The goal of this application is to understand how the t(8;21), the most frequent chromosomal translocation associated with acute myeloid leukemia (AML), sets the stage for secondary mutations to accumulate and develop into AML. Understanding how the encoded AML1-ETO fusion protein alters epigenetic wiring, is critical to finding less debilitating therapies that yield better outcomes. Both AML1 (RUNX1) and ETO/MTG family members also suffer point mutations in solid tumors, and the ETO family members Mtgr1 (CBFA2T2) and Mtg16 (CBFA2T3) are tumor suppressors in mouse models of intestinal neoplasia, so understanding how ETO contacts histone modifying enzymes has great impact outside of the t(8;21). In our preliminary data, we have used CRISPR/Cas9 technology to modify the 3’ end of the endogenous AML1-ETO with FKBP12F36V- HA or 3XFLAG tags to selectively and rapidly degrade AML1-ETO. We have coupled this system with state-of-the-art genomics such as precision nuclear run-on transcription sequencing (PROseq) and Cut&Run to establish a chemical genetic system to unambiguously define the mechanism of transcriptional control by AML1-ETO. Critically, this allows us to define the earliest, and presumably direct, changes in transcription upon inactivation of the fusion protein. Our preliminary PROseq data demonstrate that enhancers within key hematopoietic regulatory genes such as CEBPA are reactivated within 2 hr of adding dTAG47 to Kasumi-1 cell cultures. These novel reagents allow us to define changes in histone modifications and RNA polymerase dynamics to define the action of AML1-ETO at defined loci and throughout the genome. Moreover, our preliminary data already provide a paradigm shift: even though AML1-ETO bound enhancers have been repressed since the establishment of these cell lines, they were reactivated with a time course that matched the degradation of the fusion protein. Thus, continued expression of AML1-ETO is needed to maintain repression, at many loci, while other loci are more permanently silenced. Finally, we used CRISPR to allow rapid purification of AML1-ETO coupled with MUDPIT and identified a new chromatin modifying complex as potentially mediating AML1-ETO-dependent repression. We hypothesize that AML1-ETO recruits histone modifying enzymes to rewire the epigenetic landscape to suppress CEBPA, PU.1 and GFI1B to impair myeloid differentiation. This sets the stage for secondary epigenetic mutations that reinforce these changes such as inactivation of ASXL1/2. We will directly test this hypothesis by defining the molecular contacts that control AML1-ETO recruitment of repression complexes and use chemical genetics to test if these contacts are required for AML1- ETO-regulated transcription.

Narrative: We must understand how a mutation causes cancer before we can develop targeted therapeutics or design drug combinations that eliminate the leukemic cells, leaving normal cells to flourish. This proposal seeks to define how AML1-ETO alters epigenetic circuits to establish a permissive environment that allows secondary mutations to transform these cells into AML. We will focus on the recruitment of HDAC3 and PRC1.5 by AML1-ETO and test how recruitment of these histone-modifying complexes is use to set the stage for the development of AML.