BIOLOGY OF REPRODUCTION 55, 1221-1230 (1996)
Ontogeny of Estrogen Receptor Messenger Ribonucleic Acid Expression in the Postnatal Rat Uterus' Renata B. Fishman,
23
,
William S. Branham, 3 Randal D. Streck, 3 and Daniel M. Sheehan 3 ,4
Division of Reproductive and Developmental Toxicology, 3 National Center for Toxicological Research, U.S. Food and
Drug Administration, Department of Health and Human Services, Jefferson, Arkansas 72079 Department of Biochemistry,4 University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205 ABSTRACT During the first 2 wk of postnatal life, the rodent uterus undergoes a period of marked growth and differentiation. To further examine the role of the estrogen receptor (ER) in the mediation of uterine development, we analyzed the ontogeny of ER mRNA expression in the postnatal rat uterus using in situ hybridization. ER mRNA was present in the uterine stroma on the day of birth and progressively increased in abundance during the first 2 wk of postnatal life. In contrast, ER mRNA was not detectable inthe luminal epithelium at birth and did not become abundant in this region until postnatal day (P) 7. ER mRNA abundance increased in the luminal epithelium and in the invaginating and fully formed glandular epithelium during the second week of life. At P21 ER mRNA was more abundant in the glandular epithelium than in any other uterine cell type. These results are consistent with, and extend the findings of, previous studies using uterine homogenate binding assays and immunohistochemistry to define ER ontogeny in rodents. Delineation of the temporal and cell-type specific pattern of ER mRNA ontogeny in the postnatal rat uterus furthers our understanding of the molecular basis of both endogenous and exogenous estrogen effects on uterine growth and development. INTRODUCTION The developing rodent uterus undergoes a period of rapid growth and differentiation during the first 2 wk of postnatal life [1]. During this period, uterine weight increases [2, 3], luminal epithelial cells invaginate into the underlying stroma to form glands [4], longitudinal and circular muscle differentiates [1], and estrogen receptor (ER) levels rise substantially [2, 5]. This rapid uterine growth phase coincides with an elevation of serum estradiol levels beginning on postnatal day (P) 9 in the rat [6], and therefore the role of endogenous estrogen and its receptor in uterine growth and differentiation is of considerable interest. Experimental manipulations that interfere with endogenous estrogen interaction with its receptor during this period, including neonatal ovariectomy [3], administration of the antiestrogen ICI 182 780 on P10-14 [7], and creation of ER "knockout" mice [8], all lead to reductions in uterine growth, but do not affect several measures of uterine differentiation including gland genesis. In addition to the role of endogenous estrogen in normal reproductive tract development, exogenous estrogen expoAccepted July 25, 1996. Received May 22, 1996. 'R.B. Fishman was supported by a postdoctoral fellowship administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration. 2 Correspondence: Renata B. Fishman, Division of Reproductive and Developmental Toxicology, HFT-130, National Center for Toxicological Research, 3900 NCTR Road, Jefferson, AR 72079. FAX: (501) 543-7682; e-mail:
[email protected]
sure can cause abnormal development; the most striking examples in humans are the malignant (i.e., vaginal/cervical adenocarcinoma [9, 10]) and benign (i.e., uterine hypoplasia [11]) outcomes associated with fetal diethylstilbestrol (DES) exposure. In humans, the fetal uterus at gestational Day 100 corresponds developmentally to the rat uterus at birth [12]. Treatment of neonatal rats with DES leads to profound uterine hypoplasia, thus providing a model for the human outcome [13, 14]. Additionally, DES and other estrogens can inhibit uterine gland genesis in the rat, demonstrating that developmental exposure to exogenous estrogens leads to abnormalities of both growth and differentiation [15]. Since most estrogenic effects are mediated via estrogen binding to the ER, it is important to determine the ontogeny of ER in the developing uterus. Ontogeny studies using reverse-transcriptase polymerase chain reaction (RT-PCR) reveal that ER mRNA is present in the mouse blastocyst [16]. At fetal Day 17, when the uterus is first discernible in the mouse, ER can be detected in uterine mesenchyme by immunolabeling [17, 18]. In the postnatal rat, binding assays using whole uterine homogenates revealed that ER is present at relatively low levels at P1, but increases to reach maximal prepubertal levels by P10-12 [2, 5]. Because these studies used uterine homogenates, they provide no information on the cell-type distribution of ER. Postnatally, the uterus develops into a compartmentalized structure comprising a heterogeneous cell population that includes luminal and glandular epithelial cells, stromal cells, and circular and longitudinal muscle cells. These cell types differ in their functions in the adult to assure successful pregnancy, and each displays differential responses to hormone treatment. For example, in immature rats, 17,3-estradiol increases mitosis in the luminal epithelium to a greater extent than in either the stromal or muscle compartments [19] and selectively induces hypertrophy of luminal epithelial cells, but not of other cell types [20]. Given the morphological and functional diversity of uterine cell types, ER localization studies are warranted to complement and extend the results of biochemical experiments. Previous immunohistochemical studies have demonstrated both species and strain differences in the ontogenic pattern of ER protein expression in the developing rodent uterus. In mice, ER protein is present in the stroma at birth in both CD-i and BALB/c strains but is not evident in the luminal epithelium until P3-4 in CD-1 mice and not until P6 in BALB/c mice [21, 22]. Recent immunohistochemical studies in the rat (T strain) uterus reported slight ER immunolabeling in the luminal epithelium at P5 and increased labeling thereafter [23]. The postnatal ontogenic pattern of uterine ER mRNA expression has yet to be reported for any rodent species. In the present study, we examined the
1221
1222
FISHMAN ET AL.
