Abstract
Abstract
Abstract 922
SF3B1, a RNA splicing factor is frequently mutated in refractory anemia with ring sideroblasts (RARS) and RARS with thrombocytosis (RARS-T). Aside from its phenotypic importance, SF3B1 has been associated with good outcomes in MDS. Consistent with others, we found favorable survival outcomes in SF3B1 mutants. To explain the clinical phenotypes (good outcomes, anemia, and less progression to acute myeloid leukemia (AML)), we delved into the consequences of SF3B1 dysfunction in MDS. We hypothesized that SF3B1 mutations result in changes in RNA splicing of key genes in erythropoiesis resulting in anemia. Moreover, we postulated that SF3B1 is a founder mutation in RS formation but insufficient as a sole abnormality to induce a deleterious phenotype. The acquisition of genetic/ epigenetic aberrations may trigger the final phenotype. We reported abundant iron deposits in the mitochondria of SF3B1 mutants. We hypothesized that SF3B1 mutations change the composition/ chemical valence of iron leading to free radicals and DNA damage. We performed Energy-dispersive X-ray/ Electron-Energy Loss Spectroscopy on WT and SF3B1 mutants (n=2) finding that Fe2O3 is the most prevalent iron form. Changes in iron might predispose to DNA damage. We used flow cytometry to measure gamma-H2AX, a marker of DNA damage finding higher gamma-H2AX levels in bone marrow of WT (n=4; 26.35±28.36) vs SF3B1 mutants (n=8; 4.19±4.34), healthy subjects (n=4; 6.15±3.74) and other WT low-risk MDS (n=5; 7.6±1.6). Since increased DNA damage induces chromatin remodeling, we found that WT have more condensed chromatin with granular cytoplasm vs SF3B1 mutants (n=2). Further, RNA-sequencing showed alterations in histone deacetylases (HDAC1/HDAC2) and modification (ASXL1) in SF3B1 mutants. This links splicing mutations to epigenetic events and may represent the 2nd hit to induce the disease phenotype. Other targets were interrogated. Since SF3B1 interacts with members of Class II polycomb group (PcG) in Sf3b1+/− mice and PcG genes, like ASXL1/ EZH2 are mutated in MDS, we performed sequencing on PcG Class II genes (RNF2/ PCGF2). No mutation was found in RARS/RARS-T pts (n=34; SF3B1 mutant=26; WT=8). mRNA levels might be changed in these pts. We also elucidated the role of SF3B1 in the phenotype of anemia. RNA-sequencing has been instructive. Ribosomal proteins (RPS14 and RPS17) are implicated in the pathogenesis of 5q- MDS and Diamond-Blackfan anemia. Both RPS14/ RPS17 expression was different in SF3B1 mutant vs WT (n=2). Further, cytokine analysis showed a trend to higher IL-6 levels in WT vs mutants (n=3). IL-6 is associated with anemia and poor response to therapy in some diseases. We assessed the response to erythropoietin (EPO) and found that none of the WT RARS/-T (n=4) responded to EPO while 15/19 (79%) of SF3B1 mutants responded (p=0.008). It's suggested that SF3B1 mutants have better outcomes due to decreased AML transformation. To dissect this point, we looked at chromosomal defects in SF3B1 mutants (n=27) and WT (n=11), finding that 45% of WT carried complex karyotype (del(5q), del(17p) del(7), and i(17p); p=0.0009) while only 18% of mutants carried one additional abnormality (del(20q), inv3, and 16p). We also checked for concomitant mutations in genes of methylation (TET2, DNMT3A, IDH1/2), histone (ASXL1, UTX, EZH2), transcription (RUNX1, JAK2, TP53), signaling (CBL, NRAS, KRAS), and splicing (U2AF1, SRSF2). We analyzed the global mutational status in RARS/RARS-T (n=38; WT, n=11 and SF3B1 mutants, n=27) finding that 5/11 (45%) WT carried in total 6 other methylation (n=2), transcription (n=1), and splicing (n=3) gene mutations. Further, 15/27 (56%) of SF3B1 mutants carried a total of 16 mutations excluding SF3B1 in: methylation (n=9), transcription (n=4), splicing (n=1), histone (n=2). The most prevalent mutated genes (TET2, DNMT3A) predict for good treatment response. In-vitro treatment with decitabine (0.2 uM) showed changes in chromatin condensation, DNA damage, and response in SF3B1 mutants vs WT (n=2) suggesting that the pathogenetic effects of SF3B1 may be elicited through the methylation pathway. Lastly, SF3B1 predicts for good outcomes. SF3B1 may mediate effects on anemia through alterations in ribosomal function, cytokines, and epigenetic changes. Conversely, SF3B1 mutants have better survival and lower risk to AML evolution because of lower tendency to acquire poor prognostic defects.
Disclosures:
No relevant conflicts of interest to declare.
