Transcriptional Regulation of Ferroptosis in X-Linked Sideroblastic Anemia

(Background) Ferroptosis, a form of regulated cell death dependent on iron accumulation and lipid peroxidation, has been recently revealed to cause various types of oxidative stress-related diseases; however, the link between ferroptosis and hematopoietic disorders has not been fully explored. Erythropoiesis consumes large amounts of iron, and both congenital and acquired conditions associated with iron overload can cause premature erythroid death. X-linked sideroblastic anemia (XLSA), the most common form of congenital sideroblastic anemia, is the paradigmatic example of this disorder. In XLSA, germline mutations in the erythroid-specific 5-aminolevulinate synthase (ALAS2) gene results in mitochondrial iron overload and subsequent erythroid death, but the precise mechanism remains unknown. We hypothesized that ferroptosis is linked to XLSA pathophysiology.

(Aims) To explore the pathological link between ferroptosis and XLSA.

(Methods) Cellular XLSA models were previously established using human umbilical cord blood-derived erythroid progenitor (HUDEP)-2 cells (Ono et al. ASH2019). Based on CRISPR/Cas9, we introduced ALAS2 mutations substituting arginine at amino acid residue 170, one of the XLSA hot spots, with leucine (R170L) or histidine (R170H). Erythroid differentiation was induced by co-culture with OP9 cells in a sodium ferrous citrate-supplemented medium. Gene expression levels and intracellular protein concentration were evaluated using quantitative polymerase chain reactions and Western blot analysis, respectively. Cell death and lipid hydroperoxides were assessed by flow cytometry. Intracellular glutathione levels were measured colorimetrically. For gene expression profiling, Human Oligo chip 25k (Toray Industries, Tokyo, Japan) was used.

(Results) After erythroid differentiation, XLSA clones (ALAS2 R170L and R170H) showed increased numbers of ring sideroblasts. Furthermore, beta-globin gene (HBB) expression and intracellular heme concentration were significantly lower than in the wild-type controls. Under electron microscopy, XLSA clones exhibited aberrant mitochondrial iron deposits and remarkable increase in mitochondrial spheroids, which indicate response to oxidative stress. Cell viability was comparable between the XLSA clones and wild-type controls. In contrast, the XLSA clones showed a higher proportion of cell death than the wild type after treatment with erastin, a ferroptosis inducer. Lipid hydroperoxides showed significantly higher expression in XLSA clones than in the wild type; this was improved after treatment with deferoxamine, an iron chelator. These results suggested that the XLSA clones were accompanied by increased levels of oxidative stress and reduced resistance against erastin. We also studied transcription factor BTB domain and CNC homolog 1 (BACH1), a key regulator of iron metabolism and accelerator of ferroptosis. In the XLSA clones, BACH1 levels were significantly higher than in the wild type, presumably because BACH1 degradation was decreased in conditions with a low heme concentration. Concomitantly, genes repressed by BACH1 showed reduced expression, and some of the downregulated genes were associated with ferroptosis-inhibitory mechanisms such as glutathione synthesis (SLC7A11, GCLM and GCLC) and iron metabolism (SLC40A1, FTH1, and FTL). As a result, intracellular glutathione levels were significantly repressed in XLSA clones. Collectively, we think that BACH1 promoted XLSA clones prone to ferroptosis by transcriptionally regulating gene expression.

(Conclusion) Our observations supported our hypothesis that ferroptosis is associated with XLSA pathophysiology. Further study is necessary to establish a therapeutic strategy targeted against ferroptosis.