Abstract Whole-exome and targeted sequencing of MDS patients’ samples have led to discover a set of genes (SF3B1, U2AF1, SRSF2, ZRSR2 and LUC7L2) encoding mutant RNA splicing factors that alter expression patterns of protein isoforms in clonal evolution of hematopoietic cells during MDS progression. Subsequent studies found occurrences of these splicing factor mutations in AML and solid tumors. Clinical data also showed heterozygous missense mutations or hemizygous deletions of these genes are mostly mutually exclusive implying that mutations in multiple proteins in splicing regulation may be lethal in MDS cells. To date, only one compound, H3B-8800, to target SF3B1 is in phase I clinical trial to treat MDS and other myeloid malignancies. No agents targeting other mutated splicing factors are in pre-clinical development and if other mutant factors can be potential drug targets is unknown. Among them, U2AF1 mutations occurred in early MDS clones and are poor prognostic features for progression to leukemia. Studies showed U2AF1S34F hematopoietic cells and transgenic mouse models are sensitive to sudemycin that targets SF3B1. Recently, U2AF1wt was suggested to be a haplo-essential gene for the survival of cancer cells containing U2AF1 mutation. Our central hypothesis is that U2AF1wt inhibition may induce synthetic lethality to U2AF1S34F clones that depends on U2AF1wt for survival. Our rationale is that blockade of U2AF1/U2AF2 association may impair U2AF1wt function. Using fragment-based library screening, we have identified a hit compound that inhibited the U2AF1/U2AF2 binding and exhibited selective growth inhibition in K562-U2AF1S34F and human primary cells carrying U2AF1 mutations but not their wild-type counterparts. Our objective of this proposal is to develop a new class of U2AF1 inhibitors based on the hit and assess the therapeutic concept of synthetic lethality in MDS/AML disease models. Our long-term goal is to develop small-molecule therapeutics to target cancer cells defective in the spliceosome pathway. To achieve our goal, we have devised two specific aims: 1) Optimization of U2AF1 inhibitors by integrating computer-aided design with chemical syntheses; 2) Study the therapeutic effects of U2AF1 inhibitors using in vitro and in vivo U2AF1mut MDS/AML models. In Aim 1, we will improve the activity and selectivity of U2AF1 inhibitors based on the preliminary structure activity relationship and selectivity data of SF1-8 by integrating chemical syntheses, computer-aided design, biochemical assays and the co- crystal structure determination. In Aim 2, we will assess the activities of our U2AF1 inhibitors in MDS/AML cellular models, analyze transcriptome and splicing pattern changes in cell lines treated with U2AF1 inhibitors and determine the effects of our inhibitors on erythropoiesis in xenograft mouse models. Our application is innovative and significant, because it builds on our discovery of the hit to U2AF1, an emerging novel target associated with the transformation of myeloid malignancies, and has the potential to establish a new therapeutic platform to treat MDS/AML by pharmacologically targeting mutant U2AF1 cells.

Narrative: Frequent mutations in splicing factor genes (including SF3B1 and U2AF1) were discovered in Myelodysplastic Syndrome (MDS) and Acute Myeloid Leukemia (AML) and associated with disease phenotypes. Although one compound targeting SF3B1 is in clinical studies, whether other mutant splicing factors can be therapeutic targets is unknown. This project aims to develop a new class of U2AF1 inhibitors and investigate the potential of U2AF1 as a therapeutic target in myeloid malignancies.