FIG. 1. Transverse sections of uteri from rats aged P1 (A, B) and P3 (C,D) hybridized with [35SI-labeled antisense RNA complementary to ER mRNA. A, C) Darkfield illumination; B, D) brightfield illumination. Silver grains, visualized in darkfield as bright spots, indicate hybridization. Darkfield microscopy reveals the presence of ER mRNA in the stromal (s)compartment at P1. Comparison of co-embedded P1 and P3 uteri reveal that ER mRNA abundance in the stroma increases from P1 to P3. Also note that between P1 and P3, there is an increase in the region of stromal hybridization toward the mesometrial (m) region. Arrowheads indicate luminal epithelium. Scale bar = 100 p.m.
ontogeny of ER mRNA expression in the postnatal rat uterus using in situ hybridization. MATERIALS AND METHODS Animals Breeding, housing, and feeding procedures for the Sprague-Dawley rats used here have been previously described [15]. All procedures were conducted within the guidelines of the NIH and the NCTR Institutional Animal Care and Use Committee. In the present study, the day of birth is designated as postnatal day (P) 1. Uteri from female pups aged P1, 3, 5, 7, 9, 11, 13, and 21 were assessed for ER mRNA. The number of pups ranged from 7 to 12 per age group; pups of each age were obtained from a minimum of 5 litters, except pups aged P21, which were obtained from 2 litters. Preparation of Frozen Sections Female pups were killed by decapitation, and uteri were rapidly removed and fixed for 12 h in 1% paraformaldehyde in PBS at 4C. After fixation, uteri were stored in 0.5 M sucrose in PBS containing 0.02% sodium azide at 4°C [24]. Individual uteri from different age groups were then co-embedded in blocks of O.C.T. cryogenic embedding medium
(Miles Inc., Elkhart, IN) in an orientation designed to yield transverse sections. Uteri were co-embedded in groups such that a single block contained uteri from 3-6 animals, each animal being of a different postnatal age (e.g., five blocks were constructed that contained uterine tissue from animals aged P3, 5, 7 , 9, 11, and 13). Blocks were rapidly frozen on dry ice and cryostat (Hacker Instruments, Fairfield, NJ) sectioned at 4 ,um. Sections were mounted onto polylysine/ gelatin-coated glass microscope slides, and the slides were stored at -20 0 C. At least two sections of the block, each containing multiple tissues, were mounted onto each slide. After embedding, each block was treated as a single unit so that at every step (including freezing, sectioning, slide storage, hybridization, autoradiography, developing, and photography), conditions were identical for all uteri within a given block. This method allows more direct comparison of mRNA abundance among uteri from animals of different ages by eliminating procedural variability that occurs whenever uteri from animals of different ages are not co-embedded and are consequently mounted onto different slides. Probes 35
S-Labeled antisense RNA probe complementary to ER mRNA was transcribed from a 244-base pair (bp) fragment of ER cDNA [25] that had been subcloned into a Bluescript
ONTOGENY OF UTERINE ESTROGEN RECEPTOR mRNA
1223
FIG. 2. Higher magnification of the P1 (A, B) and P3 (C,D) uterine sections shown in Figure 1 reveals that, in contrast to that in the stroma (s), ER mRNA is not above background levels in the luminal epithelium (le; arrowheads). In darkfield illumination, H & E-stained cells appear orange from autofluorescence. Visualization of the band of H & E-stained luminal epithelial cells is not obscured by overlying silver grains as is the case for stromal cells. Hybridization with sense probes defines background levels of silver grains (E,F). Scale bar = 40 zm.
