2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) Disrupts Early Morphogenetic Events That Form the Lower Reproductive Tract in Female Rat Fetuses

Christopher H. Hurst*, Barbara Abbott{dagger},1, Judith E. Schmid{dagger} and Linda S. Birnbaum{ddagger}

* Curriculum in Toxicology, University of North Carolina, Chapel Hill, North Carolina 27599–7270; {dagger} Reproductive Toxicology Division (MD-67), National Health and Environmental Effects Research Laboratory, ORD, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711; and {ddagger} Experimental Toxicology Division, National Health and Environmental Effects Research Laboratory, ORD, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711

Received August 20, 2001; accepted September 27, 2001


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In female rats, in uteroexposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) during critical periods of organogenesis causes a permanent thread of tissue, consisting of a core of mesenchyme surrounded by keratinized epithelia, across the vaginal opening. The objective of the current study was to determine the earliest time after exposure to TCDD during fetal development that morphological changes in the development of the lower reproductive tract could be detected. In addition, the spatio-temporal expression of several growth factors within the developing reproductive tract was investigated to provide insight into the mechanism of action involved in TCDD-induced vaginal thread formation. Pregnant rats received a single oral dose of 1.0 µg TCDD/kg on gestation day (GD) 15. Dams were sacrificed on GD 17, 18, 19, and 21 and individual reproductive tracts were isolated from female fetuses. As early as GD 18, TCDD produced distinct abnormalities in the female reproductive tract. The width of mesenchyme separating the Mullerian ducts was significantly greater in TCDD-exposed female GD 18 and 19 fetuses and the zone of unfused Mullerian ducts was substantially increased on GD 19 and 21. TCDD induced alterations within the developing reproductive tract in the subcellular and temporal expression of transforming growth factor-ß3 (TGF-ß3) and epidermal growth factor receptor (EGFR). DNA array analysis suggested effects on several genes expressed on GD 18 and 19.

Key Words: 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD); mesenchyme; Mullerian ducts; lower reproductive tract; rat.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Exposure to TCDD and dioxin-like compounds produces a wide variety of effects in a dose-dependent manner in experimental animals, including a wasting syndrome, immunosuppression, thymic atrophy, chloracne, teratogenicity, carcinogenicity, as well as other toxic and biochemical effects (Birnbaum, 1994Go). In addition, TCDD exposure induces reproductive and developmental effects in experimental animals, and the level of exposure required is dependent on species, strain, and target organ. In rats, in utero exposure during critical periods of organogenesis can result in a wide range of adverse effects, including delayed puberty and reduced sperm counts in males and an increased incidence of cystic endometrial hyperplasia, constant estrous, reduced fecundity, and malformations in the external genitalia of females (Gray et al ., 1995Go; Gray and Ostby, 1995Go; Mably et al ., 1992Go).

A single po dose of 1.0 µg TCDD/kg on gestation day (GD) 8 or GD 15 caused structural and functional abnormalities in the female rat reproductive system, including reduced fecundity and the presence of a vaginal thread (Gray et al ., 1995Go; Gray and Ostby, 1995). A dose of 1.0 µg TCDD/kg administered on GD 8 resulted in a 27% incidence of malformations in the external genitalia of the female offspring, however the same dose on GD 15 resulted in a 79% incidence. This indicates that the critical period for this effect is closer to GD 15 (Gray and Ostby, 1995; Gray et al ., 1997Go). In Long Evans rats, a single dose of 1.0 µg TCDD/kg administered on GD 8 resulted in a fetal urogenital tract concentration of 23 pg/g on GD 16 (Hurst et al ., 1998Go). However, administration of 1.0 µg TCDD/kg during a critical period of organogenesis (GD 15) resulted in a urogenital tract concentration of 56 pg/g on GD 16 (Hurst et al ., 2000Go). In the study by Gray and Ostby (1995), no distinction was made between whether the vaginal thread was due to gestational exposure to TCDD, which altered normal fetal development, or whether TCDD interferes with normal vaginal opening postnatally. In a similar study, female offspring of Holtzman rats exposed on GD 15 to TCDD displayed a 36–44% incidence of vaginal thread (Flaws et al ., 1997Go). In addition, they showed that the vaginal thread in TCDD-exposed rats consisted of mesenchyme surrounded by epithelial cells, and that the thread was clearly visible in histological sections as early as PND 2 and was permanent (Flaws et al ., 1997Go). In a similar study, it was reported that gestational exposure to TCDD affected vaginal morphogenesis as early as GD 19 (Dienhart et al ., 2000Go). This indicates that the mechanism responsible for the occurrence of the vaginal thread involves events that occur during gestation or during the early postnatal period, rather than events at puberty during vaginal opening.

Most of the female genital tract develops from the Mullerian ducts, which form the uterine tubes, uterus, cervix, and upper 3/5 of the vagina (Cunha, 1975Go). The lower 2/5 of the vagina is derived from the endodermal urogenital sinus (Cunha, 1975Go; Forsberg, 1973Go). In females, the absence of androgens and Mullerian inhibitory substance (MIS) leads to Wolffian duct regression and Mullerian duct stabilization. The vagina develops by the down-growth of the Mullerian ducts and the fusion of the left and right Mullerian ducts with each other and the urogenital sinus (Mauch et al ., 1985Go).

The objective of this study was to investigate TCDD-induced alterations in the developing reproductive tract to better understand vaginal thread formation. In addition, DNA array analysis was performed to survey the effects of TCDD on gene expression in this tissue as this method permits evaluation of a broad range of potentially affected genes. The information presented in this article may help elucidate the mechanism of action for the TCDD-induced effects on vaginal development in the rat.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals and treatments.
Eight-week-old, timed-pregnant Long Evans rats (200–250 g) were obtained from Charles River Breeding Laboratories (Raleigh, NC) on GD 4 (GD 0, sperm positive). Animals were housed individually in clear-plastic cages with hardwood bedding (Beta Chips, Northeastern Products, Warrensburg, NY). Animals were maintained during pregnancy on Laboratory Rodent Diet (#5001, PMI Feeds, Inc., St. Louis, MO) and filtered tap water ad libitum in a room with a 12:12-h photoperiod and a temperature of 20–24°C with a relative humidity of 40–50%.

