Cooperative Effects of SRSF2 and STAG2 mutations on Development of Myelodysplastic Syndrome and Its Related Disorders

MDS and its related disorders frequently carry mutations in multiple genes implicated in RNA splicing and epigenetic regulation. Among these, mutations in SRSF2 and STAG2 co-occur more frequently than just by chance, suggesting a functional interaction between both mutations during the development of MDS. However, the leukemogenic mechanism of this combination remains unclear.

To elucidate the functional relevance of this combination of mutations, we first analyzed the mutational profile of MDS/AML harboring SRSF2 and STAG2 mutations using published data of MDS or secondary AML patients (N=3,047). In total, SRSF2 and STAG2 mutations were found in 12.0% and 6.8%, respectively, and 86 patients (2.8%) had both mutations. SRSF2 and STAG2 mutations were significantly co-mutated, with frequent additional mu­­tations in ASXL1, RUNX1, BCOR, IDH2 and CEBPA. Analysis of variant allele frequencies suggests that in typical cases, SRSF2 mutations are likely to be acquired first, followed by STAG2 mutations, where multiple independent STAG2 mutations evolved in many cases. Higher-risk MDS and secondary AML accounted for 59.3% and 26.7% of SRSF2 and STAG2 double-mutated patients, respectively. Among MDS patients, those with both mutations tended to have lower WBC and platelet counts, higher blast counts in peripheral blood, higher IPSS-R scores and a poorer overall survival. These results suggest that SRSF2 and STAG2 mutations co-occurred in MDS/AML, promoting disease progression of MDS to secondary AML.

Next, to reveal the functional mechanisms by which this combination cooperates to develop MDS in vivo, we crossed mice carrying Srsf2-P95H conditional knock-in and Stag2 conditional knockout alleles with an Mx1-Cre transgene to generate double mutant mice. Their phenotype was evaluated in the transplant setting, in which bone marrow derived from double mutant mice, as well as WT or single mutant mice, was transplanted into the lethally irradiated mice and pIpC was injected after engraftment. Compared to single mutant- or mock-transplanted mice, double-mutant transplanted mice showed lower WBC counts at 8-16 weeks after pIpC administration, with increased granulocytes/monocyte and decreased B lymphocyte subfractions. Moreover, double mutant mice showed a lower hemoglobin level and larger RDW and MCV. These results suggest that the combination of Srsf2 and Stag2 mutations leads to more severe leukopenia and macrocytic anemia than single mutations.

In bone marrow, Lin⁻Sca-1⁺Kit⁺(LSK) cells were increased in Stag2 single mutant mice and double mutant mice, compared with wild-type mice and Srsf2 single mutant mice. Within the myeloid progenitor (MP) compartment, we observed increased common myeloid progenitors (CMPs) and granulocyte-monocyte progenitors (GMPs) in Stag2 single and double mutant mice, compared to wild-type or Srsf2 single mutant mice. By contrast, megakaryocyte/erythroid progenitors (MEPs) and CD71+ Ter119+ erythroblasts strikingly decreased in double mutant mice. Double mutant mice also showed enlarged spleen, in which an increase MEP as well as CMP and GMP fractions were observed. These results suggest that the combination of Srsf2 and Stag2 mutations skews erythroid lineage toward myeloid lineage in bone marrow and causes extramedullary hematopoiesis in the spleen.

Given that Srsf2 gene is a key component of spliceosome machinery, we next assessed different splicing changes in double mutant mice. RNA sequencing of samples obtained from LSK cells showed increased aberrant splicing events of cassette exons in double mutant mice, indicating that the combination of Srsf2/ Stag2 mutations leads to the progression of MDS through aberrant splicing changes.

In conclusion, our findings suggest that SRSF2 and STAG2 mutations cooperatively induced more impaired hematopoietic differentiation compared to single mutations alone and contribute to the progression of MDS, possibly thorough aberrant splicing changes.