Carcinogenic potential of ovulatory genotoxicity

Ovulation is a rate-limiting event for the perpetuation of a species; unfortunately, it imparts a cancer risk. Reactive oxidants generated during the mechanics of ovulatory follicular rupture damage the DNA of ovarian surface epithelial cells that are located within a limited diffusion radius. Those cells that survive the trauma of ovulation, along the margins of a ruptured follicle, proliferate and migrate to reconcile the discontinuity within the ovarian epithelium created at the site of oocyte release. It is conceivable that clonal expansion of an ovarian surface epithelial cell with unrepaired DNA, but not committed to death, could be an initiating factor in the etiology of common ovarian cancer. In fact, the majority of cancers of the ovary are derived from the surface epithelium; and circumstances that avert ovulation (oral contraceptive use, pregnancy/lactation) protect against ovarian adenocarcinoma. Not surprisingly, the genotoxic potential of ovulation is exacerbated by malfunctions in tumor suppressor/cell-cycle arrest and base-excision repair mechanisms. Recent experimental evidence indicates that vitamin E and progesterone protect against ovarian metaplasia by negating the oxidative stress of ovulation and by enhancing the repair capacity (genomic integrity) of the surface epithelium, respectively. Ovarian cancer of surface epithelial origin is a deadly insidious disease because it characteristically remains asymptomatic until it has metastasized throughout the abdominal cavity; therefore, prevention is a high priority.     

Ovulation-induced DNA damage in ovarian surface epithelial cells of ewes: prospective regulatory mechanisms of repair/survival and apoptosis.

Oxidative base (8-oxoguanine) damage, DNA fragmentation, and apoptosis occurred among ovarian surface epithelial cells within the formative site of ovulation in sheep. The incidence of 8-oxoguanine adducts in surviving antiapoptotic Bcl-2/base excision repair polymerase beta-positive cells at the margins of ruptured follicles (which avoid the focal point of the ovulatory assault) was intermediate between apoptotic and outlying healthy epithelium. Cells containing perturbations to DNA expressed the tumor suppressor p53. Localized reactions of DNA injury and programmed cellular death were averted by ovulation blockade with indomethacin. Progesterone enhanced the biosynthesis of polymerase beta in ovarian surface epithelial cells exposed in vitro to a sublethal concentration of H(2)O(2). Ovulation is a putative etiological factor in common epithelial ovarian cancer. A genetically altered progenitor cell, with unrepaired DNA, but not committed to death, could give rise to a transformed phenotype that is hence propagated upon healing of the ovulatory wound; it appears that this incongruity is normally reconciled by up-regulation of the base excision repair pathway during the ensuing luteal phase.     

Morphologic responses of the mouse ovarian surface epithelium to ovulation and steroid hormonal milieu

Ovarian cancer of surface epithelial origin is an ovulation- and endocrine-related disease. It appears that a cell transformed by genotoxins generated at follicular rupture is propagated during postovulatory wound repair. A consequent steroid hormonal imbalance favoring the mitogenic estrogens is a prospective predisposing factor in ovarian neoplasia. Protection against epithelial ovarian cancer is conferred by progesterone. The objective of this study was to characterize the acute effects of ovulation and steroid hormonal exposure on morphologic responses of surface epithelial cells of mouse ovaries. Follicular development and ovulation were induced in immature animals with equine and human (=Day 0) choriogonadotropins, respectively. On Day 2 (approximately 36 hrs after ovulation), surface epithelial classifications presented in histologic sections were altered from simple (single-layered) squamous and cuboidal toward stratification; this trend was reversed (i.e., reverted to the control status) on Days 4-8. Shifts in the ovarian epithelium from simple to stratified were accentuated following postovulatory (Days 1-8) treatment with estradiol. Surface epithelia of ovaries obtained after 1 week of progesterone administration were exclusively of a simple phenotype. We conclude that the proliferative/procarcinogenic reaction of the ovarian surface epithelium to ovulation is exacerbated by estrogen and counteracted by progesterone.     