Experimental design and treatments.
Rats received a single po dose 1.0 µg TCDD/kg in 5 ml corn oil/kg or an equivalent volume of corn oil vehicle on GD 15. Dams were sacrificed on GD 17, 18, 19, and 21 and individual reproductive tracts were isolated from female fetuses. The number of litters and fetuses examined is reported for each parameter in the respective Results sections.

Histological tissue preparation.
Female GD 17, 18, 19, and 21 fetal reproductive tracts were collected and fixed in Kahle's solution (95% EtOH: 40% formalin: Glacial acetic acid: H2O; 17:6:2:28; v/v). The tissues were stained using a Feulgen nuclear staining protocol as described by Whiting (1950). The stained tissues were embedded in paraffin and serial sections were cut in cross-section (20–30 µm) to examine gross morphology, presence of mitotic figures, and localized patterns of cell death.

For immunohistochemical analysis, tissues from GD 18 and 19 fetuses were dissected, fixed in 3% paraformaldehyde for 30 min, dehydrated in ascending ethanol series from 30 to 100%, and embedded in paraffin. The tracts were positioned so that the external genitalia were cut initially in cross-section and 6 µm sections were made with a Leica Jung Supercut 2065 microtome (Deerfield, IL). Sections were arranged in a matrix so that sections for each gestational age and treatment group were present on each slide with additional sections positioned to the right of that group for use as normal serum controls.

Gross morphology.
All sections (20 or 30 µm thickness) were examined under a Leitz Laborlux D microscope (25x). For each GD, distinct morphological features, referred to as "landmarks," were identified in the caudal to cranial axis of the lower reproductive tract. The landmarks included (1) separation of the urethra from the urogenital sinus, (2) separation of the urogenital sinus into left and right Mullerian/Wolffian duct pairs, (3) appearance of distinct Mullerian and Wolffian ducts, and (4) fusion of Mullerian ducts. Of particular interest was progression of Mullerian duct fusion in the lower reproductive tract. This was assessed by measuring the length of the unfused Mullerian ducts in the vaginal region as well as determining the degree of separation of the left and right Mullerian ducts, which was measured as the width of the intervening mesenchyme. Measurements of mesenchyme thickness were taken in the first section in which the left and right Mullerian ducts were observed to be 2 distinct structures. Since serial sections were cut in a caudal to cranial direction and section thickness was known, measurements of longitudinal distances between these landmarks could be calculated. A schematic of a GD 18 reproductive tract is provided in Figure 1Go, which also shows outlines of sections representing landmark morphology (landmarks 1–4 as described above). Statistical analysis of distance separating left and right Mullerian ducts and comparisons of control and treated were performed on a litter basis using 1-way ANOVAs. The spatial position of GD 17 Mullerian ducts in the cranial-caudal axis and effects of treatment on orientation of the left and right ducts were analyzed using a mixed-effects linear model that looked at the rate of decrease in duct separation along the cranial-caudal axis and included random effects to adjust for variation due to litter and pup (nested within litter). The average separation distances are plotted along the cranial-to-caudal axis. The relative distances between specific morphological features (landmarks) were analyzed on a litter basis for treatment effect using ANOVA and proportional distances on the cranial-caudal axis and are plotted for GD 18, 19, and 21.



View larger version (22K):
[in this window]
[in a new window]
 
FIG. 1. Graphical reconstruction of a GD 18 rat reproductive tract outline of tissue as it appears in histological sections, which was adapted from Mauch et al. (1985). Below each section are the representative "landmark" numbers. In females, the Wolffian ducts degenerate cranially. Notice the region of unfused Mullerian ducts, which is present in control and TCDD-treated fetuses. Four distinct morphological features, referred to as "landmarks," were identified in the caudal to cranial axis of the lower reproductive tract. A description of the landmarks (LM) is as follows: LM 1, separation of the urethra from the urogenital sinus; LM 2, separation of the urogenital sinus into left and right Mullerian/Wolffian duct pairs; LM 3, appearance of distinct Mullerian and Wolffian ducts; and LM 4, Mullerian duct fusion. The relative position of these landmarks within the developing urogenital tract is indicated by an arrow.

 
Immunohistochemistry.
TGF-ß3 was localized with a polyclonal antibody against a peptide of human origin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The antibody was specific for TGF-ß3 and was not cross-reactive with TGF-ß1 or TGF-ß2. The stock solution of TGF-ß3 (200 µg/ml stock) was diluted 1:150 for overnight incubation at 4°C. EGFR was localized with a polyclonal antibody against EGF receptor of human origin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The stock solution of EGFR (200 µg/ml stock) was diluted 1:250 for overnight incubation at 4°C. A polyclonal antibody against laminin (Sigma Chemical Co., St. Louis, MO) that was developed in rabbit using laminin purified from mouse sarcoma cells was used at 1:200 dilution of the 200 µg/ml stock and incubated overnight at 4°C. AhR was localized with a polyclonal antibody against Ah receptor of human origin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The stock solution of Ah-R (200 µg/ml stock) was diluted 1:100 for overnight incubation at 4°C. Specificity of each primary antibody was demonstrated by competitive incubation with AhR, TGF-ß3, or EGFR peptide. Peptide was diluted to give 10x concentration relative to primary antibody. Preincubation of antibody with peptide eliminated the immunostaining patterns on sections (data not shown).

All of the above polyclonal primary antibodies were localized with a biotinylated secondary antibody in conjunction with avidin-biotin-peroxidase complex from a Vector ABC Kit, (Vector Laboratories, Burlingame, CA). The peroxidase substrate was 3,3`-diaminobenzidine (DAB) and gave a red-brown color in the regions having bound primary antibody. Additional sections on each slide were incubated with normal serum alone (no primary antibody) as controls for nonspecific staining. The normal serum was from the same species as that used to prepare the secondary antibody (Vector Laboratories, Burlingame, CA).

