Drug resistance and cancer relapse are attributed to the persistence of stem cells. The PR/SET domain protein PRDM14 regulates a cell's pluripotent potential to initiate and progress cancer through epigenetic events that are molecular signatures of stem cells. Normally, PRDM14 uses epigenetic mechanisms to establish the pluripotency of primordial germ cells, and its role in cancer is similar. PRDM14's widespread expression in many tumor types, including breast, ovarian, colon, lung, melanoma and lymphoblastic leukemia implies that it is commonly involved in epigenetic events to initiate cancer. This proposal challenges paradigms by suggesting that PRDM14 regulates a somatic cell's pluripotent potential to generate C-ICs as it promotes genomic instability, allowing it to cause tumors in many cell types. PRDM14 is a prime target for therapeutic intervention, because of its restricted expression to only a few cells in the body. Key to this project is that unique inducible mouse models will be exploited to examine events during tumor development, progression and relapse. Experiments that address three basic questions will be performed: what biochemical mechanisms does Prdm14 use to establish self-renewal to initiate cancer, how do the initiating cells maintain self-renewal in the tumor, and how does PRDM14 perturb genome integrity within these cells to progress malignancy? The first aim will use cutting edge genomic approaches to confirm Prdm14's binding targets and identify its protein partners, while it determines how Prdm14 alters chromatin to reset pluripotency in pre- leukemia cells. The second aim will determine the properties of Prdm14 pre-leukemia cells using flow cytometry, cellular and molecular profiling approaches. The third aim will determine how PRDM14 catalyzes chromosomal rearrangements, leading to copy number variation in the tumors that cause specific driver mutations such as activated NOTCH1 through recombination. These studies propose a genetic strategy to inhibit recombination and a drug treatment strategy designed against targets downstream of PRDM14's action to prevent the growth of tumors in the mouse model. The mouse models allow the behavior of cancer cells to be monitored from the beginning, mark cells for isolation, and follow their lineage potential. Data from human tumors will be mined and compared with mouse tumors throughout the study and mouse leukemia will be compared with human leukemia in Aim 3 to demonstrate clinical relevance. Although mouse leukemia models are exploited, the work will have far-reaching applications to all cancers. PRDM14's limited normal expression in adults indicates that PRDM14 or its initial regulatory targets are key candidates for a therapy directed against cells that express it abnormally. Therefore, PRDM14 could prove to be a universal target in C-ICs.