Mechanism of Estrogen Action: Lessons from the Estrogen Receptor- Knockout Mouse¹

^a Department of Veterinary Biosciences University of Illinois, Urbana, Illinois 61802 ^b Departments of Biochemistry and Child Health University of Missouri, Columbia, Missouri 65211 ^c Department of Anatomy University of California, San Francisco, California 94143

	INTRODUCTION

TOP INTRODUCTION EFFECTS OF ESTROGEN ON... SUMMARY AND CONCLUSIONS REFERENCES

 
Estradiol-17ß (E₂) stimulates uterine and vaginal epithelial proliferation in vivo [1]. E₂ also plays a critical role in other aspects of uterine and vaginal growth and adult function, and it is obligatory for normal epithelial morphogenesis, cytodifferentiation, and secretory activity in these organs. E₂ elicits its effects via the estrogen receptor (ER), which functions as a ligand-activated transcription factor to turn on target genes in E₂-responsive tissues.

For many years it was believed that E₂ acted through a single type of ER. However, the recent discovery [2, 3] of a second receptor, named ERß to distinguish it from the original receptor now termed ER, has indicated that mediation of E₂ action is more complicated than originally thought. ERß, like ER, is a member of the steroid receptor superfamily, and low but detectable levels of ERß mRNA and protein have been detected in the rodent uterus and vagina [3–5]. However, E₂ treatment of ER knockout (ERKO) mice does not produce characteristic estrogenic responses such as vaginal epithelial stratification and cornification, uterine epithelial DNA synthesis, or increases in uterine wet weight and mRNA levels for uterine progesterone receptor, glucose 6-phosphate dehydrogenase, or lactoferrin [6], despite the expression of ERß mRNA in the uterus and vagina of these animals [4]. Therefore, although the role of ERß in the female reproductive tract has not been established, ER seems to be the critical receptor for mediating vaginal and uterine responses commonly associated with E₂ treatment. This review will therefore address only the role of ER in various E₂-induced processes such as epithelial proliferation and differentiation and will not deal with ERß further.

ER is expressed in both epithelial and stromal cells of adult uterus and vagina [7, 8], and it was initially assumed that E₂ effects on epithelium and stroma were mediated directly through ER in these tissue compartments. Several findings not consistent with this seemingly obvious mechanism of E₂ action have appeared in recent years and suggest that the mitogenic effects of E₂ on neonatal uterine and vaginal epithelium, and possibly other estrogenic effects on these epithelia, could be partially or totally mediated by ER-positive stromal cells.

In this review, we will summarize earlier findings suggesting that the mitogenic and possibly other important E₂ effects on uterine and vaginal epithelium could be mediated through stromal ER. We have recently developed a new experimental methodology that allows a more definitive analysis of the respective roles of stromal and epithelial ER in E₂-induced epithelial proliferation and a variety of other epithelial responses to E₂ in the vagina, uterus, and mammary gland. Our data clearly suggest that E₂-induced epithelial mitogenesis is mediated indirectly by stromal ER in female reproductive organs. These results will be presented here, along with a discussion of preliminary results related to the role of stromal and epithelial ER in various other uterine and vaginal differentiative events.

	EFFECTS OF ESTROGEN ON REPRODUCTIVE ORGANS

TOP INTRODUCTION EFFECTS OF ESTROGEN ON... SUMMARY AND CONCLUSIONS REFERENCES

 
Does E₂ Stimulate Epithelial Proliferation through Stromal or Epithelial ER?

The critical finding that E₂ treatment stimulated proliferation in epithelium of the neonatal mouse uterus, which does not express ER, suggested that E₂ might stimulate uterine epithelial mitogenesis indirectly. Cunha et al. [9] demonstrated that ER are not detectable using steroid autoradiography in uterine epithelium (UtE) of neonatal Balb/c mice but are present in uterine stroma (UtS). This apparent lack of UtE ER in the neonatal mouse has been confirmed by other laboratories using immunohistochemical techniques [10, 11]. Despite the apparent lack of epithelial ER, injection of a synthetic estrogen, diethylstilbestrol (DES), doubled the rate of UtE proliferation [8]. Bigsby and Cunha [8] also showed that UtE remained ER negative even after DES stimulation, indicating that DES did not induce epithelial ER and then act through these ER to induce mitogenesis. That neonatal mouse UtE lacks detectable ER but responds mitogenically to E₂ suggests that the mitogenic effects of E₂ on neonatal UtE may be indirectly mediated by ER-positive stromal cells.