Expression of the protein was evaluated under the microscope for patterns and intensity. Epithelial expression was defined as basal or luminal. Mesenchymal protein expression was evaluated in 3 regions: between the urethra and ducts, between the Mullerian and Wolffian ducts, and in the region below the ducts. The level of immunohistochemical staining (protein expression) was rated as high, moderate, low, or not detectable (3, 2, 1, 0, respectively). Scoring was done blindly by 2 independent scorers and the results averaged for each sample. Statistical comparisons were on a litter basis and the average of the litter means and SEM is reported in the data tables. For GD 18 scores, 1-way ANOVAs were used to examine the effect of treatment on score. For GD 19, 2-way ANOVAs were used to adjust treatment effect for any variability between experimental sets (2 different experiments were performed to accumulate immunohistochemical data for GD 19).

DNA array.
Research Genetics Rat GeneFilters, GF300, (Research Genetics, Huntsville, AL), were used to evaluate changes in mRNA expression in the GD 18 and 19 control and TCDD-treated fetal female vaginal tissues. At each time-point, individual urogenital tracts were dissected. The bladder and uterine horns were removed leaving the lower reproductive tract, which forms the vagina. The tissues from 3 litters were pooled for each group and the number of fetuses per pool was as follows: on GD 18 there were 13 control and 14 TCDD-treated; on GD 19 there were 9 control and 10 treated. Total RNA was prepared from each pool of tissue (this means that 1 pool of total RNA was available for each treatment on each GD: 1 for GD 18 control, 1 for GD 18 treated, 1 for GD 19 control, and 1 for GD 19 treated). Total RNA was extracted using TRI® Reagent (Molecular Research Center, Inc., Cincinnati, OH). Tissues were frozen in 50 µl of TRI® reagent in 1.5 ml tubes for use with Kontes Pestles (Fisher Scientific, Raleigh, NC). After grinding, the volume was brought up to 500 µl with Tri reagent and the homogenate was extracted with 100 µl chloroform, the aqueous phase precipitated with isopropanol, pellets washed with 75% ethanol, and resuspended in 100 µl diethylpyrocarbonate (DEPC)-treated water. Radiolabeled probe was prepared using 1 µg total RNA according to the protocols provided by Research Genetics (Huntsville, AL). Briefly, the total RNA was reverse transcribed with Superscript II Reverse Transcriptase (Life Technologies, Gaithersburg, MD) in presence of Oligo dT (Research Genetics), dNTP mixture (Life Technologies), deoxycytidine 5`-triphosphate, ({alpha}:-33P), 10 mCi/ml (NEN LifeScience Products, Inc., Boston, MA), and human cot-1 DNA (Life Technologies). The probe was purified using Bio-Spin Chromatography columns (Bio-Rad Laboratories, Hercules, CA). The GF300 blots were incubated in 5 ml MicroHybTM hybridization buffer (Research Genetics, Huntsville, AL) in roller bottles in a hybridization oven at 42°C for 2–3 h. The purified probe was denatured in a boiling water bath and the entire probe was added to the roller bottles containing the blot and prehybridization buffer. The total radioactivity of the probe added to each roller bottle ranged from 2.25 to 3.99 x 108 dpm per 5 ml of hybridization buffer. Following overnight hybridization, the blots were washed twice at 50°C for 20 min in the bottles with 2X SSC, 1% SDS, and then at room temperature in 0.5X SSC, 1% SDS for 15 min. After 4 days exposure to a Phosphor Screen (Molecular Dynamics, Sunnyvale, CA), the images were captured using a Bio-Rad Molecular Imager FX (Bio-Rad Laboratories, Inc., Hercules, CA) and analyzed with Research Genetics Pathways PC version 2.01 software.

Blots were stripped with boiling 0.5% SDS solution with agitation for 1 h. Examination on the Imager revealed complete removal of all radiolabeled probe. Blots were reprobed up to 3 times and for each hybridization new probe was prepared from the original pools of total RNA (control and treated for GD 18 and 19). This experimental design evaluated consistency and reproducibility of the array data from 3 separate hybridization experiments that included 3 replicates of the GD 18 control and treated total RNA pools (once on new blots and 2 times on stripped blots) and 4 replicates of the GD 19 control and treated total RNA (2 on new blots and 2 times on stripped blots). Statistical analysis of the ratios of treated to control mRNA expression was performed using SAS software (SAS Institute, 1996). The data were examined graphically to determine if there were observable differences related to blot (new vs. stripped) or variability between preparations of probe for each experiment. The ratios of hybridization intensity of treated to control for each gene were transformed to a log (base 10) scale to reduce association of the variance with the mean and so that increase or decrease of the treated sample with respect to the control could be more easily visualized. The mean and standard error of the 3 or 4 replicates of log treated/control (t/c) was calculated for each gene and t-tests were performed, testing if each mean log (t/c) was different from 0 (equivalent to t/c different from 1). The p-values from these tests were used as descriptive indicators of which genes might have quantitatively different expression in the 2 samples. The mean log (t/c) values were ordered and those with the 50 highest and lowest values were included in the tables of significantly affected genes, if the p-values were less than 0.06.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Morphology
Mullerian duct separation and position were evaluated in fetuses collected on GD 17. There was no difference in the mean width of the mesenchyme between the Mullerian ducts (MDs) in control and TCDD-treated fetuses, however, there were subtle differences in the position and length of unfused MDs of TCDD-exposed fetuses. In Figure 2Go, the mean distance between left and right MDs is shown along the anterior-posterior axis, depicting the differences in spatial orientation between control and TCDD-treated (4 fetuses from 3 control litters and 5 fetuses from 3 TCDD-exposed litters). Progressing in a caudal to cranial direction, the MDs in control fetuses were uniformly separated in the unfused zone and fusion occurred within 275 µm. The data were analyzed with a mixed-effects model to look at the regression of the duct separation on the cranial-caudal axis and a slope was estimated for the region of unfused ducts from 60–200 µm (as shown in Fig. 2Go). In other words, the control ductal separation appeared relatively constant in that region, giving an approximately parallel ductal orientation (average slope of the control plot as estimated by the model was –0.13 ± 0.04). However, in fetuses exposed to TCDD, the orientation of the MDs was somewhat different and the ducts appeared to converge in the cranial direction. The treated group regression model estimated the slope of the 60–200 µm region to be –0.38 ± 0.03 and this differed significantly from the control (p < 0.001). The spacing of the MDs within the mesenchymal tissue was such that a profile of the orientation of the ducts was bell-shaped and MD fusion occurred within a shorter distance (~240 µm).



