Fig 1.
Cas9-RNP and ssODN-mediated targeting of the JAK2 locus in K562s and CD34+ HSPCs.
(a) Schematic of the JAK2 V617 locus with relevant genetic editing reagents annotated. SgRNAs-4 and -6 are identified with their corresponding 5’-NGG-3’ PAMs. Red arrowheads indicate cut site directed by these sgRNAs. The conversion of G>T results in the amino acid change from valine to phenylalanine (V617F, magenta) and removes the naturally occurring BsaXI restriction site. The conversion of C>T results in a silent mutation (V617V, blue) and retains the BsaXI restriction site. (b) T7 endonuclease I assay showing indel formation in cell pools edited independently with 9 different sgRNAs. (c) K562 cells were nucleofected with Cas9-RNP and the 617F ssODN. NHEJ- and HDR-mediated editing outcomes were assessed by amplicon-NGS. Data from n = 3 independent biological replicates with mean±SD graphed. (d) CD34+ HSPCs were nucleofected with Cas9-RNP and 617F or 617V ssODN. NHEJ- and HDR-mediated outcomes were assessed by amplicon-NGS. Data from n≥3 independent biological replicates with mean±SD graphed. (e) Allele spectra and corresponding percentages of alleles generated following editing with sgRNA-4 RNP and 617F or 617V ssODN in CD34+ HSPCs. Mutations causing frame-shifts are outlined.
Fig 2.
Characterizing the V617F mutation in HUDEP-2s.
(a) Experimental layout for generating clonal allelic series of JAK2 V617F and F617V alleles. Nucleofected HUDEP-2 cells are recovered for 3 days and then sorted at single cell densities into a 96 well plate based on the expression of Kusabira Orange, a marker indicative of HUDEP-2 cell viability. The isolated single cells are expanded for two weeks and subsequently screened for either V617F or F617V alleles by BsaXI restriction digest. Selected clones are validated by TA-cloning and Sanger sequencing. (b) HUDEP-2 WT and V617F cells were nucleofected with Cas9-RNP complexed with F617-sgRNA and the 617V ssODN. NHEJ- and HDR-mediated editing outcomes were assessed by amplicon sequencing and analyzed by Interference of CRISPR Editing (ICE) software. Data from n = 4 independent biological replicates with mean±SD graphed. (c) Representative immunoblot and signal intensity quantification show elevated phosphorylated STAT5 (P-STAT5) expression in homozygous and heterozygous JAK2 V617F HUDEP-2 clones. Signals were normalized to STAT5 and GAPDH. White bars, JAK2 WT clones; light pink bars, JAK2 V617F heterozygous clones; magenta bars, JAK2 V617F homozygous clones. (d) Representative immunoblot and signal intensity quantification show reduced P-STAT5 levels following reversion of the JAK2 V617F mutation back to WT. Signal intensity calculations were normalized to STAT5 and GAPDH. White bars, JAK2 WT clones; light pink bar, JAK2 V617F heterozygous clone; magenta bar, JAK2 V617F homozygous clones; light blue bar, HUDEP clone with single allele corrected (genotype V617F/F617V); dark blue bars, HUDEP clones with both alleles corrected (genotype F617V/F617V) (e) Growth curve depicting cumulative population doublings of HUDEP-2 clones measured for 5 days. F1 and F3 are V617F homozygote clones. Data from n = 5 independent biological replicates. Mean of all experiments ± SD shown. (f) Competitive/co-culture growth assay using HUDEP-2 WT and V617F clones, showing significant outgrowth of V617F clones. Equal proportions of WT and V617F clones were co-cultured in HUDEP-2 expansion media and allelic frequencies were analyzed at 6 time-point spanning 9 days by amplicon-NGS. Circles denote co-culture of F1 and WT1. Diamonds denote co-culture of F3 and WT2. Data is from n = 2 biologically independent replicates. Error bars indicate SD.*: p<0.05 by paired t-test. (g) Co-culture of WT and homozygous V617F clones grown in the presence or absence of EPO for 6 days. Amplicon-NGS time points taken at Day 0 and Day 6. Ratio of V617F:WT alleles detected in culture at day 6 were normalized to day 0 and graphed. Data is from 4 biologically independent experiments each sequenced in technical triplicate. (h) Co-culture of either WT and homozygous F617V corrected clone (circle) or homozygous V617F and corrected F617V clones (square). NGS time points taken at Day 0, and Day 6 with and without EPO. Ratio of WT:F617V or V617F:F617V alleles detected in culture at day 6 were normalized to day 0 and graphed. Data is from 2 biologically independent experiments. Mean±SD graphed.
Fig 3.
Identifying the enrichment of the V617F-edited alleles in CD34+ hematopoietic stem cells.
(a) Experimental layout for editing LT-HSCs. HSPCs were thawed and expanded for 3 days in SR1, UM171 and CC110. HSCs were sorted based on CD34+ CD38- CD45RA- CD90+ and nucleofected with appropriate editing reagents. Two days following electroporation an ‘input’ sample was taken to determine editing percentages. Cells were either maintained in liquid culture for immunophenotyping or plated in methylcellulose for further analyses. b. HSCs were nucleofected with Cas9-RNP and ssODNs encoding HBB, V617F or V617V mutations and were plated on methylcellulose as depicted in (a). HDR-mediated outcomes were assessed by NGS, two days (input) or two weeks after electroporation. Data from n = 3 independent biological replicates. Mean±SD shown. (c) NHEJ-mediated outcomes of cells in (b) were assessed by NGS, two days (input) or two weeks after electroporation. Data from n = 3 biological replicates. Mean±SD shown. (d) Pools of V167F- or V617V-edited HSCs were plated on methylcellulose with or without EPO and were scored as granulocyte, monocyte or erythroid based on morphology 14 days after plating. Data from n = 4 (for V617F) or n = 3 (for V617V) independent biological replicates. Mean±SD shown. (e) V617F- and V617V-edited HSCs were mixed and plated on methylcellulose. Input sample was taken directly after mixing prior to plating on methylcellulose. All resulting colonies were harvested two weeks later and processed for amplicon-NGS. Data from n = 4 biological replicates. Mean±SD shown. *:p<0.05 by unpaired t-test. (f) Single colonies were phenotyped as erythroid, GEMM, or GM and then genotyped by amplicon-NGS. V617F homozygous colonies were preferentially selected in the absence of EPO. Data from n≥3 biological replicates. Mean±SD shown. *:p<0.05 by unpaired t-test.
Fig 4.
Characterizing the effect of the V617F mutation in CD34+ hematopoietic stem cells.
(a, b) Immunophenotypic analysis of methylcellulose colonies cultured with or without EPO. All colonies were scraped, pooled and stained the following surface markers. Lymphoid: CD19 or CD56; Myeloid: CD33, or CD14 or CD41a; Erythroid: CD71 or GlyA. Data from n = 1 for all except GlyA samples. GlyA data from n = 3 independent biological replicates. Mean±SD shown. *:p<0.05 by paired t-test. (c) Immunoblot showing the elevated expression of B-globin in V617F edited cells grown in methylcellulose without EPO. No difference was observed in P-STAT5 expression across all samples. GAPDH served as a loading control.