| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Center for Animal Biotechnology and Genomics, Department of Animal Science, Texas A&M University, College Station, Texas 77843-2471
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(ER) by nascent endometrial glands and stroma. To test the working hypothesis that estrogen and uterine ER regulate uterine growth and endometrial gland morphogenesis in the neonatal ewe, ewes were treated daily from birth (PND 0) to PND 55 with 1) saline and corn oil as a vehicle control (CX), 2) estradiol-17ß (E2) valerate (EV), an ER agonist, 3) EM-800, an ER antagonist, or 4) CGS 20267, a nonsteroidal aromatase inhibitor. On PND 14, ewes were hemihysterectomized, and the ipsilateral oviduct and ovary were removed. The remaining uterine horn, oviduct, and ovary were removed on PND 56. Treatment with CGS 20267 decreased plasma E2 levels, whereas EM-800 had no effect compared with CX ewes. Uterine horn weight and length were not affected by EM-800 or CGS 20267 but were decreased in EV ewes on PND 56. On PND 14 and PND 56, treatment with EV decreased endometrial thickness but increased myometrial thickness. The numbers of ductal gland invaginations and endometrial glands were not affected by CGS but were lower in EM-800 ewes on PND 56. Exposure to EV completely inhibited endometrial gland development and induced luminal epithelial hypertrophy but did not alter uterine cell proliferation. Exposure to EV substantially decreased expression of ER, insulin-like growth factor (IGF) I, and IGF-II in the endometrium. Results indicate that circulating E2 does not regulate endometrial gland differentiation or development. Although ER does not regulate initial differentiation of the endometrial glandular epithelium, results indicate that ER does regulate, in part, coiling and branching morphogenesis of endometrial glands in the neonatal ewe. Ablation of endometrial gland genesis by EV indicates that postnatal uterine development is extremely sensitive to the detrimental effects of inappropriate steroid exposure.
developmental biology, estradiol, estradiol receptor, female reproductive tract, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the neonatal ewe, pituitary prolactin (PRL), estradiol-17ß E2, and uterine stromal growth factors, including fibroblast growth factors 7 and 10, hepatocyte growth factor, and insulin-like growth factor (IGF) I and IGF-II, with their respective epithelial receptors have been implicated as endocrine and paracrine regulatory systems controlling postnatal ovine endometrial adenogenesis [5, 6, 8]. Recent evidence strongly supports a primary regulatory role for PRL in endometrial gland growth and branching morphogenesis in the neonatal ovine uterus [5, 9]. Expression of both short and long forms of the PRL receptor (PRL-R) is restricted to the nascent glandular epithelium buds on PND 7 and proliferating and developing glandular epithelium from PND 14 to PND 56 [5]. Hypoprolactinemia in neonatal ewes retards endometrial gland development, whereas hyperprolactinemia increases endometrial gland development [9]. In ewes born in the fall, circulating concentrations of E2 are high at birth, increase from PND 7 to a peak on PND 28, and then decline to PND 56 [5]. The developing endometrial glands express abundant levels of estrogen receptor (ER) and progesterone receptor, an estrogen-responsive protein [5]. During this period, the stroma is also ER positive and expresses both IGF-I and IGF-II [6]. The developing glandular epithelium expresses the IGF-I receptor [6], and IGF-I can activate ER in a ligand-independent manner in other model systems [10]. Available evidence in the neonatal ewe supports the working hypothesis that uterine ER, activated by ligand-dependent or ligand-independent mechanisms, regulates endometrial gland morphogenesis in the neonatal ovine uterus [5, 6]. Expression of the ERß gene has not been detected in the neonatal ovine uterus using polymerase chain reaction techniques, in situ hybridization, or immunohistochemical analyses (unpublished results).
The precise roles of circulating E2 and uterine ER in endometrial gland morphogenesis in the neonatal ovine uterus have not been evaluated. Postnatal uterine development is accompanied by expression of ER in both the nascent and developing glandular epithelium and endometrial stroma in rodents, pigs, and sheep [5, 11–15]. Studies in rodents indicate that endometrial adenogenesis is not dependent on the ovary, adrenal gland, estrogen, or uterine ER [16–19]. In contrast, normal endometrial adenogenesis in the neonatal pig is not dependent on the ovary [15] but requires a functional ER system [20]. To test the hypotheses that circulating E2 and uterine ER regulate endometrial adenogenesis in ovine uterus, a study was conducted to determine effects of E2 valerate (EV; an ER agonist), CGS 20267 (a nonsteroidal aromatase inhibitor), and EM-800 (an ER antagonist) on uterine growth and endometrial gland development in the neonatal ewe. The antiestrogen EM-800 is a pure antagonist of ER and ERß [21] and is more potent than ICI 182,870 antiestrogen when injected subcutaneously in the mouse [22].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All experiments and surgical procedures were in accordance with the Guide for the Care and Use of Agriculture Animals in Agricultural Research and Teaching and approved by the University Laboratory Animal Care Committee of Texas A&M University.