Oxidative damage to DNA of ovarian surface epithelial cells affected by ovulation: carcinogenic implication and chemoprevention

The majority of cancers of the ovary are thought to originate from a surface epithelial cell perturbed by ovulation. Outgrowth of a follicle destined to ovulate brings it into apposition with the ovarian epithelium. Ovarian surface cells are consequently exposed, within a limited diffusion radius, to inflammatory agents and reactive oxidants generated during periovulatory processes. Cells that overlie the formative site of follicular rupture suffer irreparable damages and undergo apoptosis. Potentially mutagenic 8-oxoguanine modifications were detected in (surviving) cells circumjacent to postovulatory ovine and human follicles. It is conceivable that clonal expansion of a cell with unrepaired DNA, but not committed to death, could be an initiating factor in the etiology of malignancy, insofar as proliferative ovulatory wound-repair responses may propagate mutations. Since the prognosis for ovarian cancer patients with invasive disease is so poor, and early detection has proven elusive, it is imperative that prospective methods of chemo-prevention be explored. Ovulation-induced oxidative base damages to the ovarian epithelium of ewes were prevented by vitamin E. Oxoguanine adducts persisted and CA-125 (a phenotype of metaplastic transformation) was expressed in cultures of cells that were distressed by ovulation in which p53 synthesis was inhibited. Vitamin E negated this reaction. Ovarian cyclicity and fertility were not altered in vitamin-treated ewes. A prophylactic benefit of a supplemental antioxidant is suggested in "ovulating" individuals designated at risk (e.g., due to a tumor suppressor malfunction) for the development of ovarian cancer.     

Evidence for a role of the ovarian surface epithelium in the ovulatory mechanism of the sheep: secretion of urokinase-type plasminogen activator

The extent of dissolution of tissues within the apical wall of the preovulatory ovine follicle (formative site of rupture) is greater than that of the counterpart basal hemisphere. It has been hypothesized that proteolytic enzymes released from contiguous ovarian surface epithelial cells contribute to apical follicular weakening and ovulation. Ovulation occurs from the dominant ovarian follicle of proestrous ewes at approximately 24 h after administration of luteinizing hormone-releasing hormone (LHRH). Follicular rupture was inhibited in sheep in which the ovarian surface epithelium was surgically removed at 8 (but not at 16) h following LHRH. Plasminogen activator bioactivity was greater within the follicular apex compared to basal wall at 12 h; this difference was negated by prior removal of epithelium at 8 h after LHRH. A low M_r plasminogen activator of the urokinase-type (uPA) was secreted by epithelial cells recovered from the surface of preovulatory follicles (Western blot analysis). Ovarian epithelium, not associated with a preovulatory follicle, produced very little uPA. Finally, ovulation was suppressed by intrafollicular injection (8 h post-LHRH) of uPA antibodies. It is suggested that secretion of uPA by ovarian surface epithelium and consequent plasmin up-regulation within neighboring tunica albuginea and follicular theca is a contributing factor in the mechanism of ovulation.     

Sequence of apoptosis and inflammatory necrosis within the formative ovulatory site of sheep follicles

The aim of this study was to define the temporal and spatial patterns of apoptosis, necrosis and inflammation within preovulatory ovine follicles. A gonadotrophin surge was induced in pro-oestrous ewes by GnRH, and isolated follicles were hemisected into apical and basal segments at 0, 10, 18 and 22 h (the time of ovulatory stigma development) after GnRH. Ovarian surface epithelial and granulosa cells were isolated and assessed by fluorescence microscopy for membrane phosphatidylserine translocation-annexin V (early-stage apoptosis), oligonucleosomal DNA nick endlabelling (advanced apoptosis), and nuclear propidium iodide incorporation (necrotic membrane disruption). Thecal shells were analysed for interstitial blood cells. Preovulatory follicles were also hemisected and subjected to electrophoretic DNA degradation analysis. Annexin V binding and in situ DNA fragmentation among ovarian surface epithelial and granulosa cells along the follicular apex were high 18 and 22 h after GnRH. Propidium iodide staining of apical ovarian surface and granulosa cells was apparent at 22 h. There was a coincident increase within the apical theca as the time of ovulation approached in extravasated leucocytes (18 and 22 h) and erythrocytes (22 h). Apoptotic DNA laddering and necrotic DNA smears within the follicular apex were evident on agarose gels at 18 and 22 h, respectively. In contrast, ovarian surface epithelium not associated with the ovulation site and the basal follicular wall were largely unafflicted. It is suggested that both modalities of cellular death, apoptosis and necrosis (with acute inflammation and vascular injury), contribute progressively to follicular stigma formation and ovarian rupture.     

