Genital tract abnormalities in female rats exposed todiethylstilbestrol in utero

[r162

e To~icotog)', Vol. I, No. 3. pp. 193-202

0890-6238/87 $3.00 + .00 Copyright t~) 1988 Pergamon Prc,~s pie

printed in the U.S.A.

G E N I T A L T R A C T A B N O R M A L I T I E S IN F E M A L E R A T S E X P O S E D T O DIETHYLSTILBESTROL IN U T E R O TOVA C. ROTHSCHILD,* ROBERT E. CALHOON, a n d ELIZABETH S. BOYLAN Department of Biology, Queens College and the Graduate School, The City University of New York, Flushing, New York 11367 Abstract n Genital tract morphology in 14-month old female rats exposed prenatally to diethylstilbestrol (DES) was analyzed as part of an examination of the effects of transplacental exposure to DES on estrogen sensitive tissues. Pregnant Sprague-Da~iey rats ~ere injected ~ith sesame oil alone or ~ith DES in sesame oil on days 10 and 13 of gestation (total dose 1.2 /xg DES) or on da)s 15 and 18 (total dose !.2 /Jg or 120 p.g DES). Female offspring (9-15 per group) ~ere sacrificed at 14 months of age. Effects of DES exposure varied ~ith the dose given and ~ith the stage of differentiation of the fetal tissues. In the ovaries of rats exposed to 120 p.g of DES on days 15 and 18 of gestation, follicular elements ~ere reduced and replaced by dense sheets of stromal cells; oophoritis ~as noted in five of nine rats. Hypercellularity of oviductal stroma ,~as another common feature, as was suppurative salpingitis. Ovaries of rats exposed to 1.2 p-g DES on days 10 and 13 of gestation were more likely to contain numerous corpora lutea than the other DES-exposed groups or controls. An increased incidence of benign uterine abnormalities ~as observed in DES-exposed offspring, including squamous melaplasia and suppurative endometritis. In the cervices of all nine rats exposed to 120 p-g DES on days 15 and 18 of gestation, the epithelial surface showed a convoluted pattern, lined by stratified squamous and stratified cuboidal cells. Thus, prenatal exposure to DES, especially at the higher dose used, has long-term consequences on reproductive tract morphology in Sprague-Dawley rats. Key Words: Diethylstilbestrol, Genital tract, Prenatal, Sprague-Dawley rat, Female morphology.

INTRODUCTION

/_tg) into day 19 rat fetuses resulted in a dose-related incidence of cleft phallus, hypospadias, and incomplete oviductal coiling (6,7). Similar urogenital malformations were obtained in rat fetuses following maternal DES injection (7). However, the natural estrogen was found to be approximately 100fold less potent than DES. Studies from this laboratory have shown that prenatal exposure to low doses of DES (1.2-120/.tg per dam) leads to precocious development of the nipples in female neonates, reduced fertility, and macroscopic abnormalities of the reproductive organs in young adult Sprague-Dawley rats (8). Using rats of the ACI strain (AXC9935), we demonstrated that prenatal exposure to low doses of DES (equivalent to 1.2 and 12 ~g DES) resulted in an increased frequency of atypical uterine epithelia, cystically dilated uterine glands, and a thickened vaginal epithelium in the female offspring at 10 months of age (9). In mice, exposure to DES in utero has been correlated with a range of morphological and functional impairments including vaginal adenosis and decreased fertility; persistent vaginal cornification and absence of corpora lutea; and tumors of the ovary, uterus, vagina, and lung (10-16). In general, the incidences and severity of the anomalies in mice appeared to be correlated with time of exposure,

Development of the mammalian reproductive tract can be influenced dramatically by changes in the hormonal milieu of the mother (1). The discovery of an association between ingestion of the synthetic estrogen diethylstilbestrol (DES) by pregnant women and the appearance of vaginal and cervical adenocarcinoma in a small percentage of their teenage daughters (2) has stimulated interest in the long-term consequences of exposure to estrogenic substances during critical periods of development. The teratogenic potential of DES has been demonstrated by the finding of significant increases in the incidence of cervical ectropion and vaginal adenosis in DES-exposed young women (3). In the rat, exposure to very high doses (mg quantities) of DES or estradiol resulted in persistent vaginal cornification, urogenital malformations, gonadal hypogenesis, and tumors of the ovaries, pituitary, and endometrium in the female offspring (4,5). Direct injection of DES or estradiol (0.1-100 *Submitted in partial fulfillment of the requirements for the Ph.D. of The City University of New York. Present address: Department of Biological Sciences, Rutgers University, Newark, NJ 07102. Address correspondence to: Elizabeth S. Boylan. Rccelved 13 October 1987;Accepted 19 December 1987. 193

