Inhibition of MRP1 Induces Fetal Hemoglobin through NRF2 Activation to Protect Human Erythroid Cells from Sickling

Abstract

Introduction: Sickle cell disease (SCD) is a genetic disorder caused by a single nucleotide substitution (Glu6→Val) in the HBB gene coding for the β-globin of the hemoglobin A (HbA; α2β2) in erythroid cells. This point mutation leads to abnormal hemoglobin (HbS), which polymerizes in low oxygen condition and, ultimately, causes sickling of circulating erythrocytes leading to hemolysis, vascular inflammation, and painful vaso-occlusive crises. Hereditary persistence of fetal hemoglobin (HbF; α2γ2) in sickle cell patients significantly reduces clinical symptoms by preventing polymerization of HbS and subsequent cell sickling. Therefore, raising HbF expression in SCD patients is a promising therapeutic strategy. However, hydroxyurea, the standard of care for SCD and only FDA-approved HbF inducer, carries concerns for carcinogenicity-associated effects and the non-responders rate for this drug is relatively high (~30%). Consequently, there is a large unmet need for new therapies in SCD. Methods: To discover potential new HbF inducers, we conducted a genome-wide association study of HbF levels in 11,004 healthy donors from the INTERVAL study. We identified ABCC1, which encodes the multidrug resistant protein 1 (MRP1) and efflux pump for glutathione, as a putative causal gene modulating HbF expression. We detected a specific missense variant in ABCC1 which increases HbF expression in erythrocytes and strongly associates with plasma glutathione levels (β=0.8SD, P=3x10-38). To elucidate the role of MRP1 in HbF regulation, we performed a CRISPR-Cas9-mediated ABCC1 knockdown in erythroid HUDEP-2 cells and CD34+ hematopoietic stem cells and quantified the expression of HbF by flow cytometry, Western blot and high-performance liquid chromatography (HPLC) after erythroid differentiation (2-phase culture). We also generated HUDEP-2 clones homozygous for the p.R433S variant using CRISPR-Cas9 gene editing and measured HbF levels using the methods previously described. In CD34+ cells, MRP1 effect on HbF expression was also evaluated by using the selective MRP1-inhibitor, MK571. Transcriptomic, proteomics and phospho-proteomic analysis were performed to identify the mechanism of action involved in HbF induction after MRP1 inhibition. Expression of erythroid differentiation markers GATA1, ALAS2, LRF and Band 3 was assessed by Western blot to evaluate the impact of MRP1 inhibition on erythroid differentiation. Finally, we conducted sickling assays on MK571-treated CD34+ cells from SCD patients. Results:ABCC1 knockdown (90% gene silencing compared to control) in HUDEP-2 cells increased intracellular glutathione level (7−fold, P<0.01) and HbF (2.5−fold, P<0.05), as shown by flow cytometry, HPLC and Western blot analysis, after 7-day differentiation. Similarly, HUDEP-2 clones homozygous for the ABCC1-R433S allele presented an increase in intracellular glutathione (6-fold, P<0.05) and HbF expression levels (6-fold, P<0.01) compared to non-treated cells. In CD34+ cells knockdown for ABCC1 (95% gene silencing compared to control), HbF was increased (1.5−fold, P<0.05) after 14-day differentiation. Furthermore, pharmacological inhibition of MRP1, by MK571, in CD34+ cells led to an induction in glutathione (5-fold, P<0.01) and HbF (2-fold, P<0.05), and prevented sickling of CD34+ cells from SCD patients (2-fold decrease, P<0.05) in hypoxic condition (2% O2), without impacting erythroid differentiation. Finally, transcriptomic, proteomic, and phospho-proteomic integrated analyses in MK571-treated CD34+ cells revealed an activation of the NRF2-mediated oxidative stress pathway. Addition of ML385 (10µM), an NRF2 inhibitor, in MK571-treated CD34+ cells completely abolished HbF induction, confirming a NRF2-dependent HbF induction. Conclusion: In summary, we identified MRP1 as a HbF modulator in human erythroid cells and demonstrated that MRP1 inhibition induces HbF and prevents sickling of CD34+ cells through NRF2-mediated oxidative stress pathway activation without affecting erythroid differentiation, establishing MRP1 as a potential new therapeutic target for SCD. Future directions include replication of the ABCC1 genetic association in an independent cohort, and in vivo studies aiming to confirm the efficacy of MRP1 inhibition on improving SCD disease biomarkers.