Experimental Design
Crossbred Suffolk ewes were mated to Suffolk rams between September and November. Pregnant ewes were maintained according to normal husbandry practices. Ewes used in the following experiments were born in February and March and were randomly assigned (n = 5 ewes/treatment) at birth (PND 0) to receive daily s.c. injections from PND 0 to PND 55 of 1) saline and corn oil vehicle as a control (CX), 2) EV (50 µg/kg body weight [BW] in corn oil, 3) EM-800 (125 µg/kg BW) in saline, or 4) CGS 20267 (125 µg/kg BW) in saline.
The selected dose of EV induces general uterine hyperplasia and precocious glandular epithelium development in neonatal gilts [15, 20, 23]. The EM-800 compound was kindly provided by Fernand Labrie (Universite Lavale, PQ, Canada). The selected dose of EM-800, a pure antagonist of ER and ERß, is equivalent to the amount of ICI 182,780 used to retard endometrial gland development in neonatal pigs [20]. The CGS 20267 (Letrozole; 4,4-[1,2,3-triazol-1yl-methylene] bis-benzonitrite) compound, a highly specific nonsteroidal aromatase inhibitor selected to decrease endogenous estrogen production, was kindly provided by Norvartis Pharma AG (Basel, Switzerland). The selected dose of CGS 20267 was based on results from in vivo studies in baboons [24].
Beginning on PND 1, blood samples were collected every 4 days by jugular venipuncture into Vacutainer tubes (BD, Franklin Lakes, NJ) and then processed to obtain serum or plasma. On PND 14, the right ovarian pedicle was ligated with suture, and the ovary and oviduct were removed. The right uterine horn was ligated with suture above the intercornual ligament, and the anterior portion of the right uterine horn above the ligature was removed, fixed in fresh 4% paraformaldehyde in PBS (pH 7.2) at room temperature for 24 h, and processed for histology. On PND 56, all ewes were weighed, killed, and necropsied. The left ovary was trimmed free of the mesovarium and weighed. The uterus was obtained and trimmed free of the broad ligament, oviduct, and cervix. The entire left uterine horn was dissected from the remaining portion of the right uterine horn, weighed, and measured for length. Sections (
1 cm) from the midportion of the uterine horn were fixed in 4% paraformaldehyde and processed for histology.
Radioimmunoassay
Blood samples for serum were allowed to clot for 1 h at room temperature. Serum was then collected by centrifugation (3000 x g for 30 min at 4°C), removed, and stored at -20°C. Blood samples for plasma were placed on ice immediately after collection. Plasma was then collected by centrifugation (3000 x g for 10 min at 4°C), removed, and stored at -20°C.
Plasma concentration of estradiol was determined by a double-antibody RIA procedure (Ultra-Sensitive estradiol assay, DSL-4800; Diagnostic Systems Laboratories, Webster, TX). Duplicate aliquots of plasma (1 ml) were double-extracted in 5 ml of diethyl ether with the organic layer dried under nitrogen gas. Extraction of tritiated estradiol from 1 ml aliquots of pooled plasma gave an average recovery of 85%. Dried extracts were reconstituted in assay buffer. The RIA used rabbit anti-E2 (polyclonal) serum and iodinated estradiol. The primary antiserum cross-reacts 2.4%, 0.6%, 0.2%, 2.6%, 0.2%, and 3.4% with estrone, estriol, 17-estradiol, 17ß-estradiol-3-glucoronide, estradiol-3-sulfate, and D-equilenin, respectively. The estradiol standard curve (1.76–1600 pg/tube) was prepared from 1,3,5(10)-estratrien-3,17ß-diol (E2; E950, batch H239; Steraloids, Wilton, NH). Goat anti-rabbit gamma globulin serum and polyethylene glycol were used as the precipitating second antibody reagent. The sensitivity of the assay was 3 pg/tube. The intra- and interassay coefficients of variation were 4.3% and 5.4%, respectively. RIA data were analyzed using AssayZap 2.0 software (Biosoft, Cambridge, U.K.).
