The S878F mutation leads to hypomethylation-mediated reactivation of γ-globin. (A-B) Analysis of protein-protein interaction by coimmunoprecipitation in CD34⁺ cells with or without S878F mutation. Proteins copurified from CD34⁺ cell nuclear extracts were analyzed by western blotting (WB) with specific antibodies against target proteins (A) or anti-DNMT1 antibody (B) using isotype-matched immunoglobulin G (IgG) antibody as a negative control. All of the inputs are 5%. The WT and S878F samples are marked as WT (left column) or S878F (right column) at the top of the graphs. Each of the different target proteins examined are indicated at the right of the strips. LRF protein could be not detected in either WT or S878F cells using anti-DNMT1 antibodies, implying that an indirect interaction may exist between DNMT1 and LRF. (C) Analysis of the phosphorylation status of the serine residue at position 878 of DNMT1 in nuclear lysates of HuDEP-2 (left) or CD34⁺ cells (right). HuDEP-2 cells were transfected with WT or S878F DNMT1 constructs. Proteins copurified from constructed HuDEP-2 cells or CD34⁺ cells were analyzed by WB with the antiphosphorylated form, BAH1 Ser878 (pDNMT^S878) antibody, using anti-DNMT1 and anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibodies as controls. (D) Analysis of the protein degradation of DNMT1 in HuDEP-2 cells transfected with vectors encoding enhanced green fluorescent protein (EGFP)-tagged full-length WT (left) or mutated DNMT1 (right). Seventy-two hours after transfection, cells were treated with cycloheximide and collected at 4 time points (0 hours, 4 hours, 6 hours, and 8 hours). DNMT1 was detected by WB using anti-green fluorescent protein (GFP) antibody and GAPDH as a loading control. The degradation curve graph is shown at the bottom. (E) Assay of DNA methyltransferase activity in nuclear extracts of HuDEP-2 cells transfected with vectors encoding EGFP-tagged full-length WT or mutation DNMT1. DNA methyltransferase activity was determined using a hemi-methylated DNA-trapping assay. (F) ChIP-qPCR analysis of the relative enrichment of DNMT1 at the γ-globin promoter in CD34⁺ cells. The GAPDH promoter served as positive control (+ctrl) and the β-globin promoter served as negative control (−ctrl). Results were normalized to the positive control. (G) Effects of the S878F mutation on the HBG promoter methylation level were examined by bisulfite-sequencing method in CD34⁺ cells. Graphic location of analyzed CpG sites in the promoter regions with sequence homology in 2 HBG genes (^Gγ and ^Aγ) is shown in the left upper region. Sequence variations of 2 subjects with or without the S878F mutation are shown at the left lower section of the sequencing chromatograph. Right panel, methylation levels at 6 CpG sites (−162, −53, −50, +6, +17, and +50) obtained from tested cases with or without the S878F mutation. (H-K) Quantitative measurement of HBG mRNA expression by qPCR (H-I) and HbF production by WB (J-K) in CD34⁺ cells and CRISPR/Cas9-edited S878F mutant HuDEP-2 cells. All CD34⁺ cells were collected from HbE/β⁰-thalassemia patients. Data in the column graphs (E-I) from at least 2 independent experiments are presented as the means ± SEM (****P ≤ .0001; ***P ≤ .001; **P ≤ .01). IP, immunoprecipitation; LRF, leukemia/lymphoma-related factor; mCpG, methylated CpG.