194

Reproductive Toxicology

dosages employed, and strain under examination. Uterine adenocarcinomas and ovarian cystadenocarcinomas have also been observed in "DESlineage mice"; that is, in second generation female offspring from DES-treated mice (17). Most of the reports describing the effects of prenatal exposure to DES have focused on young animals, generally less than one-year old. McLachlan et al. (18) presented necropsy data on mice exposed to 1 to 100 /.tg/kg DES that were sacrificed between 12 and 18 months of age. Here, in addition to finding the excessive vaginal keratinization (present also at younger ages), hyperplasia of the cervical stroma was observed and a low incidence of benign and malignant tumors of the cervix and uterus was found. Walker (19) described uterine and cervical adenocarcinomas in mice exposed prenatally to 1 ~g/g DES and sacrificed between 15 and 31 months of age. The present report describes the effects of transplacental exposure to DES on reproductive tract morphology of 14-month old Sprague-Dawley rats. Several pathological conditions were found to be associated with DES exposure and were generally more severe when rats were exposed to the higher dose of DES (120 /zg) given during later stages of gestation. MATERIALS AND METHODS Sprague-Dawley derived rats (CD strain) were obtained from Charles River Breeding Laboratories (Wilmington, MA 01887) or were bred from CD stock in our animal quarters. Virgin female rats (3-5 months old) were mated to males of proven fertility. Presence of sperm in the vaginal smear was used to designate day 0 of pregnancy. Pregnant rats were housed individually in a temperature-controlled room with a 12/12 h light-dark cycle and were provided with Purina Rat Chow (Ralston-Purina Co., St. Louis, MO 63188) and tap water ad libitum. DES (Sigma, St. Louis, MO 63178) was injected subcutaneously in 0.3 rnl sesame oil either on days 10 and 13 of gestation (equal week 2 of gestation; total dose 1.2 ttg DES) or on days 15 and 18 (equal week 3 of gestation; total dose 1.2 p.g DES, or 120 #g DES). These doses are equivalent to 4 and 400 /zg/kg body weight, respectively. Control animals received 2 • 0.3 ml of vehicle only. The animals were allowed to deliver naturally and raise their pups to weaning. If a dam had not delivered on the expected date of parturition (day 22), the animal was sacrificed and examined for retained fetuses; stillbirths and neonatal deaths were recorded.

Volume1, Number 3, 1987/88 Each experimental group consisted of 15 females except for the high dose group in which there were nine rats. The offspring were housed five to a cage. At 14 months of age, the animals were sacrificed using a rodent guillotine. The entire reproductive tract was removed, fixed in 10% neutral buffered formalin, and divided as follows: one uterine horn and the vagina were transected to obtain a representative sample of each, 2 to 3 mm thick; the cervix was cut in half longitudinally. All tissues were dehydrated through tetrahydrofuran, embedded in paraffin, sectioned at 10 /.tm and stained with hematoxylin and eosin (H&E). The histologic evaluation was performed blind. Photomicrographs were taken on a Nikon Optiphot microscope. Observations on qualitative variables of reproductive tract tissues were expressed as contingency tables and were analyzed for equality of response using the log-likelihood test, which is distributed as chi-square. Where the number of responses was too few to assume approximation to chi-square, an exact test was devised based upon hypergeometric probability distribution. Probability was calculated for all possible sets of responses that summed to the observed total of positive responses and that deviated from equality as much as or more than the observed responses. The sum of these probabilities was used as a test for equality of response. Within each of the four exposure groups, all pair-wise comparisons of qualitative variables were analyzed in 2 x 2 contingency tables as tests of independence. This was a search for correlated response of these variables to prenatal exposure. Fisher's Exact Test was the test statistic. RESULTS Effect o f D E S treatment on p r e g n a n c y o u t c o m e

Table 1 shows the impact of DES treatment during week 2 or 3 of gestation on the reproductive performance of female Sprague-Dawley rats. The number of dams producing litters that survived to weaning was significantly reduced in the group treated with 120 ttg DES during week 3 of gestation. This diminished reproductive capacity was accounted for by substantial numbers of pregnant animals that failed to give birth (retention of fetuses). Some litters were born live but had 100% mortality in the neonatal period. There were no apparent differences in pregnancy outcome between the groups treated with 1.2 ttg DES (week 2 or 3 of gestation) and vehicle-treated controls. Unaffected by any of the three DES treatments were length of gestation (range 21.0-21.2 days) and number of pups per dam

DES in utero 9 T. C. ROTIISCIIILD ET AL.

