SUMMARY Genes encoding RNA splicing factors are the most common class of mutations in patients with myelodysplastic syndromes (MDS) and are also common across all other forms of myeloid malignancies. These leukemia- associated “spliceosomal mutations” primarily occur in four genes: SF3B1, SRSF2, U2AF1, and ZRSR2. In three of these four genes (SF3B1, SRSF2, and U2AF1), the mutations occur at specific amino acid residues in a heterozygous manner (so-called “mutational hotspots”) and cause gain/alteration of function. In contrast, mutations in ZRSR2 occur throughout the open reading frame and appear to confer loss of function. Moreover, ZRSR2's normal function makes it unique amongst the commonly mutated RNA splicing factors in leukemias: ZRSR2 is the only frequently mutated factor that primarily functions in the recognition of a rare class of introns known as “minor introns.” Thus, ZRSR2 mutations are significantly enriched in leukemia and exhibit a unique genetic spectrum and function amongst recurrent spliceosomal mutations, yet they are comparatively poorly studied and understood compared to mutations in SF3B1, SRSF2, and U2AF1. Here, we propose to determine the mechanistic, functional, and therapeutic consequences of ZRSR2 mutations in leukemia. Our interdisciplinary team consists of a physician-scientist with expertise in leukemia biology and patient care (Abdel-Wahab) and a basic scientist with expertise in RNA splicing and functional genomics (Bradley). As minor introns are far more conserved than are most other introns, we hypothesize that a cross-species comparisons of the effects of ZRSR2 loss will be particularly useful for understanding how molecular alterations in splicing drive malignant transformation. In addition, we hypothesize that aberrant splicing induced by ZRSR2 loss will enable novel therapeutic approaches. In preliminary experiments, we generated a Zrsr2 conditional knockout (cKO) mouse, assembled a relevant patient cohort, characterized the transcriptomes of our Zrsr2 cKO mouse and ZRSR2-mutant MDS, and performed a functional genomic screen to model and prioritize ZRSR2-regulated splicing events. These studies revealed that ZRSR2 mutations cause mis-splicing of a compact set of genes, that Zrsr2 loss promotes aberrant and increased hematopoietic stem cell self-renewal, that simultaneous ZRSR2 and TET2 collaborate to drive malignancy, and that mis-splicing of specific downstream targets of ZRSR2 promotes clonality. We propose to build on these preliminary studies as follows: Aim 1, Determine how ZRSR2 mutations dysregulate the transcriptome and proteome in leukemia; Aim 2, Determine how disruption of ZRSR2-regulated splicing events drives clonal advantage; Aim 3, Identify the functional basis for the frequent co-occurrence of ZRSR2 and TET2 mutations in leukemia. The significance of these studies is that they will elucidate mechanistic and functional connections between ZRSR2 mutations, RNA mis-splicing, and the initiation of myeloid neoplasms. The health relatedness is that the proposed work may reveal new therapies for MDS and leukemia that specifically kill ZRSR2-mutant cells.

NARRATIVE Leukemias are cancers in which the bone marrow produces too many abnormal blood cells at the expense of normal blood cells. Here we will determine how a commonly occurring change in a gene called ZRSR2, which encodes a protein that participates in the process of RNA splicing, gives rise to leukemia. We will determine how ZRSR2 mutations drive leukemia and find new ways to treat leukemias with ZRSR2 mutations.