Glucocorticoid regulation of SLIT/ROBO tumour suppressor genes in the ovarian surface epithelium and ovarian cancer cells

The three SLIT ligands and their four ROBO receptors have fundamental roles in mammalian development by promoting apoptosis and repulsing aberrant cell migration. SLITs and ROBOs have emerged as candidate tumour suppressor genes whose expression is inhibited in a variety of epithelial tumours. We demonstrated that their expression could be negatively regulated by cortisol in normal ovarian luteal cells. We hypothesised that after ovulation the locally produced cortisol would inhibit SLIT/ROBO expression in the ovarian surface epithelium (OSE) to facilitate its repair and that this regulatory pathway was still present, and could be manipulated, in ovarian epithelial cancer cells. Here we examined the expression and regulation of the SLIT/ROBO pathway in OSE, ovarian cancer epithelial cells and ovarian tumour cell lines. Basal SLIT2, SLIT3, ROBO1, ROBO2 and ROBO4 expression was lower in primary cultures of ovarian cancer epithelial cells when compared to normal OSE (P<0.05) and in poorly differentiated SKOV-3 cells compared to the more differentiated PEO-14 cells (P<0.05). Cortisol reduced the expression of certain SLITs and ROBOs in normal OSE and PEO-14 cells (P<0.05). Furthermore blocking SLIT/ROBO activity reduced apoptosis in both PEO-14 and SKOV-3 tumour cells (P<0.05). Interestingly SLIT/ROBO expression could be increased by reducing the expression of the glucocorticoid receptor using siRNA (P<0.05). Overall our findings indicate that in the post-ovulatory phase one role of cortisol may be to temporarily inhibit SLIT/ROBO expression to facilitate regeneration of the OSE. Therefore this pathway may be a target to develop strategies to manipulate the SLIT/ROBO system in ovarian cancer.     

Gonadotropin and steroid hormones stimulate proliferation of the rat ovarian surface epithelium

The ovarian surface epithelium (OSE) is a single layer of flattened or cuboidal cells covering the ovary. Ninety percent of all human ovarian malignancies arise from this layer of cells. Incessant ovulation, hyperovulation induced by infertility treatment, and hormone replacement therapy have been suggested as risk factors for ovarian cancer. In this study, two groups of rats, with and without surgically induced injury to the ovary, were treated with 17beta-estradiol, pregnant mare's serum gonadotropin (PMSG), human chorionic gonadotropin (hCG), or the combination PMSG/hCG, and the proliferative response of the OSE cells was measured using bromodeoxyuridne (BrdU) and (3)H-thymidine. All hormones, alone or in combination with ovarian surgery, were found to increase significantly the rate of proliferation of the rat OSE. These data demonstrate that hormones associated with infertility treatments and hormone replacement therapy, as well as injury- or ovulation-induced rupture of the ovarian surface, stimulate the rat OSE, and hence could have a role in the development of ovarian cancer via proliferation-associated mutagenesis, or alternatively, by promoting the rapid selection of OSE cells with accumulated mutations.     

Expression of death receptor 6 by ovarian tumors in laying hens, a preclinical model of spontaneous ovarian cancer

Tumor-associated neoangiogenesis and suppression of antitumor immunity are hallmarks of tumor development and progression. Death receptor 6 (DR6) has been reported to be associated with suppression of antitumor immunity and tumor progression in several malignancies. However, expression of DR6 by malignant ovarian epithelial tumors at an early stage is unknown. The goals of this study were to determine whether DR6 is expressed by malignant ovarian epithelial tumors at an early stage and to examine whether DR6 expression is associated with ovarian cancer (OVCA) progression in a laying hen model of spontaneous OVCA. Expression of DR6 was examined in normal and malignant ovaries, normal ovarian surface epithelial (OSE) cells, or malignant epithelial cells and in serum of 3-year-old hens. The population of microvessels expressing DR6 was significantly higher in hens with early-stage OVCA than hens with normal ovaries (P < .01) and increased further in late-stage OVCA. The results of this study showed that, in addition to microvessels, tumor cells in the ovary also express DR6 with a significantly higher intensity than normal OSE cells. Similar patterns of DR6 expression were also observed by immunoblot analysis and gene expression studies. Furthermore, DR6 was also detected in the serum of hens. In conclusion, DR6 expression is associated with OVCA development and progression in laying hens. This study may be helpful to examine the feasibility of DR6 as a useful surrogate marker of OVCA, a target for antitumor immunotherapy and molecular imaging and thus provide a foundation for clinical studies.     

