* Reproductive Toxicology Division, Endocrinology Branch, MD 72, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, 86 Alexander Drive, Research Triangle Park, North Carolina 27711;
Department of Environmental and Molecular Toxicology, North Carolina State University, Raleigh, North Carolina 27695; and
USEPA/NCSU Cooperative Training Agreement, Raleigh, North Carolina 27695
Received June 20, 2001; accepted September 25, 2001
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: testosterone propionate; androgens; masculinization; prenatal exposure; sexual differentiation; anogenital distance; agenesis of the lower vagina; Sprague-Dawley rats.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In addition to their effects on sexual development, prenatal exposure to androgens can result in other adverse effects. High levels of T or other androgens administered to the dam can induce infanticide in rats (Rosenberg and Sherman, 1974) and mice (Mann and Svare, 1983
) and can produce negative effects on the offspring such as low body weight (Fritz et al., 1984
; Slob et al., 1983
) and reduced litter size or low pup viability (Fritz et al., 1984
; Popolow and Ward, 1978
; Rosenberg and Sherman, 1974
), and decreased reproductive capacity of the dam including delayed parturition and a greater number of resorptions (Greene et al., 1939
; Swanson and van der Werff ten Bosch, 1965
). In addition, women with hyperandrogenism often have low reproductive success (Gustafson et al., 1996
; Redmond, 1995
).
There is a growing awareness that androgenic chemicals are present in the environment. Female mosquito fish downstream of kraft pulp mill effluent were found with anal fins that had developed into a male-like gonopodium (Davis and Bortone, 1992) and female marine animals and bears have been found masculinized (Vos et al., 2000
). More recently, the drug trenbolone acetate, given to beef cattle to improve weight gain and feed, was shown to cause masculinizing effects in female rat offspring including clitoral enlargement and increased AGD (FDA summary, 2000). The above information provides examples of how environmental androgens might alter development.
In light of this information, it is possible that the female reproductive system would be altered by environmental androgen exposure, especially during the prenatal period. The fetus is acutely susceptible to the effects of environmental endocrine disruptors (Gray et al., 1999; Wolf et al., 2000
). Therefore, the study of the effects of prenatal testosterone propionate (TP) on male and female offspring will prove a useful model for the study of environmental androgens and allow us to identify sensitive endpoints to monitor in future studies. In the present study, we sought to characterize the effects of prenatal TP in the male and female rat and to define developmental endpoints that are sensitive to disruption of sexual differentiation by TP. In 1 experiment, presented first, we describe in vivo effects of various doses of prenatally administered TP in the adult offspring. In a second experiment, we measured the levels of T in the serum of the dam and in the gestational day (GD) 19 male and female whole fetuses after administration of TP to the dam to determine the levels in the fetus responsible for the physical effects of TP observed in the adult rat offspring. This information will be used to determine androgenic, nontoxic doses of TP suitable for coadministration with antiandrogens such as vinclozolin during the gestational period.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Maternal weight was monitored throughout the dosing period. Day of delivery was recorded and GD 23 was designated postnatal day (PND) 1 for all litters, including those that actually delivered on a later day. On PND 2, pups were counted, weighed, sexed if possible (sexing of pups could not be reliably performed in the higher dose groups), and anogenital distance (AGD) was measured. AGD was measured on each pup in a blind fashion using a dissecting microscope fitted with an ocular micrometer reticle. On PND 15, pups were reexamined for sexual phenotype, their sex confirmed or reassigned if necessary, and males and females were checked for areolas in a blind fashion. An areola or a nipple was considered an areola and areola counts were based upon the consensus of 2 technicians. Areolas were described as either faint, smaller than normal, or normal, meaning prominent and easily identified.
On PND 22, pups were weaned, counted, sexed, and measured for AGD in a blind fashion using micron rotary dial calipers (Manostat). Litter mates were assigned an individual identification marking with picric acid stain, distributed 23 per cage, and housed under the same conditions as described for dams except that feed provided was switched to Purina LabDiet 5001 (standard rodent diet). Runts (3 animals) were weaned on PND 28, when they were large enough to reach the water dispenser. Dams were sacrificed 1 day after weaning (pup PND 23) by CO2 asphyxiation followed by decapitation, and uterine implantation sites were counted by visual examination. Female offspring were checked for vaginal opening (VO) from PND 2938 and male offspring were monitored for preputial separation (PPS) from PND 3747 as indicators of puberty. From PND 88 to necropsy, females were sacrificed and gross necropsy was performed if they became bloated, weak, and in poor health.
On PND 112158, females (3 per litter where possible) were euthanized by CO2 asphyxiation followed by decapitation, shaved on the ventral and lateral surfaces of the trunk for viewing nipples, and necropsied in blocked fashion by treatment group. Blood was collected in sterile 13 ml vacutainer serum separation tubes (Becton Dickinson, Lincoln Park, NJ), centrifuged at 1000 x g at 8°C for 15 min and serum collected and stored at 70°C for subsequent measurement of estradiol (E2) levels. Most of the endpoints measured that showed statistical significance at necropsy are summarized in Table 1. Phallus width and length, not included in table, were significant in some dose groups (measured on 3 per litter). Other endpoints, all of which were not significant, include weights of liver, right and left kidney, and paired adrenals (measured on 2 per litter), pituitary weight, and weight of filled or drained uterus (continued on 3 per litter). Uteri were classified as normal or having hydrometrocolpos, a condition marked by severe distention and fluid retention of both the uterus and upper vagina. Ovaries were observed fresh for the presence of corpora lutea (CL). Observation of bulbourethral glands (BUG) was included in the necropsy when this structure was noticed, after 1 female from each of the 0, 0.1, 0.5, and 1 mg TP dose groups had been necropsied. Considering the nonexistent incidence of BUG in the low dose groups, probably no BUG were overlooked.
|
Reproductive tissues (uterus, ovary, phallus, prostate, levator ani [LA], seminal vesicle [SV], BUG, any male structures, vaginal threads, pituitary) were stored in Bouin's fixative for 24 h and rinsed and stored in 70% ethanol for histological assessment (see histopathology section).