Further data consistent with this idea have come from Yamashita et al. [12], who used double-labeling studies to simultaneously examine ER expression and proliferation in neonatal UtE. Using the H-222 antibody against the ER, which has subsequently been shown to recognize only ER [3], they found that the UtE of neonatal CD-1 mice contained both ER-positive and ER-negative cells at 4 days postpartum. Significantly, DES stimulated proliferation in both ER-negative and ER-positive uterine epithelial cells, and the magnitude of proliferation in ER-negative epithelial cells was almost identical to that in ER-positive uterine epithelial cells, again suggesting that E₂ stimulated epithelial mitogenesis indirectly through the stroma.

In addition to these in vivo studies, other studies have shown that E₂ is not mitogenic for isolated uterine or vaginal epithelial cells in vitro [13, 14]. However, when cultured vaginal epithelium (VE) or UtE is recombined with its respective stroma and grafted in vivo, E₂ is again mitogenic for the epithelium [15]. Furthermore, in cocultures of UtS and UtE, E₂ increased epithelial DNA content, an effect not observed in pure epithelial cultures [16]. This further indicates that normal mitogenic responses to E₂ might be stromally mediated.

The evidence that E₂ induced epithelial proliferation indirectly through stromal ER was suggestive, but not definitive. Though neonatal Balb/c UtE lacks detectable ER, these cells could express ER at levels below the limit of detection of present techniques. Such low levels of ER could be sufficient to directly mediate mitogenic effects of E₂ [12]. Therefore, to definitively test the hypothesis that E₂ stimulates epithelial mitogenesis or other processes indirectly through the stroma, another experimental approach was needed that would yield definitive data.

A New Model System to Study the Mechanism of E₂-Induced Effects on Reproductive Epithelium

The advent of gene knockout technology, in which a specific gene can be inactivated or "knocked out" by a variety of techniques, has provided a powerful new tool for examining the role of various genes and their corresponding proteins. The development of the ERKO mouse [17], in which the ER gene has been rendered nonfunctional by gene targeting, provided a unique opportunity to examine the phenotypic and functional alterations in reproductive and other organs in the absence of ER expression. The ERKO mouse has been used to examine the role of ER in angiogenesis [18], in male and female behavior [19, 20], in the vascular injury response in females [21], and in male and female reproductive organs [6, 22–24], as well as in many aspects of development and function of the pituitary [25].

We have recently developed a new experimental system that utilizes tissues from the ERKO mouse to study the mechanism of E₂ action on female genital tract epithelia [26]. The crucial feature of this system involves enzymatically separating and recombining either uterine or vaginal tissue from the ERKO mouse with that of the wild-type ER-positive Balb/c mouse. This tissue separation/recombination technique provides a unique method for experimentally controlling ER status of both stroma and epithelium and allows preparation of tissue recombinations that lack ER in both stromal and epithelial compartments, express ER in either epithelium or stroma, or express ER in both epithelium and stroma. These tissue recombinants are then grafted into host animals and their responsiveness to E₂ is tested. Through analysis of effects of a lack of stromal and/or epithelial ER on E₂ responses such as epithelial mitogenesis or secretory protein production, the role of ER in each tissue compartment can be definitively determined.

E₂ Induces Epithelial Proliferation in the Uterus Indirectly through Stromal ER

The procedures used for the preparation, grafting, and analysis of the tissue recombinants derived from ERKO and Balb/c uteri have been described [26, 27]. Briefly, uteri were removed from adult (90–120 day) ERKO and neonatal (0–3 days old) Balb/c mice. Uterine gland formation in Balb/c mice begins at approximately 1 wk of age [8] and continues throughout the neonatal period. Once glandular invasion of the underlying stroma has become extensive, it is not possible to isolate stromal tissue free of contaminating glandular epithelium, and thus neonatal Balb/c mice were used for all studies described here. ERKO UtE also forms glands [24], but these glands are rudimentary and do not preclude effective separation of the stromal and epithelial components of the ERKO uterus. Therefore, adult ERKO females were used for all experiments in order to maximize the amount of uterine tissue that could be obtained from these animals. We have previously shown that tissue recombinations of uterus and vagina prepared with combinations of neonatal and adult tissue grow well and manifest normal physiological responses [15, 27].

Uteri from neonatal and adult mice were removed and then cut into small pieces, trypsinized, and separated into epithelial and stromal components, as described previously [28]. As shown in Figure 1, the following tissue recombinations were prepared by culturing recombined stroma and epithelium on agar plates overnight (wt, wild type; ko, knockout): 1) wt-S+wt-E, 2) wt-S+ko-E, 3) ko-S+wt-E, and 4) ko-S+ko-E. Grafts were transplanted under renal capsules of intact adult (8–12 wk old) female nude mice. Grafts were grown for approximately 1 mo, and then all hosts were ovariectomized.