Differential effects of indomethacin on the sheep ovary: Prostaglandin biosynthesis, intracellular calcium, apoptosis, and ovulation

Cells of the apical wall of the dominant follicle and contiguous ovarian surface epithelium become apoptotic with the approach of ovulation in the sheep. It was hypothesized that indomethacin, an established inhibitor of prostaglandin biosynthesis and ovulation, would protect apical ovarian cells from programmed death. The anovulatory potencies of two systemic doses of indomethacin (200 and 800 mg) were tested in gonadotropin-stimulated ewes. A complete blockade of ovulation occurred at the higher dose of indomethacin. Ovulation was not inhibited by 200 mg indomethacin. Both doses of drug suppressed follicular prostaglandin production below pregonadotropin levels. Immunofluorescence detection of digoxigenin end-labeled (fragmented) DNA was used as a marker of apoptosis among ovarian surface epithelial and granulosa cells recovered from the apical hemisphere of preovulatory ovine follicles. Cellular DNA fragmentation was averted in animals given 800 mg indomethacin, whereas apoptosis ensued after 200 mg. A sustained increase in cytosolic calcium is generally a prerequisite to apoptotic DNA fragmentation and cell death. Indeed, intracellular calcium, detected by fluorescence of fura-2, was elevated in ovarian cells of animals destined to ovulate (controls, 200 mg indomethacin) in comparison to (safeguarded) cells of anovulatory ewes (800 mg indomethacin). These observations provide circumstantial evidence that apical ovarian cell degeneration by calcium-mediated apoptosis is a determinant of follicular instability and rupture, but that these events are unrelated to the gonadotropin-induced rise in prostanoid production characteristic of preovulatory follicles.     

Comparison of expression profiles in ovarian epithelium in vivo and ovarian cancer identifies novel candidate genes involved in disease pathogenesis

Molecular events leading to epithelial ovarian cancer are poorly understood but ovulatory hormones and a high number of life-time ovulations with concomitant proliferation, apoptosis, and inflammation, increases risk. We identified genes that are regulated during the estrous cycle in murine ovarian surface epithelium and analysed these profiles to identify genes dysregulated in human ovarian cancer, using publically available datasets. We identified 338 genes that are regulated in murine ovarian surface epithelium during the estrous cycle and dysregulated in ovarian cancer. Six of seven candidates selected for immunohistochemical validation were expressed in serous ovarian cancer, inclusion cysts, ovarian surface epithelium and in fallopian tube epithelium. Most were overexpressed in ovarian cancer compared with ovarian surface epithelium and/or inclusion cysts (EpCAM, EZH2, BIRC5) although BIRC5 and EZH2 were expressed as highly in fallopian tube epithelium as in ovarian cancer. We prioritised the 338 genes for those likely to be important for ovarian cancer development by in silico analyses of copy number aberration and mutation using publically available datasets and identified genes with established roles in ovarian cancer as well as novel genes for which we have evidence for involvement in ovarian cancer. Chromosome segregation emerged as an important process in which genes from our list of 338 were over-represented including two (BUB1, NCAPD2) for which there is evidence of amplification and mutation. NUAK2, upregulated in ovarian surface epithelium in proestrus and predicted to have a driver mutation in ovarian cancer, was examined in a larger cohort of serous ovarian cancer where patients with lower NUAK2 expression had shorter overall survival. In conclusion, defining genes that are activated in normal epithelium in the course of ovulation that are also dysregulated in cancer has identified a number of pathways and novel candidate genes that may contribute to the development of ovarian cancer.     

