Characterization of diethylstilbestrol-induced hypospadias in female mice

Abstract

The urethral duct and vagina are formed from the urogenital sinus (UGS) during the early neonatal period in mice. Neonatal estrogen exposure results in hypospadias, or the malpositioning of vaginal and urethral openings, with wide cleft clitoris. We sought to characterize diethylstilbestrol (DES) influence on UGS morphogenesis and hypospadias formation. Newborn (day 0) and 1–4-day-old female mice (ICR/Jcl) were given (s.c.) oil or 3.0 μg DES. Animals were killed 24 hr later; then hypospadias formation and epithelial apoptosis and proliferation within the developing UGS were assessed. DES did not alter normal UGS morphogenesis by day 1, in comparison with controls. However, hypospadias formation was observed in DES-treated mice by day 3. In these mice, the distal dorsal urethral duct appeared to fuse with and open into the lower vaginal solid cord region. Further, DES treatment produced a gradual significant increase in dorsal urethral epithelial apoptosis (P < 0.05) just prior to and during fusion and hypospadias formation. DES-induced urethral epithelial and sinus cord proliferation appeared significantly increased (P < 0.05) and unchanged, respectively, just prior to fusion. By day 5, DES-treated mice exhibited wide cleft clitoris. In addition, if DES was given on day 3 or 5, a gradual, distinct caudal shift in the vaginal-urethral junction was observed compared to mice treated on days 0–2. Although hypospadias was not induced when neonates were given DES on day 7, these mice continued to display early vaginal opening. Dose-response analysis indicated that 0.03 μg DES for 5 days is the lowest known critical dose for hypospadias induction. We have shown for the first time that DES-induced hypospadias onset may primarily be the result of changes in developing dorsal urethral epithelial cell apoptotic and proliferative activity, and that the location of DES-induced hypospadias formation is dependent on age at time of exposure. Anat Rec 266:43–50, 2002. © 2002 Wiley-Liss, Inc.

The progression of normal reproductive tract morphogenesis involves a delicate balance of cell growth, proliferation, differentiation, and apoptosis. Neonatal exposure of female mice to estrogens, including diethylstilbestrol (DES), results in various developmental, functional, and pathological abnormalities of the vaginal tissues (for reviews see Takasugi, 1976; Iguchi, 1992). Such abnormalities include persistent epithelial proliferation and cornification (Takasugi et al., 1962; Takasugi, 1963; Takasugi and Bern, 1964; Iguchi et al., 1988); hyperplastic lesions; carcinogenesis (Dunn and Green, 1963); and, of particular interest, hypospadias, the formation of a common urethral-vaginal canal accompanied by wide cleft clitoris (Takasugi and Bern, 1962; Forsberg and Lannerstad, 1968).

In normal neonatal female mice, the vagina consists of a proximal part (cranial 3/5) derived from the Müllerian duct, and a distal part (caudal 2/5) which, along with the urethra, arises from the urogenital sinus (UGS) (Forsberg, 1965). The urethral canal, which is approximately parallel to the vagina, is already open to the skin by birth. The distal vaginal rudiment is composed of the sinus cord and is not open to the skin at birth. However, the developing vaginal lumen progressively extends distally during the neonatal and early postnatal periods. Under normal circumstances, vaginal opening finally occurs around 5 weeks of age following a hormonally-triggered apoptotic event (Rodriguez et al., 1997). Early vaginal opening in rats can also be induced by exogenous 17β-estradiol or xenoestrogens (Ashby and Tinwell, 1998). For example, rats given an oral dose of the isoflavonoid coumestrol exhibit early vaginal opening and precocious sexual maturation (Whitten and Naftolin, 1992). Postnatal exposure to the chemical bisphenol-A (BPA), which is a monomer of polycarbonate plastics used in food packaging and is a food contaminant shown to have estrogenic activity (vom Saal et al., 1998), induces early vaginal opening in rats (Ashby and Tinwell, 1998).

Hypospadias in the female mouse is a malpositioning of the urethral and vaginal openings that is induced perinatally by estrogens or neonatally by androgens (Iguchi and Takasugi, 1976; Ozawa et al., 1991). Forsberg and Lannerstad (1968) observed hypospadias formation in female mice with inhibition of UGS differentiation after estrogen administration. Although hypospadias is a serious abnormality caused by developmental estrogen exposure, the cellular and molecular changes leading to hypospadias formation remains poorly understood. In the present study, we investigated the cellular mechanisms and morphological changes leading to hypospadias as induced by DES. Cell proliferation, apoptosis, and hypospadias formation were monitored in the developing vaginal and urethral mouse tissues following neonatal exposure to DES. The effect of the known estrogenic contaminant, BPA, on hypospadias formation was also examined.

