PROJECT SUMMARY Acute myeloid leukemia (AML), diagnosed in over 20,000 US citizens each year, has a 5-year survival of only 20-30%. Despite decades of research, the effective treatment of AML remains a significant unmet clinical need, with standard of care underscored by high relapse and mortality rates due to the persistence of treatment- resistant leukemic stem cells (LSC) that drive the disease. Work by us and others, and the existence of a large genomic subgroup represented by mutations in RNA splicing genes, in combination present compelling evidence to suggest that deregulated control at the level of RNA metabolism is a feature that defines and mediates aspects of disease pathology. RNA binding proteins (RBPs) are core effectors of post-transcriptional regulation that execute precise control of gene expression by modulating RNA properties that include splicing, polyadenylation, localization, degradation and translation, and as a class can enforce an extremely diverse network of regulatory pathways. One specific class of RBPs functions within stress granules, membraneless cytoplasmic structures that are composed of RBPs and mRNAs to orchestrate post-transcriptional programs mediating adaptive cellular responses. Through a combination of genetic screens and in vivo studies, we recently discovered that LSCs selectively depend on the ability to form stress granules for their survival and proliferation. We therefore hypothesize that stress granules play a critical role in AML pathogenesis and that stress granule proteins represent a vulnerability that can be exploited in novel targeted therapeutic strategies. In this project, we aim to define the dynamics and functional requirements for stress granules in the leukemic stem cells of highly treatment refractory MLL-rearranged (MLL-r) AML. We will first perform a series of in vitro stress granule formation and in vivo cell engraftment studies assays using acutely isolated MLL-r AML and normal HSCs to identify stress granule dynamics and characteristics across primitive to more committed AML and normal hematopoietic cells. We will uncover mRNAs bound selectively in stress granules, define and functionally validate putatively key LSC- specific stress granule-interacting proteins, including one candidate we have already identified, and map the role of novel LSC-specific stress granule proteins and RNAs sequestered in them, in LSC function. Lastly, we will systematically screen for novel stress granule-relevant RBPs that underlie the function of relapsed LSCs. Our study will thus define the role of stress granule RNA and protein components and the gene expression networks they control to promote leukemia pathogenesis, thereby identifying novel potential therapeutic targets in AML.