View larger version (38K): [in this window] [in a new window]   FIG. 1. Tissue separation/recombination procedure used in experiments to determine the roles of stromal and epithelial ER in E2-induced epithelial responses. Reproductive organs (uterus, vagina, or mammary gland) from neonatal Balb/c (wt) and adult ERKO (ko) mice were trypsinized, and the epithelial (E) and stromal (S) cells were separated. Four types of tissue recombinations were then prepared; these tissue recombinants expressed ER in both epithelium and stroma (wt-S+wt-E), lacked ER in both stroma and epithelium (ko-S+ko-E), or expressed ER in only the epithelium (ko-S+wt-E) or stroma (wt-S+ko-E).

To examine the effects of E₂ on epithelial mitogenesis in the tissue recombinants, 7 days after ovariectomy some hosts were given 100 ng of E₂ while others were given vehicle alone. To determine whether stimulation of epithelial proliferation seen in response to E₂ was mediated through ER, some hosts were given daily s.c. injections of 1 mg/kg of the ER antagonist ICI 182,780 on Days 5–7 postovariectomy and then given E₂ on Day 7 postovariectomy. Epithelial proliferation was then examined 16 h after E₂ or oil treatments by tritiated-thymidine autoradiography [26]. Epithelial labeling index in various tissue recombinants was measured as [³H]thymidine-labeled cells per total cells.

Tritiated-thymidine autoradiography of the various types of tissue recombinants indicated that an ER-negative epithelium can respond mitogenically to E₂ when associated with stroma that expresses ER. Epithelial labeling index in tissue recombinants composed of wt-S+wt-E and wt-S+ko-E was increased several-fold in E₂- versus oil-treated specimens (Fig. 2). In addition, the magnitude of the E₂-induced increase in epithelial mitogenesis in the wt-S+ko-E tissue recombinants was not statistically different from that observed in the wt-S+wt-E tissue recombinants (Fig. 2), indicating that the epithelial mitogenic response was both qualitatively and quantitatively similar in the two groups of tissue recombinants. In contrast, in ko-S+wt-E and ko-S+ko-E uterine tissue recombinants, the tritiated-thymidine epithelial labeling index was low and showed no statistical difference in E₂- versus oil-treated specimens (Fig. 2). Therefore, E₂ does not stimulate epithelial proliferation in tissue recombinants that lack stromal ER, even when epithelial ER are present. Taken together, these results indicate that epithelial ER are neither necessary nor sufficient to mediate a mitogenic response to E₂, while stromal ER are obligatory.

View larger version (33K): [in this window] [in a new window]   FIG. 2. Labeling index (L.I.) of epithelium in uterine tissue recombinants (wt-S+wt-E, wt-S+ko-E, ko-S+wt-E, and ko-S+ko-E). Grafts were grown for 1 mo in female nude mouse hosts, and then hosts were ovariectomized. One week later, hosts were injected with either 100 ng of E2 or vehicle alone. Sixteen hours later, cell proliferation was assessed by tritiated-thymidine autoradiography. One group of hosts receiving E2 was also treated with the estrogen receptor antagonist ICI 182,780 (1 mg/kg) on Days 5–7 postovariectomy.

It was critical to establish that the epithelial proliferation in wt-S+ko-E tissue recombinants was induced by E₂ acting through stromal ER rather than by some other mechanism such as E₂ interacting with another receptor or a nonreceptor-dependent action of E₂. The ability of pretreatment with the specific ER antagonist ICI 182,780 (Fig. 2) to block the proliferative effect of E₂ on epithelium in wt-S+ko-E tissue recombinants shows that the normal mitogenic effect of E₂ in these tissue recombinants is indeed acting through stromal ER.

On the basis of earlier findings and our work described here, we recently proposed [26] the following model to explain estrogenic stimulation of uterine epithelial mitogenesis (Fig. 3). Estrogens bind to ER in uterine mesenchymal/stromal cells and trigger the production of paracrine factors, which then act on UtE to stimulate mitogenesis. The nature of the E₂-induced stromal signal that induces epithelial proliferation is unknown but likely may involve growth factors. A variety of growth factors are produced by stroma from the uterus and other estrogen target organs, and in many cases, epithelial growth factor receptors and/or biological actions of growth factors on uterine epithelial cells have been described. For example, E₂ increases expression of epidermal growth factor (EGF) and its receptor in the uterus [29, 30]. In addition, many in vivo effects of E₂ can be elicited in UtE of ovariectomized mice by exogenous EGF [31]. Recent evidence also indicates that heparin-binding epidermal growth factor, an EGF-like growth factor that normally acts through the EGF receptor, may also be involved in the mediation of the mitogenic effects of E₂ on UtE [32]. Insulin-like growth factor-1 is produced in relatively high amounts in uterus, preferentially in stroma [33]. Insulin-like growth factor-1 is a potent epithelial mitogen that is regulated by E₂ in the uterus [33] and could function as a paracrine mediator of mitogenic effects of E₂ on UtE. Other growth factors such as hepatocyte growth factor (HGF) and keratinocyte growth factor (KGF) could also be involved in uterine stromal-epithelial interactions. HGF and KGF are produced by mesenchymal/stromal tissue and have receptors and/or mitogenic effects on epithelium [34, 35]. In all likelihood, multiple paracrine growth factor pathways are involved in various aspects of stromal-epithelial communication in the uterus and other reproductive organs.