195

Table 1. Impaired p r e g n a n c y o u t c o m e following D E S t r e a t m e n t

No. pregnant rats injected

Treatment a Vehicle 1.2/.tg DES (days 10 and 13) 1.2/ag DES (days 15 and 18) 120/.~g DES (days 15 and 18)

Live-born litters t~

Litters with neonatal deaths ~

Retention of fetuses d

Length of gestation (days)

16 16

14 (88%) 16 (100%)

0 0

2 0

21.2 21.0

15

12 (80%)

1

2

21.0

24

I0 (41.7%) e

2

12

20.8

~Dams injected with vehicle, 1.2 ~ug or 120/~g DES during second or third weeks of gestation. bNumber of rats producing at least one offspring that survived to weaning. CNumber of rats whose entire litter died within one week of parturition. dNumber of rats that did not deliver live offspring. At necropsy, retained fetuses were found in uteri. ~Significantly different from vehicle-exposed controls (p <~ 0.01).

(range 8.8-10.5 pups). There was a slight prepond e r a n c e o f m a l e o f f s p r i n g in all f o u r e x p e r i m e n t a l g r o u p s ( v e h i c l e : 4 6 . 8 % f e m a l e s ; 1.2/.tg D E S w e e k 2: 4 8 . 1 % f e m a l e s ; 1.2 # g D E S w e e k 3: 4 8 . 2 % f e m a l e s ; 120/.tg D E S w e e k 3: 3 7 % f e m a l e s ) .

O b s e r v a t i o n s at n e c r o p s y B o d y w e i g h t s at s a c r i f i c e d i f f e r e d s i g n i f i c a n t l y a m o n g t h e g r o u p s o f 1 4 - m o n t h o l d r a t s ( T a b l e 2),

w i t h r a t s e x p o s e d to e i t h e r d o s e o f D E S d u r i n g week 3 of gestation weighing an average of approxi m a t e l y 60 g m o r e t h a n c o n t r o l s o r r a t s e x p o s e d t o D E S d u r i n g w e e k 2 o f g e s t a t i o n . In t h e h i g h d o s e g r o u p (120 /.tg D E S in w e e k 3 o f g e s t a t i o n ) , t h e mean wet weight of the ovaries was increased while that of the uterus was reduced compared to groups e x p o s e d to 1 . 2 / . t g D E S o r v e h i c l e o n l y ( T a b l e 2). Furthermore, the high-dose group displayed a high

Table 2. Data on body weights and reproductive organs of 14-month old rats Microscopic abnormalities of uteri Organ Weights (g) Gross abnormalities of reproductive organs

Squamous metaplasia n (%)

Prenatal exposure

Body weight (g)

Uterus ~

Ovaries h

Pyometra n (%)

Vehicle n -- 15 i.2 #g DES days 10 and 13 n = 15 1.2/.tg DES days 15 and 18 n = 15 120/.tg DES days 15 and 18 n = 9

333.6 __ 54.7

0.91 • 0.27

0.13 • 0.04

None

7(46) c

340.7 • 39.0

0.99 • 0.23

0.11 • 0.02

None

11(73) r

2(13) 't

398.3 • 80.7"

0.96 • 0.19

0.12 • 0.02

12(80)c.d

3(20) d

400.1 • 70.9 L"

0.60 • 0.19 r

0.39 • 0.3U

one uterine horn above cervix, filled with necrotic material seven enlarged ovaries (4 bilateral, 3 unilateral) five enlarged loosely coiled oviducts (3 bilateral, 2 unilateral) one uterus distended with yellow, milky. fluid one uterus with several masses

9(100)d.g

4(44) d

0(0)

aWet weights of both uterine horns anterior to cervix. bIncludes wet weights of both ovaries and oviducts. cSquamous metaplasia lined endometrial glands ofily. dDES groups collectively are significantly different from vehicle-exposed control (p ~< 0.05). eSignificantly different from groups exposed in utero to vehicle or to 1.2/.tg DES days l0 and 13 (p ~< 0.01). tSignificantly different from vehicle-exposed controls and.from groups exposed prenatally to 1.2 p.g DES (p ~< 0.01). ~Squamous metaplasia lined entire uterine cavity.

196

Reproductive Toxicology

.

.

.

.

.

.

.

.

Volume 1, Number 3, 1987/88

~ q

Fig. 1 a. Cross section of ovary and oviduct from a 14month old control rat. Note follicles in various stages of development or atresia and oviducts with minimal connective tissue surrounding the layer of smooth muscle. H & E stain. Bar = 232/zm. b. Cross section of ovary and oviduct from a rat exposed to 120 ~g DES in week 3 of gestation. Note paucity of ovarian follicles, the absence of corpora lutea, and the dense sheets of stromal cells. Also visible is the prominent increase in area occupied by oviductal connective tissue. H & E stain. Bar = 232/tm. e. Cross section of ovary and oviduct from rat exposed to 120 #g DES in week 3 of gestation. Note extensive dilation of oviducts and accumulation of neutrophils (*) with the epithelium of some parts of the oviducts completely destroyed by infection (arrows). Ov = oviduct. H & E stain. Bar = 270/.tin.

frequency of macroscopic abnormalities of the reproductive tract as itemized in Table 2. Particularly prominent were instances where the ovaries were enlarged up to four times normal size and the oviducts were large and loosely coiled, appearing draped over the ovary rather than tightly coiled as usual. Evidence of an inflammatory exudate distending the uterine lumen was also present in two DES-exposed rats.