MATERIALS AND METHODS

Animals

Female mice of the ICR/Jcl strain, kept under 12 hr light/12 hr dark at 23–25°C, were fed laboratory chow (CE-2, CLEA, Tokyo, Japan) and tap water ad libitum. All procedures were carried out according to the NIH Guide for the Care and Use of Laboratory Animals.

Treatments, Histology, and Morphometry

To determine the critical time point of hypospadias induction and assess histologic changes, female mice were injected (s.c.) for 4 days beginning on day 0 (day of birth) with 3.0 μg DES (Sigma Chemical, St. Louis, MO) dissolved in 0.02 ml sesame oil or vehicle alone, and killed 24 hr later (Fig. 1A). Some animals were killed 12 hr after the three daily DES injections (day 2.5). To relate the approximate anatomical position of hypospadias formation to time of exposure, four groups of mice received one injection of DES (3.0 μg) or vehicle every 24 hr for 5 days, beginning at 0, 3, 5, or 7 days of age. All four groups were killed on day 12 (Fig. 1B). To assess the critical dose for hypospadias induction by DES and compare the effect of BPA, five groups of mice were given a daily single injection of either DES (0.0003, 0.003, 0.03, or 0.3 μg), or 300 μg BPA (Tokyo Kasei, Tokyo, Japan) for 5 days beginning on day 0, and killed 24 hr later (day 5; Fig. 1C). As a way to confirm DES exposure, the extent of vaginal epithelial stratification was also assessed in these animals. For histologic and morphometric analyses, the caudal half of some mice were fixed in Bouin's solution, paraffin embedded, and 8-μm sections were stained with hematoxylin and eosin. Positional shifts in the location of hypospadias formation were determined as a percent shift in distance from the cervix to the vaginal opening at the epidermal surface. Distance in shift was measured through a WT10X eyepiece using an Olympus Optical, BH microscope (Olympus Optical, Tokyo, Japan). More than five mice were used for each experiment group. Unless otherwise mentioned, materials were purchased from Wako Pure Chemical, Osaka, Japan.

Figure 1. Treatment timelines. A: The earliest time points of hypospadias formation and histologic changes were assessed in female mice for 5 days following one s.c. injection of oil or 3.0 μg DES beginning on day 0 (day of birth). Animals were killed 24 hr after treatment. B: The position of hypospadias formation based on age at time of DES exposure was determined after 3.0 μg DES or vehicle every 24 hr for 5 days, beginning at 0, 3, 5, or 7 days of age. All animals were killed on day 12. C: The critical DES dose for hypospadias induction and the effect of BPA were assessed in 5-day-old mice following one injection every 24 hr for 5 days, beginning on day 0, with 0.0003, 0.003, 0.03, or 0.3 μg DES, or 300 μg BPA.

In Situ 3′-DNA Nick End Labeling

Apoptosis was examined in the dorsal urethral wall, ventral vaginal wall, and sinus cord from 3.0 μg DES-exposed and control mice at 0, 1, 2, 2.5, and 3 days of age (Fig. 5A). Tissues were fixed in 10% neutral buffered formalin solution, paraffin embedded, sectioned at 6 μm, and mounted on glass slides coated with 3-aminopropyltriethoxysilane (2%; Shinetsu, Nagano, Japan) in acetone. 3′-DNA nick end labeling of tissue sections was carried out by the method of Gavrieli et al. (1992) with minor modifications. To strip cell nuclei of proteins, tissues were deparaffinized, rehydrated, and then incubated in 20 μg/ml proteinase K in TE buffer (10 mM Tris-HCl, 10 mM EDTA, pH 8.0) for 15 min and washed for 3 min in TE. Endogenous peroxidase was inactivated with 1.5% H₂O₂ in TE and sections were rinsed with terminal deoxynucleotidyl transferase (TdT) buffer (30 mM Tris, 140 mM sodium cacodylate, 1 mM cobalt chloride, pH 7.4) and preincubated for 15 min. Tissues were covered with TdT (0.3 U/ml; Takara, Ohtsu, Japan) and biotinylated dATP (4 μM; GIBCO BRL, Grand Island, NY) in TdT buffer and incubated in a humidified chamber at 37°C for 90 min. For negative controls, TdT enzyme was omitted from the reaction mixture. The reaction was stopped in termination buffer (300 mM NaCl, 30 mM sodium citrate). Tissues were rinsed with 10 mM PBS (pH 7.4) and covered with prewarmed extra-avidin peroxidase (Sigma) diluted 1:1,000 in PBS for 30 min at 37°C. For color development, diaminobenzidine tetrahydrochloride (DAB; Sigma) in imidazole-containing buffer was used. Finally, tissues were counterstained with hematoxylin and eosin. The number of apoptotic epithelial cells was counted in the distal and dorsal urethral wall, sinus cord, and distal ventral vaginal wall, separately. The apoptotic index was estimated by counting the number of apoptotic cells per 200 cells in each region. Values were compared between treatment groups using the Dunnett's test or Dunnett type mean rank test, followed by Bartlett's test; differences with a P < 0.05 were considered significant.