View larger version (43K): [in this window] [in a new window]   FIG. 3. Proposed mechanisms of E2-stimulated epithelial proliferation and differentiation. On the basis of the mitogenic response of ER-negative uterine, vaginal, and mammary epithelium to E2 when associated with their respective ER-positive stromas and the ability of the anti-estrogen ICI 182,780 to antagonize this effect, mitogenic effects of E2 on epithelial proliferation appear to be mediated through stromal ER. In contrast, E2 acts through ER in both stroma and epithelium to induce secretory product production in UtE or to induce normal stratification and initiate cornification in VE.

Is E₂-Induced Uterine Epithelial Secretory Protein Production Mediated by Stromal or Epithelial ER?

In addition to its role as the major mitogen for uterine and vaginal epithelium, E₂ is essential for the production of major uterine epithelial secretory proteins. A number of these proteins have been identified, such as lactoferrin (LF; [36]), complement component C3 [37], Muc-1 [38], and others. The inability of ERKO mice lacking ER to produce LF in response to E₂ treatment [6] indicates that it is normally ER, rather than ERß, that mediates the effects of E₂ on epithelial cells. Though epithelial ER does not appear to be involved in the induction of epithelial proliferation by E₂, some work has indicated that epithelial ER may be obligatory for production of E₂-regulated secretory products by UtE. Yamashita et al. [12] found that in neonatal mice, DES induced production of LF, the most abundant uterine epithelial secretory protein. The UtE of neonatal CD-1 mice contained both ER-positive and ER-negative cells, but LF production was detected only in UtE cells containing ER. These results suggest that epithelial ER expression could be a prerequisite for secretory protein production in UtE and that E₂ effects on UtE secretory proteins may be at least partially direct.

We have recently begun experiments to directly determine the role of epithelial and stromal ER in the production of uterine secretory proteins and their mRNAs [39]. The same four types of uterine tissue recombinants used in the earlier studies of E₂-induced epithelial proliferation were prepared: wt-S+wt-E, wt-S+ko-E, ko-S+ko-E, and ko-S+wt-E. These were grafted into intact female nude mice, which were subsequently ovariectomized and injected daily with oil or E₂ for up to 3 days. Grafts were recovered and used for either LF or ER immunohistochemistry or Northern blotting for LF. Preliminary results indicate that in wt-S+wt-E tissue recombinants from E₂-treated hosts, the epithelium stained intensely for LF, while similar tissue recombinants from oil-treated hosts showed minimal staining. Conversely, LF staining was minimal in wt-S+ko-E tissue recombinants from both oil- and E₂-treated hosts. LF staining was also minimal in all tissue recombinants prepared with ko-S (ko-S+ko-E, ko-S+wt-E) in both oil- and E₂-treated hosts. A strong single 2.6-kilobase band of LF mRNA was detected by Northern blot in wt-S+wt-E tissue recombinants in E₂-treated hosts, but LF mRNA was below the limit of detection in all other tissue recombinants. Similar results have been obtained using another E₂-dependent uterine secretory protein, complement component C3 (unpublished results). These data suggest that both stromal and epithelial ER are required for production of E₂-dependent epithelial secretory proteins such as LF and C3, in contrast to E₂-induced epithelial mitogenesis, which requires only stromal ER (Fig. 3).

Role of Epithelial and Stromal ER in Mediating E₂ Effects on VE

Proliferation, stratification, and cornification of VE are regulated by E₂. We are presently using the tissue separation/recombination methodology described above for the uterus with vaginal tissue to determine the respective roles of epithelial and stromal ER in these E₂-induced events [40]. ERKO and Balb/c mice were used to prepare the following vaginal tissue recombinants (Fig. 1): wt-S+wt-E, wt-S+ko-E, ko-S+ko-E, and ko-S+wt-E. These were grown as subrenal capsule grafts, and their responsiveness to E₂ was tested as described above for uterine grafts. E₂-treated vaginal tissue recombinants prepared with wt-S (wt-S+wt-E and wt-S+ko-E) showed similar large increases in epithelial labeling index over oil-treated controls, indicating that E₂ stimulates epithelial proliferation despite the lack of epithelial ER in wt-S+ko-E tissue recombinants; these results are similar to those previously observed using uterine tissue recombinants and indicate that the mediation of epithelial mitogenesis by stromal ER may be a phenomenon common to other organs where E₂ stimulates epithelial proliferation. Epithelial labeling index was low in all grafts prepared with ko-S, again confirming our uterine results, indicating that epithelial ER were not sufficient to allow an epithelial mitogenic response to E₂ in the absence of stromal ER [40].