Ovarian and oviductal pathology Median samples of ovarian tissue from 54 ovaries representing one ovary from each rat under study were selected for histological analysis; 12 to 36 serial sections were analyzed from each sample. Ovaries from 14-month old control rats and rats exposed to the low dose of DES (week 2 or week 3 of gestation) contained numerous cystic and atretic follicles and a few growing follicles (Figure la). Corpora lutea were particularly evident in the group exposed to 1.2/xg DES in week 2 of gestation (Figure 2). There was no evidence of inflammati(n of the ovarian tissue in rats from these three groups. In contrast, in several rats exposed to 120 /~g DES

during week 3 of gestation, the ovaries were characterized by extensive stromal hypercellularity, oophoritis, and hemosiderin deposition (Figures lb and lc). A few empty follicles lacking any signs of ova development were also seen in some animals. Polyovular follicles or polynuclear oocytes were not apparent in any rats in this study. Concomitant with these ovarian changes, oviducts from rats exposed to the high dose of DES showed extensive areas of stromal hypercellularity (Figure lb). A higher magnification of this feature is shown in Figure 3, in which the area occupied by oviductal connective tissue appears markedly increased. Another oviductal anomaly associated with DES exposure was suppurative salpingitis; this was seen in 4/9 rats exposed to the high dose of DES during week 3 of gestation (Figure lc). Two o f these rats also had diffuse chronic endometritis. In some rats the oviducts were completely occluded by cellular debris and polymorphonuclear leucocytes (Figure lc). In one rat the oviductal lumen, which is normally lined by low columnar epithelium (Figure 4a), was compressed by disorganized stromat cells (Figure 4b).

DES in utero 9 T. C. ROTHSCHILDET AL.

197

Fig. 2. Area of ovarian tissue from rat exposed to 1.2/.tg DES in week 2 of gestation. Note a few growing follicles and predominance of healthy-looking corpora lutea. H & E stain. Bar = 131 ttm,

Fig. 3. Higher magnification of lb. Note extensive hypercellularity of stromal elements and hemosiderin deposition (arrows). Ov = oviduct. H & E stain. Bar = 131/.tm.

Uterhze pathology T h e histology of uterine cross-sections was analyzed by examination of 15 to 36 serial sections obtained from a representative portion o f one uterine horn per animal; tissue from each rat in the four experimental groups was examined. Squamous metaplasia o f the uterine epithelium was noted in all groups o f rats in this study, but its prevalence differed significantly among the groups (Table 2 ) . Compared to the vehicle-exposed group (Figure 5a), the incidence of squamous metaplasia in the groups exposed to DES was increased 27 to 54%. In both groups exposed to 1.2 # g DES (week 2 or 3 o f gestation), foci o f squamous cells replaced the columnar epithelium o f the endometrial glands

Fig. 4 a. Cross section of oviduct from a vehicle-exposed control rat. Note uniformity of cuboidal-low columnar cells lining luminal epithelium, and the sparse connective tissue surrounding the tunica muscularis. H & E stain. Bar = 63 #m. b. Cross section of oviduct from a rat exposed to 120/.tg DES in week 3 of gestation. Note virtually complete compression of oviductal lumen (arrows) inside the muscle layer and increased stromal tissue extending beyond the sheath of smooth muscle. H & E stain. Bar = 63 # m .

only. In contrast, e x p o s u r e to the higher dose o f DES in utero resulted in complete or partial replacement o f both luminal and glandular epithelium with approximately 4 to 6 layers of stratified squamous epithelium (Figure 5b). The metaplastic transformation o f t h e glandular epithelium was often accompanied by cystic dilation o f the uterine glands (Figure 5c). In addition to a higher incidence o f squamous metaplasia o f the uterine epithelium in the DES-exposed animals, the higher dose group had an unusual case of a hemorrhagic endometrial polyp that projected into and almost obliterated the uterine lumen (Figure 5d); the uterine horn was grossly enlarged at this point. The four prenatal exposure groups also differed

198

Reproductive Toxicology

Volume 1, Number 3, 1987/88

Fig. 5 a. Cross section of uterine horn from a vehicleexposed control rat. Note luminal epithelium of uniform thickness. Endometrial glands are not dilated. H & E stain. Bar = 312 /.tin. b. Cross section of uterine horn from a rat exposed to 120/.tg DES in week 3 of gestation. Note replacement of both luminal and glandular epithelium with thick stratified squamous epithelium. Glands are dilated with cellular debris. H & E stain. Bar = 312 txm. e. Small area of uterine tissue from a rat exposed to 120 tzg DES in week 3 of gestation showing stratified squamous epithelium of lumen (lu) and glands (gl). Note cystic dilation of uterine glands and accumulation of polymorphonuclear leucocytes (arrows) and debris in lumina. H & E stain. Bar = 16/.tm.d. Section of hemorrhagic polyp from rat exposed to 120/.tg DES in week 3 of gestation. Uterine lumen (lu) is displaced by polyp containing large cystic glands. Note magnification of uterine horn is same as in Figures 5a,b. H & E stain. Bar = 312/zm.