Bromodeoxyuridine-Labeling and Immunostaining

The dorsal urethral wall, ventral vaginal wall, and sinus cord were also examined for bromodeoxyuridine (BrdU) uptake in the 3.0 μg DES-exposed and control mice at 0, 1, 2, 2.5, and 3 days of age. A single injection of 20 mg BrdU (Sigma)/100 g body weight was given (s.c.) to mice 3 hr before they were killed. Tissues fixed in 10% formalin neutral buffered solution were embedded in paraffin and sectioned at 6 μm. Sections were mounted on 2% 3-aminopropyltriethoxysilane-coated glass slides, deparaffinized, and rehydrated. The immunohistochemical procedure was performed as described previously (Ohta et al., 1994) with minor modifications. Sections were washed three times in PBS and digested with 0.1% trypsin (Sigma) in 0.1% CaCl₂ (pH 7.8) for 30 min at 37°C. After washing in PBS, endogenous peroxidase was inactivated by 0.3% H₂O₂ diluted with methanol for 30 min, followed by a PBS wash. Tissues were then incubated with anti-BrdU monoclonal antibody or PBS containing 10 units/ml nuclease (Amersham Pharmacia Biotech, Buckinghamshire, UK) for 60 min. After washing in PBS, sections were incubated with the anti-mouse HRP-F(ab′)2 fragment (Amersham) for 30 min. DAB reaction was carried out for 30 min. Tissues were counterstained with hematoxylin and eosin. BrdU-labeling index was estimated by counting the number of BrdU-incorporated cells per 200 cells in the distal dorsal urethral, sinus cord, and ventral vaginal regions. Values were compared between treatment groups using the Dunnett's test or Dunnett type mean rank test, followed by Bartlett's test; differences with a P < 0.05 were considered significant.

RESULTS

Histologic and Morphometric Analyses

On day 0, epithelia of the distal vaginal rudiment in control mice showed normal morphogenesis consisting of the solid cord composed of cuboidal cells without lumen (Fig. 2A). Within this region, urethral and sinus cord epithelial cells could not be distinguished from one another morphologically, and the dorsal vaginal part and ventral urethral part of the UGS were not completely separated by the stromal tissue (Fig. 2A and D), while urethral and vaginal epithelial cells were completely separated by stroma in the proximal region (Fig. 2B and C). Vaginal lumen formation in day 0 of control mice gradually proceeded caudally through day 4. Until day 5, the distal solid cord remained unchanged from birth and showed no separation from the urethral epithelium. Thus, at this stage, the vaginal lumen was not yet formed distally; no vaginal opening was observed. DES (3.0 μg) had no apparent morphological effect on lower reproductive tract development by day 2, compared to controls. However, by 3 days of age the vaginal and urethral lumen fused, resulting in the formation of a common vaginal-urethral lumen (Fig. 3B). By 5 days of age, the area where the vagina would normally open still opened to the outside as part of the bigger common urethral-vaginal opening with wide cleft clitoris (Fig. 3C).

Figure 2. A: Longitudinal section of the Müllerian- and UGS-derived developing vaginal and urethral tissues at day 0 (×48 magnification). Cross sections of developing vagina and urethra from day 0 neonates in (B) proximal, (C) middle, and (D) distal regions. Note the mesenchymal separation of the developing vaginal and urethral lumens in the proximal and middle, but not distal, regions. ×98 magnification for B, C, and D. v, vagina; u, urethra; s, sinus cord; r, rectum.