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 2. Spatial orientation of Mullerian ducts on GD 17. Measurements of Mullerian duct separation were made from the section in which 2 distinct Mullerian ducts were present until Mullerian duct fusion. Values represent the mean of 4–5 fetuses. (A) Control. (B) TCDD.

 
Prenatal exposure to TCDD produced distinct abnormalities in the female reproductive tract as early as GD 18. The separation of the left and right MDs was significantly increased in GD 18 and GD 19 treated fetuses relative to controls (Table 1Go). When the relative distance between specific morphological features of the developing vaginal tract was evaluated on the caudal-cranial axis (4 fetuses from 3 control and 4 fetuses from 3 TCDD-treated litters), the length of the vaginal tract from the point of separation from the urethra to the first appearance of the Mullerian and Wolffian ducts (WD; landmark 1–2) appeared to be considerably shorter in TCDD-exposed fetuses (Fig. 3Go). Although the proportional distances were not significantly different on GD 18, TCDD appeared to alter the relative positioning of distinguishable landmarks within the developing reproductive tract.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Distance between Left and Right Mullerian Ducts
 


View larger version (45K):
[in this window]
[in a new window]
 
FIG. 3. Relative length between specific morphological features on the caudal-cranial axis on GD 18. LM 1, urethra-vaginal separation. LM 2, separation of left and right Mullerian ducts (MDs). LM 3, appearance of distinct MDs and Wolffian ducts. LM 4, MD fusion.

 
Similar relative measures of vaginal tract development were performed for GD 19 (4 control fetuses from 3 litters and 4 TCDD-exposed fetuses from 2 litters) and GD 21 (4 control fetuses from 2 litters and 4 TCDD-exposed fetuses from 2 litters). On GD 19, the distance from the initial vaginal separation to the appearance of separate left and right MDs, landmark (LM) 1–2, was significantly reduced (p < 0.01) by TCDD exposure. On GD 19, the length of unfused ducts (LM 3–4, Fig. 4Go) in TCDD-treated fetuses was 68 µm (31% of length measured from the separation of the urethra from the vagina to MD fusion) compared to 38 µm (13%) in control (Fig. 4Go). Similar patterns were seen on GD 21 (Fig. 5Go), as the region of unfused ducts (LM 3–4) was significantly (p < 0.05) increased in treated fetuses (67 µm and 4.7% of the total tract distance in control fetuses vs. 315 µm or 23% in exposed fetuses).



View larger version (48K):
[in this window]
[in a new window]
 
FIG. 4. Relative length between specific morphological features on the caudal-cranial axis on GD 19. Landmark (LM) 1, urethra-vaginal separation. LM 2, separation of left and right Mullerian ducts (MDs). LM 3, appearance of distinct MDs and Wolffian ducts. LM 4, MD fusion. Exposure to TCDD significantly (p < 0.01) reduced the relative distance from initial vaginal separation to appearance of separate MDs (LM 1–2).

 


View larger version (45K):
[in this window]
[in a new window]
 
FIG. 5. Relative length between specific morphological features on the caudal-cranial axis on GD 21. (Note: morphological definitions of landmarks for GD 21 differ slightly from GD 18 and 19). LM 1, urethra and surrounding mesenchyme appear "bell-shaped." LM 2, urethra-vaginal separation. LM 3, appearance of left and right Mullerian ducts. LM 4, Mullerian duct fusion. LM 5, oval vaginal lumen, which is filled with epithelial plug of cells. LM 6, oval vaginal lumen, but epithelial plug is not present. TCDD-exposed fetuses showed significant (p < 0.05) increase in the distance occupied by unfused Mullerian ducts (LM 3–4).

 
The separation of the MDs (width of interductal mesenchyme) was evaluated at the most distal position in which these ducts appeared as unfused and separate from WD (GD 18: 4 fetuses from 3 litters each of control and TCDD-exposed; GD 19: 4 fetuses from 3 control litters and 4 TCDD-exposed fetuses from 2 litters; GD 21: 4 fetuses from 2 litters each of control and TCDD-treated). On GD 18 and GD 19, the width of mesenchyme separating the MDs was significantly (p < 0.05) greater in fetuses exposed to TCDD versus control fetuses (Figs. 6A and 6BGo; Table 1Go). Although the mesenchyme width decreased in control and treated reproductive tracts over time (GD 18, 19, and 21), TCDD-exposed fetuses had more mesenchyme between the MDs relative to controls (Figs. 6A –6FGo; Table 1Go). The Feulgen stained sections permitted detection of regions of cellular proliferation through the presence of mitotic figures (Fig. 6AGo, Inset 1). The cellular debris typical of necrosis and/or programmed cell death could also be observed (Fig. 6AGo, Inset 2). No differences between control and treated tissues were observed in location or extent of either proliferation or cell death during the microscopic evaluations. The vaginal thread may be the retained zone of mesenchyme separating the unfused MDs.