Males (2 per litter) were sacrificed by decapitation and necropsied on PND 161172. Endpoints measured were weights of glans penis, pituitary, liver, right and left kidneys, paired adrenals, right and left testes, left epididymis, right caput + corpus epididymis, right cauda epididymis, ventral prostate (VP), SV, and LA + bulbocavernosus (LA/BC). The remaining males were necropsied on PND 177 and 178 for 1 endpoint that showed a statistically significant difference, the glans penis weight, and these data represented all males, or 36 males per litter per dose group.
Radioimmunoassay.
Adult female offspring serum from the necropsy was assayed for E2 using the E2 RIA kit #TKE21 (Diagnostic Products Company, Los Angeles, CA) with the supplied protocol.
Histopathology of female tissues.
Reproductive tracts were dissected from the carcass intact with associated accessory organs (prostate, SV, phallus, and LA) attached, fixed in Bouin's solution for 24 h, rinsed with water, and stored in 75% ethanol. Accessory organs were observed in situ and trimmed by the pathologist. All tissues (distended uteri [n = 5], malformed uteri [n = 3], BUG [n = 8] and ovaries, prostates, LAs, SVs, [1 per litter each]) were paraffin embedded with VIP tissue processor, sectioned 3 microns thick to produce 1 section, stained with Harris hematoxylin and eosin Y (Anatach, Battle Creek, MI), and evaluated for pathologies and the identity of male organs by the pathologist. CLs were identified in ovarian sections (1 section per ovary, 1 ovary per female, 1 female per litter) as large round masses of unorganized, luteal granulosa cells. Ovarian sections were scored for CL abundance and for antral follicle abundance in a blind fashion. Scoring system was designed as follows: no CLs (or antral follicles), score = 0; if 1 to 2, score = 1; if 3 to 6, score = 2; if 7 to 10, score = 3; if 11, score = 4.
Study 2: Maternal/Fetal Testosterone Level Experiment in Vivo
Eight timed-pregnant Sprague-Dawley rats (Charles River Labs, Raleigh, NC) were received on GD 4 (GD 1 = day of sperm positive smear) and housed 1 per cage. Conditions were the same as described for the above experiment during the gestational period. On GD 13, dams were weighed, weight ranked, and randomly assigned to treatment groups that were equilibrated with respect to treatment group. Dams were dosed on GD 1419 by sc injection in the nape of the neck with 0.1 ml of 0 (n = 3), 0.5 (n = 2), or 1 (n = 2) mg TP/0.1 ml corn oil dose solution. These doses were selected based on their masculinizing effects in the female offspring without any toxicity in the dam. On GD 19, 1 h after dosing, dams were euthanized in blocked fashion by CO2 asphyxiation followed by decapitation. Blood and fetuses were collected not sooner than 1 h after dosing to allow TP to undergo distribution and metabolism and reach probable peak T levels in both dam and fetus, as TP has a longer half-life than that of T (Rhees et al., 1997; Sommerville and Tarttelin, 1983
). Each dam was euthanized and its fetuses collected before the next dam was euthanized. Fetuses were removed from the uterus, held on ice in a small plastic petri dish, sexed by opening of their abdominal wall under a dissecting microscope and viewing of internal reproductive organs, and saved in 15 ml plastic round-bottomed Falcon tubes (Becton-Dickinson, Lincoln Park, NJ). Fetuses were stored at 20°C for
1 week until extracted and assayed for T levels. To avoid interference of the fetal carcass collection protocol, maternal blood was collected on a separate set of dams (n = 6; 2 per dose group) at a later date. To avoid contamination of blood samples with the TP dose solution from the neck, blood was collected by cardiac puncture. Dams were maintained under the same housing conditions described above. Dams were selected for blood collection in blocked fashion by treatment group (1 representative of each dose group constitutes a block). Blood was collected by heart puncture while dam was under halothane anesthesia and dams died by exsanguination. Maternal blood was centrifuged and serum was stored at 70°C for 1 week until assayed for T levels. All chemicals were purchased from Sigma-Aldrich Co. (St. Louis, MO) unless otherwise noted.
Fetal testosterone extraction.
T was extracted from fetal tissue as described previously (Parks et al., 2000). Fetuses were thawed and homogenized individually in 500 µl distilled deionized water with a Polytron homogenizer (Brinkmann Instruments, Westbury, NY). After homogenization, 2 ml ethyl ether (Fisher Scientific, Pittsburgh, PA) were added to each tube, tubes were vortexed for 30 s, and centrifuged at 2000 rpm (1000 x g) at 8°C for 10 min. Following centrifugation, each tube was held 1 at a time in an acetone/dry ice bath until the bottom aqueous layer froze, and the supernatant (ether layer) was then transferred to a 12 x 75 mm glass tube. The ether extraction was performed twice. Glass tubes of ether extract were dried in a fume hood overnight. Tubes were stored for up to 2 weeks until analyzed by radioimmunoassay.
Radioimmunoassay (RIA) of dams and fetuses.
Each tube of dried fetal extract was resuspended by vortexing for 30 s in 70 µl of 0 standard buffer provided in the Coat-A-Count Total Testosterone RIA kit #TKTT5 (Diagnostic Products Company, Los Angeles, CA). Fifty µl of the 70 µl fetal resuspension were transferred to the antibody-coated tubes in the RIA kit and T levels were determined according to the manufacturer's protocol. Tubes were read for 1 min each in a gamma counter (CliniGamma 1272, LBK-Wallac, Finland). Counts are programmed to report a ng/ml value. This value was adjusted for volume of fetal extract by multiplying by 0.07 (70 µl/1 ml). Maternal serum was vortexed and 50 µl of straight and 50 µl of 5x diluted serum was assayed in duplicate by RIA for T following the DPC Total Testosterone RIA kit protocol.
Statistics.