Figure 3. Longitudinal sections of (A) day 2.5 control, (B) day 2.5 DES (3.0 μg;, ×140 magnification), and (C) day 5 DES (3.0 μg;, ×60 magnification). v, vagina; u, urethra; s, sinus cord.

The position of DES-induced hypospadias formation was assessed in mice at 12 days and related to age of exposure. Control mice did not exhibit hypospadias or early vaginal opening (Fig. 4A; Table 1). However, all mice treated with 3.0 μg DES for 5 days beginning on day 0 exhibited hypospadias by day 12 (Fig. 4B; Table 1). Daily DES injections beginning on day 3 or 5 resulted in a gradual but distinct shift of the junctional position between the vaginal and urethral lumens in the caudal direction by the day the mice were killed (Table 1; compare Fig. 4B and C). Interestingly, the incidence of hypospadias formation was minimal compared to controls when DES administration was started at 7 days of age, although these mice displayed early vaginal opening by day 12 (Table 1; Fig. 4D).

Figure 4. Day 12 distal vagina after (A) one injection per day of oil or 3.0 μg DES for 5 days beginning on (B) day 0, (C) 5, or (D) 7. ×35 magnification. v, vagina; u, urethra.

Table 1. Diethylstilbestrol (DES; 3.0 μg)-induced hypospadias formation and age-dependent directional shift through 12 days of age

Treatment period (days of age)	Number of mice	Mice with hypospadias induction (%)	Mice with early vaginal opening (%)	Percent shift from cervix
Neonatal mice received daily a single s.c. injection for 5 days beginning on either the day of birth (day 0), day 3, 5, or 7. An additional group received oil for 5 days beginning on day 0. Tissues were collected from all animals on day 12.
0–5 DES	5	100	100	36.8 ± 1.6
3–8 DES	5	100	100	84.1 ± 2.2
5–10 DES	5	80	100	86.9 ± 4.0
7–12 DES	6	17	100	96.2
0–5 Oil	6	0	0	—

Threshold Dose Analysis and Vaginal Epithelial Stratification

The inductive effect of DES on hypospadias formation was determined over a range of concentrations. Neonates given the two lowest DES doses (0.003 or 0.0003 μg) exhibited normal development, similar to controls. However, of the animals treated with 0.3 μg or 0.03 μg, all developed hypospadias by day 5. Therefore, we conclude that 0.03 μg DES/day is the lowest known critical dose for hypospadias induction (Table 2). The percent shift in distance from the cervix to the vaginal-urethral junction was similar in 3.0, 0.3, and 0.03 μg DES-exposed mice by 5 days (Table 2). In these groups, vaginal epithelia were stratified with six to 10 cell layers, while vaginal epithelia from groups exposed to 0.003 μg was four to six cell layers by day 5. Vaginal epithelial stratification in 0.0003 μg DES-treated and control mice was similar, with only two cell layers. Neither hypospadias nor increased epithelial stratification were observed in neonates exposed to 300 μg BPA.

Table 2. Threshold effect of diethylstilbestrol (DES) on hypospadias induction in neonates

Treatment (μg)	Number of mice	Mice with hypospadias induction (%)	Mice with early vaginal opening (%)	Percent shift from cervix
Neonatal mice received daily a single s.c. injection for 5 days beginning on the day of birth (day 0). Tissues were collected from all animals on day 5.
3.0 DES	6	100	100	61.7 ± 6.1
0.3 DES	5	100	100	60.5 ± 3.3
0.03 DES	6	100	100	68.1 ± 5.8
0.003 DES	7	0	0	—
0.0003 DES	7	0	0	—
Oil	6	0	0	—

Apoptotic and BrdU-Labeling Indices

The areas examined for apoptosis and BrdU labeling are shown in Fig. 5A. Overall, the number of apoptotic cells was low in the developing urethra and vagina. Nonetheless, significant differences between groups were detected. In DES-exposed mice, the sinus cord apoptotic index began to rise by day 2 but was not significantly increased until day 3 (P < 0.05; Fig. 5). However, the apoptotic index in dorsal urethral epithelial cells was significantly increased (P < 0.05) as early as day 2, compared to controls, and continued to rise through day 3 (Fig. 5). Ventral vaginal epithelial apoptotic index did not differ from controls through day 1, but was significantly decreased (P < 0.05) by day 2 and remained decreased through day 3. Apparently, there were regional differences in the distribution of apoptotic cells in the sinus cord. The number of apoptotic epithelial cells appeared to be increased at the junction of the vaginal and urethral lumens on day 3 by DES (3.0 μg) compared to controls (Fig. 6). Apoptotic indices in urethral stromal cells were not altered by DES injection (data not shown). BrdU-labeling index was similar after 3.0 μg DES injection throughout days 0–2.5 in sinus cord epithelia. However, DES significantly decreased the BrdU-labeling index in sinus cord epithelia (P < 0.05) by day 3 (Fig. 5). In the dorsal urethral epithelium of DES-exposed mice, the proliferative index increased at day 2, but returned to levels similar to those of controls by day 3 (Fig. 5). The BrdU-labeling indices of urethral stromal cells were not altered by DES (data not shown).

