules with hematoxylin and eosin (H&E), 2) and mild dysplastic changes (Fig. GROSS MORPHOLOGY and by silver-methenamine stain. of rete testis epithelium were All the mice belonging to the observed at 1 y of age. ELECTRON MICROSCOPY Table I shows the treatment effect estrogen-treated groups showed testes Those testes that showed lesions by normally descended as in the negative on the diameter means of seminiferlight microscopy were analyzed by control group. No gross malformation ous tubules in each day of age studelectron microscopy. Tissues were was observed in any of the 3 groups ied. Two-way analysis of variance found significant difference in this fixed by 2.5% glutaraldehyde in studied. Table I summarizes the effects of variable depending on factor "time" 0. 1 M phosphate buffer (pH 7.4-7.6) and postfixed in 1% osmium tetroxide zeranol or DES on body weight and effect (P < 0.001) and "time/ prior to dehydration and embedding in testicular weight and density in each treatment" interaction effect (P < Epon 812 resin. Ultrathin and 1 ti-thick day of age studied. To calculate the 0.05). Regressive changes in tubular semithin sections were cut with ultra- testicular parameters, the average epithelium that affected both germinal microtome (LKB Bromma 2088) value of both gonads for each mouse and Sertoli's cells were observed (LKB, Switzerland). Semithin sec- was used since there was no signifi- from 45 d of age, in zeranol-treated tions were stained with 1% toluidine cant difference between right and left and DES-treated mice. The germinal blue for selection of fields. Ultrathin testes using Student's t-test. The cells appeared degenerated and free in sections were stained with uranyl results obtained by two-way ANOVA the lumen forming a syncytium acetate and lead citrate and were showed significant difference in body whereas Sertoli's cells showed observed with a JEOL JEM 1010 weight depending on factor "time" optically empty vacuoles in their transmission electron microscope effect (Pe< 0.001) and "time/ cytoplasm (Fig. 3). These vacuoles treatment" interaction effect (P < ultrastructurally corresponded to (JEOL, Japan) under 60 kV. 96

dilatations of the smooth endoplasmic reticulum (SER) (Fig. 4a). Likewise, the Sertoli-to-Sertoli cell junctional specializations (complexes containing extensive tight junctions), which are essential to the blood-testis barrier, did not maintain their structural integrity, showing expanded intercellular spaces in the estrogen-treated groups (Fig. 4b) in comparison to the negative control group (Fig. 5). All these alterations, in older animals, were observed in a greater number of tubules of mice treated with estrogens. At 365 d, the testicular parenchyma (about 50% in DES-treated mice; about 30% in zeranol-treated animals and about 10% in negative control mice) showed regressive changes which were namely: seminiferous tubules with reduced spermatogenesis and giant multinucleated cells free in their lumen as well as atrophic seminiferous tubules with thickened basement membranes and almost exclusively made up of Sertoli's cells with many vacuoles, and macrophages filled with cellular debris (Fig. 6). Infiltrates of mononuclear cells (lymphocytes, monocytes and macrophages) were found around the most disordered tubules. Localized accumulations of Leydig's cells around atrophied tubules were present from 45 d of age onwards in both estrogen-treated groups (Fig. 6). On the other hand, sheet-like accumulations of Leydig's cells were observed in 45-day-old mice treated with zeranol and in 3-mo-old mice treated with DES. These cells, morphologically immature, showed a variable nuclear size, with irregular profile and heterochromatin attached to nuclear membrane. Ultrastructurally, the cytoplasm contained an abundant SER whose tubules, concentrically arranged around a non electron-dense area, presented dilatations in the outer region (Fig. 7). From 45 d of age onwards, the DES-treated mice showed rete testis dilatation, which on occasion gave place to cysts covered by a simple flattened epithelium with free epithelial cells, spermatogenic cells and macrophages in its lumen. These alterations were more pronounced from 6 mo of age onwards when an

epithelial papillomatous dysplasia with a loss in cellular polarity was also observed. These cystic and dysplastic changes began to be observed at 6 mo of age in the zeranol-treated mice (Fig. 8), and at 1 y in the negative control animals.