Data were analyzed by ANOVA on a litter means basis using the PROC GLM (general liner models) function on SAS (for Windows 95, version 3.0.554, Cary, NC). Weights and distance measurements (kidney to ovary distance, Kd-Ov; AGD, ano-vaginal distance, AVD; phallus length) were analyzed with and without body weight as a covariate. Percentage data (% late delivery, % with nipples) were performed with arcsine transformation of the individual (dam) means or litter (pup) means. When significant differences were found for a main effect, a 2-tailed t-test was used to test differences between treatment groups using least square means. Maternal endocrine data were log transformed to reduce heterogeneity of the variance. Uteri were classified as normal or having hydrometrocolpos both subjectively based on size and quantitatively based on weight. Counts and categorical data on malformations (cleft phallus, vaginal thread, absence of vaginal orifice, vaginal orifice-phallic cleft not separate, presence of prostate, SV, LA, BUG, hydrometrocolpos) and number of females dead before necropsy were analyzed on an individual basis using Fisher's exact test or chi square as appropriate.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
Weaning, puberty, and viability.
Viability from birth to weaning (PND 22) and body weight at weaning were unaffected in either sex by any dose of TP (Table 2).
The ages at PPS in all males and VO in females in the 0.1 and 0.5 mg dose groups were unaffected by TP treatment (F[6,15] = 0.80; F[3,7] = 2.03, respectively; Table 3). However, inspection of females for VO revealed an absence of the vaginal orifice in 100% of females in the 1 through 10 mg TP dose groups (Fig. 2
; Table 5
), absence of the vaginal orifice in 1 female in the 0.5 mg dose group, and a vaginal orifice so small in 1 female in the 0.5 mg TP dose group VO could not be determined. Shortly after puberty, female offspring from the middle range of TP dose groups (in this case the 1, 2, and 5 mg TP groups) appeared to have distended abdomens and began dying (11 in the 1 mg group, 10 in the 2 mg dose group, and 1 in the 5 mg TP dose group died before scheduled necropsy). Thereafter, females with distended abdomens that appeared lethargic were sacrificed for gross necropsy (3 females in the 1 mg dose group, 1 female in the 2 mg dose group; Fig. 3a
). Necropsy of each female revealed an extremely large, distended, fluid-filled uterus and upper vagina (hydrometrocolpos), some uteri weighing as much as 94 grams, with no other gross abnormalities aside from the absence of a vaginal orifice (Fig. 2
). This condition was suspected of being the cause of death of the females that had died.
|
|
|
Females that had no vaginal orifice did have a cervix and upper vagina. In the middle dose groups (1, 2, and 5 mg TP), the upper vagina ended blindly alongside the dorsal aspect of the urethra. In the higher dose groups (mostly 5 and 10 mg TP), the formation of the end of the upper vagina was difficult to determine visually, but the vagina was apparently continuous with and opened into the urethra, as explained subsequently. A test was performed on rats in 2 middle dose groups (2 and 5 mg TP) to determine whether the fluid in the uterus could escape. Saline injected into the uterus of 1 female in the 2 mg TP dose group did not escape but further distended the uterus (uterus initially appeared to be only slightly distended, not having hydrometrocolpos). Saline injected into the uterus of other females without hydrometrocolpos in the 2 mg (1 female) and 5 mg TP (1 female) dose groups passed into the bladder and exited via the urethra to the tip of the phallus, indicating that fluid in the uterus could escape in some females in the high dose groups with no vaginal orifice by exiting through the urethra. In addition to the hydrometrocolpos observed in the middle dose groups, 3 females in the high dose groups (1 in the 5 mg and 2 in the 10 mg TP dose group) had malformed uteri such that the ends of the horns were hard and curled or crumpled, or in 1, the portion of the uterus in which the 2 horns meet had become hardened and enlarged.
Every female in the low dose groups (0, 0.1, and 0.5 mg TP) had nipples and had nearly all 12 nipples. However, at 0.5 mg TP the number of nipples was significantly reduced to 11.25. At 1 mg TP and higher doses the percentage of females having nipples and the mean number of nipples per rat was drastically reduced to near 0 (Table 4).
AGD at necropsy increased in a dose-dependent fashion and was significant (by covariate analysis with body weight) at 1 mg TP and higher doses (Table 4). Although AGD was not increased in the 0.5 dose group, many females displayed an array of genital malformations such as cleft phallus, vaginal thread, and a joined vaginal orifice-cleft phallus (Table 5
). In the latter case, the perimeter of the vaginal orifice ran continuous with the cleft of the phallus so that the 2 were nearly indistinguishable. In those females that did have a complete vaginal orifice, the orifice appeared closer to the phallus than in control females, less well defined, and smaller in diameter (not quantitated). The measurement of AVD (Table 1
) in the 0.5 mg dose group reflected this trend (p = 0.0532), and the difference between the AGD and the AVD, or the vaginal-genital distance (VGD), indicating the distance from the vaginal orifice to the phallus, was significantly smaller at 0.5 mg TP (p < 0.01; Tables 1 and 4
). These malformations were unique to the 0.5 mg dose group, with the exception of a partially cleft phallus that was observed in 1 animal in the 1 mg TP dose group (Table 5
).
Phallus length was significantly increased in the middle dose groups (2 and 5 mg TP; Table 4). Phallus width was increased in the 1 and the 10 mg dose groups only, with or without body weight as a covariatethese results did not reveal a pattern and were variable, possibly due to body fat, and thus were considered less indicative of a response (Table 4
). The internal shaft of the phallus appeared thicker and more developed, more masculine, with increasing dose of TP and included penile bulbs at their base to which LA were usually attached.
Male structures such as prostate, SV, LA, and BUG or cowper's glands, were present in females in the higher dose groups to a significant degree (Table 6, Fig. 4
). Prostatic tissue appeared on the ventral side of the urethra close to the base of the bladder, a location in which it is found in the male, and its incidence was significant at doses as low as 0.5 mg TP. The prostate appeared to increase in size with increasing dose (not quantitated) and to acquire dorsolateral lobes. The SV appeared to be attached to either side of the cervical area of the uterus, and the BUG resided in a pocket within the perineal muscles at location corresponding to the location of the BUG in the male. The identity of these structures was confirmed histologically (1 per litter; Figs. 4c4f
). Some females in the 1 mg TP and higher dose groups also had the appearance of a gubernacular cord upon dissection, or at least the presence of a stream of fat and other connective tissue issuing out from an invagination in the muscle wall of the perineal region on either side of the upper vagina, an area corresponding to the scrotum.
|
|
|
E2 levels from female offspring at necropsy were unaffected (F[6,16] = 2.63; Table 6). E2 levels and visual inspection of uteri and ovaries suggest females from every dose group had estrous cycles.
