Targeted Base Editing Strategies for Beta-Hemoglobinopathies

Abstract

β-hemoglobinopathies are genetic anemias caused by a reduced or abnormal synthesis of the adult hemoglobin β-chain. In β-thalassemia, the reduced (β+) or absent (β0) production of β-chains causes α-globin precipitation and death of red blood cell (RBC) precursors. In sickle cell disease (SCD), a single amino acid change (β6Glu→Val) in the adult hemoglobin (Hb) βS-chain causes Hb polymerization with consequent RBC sickling, vaso-occlusive crises, organ damage and reduced life expectancy. Transplantation of autologous, genetically modified hematopoietic stem/progenitor cells (HSPCs) is an attractive therapeutic option. However, current gene therapy strategies based on the use of lentiviral vectors or CRISPR/Cas9 nuclease are not equally effective in all the patients and/or raise safety concerns. Base editing (BE) allows the introduction of point mutations (C>T by cytidine base editors, CBEs; A>G by adenine base editors, ABEs) without generating dangerous double strand breaks. In this study, we have developed BE strategies aiming either to correct the genetic defect or to target disease modifiers in patient cells, and provide a safe and effective treatment for β-hemoglobinopathies. First, we exploited the capacity of ABEs in combination with single guide RNAs (gRNAs) to correct prevalent β-thalassemic mutations (the CD39 and IVS2-1 β0 mutations and the IVS1-110 β+ mutation). RBCs derived from edited HSPCs exhibited high β-globin levels. The delayed erythroid differentiation typically observed in β-thalassemic cell cultures was corrected by our treatment. Measurement of Annexin+ cells by flow cytometry showed a substantially reduced apoptotic rate in edited β-thalassemic samples compared to untreated controls. Finally, off-target activity was minimal and by-stander edits (additional base conversions occurring close to the target base) did not impact the β-globin expression. In parallel, we used BE to target disease modifiers such as the fetal γ-globin genes. In fact, the co-inheritance of genetic mutations causing a sustained fetal β-like γ-globin chain production in adult life (hereditary persistence of fetal hemoglobin (HbF), HPFH) reduces the clinical severity of β-hemoglobinopathies. HPFH mutations in the promoter of the two γ-globin genes, HBG1 and HBG2 eitherdisrupt the binding sites (BS) for transcriptional repressors (e.g., LRF) or generate BS for transcriptional activators (e.g., KLF1). BE allowed the introduction of HPFH mutations in the HBG promoters, which disrupt the LRF BS (via CBE) or generate the KLF1 BS (via ABE). This led to therapeutically relevant HbF levels in RBCs differentiated from edited β-thalassemia or SCD HSPCs, and rescued the pathological phenotypes (correcting ineffective erythropoiesis in β-thalassemia and the sickling phenotype in SCD RBCs). Finally, serial xenotransplantation experiments showed BE in long-term HSCs; however, the efficiency was reduced compared to input HSPCs. RNA-seq analysis showed no alteration for ABE-treated HSPCs, but activation of the p53 pathway and dysregulation of genes involved in HSC biology in CBE-treated samples. To increase the fitness of edited HSCs and editing efficiency, we optimized the RNA delivery of the BE machinery and used histone deacetylase, RNase and apoptosis inhibitors. Moreover, we developed a strategy to enrich edited cells. By way of example, two rounds of mRNA electroporation led to increased BE efficiency with minimal toxicity. Furthermore, highly processive base editors outperformed the original enzymes even upon only one round of electroporation. In conclusion, we developed efficient BE approaches to either correct highly prevalent mutations or induce HbF expression. Overall, our treatments led to correction of the pathological phenotypes and optimal BE efficiencies to finally enable the clinical development of base-edited HSPCs for β-hemoglobinopathies.