E₂ treatment induced extensive epithelial cornification in wt-S+wt-E grafts, while in all other tissue recombinants the epithelium failed to cornify. E₂-treated wt-S+wt-E vaginal tissue recombinants had 9–14 layers of epithelial cells while epithelial thickness was greatly reduced in wt-S+ko-E tissue recombinants (4–6 layers); epithelium was atrophic (2–3 layers) in all other tissue recombinants. These results indicate that E₂ induction of vaginal, as well as uterine, epithelial proliferation is a paracrine event requiring ER in the stroma. Conversely, both epithelial and stromal ER are required for normal E₂-induced epithelial stratification and cornification (Fig. 3). The obligatory role for epithelial ER in mediating normal E₂-induced vaginal epithelial stratification and cornification, as well as the E₂-induced production of uterine epithelial secretory products discussed in the preceding section, are the first known functions attributed to epithelial ER in vivo. Furthermore, the demonstration that both stromal and epithelial ER are necessary for cornification and normal epithelial stratification in the vagina, as well as for the production of uterine epithelial secretory proteins such as LF, represents the first time any epithelial response to E₂ has been shown to require ER in both the stromal and epithelial compartments [40].

Mesenchymal-Epithelial Recombination Studies Elucidate a Role for Mesenchymal ER in Mammary Ductal Growth

The mammary fat pad, along with the mammary mesenchyme, has been thought to play a critical role in mammary epithelial development, and the mammary fat pad appears to be the tissue responsible for inducing morphogenesis of the ductal pattern typical of the mammary gland (reviewed in [41]). To explore the role of estrogen-dependent paracrine cell-cell interactions in the mammary gland, new techniques were developed for constructing mammary tissue recombinants with postnatal mammary tissues. For this purpose, all four possible tissue recombinants were prepared with mammary fat pad (FP) and mammary epithelium (E) from wild-type and ERKO mice (wt-FP+wt-E, wt-FP+ko-E, ko-FP+ ko-E, ko-FP+wt-E) as represented in Figure 1. After overnight culture to allow the tissues to adhere, the mammary tissue recombinants were grafted under the renal capsules of athymic female nude mice and allowed to grow for 4 wk. Harvested tissue recombinants were fixed, prepared as whole mounts, and stained with hematoxylin. As opposed to procedures in the uterine and vaginal studies described above, the developmental endpoints assessed in the whole mounts were total epithelial mass and ductal branching morphogenesis. Tissue recombinants composed of wild-type stroma and epithelium (wt-FP+wt-E) contained an extensively branched ductal network that completely filled the fat pad, while ductal growth was meager in tissue recombinants composed solely of ERKO tissues (ko-FP+ko-E). When ERKO mammary epithelium was grown in association with wild-type fat pad (wt-FP+ko-E), the ERKO mammary epithelium underwent extensive ductal growth and branching morphogenesis. Although there was some variation in the amount of ductal growth between individual tissue recombinants, ductal growth and branching morphogenesis was always more extensive in tissue recombinants composed of wt-FP+wt-E or wt-FP+ko-E versus ko-FP+ko-E. Strikingly, in tissue recombinants composed of ko-FP+wt-E, ductal tissue could never be recognized in whole mounts. Examination of serial sections of these ko-FP+wt-E tissue recombinants indicated that all (n = 30) did indeed contain small numbers of mammary ducts. Thus, the results of the mammary gland study are in complete agreement with the uterine and vaginal studies and indicate a critical role of stromal ER in estrogen-dependent epithelial proliferation. On the basis of previous studies with ovariectomized pubertal mice and more recent observations on intact ERKO mice, it is evident that estrogen is a morphogen acting through the ER to drive mammary ductal growth and ductal branching morphogenesis. The ERKO/wild-type tissue recombinant studies provide strong evidence that estrogen elicits ductal growth and branching morphogenesis in the mammary gland through paracrine mechanisms in which stromal ER is the key estrogen target [42].

	SUMMARY AND CONCLUSIONS

TOP INTRODUCTION EFFECTS OF ESTROGEN ON... SUMMARY AND CONCLUSIONS REFERENCES

 
The importance of the interaction of epithelium with stroma or its embryonic/fetal precursor, mesenchyme, in the differentiation, growth, and morphogenesis of many organs has been well documented [43]. The results discussed here emphasize that stromal ER and stromal-epithelial interactions also play a critical role in epithelial responses to E₂ in female reproductive organs. A complete understanding of the mechanism of E₂ action on the epithelium in these organs will necessitate an elucidation of how stroma normally communicates with epithelium and how this pattern of communication is altered by E₂ binding to stromal ER. The experimental system described here provides a potentially valuable tool to investigate this question and may be useful in furthering our understanding of the mechanism by which E₂ regulates epithelial mitogenesis, secretory protein production, and other E₂-regulated epithelial processes such as apoptosis and the regulation of the epithelial progesterone receptor.