Histology and Morphometry
After fixation, uterine tissues were changed to 70% ethanol for 24 h and then dehydrated and embedded in Paraplast Plus (Oxford Labware, St. Louis, MO). Uteri were sectioned (5 µm) and stained with hematoxylin and eosin as described previously [3]. Uterine sections (n = 4) from each ewe were photographed, and images were analyzed using Scion Image software (Scion Corporation, Frederick, MD) as described previously [9]. Measurements were standardized using the image of a stage micrometer at the same magnification. The number of superficial ductal invaginations of glandular epithelium from luminal epithelium into the stroma was determined. The criterion for a ductal gland invagination was an invagination of the glandular epithelium into the underlying stroma with a length of at least 15–20 µm that could be visibly tracked to a cross section of a gland. Endometrial gland number was determined by counting the total number of uterine glands in a complete cross section of the uterine horn. A gland cross section with a visible open lumen was counted as a single uterine gland. Endometrial gland density was determined by counting the number of glands in a 200-µm2 area of the stratum compactum and stratum spongiosum of the intercaruncular endometrium. The numbers of ductal gland invaginations and endometrial glands and estimates of gland density were generated for at least three areas within five nonsequential sections from each uterine horn. Intra- and inter-section repeatability estimates for determination of ductal gland invagination number and endometrial gland number by a single observer were 0.85 and 0.8, respectively. In the endometrium of uteri from CX and EV-treated ewes, the thickness or width of the endometrium and myometrium (inner circular and outer longitudinal layers) and the luminal epithelium cell height were measured using the Scion Image software from multiple points (n = 3 or 4) of at least 10 nonsequential uterine sections.
In Situ Hybridization Analysis
Location of mRNA in uterine tissue sections was determined by in situ hybridization as described previously [25]. Deparaffinized, rehydrated, and deproteinated sections (5 µm) of the uterus from each CX and EV-treated ewe were hybridized with radiolabeled sense or antisense cRNA probes generated from linearized plasmid templates using in vitro transcription with [-35S]UTP. Plasmid templates were partial cDNAs for ovine ER [26], ovine IGF-I, IGF-II, and IGF-I receptor [6], and bovine long PRL-R [27]. After hybridization and digestion with RNase A, slides were exposed overnight to BioMax x-ray film (Kodak, Rochester, NY). Slides were then dipped in Kodak NTB-2 liquid photographic emulsion, exposed at 4°C for 1–4 wk depending on signal intensity as judged from autoradiographs, developed in Kodak D-19 developer, counterstained with hematoxylin, dehydrated, and protected with coverslips.
Immunohistochemistry
Immunoreactive proliferating cell nuclear antigen (PCNA) and ER proteins were localized in cross sections (5 µm) of the uterus from each CX and EV-treated ewe using the appropriate mouse antibodies and a Super ABC Mouse/Rat IgG Kit (Biomeda, Foster City, CA) as described previously [5]. Mouse monoclonal antibody to PCNA (M0879, clone PC10) was purchased from DAKO (Carpinteria, CA). Rat monoclonal antibody to human ER (H222) was kindly provided by Dr. Geoffrey Greene (University of Chicago, Chicago, IL). The final working antibody concentration was 2 µg/ml for PCNA and 5 µg/ml for ER. Antigen retrieval utilizing boiling citrate buffer was performed as described previously for PCNA detection [5, 8]. Antigen retrieval using limited pronase digestion was performed as described previously for ER detection [26]. The chromagen used for peroxidase localization was 3,3'-diaminobenzidine tetrahydrochloride (Sigma Chemical Co., St. Louis, MO). Negative controls were performed in which the primary antibody was substituted with the same concentration of purified normal mouse IgG (Sigma). Multiple tissue sections from each ewe were processed as sets within an experiment.
Relative hybridization signal intensity for mRNA expression (IGF-I and IGF-II) and staining intensity for immunoreactive protein expression (PCNA and ER) was assessed visually in uterine sections from each ewe by two independent observers as described previously [6] and scored as follows: absent (-, no staining stronger than that of IgG control), weak (+), moderate (++), or strong (+++). When histologically discernable, intercaruncular endometrial tissues (including luminal epithelium, stroma, and glandular epithelium) and myometrium were scored. The glandular epithelium and stroma was separated into shallow (stratum compactum) and deep (stratum spongiosum).
Photomicroscopy
Representative photomicrographs of tissues analyzed by in situ hybridization or immunohistochemistry were taken using an Eclipse 1000 photomicroscope (Nikon Instruments, Lewisville, TX) fitted with a Nikon DXM1200 digital camera. Digital images were captured and assembled using Adobe Photoshop (Adobe Systems, Seattle, WA).