significantly with respect to the incidence o f suppurative endometritis; that is, p y o m e t r a (Table 2). Significantly m o r e cases o f suppurative endometritis (Figure 5c) were found in the group e x p o s e d to 120 # g DES in w e e k 3 o f gestation (19 ~< 0.05). A l o w e r incidence o f this severe f o r m o f endometritis was found in b o t h groups e x p o s e d to 1.2 # g D E S , and no cases w e r e found a m o n g the vehicle-

e x p o s e d controls. M o s t animals in b o t h the control and the e x p e r i m e n t a l groups displayed a low grade uterine i n f l a m m a t o r y response, characterized by mild to m a r k e d l y m p h o c y t i c infiltrate in the s t r o m a and an a c c u m u l a t i o n of basophils in the uterine glands and lumen. A p a r t from this i n f l a m m a t o r y response, no morphological alterations were a p p a r e n t in the uterine s t r o m a o f the D E S - e x p o s e d rats.

DES in utero s T. C. ROTIISCIIILDET AL.

199

Fig. 6 a. Longitudinal section through cervix of rat exposed prenatally to 120/zg DES in week 3 of gestation. Note extreme convolution of cervical epithelium creating cystically dilated crypts (arrows). H & E stain. Bar = 306/.tm. b. Longitudinal section of cervix from vehicle-exposed control rat. Note absence of crypts or deep convolutions (cf. Fig. 6a). H & E stain. Bar = 306/xm.

Cervical and vaghtal pathology In the cervix, 12 to 14 median longitudinal serial sections, 10 /zm each, were analyzed for the presence of adenosis and other abnormalities. F o r a slide o f the cervix to be included in the analysis, the squamo-columnar junction in both cervical canals must have been present; therefore the analysis was based on 41 of the possible 54 offspring. Intrauterine exposure to 120/.tg D E S in week 3 o f gestation had a p r o n o u n c e d effect, as the cervical epithelium of all rats in this group showed a convoluted pattern, with numerous crypts and nests lined by stratified squamous and stratified cuboidal cells

(Figure 6a). S o m e of these structures also showed cystic dilation. F o r comparison purposes, a cervix from a typical control rat is shown in Figure 6b. The cervix of one control rat was categorized as having a moderate pattern o f convoluted cervical epithelium. Extensive areas of submucosal lymphocytic infiltration were observed in two rats exposed to 1.2/.tg DES in week 2 of gestation and in one rat from each o f the two groups exposed to DES in week 3 of gestation. The cervical epithelium o f one rat in the group exposed to 120 p.g D E S in week 3 o f gestation was characterized by intense vacuolation (Figure 7a).

Fig. 7 a. Longitudinal section through cervix of rat exposed to 120 /.tg DES in week 3 of gestation. Note prominent vacuolation of cervical epithelium. H & E stain. Bar = 384 tim. b. Higher magnification of a. Note intraepithelial vacuolations occupied by pycnotic cells and cell debris. H & E stain. Bar = 35/~m. c. Small area of cervical epithelium showing vacuolations and lightly staining cells distended with dense amorphous material (arrows). H & E stain. Bar = 17 ~m.

200

Reproductive Toxicology

The intra-epithelial spaces contained pycnotic cells and cell debris (Figure 7b). When examined under higher magnification, some cells appeared distended with a dense amorphous material (Figure 7c). Histology of the lower vagina was analyzed by examination of 14 to 27 serial cross-sections obtained from a representative sample of tissue; specimens of vaginal tissue from all 54 rats in this study were examined. Vaginas from all of the rats in both the experimental and control groups appeared normal and there was no evidence of glandular tissue associated with the vaginal epithelium. From inspection of the slides, it was clear that animals exposed to DES in utero were found in all stages of the estrous cycle. DISCUSSION The results of the present investigation confirm our previous findings (8,9) that treatment of dams with relatively low doses of DES during week 3 of gestation significantly impairs pregnancy outcome. Interestingly, prenatal administration of the higher dose of DES increased perinatal mortality in spite of the absence of any effect on gestation length. Rands et al. (20) have reported an inverse correlation between gestation length and perinatal mortality in Sprague-Dawley rats that were dosed orally by gavage from gestation day 6 through day 18 with either 10 /xg/kg/day DES or 45 ~g/kg/day DES. Thus, while the efficacy of DES treatment in preventing miscarriages in pregnant women, at best, remains controversial (21), in rats DES appears to interfere with term pregnancy and survival of offspring. In the current experiment, transplacental exposure to DES in Sprague-Dawley rats resulted in an array of reproductive tract abnormalities in the female offspring at 14 months of age. The severity of the effects noted varied with the time of exposure and with the dose of the synthetic estrogen given during pregnancy. We have shown that prenatal exposure to 1.2 p.g DES was not associated with a permanent estrus condition in 6 to 7-week old Sprague-Dawley rats (22) or in ACI rats examined at 12 weeks of age (9). Even Sprague-Dawley rats exposed to 1.2/.tg DES prenatally and treated postnatally with 7,12-dimethylbenz(a)anthracene (DMBA) were found to be comparable to controls in the percent Of animals in proestrus-estrus vs metestrus-diestrus at nine months of age (22). And, DES-exposed ACI rats examined at 10 months of age were found in all stages of the estrous cycle (9). While fertility and the ability of dams to raise offspring to weaning was compromised in rats exposed to 120/.tg DES in week 3 of gestation, the lower