plasmid vector (gift of Dr. PJ. Shughrue, Wyeth-Ayerst Research, Philadelphia, PA [26]). The 244-bp fragment of ER cDNA included the 5' untranslated region and the first 27 bp of the encoding region. The 5' untranslated region is thought to have little sequence homology with that of other steroid receptor genes. Labeled RNA probe was generated by incubating linearized plasmid in the presence of ATP, CTP, GTP, [ 5S]UTP (Riboprobe Gemini System II, Promega, Madison, WI), and T3 RNA polymerase (Promega),
and purified on a Sephadex G-50 spin column (BoehringerMannheim, Indianapolis, IN). In Situ Hybridization and Autoradiography All procedures were performed as previously described [24]. Briefly, sections were pretreated to minimize nonspecific hybridization and then hybridized with RNA probes overnight at high stringency (50°C and 50% formamide;
1224
FISHMAN ET AL.
FIG. 3. Transverse sections of uteri from rats aged P5-13 co-embedded in a single block. A, C, E,G, I) Darkfield illumination; B, D, F,H, ) brightfield illumination. Sections are hybridized with 35S-labeled antisense RNA complementary to ER mRNA. Overall ER mRNA increases from P5-P13, and at low magnification this increase is most evident in the stromal (s)compartment. However, ER mRNA abundance in the outer stroma appears to decrease somewhat between P11 and P13. At each age, note that signal intensity is less in the mesometrial (m)region than in the adjacent stroma and musculature (when present). At P11, the luminal epithelium (arrow) adjacent to the mesometrial region demonstrates increased hybridization signal. At P13, glands are clearly evident, and ER mRNA is clearly more abundant in glandular epithelium (ge) than in the surrounding stroma and, in this case, than in the luminal epithelium. At ages when the circular muscle (c) is clearly differentiated (P7-13), ER mRNA is evident in the circular muscle, and levels are comparable to those of the adjacent stroma. In contrast, ER mRNA is at low levels in the longitudinal muscle (I).Arrowheads indicate luminal epithelium. Scale bar = 100 Im.
Boehringer-Mannheim), washed at high stringency in 50% formamide plus 10 mM dithiothreitol (Sigma Chemical Co., St. Louis, MO) in single-strength saline-sodium citrate buffer (SSC [single-strength SSC = 0.15 M sodium chloride, 0.015 M sodium citrate]; Sigma) for 30 min at 50°C,
treated with RNase A (Sigma) to remove unhybridized probe, and further washed in 0.2-strength SSC at 60 0C for 2 h. All solutions used in procedures before, and including, hybridization were treated with diethylpyrocarbonate (DEPC) or made with DEPC-H 2 0. Each slide contained at
ONTOGENY OF UTERINE ESTROGEN RECEPTOR mRNA
1225
FIG. 3. (Continued).