View larger version (178K):
[in this window]
[in a new window]
 
FIG. 6. Feulgen staining of female rat reproductive tracts after GD 15 administration of 1.0 µg TCDD/kg showing width of interductal mesenchyme. Location of the mesenchyme between Mullerian ducts and degree of separation is indicated by double arrow. (A, C, E) GD 18, GD 19, and GD 21 control reproductive tracts, respectively. (B, D, F) GD 18, GD 19, and GD 21 TCDD-treated reproductive tracts. In TCDD-treated fetuses, the mesenchyme width was significantly greater than controls. In Feulgen stained sections, cell proliferation and cell death were detected as mitotic figures (Inset 1) and condensed nuclear chromatin (Inset 2), respectively. Scale bar, 100 µm. M, Mullerian duct; W, Wolffian duct.

 
Immunohistochemistry
The lower reproductive tracts from female rat fetuses from GD 18 and GD 19 were evaluated for expression of protein for AhR, EGFR, TGF-ß3, and laminin. The results are presented for each protein and on each of the different gestation days.

Ah Receptor
In GD 18 controls, AhR was expressed in epithelial and mesenchymal cells; however, the intensity and pattern of expression varied by region and cell type. The epithelial cells of the urethra and WDs displayed prominent nuclear staining (Fig. 7AGo, Inset W). In contrast, in the epithelium of the MDs, AhR was expressed uniformly across the nucleus and cytoplasm (Fig. 7AGo, Inset M). Prenatal exposure to TCDD did not significantly affect AhR protein expression in the epithelium or mesenchyme of the reproductive tract (Table 2Go; GD 18 mean values from 4 control litters with a total of 8 fetuses and 5 TCDD-exposed litters for a total of 8 fetuses). On GD 19, expression of AhR protein was greater in the nucleus compared to the cytoplasm in the epithelia of the urethra, MD, and WD and there was no treatment effect on AhR protein levels or distribution (Table 2Go; GD 19 mean values of 6 control litters with a total of 9 fetuses and 5 TCDD-exposed litters with a total of 10 fetuses).



View larger version (65K):
[in this window]
[in a new window]
 
FIG. 7. (A) Expression of AhR in GD 18 control fetuses localized to epithelia and mesenchyme. The epithelia of the Wolffian ducts (Inset W) displayed prominent nuclear staining, whereas the epithelia of the Mullerian ducts displayed uniform staining across the nucleus and cytoplasm (Inset M). (B) EGFR expression in the GD 18 control was perinuclear in the epithelia of the Mullerian ducts (Inset M) and was evenly distributed between the nucleus and cytoplasm of the Wolffian ducts (Inset W). (C) The GD 18 control fetuses exhibited perinuclear localization of TGF-ß3 in the epithelia of the Mullerian ducts (Inset M), however in the Wolffian ducts, TGF-ß3 was evenly distributed between the nucleus and cytoplasm (Inset W). M, Mullerian duct; W, Wolffian duct. Scale bar 25 µm.

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Immunohistochemical Staining Scores for AhR
 
Laminin
Laminin, a major constituent of basement membranes, was strongly expressed in the basement membranes surrounding the basal epithelial cells of the urethra, MDs, and WDs (Figs. 8A and 8BGo). On GD 18 and GD 19, no significant effects of TCDD treatment were detected on the expression of laminin (Table 3Go; GD 18 mean of 4 control litters, 8 fetuses, and 5 TCDD-exposed litters, 8 fetuses total; GD 19 values from 6 control litters, 11 fetuses, and 5 TCDD-treated litters, 11 fetuses).



View larger version (145K):
[in this window]
[in a new window]
 
FIG. 8. (A, B) Laminin was expressed in GD 19 control and treated fetuses and on GD 19 expression in basement membrane surrounding Wolffian duct appeared to increase in treated tracts (B). (C, D) EGF-R was strongly expressed in epithelial structures as shown for GD 18 of control and treated. (D) TCDD treatment decreased expression of EGF-R in the epithelia of the urethra, Mullerian, and Wolffian ducts. (E, F) TGF-ß3 localized to both epithelial and the mesenchymal cells of control and treated tissues. (F) In TCDD-treated tissues the expression of TGF-ß3 increased relative to controls in the urethra, Mullerian, and Wolffian ducts. U, urethra; M, Mullerian duct; W, Wolffian duct. Scale bar 50 µm.

 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Immunohistochemical Staining Scores for Laminin
 
EGFR
On GD 18, EGFR protein was localized to the epithelia of the urethra, MD, and WD (Figs. 8C and 8DGo); however, the specific subcellular localization of EGFR in these structures differed between MD and WD (Fig. 7BGo). In the urethra, nuclear staining predominated (Fig. 8CGo), while in the MD, EGFR was perinuclear (Fig. 7BGo Inset for M). In the WDs, EGFR was evenly distributed between the nucleus and the cytoplasm (Fig. 7BGo Inset for W). On GD 18, TCDD treatment significantly decreased EGFR protein expression within all of these epithelia (Fig. 8DGo; Table 4Go: GD 18, 4 control litters, 8 fetuses and 5 TCDD-exposed litters, 6 fetuses). However, on GD 19, in TCDD-exposed fetuses, EGFR protein levels were not different in any region (Table 4Go: GD 19, 6 control litters, 11 fetuses and 5 TCDD-exposed litters, 10 fetuses). On GD 19, subcellular localization shifted (compared to GD 18) in the WD to become perinuclear and in the urethra, expression was uniform in the nucleus and cytoplasm.


View this table:
[in this window]
[in a new window]
 
TABLE 4 Immunohistochemical Staining Scores for EGFR
 
TGF-ß3
On GD 18 and GD 19, TGF-ß3 protein localized to both epithelial and the mesenchymal cells; however, the level of expression was much greater in the epithelia (Figs. 8E and 8FGo). Within the urethra, the subcellular localization was predominantly nuclear. This is in contrast to the MDs in which the protein was mainly perinuclear (Fig. 7CGo, Inset M) or the WDs that exhibited uniform expression across cytoplasm and nucleus (Fig. 7CGo, Inset W). On GD 18, TCDD did not significantly alter the level of expression (Table 5Go; GD 18, 4 control litters, 7 fetuses and 5 TCDD-exposed litters, 9 fetuses), however exposure appeared to shift the expression from perinuclear in control MDs to a more uniform nuclear and cytoplasmic pattern in the TCDD-treated tissues (not shown). On GD 19 (Table 5, 6GoGo control litters, 10 fetuses and 5 TCDD-exposed litters, 11 fetuses), there was a significant increase in TGF-ß3 in the TCDD-treated urethra, MDs, WDs, and mesenchyme surrounding these structures (Fig. 8FGo, Table 5Go). In addition, within the MDs, TCDD treatment shifted the subcellular localization from perinuclear in controls to mainly nuclear in the treated samples (not shown).


