Louis Dubeau - US grants

Ovarian epithelial tumors are subdivided into design (cystadenomas) and malignant (carcinomas) categories. There is also an intermediate entity known as tumors of low malignant potential (LMP). The principal investigator's previous grant focused on identifying frequent molecular genetic differences between these different tumor subtypes in order to obtain insights into the molecular determinants for their phenotypic differences. Recently, the principal investigator's laboratory has succeeded in immortalizing several ovarian cystadenoma cell lines in culture as well as a cell line derived from a LMP tumor. The principal investigator now proposes to pursue studies directed at understanding the molecular genetic determinants of ovarian epithelial tumor development by taking advantage of the above-mentioned cell lines to address the functional significance of specific molecular alterations. Three specific aims are proposed. The first aim will examine the functional significance of expression of steroid and gonadotropin hormone receptors. The levels of expression of such receptors vary among different ovarian tumors and the significance of those differences as well as the potential role of hormone therapy in the management of these tumors is presently unclear. The second specific aim is based on findings from the first grant period that losses of heterozygosity are frequent molecular genetic abnormalities in ovarian carcinomas but are apparently absent in cystadenomas. The principal investigator proposes that the 2 groups of ovarian tumors develop via fundamentally distinct mechanisms and that the presence of loss of heterozygosity, which he regards as a measure of aneuploidy, strongly predisposes to malignant development. The principal investigator will test this hypothesis using 2 different approaches to create mitotic errors in cultured cystadenomas. Such mitotic errors are expected to result in aneuploidy and loss of heterozygosity. The rate of malignant transformation will be compared in treated and control cells. Finally, in the third specific aim, the principal investigator will pursue his recent findings suggesting that a gene which may escape X chromosome inactivation is important for the control of ovarian LMP tumors. This candidate gene will be further localized on the X chromosome and a function for its role in ovarian tumorigenesis may be ascribed to it using microcell-mediated chromosome transfer technologies.

This is a Shannon Award providing partial support for research projects that fall short of the assigned institute's funding range but are in the margin of excellence. The Shannon award is intended to provide support to test the feasibility of the approach; develop further tests and refine research techniques; perform secondary analysis of available data sets; or conduct discrete projects that can demonstrate the PI's research capabilities or lend additional weight to an already meritorious application. Further scientific data for the CRISP System are unavailable at this time.

molecular biology; environmental health; biomedical facility; genetic techniques; tissue /cell preparation;

Novel strategies, based on recent advances in our understanding of tumor development at the molecular genetic level, and aimed at improving our abilities to detect tumor cells in cancer patients, can be important not only for early diagnosis, but also during the period following completion of therapeutic protocols. Such techniques can help distinguish disease-free patients from those harboring residual tumor cells and thus likely to benefit from further therapeutic interventions. Ovarian carcinoma, one of the deadliest of all gynecologic cancer, is particularly amenable to such strategies because these tumors usually remain confined to the abdominal and pelvic cavities, even at advanced disease stages. Currently, 24-54% of women treated for advanced ovarian carcinoma and thought to be free of disease based on the most sensitive techniques currently available eventually experience disease recurrences, highlighting the poor sensitivity of these conventional approaches. We hypothesize that testing for the presence of an enzyme called telomerase, which is strongly associated with the cancer phenotype, can provide a means of increasing our sensitivity to detect residual viable ovarian cancer cells in abdominal washing from patients after completion of adjuvant chemotherapy. This hypothesis is strongly supported by our recent published data We will test this hypothesis in 500 abdominal washings obtained from patients undergoing second-look laparotomies at either USC or six other collaborating institutions across the US. Factors affecting the sensitivity and specificity of this approach for ovarian cancer detection and for predicting recurrences will be defined and characterized. We will also examine the role of steroid and gonadotropin hormones as well as of chemotherapeutic agents on regulation of telomerase expression. The potential influence of these agents on the sensitivity of testing for telomerase as a means of detecting ovarian cancer cells will be examined.

Description: The Molecular Biology and Sample Processing (MBSP) Core is combined from the predecessor Molecular Biology and Sample Processing and Storage Cores. Experience showed that the functions of the two Cores were inter-related as stored samples were often processed for molecular biological tests after receipt and storage. Also, it was an opportunity to integrate epidemiologists and molecular biologists to obtain better interaction. It is housed in a new, well-equipped 600 square foot laboratory within the clinical pathology laboratories at USC. The Cores functions are to aid in storage, processing, histological examination, and microdissection of biological samples including blood, tissue, urine, scrapings, etc. The Core extracts DNA or RNA from the samples and use it for PCR, and other molecular biological techniques. The Core helps to function as an education/consultation resource for Pilot Projects.