least one section of a block hybridized with antisense probe and one section of a block hybridized with negative control (sense) probe. Rubber cement rings were applied around each section of a block before probe application to prevent cross-contamination between antisense and sense probes. Hybridization with sense RNA probe defined background levels of silver grains. After autoradiography and development of the slides, the sections were counterstained with hematoxylin and eosin (H & E; Sigma). RESULTS Using in situ hybridization, we analyzed the postnatal ontogenic pattern of ER mRNA expression in the rat uterus. Darkfield microscopic examination of rat uterine cross sections hybridized with ER riboprobe revealed the presence of ER mRNA (visualized as bright silver grains) at P1 in the stroma (Fig. 1). Comparison of co-embedded P1 and P3 uteri revealed that the overall abundance of ER mRNA in the stroma increased from P1 to P3. One interesting aspect of this increase was that there was an apparent ontogenic gradient of stromal hybridization signal toward the developing mesometrial region. At P1 there were only background levels of hybridization in the mesometrial region and low or background levels in the stroma adjacent to the mesometrial region but substantial hybridization signal in the remaining stroma. By P3 there was a noticeable increase in stromal hybridization signal progressing toward the mesometrial region. In contrast to that in the stroma,
ER mRNA expression in the luminal epithelium at P1 and P3 was not above background (sense) hybridization levels (Fig. 2). A band of H & E-stained luminal epithelial cells was clearly discernible such that visualization of individual epithelial cells was not obscured by overlying silver grains, as was the case in the stromal cells. We further analyzed the ontogenic pattern of uterine ER mRNA expression during the first two postnatal weeks by examining uterine sections from female rats aged P3, 5, 7, 9, 11, and 13, which had been co-embedded in a single block. Qualitative analysis of the overall pattern and density of silver grains showed that ER mRNA abundance in the stroma increased with age from P3 (not shown) through P11, but that by P13 ER mRNA in the outer stroma was somewhat decreased from P11 levels (Fig. 3). At P11 ER mRNA was more abundant in the stromal compartment subjacent to the luminal epithelium than in the outer stroma, perhaps because of the increased density of stromal cells in this region. At all ages examined, ER mRNA abundance was lower in the mesometrial region of the uterus than in the adjacent stroma and musculature. This regional decrease in hybridization signal in the mesometrial area was evident in both stromal and circular muscle cells. The apparent lateral progression of increased stromal hybridization signal toward the mesometrial region observed from P1 to P3 continued over the first two postnatal weeks. At P11 an area of the luminal epithelium adjacent to the mesometrial region (perhaps developing glandular epithe-
1226
FISHMAN ET AL.
FIG. 4. Higher magnification of the uteri shown in Figure 3 reveals that at P5 (A, B), ER mRNA is present in the stroma (s)but does not appear to be above background (sense; K, L) levels in the luminal epithelium (le; arrowheads). In contrast, by P7 (C,D), ER mRNA is abundant in luminal epithelial cells. Hybridization signal in the luminal epithelium and stroma increases over P9 (E,F) and P11 (G, H). Again, at P13 (I, ), glands are clearly evident, and ER mRNA appears to be more abundant in the glandular epithelium (ge) than in the luminal epithelium. Both epithelial cell types show more intense hybridization signal than the adjacent stroma. Scale bar = 40 Rm.
lium) appeared to demonstrate increased hybridization signal. By P13, hybridization signal was more intense in both the luminal epithelium and in the developing and fully formed glandular epithelium than in the surrounding stroma. At all ages in which circular musculature (identified as a band of spindle-shaped cells with elongated nuclei surrounding the outer stroma) was present (P7 and older), ER
mRNA was evident in the circular muscle. In general, hybridization levels in the circular muscle were comparable to those observed in the adjacent stroma. At P13, when ER mRNA abundance was diminished in the outer, compared to the inner, stroma, the circular muscle was comparable in hybridization signal levels to the adjacent outer stroma. In contrast, ER mRNA was present at relatively low levels in the longitudinal muscle. High-magnification microscopy
ONTOGENY OF UTERINE ESTROGEN RECEPTOR mRNA
1227
FIG. 4. (Continued).
was used to confirm these patterns of hybridization (not shown). At higher magnification, it was evident that, in contrast to the stroma, ER mRNA was not yet consistently above background levels in the uterine luminal epithelium of rats aged P5 (Fig. 4). However, by P7, ER mRNA was relatively abundant in the luminal epithelium. Examination of 12 P7 uteri that had been co-embedded with P5 uteri (allowing direct comparison of the silver grain patterns at the two ages) revealed that 5 of the 12 P7 uteri had silver grains distributed over most cells of the luminal epithelium at a density of silver grains sufficient to obscure covisualization
of H & E-stained epithelial cells. Among the remaining 7 P7 uteri examined, 5 showed that ER mRNA qualitatively appeared to be above background and at a greater level than that observed at P5, and 2 showed signal intensity equivalent to that at P5 (Table 1). This variation probably reflected individual animal differences in development of ER and slight differences in maturation among P7 animals. At P9, P11, and P13, all uteri examined demonstrated the presence of ER mRNA in the luminal epithelium, and the levels increased as uterine development proceeded. By P13 hybridization signal in the luminal and glandular epithelium was more intense than that in the surrounding stroma. At
1228
FISHMAN ET AL.
FIG. 5. Transverse sections of a uterus from a rat aged P21. A) Darkfield illumination; B)brightfield illumination. ER mRNA is more abundant in the luminal epithelium (le; arrowheads) and glandular epithelium (ge) than in the surrounding stroma (s). Hybridization signal is higher in the glandular epithelium than in the luminal epithelium. Scale bar = 50 m.