Volume 1, Number 3, 1987/88

dose (1.2/~g DES) had no apparent effect on fertility and pregnancy outcome in 3 to 5-month old SpragueDawley rats that had been exposed to DES in week 2 or 3 of gestation (8). Thus, some aspects of reproductive tract structure and function were not disturbed in either Sprague-Dawley or ACI rats up to 9 to 10 months of age following prenatal exposure to 1.2/.tg DES. In the present study, DES teratogenicity was seen as extensive hyperplasia of stromal elements in the ovaries and oviducts of rats exposed to 120/zg DES in week 3 of gestation. Other prominent alterations in ovarian and oviductal morphology have been reported in 4-week to 10-month old CD-I mice exposed to 100 #g/kg DES on days 9 to 16 of gestation (12) and in 7-month old Swiss mice exposed to a single low dose (I0 /~g/kg) of DES on day 15 of gestation (16). The current findings are particularly noteworthy in light of the report by Cunha and Fujii (23) that suggests that the genital stroma may be the principal site of DES action in the developing fetus. Furthermore, microscopic examination revealed that most of the ovaries from 14-month old offspring contained numerous empty follicles but generally lacked young, enlarging follicles. As expected, most ovaries of these old rats had no evidence of recent ovulations, except for the presence of significant numbers of " n e w " corpora lutea in 5 of 15 females exposed to 1.2/xg DES in week 2 of gestation. This may indicate that DES exposure during week 2 of gestation delayed cessation of reproductive cycling in certain animals. In the uterus, squamous metaplasia restricted to the uterine glands was found in 46% of the vehicle-exposed rats at 14 months of age. Squamous metaplasia of the uterine epithelium has been frequently seen in adult rats following prolonged estrogen administration (24) and in aged mice exposed to DES in utero (17,18,25). The appearance of the squamous metaplasia in almost half of the vehicleexposed controls in this study suggests that the replacement of glandular epithelum with squamous cell layers in an age-related process; it may be potentiated by either additional estrogenic stimulation postnatally (9) or, as shown here, by exposure to the synthetic estrogen during a critical period of differentiation in utero. Suppurative endometritis was seen only in DES-exposed offspring in this study. This condition has also been observed in 5-month old SpragueDawley rats implanted with a 5 mg DES pellet at puberty (26), but not in 10-month old DES-exposed ACI rats or in 10-month old ACI rats implanted with a 2.5 mg DES pellet at 12 weeks of age (9).

DES in utero 9 T. C. ROTHSCHILD ET AL.

Leucocytic infiltration of uterine glands and stroma was observed in 3-month old Sprague-Dawley rats that had been exposed to clomiphene citrate in utero (27). Thus, the Sprague-Dawley rat appears to have a special sensitivity to estrogen in terms of its susceptibility to reproductive tract infection and inflammation, compromising its usefulness as a model to study long-term carcinogenic effects associated with DES exposure in utero. One uterine polyp was observed in a rat exposed to 120 #g DES in week 3 of gestation. While there have been no other reports of uterine tumors in rats exposed prenatally to such a relatively low dose of DES, Napalkov and Anisimov (4) described three endometrial polyps and one uterine fibroma among 18 rats exposed to a high dose (1 mg/kg) of DES on day 19 of gestation. A total of 19 uterine and cervical adenocarcinomas were obtained in 143 CD-I mice exposed to a single injection of 1 ~g/g DES on gestation days 16V2, 17, or 17V, and necropsied at ages 15 to 31 months (18). Also, one uterine adenocarcinoma was seen among 20 female offspring of CD-1 mice that had received injections of 100 ~g DES/kg between days 9 to 16 of gestation and two leiomyomata and one stromal cell sarcoma of the uterus were found among 74 other mice exposed to lower doses of DES (2.5-10/.tg/kg/day) (18). In the current experiment, the cervical epithelium had a markedly convoluted appearance in rats exposed to 120 p.g DES in week 3 of gestation. A similar morphologie abnormality was also seen in ACI rats implanted with a DES pellet postnatally (9). Bern et al. (28) have described "epithelial pegs" and "downgrowths" in the cervices of aged mice treated neonatally with low doses of either DES alone or estradiol plus prolactin. These epithelial alterations appeared to progress to cervicovaginal squamous cell tumors when neonatal treatment was prolonged or consisted of higher estrogen doses. Miller et al. (29) have reported cervicovaginal adenocarcinomas and adenosquamocarcinomas in Wistar rats exposed prenatally to relatively high doses of DES (0.1-50 mg/kg DES) on days 18, 19, and/or 20 of gestation and necropsied after four months of age. When viewed together these experiments suggest a possible progression from the benign cervicovaginal alterations such as those reported here to squamous cell tumors; this may require either (a) a higher dosage of the synthetic estrogen, (b) an additional stimulus, and/or (c) a longer latency to promote tumor formatioh. Data from the experiment reported here provide no evidence that prenatal exposure to DES resulted in classical adenosis in the cervix or vagina of

