Lords of the Iron Rings: Coordinated Missplicing of TMEM14C and ABCB7 Drives Ring Sideroblast Formation in SF3B1-Mutant MDS

SF3B1 is a key component of the spliceosome and is the most commonly mutated splicing factor in myelodysplastic syndromes (MDS), and frequently mutated in solid tumors. MDS with SF3B1 mutations is associated with the presence of bone marrow “ring sideroblasts” (MDS-RS) — a mysterious phenomenon in which iron-positive granules surround the erythroblast nucleus like a ring. MDS-RS results in anemia but usually has a relatively indolent disease course. SF3B1 mutations were recently classed as an MDS-defining genetic lesion in the current, 5th edition of the World Health Organization’s hematolymphoid classification.¹  The mutated form of SF3B1 introduces an upstream cryptic 3’ splice site causing alternative 3’ splicing. This produces abnormal transcripts that are often subject to degradation, resulting in reduced protein expression.^{2, 3}  Inherited sideroblastic anemias are often caused by mutations in genes required for iron and heme synthesis such as ALAS2 and ABCB7. Aberrant splicing of genes required for these pathways (e.g., ABCB7, TMEM14C, and PPOX)^{3, 4}  has previously been identified in SF3B1+ MDS, suggesting that they may be relevant to MDS-RS pathobiology.

Until now, disease models of MDS-RS have not successfully replicated the ring sideroblast phenotype. For example, SF3B1 K700E knock-in mice develop anemia and ineffective erythropoiesis but no ring sideroblasts.⁵  In a recent article published in Blood, Dr. Courtney Clough and colleagues report the development of the first disease model using induced pluripotent stem cells (iPSC) that successfully recapitulates the pathognomonic feature of ring sideroblasts in SF3B1-mutated MDS. The iPSCs were re-programmed from cells taken from a patient with SF3B1^G742D/1 and EZH2^R685H/1 mutations who had bone marrow ring sideroblasts of 40 to 50 percent, and the team created iPSC lines that were wild-type, SF3B1 single-mutant and SF3B1 EZH2 double-mutant, enabling study of the impact of each mutation individually and combined. When all three lines were subject to in vitro erythroid differentiation, a slight increase in mature erythroblast proportions was detected in SF3B1-only iPSCs, while SF3B1+EZH2+ had a significantly (twofold) reduced erythroid differentiation. Clear ring sideroblasts were seen in erythroblasts from SF3B1+ iPSCs at terminal differentiation.

Missplicing events in SF3B1+ iPSCs were detected in approximately 100 genes, including TMEM14C, PPOX, ABCB7, and MAP3K7. This led the authors to explore a potential role for these genes in ring sideroblast formation using “rescue” experiments, re-inducing expression of the misspliced genes to the parent iPSC cell line. Lentiviral vectors encoding each of these genes were transduced into the iPSC lines, to explore the impact on erythropoiesis. Reintroduction of the gene had no effect on erythroid differentiation for any of the targets assessed. Expression of MAP3K7 did not ameliorate ring sideroblast formation, and correction of PPOX resulted in only a marginal decrease. However, TMEM14C and ABCB7 both led to reduced ring sideroblast formation, and these two genes had an additive effect when combined, with the authors identifying ABCB7 is the dominant driver of ring sideroblast formation and a contributory role for TMEM14C. Therefore, they concluded that coordinated missplicing of TMEM14C and ABCB7 causes ring sideroblast formation in MDS.