The data shown here indicate that stromal ER appears to be the sole mediator of E₂-induced epithelial mitogenesis in the uterus, vagina, and mammary gland. However, other epithelial responses to E₂, such as epithelial secretory protein production in the uterus and epithelial cornification and full stratification in the vagina, are more complex in that they appear to require both stromal and epithelial ER. Therefore, there does not appear to be a universal pattern in which E₂ actions on uterine or vaginal epithelium are mediated directly through the epithelial ER or indirectly through the stromal ER. Rather, the vast number of epithelial effects induced by E₂ probably represent a combination of effects mediated by stromal ER alone and by stromal and epithelial ER working in concert, and quite possibly some processes that involve only direct and exclusive action through epithelial ER.

The tissue separation/recombination methodology utilized here is an adaptation of a methodology previously used with male reproductive tissue derived from the naturally occurring testicular feminization mouse, which lacks functional androgen receptor. These experiments demonstrated that androgen-induced epithelial proliferation in the male reproductive tract was mediated through stromal androgen receptors (reviewed in [44]), and provided both the methodological and conceptual framework for the present experiments in that they suggested the possibility that mitogenesis induced by sex steroids in reproductive organs might occur via paracrine mechanisms.

In addition to this work on androgen-induced proliferation in male reproductive epithelia, recent experiments utilizing tissue from the progesterone receptor knockout mouse [45, 46] have indicated that the inhibitory effect of progesterone on estrogen-induced epithelial proliferation is mediated through stromal progesterone receptors (unpublished results). Taken together with our data presented here, these findings suggest that epithelial growth regulation by androgens, estrogens, and progestins in male and female reproductive organs may occur through common paracrine mechanisms mediated by stromal hormone receptors.

An increased understanding of how stromal-epithelial interactions are involved in E₂-induced epithelial responses may have clinical relevance. Cancers of the breast, uterine endometrium, and other female reproductive organs are serious human health problems, and E₂ is involved at least as a permissive agent in the initiation and progression of these diseases. Epithelial cells from these tumors frequently show direct mitogenic responses to E₂ in vitro, in contrast to primary epithelial cultures of these organs, where E₂ is not mitogenic. When viewed in light of our present results, it is likely that the tumor cells have acquired the capacity to directly respond mitogenically to E₂, a capacity not expressed by normal epithelium from these organs. This transition may represent a crucial step in carcinogenesis, and understanding the mechanism of this process may yield important insights into the etiologies of these diseases.

	FOOTNOTES

 
¹ This work was supported by NIH grants HD 29376, AG 15500 (to P.S.C.), ES 08272 (to D.B.L.), and CA 05388 and AG 13784 (to G.R.C.).

² Correspondence: Paul Cooke, Department of Veterinary Biosciences, 2001 S. Lincoln Ave., University of Illinois, Urbana, IL 61802. FAX: (217) 244–1652; p-cooke{at}uiuc.edu

Accepted: March 26, 1998.

Received: January 23, 1998.

	REFERENCES

TOP INTRODUCTION EFFECTS OF ESTROGEN ON... SUMMARY AND CONCLUSIONS REFERENCES

 