Statistical Analyses
All quantitative data were subjected to least-squares ANOVA using the general linear models procedures of the Statistical Analysis System [28]. Plasma E2 and serum PRL levels were log transformed prior to least-squares regression analyses. Ovarian weight, uterine horn weight, and uterine horn length data were analyzed using BW as a covariate. Histomorphometrical data were analyzed using an overall model that included main effects of treatment, day, section, and area and the treatment by day interaction. If a treatment by day interaction was detected, data were reanalyzed within day to determine effects of treatment. Preplanned comparisons used to determine treatment effects on a within-day basis were CX versus EM-800, CX versus CGS 20267, and CX versus EV. In all analyses, error terms used in tests of significance were identified according to the expectation of the mean squares for error. Data are presented as least-square means (LSM) of untransformed values with overall SEMs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Circulating concentration of E2 in plasma was affected (P < 0.01) by treatment, day, and their interaction. In CX ewes (Fig. 1), E2 levels were highest between PND 1 and PND 9, declined by PND 13, increased slightly by PND 29, and then declined slightly by PND 56 (cubic effect of day, P = 0.03). Overall, E2 levels in EM-800 ewes were not different (P > 0.10) from those in CX ewes. However, circulating concentrations of E2 were lower (effect of treatment, P = 0.07) in ewes receiving CGS 20267, particularly between PND 5 and PND 13 (quadratic effect of day, P = 0.09). As expected in EV ewes, plasma E2 was higher than that in CX ewes on PND 1 and increased thereafter (quadratic effect of day, P = 0.02).
|
Treatment with EM-800 Antiestrogen Retards Endometrial Gland Morphogenesis
Uterine weight and horn length on PND 56 were not affected (P > 0.10) by EM-800 or CGS treatments (Table 1). Treatment of ewes with EV from birth decreased (P < 0.01) uterine weight, uterine horn length, and ovarian weight on PND 56. Ovarian weight was 200% greater (P < 0.01) in CGS ewes but was not different (P > 0.10) in EM-800 compared with CX ewes. The ovaries from CX and EM-800 ewes contained numerous small vesicular follicles (data not shown). The ovaries of CGS ewes also contained follicles and one or more corpora lutea in three of the five treated ewes. In contrast, the ovaries of EV ewes were small and devoid of vesicular follicles.
|
On PND 14, the intercaruncular endometrium in CX ewes contained tubular and slightly coiled endometrial glands radiating from the uterine lumen into the upper stroma (Fig. 2). The endometrial luminal epithelium appeared to be pseudostratified and columnar. On PND 56, the intercaruncular endometrium contained numerous coiled and branched glands in the intercaruncular endometrium extending from the luminal epithelium through the stroma to the inner circular layer of smooth muscle of the myometrium. In CX ewes, the number of superficial ductal gland invaginations and endometrial glands increased (P < 0.01) between PND 14 and PND 56 (Table 2). Treatment of ewes with CGS 20267 from birth did not result in detectable effects on uterine development or endometrial adenogenesis (Fig. 2). The numbers of ductal gland invaginations and endometrial glands were not different (P > 0.10) in CGS and CX ewes on either PND 14 or PND 56.
|
|
Treatment of ewes with EM-800 from birth altered endometrial gland morphogenesis (Fig. 2). On PND 14, uteri from EM-800 ewes appeared histologically similar to those of CX ewes, except for a slight reduction in coiled endometrial glands in the stratum compactum. The numbers of ductal gland invaginations and endometrial glands were not different (P > 0.01) in EM-800 and CX ewes on PND 14 (Table 2). However, the intercaruncular endometrium of EM-800 ewes on PND 56 tended to have fewer (P < 0.10) endometrial glands that also were less branched, and there were fewer ductal gland invaginations (P < 0.01) than in CX ewes.
Endometrial gland density in the stratum compactum area of the intercaruncular endometrium adjacent to the luminal epithelium was affected by treatment (P < 0.01) but not day (P > 0.10) (Table 2). On PND 14 and PND 56, gland density was not different (P > 0.10) in CX and CGS ewes. In contrast, treatment of ewes from birth with EM-800 decreased (P < 0.01) gland density in the stratum compactum on PND 14 and PND 56. Endometrial gland density in the stratum spongiosum area of the intercaruncular endometrium adjacent to the inner circular layer of myometrium was not affected (P > 0.10) by CGS treatment, but treatment with EM-800 decreased (P < 0.01) endometrial gland density on PND 56.