201

14-month old DES-exposed Sprague-Dawley rats. While the absence of uterine-type glandular elements in these old DES-exposed rats could be explained by the replacement of tissue through metaplastic healing processes similar to those described in humans (30), other data we have obtained does not support this supposition. Adenosis was never observed in cross sections of the vagina or in median longitudinal sections through the cervix in 19 DES-exposed Sprague-Dawley rats at two months of age or in 28 DES-exposed rats of the same strain at nine months of age; both groups had been exposed to 1.2/.tg DES in week 3 of gestation and the latter had been treated postnatally with DMBA (22). Similarly, uterine-type glands in the vagina were not evident in 78 10-month old ACI rats that had been exposed to DES prenatally and implanted with a cholesterol pellet at 12 weeks of age (9). The incidence of vaginal adenosis of neonatally-treated mice varies considerably depending upon the strain of mouse, the dose and time of treatment with DES, and on the age of the animal when the tissues are examined ( 14,18,25,31,32). The present study together with those referred to above suggest that cervicovaginal adenosis resuiting from DES exposure is not a common abnormality in rats and some strains of mice. The unusual morphology of the oviducts and ovaries Of these DES-exposed rats points to an additional example of species-specific sensitivity of reproductive tract tissues to the effects of this teratogen.

Acknowledgements- The authors gratefully acknowledge the contributions of Ms. J. Doody to all aspects of this project; the cooperation of Dr. E. H. Fowler, pathology consultant, is recognized with appreciatior~. We thank Dr. ]. Szalay for her critical reading of the manuscript. Supported by USPHS grants CA-18458 to ESB and RR07064 to Queens College, and by a grant from the PSC-CUNY Award Program, The City University of New York. Computer processing of the data was supported by grants from the CUNY Computer Center.

REFERENCES I. Ratzon SK, Weldon VV. Exposure to endogenous and exogenous hormones during pregnancy. In: Nitzan M, ed. The

h~fluence o f maternal hormones on the fetus and newborn. New York: Karger; 1979:174-190. 2. Herbst A, Ulfelder H, Poskanzer D. Adenocarcinoma of the vagina. Association of maternal stilbestrol therapy with t u m o r appearance in young women. New EnglJ Med 1971; 284:878-888. 3. Seully RE, Welch WR. Pathology of the female genital tract after prenatal exposure to diethylstilbestrol. In: Herbst AL, Bern HA, eds. Developmental effects of diethylstilbestrol (DES) in pregnancy. New York: Thieme-Stratton; 1981:26-46.