1. Galand P, Leroy F, Chretien J. Effect of oestradiol on cell proliferation and histological changes in the uterus and vagina of mice. J Endocrinol 1971; 49:243–2521.[Medline]
2. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-A. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 1996; 93:5925–5930.[Abstract/Free Full Text]
3. Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson J-A. Comparison of ligand binding specificity and transcript tissue distribution of estrogens receptors and ß. Endocrinology 1997; 138:863–870.[Abstract/Free Full Text]
4. Couse JF, Lindzey J, Grandien K, Gustafsson J-A. Tissue distribution and quantitative analysis of estrogen receptor- (ER) and estrogen receptor-ß (ERß) messenger ribonucleic acid in the wild-type and ER-knockout mouse. Endocrinology 1997; 138:4613–4621.[Abstract/Free Full Text]
5. Saunders PTKL, Maguire SM, Gaughan J, Millar MR. Expression of oestrogen receptor beta in multiple rat tissues visualized by immunohistochemistry. J Endocrinol 1997; 154:R13-R16.
6. Couse JF, Curtis SW, Washburn TF, Lindzey J, Golding TS, Lubahn DB, Smithies O, Korach KS. Analysis of transcription and estrogen insensitivity in the female mouse after targeted disruption of the estrogen receptor gene. Mol Endocrinol 1995; 9:1441–1454.[Abstract]
7. Stumpf W, Sar M. Autoradiographic localization of estrogen, androgen, progestin, and glucocorticosteroid in "target tissues" and "non-target tissues". In: Pasqualini J (ed.), Receptors and Mechanism of Action of Steroid Hormones. New York, NY: Marcel Dekker Inc.; 1976: 41–84.
8. Bigsby RM, Cunha GR. Estrogen stimulation of deoxyribonucleic acid synthesis in uterine epithelial cells which lack estrogen receptors. Endocrinology 1986; 119:390–396.[Abstract]
9. Cunha GR, Shannon JM, Vanderslice KD, Sekkingstad M, Robboy SJ. Autoradiographic analysis of nuclear estrogen binding sites during postnatal development of the genital tract of female mice. J Steroid Biochem 1982; 17:281–286.[CrossRef][Medline]
10. Greco TL, Duello TM, Gorski J. Estrogen receptors, estradiol, and 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.
11. Li S. Relationship between cellular DNA synthesis, PCNA expression and sex steroid hormone receptor status in the developing mouse ovary, uterus and oviduct. Histochemistry 1994; 102:405–413.[CrossRef][Medline]
12. Yamashita S, Newbold RR, McLachlan JA, Korach KS. The role of the estrogen receptor in uterine epithelial proliferation and cytodifferentiation in neonatal mice. Endocrinology 1990; 127:2456–2463.[Abstract]
13. Casimiri V, Tath NC, Parvey H, Psychoyos A. Effect of sex steroids on rat endometrial epithelium and stroma cultured separately. J Steroid Biochem 1980; 12:293–298.[CrossRef][Medline]
14. Iguchi T, Uchima F-DA, Ostrander PL, Bern HA. Growth of normal mouse vaginal epithelial cells in and on collagen gels. Proc Natl Acad Sci USA 1983; 80:3743–3747.[Abstract/Free Full Text]
15. Cooke PS, Uchima F-DA, Fujii DK, Bern HA, Cunha GR. Restoration of normal morphology and estrogen responsiveness in cultured vaginal and uterine epithelia transplanted with stroma. Proc Natl Acad Sci USA 1986; 83:2109–2113.[Abstract/Free Full Text]
16. Inaba T, Wiest WG, Strickler RC, Mori J. Augmentation of the response of mouse uterine epithelial cells to estradiol by uterine stroma. Endocrinology 1988; 123:1253–1258.[Abstract]
17. Lubahn DB, Moyer JS, Golding TS, Couse JF, 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.[Abstract/Free Full Text]
18. Johns A, Freay AD, Fraser W, Korach KS, Rubanyi GM. Disruption of estrogen receptor gene prevents 17ß estradiol-induced angiogenesis in transgenic mice. Endocrinology 1996: 137:4511–4513.
19. Ogawa S, Lubahn DB, Korach KS, Pfaff DW. Behavioral effects of estrogen receptor gene disruption in male mice. Proc Natl Acad Sci USA 1997; 94:1476–1481.[Abstract/Free Full Text]
20. Rissman EF, Early AH, Taylor JA, Korach KS, Lubahn DB. Estrogen receptors are essential for female sexual receptivity. Endocrinology 1997; 138:507–510.[Abstract/Free Full Text]
21. Iafrati MD, Karas RH, Aronovitz M, Kim S, Sullivan TR Jr, Lubahn DB, O'Donnell TR Jr, Korach KS, Mendelsohn ME. Estrogen inhibits the vascular injury response in estrogen receptor- deficient mice. Nat Med 1997; 3:545–548.[CrossRef][Medline]
22. Eddy EM, Washburn TF, Bunch DO, Goulding EH, Bladen BC, Lubahn DB, Korach KS. Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 1996; 137:4796–4805.[Abstract]
23. Hess, RA, Bunick D, Lee K-H, Bahr J, Taylor JA, Korach KS, Lubahn DB. A role for estrogens in the male reproductive system. Nature 1997; 390:509–512.[CrossRef][Medline]
