Adenine base editor-mediated correction of the common and severe IVS1-110 (G>A) β-thalassemia mutation

Abstract

β-Thalassemia (BT) is one of the most common genetic diseases worldwide and is caused by mutations affecting β-globin production. The only curative treatment is allogenic hematopoietic stem/progenitor cells (HSPCs) transplantation, an approach limited by compatible donor availability and immunological complications. Therefore, transplantation of autologous, genetically-modified HSPCs is an attractive therapeutic option. However, current gene therapy strategies based on the use of lentiviral vectors are not equally effective in all patients and CRISPR/Cas9 nuclease-based strategies raise safety concerns. Thus, base editing strategies aiming to correct the genetic defect in patients' HSPCs could provide safe and effective treatment. Here, we developed a strategy to correct one of the most prevalent BT mutations (IVS1-110 [G>A]) using the SpRY-ABE8e base editor. RNA delivery of the base editing system was safe and led to ∼80% of gene correction in the HSPCs of patients with BT without causing dangerous double-strand DNA breaks. In HSPC-derived erythroid populations, this strategy was able to restore β-globin production and correct inefficient erythropoiesis typically observed in BT both in vitro and in vivo. In conclusion, this proof-of-concept study paves the way for the development of a safe and effective autologous gene therapy approach for BT.

Graphical abstract

Figure 1.

A base editing strategy to efficiently correct the IVS1-110 (G>A) mutation. (A) gRNAs 1 to 6 were manually designed to place the IVS1-110 (G>A) mutation in positions 3 to 8 of the protospacer. The mutation is highlighted with a grey box. (B) Overview of the cell collection for testing the ability of gRNA/BE combinations to correct the IVS1-110 (G>A) mutation. (C) Frequency of corrected alleles (normalized to the frequency of GFP⁺ cells) and indels as assessed by Sanger sequencing in T cells electroporated with different combinations of synthetic gRNAs and ABE mRNAs. Data are expressed as mean ± standard error of the mean (SEM) (n = 3, 2 donors). (D) Representative percent composition of Sanger sequencing traces measured to be significantly different from noise (in red), as assessed by EditR following Sanger sequencing in T cells electroporated with gRNA1/SpRY-ABE8e (cor) or with TE buffer (TE). The target base position is outlined with a red box and the observed bystander edits with black boxes. Single nucleotide polymorphisms rs777028217 (G>A) and rs1480884739 (T>C) mapped to positions 6 and 8 are not associated with a clinical phenotype. PBMC, peripheral blood mononuclear cells; cor, corrected.

Figure 2.

ABE-mediated correction of the IVS1-110 (G>A) mutation in BT HSPCs restores normal Hb production in their erythroid progeny. (A) The frequency of corrected alleles (BE) and indels as assessed by Sanger sequencing in BTHSPCs. Data are expressed as mean ± SEM (3 donors). The frequency of corrected alleles in the cells obtained from the compound heterozygous patient (BT2) was corrected considering only the alleles harboring the IVS1-110 (G>A) mutation. The dotted line represents the maximum background noise of indels calculated by TIDE. (B) RT-qPCR using primers detecting exclusively correctly spliced β-globin mRNAs in erythroid cells derived from BT HSPCs (cor). β-globin expression was normalized to α-globin. Data are expressed as mean ± SEM. The dotted lines indicate the maximum and minimum values observed in HD cells. (C) RT-PCR using primers amplifying a region spanning the HBB exon 1-exon 2 junctions. cDNA was obtained from erythroid cells derived from BT HSPCs (cor). (D) Expression of β-, ^Gγ-, ^Aγ- and δ- globin chains measured by RP-HPLC in BT and HD RBCs. β-like-globin expression was normalized to α-globin. The α-/non–α-globin ratio is reported on top of the graph. RBCs were obtained from BT HSPCs (cor). Data are expressed as mean ± SEM. (E) Analysis of HbA, HbF, and HbA₂ by CE-HPLC in BT and HD RBCs. We calculated the percentage of each Hb type over the total Hb tetramers. RBCs were obtained from corrected BT HSPCs (cor). (B-C) As controls, we used erythroid cells derived from BT or HD HSPCs electroporated only with SpRY-ABE8e mRNA (BE) (3 patients with BT and 2 HDs). (D-E) As controls, we pooled data obtained in RBCs derived from BT or HD HSPCs electroporated with TE or with SpRY-ABE8e mRNA only (3 patients with BT and 2 HDs). Data are expressed as mean ± SEM. cor, corrected; ctr, control.

Figure 3.