Male necropsy.
Male necropsy revealed only a reduction in glans penis weight, highly significant in the middle dose groups (in this case 0.5, 1, and 2 mg TP) and slightly significant at 10 mg TP (Table 7), graphically approximating a U-shaped dose-response curve. No other effect was found in the male offspring at necropsy (Table 7
).
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A major outcome of this study is the identification of endpoints in the female sensitive to TP that can be used to detect in utero exposure to androgenic chemicals (Table 1). Endpoints in the female that were most sensitive to maternal sc TP administration, found at the 0.5 mg TP dose, include malformations of the external genitalia, inhibition of areolar and nipple development, and prostate development. Less sensitive endpoints, found in the middle dose ranges, include AGD, complete absence of nipples, complete absence of a vaginal orifice, precocious death and hydrometrocolpos, and LA development. Least sensitive endpoints, induced only at the 2, 5, and 10 mg TP dose levels, include SV and BUG development, and elongation of the ovarian ligament. In addition, our data revealed a remarkable inverted U-shaped dose-response curve for uterine condition that resulted in a similar level of mortality at the corresponding dose levels. A similar cascade of effects was induced by various doses of TP in the early study by Greene et al. (1939). In addition to those endpoints, we included AGD, areola and nipple count and incidence, and hormone levels, and we used a larger number of offspring for the study.
Sensitive Endpoints in Female Offspring
Prostate.
Prostatic tissue was displayed in roughly half the female offspring at 0.5 mg TP, while 0.1 mg TP was without effect. It is possible that some effects, such as presence of prostate, could have been seen at 0.1 mg TP if a greater number of litters was used, however, determining a no observed adverse effect level (NOAEL) was not the primary interest in this study. Earlier studies involving prenatal TP exposure have also reported presence of prostate tissue at doses lower than those capable of inducing development of other male sex accessory glands (Greene et al., 1939; Hamilton and Gardner, 1937
). Development of the prostate is DHT-dependent (Imperato-McGinley et al., 1992
;Schultz and Wilson, 1974
) and the high induction of prostate tissue compared to seminal vesicle or other T-dependent Wolffian derived tissues may result from the greater potency of DHT as compared to T.
Nipple development.
Formation of the external portion of the mammary anlagen (i.e., the areola and nipple) during sexual differentiation in the rat is prevented by the presence of DHT (Goldman et al., 1976; Imagawa et al., 1994
; Imperato-McGinley et al., 1992
; Topper and Freeman, 1980
). The drastic reduction in areola or nipple number observed at 1 mg TP coincided with elevated T levels observed in the female fetus and thus T or its metabolites may have been responsible for the suppression of areola and nipple formation. The appearance and reduced number of areolas and nipples proved to be a sensitive endpoint for androgenicity in the female as evidenced by its significance at 0.5 mg TP and the drastic reduction at 1 mg TP. The presence of areolas in the preweaning male rat is a sensitive indicator of antiandrogenicity as well (Clark et al., 1990
; Gray et al., 1999
; Ostby et al., 1999
; Wolf et al., 2000
).
External genitalia.
Another set of effects seen at the 0.5 mg TP dose is altered morphology of the genitalia. These effects include decreased size of the vaginal orifice, reduced distance from the phallus, absence of the vaginal orifice in 1 female in this dose group, and persistent cleaving of the phallus. These malformations have been reported previously in the rat prenatally exposed to TP (Greene et al., 1939; Swanson and van der Werff ten Bosch, 1965
). These anatomical alterations are similar to the underdeveloped state of the reproductive tracts and urogenital sinus early in development. The vagina and urethra are nearer each other and open into the invaginated urogenital sinus, and the phallus is still cleft, or hypospadiac. We also observed the presence of vaginal thread, or persistent isthmus of tissue across the diameter of the vaginal orifice. The female cleft phallus and the vaginal thread have been associated with prenatal estrogen exposure (Henry et al., 1984
; Vannier and Raynaud, 1980
) and with TCDD or PCB exposure (Flaws et al., 1997
; Gray et al., 1997
; Wolf et al., 1999
) as well. Cleft phallus is also an effect of antiandrogens in the male (Gray et al., 1994
; Imperato-McGinley et al., 1992
;Ostby et al., 1999
; Wolf et al., 2000
). Therefore, the presentation of cleft phallus and vaginal thread alone is not indicative of androgen action, but of endocrine disruption, and must be considered with other endpoints in order to identify the mode of action of the disrupter.
Anogenital distance.
The effect of TP treatment on female AGD was permanent, being increased throughout life. The significant increase at weaning in the female at 1 mg TP, or at 0.5 mg TP when adjusted for body weight, attest to the sensitivity of this effect (Table 3).
Internal Reproductive Effects
Female urogenital tract.
A threshold for several developmental processes was apparent at 1 mg TP. The most unusual effect was hydrometrocolpos, or fluid retention and gross distention of the uterus and upper vagina, an effect associated with vaginal atresia (Fig. 2). This condition has been reported in humans at birth and was associated with vaginal atresia (Janus and Godine, 1986
; Nguyen et al., 1984
). Moderately masculinized female rats in early studies (Greene et al., 1939
; Hamilton and Gardner, 1937
) were found to have a greatly distended, fluid filled uterus. With moderate masculinization, development of the external portion of the vagina does not occur while the upper vagina ends blindly. With further masculinization, development of the urethra and associated genital ducts is directed by androgens in what can be considered a male-like fashion, joining with the urethra. Greene et al. (1939) showed with histology that the upper vagina of affected females was connected to the urethra via a fistula through which the contents of the uterus and vagina could flow freely. A similar fistula, between the vagina and the bladder, has also been reported in a human infant presenting absent vaginal orifice and mild hydrometrocolpos (Takeda et al., 1997
). In the case of Greene et al. the uterus and upper vagina were not distended as the fluid can escape through this fistula and out the urethra, to the tip of the masculinized phallus. Indeed, in our study, saline injected into the uterus in some females with vaginal atresia but no hydrometrocolpos exited through the urethra.