DISCUSSION The possible association between the exposure to estrogenic substances during the prenatal period and the development of urogenital anomalies and infertility in male offspring is a controversial topic (4,15). In the present study, it was confirmed that zeranol or DES treatment at the time of pregnancy does not have any influence on gross testicular parameters, a fact that has previously been pointed out for this strain (10). However, histological study showed morphological alterations. As for the diameter of seminiferous tubules, a difference between both treatments was the significant increase in tubular diameter induced by DES from 180 d of age onwards which was not observed in zeranol-treated group. The reason for this is not known but a possible explanation could be a lower pathogenic activity of the latter compound. The vacuolization observed in the Sertoli's cells of estrogen-treated mice from 45 d of age and of negative control animals from 6 mo of age has been also described in 6 mo-old control mice (16). Ultrastructurally, we confirmed that it deals with dilatations of the smooth endoplasmic reticulum of these cells (17). These morphological alterations of Sertoli's cells could result in functional disorders which could favour the liberation of germinal cells in development, which under normal conditions establish close relationships with Sertoli's cells (18). Likewise, the alteration of the blood-testis barrier resulting from the separation of the junctions of Sertoli-to-Sertoli cells, ultrastructurally observed in this work, could explain the infiltrates which were present around the damaged seminiferous tubules. This anomaly of the barrier could result in the production of antibodies to germinal cells which could be involved in an autoimmune reaction, which has previously been

Figure 6. 365-day-old prenatally zeranoltreated mouse. Degenerative changes in the testicular parenchyma and numerous Leydig's cells around the atrophied tubules which present thickened basement membrane (arrow). Silver methenamine stain. (Bar = 350 pm) '

.4 ^t

.. * i

,.s

a\s.;'.i' 't w m *s^ w S 'g' ' t-. &

'

.

W.S

'ioto.-. I'l

Figure 7. 90-day-old prenatally zeranoltreated mouse. Electron micrograph of a Leydig cell showing folded nucleus (N); concentric pattern of smooth endoplasmic reticulum around a central non electron-dense region (SER). (Bar = 1 ,um)

Figure 8. Rete testis of a 180-day old prenatally zeranol-treated mouse. Dysplasia of its epithelium. H&E. (Bar = 125 ,m)

pointed out in nontreated mice of an advanced age (16). In this study the lesions induced in the germinal cells at 45 d of age by prenatal exposure to zeranol were similar to those observed in the positive control (at 45 d of age) and negative control (at 6 mo of age) groups. These 97

modifications have also been described in mice after prenatal DES (19) and postnatal zeranol (10) or DES (20) exposures and after 6 mo of age in nontreated mice (16). In the literature, these alterations have been explained taking into account that DES and its metabolites are capable of a covalent binding to DNA of germinal cells and modifying it (2). The increase in the population of Leydig's cells induced by the administration of zeranol or DES in this work has been previously cited in other strains of mice exposed to DES (21), but, to the authors' knowledge, there are no references on prenatal exposure to zeranol. The increase in Leydig's cells around the damaged tubules found in estrogen-treated groups has been explained by some authors (22) as a result of the emission of some diffusable agents from the tubules to the interstitium which would induce the proliferation of these cells. The immature features of Leydig's cells observed in both treatments have previously been cited in humans (23) and sheep (24), but when exposures were performed after birth. Nevertheless, these studies do not describe the concentric arrangement adopted by SER, which was observed in this assay. Dilatation and dysplasia of rete testis produced by the administration of zeranol or DES (at 6 mo of age) following our protocol and observed in negative controls (at 1 year of age) have only been described in prenatal DES- and non-treated control mice from 10 mo of age (21). The authors do not know the origin of these lesions in negative controls and they have found no reference in the literature. However, it has been suggested that a possible origin of these lesions in prenatal DES-exposure mice may occur during the embryological development of the rete testis because, in addition to the mesonephric tubules (25), the coelomic epithelium has also been suggested as a precursor for the gonadal rete system (26). This is important because coelomic epithelium also gives rise to the uterus which is a target for DES dysmorphogenesis and carcinogenesis (27). To sum up, the data presented in this report suggest that prenatal expo-

98

sure to zeranol or DES accelerates the development of testicular alterations, which were consistent with usual agerelated changes in negative control mice. On the contrary, Leydig's cells of estrogen-treated groups presented alterations which were not observed in negative control mice at different ages studied. However, no neoplasms could be confirmed. Thus, our results suggest prenatal zeranol exposure might decrease fertility due to a premature aging of testis although further studies should be done to evaluate the functional effects of zeranol exposure in utero on testis.

16.

REFERENCES

17.

1. ARAI Y, MORI T, SUZUKI Y, BERN HA. Long-term effects of perinatal exposure to sex steroids and diethylstilbestrol on the reproductive system of male mammals. Int Rev Cytol 1983; 84: 235-268. 2. GLADEK A, LIEHR JG. Transplacental genotoxicity of diethylstilbestrol. Carcinogenesis 1991; 12: 773-776. 3. BURKE L, ANTONIOLI D, FRIEDMAN EA. Evolution of DES-associated genital tract lesions. Obstet Gynecol 1981; 57: 79-84. 4. DEPUE RH. Maternal and gestational factors affecting the risk of cryptorchidism and inguinal hernia. Int J Epidemiol 1984; 13: 331-335. 5. KAHN-NATHAN J. Les effets de l'exposition in utero au DES chez les filles. Contr Fertil Sex 1989; 17: 243-245. 6. BULLOCK BC, NEWBOLD RR, McLACHLAN JA. Lesions of testis and epididymis associated with prenatal diethylstilbestrol exposure. Environmental HIth Perspect 1988; 77: 29-31. 7. GIMENO EJ. Informe zeranol: Referencias sobre su seguridad y eficacia. Rev Med Vet 1986; 1-31. 8. ISAACSON WK, JONES SJ, KRUEGER

9.