P21 ER mRNA continued to be more abundant in the epithelium than in the stroma and was most pronounced in the glandular epithelium (Fig. 5). There was a striking increase in mRNA expression in the glandular epithelium relative to the luminal epithelium, despite the fact that luminal and glandular epithelia constitute a continuous sheet one cell thick. In addition, hybridization signal in the circular muscle was comparable to that of the adjacent outer stroma. Nonspecific background labeling as revealed by sense controls was low and uniform across all uterine cell types at each age. These results are summarized in Table 1. DISCUSSION As most estrogen-induced effects are mediated via the ER, it is important to define the ontogeny of ER in the developing rat uterus in order to understand the role of estrogen in both normal and abnormal uterine development. In the study reported here, we used in situ hybridization to show that ER mRNA was already present in the stromal cell compartment at the earliest age examined, P1. In contrast, in the luminal epithelium, ER mRNA expression was
not consistently detectable above background levels at P15 but was relatively abundant by P7. Expression of ER mRNA in the epithelium increased during the second week of life, coincident with the differentiation of luminal epithelium into glandular epithelium. Also, as uterine glands continued to develop, ER mRNA expression in the glandular epithelium was usually higher than in any other uterine cell type. These results demonstrate a complex pattern of ER mRNA expression characterized by developmental stage and cell-type specificity, as well as regional specificity upon comparison of the mesometrial region to the adjacent stroma and/or musculature. In general, these results are consistent with studies that examined the ontogeny of ER protein in rat uterine homogenates by binding assays [2, 5], as well as with immunohistochemical data from mice [21, 22] and rats [23]. This suggests that the ER mRNA changes we observed are responsible for the ontogenic pattern changes in ER binding and protein expression. However, there are slight differences between our results and immunohistochemical reports in the rat by Ohta et al. [23]. From the present study,
TABLE 1. Summary of the ontogeny of ER mRNA expression in the postnatal rat uterus. Uterine cell subpopulations* Age (P) 1 3 5 7 9 11 13 21
Stroma + + +/+ +++ +++ +++ ++
Longitudinal muscle
Luminal epithelium
Glandular epithelium
Circular muscle
-
NP NP NP NP
NP NP + +
NP NP +/-
NP ++ +++ +++
++ ++ +/+
+/+/+++ +/-
+ (n = 5) +/- (n = 5) - (n = 2) ++ +++ ++ ++
+/-
*The number of plus signs (+) corresponds to relative ER mRNA abundance as
estimated by the relative levels of grain density above background (relative numbers should only be compared within a single column to reflect relative ER mRNA abundance over time); minus signs (-) indicate hybridization levels not above background; NP indicates that glands or musculature is not yet present at this age; n = number of P7 uteri co-embedded with P5 uteri.
ONTOGENY OF UTERINE ESTROGEN RECEPTOR mRNA we cannot conclude that ER mRNA abundance is consistently above background levels in the luminal epithelium at P5, whereas Ohta et al. [23] reported slight ER immunolabeling in the luminal epithelium at P5. Our conclusion that ER mRNA is not consistently above background at P5 is a conservative one based on qualitatively comparing the relative abundance of silver grains at P5 to background levels of silver grains. However, it is possible that ER mRNA is present in the luminal epithelium at P5 at very low levels, which would account for the slight immunolabeling that Ohta et al. [23] observed. It is possible that more sensitive methods of mRNA detection, such as in situ PCR, might detect very low levels of ER mRNA in the luminal epithelium at P5 or even earlier. Furthermore, in our study, the day of birth was designated as P1, whereas in the Ohta et al. [23] study, the day of birth was considered to be P0. Therefore, P5 in the Ohta et al. [23] study would be equivalent to our P6, which we did not examine. In addition, differences may be due to the strain of rat used: Ohta et al. [23] used T strain rats, whereas we used Sprague-Dawley CD rats. Strain differences have been noted in ER immunohistochemistry experiments in mice [21, 22]. Another difference between the two studies is that we clearly detected ER mRNA in the stroma on the day of birth, whereas Ohta et al. [23] did not report ER protein in the stroma until one day after birth. This discrepancy could be due to differences in methodology and/or due to the fact that transcription precedes translation. Furthermore, we found that ER mRNA abundance in the stroma clearly increased during the first 2 wk of life, consistent with the ontogenic pattern seen from binding assay studies using uterine homogenates [2, 5]. In contrast, Ohta et al. [23] did not report any changes in the intensity of ER immunolabeling in the stroma during this period. Given the biological variability in ER ontogeny, in terms of species, strain, and individual animal differences, it is necessary to minimize the experimental variables that are not associated with animal heterogeneity. Our method of co-embedding [27] uteri