Human ovarian epithelial tumors represent a particularly good model for the study of the genetic determinants of tumor progression. Our recent work has shown that ovarian carcinomas frequently showed losses of constitutional heterozygosity at genetic loci on chromosomal segments 3p, 6q, and 11p and work from other laboratories demonstrated that the c-myc, k-ras, and HER-2/neu proto-oncogenes were frequently amplified in those tumors. We now want to expand our studies on losses of heterozygosity to include loci from other somatic chromosomal arms and we plan to examine the state of amplification of the above 3 proto-oncogenes in the same tumors. These studies will include benign, borderline, and malignant tumors because these entities can be regarded as representing specific stages of tumor development. The sequence of molecular changes taking place during malignant tumor development will also be examined by taking advantage of the availability of multiple specimens from the same patients obtained from repeat operations (second look laparatomies). In addition, we will test the hypothesis that loss of function of an alternatively spliced H-ras gene is an important cause of increased tumorigenicity in tumors with loss of heterozygosity at this locus by examining the frequencies of point mutations within the alternatively spliced exon and its flanking splice motifs in tumors with such genetic losses. Another important part of this proposal is aimed at using in vitro and animal models to examine the role of specific molecular changes in ovarian tumorigenesis. We will use microcell-mediated chromosome transfer techniques to introduce intact human chromosomes 6 and 11, either together or separately, in ovarian carcinoma cell lines with known defects on these chromosomes and examine the consequences of such transfers on tumorigenicity and in vitro growth requirements and kinetics. This will initially be done in cell lines with reported cytogenetic abnormalities on chromosomes 6 and 11. However, we also plan to isolate new cell lines from tumors with known losses of heterozygosity on those chromosomes and such cell lines will be particularly useful to investigate the role of these loci using the above genetic transfer techniques. We will also further characterize the determinants of progression in 2 established human ovarian carcinoma cell lines, called HOC-1 and HOC-7, which were obtained from the same patient at 2 different time points during the disease course. We have recently determined that HOC-7 is the only one of the 2 lines that can be cultured in serum-free medium and we will fuse the 2 cell lines to determine if such independence from serum factors is due to recessive or dominant genetic events.