Proliferating cell nuclear antigen immunoreactivity in the ovarian surface epithelium of mice of varying ages and total lifetime ovulation number following ovulation

Everytime an oocyte is released at ovulation, the ovarian surface epithelium (OSE) is ruptured and must be restored by epithelial cell proliferation. Ovulation site closure was studied in mice of various ages along with total lifetime ovulation number to investigate the known association of these factors with the risk of epithelial ovarian cancer. Ovaries from Swiss Webster mice were collected at various time points postovulation from 3-mo virgin animals (estimated median total lifetime ovulation number, 92; n = 40 mice), 8-mo virgin animals subject to incessant ovulation (estimated median total lifetime ovulation number, 652; n = 15 mice), and 12-mo breeders (estimated median total lifetime ovulation number, 208; n = 35 mice). Diameters of ovulation sites were estimated by scanning electron microscopy. No differences were found in the rate of ovulation site closure between the groups. Sections of ovaries were analyzed using immunohistochemistry for proliferating cell nuclear antigen (PCNA). The highest density of immunoreactive cells was observed in all animal groups in the cuboidal cells around the rupture site the day after ovulation. Despite the similarity in ovulation site closure rates between groups, the total number of OSE cells that were positive for PCNA in both the 8- and 12-mo animals was significantly reduced, so the number of stained cells appeared to be insufficient to cover the ovulation site. These data suggest that other mechanisms, such as proliferation of the extraovarian mesothelium, may play a role in the re-epithelialization of the ovary.     

The effect of indomethacin on ovarian prostaglandin release in hens

Indomethacin, an inhibitor of prostaglandin (PG) synthetase, will block uterine muscle electromyographic activity (EMG activity) and oviposition at a midsequence oviposition and ovulation in domestic hens, but does not block the increase in EMG activity associated with the first ovulation of a sequence. To assess the potential relationship between prostaglandin release from the ovarian follicles and EMG activity in egg-laying hens, we determined the concentrations of PGF2 alpha, 13,14-dihydro-15-keto-PGF2 alpha (PGFM), and PGE2 in brachial, ovarian follicular and uterine venous plasma and tissues in relation to uterine muscle EMG activity at the first ovulation and at a midsequence oviposition. The concentrations were measured after an i.m. injection (25 mg/hen) of indomethacin. In control hens sampled hourly, beginning 4 h before the peak of EMG activity at the first ovulation of a sequence, there was a sharp increase (p less than 0.05) in concentrations of PGF2 alpha and PGFM in brachial vein plasma coincident with the increase (p less than 0.05) in uterine EMG activity. Hens pretreated with indomethacin also had increased plasma PGF2 alpha and PGFM levels (p less than 0.05) in brachial vein plasma and increased uterine EMG activity (p less than 0.05) at this time. Indomethacin treatment lowered but did not eliminate mean levels of PGF2 alpha in the venous effluent from the largest preovulatory follicle at the first ovulation (36.0 +/- 9.9 ng/ml vs. 14.4 +/- 1.8 ng/ml).(ABSTRACT TRUNCATED AT 250 WORDS)     

Evidence for Follicle Wall Involvement in Ovulation and Progesterone Production by Frog (Rana pipiens) Follicles in vitro

Involvement of different cellular investments of the amphibian ovarian follicle wall in the ovulatory process, progesterone production, and oocyte maturation was investigated. Following microdissection, to selectively remove one or more layers (surface epithelium, theca, follicle cells) of the follicle wall, dissected and undirected ovarian follicles were treated with frog pituitary homogenate (FPH) or progesterone. Intact follicles ovulated in response to pituitary homogenate and this was associated with contractions of the follicle wall. Ovulation and follicular contractions were not observed following removal of the surface epithelium without removing the thecal layer. Oocyte maturation occured in response to FPH following removal of the surface epithelium alone or together with the theca, but not in the absence of the follicle cells. Intact follicles were most responsive to FPH with respect to progesterone production, and removal of all somatic cells from oocytes obliterated FPH stimulated progesterone production. Oocytes, regardless of wether any or all follicular wall layers were removed, matured but did not ovulate following exposure to progesterone. The results suggest that the surface epithelium, but not the theca, is required for FPH-induced extrusion (ovulation) of the oocyte from ovarian follicle wall. Additionally, the somatic tissue rather than the oocyte appears to be the cells producing progesterone following FPH treatment. The results indicate that separate cellular layers (individually and/or as a result of interactions) of the follicle wall carry out different functions during follicular differentiation and mediation of ovulation. Data provide functional evidence for a role of the surface epithelium in controlling the process of ovulation and follicular contraction.     