10.

11.

12.

RJ. Testosterone, dihydrotestosterone, trenbolone acetate, and zeranol alter the synthesis of cortisol in bovine adrenocortical cells. J Anim Sci 1993; 71: 1771-1777. O'CALLAGHAN D, QUIRKE JF, BOURKE S, ROCHE JF. The effect of implantation with zeranol or estradiol benzoate alone or with progesterone on growth of steers and growth, carcass weight and testicular weight of bulls. Irish J Agric Res 1988; 27: 13-20. PYLKKANEN L, JAHNUKAINEN K, PARVINEN M, SANTTI R. Testicular toxicity and mutagenicity of steroidal and non-steroidal estrogens in the male mouse. Mutat Res 1991; 147: 79-95. WALKER BE. Animal models of prenatal exposure to diethylstilbestrol. Peri Multigen Carcinogenesis 1989; 25: 349-364. REEL JR, LAWTON AD, WOLKOWSKI-TYL R, DAVIS GW, LAMB IV JC.

13.

14.

15.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

Evaluation of a new reproductive toxicology protocol using diethylstilbestrol (DES) as a positive control compound. J Amer Coll Toxicol 1985; 4: 147-162. CANADIAN COUNCIL ON ANIMAL CARE. Guide to the Care and Use of Experimental Animals. Vol 1. Ottawa: CCAC, 1980. Vol 2. Ottawa: CCAC, 1984. MARTIN ANDRES A, LUNA DEL CASTILLO JD. Andlisis de varianza. In: Martin Andres A, Luna del Castillo JD, eds. Bioestadfstica para las ciencias de la salud. Madrid: Norma S.A., 1990: 383-444. LEARY FJ, RESSEQUIE LJ, KURLAND LT, O'BRIEN PC, EMSLANDER RF, NOLLER KL. Males exposed in utero to DES. J Am Med Ass 1984; 252: 2984-2989. TAKANO H, ABE K. Age-related histologic changes in the adult mouse testis. Arch Histol Jap 1987; 50: 533-544. KIERSZENBAUM AL, MANCINI RE. Structural changes manifested by Sertoli cells during experimental allergic orchitis in guinea pigs. J Reprod Fertil 1973; 33: 119-122. SEGRETAIN D, DECROSSAS B. The Sertoli-spermatid processes in the mouse: New cellular structure. Biol Cell 1991; 71: 321-324. NEWBOLD RR, MCLACHLAN JA. DES associated defects in murine genital tract development. In: McLachlan JA, ed. Estrogens in the Environment. New York: Elsevier/North-Holland Biomedical Press, 1985: 288-318. GODOWICZ B, GASINSKA A, PAWLUS M. Effect of stilboestrol dipropionate on the rate of spermatogenesis in KP and C57 mice and their hybrids. Folia biol (Krakow) 1984; 32: 63-70. NEWBOLD RR, BULLOCK BC, MCLACHLAN JA. Testicular tumors in mice exposed in utero to diethylstilbestrol. J Urol 1987; 138: 1446-1450. AOKI A, FAWCETT DW. Is there a local feedback from the seminiferous tubules affecting activity of the Leydig cell? Biol Reprod 1978; 19: 144-158. SCHULZE C. Response of the human testis to long term estrogen treatment: Morphology of Sertoli cells, Leydig cells and spermatogonial stem cells. Cell Tissue Res 1988; 251: 31-43. FERRERAS MC, MARTINEZ JM, GARCIA MJ, BRAVO AM, PEREZ C, ESCUDERO A. Respuesta del testiculo ovino al tratamiento con DES y zeranol: Estudio morfopatol6gico. Rev Toxicol 1990; 7: 277-299. BYSKOV AG. The anatomy and ultrastructure of the rete system in the fetal mouse ovary. Biol Reprod 1978; 19: 720-735. AREY LB. The genital system. In: LB Arey, ed. Developmental Anatomy. Philadelphia: WB Saunders Company, 1954: 312-339. BARTER JF, AUSTIN JM, SHINGLETON HM. Endometrial adenocarcinoma after in utero diethylstilbestrol exposure. Obstet Gynecol 1986; 67: 84-85.