Recent genetic studies have revealed frequent and specific pathway mutations involving multiple components of the RNA splicing machinery in myelodysplasia. Among these, SRSF2 mutations are more prevalent in CMML subtype and are associated with poor prognosis. Mutations showed a prominent hotspot involving proline 95, causing either P95H, P95L, or P95 conversion. Comprehensive analysis in our large cohort of MDS revealed that SRSF2 mutations showed a significant trend to coexist with TET2, STAG2, ASXL1 and RUNX1 mutations, while being mutually exclusive with EZH2 mutations. On the other hand, the molecular mechanism by which SRSF2 mutations lead to myelodysplasia remains largely unknown.
To elucidate the role of SRSF2 mutations in the development of myelodysplasia, we generated a heterozygous conditional knock-in mouse model of Srsf2 P95H mutation and crossed them with Vav1-Cre transgenic mice. Srsf2 P95H mutant mice exhibited macrocytic anemia, otherwise no significant changes in total peripheral blood (PB) cell counts compared to wild-type mice at 8-15 weeks after birth. There was no significant difference in lineage composition as well as blood cell morphology between wild-type and mutant mice in both bone marrow (BM) and PB. Flow cytometry of BM cells showed significant decrease of the number of hematopoietic stem cells (HSCs) and multipotent progenitor cells defined as Lin-Sca-1+Kit+ (LSK) fractions in Srsf2 P95H mice compared to wild-type mice. On the other hand, there were no significant differences in the number of more differentiated progenitor cells including common myeloid progenitors (CMPs), granulocyte/macrophage lineage-restricted progenitors (GMPs), megakaryocyte/erythrocyte lineage-restricted progenitors (MEPs), and common lymphoid progenitors (CLPs) between Srsf2 P95H and wild-type mice. These observations suggested that heterozygous Srsf2 mutation led to deregulation of hematopoietic stem cells, which however, is not sufficient for the development of MDS.
We next performed noncompetitive transplantation experiments to assess the cell intrinsic effects of Srsf2 P95H mutations. In PB, decreased white blood cell counts and progressive anemia were observed in mutant mice, which were evident as early as one month after transplantation. Cytological analysis of PB revealed morphological abnormalities in mice reconstituted with Srsf2 mutated cells, including hypersegmentation in neutrophils and dysplasia in the erythroid lineage. Srsf2 mutant-reconstituted mice showed normo-to-hypercellular marrow, where abnormal megakaryocyte distribution adjacent to trabecular bone and erythroid dysplasia was observed. Flow cytometrical analysis revealed decreased numbers of HSCs, LSK fractions and CMPs, whereas there was no significant change in the number of MEPs, GMPs and CLPs in BM. The BM erythroid progenitors were decreased in mutant-reconstituted mice, whereas the mutant mice showed splenic erythropoiesis with increased erythroid progenitors, suggesting the presence of extramedullary hematopoiesis, which was not seen in wild-type Srsf2 transduced mice. These observations suggested that the Srsf2 mutation led to ineffective hematopoiesis and morphological abnormalities, which seemed to recapitulate the phenotype of MDS.
Subsequently, we assessed the reconstitution capacity of whole BM cells from Srsf2 mutant mice in competitive transplantation experiments. The donor chimerism of Srsf2 P95H-derived cells in PB was significantly lower than that of wild-type cells. At 4 months post transplantation, the chimerism of Srsf2 P95H-derived cells was remarkably lower than that of wild-type cells in the fractions of HSCs, MPPs, CMPs, MEPs, GMPs and CLPs in BM. Furthermore, the reduced donor chimerism for Srsf2 P95H mutants was recapitulated in secondary transplantation experiments.
In summary, our results demonstrated that heterozygous P95H mutation of Srsf2 led to deregulation of hematopoietic stem cells that was evident from reduced competitive repopulation and impaired hematopoietic differentiation. Whereas mice reconstituted with Srsf2 mutant BM cells developed MDS-like phenotype in non-competitive transplantation setting, Srsf2 mutation by itself does not seem to be sufficient to develop MDS without transplantation, raising the possibility that an additional genetic and/or epigenetic events was required for overt MDS phenotype.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.