We will perform insertional mutagenesis using murine stem cell virus (MSCV) to activate genes in ES cells using the viral LTR. The advantage of using retroviruses for insertional mutagenesis is that cloning of the insertion sites is relatively simple. ES cells with randomly inserted MSCV will be selected for viable clones after both copies of the functional Brca1 or Brca2 are disrupted. The advantage of using MSCV is that, unlike other retroviral vectors, its LTR can drive high levels of target gene expression in ES cells. In a complementary approach, we will also use an siRNA library-based screen to identify genes whose loss of function can rescue the lethality of BRCA2-deficient ES cells. We will take advantage of the intact DNA damage response machinery and cell-cycle checkpoint genes of ES cells to perform an siRNA-based screen to identify genes whose loss can rescue the lethality of Brca2 null ES cells. Identification of such genes will provide clues to the genes that allow BRCA2-deficient somatic cells to survive and develop into tumors. To allow rapid high-throughput screening, we have generated a mouse ES cell line that expresses a mutant allele of BRCA2 that is defective in homologous recombination and renders the cells hypersensitive to PARP inhibitors. In the primary screen we will identify siRNAs that allow these ES cells to survive in the presence of PARP inhibitors. The results obtained will provide clues to the genes may compensate for the loss of BRCA2-mediated homologous recombination. To further address the specific compensation for the loss of function of BRCA2, we will perform a secondary screen, in which we will test the siRNAs for their ability to rescue the lethality of Brca2-null ES cells. For this screen, we have generated a line of mouse ES cells that have one functionally null allele of Brca2 and the second allele is conditional and flanked by two loxP sites. The conditional allele can be deleted by Cre-mediated recombination. We will examine the expression of candidate genes in human BRCA2-deficient tumors to identify those that are silenced or down regulated. We believe that genes that contribute to BRCA2-mediated tumorigenesis can be potential targets for therapeutic intervention if their expression in tumor cells can lead to cell death or cell cycle arrest.