Tetraploid cells from cytokinesis failure induce aneuploidy and spontaneous transformation of mouse ovarian surface epithelial cells

Most ovarian cancers originate from the ovarian surface epithelium and are characterized by aneuploid karyotypes. Aneuploidy, a consequence of chromosome instability, is an early event during the development of ovarian cancers. However, how aneuploid cells are evolved from normal diploid cells in ovarian cancers remains unknown. In the present study, cytogenetic analyses of a mouse syngeneic ovarian cancer model revealed that diploid mouse ovarian surface epithelial cells (MOSECs) experienced an intermediate tetraploid cell stage, before evolving to aneuploid (mainly near-tetraploid) cells. Using long-term live-cell imaging followed by fluorescence in situ hybridization (FISH), we demonstrated that tetraploid cells originally arose from cytokinesis failure of bipolar mitosis in diploid cells, and gave rise to aneuploid cells through chromosome mis-segregation during both bipolar and multipolar mitoses. Injection of the late passage aneuploid MOSECs resulted in tumor formation in C57BL/6 mice. Therefore, we reveal a pathway for the evolution of diploid to aneuploid MOSECs and elucidate a mechanism for the development of near-tetraploid ovarian cancer cells.     

Tetraploid cells from cytokinesis failure induce aneuploidy and spontaneous transformation of mouse ovarian surface epithelial cells

Most ovarian cancers originate from the ovarian surface epithelium and are characterized by aneuploid karyotypes. Aneuploidy, a consequence of chromosome instability, is an early event during the development of ovarian cancers. However, how aneuploid cells are evolved from normal diploid cells in ovarian cancers remains unknown. In the present study, cytogenetic analyses of a mouse syngeneic ovarian cancer model revealed that diploid mouse ovarian surface epithelial cells (MOSECs) experienced an intermediate tetraploid cell stage, before evolving to aneuploid (mainly near-tetraploid) cells. Using long-term live-cell imaging followed by fluorescence in situ hybridization (FISH), we demonstrated that tetraploid cells originally arose from cytokinesis failure of bipolar mitosis in diploid cells, and gave rise to aneuploid cells through chromosome mis-segregation during both bipolar and multipolar mitoses. Injection of the late passage aneuploid MOSECs resulted in tumor formation in C57BL/6 mice. Therefore, we reveal a pathway for the evolution of diploid to aneuploid MOSECs and elucidate a mechanism for the development of near-tetraploid ovarian cancer cells.     

Animal models of ovarian cancer

Ovarian cancer is the most lethal of all of the gynecological cancers and can arise from any cell type of the ovary, including germ cells, granulosa or stromal cells. However, the majority of ovarian cancers arise from the surface epithelium, a single layer of cells that covers the surface of the ovary. The lack of a reliable and specific method for the early detection of epithelial ovarian cancer results in diagnosis occurring most commonly at late clinical stages, when treatment is less effective. In part, the deficiency in diagnostic tools is due to the lack of markers for the detection of preneoplastic or early neoplastic changes in the epithelial cells, which reflects our rather poor understanding of this process. Animal models which accurately represent the cellular and molecular changes associated with the initiation and progression of human ovarian cancer have significant potential to facilitate the development of better methods for the early detection and treatment of ovarian cancer. This review describes some of the experimental animal models of ovarian tumorigenesis that have been reported, including those involving specific reproductive factors and environmental toxins. Consideration has also been given to the recent progress in modeling ovarian cancer using genetically engineered mice.     

Estrogen stimulation of ovarian surface epithelial cell proliferation

Summary Ovarian cancer is the leading cause of gynecological cancer mortality, and 85–90% of this malignancy originates from the ovarian surface epithelium (OSE). The etiology of ovarian epithelial cancer is unknown but a role for estrogens has been suspected. However, the effect of estrogens on OSE cell proliferation remains to be determined. Using the rabbit model, our studies have demonstrated that 17β-estradiol stimulates OSE cell proliferation and the formation of a papillary ovarian surface morphology similar to that seen in human ovarian serous neoplasms of low malignant potential. Immunohistochemical staining of ovarian tissue sections with an antibody to the estrogen receptor α demonstrates its expression in both OSE cells and stromal interstitial cells. In primary ovarian cell cultures, the proliferative response of the epithelial cells to 17β-estradiol depends on the expression of the estrogen receptor α in the epithelial cells. However, when the epithelial cells are grown together with ovarian stromal cells, their proliferative response to this hormone is greatly enhanced, suggesting the involvement of stromal-epithelial interactions. These studies suggest a role for estrogens and the estrogen receptor α in OSE growth.