Figure 5. Schematic illustration of the region examined for (A) apoptosis and BrdU labeling, and (B) apoptotic and BrdU-labeling indices. Indices are presented using percent values based on the number of apoptotic or BrdU-incorporated cells per 200 cells in the distal dorsal urethra, sinus cord, and ventral vaginal regions of control and DES-exposed mice. The bars indicate standard error (n = 6–8). Points of statistical significance vs. control are indicated by asterisks (*), with P < 0.05.

Figure 6. Apoptotic epithelial cells at the vaginal-urethral junction in the distal reproductive tract of day 3 (A) oil control and (B) DES-treated (3.0 μg) neonates. Apoptotic cells, which are prominent around the junctional region of DES-exposed animals, are apparent by the accumulation of dark brown reaction product. ×280 magnification. u, urethra; s, sinus cord.

DISCUSSION

Organogenesis of the reproductive tract is not complete by birth in mice, and as such it is an increasingly unprotected target for chemical action. Hypospadias formation with wide cleft clitoris and early vaginal opening are archetypal endpoints of neonatal DES exposure (Takasugi and Bern, 1962; Whitten and Naftolin, 1992; Ashby and Tinwell, 1998). However, the cellular and morphogenic changes which culminate in female hypospadias formation are poorly understood. Furthermore, dose dependence and developmental sensitivity remain undefined. Thus, we sought to characterize the DES effect on UGS morphogenesis and hypospadias formation.

Apoptosis, which is defined as the rapid elimination of apoptotic bodies by ingestion from surrounding cells without inflammatory reaction (Kerr et al., 1972), is necessary to eliminate undesirable cells and is an essential event of morphogenesis. For example, vaginal opening in mice occurs following apoptosis within the hymen at about 5 weeks of age (Rodriguez et al., 1997). In the present study, the epithelial apoptotic and proliferative indices were examined in the vaginal and urethral regions of DES-exposed mice in order to further understand the cellular mechanisms underlying hypospadias formation. Thus, we examined cell proliferation and death during and after the period of hypospadias onset (days 0–3).

Since the urethral duct is normally open at the skin surface by birth, it seems likely that an induced alteration in the caudal progression of vaginal lumen formation would result in hypospadias. Vaginal lumen formation would deviate from its normal path and encroach upon the urethral duct. The vagina opened into the urethra, but also formed its normal opening at the skin surface, resulting in a larger common urethra-vaginal opening with wide cleft clitoris. Following DES, not only the vagina and sinus cord but also the urethra had an effect on hypospadias formation. In support of this, DES-induced dorsal urethral epithelial apoptosis was significantly increased by day 2, prior to hypospadias formation. DES treatment also increased dorsal urethral epithelial proliferative activity on day 2. However, while proliferation decreased thereafter in the dorsal urethral epithelium to control levels, apoptosis continued to rise into day 3. Thus, the increased apoptotic and proliferative activities in the dorsal urethral epithelia preceded and accompanied hypospadias onset and formation. In contrast, we did not observe any changes in apoptotic or proliferative activity in the sinus cord of DES-treated neonates until day 3, after hypospadias formation. Massive apoptotic cells were not detected in the sinus cord, and the lack of apoptotic activity prior to hypospadias formation suggests that any alteration in sinus cord differentiation by DES may not be contributory to hypospadias onset. Thus, DES-induced alterations in urethral, not sinus cord, cell activity appear to be primarily responsible for hypospadias formation. These observations not only suggest that a DES effect on urethral cells appears to be important in hypospadias induction, but that the mechanistic cellular effects of DES within the sinus cord and urethra are distinct and independent. Nevertheless, further evidence regarding the involvement of urethral vs. sinus cord cells in hypospadias is warranted, and additional DES-induced alterations in cellular functioning of the sinus cord cannot be excluded. As described by Forsberg (1979), ventral vaginal epithelial apoptosis and proliferation were decreased by DES treatment.