[unreadable] DESCRIPTION (provided by applicant): Women with germline mutations in BRCA1 are strongly predisposed to cancers of the ovary and fallopian tubes. Extensive epidemiological data have demonstrated a strong link between menstrual activity and risk of ovarian cancer in the general population. This led us to hypothesize that BRCA1 might predispose to ovarian cancer indirectly, by influencing ovarian granulosa cells, which play an important role in controlling menstrual cycle progression through their secretion of sex steroid hormones. Such hormones, as well as the peptide hormone mullerian inhibiting substance (MIS), also secreted by granulosa cells, can regulate ovarian epithelial tumor cell growth. We used the Cre-lox system to inactivate the mouse Brca1 gene in granulosa cells in order to test this hypothesis. A truncated form of the FSH receptor promoter, which directs expression exclusively in granulosa cells, was used as Cre driver. A majority (40/59) of Brca1 knockout mice had grossly visible cystic tumors either attached to the ovary, to the uterine horns, or with no demonstrable attachment to either of these organs. All tumors except one resembled human serous cystadenomas, which are benign tumors made up of the same cell type as ovarian serous carcinomas. A single tumor was histologically compatible with malignancy, although no sign of invasive or metastatic ability could be demonstrated. Strikingly, these tumors carried only the non-recombined (wild type) form of the floxed BRCA1 allele while the recombined (mutant) form was present only in granulosa cells in strong support of our hypothesis. Our proposal is focused on the elucidation of the mechanism of tumor predisposition in this experimental model, which we believe to be relevant to the strong tumor predisposition seen in women with germline BRCA1 mutations. We will identify the nature and tissue distribution of the ovarian and uterine cells affected by an absence of functional Brca1 in granulosa cells and will evaluate the reversibility of the phenotypic consequences of such defect and the protective role of pregnancy in Aim 1. We will characterize the consequences of such absence on the timing and hormonal activity of the various phases of the ovulatory cycle in Aim 2. The consequences of absence of a functional Brca1 on the spectrum of gene expression in ovarian granulosa cells will be examined in the last aim (Aim 3). Once the mechanisms of tumor predisposition are understood in this experimental model, we will be in a position to investigate its relevance to humans carrying a germline BRCA1 mutation and to use this knowledge to develop novel approaches to the clinical management of these individuals. [unreadable] [unreadable] [unreadable]

Applications Grants; Biological Assay; Characteristics; Core Facility; cost; Data; design; Development; Environmental Health; Genetic Techniques; Goals; health science research; Human Resources; member; Methods; Mission; Molecular Biology; Molecular Genetics; Pilot Projects; Preparation; Process; programs; Protocols documentation; Research; Research Infrastructure; Research Personnel; Research Project Grants; sample collection; Sampling

DESCRIPTION (provided by applicant): All cancer screening tests that have had a significant impact on disease morbidity and mortality are aimed not at detecting early stage cancers, but pre-cancerous or pre-invasive lesions. Searching for the precursor lesion of ovarian epithelial tumors implies knowledge of their actual site of origin. Although the currently favored theory states that ovarian epithelial tumors arise from the mesothelial cell layer that lines the ovarian surface (ovarian coelomic epithelium), this theory does not account for the well-known fact that these tumors show features of tissues embryologically derived from the mullerian ducts and for the fact that preneoplastic lesions found in surgical specimens of women undergoing prophylactic surgeries for familial ovarian cancer predisposition are most often located in the fimbriae of the fallopian tubes. This proposal is based on Dr. Dubeau's longstanding hypothesis that ovarian carcinomas do not arise from the ovarian coelomic epithelium, but from tissues that are embryologically derived from the mullerian ducts. Aim #1 will use transgenic mouse models to directly determine the exact contribution of the mullerian ducts to the adult female reproductive tract and test the hypothesis that the ovarian surface epithelium is not of mullerian origin while specific para-ovarian and para-uterine structures such as rete ovarii and others have such an origin. The possibility that portions of the coelomic epithelium can re-program their differentiation state to become mullerian-like, which is an important component of the hypothesis that ovarian tumors are of coelomic origin, will also be investigated. Aim #2 will take advantage of a mouse model recently developed in Dr. Dubeau's laboratory that suggests that inactivation of Brca1 in ovarian granulosa cells stimulate neoplastic transformation in tissues that are embryologically derived from the mullerian ducts. We will use technologies similar to those used in aim #1 to determine whether the tumors that develop in these mice are of mullerian or coelomic origin. We will also knockout the p53 and Brca1 genes either in derivatives of the mullerian ducts or in coelomic epithelium and will cross the resulting mouse lines with our line carrying a knockout of Brca1 in granulosa cells. We anticipate that the tumors that will develop in these double mutants will have increased malignant potential and we will test the hypothesis that tumors in mice in which the additional mutations are targeted to the mullerian tract will most closely resemble human tumors morphologically and immunohistochemically. We will determine which portions of either the normal mullerian tract or coelomic cavity shows the most resemblance to those tumors based on their overall gene expression profile and the manner in which they respond to different phases of the estrus cycle, the equivalent of the human menstrual cycle. We will also test the hypothesis that tumors located in the para-uterine region are molecularly similar to those arising in the oviductal/ovarian region but that their anatomical location determines the likelihood that they will be cystic, which is a frequent characteristic of human ovarian tumors.