Treatment with EV Inhibits Uterine Growth and Endometrial Gland Differentiation
Treatment of ewes from birth with EV affected uterine development and endometrial adenogenesis. The endometrium from uteri of EV ewes on both PND 14 and PND 56 did not contain any histologically discernable endometrial glands (Fig. 2). The luminal epithelium of the presumptive intercaruncular endometrial areas appeared hypertrophic, columnar, and more ruffled compared with CX ewes. In uteri from EV ewes, the stroma appeared more compact on PND 14 and was markedly denser on PND 56 compared with CX ewes. The dense endometrial stroma in EV ewes on PND 56 lacked the characteristic stratum compactum and stratum spongiosum layers of the stroma observed for the intercaruncular endometrium of CX ewes.
The effect of EV exposure on different uterine tissues was determined using histomorphometry. Overall, endometrial thickness was affected (P < 0.001) by day, treatment, and their interaction. In CX ewes, endometrial thickness increased between PND 14 and PND 56. In contrast, EV treatment decreased (P < 0.01) endometrial thickness on PND 14 (CX vs. EV, 219 vs. 178 ± 5 µm) and PND 56 (318 vs. 138 ± 5 µm). In CX ewes, myometrial thickness increased (P < 0.05) between PND 14 and PND 56. Treatment with EV increased (P < 0.01) myometrial thickness on both PND 14 (CX vs. EV, 224 vs. 335 ± 11 µm) and PND 56 (291 vs. 355 ± 11 µm). Overall, height of the luminal epithelium was affected by treatment (P < 0.0001), day (P = 0.05), and their interaction (P < 0.01). In uteri from PND 14, luminal epithelium height was greater (P < 0.01) in EV than in CX ewes (CX vs. EV, 2.9 vs. 5.0 ± 0.1 µm). Similarly, treatment with EV from birth increased (P < 0.01) height of the luminal epithelium on PND 56 (CX vs. EV, 2.4 vs. 5.1 ± 0.1 µm).
Treatment with EV Does Not Affect Uterine Cell Proliferation
To determine how EV affects postnatal ovine uterine growth and differentiation, concentration of PCNA protein was determined in uteri from CX and EV ewes (Fig. 3 and Table 3). PCNA is a highly conserved DNA polymerase accessory protein essential for DNA synthesis, is expressed during late G1 and S phases of the cell cycle, and is a marker of cell proliferation [29]. In CX ewes, immunoreactive PCNA was present in all cell types of PND 14 uteri. Highest concentrations of PCNA protein were detected in the luminal epithelium, glandular epithelium, and stroma of the intercaruncular endometrial areas. By PND 56, overall concentrations of PCNA were lower in endometrium of CX ewes. In EV ewes, concentration of PCNA protein was not different on PND 14 than that in CX ewes. On PND 56, PCNA protein concentration was higher in the endometrial luminal epithelium in the presumptive intercaruncular endometrial areas of uteri from EV ewes.
|
|
Treatment with EV Suppresses PRL-R and ER Expression
The PRL-R is a specific and restricted marker of glandular epithelium phenotype in the neonatal and adult ovine uterus [5, 30]. Therefore, PRL-R mRNA expression was measured in CX and EV uteri (Fig. 4). The uteri of many ewes contained darkly pigmented melanocytes that appear white in darkfield photomicrographs. In uteri from CX ewes, PRL-R mRNA was detected only in the nascent budding and tubular glands on PND 14 and only in the coiled and branched glands on PND 56. PRL-R mRNA was absent from the superficial ductal glands in uteri from CX ewes on PND 56 (Fig. 4). In uteri from EV ewes, PRL-R mRNA expression was not detected in any uterine cell type.
|
On PND 14 in CX ewes, expression of ER mRNA and protein was most abundant in the endometrial glandular epithelium, with lower levels in luminal epithelium and stroma (Fig. 4 and Table 3). In contrast, ER mRNA expression was higher in the epithelium and lower in the stroma on PND 56 in CX ewes, whereas ER protein concentrations were lower. In uteri from EV ewes, ER expression was markedly reduced in the endometrial luminal epithelium and essentially absent in stroma and myometrium on both PND 14 and PND 56.