202

Reproductive Toxicology

4. Napalkov Y, Anisimov VN. Transplacental effect of diethylstilbestrol in female rats. Cancer Lett 1979; 6:107-114. 5. Vannier B, Raynaud JP. Long-term effects of prenatal estrogen treatment on genital tract morphology and reproductive function in the rat. J Reprod Fertil 1980; 59:43--49. 6. Henry EC, Miller RK, Baggs RB. Direct fetal injections of diethylstilbestrol and 17fl-estradiol: a method for investigating their teratogenicity. Teratology 1984; 29:297-304. 7. Baggs RB, Miller RK, Garman R, McKenzie RC. Teratogenic and neoplastic lesions of the genitourinary system in Wistar rats exposed in utero to diethylstilbestrol (DES). Teratology 1980; 21:26A. 8. Boylan ES. Morphological and functional consequences of prenatal exposure to diethylstilbestrol in the rat. Biol Reprod 1978; 19:854-863. 9. Rothschild TC, Calhoon RE, Boylan ES. Effects of diethylstilbestrol exposure in utero on the genital tracts of female ACI rats. Exper Mol Pathol 1988; 48:59-76. 10. Lamb JC, Newbold R, McLachlan J. Visualization by light and scanning electron microscopy of reproductive tract lesions in female mice treated transplacentally with diethylstilbestrol. Cancer Res 1981; 41:4057-4062. 11. McLachlan JA, Shah HC, Newbold RR, Bullock BC. Effect of prenatal exposure of mice to diethylstilbestrol on reproductive tract function in the offspring. Toxicol Appl Pharmacol 1975; 33:190. 12. Newbold RR, Bullock BC, McLachlan JA. Exposure to diethylstilbestrol during pregnancy permanently alters the ovary and oviduct. Biol Reprod 1983; 28:735--744. 13. Nomura T, Masuda M. Carcinogenic and teratogenic activities of diethylstilbestrol in mice. Life Sciences 1980; 26:1955-1962. 14. Nomura T, Kanzaki T. Induction of urogenital anomalies and some tumors in the progeny of mice receiving diethylstilbestrol during pregnancy. Cancer Res 1977; 37:10991104. 15. Walker BE. Reproductive tract anomalies in mice after prenatal exposure to DES. Teratology 1980; 21:313-321. 16. Wordinger RJ, Highman B. Histology and ultrastructure of the adult mouse ovary following a single prenatal exposure to diethylstilbestrol. Virchows Arch Cell Pathol 1984; 45:241-253. 17. Walker BE. Tumors of female offspring of mice exposed prenatally to diethylstilbestrol. J Natl Cancer lnst 1984; 73:133-137. 18. McLachlan JA, Newbold RR, Bullock BC. Long-term effects on the female mouse genital tract associated with prenatal exposure to diethylstilbestrol. Cancer Res 1980; 40:3988-3999. 19. Walker BE. Uterine tumors in old female mice exposed pre-

Volume 1, Number 3, 1987/88 natally to diethylstilbestrol. J Natl Cancer hzst 1983; 70:477--484. 20. Rands P.L., White RD, Carter MW, Allen SD, Bradshaw WS. Indicators of developmental toxicity following prenatal administration of hormonally active compounds in the rat. I. Gestational Length. Teratology 1982; 25:37-43. 21. Dieckmann WE, David ME, Rynkiewicz SM, Pottinger RE. Does the administration of diethylstilbestrol during pregnancy have therapeutic value? Am J Obstet Gynecol 1953; 66:1062-1068. 22. Boylan ES, Calhoon RE, Vonderhaar BK. Transplacental action of diethylstilbestrol on reproductive and endocrine organs, mammary glands, and serum hormone levels in two- and nine-month old female rats. Cancer Res 1983; 43:4872-4878. 23. Cunha GR, Fujii H. Stromal-parenchymal interactions in normal and abnormal development of the genital tract. In: Herbst AL, Bern HA, eds. Developnzental effects ofdiethylstilbestrol (DES) in pregnancy. New York: ThiemeStratton, 1981:179-193. 24. Gitlin G. On the mode of development of stratified squamous epithelium in the rat's uterus following prolonged estrogen administration. Anat Rec 1954; 120:637-655. 25. Newbold RR, McLachlan JA. Vaginal adenosis and adenocarcinomas in mice exposed prenatally or neonatally to diethylstilbestrol. Cancer Res 1982; 42:2003-2011. 26. Shellaberger C J, Stone JP, Holtzman S. Rat differences in mammary tumor induction with estrogen and neutron radiation. J Natl Cancer hzst 1978; 61:1505-1508. 27. McCormack S, Clark JH. Clomid administration to pregnant rats causes abnormalities of the reproductive tract in offspring and mothers. Science 1979; 204: 629-631. 28. Bern H, Jones L, Mills K, Kohrman A, Mori T. Use of the neonatal mouse in studying long-term effects of early exposure to hormones and other agents. J Toxicol Environ Health (Suppl.) 1976; 1:103-116. 29. Miller RK, Baggs RB, Odoroff CL, McKenzie RC. Prenatal exposure to diethylstilbestrol (DES): vaginal tumors and malformations in the rat. Pharmacologist 1982; 24:236-241. 30. Fu YS, Reagan JW, Richart RM. Cytologic diagnosis of diethylstilbestrol-related genital tract changes and evaluation of squamous cell neoplasia. In: Herbst AL, Bern HA, eds. Developmental effects of diethystilbestrol (DES) Dr pregnancy. New York: Thieme-Stratton; 1981:46-62. 31. Plapinger L, Bern H. Adenosine-like lesions and other cervicovaginal abnormalities in mice treated perinatally with estrogen. J Natl Cancer lnst 1979; 63:507-515. 32. Forsberg J.-G, Kalland T. Neonatal estrogen treatment and epithelial abnormalities in the cervicovaginal epithelium of adult mice. Cancer Res 1981; 41:721-734.