SF3B1-mutant models of RNA mis-splicing uncover UBA1 as a therapeutic target in myelodysplastic neoplasms

Abstract

Myelodysplastic syndromes with somatic mutations in the splicing factor SF3B1 gene (MDS-SF3B1) result in RNA mis-splicing, erythroid dysplasia and ultimately refractory anemia. Precision medicine approaches for MDS-SF3B1 remain challenging due to both the complexity of the mis-splicing landscape and its evaluation in disease-accurate models. To uncover novel RNA mis-splicing events, isogenic SF3B1^K700E and SF3B1^WT iPSC lines from an MDS-SF3B1 patient were differentiated into hematopoietic cells and analyzed via unsupervised splicing event profiling using full-length RNA sequencing. This identified SF3B1^K700E-specific mis-splicing of ubiquitin-like modifier activating enzyme 1 (UBA1), which encodes a key E1 protein at the apex of the ubiquitination cascade. UBA1 mis-splicing (UBA1^ms) introduced protein instability and decreased total UBA1 levels, rendering mutated cells susceptible to the small-molecule UBA1 inhibitor TAK-243. Analysis of CD34⁺ RNA sequencing data from an MDS patient cohort confirmed unique and ubiquitous UBA1^ms in MDS-SF3B1 patients, absent in other splicing factor-mutated MDS cases or healthy controls. TAK-243 selectively targeted MDS-SF3B1 primary CD34⁺ cells and reduced mutant cell numbers in colony-forming assays. In contrast, normal hematopoietic progenitor cells were unaffected. Altogether, we here define UBA1^ms as a novel therapeutic vulnerability in SF3B1-mutant cells, introducing UBA1 inhibition as a potential avenue for future MDS-SF3B1 treatments.