View this table:
[in this window]
[in a new window]
 
TABLE 5 Immunohistochemical Staining Scores for TGFß3
 

View this table:
[in this window]
[in a new window]
 
TABLE 6 Gene Expression on GD 18 in Response to TCDD
 
Gene Expression
DNA arrays were used to evaluate changes in mRNA expression in the reproductive tract of GD 18 and 19 female fetuses. There were 1692 identified genes on the array membrane, as well as a large number of ESTs (expressed sequence tags) or unidentified genes and control genomic DNA spots, giving a total of 5147 hybridization targets on the membrane. The known genes whose expression was induced or supressed by TCDD treatment, and which were among the 50 highest or lowest t/c ratios, are presented in Tables 6 and 7GoGo, for GD 18 and 19 respectively. Statistical analysis of the ratios of control to treated mRNA expression indicated that gestational exposure to TCDD induced several genes that may be involved in the TCDD-induced vaginal dysmorphogenesis. On GD 18, bone morphogenetic protein was increased in response to TCDD (Table 6Go). TCDD exposure suppressed Smad 4 protein (p = 0.06) and matrix Gla protein, p = 0.06 (Table 6Go). On GD 19, TCDD induced the expression of STAT5b and inhibited genes shown in Table 7Go.


View this table:
[in this window]
[in a new window]
 
TABLE 7 Gene Expression on GD 19 in Response to TCDD
 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
This study shows that prenatal exposure to TCDD produced subtle alterations in the normal development of the lower reproductive tract by 48 h after GD 15 administration. By GD 18, TCDD produced distinct abnormalities in the female reproductive tract, such as altering the spatial orientation and distance between specific morphological features. In addition, TCDD-exposed fetuses contained more mesenchyme between the MDs, which may be a precursor of the vaginal thread. The presence of interductal mesenchyme may prevent MD fusion, which is critical in the formation of the vaginal tract. Immunohistochemical analysis indicated that EGFR and TGF-ß3 may be important in the formation of this genital abnormality. These results are in agreement with those of Flaws and coworkers (1997) in which they reported the formation of the vaginal thread as a prenatal event.

To explain the TCDD-induced vaginal thread formation, we hypothesized that TCDD may be increasing proliferation and/or inhibiting apoptosis. Chromatin staining by the Feulgen stain allowed an assessment of these cellular processes. Although cells undergoing mitosis were clearly visible, an increase in proliferation was not seen in the zone of mesenchyme between the MDs. Furthermore, apoptotic bodies (indicative of cell death) were not seen in the region of unfused MDs in the TCDD-treated fetuses, which may support our hypothesis that TCDD suppresses or inhibits apoptosis. Although appropriate tests for quantitating apoptosis are needed to confirm these observations, our findings suggest that other cellular processes may be involved in the pathogenesis of vaginal thread formation.

One of the most striking characteristics of the vaginal thread was the retained zone of unfused MDs, including the mesenchymal core. This was consistent with observations of Dienhart et al. (2000). During normal MD fusion, mesenchyme removal may involve migration of these cells out of the zone between the left and right MDs. Processes involved in disruption of fusion could include failure of cells to migrate out of the zone or stimuli that recruit cells into the zone. Growth factor expression as well as synthesis and remodeling of extracellular matrix are potential regulators of these processes and members of the TGF-ß3 family are known to regulate such events. For example, TCDD disrupts growth factor expression in the embryonic urinary tract affecting epithelial cell proliferation and matrix protein expression (Abbott and Birnbaum, 1990aGo; Abbott et al ., 1987Go). In developing embryonic tissues, TCDD alters expression of TGF-{alpha}:, EGF, and TGF-ß (Abbott and Birnbaum, 1990aGo,bGo). The TGF-ßs can inhibit the growth of epithelial cells, but can also stimulate or inhibit the proliferation of mesenchymal cells depending on the entire set of expressed growth factors (Sporn et al ., 1994Go).

The present study demonstrated that the epithelial and mesenchymal cells of the developing reproductive tract expressed the AhR with a specific subcellular and time-dependent localization from GD 18–19. TCDD exposure did not significantly alter the AhR protein expression. This is in agreement with the work of Bryant and coworkers (1997) in which TCDD did not produce changes in the concentration of AhR protein or mRNA in embryonic mouse urinary tract. However, other data suggest a down-regulation of AhR due to TCDD exposure and the detection of this response may be dependent on the time after exposure at which the tissues are examined. (Abbott et al ., 1994Go; Pollenz et al ., 1998Go; Prokipcak and Okey, 1991Go; Swanson and Bradfield, 1993Go). Within the developing MDs, there was a shift in the localization of AhR protein from GD 18 to GD 19. On GD 18, the epithelial cells of the MD exhibited equal nuclear and cytoplasmic staining. However, by GD 19, this pattern shifted to a greater expression in the nucleus. A temporal change in the cellular localization of AhR protein may indicate that the AhR plays a critical role in the development of the reproductive system.