Efficient reversion of the IVS1-110 (G>A) mutation in BT HSPCs corrects ineffective erythropoiesis. (A-C) Frequency of CD36⁺ (A), CD71⁺ (B), and GPA⁺ (C) cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis. Control BT or HD cells were either electroporated with TE buffer (TE), or only with SpRY-ABE8e mRNA (BE). Data are expressed as mean ± SEM (3 patients with BT and 3 HDs). ∗P ≤ .05; ∗∗P ≤ .01 (paired t test; BT-BE vs BT-cor). (D) Mean Fluorescence Intensity (MFI) in ROS-containing cells (DCFDA⁺) for control and edited (cor) BT samples. For the control, we pooled data obtained in RBCs derived from BT HSPCs electroporated with TE or with SpRY-ABE8e mRNA only (3 patients with BT). ∗P ≤ .05 (paired t test). (E) Flow cytometry histograms showing the frequency of apoptotic cells (annexin V⁺ cells) in the 7AAD⁻ cell population in unstained (Uns), BT and HD samples at day 13 of erythroid differentiation (3 patients with BT and 3 HDs). (F) Frequency of enucleated cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis of cells stained with the DRAQ5 nuclear dye (3 patients with BT and 3 HDs). Data for HD samples are expressed as mean ± SEM. (G) Cell size of enucleated cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis of the median FSC-A intensity (3 patients with BT and 3 HDs). Data for HD samples are expressed as mean ± SEM. ∗P ≤ .05; ∗∗P ≤ .01 (paired t test; BT-BE vs BT-cor). (H) Single RBC parameters (perimeter, surface, optical volume, dry mass, and surface density) evaluated by quantitative phase image microscopy in RBCs obtained from BT HSPCs (cor). As controls, we used RBCs derived from BT or HD HSPCs electroporated only with SpRY-ABE8e mRNA (BE). Data are expressed as mean ± SEM (3 patients with BT and 2 HDs). ∗∗∗∗P ≤ .0001 (Ordinary one-way ANOVA). cor, corrected; ctr, control; D, day.

Figure 3.

Efficient reversion of the IVS1-110 (G>A) mutation in BT HSPCs corrects ineffective erythropoiesis. (A-C) Frequency of CD36⁺ (A), CD71⁺ (B), and GPA⁺ (C) cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis. Control BT or HD cells were either electroporated with TE buffer (TE), or only with SpRY-ABE8e mRNA (BE). Data are expressed as mean ± SEM (3 patients with BT and 3 HDs). ∗P ≤ .05; ∗∗P ≤ .01 (paired t test; BT-BE vs BT-cor). (D) Mean Fluorescence Intensity (MFI) in ROS-containing cells (DCFDA⁺) for control and edited (cor) BT samples. For the control, we pooled data obtained in RBCs derived from BT HSPCs electroporated with TE or with SpRY-ABE8e mRNA only (3 patients with BT). ∗P ≤ .05 (paired t test). (E) Flow cytometry histograms showing the frequency of apoptotic cells (annexin V⁺ cells) in the 7AAD⁻ cell population in unstained (Uns), BT and HD samples at day 13 of erythroid differentiation (3 patients with BT and 3 HDs). (F) Frequency of enucleated cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis of cells stained with the DRAQ5 nuclear dye (3 patients with BT and 3 HDs). Data for HD samples are expressed as mean ± SEM. (G) Cell size of enucleated cells at days 13, 16, and 20 of erythroid differentiation, as measured by flow cytometry analysis of the median FSC-A intensity (3 patients with BT and 3 HDs). Data for HD samples are expressed as mean ± SEM. ∗P ≤ .05; ∗∗P ≤ .01 (paired t test; BT-BE vs BT-cor). (H) Single RBC parameters (perimeter, surface, optical volume, dry mass, and surface density) evaluated by quantitative phase image microscopy in RBCs obtained from BT HSPCs (cor). As controls, we used RBCs derived from BT or HD HSPCs electroporated only with SpRY-ABE8e mRNA (BE). Data are expressed as mean ± SEM (3 patients with BT and 2 HDs). ∗∗∗∗P ≤ .0001 (Ordinary one-way ANOVA). cor, corrected; ctr, control; D, day.

Figure 4.

