Fig 1.
Targeting the rs58923657 haplotype block for CRISPR/Cas9-mediated SNP microdeletions in human LCLs.
(A) Epigenetic profile of the region harboring the rs58923657 haplotype block. Tracks show the position of the IKZF1 3’ UTR (dark blue rectangle); the positions of rs58923657 (blue) and SNPs targeted for deletion (red); GM12878 chromatin state (ChromHMM); DNase-seq signal from GM12878 with signal strength corresponding to chromatin accessibility; and transcription factor ChIP-seq data from GM12878 and other ENCODE LCLs with darkness corresponding to signal strength. Data are provided by the ENCODE project and displayed in the UCSC genome browser (http://genome.ucsc.edu/). (B) Regions targeted for CRISPR/Cas9 deletion. The wild-type (WT) sequences correspond to the major allele and show the positions of gRNA target sequences (blue underlined), PAM sequences (blue bold), gRNA-directed Cas9 cut sites (red arrows), and SNPs (red). The edited allele sequences show the expected locations and sizes of DNA fragment deletions. (C) PCR analysis demonstrating SNP microdeletions in bulk transfected LCLs. PCR products from non-transfected LCLs were analyzed for comparison (WT). Bulk populations with at least 1 band at the expected size were used for single cell cloning and are marked with an asterisk. Expected PCR product sizes (WT/Δ bp) rs62445866, 782/745; rs6964969, 765/716; rs6944602, 527/487; rs10264390, 772/731, 698/658. Panels are from images of either the same gel, or gels run separately that have been rearranged to reorder lanes.
Fig 2.
Homozygous microdeletions of SNPs in the rs58923657 non-risk haplotype result in significantly decreased IKZF1 expression and increased proliferation in human LCLs.
Expression of IKZF1 mRNA in (A) non-DS and (B) DS LCLs with CRISPR/Cas9 induced microdeletions of regions at rs62445866, rs6964969, rs6944602, rs10264390, and rs17133807. Dot plots show the log2 fold change of IKZF1 mRNA expression in SNP deletion clones normalized to WT clones in LCLs with non-risk allele status for each SNP. Results are mean ± SEM (n = 8 deletion clones per group). The differences between WT and SNP deletion groups were analyzed by one-way ANOVA. *P < 0.01; **P < 0.0001. (C) The effect on IKZF1 mRNA expression was compared for each SNP deletion between non-DS and DS clones. The differences were analyzed by one-way ANOVA. #P < 0.05. (D) Cellular proliferation in SNP deletion versus WT clones for 2 non-DS and 2 DS LCLs. Results are mean cell counts ± SEM (n = 3 clones per SNP deletion and 3 WT clones). Differences between WT and SNP deletion clones were analyzed by two-way ANOVA. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.
Fig 3.
The IKZF1 enhancer overlapping rs17133807 is conserved in mammalian species.
(A) Epigenetic profile showing mammalian sequence conservation across the rs58923657 haplotype block. Tracks show IKZF1 3’ UTR position; GM12878 chromatin states; DNase-seq signal from GM12878; and mammalian sequence conservation with higher conservation scores above the axis reflecting increasing sequence conservation (see Fig 1A for additional details of track annotation). The conserved region encompassing rs17133807 is highlighted yellow. (B) Epigenetic profile of the orthologous mouse locus. Tracks show DNase I HS sites in mouse CD19+ and CD43- B cells and the murine B lymphoblastoid cell line CH12; histone modification ChIP-seq data (H3K4Me1, H3K4Me3, H3K27ac) from mouse bone marrow and CH12; TF ChIP-seq data from CH12; mammalian sequence conservation; and the positions of mouse SNP rs263378223 (blue) and gRNA targets (red). The conserved region is highlighted yellow. All data are displayed in the UCSC genome browser (http://genome.ucsc.edu/).
Fig 4.
Homozygous deletion of the orthologous murine Ikzf1 enhancer results in significantly decreased Ikzf1 expression, increased cell proliferation, and altered B cell differentiation.
(A) Schematic of the orthologous enhancer region on mouse chr11 showing gRNA-directed Cas9 cut sites (blue scissors) and primer binding sites (black arrows). (B) PCR analysis of CRISPR/Cas9mediated deletions at the orthologous enhancer in WT and Dp16 BM HSCs. Samples with at least 1 band at the expected size and no unedited band (*) were sequenced. (C) Alignment of representative Sanger sequencing results from edited Dp16 and WT samples. The positions of gRNA target sequences (blue underlined), PAM sequences (blue bold), and gRNA-directed Cas9 cut sites (red arrows) are shown. (D) Representative chromatograms from samples in (C) are shown with the repaired DNA junctions marked (vertical line). (E) Relative Ikzf1 mRNA expression in c-Kit+Sca-1+CD34- HSCs with homozygous enhancer deletion (Δ) was analyzed by qRT-PCR. Bars show mean ± SEM normalized to non-targeting (NT) gRNA-expressing controls. Results were analyzed by Student’s two-tailed t-test. (F) Pre-B lymphoid colony forming assays were performed using transduced BM HSCs. ZsGreen+RFP+ colonies for NT and Δ samples were counted after incubating for 7 days in pre-B lymphoid-promoting methylcellulose. Bars show mean colony counts ± SEM. (G) B cell differentiation was analyzed in Δ and NT HSCs by Hardy fraction analysis. Bars show mean ± SEM percentages of cells in each Hardy fraction. (H) The percentage of B220+ lymphocytes in Δ and NT HSC co-cultures was measured by flow cytometry. Bars show mean ± SEM percentages of B220+ cells. Results from (E)–(H) are from independently transduced HSC harvests from n = 3 mice. Results in (E) and (H) were analyzed by Student’s t-test. Results in (F) and (G) were analyzed by two-way ANOVA with Tukey's multiple comparisons test. * P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001.