Laminin localized predominantly to the basement membranes surrounding epithelial structures. Regression of the MD in male rats is accompanied by apoptosis of epithelial and mesenchymal cells, degradation of the basement membrane, and loss of distinction between the epithelium and the mesenchyme (Trelstad et al ., 1982Go). Our results indicated that the region of unfused MDs was substantially longer and contained more interductal mesenchyme in the TCDD-treated fetuses. In the present study, cell death did not appear to contribute to the removal of the mesenchyme between the MDs or to the fusion of the MDs. In male rats, MD morphogenesis is dependent on the interaction between the epithelia and the mesenchyme (Cunha and Lung, 1979Go), and dissolution of the ductal membrane was shown to be associated with several biochemical events (Hayashi et al ., 1982Go). Transformations of epithelia-to-mesenchyme and mesenchyme-to-epithelia commonly occur in development and this process is believed to have a role in regression of the MD in the male (Trelstad, 1977Go; Trelstad et al ., 1967Go). The remodeling of the ductal system in both male and female requires multiple, complex interactive processes to occur and the potential exists for TCDD to disrupt several of these, including epithelial to mesenchymal transformations, remodeling of the extracellular matrix and basement membrane, and removal of cells via cell migration, proliferation, or cell death. The use of DNA array methodology provided some support for the involvement of remodeling and transformation through detection of altered expression of genes mechanistically involved in these pathways.

Gestational TCDD exposure significantly increased mRNA expression of bone morphogenetic protein (BMP-2; Table 6Go). BMPs are members of the TGF-ß superfamily. TGF-ß family members initiate signaling through interactions with type I/type II TGF-ß receptor complexes (Wrana et al ., 1994Go). In early development, BMPs are involved in the formation of early mesoderm, endoderm, and epidermis (Gerhart, 1999Go). Downstream targets of BMP-2 are Smad 1, 5, and 8, which upon phosphorylation, form a complex and translocate to the nucleus to activate gene transcription (Whitman, 1998Go). On GD 18, TCDD exposure decreased Smad 4 mRNA. Smads interact with activated TGF-ß receptors and are important in TGF-ß signal transduction (Itoh et al ., 2000Go). TGF-ß responsive genes include collagen, fibronectin, and plasminogen activator inhibitor-1 (PAI-1), which are ECM proteins that are critical in tissue remodeling and repair (Roberts et al ., 1992Go). TCDD-induced alterations in these proteins may disrupt basement membrane degradation and/or ECM remodeling, events that may be necessary for fusion of epithelial structures such as the MDs. On GD 19, TCDD increased the expression of STAT5b. Signal transducers and activators of transcription (STATs) are transcription factors that are mediators of cytokine signaling. Activation of STATs results in genes that are important in controlling cell proliferation, survival, differentiation, and development (Bowman et al ., 2000Go). Aberrant STAT signaling in response to TCDD may contribute to cellular transformation by promoting cell cycle progression and/or cell survival.

The power of the DNA array approach allows the evaluation of treatment effects on a very large number of genes across multiple physiological pathways and potentially identifies genes and/or pathways that might not have been revealed without this analysis. However, in the present application this analysis required pooling of multiple dissected specimens to provide sufficient total RNA for each analysis. The reported changes in gene expression should be confirmed using quantitative methods; unfortunately, the RNA and other resources were not available to conduct these studies. However, the DNA array data can be appropriately used to provide guidance for future directions of TCDD mechanistic research. The experiments identified genes and gene families that participate in remodeling pathways during development and the literature regarding the roles of these pathways supports the proposal that these mechanisms are important in vaginal thread formation.

In conclusion, prenatal exposure to TCDD produces subtle effects on the female reproductive tract, which results in persistent vaginal abnormalities, as early as 48 h after exposure. Changes in the spatial and temporal expression of growth factors in response to TCDD may be implicated in the vaginal thread pathogenesis. Careful analysis of DNA array data may provide insight regarding gene pathways involved in the responses to TCDD that ultimately lead to TCDD-induced vaginal thread formation.


    NOTES
 
The research described in this article has been funded in part by the U.S. Environmental Protection Agency Cooperative Training Agreement (ES07126) with the University of North Carolina at Chapel Hill, NC 27599-7270. The contents do not necessarily reflect the views and policies of the EPA, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

1 To whom correspondence should be addressed. Fax: (919) 541-4017. E-mail: abbott.barbara{at}epa.gov. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Abbott, B. D., and Birnbaum, L. S. (1990a). Effects of TCDD on embryonic ureteric epithelial EGF receptor expression and cell proliferation. Teratology 41, 71–84.[ISI][Medline]

Abbott, B. D., and Birnbaum, L. S. (1990b). Rat embryonic palatal shelves respond to TCDD in organ culture. Toxicol. Appl. Pharmacol. 103, 441–451.[ISI][Medline]

Abbott, B. D., Morgan, K. S., Birnbaum, L. S., and Pratt, R. M. (1987). TCDD alters the extracellular matrix and basal lamina of the fetal mouse kidney. Teratology 35, 335–344.[ISI][Medline]

Abbott, B. D., Probst, M. R., and Perdew, G. H. (1994). Immunohistochemical double-staining for Ah receptor and ARNT in human embryonic palatal shelves. Teratology 50, 361–366.[ISI][Medline]

Birnbaum, L. S. (1994). The mechanism of dioxin toxicity: Relationship to risk assessment. Environ. Health Perspect. 102(Suppl.), 157–167.[ISI][Medline]

Bowman, T., Garcia, R., Turkson, J., and Jove, R. (2000). STATs in oncogenesis. Oncogene 19, 2474–2488.[ISI][Medline]

Bryant, P. L., Clark, G. C., Probst, M. R., and Abbott, B. D. (1997). Effects of TCDD on Ah receptor, ARNT, EGF, and TGF-{alpha}: expression in embryonic mouse urinary tract. Teratology 55, 326–337.[ISI][Medline]

Cunha, G. R. (1975). The dual origin of vaginal epithelium. Am. J. Anat. 143, 387–392.[ISI][Medline]

Cunha, G. R., and Lung, B. (1979). The importance of stroma in morphogenesis and functional activity of urogenital epithelium. In Vitro 15, 50–71.[ISI][Medline]