It appears that the females with hydrometrocolpos displayed estrous cyclicity despite the masculinization of their external genitalia. Estrous cyclicity is evidenced by the apparent uterine activity and the presence of CLs observed in fresh and histological sections of ovaries. Doses of prenatally administered TP that induce a similar degree of masculinization of the external genitalia in the rat do not alter the pattern of gonadotropin release (Rhees et al., 1997; Swanson and van der Werff ten Bosch, 1965
). It also appears that females without hydrometrocolpos, specifically those in the higher dose groups, also displayed estrous cyclicity. Evidence of estrous cyclicity in these females is provided by the presence of CLs in fresh ovaries and the unaffected CL and antral follicle score from ovarian histological sections. In addition, serum E2 levels and ovarian weights in adult female offspring were not significantly different among dose groups. The sensitive developmental period for masculinization of gonadotropin release is not the prenatal period but the early postnatal period (Diaz et al., 1995
; Rhees et al., 1997
). Prenatal androgen treatment is less effective than neonatal androgen treatment in inducing early androgen syndrome, an anovulatory syndrome (Huffman and Hendricks, 1981
; Slob et al., 1983
; Swanson and van der Werff ten Bosch, 1965
). Neonatal androgen treatment can cause delayed anovulatory syndrome (DAS; Hendricks et al., 1977
) characterized by lack of estrous cyclicity and reduced numbers of ovulatory follicles associated with reduced ovarian weight due to lack of CL in the adult female offspring, effects not observed in this study.
The number of females in this study who died before necropsy by dose group is reflected by the number of females with hydrometrocolpos by dose group. Both plots depict an inverted U-shaped dose response curve. We hypothesize that the uteri of females in the middle dose groups were responding normally to circulating estrogens by cycling and producing fluid but that the fluid had no outlet for drainage and accumulated, leading to death. The inverted U-shape of this graph thus can be explained by a continuous increase in masculinizationthe partial masculinization of the genitalia at middle dose groups, characterized by agenesis of a vaginal orifice with no outlet for drainage of uterine fluid, and a more complete masculinization of the genitalia at higher doses, characterized by a severely reorganized urogenital tract permitting drainage of uterine fluid and alleviation of the hydrometrocolpos. U-shaped doseresponse curves generated from in vivo experiments are rare, but have been reported from studies on gestationally administered estrogens, DES and E2, in mice (vom Saal et al., 1997). The issue of in vivo U-shaped dose response curves is currently under discussion (NTP, 2000
).
Elongation of ovarian ligament, as measured by kidney-to-ovary distance, occurred only in the high dose groups. This measurement was increased in a previous study in which females were exposed in utero to androgens (Lee and Hutson, 1999) and indicates masculinization, although only with high levels of androgen.
Male reproductive tissues in female offspring.
The pattern or degree of induction of male reproductive tissues in females by prenatal androgen exposure in this study closely follows that observed in past studies (Greene et al., 1939; Hamilton and Gardner, 1937
). However, the induction of the LA and histology of these male tissues was not studied before. The LA is present in the male rat and is located in the perineal region attached to the penile shaft, but is not present in the normal female rat. LA development is T dependent (Tobin and Joubert, 1991
). The LA has been induced in the female by neonatal T treatment in a previous study (Tobin and Joubert, 1991
). In our study, the LA developed only in those females whose phallus had been sufficiently masculinized to form a penile shaft and penile bulbs from which the LA muscle is attached. The bulbocavernosus muscle (BC), another T dependent muscle of the male rat anatomically associated with the LA, did not appear in treated females at any dose. Thus, our study shows organization and development of the LA and BC muscles appear to be differentially regulated with the LA more sensitive to T than is the BC. This was shown elsewhere by differences in AR levels in these 2 muscles altered by castration (Antonio et al., 1999
).
The SV appeared rudimentary in all but 1 observed case. In previous studies, prenatal androgens induced well developed SV, but only at doses capable of inducing the vas deferens and epididymides as well (Wistar rat; Greene et al., 1939). In the current study, vas deferens and epididymis were not observed in any female at any dose. Bulbourethral glands have been reported at doses that induce SV (Greene et al., 1939
), as they were in the current study.
Masculine development of the phallus and internal penile shaft in the female offspring occurred in association with vaginal atresia. The increase in phallus length itself at 2 and 5 mg TP, although not a robust effect, was indicative of masculinization. Phallus length was not increased at 1 or 10 mg TP, although this may be due to a lower number of individuals in these 2 dose groups.
Male Offspring
The decrease in male AGD on PND 2 at 0.5 mg TP was not only a transient effect, but it was not significantly reduced at this dose when analyzed by covariance with body weight. The body weight of all pups at 0.5 mg and higher doses of TP was reduced. However, the body weight of male pups on PND 2 did not further decrease in higher TP dose groups from that in the 0.5 mg TP dose group, while AGD did further decrease at 1 mg TP and again at 2 mg TP, indicating AGD is independent of body weight and the reduction in AGD is most likely a true effect. AGD could not be determined by sex at higher doses of TP and thus male AGD could not be analyzed by covariate analysis with body weight at these doses. Interestingly, decreased AGD can also be produced in the male offspring of antiandrogen-treated dams (Gray et al., 1994; Ostby et al., 1999
; Wolf et al., 2000
). However, these males also displayed nipples and other malformations not observed in our TP exposed males.