of animals of various ages in the same frozen block allows us to make direct comparisons of relative uterine ER mRNA abundance among animals of different ages. The results of in situ hybridization studies that use co-embedded tissues are not subject to the inherent experimental variability of section and slide preparation, specific activity of probes, hybridization conditions, dipping in photographic emulsion, exposure time, and developing that occurs when uteri of different ages are not co-embedded and are consequently processed on different slides. Much attention has focused on the role of estrogen in reproductive tract development. Early studies by Jost [28] showed that fetal ovariectomy did not prevent sexual differentiation of the female rabbit reproductive tract. In the postnatal rat, serum estradiol levels are elevated beginning on P9 [6], and this corresponds to a period of uterine growth [4]. The results of several studies demonstrate that endogenous estrogens may be more important for postnatal uterine growth than for several measures of differentiation. For example, administration of ICI 182 780, a potent antiestrogen that does not exhibit estrogen agonism, to female rats on P10-14 resulted in uteri that were diminished in weight but were no different from control uteri in the number of uterine glands, epithelial cell height, or general histological appearance [7]. Similarly, removal of endogenous estrogen by combined ovariectomy and adrenalectomy on P6 resulted in decreased uterine weight after P8 but did not
1229
alter differentiation as described above [3]. In ER "knockout" mice, uteri were present but were smaller than uteri of wild-type mice [8]. Like the uteri of neonatally ovariectomized rats or antiestrogen (ICI 182 780)-treated rats, uterine differentiation in ER "knockouts" proceeded normally, since all uterine cell types and glands were present. Taken together, these results suggest that endogenous estrogens of ovarian origin are the primary mediators of uterine growth, but not differentiation, in the second week of life in the rodent. Defining the ontogeny of ER is crucial to understanding the deleterious effects of exogenous estrogen exposure. Although the most striking example of developmental toxicity following estrogen exposure is the collection of adverse outcomes associated with DES exposure in humans [9, 10] and rodents [13, 29], there is increasing evidence that a host of synthetic and environmental estrogens have the capacity to induce abnormalities in reproductive tract development and to impair fertility [30]. Although endogenous estrogens do not appear to be required for early postnatal uterine differentiation, administration of exogenous estrogens during the first 2 wk of life does affect differentiation [13-15]. The presence of ER mRNA during critical periods of reproductive tract development provides a molecular mechanism for exogenous estrogens to cause uterine abnormalities. For example, high levels of ER mRNA expression in luminal and, particularly, glandular epithelial cells preceding and during gland genesis correspond to the time during which gland genesis can be inhibited by estrogen treatment [4, 13, 14]. Exogenous estrogens and antiestrogens not only exert their effects by interacting with the ER protein already present in a developing tissue, they also have been shown to alter ER mRNA levels [7, 31-34]. One long-standing paradox is the inability of the rat uterus to show a full uterotropic response to a single estradiol injection before P14-P15 [3, 35], despite the fact that the present study and others show that ER is present at significant levels during this period [2, 5]. However, daily estradiol injections on P1-5 induce a robust uterotropic response [36], suggesting an estradiol induction of competence (priming) followed by responsiveness. Recent studies demonstrating estrogen-independent activation of ER to a transcriptionally active form by insulin-like growth factor-I [37], and related results with epidermal growth factor [38], suggest the possibility that growth factors may be involved in the acquisition of uterine responsiveness to estradiol on P14-P15 and in the premature responsiveness induced by five daily injections of estradiol. It has traditionally been thought that estrogen exerts its effects by binding to its high-affinity receptor in the nucleus; ER in turn binds to DNA and subsequently initiates a series of changes in gene expression. More recently, there is evidence that estrogens can elicit rapid cellular responses by acting via extracellular, membrane-bound receptors. Watson et al. [39] reported the presence of membranebound ERs on a subpopulation of GH3 pituitary cells that are involved in rapid secretion of prolactin in response to estrogen exposure. It will be interesting to determine what role these receptors might play in both normal and abnormal reproductive tract development. In conclusion, we examined the ontogeny of ER mRNA expression in the postnatal rat uterus using in situ hybridization and found a complex expression pattern across age, cell-types, and regions of the uterus. ER mRNA is present in the stroma at birth but is not highly abundant in the luminal epithelium until the end of the first week of life.