24. Korach KS, Couse JF, Curtis SW, Washburn TF, Lindzey J, Kimbro KS, Lubahn DB, Eddy EM, Snedeker SM, Schomberg DW, Smith EP. Estrogen receptor gene disruption: molecular characterization, experimental and clinical phenotypes. Insights from the study of animals lacking functional estrogen receptor. Rec Prog Horm Res 1996; 51:159–188.
25. Scully KM, Gleiberman AS, Lindzey J, Lubahn DB, Korach KS, Rosenfeld MG. Role of estrogen receptor- in the anterior pituitary gland. Mol Endocrinol 1997; 11:674–681.[Abstract/Free Full Text]
26. Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Taylor J, Korach KS, Lubahn DB, Cunha GR. Stromal estrogen receptors (ER) mediate the mitogenic effect of estradiol on uterine epithelium. Proc Natl Acad Sci USA 1997; 94:6535–6540.[Abstract/Free Full Text]
27. Cunha GR. Stromal induction and specification of morphogenesis and cytodifferentiation of the epithelia of the Müllerian ducts and urogenital sinus during development of the uterus and vagina in mice. J Exp Zool 1976; 196:361–370.[CrossRef][Medline]
28. Bigsby RM, Cooke PS, Cunha GR. A simple, efficient method for separating murine uterine epithelial and mesenchymal cells. Am J Physiol 1986; 251:E630–636.
29. Mukku VR, Stancel GM. Receptors for epidermal growth factor in the rat uterus. Endocrinology 1985; 117:149–154.[Abstract]
30. Mukku VR, Stancel GM. Regulation of epidermal growth factor receptor by estrogen. J Biol Chem 1985; 260:9820–9824.[Abstract/Free Full Text]
31. Nelson KG, Takahashi T, Bossert NL, Walmer DK, McLachlan JA. Epidermal growth factor replaces estrogen in the stimulation of female genital-tract growth and differentiation. Proc Natl Acad Sci USA 1991; 88:21–25.[Abstract/Free Full Text]
32. Zhang Z, Laping J, Glasser S, Day P, Mulholland J. Mediators of estradiol-stimulated mitosis in the rat uterine luminal epithelium. Endocrinology 1998; 139:961–966.[Abstract/Free Full Text]
33. Murphy LJ, Ghahary A. Uterine insulin-like growth factor-1: regulation of expression and its role in estrogen induced uterine proliferation. Endocr Rev 1990; 11:443–453.[Medline]
34. Aaronson SA, Bottaro DP, Miki T, Ron D, Finch PW. Keratinocyte growth factor: a fibroblast growth factor family member with unusual target cell specificity. Ann NY Acad Sci 1990; 638:62–77.
35. Zarnegar R, Michalopoulos G. The many faces of hepatocyte growth factor: from hematopoiesis to hematopoiesis. J Cell Biol 1995; 129:1177–1180.[Free Full Text]
36. Pentecost BT, Teng CT. Lactotransferrin is the major estrogen inducible protein of mouse uterine secretions. J Biol Chem 1987; 262:10134–10139.[Abstract/Free Full Text]
37. Sundstrom SA, Komm BS, Ponce-de-Leon H, Yi Z, Teuscher C, Lyttle CR. Estrogen regulation of tissue-specific expression of complement. J Biol Chem 1989; 264:16941–16947.[Abstract/Free Full Text]
38. Surveyor GA, Gendler SJ, Pemberton L, Das SK, Chakraborty I, Julian J, Pimental RA, Wegner CC, Dey SK, Carson DD. Expression and steroid hormonal control of Muc-1 in the mouse uterus. Endocrinology 1995; 136:3639–3647.[Abstract]
39. Setiawan T, Buchanan D, Taylor JA, Young P, Lubahn DB, Cunha GR, Cooke PS. Role of stromal and epithelial estrogen receptors (ER) in uterine epithelial secretory protein production. Biol Reprod 1997; 56(suppl 1):83.
40. Buchanan DL, Kurita T, Taylor JA, Lubahn DB, Cunha GR, Cooke PS. Role of stromal and epithelial estrogen receptors in vaginal epithelial proliferation, stratification and cornification. Endocrinology 1998; (in press).
41. Cunha GR, Hom YK. Role of mesenchymal interactions in mammary gland development. J Mammary Gland Biol Neoplasia 1996; 1:21–35.[CrossRef][Medline]
42. Cunha GR, Young P, Hom YK, Cooke PS, Taylor JA, Lubahn DB. Elucidation of a role of stromal steroid hormone receptors in mammary gland growth and development by tissue recombination experiments. J Mammary Gland Biol Neoplasia 1997; 2:393–402.[CrossRef][Medline]
43. Wessells NK. Tissue Interactions and Development. Menlo Park, CA: WA Benjamin, Inc.; 1977.
44. Cunha GR, Alarid ET, Turner T, Donjacour AA, Boutin EL, Foster BA. Normal and abnormal development of the male urogenital tract: role of androgens, mesenchymal-epithelial interactions and growth factors. J Androl 1992; 13:465–475.[Abstract/Free Full Text]
45. Lydon JP, DeMayo FJ, Conneely OM, O'Malley BW. Reproductive phenotypes of the progesterone receptor null mutant mouse. J Steroid Biochem Mol Biol 1996; 56:67–77.[CrossRef][Medline]
46. Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Shyamala G, Conneely OM, O'Malley BW. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 1995; 9:2266–2278.[Abstract/Free Full Text]