Mechanisms by Which Estrogen Accelerates Ovarian Tumor Initiation.

Abstract

Ovarian cancers present as three different types, epithelial, germ cell, and stromal, with epithelial cancers consisting ~90% of cases. As a result, it is believed that the majority of cancers that develop in the ovary arise from the ovarian surface epithelium. Previous studies have shown that exposure of mice to exogenous estradiol (E2) causes preneoplastic changes in ovarian surface epithelial (OSE) cell morphology and accelerates ovarian tumor onset. Clinically, estrogenic compounds are used for treatments such as hormone replacement therapy for relief of menopausal symptoms or as a form of contraceptive such as birth control pills. However, women using estrogen-only hormone replacement are at higher risk of ovarian cancer. Thus, the aim of this project is to determine the mechanisms by which estrogen has detrimental effects that promote an increased risk of ovarian cancer, specifically by elucidating the actions of E2 on the OSE. The pathways controlling the premalignant transformation are unknown, but exogenous E2 promote stratification and apparent loss of cellular polarity in the OSE layer. We hypothesized that this change in phenotype may be due to decreases in disabled-2 (Dab2), as Dab2 is required for epithelial cells to maintain polarity and a genome-wide screen has identified estrogen response elements in regions proximal to the transcriptional start site of Dab2 in mice and humans. Dab2 is an adaptor protein that is found in a variety of tissues and is highly expressed in breast and ovarian tissue but its expression is lost in the majority of ovarian carcinomas. To determine if Dab2 mediates the actions of E2, we examined the effects of E2 exposure on Dab2 expression (by qPCR) in cultures of normal mouse OSE (mOSE), SV40 T antigen-immortalized mOSE, and in a mouse ovarian cancer cell line (MASE). None of the cell lines showed a change in Dab2 expression in response to E2 over a range of doses (0-1000 nM) and treatment periods (0-72 hours). To determine the effects of E2 exposure on Dab2 expression in vivo, SCID mice were intraperitoneally injected with MASE cells and inserted subcutaneously with a 6o-day timed release 0.25 mg E2 pellet. Tumors showed a significant 8-fold decrease in Dab2 mRNA expression in response to E2 treatment. Tumors from the tgCAG-LS-TAg transgenic model of ovarian cancer also showed a trend for E2 suppression of Dab2 expression. Immunohistochemical detection of Dab2 shows that it is present in normal human and mouse OSE, but is lost in areas of stratified mOSE that occur at higher incidence after 60 days of exogenous E2 treatment. These results demonstrate that E2 exposure decreases Dab2 expression in both OSE and ovarian cancers in vivo, but not in vitro. The lack of response in vitro suggests that growing OSE cells on plastic is not a good model to study changes in mOSE polarity, a result that is consistent with previous studies using mammary epithelial cells. Current experiments are investigating alternative culture systems to develop a better model for the study of OSE cell polarity and epithelial cell architecture.