Understanding and targeting spliceosomal gene mutations in cancer

Abstract

Mutations in the spliceosomal genes SRSF2, U2AF1 and SF3B1 are commonly found in leukemia and are among the most common class of genetic alterations in myelodysplastic syndromes (MDS) and chronic lymphocytic leukemia. These mutations occur at highly restricted amino acid residues, are always heterozygous, and never co‐occur with one another. These data suggest that splicing mutations confer an alteration of function and/or that cells may only tolerate a certain degree of splicing modulation. We previously showed that mice expressing heterozygous Srsf2P95H mutation develop MDS‐like features due to altered RNA binding and splicing preference of the mutant protein. These features of the mutant SRSF2 protein are distinct from those seen with loss of 1 or copies of SRSF2, indicating that SRSF2 mutations confer an alteration of function. Specifically, mutations in SRSF2 alter its binding to exonic splicing enhancers (ESEs) such that the mutant protein recognizes C‐rich ESE sequences over G‐rich ESEs whereas the wildtype protein recognizes C‐ and G‐rich ESEs similarly. Recent work from others has revealed that mutations in the core spliceosomal protein U2AF1 also result in altered RNA biding and splicing preference based on the nucleotide sequence immediately surrounding the 3′ splice site. Thus, overall, each of the main mutations in RNA splicing factors alter splicing in a sequence‐specific manner but in a manner that are each distinct from one another.While the above data suggest that mutations in spliceosomal proteins confer an alteration of function, they do not explain why these mutations are heterozygous or why they are mutually exclusive with one another. In order to address these questions, we performed a series of murine genetic experiments to study the effect of (1) loss of the wildtype allele while concomitantly expressing a mutant splicing factor and (2) expressing 2 mutant splicing factors within the same cell concomitantly. Overall our murine genetic experiments suggest that cells bearing spliceosomal mutations do not tolerate additional perturbations to splicing function.Based on the above observations identifying a requirement for wildtype splicing function in cancer cells bearing a spliceosomal gene mutation, we hypothesized that splicing mutant leukemias might display greater sensitivity towards pharmacologic inhibition of splicing function than spliceosomal wildtype counterparts. To address this hypothesis, we tested the therapeutic efficacy of the spliceosome inhibitor E7107 in myeloid leukemias with and without spliceosomal mutations. More recently, in collaboration with H3 Biomedicine Inc., we have characterized a novel, orally bioavailable modulator of splicing, H3B‐8800. H3B‐8800 binds to the SF3b complex of the core spliceosome and modulates splicing in a time‐ and dose‐dependent manner. RNA‐seq analyses of the effects of H3B‐8800 relative to E7107 identify that H3B‐8800 preferentially results in intron retention of introns that are short and GC‐rich in content. Interestingly, mRNAs encoding splicing factors are highly enriched in such introns and are preferentially mis‐spliced upon exposure to H3B‐8800. This results in preferential effects of H3B‐8800 on spliceosomal mutant cancers including multiple models of MDS and AML. These findings have now resulted in the initation of a phase I clinical trial of H3B‐8800 for spliceosomal mutant leukemia which is underway now.This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.