Dienhart, M. K., Sommer, R. J., Peterson, R. E., Hirshfield, A. N., and Silbergeld, E. K. (2000). Gestational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin induces developmental defects in the rat vagina. Toxicol. Sci. 56, 141–149.[Abstract/Free Full Text]

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, 351–362.[ISI][Medline]

Forsberg, J. G. (1973). Cervicovaginal epithelium: Its origin and development. Am. J. Obstet. Gynecol. 115, 1025–1043.[ISI][Medline]

Gerhart, J. (1999). 1998 Warkany lecture: Signaling pathways in development. Teratology 60, 226–239.[ISI][Medline]

Gray, L. E., Jr., Kelce, W. R., Monosson, E., Ostby, J. S., and Birnbaum, L. S. (1995). Exposure to TCDD during development permanently alters reproductive function in male Long Evans rats and hamsters: Reduced ejaculated and epididymal sperm numbers and sex accessory gland weights in offspring with normal androgenic status. Toxicol. Appl. Pharmacol. 131, 108–118.[ISI][Medline]

Gray, L. E., Jr., and Ostby, J. S. (1995). In utero 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) alters reproductive morphology and function in female rat offspring. Toxicol. Appl. Pharmacol. 133, 285–294.[ISI][Medline]

Gray, L. E., Wolf, C., Mann, P., and Ostby, J. S. (1997). In utero exposure to low doses of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) alters reproductive development of female Long Evans hooded rat offspring. Toxicol. Appl. Pharmacol. 146, 237–244.[ISI][Medline]

Hayashi, A., Donahoe, P. K., Budzik, G. P., and Trelstad, R. L. (1982). Periductal and matrix glycosaminoglycans in rat Mullerian duct development and regression. Dev. Biol. 92, 16–26.[ISI][Medline]

Hurst, C. H., Abbott, B. D., DeVito, M. J., and Birnbaum, L. S. (1998). 2,3,7,8-Tetrachlorodibenzo-p-dioxin in pregnant Long Evans rats: Disposition to maternal and fetal tissues. Toxicol. Sci. 45, 129–136.[Abstract]

Hurst, C. H., DeVito, M. J., and Birnbaum, L. S. (2000). Tissue disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in maternal and developing Long Evans rats following subchronic exposure. Toxicol. Sci. 57, 275–283.[Abstract/Free Full Text]

Itoh, S., Itoh, F., Goumans, M. J., and Ten Dijke, P. (2000). Signaling of transforming growth factor-ß family members through Smad proteins. Eur. J. Biochem. 267, 6954–6957.[Abstract/Free Full Text]

Mably, T. A., Bjerke, D. L., Moore, R. W., Gendron-Fitzpatrick, A., and Peterson, R. E. (1992). In utero and lactational exposure of male rats to 2,3,7,8-tetrachlorodibenzo-p-dioxin. 3. Effects on spermatogenesis and reproductive capability. Toxicol. Appl. Pharmacol. 114, 118–126.[ISI][Medline]

Mauch, R. B., Thiedemann, K. U., and Drews, U. (1985). The vagina is formed by downgrowth of Wolffian and Mullerian ducts. Graphical reconstructions from normal and Tfm mouse embryos. Anat. Embryol. 172, 75–87.[ISI][Medline]

Mori, Y., Chen, S. J., and Varga, J. (2000). Modulation of endogenous Smad expression in normal skin fibroblasts by transforming growth factor ß. Exp. Cell Res. 258, 374–383.[ISI][Medline]

Pollenz, R. S., Santostefano, M. J., Klett, E., Richardson, V. M., Necela, B., and Birnbaum, L. S. (1998). Female Sprague-Dawley rats exposed to a single oral dose of 2,3,7,8-tetrachlorodibenzo-p-dioxin exhibit sustained depletion of aryl hydrocarbon receptor protein in liver, spleen, thymus, and lung. Toxicol. Sci. 42, 117–128.[Abstract]

Prokipcak, R. D., and Okey, A. B. (1991). Physiochemical characterization of the nuclear form of the Ah receptor from mouse hepatoma cells exposed in culture to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch. Biol. Biophys. 261, 6352–6365.

Roberts, A. B., McCune, B. K., and Sporn, M. B. (1992). TGF-ß: Regulation of extracellular matrix. Kidney Int. 41, 557–559.[ISI][Medline]

SAS Institute. (1996). SAS/STAT Software: Changes and Enhancements for Release 6.12. SAS Institute, Cary, NC.

Sporn, M. B., Roberts, A. B., Wakefield, L. M., and Assolan, R. K. (1994). Modulation of growth factor expression by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Exp. Clin. Immunogenet. 11, 142–148.[ISI][Medline]

Swanson, H. I., and Bradfield, C. A. (1993). The Ah-receptor: Genetics, structure and function. Pharmacogenetics. 3, 213–230.[ISI][Medline]

Trelstad, R. L. (1977). Mesenchymal cell polarity and morphogenesis of chick cartilage. Dev. Biol. 59, 153–163.[ISI][Medline]

Trelstad, R. L., Hay, E. D., and Revel, J. P. (1967). Cell contact during early morphogenesis in the chick embryo. Dev. Biol. 16, 78–106.[ISI][Medline]

Trelstad, R. L., Hayashi, A., Hayashi, K., and Donahoe, P. (1982). The epithelial-mesenchymal interface of the male rat Mullerian duct: Loss of basement membrane integrity and ductal regression. Dev. Biol. 92, 27–40.[ISI][Medline]

Whiting, A. R. (1950). A modification of the Schmuck-Metz whole-mount technique for chromosome study. Stain Technol. 25, 21–22.[ISI][Medline]

Whitman, M. (1998). Smads and early developmental signaling by the TGF-ß superfamily. Genes Dev. 12, 2445–2462.[Free Full Text]

Wrana, J. L., Attisano, L., Wieser, R., Ventura, F., and Massague, J. (1994). Mechanism of activation of the TGF-ß receptor. Nature 370, 341–347.[ISI][Medline]