The only permanent effect seen in the male was a reduction in glans penis weight at 0.5, 1, 2, and 10 mg TP, but not at 5 mg TP. The graphical representation of these data approximates a U-shaped dose-response curve. Reduction in glans penis weight or phallus size is often associated with antiandrogenicity (Clark et al., 1990; Ostby et al., 1999
; Prahalada et al., 1997
), although it was reported in male offspring of dams fed androgenic Trenbolone acetate prior to pregnancy (FDA, 2000
). However, this effect could not be replicated in more recent studies by the authors of this manuscript (unpublished data). As this effect was the only permanent response seen in any of the androgen-dependent tissues in the male offspring, this response needs to replicated before one can conclude that it is not spurious statistical vagary.
Maternal Toxicity
Adverse effects on reproduction were observed in the dam and neonates at high dose levels of TP. At the 2, 5, and 10 mg TP dose levels, significant adverse effects on maternal reproductive capacity were evident in the decreased weight gain through the dosing period, the delay in parturition, and the decreased litter size. These are well known effects of pre- or perinatal androgen treatment (Fritz et al., 1984; Greene et al., 1939
; Huffman and Hendricks, 1981
; McCoy and Shirley, 1992
; Popolow and Ward, 1978
; Rhees et al., 1997
; Rosenberg and Sherman, 1974
; Swanson and van der Werff ten Bosch, 1965
). The decrease in live litter size is due not to reduced number of implantation sites, but to late fetal resorption (Greene et al., 1939
; our unpublished observations) and pup death at birth as evidenced by the pup carcass remnants on PND 1.
T Levels
Fetal body T concentration was significantly increased at 1 mg TP in the GD 19 female fetus, the dose associated with the profound alterations in sexual development. In a similar study, Hotchkiss (2001) found increased fetal T levels in GD 18 females exposed to 1.5 and to 2.5 mg/kg TP, doses similar to our 0.5 and 1 mg doses. Fetal T level was not elevated in the GD 19 female fetus in our study at 0.5 mg TP, a dose of TP that elicited androgenic responses in the female. However, maternal serum T concentrations rose to 10 times normal levels in this dose group. The difference in T level elevation between dams and fetuses may be explained by the finding that T administered to the rat dam is not delivered to the fetus at equivalent levels, but is metabolized or blocked at the placenta (Vreeburg et al., 1981). In addition, TP given to the rat dam is metabolized by the dam and placenta to other androgens, of which androsterone is the most abundant in the fetus, followed by 3
-androstanediol and epiandrosterone (Slob et al., 1983
). Greene et al. (1939) showed that sc injection of androsterone to the dam can masculinize female rat offspring, thus androsterone may directly or indirectly be a hormone responsible for the malformations of the female genitalia, undetected by the T radioimmunoassay. However, androsterone has low affinity for the androgen receptor and may be converted in the fetus to a more potent androgen. T may also be metabolized to DHT, an androgen known to mediate the tissues and processes affected in the female offspring in this study. DHT levels in the fetus were not analyzed in this study, but will be addressed in future studies.
We have concluded that 0.5 mg and 1 mg TP given on GD 1419 are effectively androgenic while not severely compromising to litter size, viability, or pregnancy and parturition, and illustrate sensitive endpoints to monitor for identification of androgenic chemicals.
It is noteworthy that few of the above endpoints presented herein are included in the standard multigenerational protocols. The standard multigenerational protocols do not include evaluation for the presence of nipples in male or female offspring or examination of female offspring for ventral prostate, levator ani, or other male tissues. In addition, inclusion of AGD is restricted to the F2 generation only after an alteration of puberty is found in the F1 generation. We suggest consideration be given to including some of the more sensitive endpoints in the USEPA Tier 2 testing phase for chemicals that display androgenic activity in Tier 1 screening.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
1 To whom correspondence should be addressed. Fax: (919) 541-4017. E-mail: gray.earl{at}epa.gov.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Antonio, J., Wilson, J. D., and George, F. W. (1999). Effects of castration and androgen treatment on androgen-receptor levels in rat skeletal muscles. J. Appl. Physiol. 87, 20162019.
Bentvelsen, F. M., Brinkmann, A. O., van der Schoot, P., van der Linden, J. E., van der Kwast, T. H., Boermsa, W. J., Schroder, F. H., and Nijman, J. M. (1995). Developmental pattern and regulation by androgens of androgen receptor expression in the urogenital tract of the rat. Mol. Cell. Endocrinol. 113, 245253.[ISI][Medline]
Berenbaum, S. A., Duck, S. C., and Bryk, K. (2000). Behavioral effects of prenatal versus postnatal androgen excess in children with 21-hydroxylase-deficient congenital adrenal hyperplasia. J. Clin. Endocrinol. Metab. 85, 727733.
Clark, R. L., Antonello, J. M., Grossman, S. J., Wise, L. D., Anderson, C., Bagdon, W. J., Prahalada, S., MacDonald, J. S., and Robertson, R. T. (1990). External genitalia abnormalities in male rats exposed in utero to finasteride, a 5-reductase inhibitor. Teratology 42, 91100.[ISI][Medline]
Cunha, G. R., Cooke, P. S., Bigsby, R., and Brody, J. R. (1991). Ontogeny of sex steroid receptors in mammals. In Nuclear Hormone Receptors (M. Parker, Ed.), pp. 235268. Academic Press, New York.
Davis, W. P., and Bortone, S. W. (1992). Effects of mill effluent on fish sexuality. In Advances in Modern Environmental Toxicology. Chemically Induced Alterations in Sexual and Functional Development: The Wildlife/Human Connection (T. Colborn and C. Clements, Eds.), Vol. XXI, pp. 113127. Princeton Scientific, Princeton, NJ.
Diaz, D. R., Fleming, D. E., and Rhees, R. W. (1995). The hormone-sensitive early postnatal periods for sexual differentiation of feminine behavior and luteinizing hormone secretion in male and female rats. Dev. Brain Res. 86, 227232.[ISI][Medline]
FDA (2000). NADA summary #138612. Food and Drug Administration. Available at: http://www.fda.gov/cvm/efoi/section1/138612.html. Accessed October 19, 2001.
