Can Leuk-warm Be Hot? Heating Up the Debate of Non-canonical Hotspot Spliceosomal Mutations in Myeloid Pathogenesis

Genes encoding RNA splicing factors, including SF3B1, SRSF2, and U2AF1, are frequently affected by somatic mutations in hematologic diseases such myelodysplastic syndromes (MDS), acute myeloid leukemia (AML), and chronic lymphocytic leukemia (CLL), as well as some solid malignancies. These missense mutations occur almost entirely in a highly specific set of “hotspot” residues in the gene locus.¹  The mutations more commonly occur in older individuals and are thought to drive clonal expansion of the aging hemopoietic system.²  Furthermore, functional studies indicate these perturbations drive myeloid disease pathogenesis and have been shown to confer an increased risk of transformation to acute leukemia.³ 

While the bulk of spliceosome gene mutations affect few hotspot locations, a minority of patients with hematologic malignancies do express non-hotspot mutations, the relevance of which remain unclear but often are presumed not pathogenic. Motivated by the observation that novel SF3B1 in-frame deletions mimicked splicing profiles of patients with hotspot SF3B1 mutations in CLL,⁴  in the current study Dr. Joseph Pangallo and colleagues tested this assumption by systematic characterization of diverse rare and private spliceosomal mutations to infer their disease relevance. They hypothesized that rare or private SRSF2 and U2AF1 mutations that phenocopied hotspot-induced changes in splicing were candidate drivers, while mutations that induced few or no splicing changes were likely passengers.

The authors queried the COSMIC database to identify non-hotspot somatic mutations in SRSF2 and U2AF1 and established isogenic cell lines of 14 selected rare and private mutations via transgenic expression in K562 cells. High-coverage RNA sequencing of each of these cell lines induced distinct alterations in RNA splicing. The investigators then proceeded to perform experiments demonstrating that both rare and hotspot SRSF2 mutations alter SRSF2’s RNA-binding affinity and avidity to induce sequence-specific changes in exonic splicing enhancer preference. These findings were further validated in primary patient samples, which recapitulated spatially restricted recognition of C- versus G-rich SSNG motifs observed from their cell culture results. The authors next queried rare and private mutations in U2AF1, which expresses two hotspot mutations, giving rise to two distinct changes in RNA-binding specificity and 3′ splice site recognition. They observed both complete and dual phenocopy of altered 3′ splice site recognition in several rare and private mutations and validated these findings in mutation-matched patient samples. Lastly, there was the identification of only three genes that were differentially spliced in association with both SRSF2 and U2AF1 mutations. These included H2AFY and IRAK4, both with known involvement in hematologic disease.^5,6