Treatment with EV Reduces Stromal IGF-I and IGF-II mRNA in the Endometrium
In CX ewes, IGF-I mRNA was detected only in the uterine stroma and was predominantly expressed in stroma of the intercaruncular endometrium (Fig. 5 and Table 3). Overall levels of IGF-I mRNA expression increased in uteri from CX ewes between PND 14 and PND 56. In EV ewes, stromal IGF-I mRNA expression was markedly decreased in PND 14 uteri and completely absent in PND 56 uteri as compared with the sense control.
|
Expression of IGF-II mRNA was also detected only in the endometrial stroma and increased between PND 14 and PND 56 in uteri from CX ewes (Fig. 5 and Table 3). In uteri from PND 14 EV ewes, IGF-II mRNA expression was not detected in any uterine cell type. However, IGF-II mRNA expression was detected in the stroma of uteri from PND 56 EV ewes. Compared with uteri from PND 56 CX ewes, the expression pattern of IGF-II mRNA was not as abundant and uniform and appeared patchy in nature. Expression of IGF-I receptor mRNA was detected in all endometrial cell types on both PND 14 and PND 56 and was not affected by treatment with EV (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
25 pg/ml) and ranged from a low of 6 pg/ml on PND 13 to 16 pg/ml on PND 29 in CX ewes. The circulating concentrations of E2 reported here for ewes born in the spring are lower than those previously reported for fall-born ewes [5]. This difference may be associated with season; marked differences that reflected in utero photoperiod were found in circulating concentrations of PRL at birth and from PND 0 to PND 35 in neonatal ewes [31]. Treatment with CGS 20267, a nonsteroidal aromatase inhibitor, decreased circulating concentrations of E2 in plasma. Although circulating levels of E2 were not completely suppressed in CGS ewes, CGS 20267 decreased circulating concentrations of E2 in baboons from 1–2 ng/ml to 0.1 ng/ml within 48–72 h posttreatment [32]. In the present study, ovarian weight was 200% greater in CGS than in CX ewes on PND 56, which can be attributed to the presence of corpora lutea on the ovaries of three of the five ewes treated with CGS 20267. In women with anovulatory infertility, oral administration of 2.5 mg CGS 20267 was effective in inducing ovulation [33]. The ovary of the neonatal ewe normally contains large numbers of growing and vesicular follicles; these numbers increase from birth to PND 28 and then decline significantly by PND 84 [7]. The ovaries of spring-born ewes do not usually contain corpora lutea because these animals are prepubertal and born during the period of seasonal anestrus. Treatment of ewes with CGS 20267 did not affect ovine uterine gland morphogenesis between birth and PND 56. These results do not support our initial working hypothesis that changes in circulating E2 regulate endometrial gland differentiation and morphogenesis in the neonatal ovine uterus. In the ewe, removal of the ovary at birth does not affect uterine gland development on PND 14 [34]. Similarly, the ovary and E2 do not play a role in endometrial gland development in either the rodent or pig. Although circulating estrogens increase between PNDs 9 and 11 in the rat [35], initial postnatal uterine growth and endometrial adenogenesis are independent of both ovary and adrenal gland [16, 18]. In the neonatal gilt, ovariectomy at birth inhibits uterine growth after PND 56 but does not affect genesis of uterine glands or related endometrial morphogenetic events prior to PND 120 [15]. Available evidence indicates that prepubertal uterine development and endometrial adenogenesis is an estrogen-independent process in mammals.
Genesis of endometrial glands in pig [15], rodent [11–14], and sheep [5] uteri involves coordinated changes in epithelial phenotype that are marked by ER expression in nascent and proliferating endometrial glands. In the present study, treatment of neonatal ewes with EM-800, an antagonist of both ER and ERß, did not affect the initial stage of gland genesis between birth and PND 14 that involves budding differentiation of the glandular epithelium from the luminal epithelium and formation of tubules. However, EM-800 treatment retarded the coiling and branching of the endometrial glands that usually occurs between PND 14 and PND 56. Further, histomorphometry indicated that EM-800 treatment prevented the increase in ductal gland invaginations that occurs between PND 14 and PND 56 in CX ewes. These results support a portion of our working hypothesis that expression and functional activation of ER regulate endometrial gland morphogenesis in the ewe, but we modified our hypothesis to indicate a specific role(s) in gland genesis and coiling and branching morphogenesis between PND 14 and PND 56 rather than the initial stages of gland tubulogenesis occurring between birth and PND 14. Daily administration of the antiestrogen ICI 182,870 (125 µg/kg BW) to ewes from birth did not affect endometrial adenogenesis on PND 14 (unpublished results). Similarly, inhibition of ER activation by ICI 182,870 in neonatal rats from PND 10 to PND 14 had no effect on endometrial adenogenesis [19]. Homozygous ER null mice have hypoplastic uteri that contain all characteristic cell types in reduced proportions [17]. In contrast to rodents, ER regulates in part endometrial gland morphogenesis in pigs; administration of ICI 182,780 to gilts from birth retarded endometrial adenogenesis on PND 7 and PND 14 [20]. The lack of involvement of uterine ER in endometrial gland morphogenesis in rodents may be attributed to the lack of tightly coiled and branched endometrial glands in their uteri as compared with the sheep and pig.