1230
FISHMAN ET AL.
Abundance of ER mRNA increases in the uterine stroma during the first 2 wk of life and in the luminal and glandular epithelium during the second week of life. These results are consistent with studies showing that endogenous estrogens play a role in mediation of uterine growth in the second week of life and provide a molecular mechanism to mediate estrogen toxicity in response to exogenous estrogen exposure. ACKNOWLEDGMENTS The authors gratefully acknowledge Dr. Paul J. Shughrue for providing the ER subclone. We also thank Ms. Rebecca Webb for her expert assistance with cryostat sectioning and Ms. Peggy Webb for technical assistance.
19. 20.
21.
22.
23.
REFERENCES 1. Wiesner BP. The post-natal development of the genital organs in the albino rat. J Obstet Gynaecol Br Emp 1934; 41:867-922. 2. Medlock KL, Sheehan DM, Branham WS. The postnatal ontogeny of the rat uterine estrogen receptor. J Steroid Biochem 1981; 15:285288. 3. Branham WS, Sheehan DM. Ovarian and adrenal contributions to postnatal growth and differentiation of the rat uterus. Biol Reprod 1995; 53:863-872. 4. Branham WS, Sheehan DM, Zehr DR, Ridlon E, Nelson CJ. The postnatal ontogeny of rat uterine glands and age-related effects of 17[3estradiol. Endocrinology 1985; 117:2229-2237. 5. Clark JH, Gorski J. Ontogeny of the estrogen receptor during early uterine development. Science 1970; 169:76-78. 6. Dohler KD, Wuttke W. Changes with age in levels of serum gonadotropins, prolactin and gonadal steroids in prepubertal male and female rats. Endocrinology 1975; 97:898-907. 7. Branham WS, Fishman R, Streck RD, Medlock KL, DeGeorge JJ, Sheehan DM. ICI 182,780 inhibits endogenous estrogen-dependent rat uterine growth and tamoxifen-induced developmental toxicity. Biol Reprod 1996; 54:160-167. 8. Lubahn DB, Moyer JS, Golding TS, Couse JE Korach KS, Smithies O. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA 1993; 90:11162-11166. 9. Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina. Association of maternal stilbestrol therapy with tumor appearance in young women. N Engl J Med 1971; 284:878-881. 10. Herbst AL. Summary of the changes in the human female genital tract as a consequence of maternal diethylstilbestrol therapy. J Toxicol Environ Health Suppl 1976; 1:13-20. 11. Kaufman RH, Binder GL, Gray PM Jr, Adam E. Upper genital tract changes associated with exposure in utero to diethylstilbestrol. Am J Obstet Gynecol 1977; 128:51-59. 12. Hoar RM. Comparative developmental aspects of selected organ systems. II. Gastrointestinal and urogenital systems. Environ Health Perspect 1976; 18:61-66. 13. Medlock KL, Sheehan DM, Nelson CJ, Branham WS. Effects of postnatal DES treatment on uterine growth, development, and estrogen receptor levels. J Steroid Biochem 1988; 29:527-532. 14. Medlock KL, Branham WS, Sheehan DM. Long-term effects of postnatal exposure to diethylstilbestrol on uterine estrogen receptor and growth. J Steroid Biochem Mol Biol 1992; 42:23-28. 15. Branham WS, Zehr DR, Chen JJ, Sheehan DM. Uterine abnormalities in rats exposed neonatally to diethylstilbestrol, ethynylestradiol, or clomiphene citrate. Toxicology 1988; 51:201-212. 16. Hou Q, Gorski J. Estrogen receptor and progesterone receptor genes are expressed differentially in mouse embryos during preimplantation development. Proc Natl Acad Sci USA 1993; 90:9460-9464. 17. Greco TL, Furlow JD, Duello TM, Gorski J. Immunodetection of estrogen receptors in fetal and neonatal female mouse reproductive tracts. Endocrinology 1991; 129:1326-1332. 18. Greco TL, Duello TM, Gorski J. Estrogen receptors, estradiol, and
24.
25. 26.
27.
28. 29. 30. 31.
32. 33.
34.
35. 36. 37.
38.
