Fig 2: Functional effects of CRISPR/Cas9-mediated ASXL1 mutation correction(A) Evaluation of ASXL1 protein expression by Western blotting. Left-hand side: ASXL1 protein expression in the SET2 leukemia cell line (wild-type for ASXL1), uncorrected KBM5 cells, the K562 leukemia cell line (carrying the Y591X heterozygous ASXL1 mutation), and KBM5 clones (labeled 1–5) with heterozygous precise correction of the ASXL1 mutation. Right-hand side: ASXL1 protein expression in the SET2 leukemia cell line (wild-type for ASXL1), uncorrected KBM5 cells, and KBM5 clones (labeled 1–5) with homozygous precise correction of the ASXL1 mutation. β-actin was used as loading control. (B) Evaluation of the expression levels of HOXA genes using quantitative real-time PCR (q-RT-PCR). The expression levels of HOXA5, 6, 7, 9, 10 and 13 were measured in three ASXL1 homozygous corrected KBM5 clones compared with uncorrected cells. The results shown were obtained from six independent experiments for each clone. The values in ASXL1 homozygous corrected cells are relative to the uncorrected cells. Bar graphs show mean + standard error of the mean (s.e.m.) (* = P < 0.05, ** = P < 0.01, *** = P < 0.001, paired t-test). (C) Evaluation of H3K27me3 levels and expression of PRC2 components by Western blotting in uncorrected KBM5 cells and three KBM5 clones with homozygous correction of the ASXL1 mutation. H3K27me3 levels and total H3 levels were evaluated using purified histone fractions. The expression levels of two PRC2 components (EZH2, SUZ12) were determined using whole cell lysates. β-actin was used as loading control. (D) Immunoprecipitation of BAP1 in the SET2 leukemia cell line (wild-type for ASXL1), uncorrected KBM5 cells, and three KBM5 clones with homozygous correction of the ASXL1 mutation. The BAP1 protein fraction was immunoprecipitated using a BAP1 antibody and stained for ASXL1 and BAP1.

Fig 3: CRISPR/Cas9-mediated correction of ASXL1 mutations in the CML cell line KBM5(A) Structure of the ASXL1 gene. ASXL1 maps to the chromosome region 20q11 and comprises 12 exons. The ASXL1 G710X mutation found in KBM5 cells is located in exon 12. (B) Design of sgRNAs and ssODN repair template used for the CRISPR/Cas9-mediated ASXL1 mutation correction. Left-hand side: sequences of three sgRNAs (sgRNA#1, sgRNA#5, sgRNA#25), identified using the crispr.mit.edu online resource. Right-hand side: alignment of the three sgRNAs to the genomic region containing the ASXL1 mutation (indicated in red) in KBM5 cells. Each site comprises 20 nt followed by a trinucleotide (5′-NGG-3′) protospacer adjacent motif (PAM), highlighted in bold, which is required for Cas9 activity (DNA double-strand break). Bottom: sequence of the ssODN used as repair template in the HDR. The G nucleotide, which corrects the mutated T nucleotide in KBM5 cell line, is highlighted in red; five silent nucleotides changes (i.e. not causing amino acid changes in the resulting ASXL1 protein) were introduced in the ssODN sequence (highlighted in green) to avoid undesired Cas9 activity in mutation-corrected cells. (C) Evaluation of ASXL1 mutation correction in KBM5 cells using Sanger sequencing. Top trace: sequencing trace showing the presence of homozygous ASXL1 point mutation (GGA > TGA, p.G710X); the mutated nucleotide (G > T) is highlighted in red. Middle trace: representative sequencing trace showing heterozygous correction of the ASXL1 mutation; KBM5 clones with heterozygous correction retain the mutant allele (G / T), highlighted in red in the sequencing trace. Bottom trace: representative sequencing trace showing homozygous correction (G) of the ASXL1 mutation. The red arrows in the middle and bottom panel indicate the silent nucleotides changes that were introduced in the ssODN sequence.

Fig 4: Effects of CRISPR/Cas9-mediated ASXL1 mutation correction on cell growth and myeloid differentiation(A) Cell growth analysis. Viable cells were identified by trypan blue exclusion at different time points in culture. Uncorrected KBM5 cells and three KBM5 clones with homozygous correction of the ASXL1 mutation were cultured at an initial density of 1 × 104/mL and viable cells were counted over a 7-day period. The results shown were obtained from three independent experiments for each clone and represent mean ± s.e.m. (*** = P < 0.001, two-way ANOVA). (B) Flow cytometry contour plots showing the expression of the two myeloid surface markers CD33 and CD15 in KBM5 clones with homozygous correction of the ASXL1 mutation (labeled 1–5) compared to uncorrected KBM5 cells. Cells stained with anti-CD33 and anti-CD15 antibodies are shown in red. A total of nine KBM5 clones with homozygous correction of the ASXL1 mutation were analyzed (all showing a homogenous CD33+CD15hi cell population) and data from five representative clones are shown. (C) Percentage of CD15+ cells in KBM5 clones with homozygous correction of the ASXL1 mutation (labeled 1–5) compared to uncorrected KBM5 cells, assessed by flow cytometry. The results shown were obtained from six independent experiments for each clone. Bar graphs show mean ± s.e.m (* = P < 0.05, Wilcoxon signed-rank test). (D) Histogram plots showing expression of CD15+ cells in representative KBM5 clones with homozygous correction of the ASXL1 mutation (labeled 1–5) compared to uncorrected KBM5 cells, assessed by flow cytometry. (E) Representative images of May-Grünwald/Giemsa stained uncorrected KBM5 cells and KBM5 clones with homozygous correction of the ASXL1 mutation. The red arrows indicate cells with a decreased nucleus to cytoplasm ratio and condensed nuclei, which are signs of myeloid maturation. Scale bar indicates 25 μm.