Aside from DES, other exogenous estrogens have also been shown to alter sexual development. The synthetic estrogenic chemical BPA causes various abnormalities in mouse reproductive organs (vom Saal et al., 1997), and 150 μg BPA (exposed for 5 days from the day of birth) has also been shown to induce ovary-independent vaginal changes and polyovular follicles (Suzuki et al., unpublished results). Therefore, we examined the affect of BPA on hypospadias formation and a representative vaginal epithelial differentiative event, stratification. The relatively high concentration of BPA (300 μg) we used did not induce hypospadias or alter vaginal epithelial stratification compared to oil, indicating that the effects of BPA may not be specific for these endpoints. However, Laws et al. (2000) reported that environmental estrogens, including BPA, showed different estrogenic activities in different organs, suggesting that alternative BPA exposures could indirectly influence vaginal development. Notably, the effect of in utero exposure to BPA on neonatal UGS differentiation was not investigated in the present study. Thus, the present findings cannot exclude other potential effects of BPA on normal neonatal vaginal development.

DES exposure as low as 0.03 μg induced hypospadias in neonatal mice until about 5 days of age, while a DES dose that was 100 times greater (3.0 μg) was not effective for hypospadias induction if administered beginning on day 7 or later. Although sinus cord differentiation is not yet fully complete until 4–5 weeks of age, just prior to vaginal opening, the state of cellular differentiation achieved by day 7 in the lower reproductive tract may be sufficient to halt hypospadias induction by DES. These findings are consistent with differentiative responses associated with development in the uterus (Cunha, 1976) and indicate that distal dorsal urethral differentiation is essentially complete by about day 5, while vaginal sinus cord differentiation possibly has achieved a developmental state that is tolerant to DES for hypospadias by this age.

Estrogen-induced vaginal epithelial proliferative and differentiative events are dependent on stromal estrogen receptors in the mature animal (Cunha et al., 1985; Buchanan et al., 1998). Although vaginal and urethral morphogenesis are so far not described to be estrogen dependent, estrogen receptors are present throughout the lower neonatal reproductive tract (Sato et al., 1996). Interestingly, we also observed estrogen receptor α immunoreactivity in urethra (data not shown), suggesting that DES could directly affect morphogenesis via the urethral estrogen receptor. Further, normal vaginal epithelial morphogenesis is dependent on mesenchymal function (Boutin and Cunha, 1997). Taken together, these and the present findings raise the possibility that DES may induce hypospadias via alterations in mesenchymal activity. In the present study, DES did not alter stromal proliferative or apoptotic activity, but increased dorsal urethral and sinus cord epithelial apoptosis and decreased sinus cord proliferation by day 3. Normal dorsal urethral and sinus cord epithelial differentiation may occur in response to estrogen-independent mesenchymal activity, and DES may somehow disturb normal mesenchymal-epithelial interactions in these tissues. Since the effect of DES on the mesenchyme of the UGS, and the role of mesenchymal cells in hypospadias formation remain to be clarified, the DES effect in lower reproductive tract development presents an intriguing model system for further study of stromal/mesenchymal-epithelial cell interactions.

In conclusion, the morphogenic events that contribute to normal vaginal and urethral development and function are altered by DES. We have demonstrated here for the first time that DES treatment in neonatal mice induces a series of alterations in dorsal urethral cell function which lead to an apparent invasion by urethral cells into the sinus cord region and result in hypospadias formation by 3 days of age. In addition, we determined that DES is ineffective for hypospadias if administered at or after 7 days of age, indicating differentiative activity has concluded in the dorsal urethral and possibly the ventral vaginal sinus cord regions by this developmental stage. We have characterized the lowest critical dose for hypospadias induction, and demonstrated that the events leading to hypospadias formation involve cellular changes between the dorsal urethral epithelia and sinus cord at a position consistently distal to vaginal lumen formation. In addition, hypospadias formation is associated with alternating apoptotic and proliferative events in dorsal urethral epithelia and sinus cord, suggesting independent DES effects within these cellular regions. Finally, we found that DES can induce a distinct age-dependent shift in the location of hypospadias formation. Thus, DES-induced hypospadias formation in the female mouse is a regional cellular response within the developing distal female reproductive tract.

Acknowledgements

The authors are grateful to Professor Emeritus Noboru Takasugi of Yokohama City University for his valuable advice and critical reading of this manuscript.