Flaws, J. A., Sommer, R. J., Silbergeld, E. K., Peterson, R. E., and Hirshfield, A. N. (1997). In utero and lactational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induces genital dysmorphogenesis in the female rat. Toxicol. Appl. Pharmacol. 147, 351362.[ISI][Medline]
Fritz, H., Giese, K., and Suter, H. P. (1984). Prenatal and postnatal development of rats following the maternal treatment with testosterone during the late period of embryogenesis. Arzneimittelforschung 34, 780782.[Medline]
Gandelman, R., Simon, N. G., and McDermott, N. J. (1979). Prenatal exposure to testosterone and its precursors influences morphology and later behavioral responsiveness to testosterone of female mice. Physiol. Behav. 23, 2326.[ISI][Medline]
Goldman, A. S., Shapiro, B., and Neumann, F. (1976). Role of testosterone and its metabolites in the differentiation of the mammary gland in rats. Endocrinology 99, 14901495.[Abstract]
Gray, L. E., Jr., Ostby, J. S., and Kelce, W. R. (1994). Developmental effects of an environmental antiandrogen: The fungicide vinclozolin alters sex differentiation of the male rat. Toxicol. Appl. Pharmacol. 129, 4652.[ISI][Medline]
Gray, L. E., Jr., Ostby, J. S., Monosson, E., and Kelce, W. R. (1999). Environmental antiandrogens: Low doses of the fungicide vinclozolin alter sexual differentiation of the male rat. Toxicol. Ind. Health 15, 4864.[ISI][Medline]
Gray, L. E., Jr., Wolf, C., Mann, P., and Ostby, J. S. (1997). In utero exposure to low doses of 2,3,7,8-tetrachlorodibenzo-p-dioxin alters reproductive development of female Long Evans hooded rat offspring. Toxicol. Appl. Pharmacol. 146, 237244.[ISI][Medline]
Greene, R. R., Burrill, M. W., and Ivy, A. C. (1939). Experimental intersexuality. The effect of antenatal androgens on sexual development of female rats. Am. J. Anat. 65, 415469.
Goy, R. W., Bercovitch, F. B., and McBrair, M. C. (1988). Behavioral masculinization is independent of genital masculinization in prenatally androgenized female rhesus macaques. Horm. Behav. 22, 552571.[ISI][Medline]
Gustafson, O., Nylund, L., and Carlstrom, K. (1996). Does hyperandrogenism explain lower in vitro fertilization (IVF) success rates in smokers? Acta Obstet. Gynecol. Scand. 75, 149156.[ISI][Medline]
Hamilton, J. B., and Gardner, W. U. (1937). Effects in female young born of pregnant rats injected with androgens. Proc. Soc. Exp. Biol. Med. 37, 570572.
Hendricks, S. E., McArthur, D. A., and Pickett, S. (1977). The delayed anovulation syndrome: Influence of hormones and correlation with behaviour. J. Endocrinol. 75, 1522.[Abstract]
Henry, E. C., Miller, R. K., and Baggs, R. B. (1984). Direct fetal injections of diethylstilbestrol and 17 ß-estradiol: A method for investigating their teratogenicity. Teratology 29, 297304.[ISI][Medline]
Hotchkiss, A. K. (2001). The effect of androgens and antiandrogens on sexual differentiation in male and female rats. Ph.D. dissertation. North Carolina State University, Raleigh, NC.
Huffman, L., and Hendricks, S. E. (1981). Prenatally injected testosterone propionate and sexual behavior of female rats. Physiol. Behav. 26, 773778.[ISI][Medline]
Imagawa, W., Yang, J., Guzman, R., and Nandi, S. (1994). Control of mammary gland development. In The Physiology of Reproduction (E. Knobil and J. D. Neill, Eds.), pp. 10331063. Raven Press, New York.
Imperato-McGinley, J., Sanchez, R. S., Spencer, J. R., Yee, B., and Vaughan, E. D. (1992). Comparison of the effects of the 5-reductase inhibitor finasteride and the antiandrogen flutamide on prostate and genital differentiation: Dose-response studies. Endocrinology 131, 11491156.[Abstract]
Ito, S., Murakami, S., Yamanouchi, K., and Arai, Y. (1986). Prenatal androgen exposure, preoptic area and reproductive functions in the female rat. Brain Dev. 8, 463468.[ISI][Medline]
Janus, C., and Godine, L. (1986). Newborn with hydrometrocolpos and ambiguous genitalia: Clinical significance. J. Clin. Ultrasound 14, 739741.[ISI][Medline]
Josso, N., Picard, J. Y., and Tran, D. (1977). The anti-Mullerian hormone. Recent Progr. Hormone Res. 33, 117160.
Kawashima, K., Nakaura, S., Nagao, S., Tanaka, S., Kuwamura, T., and Omori, Y. (1978). Virilizing effect of methyltestosterone on female descendants in the rat. Endocrinol. Jpn. 25, 16.[Medline]
Lee, S. M., and Hutson, J. M. (1999). Effect of androgens on the cranial suspensory ligament and ovarian position. Anat. Rec. 255, 306315.[ISI][Medline]
Mann, M. A., and Svare, B. (1983). Prenatal testosterone exposure elevates maternal aggression in mice. Physiol. Behav. 30, 503507.[ISI][Medline]
McCoy, S. J., and Shirley, B. A. (1992). Effects of prenatal administration of testosterone and cortisone on the reproductive system of the female rat. Life Sci. 50, 621628.[ISI][Medline]
NTP (2000). Endocrine Disruptors Low Dose Peer Review, Meeting Follow-up. National Toxicology Program. Available at: http://ntp-server.niehs.nih.gov/hotdoc/liason/LowDoseWebPage.html. Accessed June 4, 2000.
New, M. I., and Wilson, R. C. (1999). Steroid disorders in children: Congenital adrenal hyperplasia and apparent mineralocorticoid excess. Proc. Nat. Acad. Sci. U.S.A. 96, 1279012797.
