Abstract
Myelodysplastic syndromes (MDS) and leukemias require the acquisition of multiple mutations during disease development resulting in clonal diversity and different responses. Splicing factors, transcription factors, epigenetic regulators, and cell signaling proteins are the common molecular events mutated during disease evolution and those events rarely occur alone. However, it remains unclear how the combinations of mutations in different categories may have cooperative effects in gene regulation and disease etiology. Mutations in the splicing factor SRSF2 and the transcription factor RUNX1 are closely associated in MDS patients, and their co-existence is linked to poor prognosis.
To understand the functional contribution of the coexistence in vivo, we utilized Mx1-Cre based conditional knock-in Srsf2-P95H mutation (P95H/+) mice, and Mx1-Cre based Runx1 conditional knockout mice (Runx1 f/f). We crossed these two strains to establish a new mouse model with inducible double mutations (Srsf2 P95H/+ Runx1Δ/Δ). Double mutant mice showed pancytopenia with MDS features including severe leukopenia in multiple lineages, macrocytic anemia, thrombocytopenia, and dysplastic morphology in peripheral blood. Double mutant mice also displayed more dramatic skewing toward the myeloid lineage at the expense of the B cell lineage when compared to single mutant mice. In competitive bone marrow transplantation assays, SRSF2 P95H cooperated with RUNX1 deficiency to confer a competitive disadvantage in vivo.
To investigate the mechanistic basis of this cooperation, differential splicing and gene expression were assessed by RNA sequencing of Lineage- c-kit+ cells isolated from WT, SRSF2 P95H, RUNX1 KO, and Double mutant bone marrow cells. Interestingly, deletion of the Runx1 gene alone resulted in significant changes to RNA splicing in 1120 genes, while the SRSF2 P95H mutation itself induced splicing changes in 935 genes. Furthermore, 2468 splice junctions in 1677 genes showed splicing changes in double mutant samples compared to wildtype controls. Among these altered splicing events, intriguingly, exon skipping was the major alteration in single and double mutants. Furthermore, the double mutant demonstrated increased aberrant splicing events when compared to the single mutants alone. We performed pathway analysis using the differentially spliced genes identified in double mutant cells. Pathways in cancer, DNA replication/repair, cell death and survival, hematological disease and inflammatory response were enriched. Splicing changes were detected in genes recurrently mutated in blood malignancies, including Fanca, Fance, Fancl, Ezh2, Atm, Gnas, Braf, Bcor, Fyn, and Wsb1 as well as in genes critical for splicing regulation, such as Srsf6, Fus, Hnrnpa2b1, and Srrm2. Gene expression analysis revealed 869 significantly differentially expressed genes in double mutant cells. Within the events in the double mutant population, 60% of the differentially expressed genes were also observed in RUNX1 single mutant cells, while only 2% of the differentially expressed genes were observed in SRSF2 single mutant cells, and 38% of the differentially expressed genes were uniquely presented in the double mutant cells. These results suggest that the gene expression program is heavily affected by loss of RUNX1 and the coexistence of an SRSF2 mutation contributes to certain synergistic effects in transcriptional regulation. Furthermore, we identified 101 genes that showed both differential splicing and expression, including Jak3, Jag2, Csf3r, Fcer1g, CD244 which are important in hematologic disorders.
Together, these results suggest that the deficiency of compound RUNX1 and SRSF2 P95H mutations impairs multi-lineage hematopoiesis and exacerbates the disease phenotypes caused by single mutations alone. At the genome-wide level, loss of the transcription factor RUNX1 itself dysregulates splicing outcomes and cooperates with the splicing factor SRSF2 P95H mutation to further perturb the expression and splicing of key regulators involved in hematopoietic stem/progenitor cell development, inflammatory responses, DNA damage, and RNA splicing.
Disclosures
No relevant conflicts of interest to declare.
The Caenorhabditis elegans LIN-26 protein is required to specify and/or maintain all non-neuronal ectodermal cell fates