39.
diethylstilbestrol in early development: the mouse as a model for the study of estrogen receptors and estrogen sensitivity in embryonic development of male and female reproductive tracts. Endocr Rev 1993; 14:59-71. Kirkland JL, LaPointe L, Justin E, Stancel GM. Effects of estrogen on mitosis in individual cell types of the immature rat uterus. Biol Reprod 1979; 21:269-272. Branham WS, Zehr DR, Sheehan DM. Differential sensitivity of rat uterine growth and epithelium hypertrophy to estrogens and antiestrogens. Proc Soc Exp Biol Med 1993; 203:297-303. Yamashita S, Newbold RR, McLachlan JA, Korach KS. Developmental pattern of estrogen receptor expression in female mouse genital tracts. Endocrinology 1989; 125:2888-2896. Bigsby RM, Li AX, Luo K, Cunha GR. Strain differences in the ontogeny of estrogen receptors in murine uterine epithelium. Endocrinology 1990; 126:2592-2596. Ohta Y, Fukazawa Y, Sato T, Suzuki A, Nishimura N, Iguchi T. Effect of estrogen on ontogenic expression of progesterone and estrogen receptors in rat uterus. Zool Sci 1996; 13:143-149. Streck RD, Wood TL, Hsu MS, Pintar JE. Insulin-like growth factor I and II and insulin-like growth factor binding protein-2 RNAs are expressed in adjacent tissues within rat embryonic and fetal limbs. Dev Biol 1992; 151:586-596. Koike S, Sakai M, Muramatsu M. Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Res 1987; 15: 2499-2513. Shughrue PJ, Bushnell CD, Dorsa DM. Estrogen receptor messenger ribonucleic acid in female rat brain during the estrous cycle: a comparison with ovariectomized females and intact males. Endocrinology 1992; 131:381-388. Streck RD, Rajaratnam VS, Fishman RB, Webb PJ. Effects of maternal diabetes on fetal expression of insulin-like growth factor and insulin-like growth factor binding protein mRNAs in the rat. J Endocrinol 1995; 147:R5-R8. Jost A. Problems of fetal endocrinology: the gonadal and hypophyseal hormones. Recent Prog Horm Res 1953; 8:379-418. Bern HA, Jones LA, Mills KT Use of the neonatal mouse in studying long-term effects of early exposure to hormones and other agents. J Toxicol Environ Health Suppl 1976; 1:103-116. McLachlan JA, Korach KS. Introduction: symposium on estrogens in the environment, II. Environ Health Perspect 1995; 103(suppl 7):34. Read LD, Greene GL, Katzenellenbogen BS. Regulation of estrogen receptor messenger ribonucleic acid and protein levels in human breast cancer cell lines by sex steroid hormones, their antagonists, and growth factors. Mol Endocrinol 1989; 3:295-304. Shupnik MA, Gordon MS, Chin WW. Tissue-specific regulation of rat estrogen receptor mRNAs. Mol Endocrinol 1989; 3:660-665. Rosser M, Chorich L, Howard E, Zamorano P, Mahesh VB. Changes in rat uterine estrogen receptor messenger ribonucleic acid levels during estrogen- and progesterone-induced estrogen receptor depletion and subsequent replenishment. Biol Reprod 1993; 48:89-98. Sato T, Ohta Y, Okamura H, Hayashi S, Iguchi T. Estrogen receptor (ER) and its messenger ribonucleic acid expression in the genital tract of female mice exposed neonatally to tamoxifen and diethylstilbestrol. Anat Rec 1996; 244:374-385. Somjen D, Kaye AM, Lindner HR. Postnatal development of uterine response to estradiol-173 in the rat. Dev Biol 1973; 31:409-412. Sheehan DM, Branham WS, Medlock KL, Olson ME, Zehr DR. Uterine responses to estradiol in the neonatal rat. Endocrinology 1981; 109:76-82. Ignar-Trowbridge DM, Pimentel M, Parker MG, McLachlan JA, Korach KS. Peptide growth factor cross-talk with the estrogen receptor requires the A/B domain and occurs independently of protein kinase C or estradiol. Endocrinology 1996; 137:1735-1744. Ignar-Trowbridge DM, Nelson KG, Bidwell MC, Curtis SW, Washburn TF, McLachlan JA, Korach KS. Coupling of dual signaling pathways: epidermal growth factor action involves the estrogen receptor. Proc Natl Acad Sci USA 1992; 89:4658-4662. Watson CS, Pappas TC, Gametchu B. The other estrogen receptor in the plasma membrane: implications for the actions of environmental estrogens. Environ Health Perspect 1995; 103:41-50.