Nguyen, L., Youssef, S., Guttman, F. M., Ahlgren, L. S., and Schlechter, R. (1984). Hydrometrocolpos in neonate due to distal vaginal atresia. J. Pediatr. Surg. 19, 510514.[ISI][Medline]
Ostby, J. S., Kelce, W. R., Lambright, C. R., Wolf, C. J., Mann, P., and Gray, L. E., Jr. (1999). The fungicide procymidone alters sexual differentiation in the male rat by acting as an androgen-receptor antagonist in vivo and in vitro. Toxicol. Ind. Health 15, 8093.[ISI][Medline]
Parks, L. G., Ostby, J. S., Lambright, C. R., Abbott, B. D., Klinefelter, G. R., Barlow, N. J. and Gray, L.E., Jr. (2000). The plasticizer diethylhexyl phthalate induces malformations by decreasing fetal testosterone synthesis during sexual differentiation on the male rat. Toxicol. Sci. 58, 339349.
Phoenix, C. H., Goy, R. W., Gerall, A. A., and Young, W. C. (1959). Organizing action of prenatally administered testosterone propionate on the tissues mediating mating behavior in the female guinea pig. Endocrinology 65, 369382.[ISI]
Popolow, H. B., and Ward, I. L. (1978). Effects of perinatal androstenedione on sexual differentiation in female rats. J. Comp. Physiol. Psychol. 92, 1321.[ISI][Medline]
Prahalada, S., Tarantal, A. F., Harris, G. S., Ellsworth, K. P., Clarke, A. P., Skiles, G. L., MacKenzie, K. I., Kruk, L. F., Ablin, D. S., Cukierski, M. A., Peter, C. P., vanZwieten, M. J., and Hendrickx, A. G. (1997). Effects of finasteride, a type 2 5- reductase inhibitor, on fetal development in the rhesus monkey (Macaca mulatta). Teratology 55, 119131.[ISI][Medline]
Redmond, G. P. (1995). Androgenic disorders of women: Diagnostic and therapeutic decision making. Am. J. Med. 98, 120S129S.[Medline]
Rhees, R. W., Kirk, B. A., Sephton, S., and Lephart, E. D. (1997). Effects of prenatal testosterone on sexual behavior, reproductive morphology and LH secretion in the female rat. Dev. Neurosci. 19, 430437.[ISI][Medline]
Rosenberg, K. M., and Sherman, G. F. (1974). Testosterone induced pup-killing behavior in the ovariectomized female rat. Physiol. Behav. 13, 697699.[ISI][Medline]
Schardein, J. L. (1993). Hormones and hormonal antagonists. In Chemically Induced Birth Defects (J. Schardein, Ed.), pp. 271339. Marcel Dekker, New York.
Schultz, F. M., and Wilson, J. D. (1974). Virilization of the Wolffian duct in the rat fetus by various androgens. Endocrinology 94, 979986.[ISI][Medline]
Shapiro, E., Huang, H. Y., and Wu, X. R. (2000). Uroplakin and androgen receptor expression in the human fetal genital tract: Insights into the development of the vagina. J. Urol. 164, 10481051.[ISI][Medline]
Slob, A. K., den Hamer, R., Woutersen, P. J. A., and van der Werff ten Bosch, J. J. (1983). Prenatal testosterone propionate and postnatal ovarian activity in the rat. Acta Endocrinol. 103, 420427.[ISI][Medline]
Sommerville, E. M., and Tarttelin, M. F. (1983). Plasma testosterone levels in adult and neonatal female rats by testosterone propionate-filled silicone elastomer capsules for varying periods of time. J. Endocrinol. 98, 365371.[Abstract]
Swanson, H. E., and van der Werff ten Bosch, J. J. (1965). The "early androgen" syndrome; effects of pre-natal testosterone propionate. Acta Endocrinol. (Copenh.) 50, 379390.[ISI][Medline]
Takeda, A., Konda, R., Sakai, K., Ota, S., and Orikasa, S. (1997). Hydrometrocolpos with ectopic vaginal opening to the bladder. A case report. Urol. Int. 59, 5758.[ISI][Medline]
Tobin, C., and Joubert, Y. (1991). Testosterone-induced development of the rat levator ani muscle. Dev. Biol. 146, 131138.[ISI][Medline]
Topper, Y. J., and Freeman, C. S. (1980). Multiple hormone hormone interactions in the developmental biology of the mammary gland. Physiol. Rev. 60, 10491106.
Vannier, B., and Raynaud, J. P. (1980). Long-term effects of prenatal oestrogen treatment on genital morphology and reproductive function in the rat. J. Reprod. Fertil. 59, 4349.[Abstract]
vom Saal, F. S., Timms, B. G., Montano, M. M., Palanza, P., Thayer, K. A., Nagel, S. C., Dhar, M. D., Ganjam, V. K., Parmigiani, S., and Welshons, W. V. (1997). Prostate enlargement in mice due to fetal exposure to low doses of estradiol or diethylstilbestrol and opposite effects at high doses. Proc. Nat. Acad. Sci. U.S.A. 94, 20562061.
Vos, J. G., Dybing, E., Greim, H. A., Ladefoged, O., Lambre, C., Tarazona, J. V., Brandt, I., and Vethaak, A. D. (2000). Health effects of endocrine-disrupting chemicals on wildlife, with special reference to the European situation. Crit. Rev. Toxicol. 30, 71133.[ISI][Medline]
Vreeburg, J. T. M., Woutersen, P. J. A., Ooms, M. P., and van der Werff ten Bosch, J. J. (1981). Androgens in the fetal guinea-pig after maternal infusion of radioactive testosterone. J. Endocrinol. 88, 916.[Abstract]
Wolf, C. J., LeBlanc, G. A., Ostby, J. S., and Gray, L. E., Jr. (2000). Characterization of the period of sensitivity of fetal male sexual development to vinclozolin. Toxicol. Sci. 55, 152161.
Wolf, C. J., Ostby, J. S., and Gray, L. E., Jr. (1999). Gestational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) severely alters reproductive function of female hamster offspring. Toxicol. Sci. 51, 259264.[Abstract]