Correction of the IVS1-110 (G>A) mutation in repopulating HSCs. (A) Overview of the experimental protocol of HSPC xenotransplantation. BT HSPCs were subjected to RNA-mediated base editing and xenotransplanted into NBSGW immunodeficient mice. HD and BT HSPCs electroporated with TE buffer or only with SpRY-ABE8e mRNA were injected as controls. Peripheral blood analysis was performed at weeks 9 and 16. Mice were euthanized 16 weeks after engraftment, after Clo-Lip injection, and their hematopoietic tissues and organs were collected and analyzed. (B) Engraftment of human cells in NBSGW mice transplanted with HD or BT control (HD-ctr; BT-ctr) or corrected (BT-cor) HSPCs 16 weeks post-transplantation (HD-ctr, n = 4; BT-ctr, n = 4; BT-cor, n = 5). Engraftment is represented as the percentage of human CD45⁺ cells in the total murine and human CD45⁺ cell population in peripheral blood, bone marrow, spleen, and thymus. Each data point represents an individual mouse. The mouse with the lowest chimerism is indicated with the symbol ◓. Data are expressed as mean ± SEM. (C) Frequency of human T (CD3) and B (CD19) lymphoid, myeloid (CD14, CD15, and CD11b), erythroid (GPA, CD36, CD71), and HSPC (CD34) cells in BM 16 weeks after the transplantation (HD-ctr, n = 4; BT-ctr, n = 4; BT-cor, n = 5). Each data point represents an individual mouse. Data are expressed as mean ± SEM. (D) Human hematopoietic progenitor content in BM human CD45⁺ cells derived from mice transplanted with control and edited HSPCs (HD-ctr, n = 4; BT-ctr, n = 4; BT-cor, n = 5). We plotted the percentage of human CD45⁺ cells giving rise to BFU-E and CFU-GM. Data are expressed as mean ± SEM. (E) Base editing efficiency, calculated by the EditR software, in input, peripheral blood-, bone marrow- and spleen-derived HD, and BT human samples subjected to Sanger sequencing. Data are expressed as mean ± SEM (BT-cor, n = 5). The frequency of base editing in the input was calculated in cells cultured in the HSPC medium (▲), in liquid erythroid cultures (▼), BFU-E (■) and CFU-GM (◆) colonies. Each data point represents an individual mouse. Data are expressed as mean ± SEM. ctr, control. BFU-E, burst forming units-erythroid; CFU-GM, colony forming units-erythroid.

Figure 5.

Correction of the IVS1-110 (G>A) mutation in xenotransplanted BT HSPCs rescues the ineffective erythropoiesis in vivo. (A) Frequency of ROS-containing (DCFDA⁺) human GPA⁺ erythroid cells derived from the bone marrow of mice transplanted with HD or BT control (HD-ctr; BT-ctr) or corrected (BT-cor) HSPCs 16 weeks after the transplantation (HD-ctr, n = 3; BT-ctr, n = 4; BT-cor, n = 4). ∗∗P ≤ .01 (unpaired t test; BT-ctr vs BT-cor). We plotted the fold change relative to BT-ctr samples. (B) Frequency of enucleated cells as measured by the flow cytometry analysis of cells stained with the DRAQ5 nuclear dye in human GPA⁺ erythroid populations from the bone marrow of mice transplanted with HD or BT control (HD-ctr; BT-ctr) or corrected (BT-cor) HSPCs 16 weeks after the transplantation (HD-ctr, n = 3; BT-ctr, n = 4; BT-cor, n = 4). ∗P ≤ .05 (unpaired t test; BT-ctr vs BT-cor). (C) Representative RP-HPLC chromatograms from sorted human GPA⁺ bone marrow erythroid cells 16 weeks posttransplantation. (D) α/non–α ratio calculated based on RP-HPLC data from sorted human GPA⁺ bone marrow erythroid cells obtained from mice transplanted with HD or BT control (HD-ctr; BT-ctr) or corrected (BT-cor) HSPCs 16 weeks after the transplantation (HD-ctr, n = 4; BT-ctr, n = 4; BT-cor, n=5). The dotted lines indicate minimum and maximum values observed in HD-ctr samples. ∗P ≤ .05 (unpaired t test; BT-ctr vs BT-cor). (E) Frequency of human RBCs in total peripheral blood 4 days after Clo-Lip injection in mice transplanted with HD or BT control (HD-ctr; BT-ctr) or corrected (BT-cor) HSPCs 16 weeks after the transplantation (HD-ctr, n = 4; BT-ctr, n = 4; BT-cor, n = 5). (F) Representative RP-HPLC chromatograms from sorted human circulating RBCs 16 weeks after the transplantation (BT-ctr, n = 4 pooled samples; BT-cor, n = 1 representative graph; HD-ctr, n = 1 representative graph). cor, corrected; ctr, control.