Results from the present study indicated that the uterine ER system regulates in part the coiling and branching of the endometrial glands following the initial period of gland genesis that occurs between birth and PND 14 in the ewe. The activation of ER appears to be ligand independent rather than ligand dependent (for review see [10]). Activation of ER by growth factors such as IGF-I involves direct serine phosphorylation by mitogen-activated protein kinase (MAPK) pathways [36]. In the neonatal ovine uterus, both IGF-I and IGF-II mRNAs were expressed in the endometrial stroma and myometrium, with IGF-I predominantly present in the intercaruncular endometrial stroma [6]. IGF-I mRNA is nearly undetectable in ovine endometrium at birth and becomes strikingly abundant in the intercaruncular stroma surrounding the differentiating, developing, and proliferating endometrial glands between PND 7 and PND 56 [6]. In the neonatal ovine uterus, the IGF-I receptor is expressed in all uterine cell types, and phosphorylated extracellular regulated (ERK1 and ERK2) MAPKs are particularly abundant in the endometrial glandular epithelium [6]. Therefore, IGF-I and perhaps IGF-II may regulate endometrial gland coiling and branching morphogenesis via ligand-independent activation of ER. The effects of EM-800 observed in the present study may have been due to inhibition of IGF-mediated activation of ER in the developing endometrial glands.
The present study is the first to demonstrate that inappropriate exposure of the developing neonatal ovine uterus to EV from birth disrupts uterine development and ablates endometrial gland morphogenesis. Transcription of the PRL-R gene is localized exclusively to nascent and developing endometrial glands, where it plays a role in proliferation and development of the glandular epithelium [5, 9]. Therefore, absence of PRL-R mRNA in EV-treated uteri clearly suggests that EV inhibited the genesis of endometrial glands. The effects of EV in the present study were similar to those observed in rodents but opposite those reported for pigs. The same dose of EV administered to gilts from birth to PND 6 or PND 13 increased uterine weight and induced precocious development of endometrial glands [20, 22]. In gilts, thickness of the endometrium, but not the myometrium, was increased by EV treatment. In contrast, treatment of ewes with EV from birth decreased endometrial thickness and increased myometrial thickness on both PND 14 and PND 56. The antiadenogenic effect of EV in the neonatal ovine uterus is similar to that reported for rats; administration of estrogens to neonatal rats during the period of normal gland genesis (PNDs 10–14) induced a dose-related delay in the onset of gland differentiation [37]. Similarly, administration of tamoxifen, a mixed ER agonist/antagonist, to neonatal rats on PNDs 1–5 or PNDs 10–14 elicited a dose-related inhibition of uterine gland genesis that persisted to PND 26 and PND 60, respectively [38]. However, tamoxifen administered on PNDs 20–24, after the age of normal gland genesis in rats, did not alter the number of preexisting glands. The observed differences in effects of inappropriate exposure to estrogen on uteri of sheep and rodents as compared with pigs may be due to differences in effects of ovarian steroids on steroid receptor gene expression in endometrial epithelia and myometrium of the adult uterus [39]. Collectively, studies in domestic and laboratory animals support the hypothesis that during the critical period of postnatal uterine development endocrine disruptors can exert irreversible negative effects on uterine development and in turn on uterine function in the adult.
The mechanism whereby exposure to EV inhibits endometrial adenogenesis is not fully understood. In the present study, treatment with EV from birth almost completely suppressed ER expression in the uterus. Similarly, treatment of ewes with a 19-norprogestin from birth ablated endometrial adenogenesis and suppressed ER expression [8]. However, results from the present study indicate that expression and activation of a functional ER system is not required for endometrial adenogenesis between birth and PND 14. Results from the analyses of PCNA protein concentration indicates that the antiadenogenic effects of EV are not due to inhibition of cell proliferation in the endometrial epithelia or stroma. PCNA concentration in endometrium of EV ewes on PND 56 appeared to be greatest in the subepithelial stroma and was correlated with PCNA staining in the deep stroma surrounding the proliferating and differentiating glandular epithelium during normal uterine development [5]. In developing lung epithelium, bud outgrowth is not accompanied by induction of localized cell proliferation [40] but appears to involve remodeling of the basement membrane [41]. Thus, EV may suppress alterations in the epithelial-mesenchymal interface or extracellular matrix that have been proposed to support initiation of endometrial adenogenesis [2, 34, 42, 43].
Treatment of neonatal rats and mice with EV induces hypertrophy of the endometrial luminal epithelium [19, 44]. Similarly, EV exposure in the present study induced luminal epithelium hypertrophy combined with compaction of the stroma, resulting in a 66% decrease in endometrial thickness by PND 56 in EV ewes. The disorganized, patchy nature of IGF-II expression in the compacted stroma of uteri from PND 56 EV ewes is distinctly different from the temporal and spatial pattern of IGF-II expression during normal development of the intercaruncular endometrial stroma into two layers (the stratum compactum and stratum spongiosum) [5]. Given that the first step in uterine gland development involves differentiation of glandular epithelium and invagination of nascent glands into underlying stroma, the extreme hypertrophic state of luminal epithelium induced by ER agonists such as EV and tamoxifen may prevent invagination of glandular epithelium physically, as a consequence of alterations in cell shape and associated changes in cell-cell, and cell-extracellular matrix relationships that would otherwise support this process [45, 46]. Under such conditions, epithelial cells could be unable to recognize, integrate, and respond normally to cooperative signals that normally drive gland genesis [19, 37, 47]. In the present study, EV treatment disrupted normal patterns of IGF-I and IGF-II expression in the neonatal ovine uterus. IGF-I is expressed in the periglandular stroma, but the IGF-I receptor is expressed only in the endometrial epithelia [6]. IGF-I and IGF-II are considered stromal mediators of epithelial cell function in the uterus and other epitheliomesenchymal organs such as the mammary gland [48, 49]. Therefore, the antiadenogenic effects of EV on the neonatal ovine uterus may be due to inhibition of IGF-I and IGF-II expression.
The process of uterine morphogenesis is governed by a variety of hormonal, cellular, and molecular mechanisms, many of which remain to be defined (for review see [2, 4]). Results of our previous studies supported the hypothesis that E2 and uterine ER comprise a regulatory system to stimulate and maintain endometrial gland morphogenesis in the neonatal ovine uterus [5, 6, 8]. Results from the present study do not support a role of E2 as a regulator of endometrial gland morphogenesis but clearly support the hypothesis that expression and functional activation of ER by ligand-independent mechanisms is required for proper endometrial gland coiling and branching in the neonatal ewe. Results from EV-treated ewes support the hypothesis that epithelial-stromal interactions are important for endometrial gland differentiation. Future studies will be directed at determining the roles of these critical interactions and stroma-derived factors, such as IGF-I and IGF-II, in endometrial gland morphogenesis using the neonatal ewe as a model system.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Thomas E. Spencer, Center for Animal Biotechnology and Genomics, 442 Kleberg Center, 2471 TAMU, Texas A&M University, College Station, TX 77843-2471. FAX: 979 862 2662; tspencer{at}tamu.edu
Received: 31 January 2003.
First decision: 26 February 2003.
Accepted: 1 April 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Hayashi, A. R O'Connell, J. L Juengel, K. P McNatty, G. H Davis, F. W Bazer, and T. E Spencer Postnatal uterine development in Inverdale ewe lambs Reproduction, March 1, 2008; 135(3): 357 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hayashi and T. E. Spencer WNT Pathways in the Neonatal Ovine Uterus: Potential Specification of Endometrial Gland Morphogenesis by SFRP2 Biol Reprod, April 1, 2006; 74(4): 721 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hayashi and T. E. Spencer Estrogen Disruption of Neonatal Ovine Uterine Development: Effects on Gene Expression Assessed by Suppression Subtraction Hybridization Biol Reprod, October 1, 2005; 73(4): 752 - 760. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hayashi, K. D Carpenter, T. H Welsh Jr, R. C Burghardt, L. J Spicer, and T. E Spencer The IGF system in the neonatal ovine uterus Reproduction, March 1, 2005; 129(3): 337 - 347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hayashi, K. D. Carpenter, and T. E. Spencer Neonatal Estrogen Exposure Disrupts Uterine Development in the Postnatal Sheep Endocrinology, July 1, 2004; 145(7): 3247 - 3257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. E. Spencer and F. W. Bazer Uterine and placental factors regulating conceptus growth in domestic animals J Anim Sci, January 1, 2004; 82(13_suppl): E4 - 13. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. D. Carpenter, K. Hayashi, and T. E. Spencer Ovarian Regulation of Endometrial Gland Morphogenesis and Activin-Follistatin System in the Neonatal Ovine Uterus Biol Reprod, September 1, 2003; 69(3): 851 - 860. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
