Scarless genome editing in human pluripotent stem cells (hPSCs) represents a goal for both precise research applications and clinical translation of hPSC-derived therapies. Here we established a versatile and efficient method that combines CRISPR–Cas9-mediated homologous recombination with positive–negative selection of edited clones to generate scarless genetic changes in hPSCs.
This is a preview of subscription content
Subscribe to Journal
Get full journal access for 1 year
only $9.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
The datasets generated and analyzed in this study are available from the corresponding author upon reasonable request.
Porteus, M. Annu. Rev. Pharmacol. Toxicol. 56, 163–190 (2016).
Yadegari, H. et al. Blood 128, 2144–2152 (2016).
Duan, J. et al. Hum. Mol. Genet. 12, 205–216 (2003).
Kimchi-Sarfaty, C. et al. Science 315, 525–528 (2007).
Paquet, D. et al. Nature 533, 125–129 (2016).
Miyaoka, Y. et al. Nat. Methods 11, 291–293 (2014).
Steyer, B. et al. Stem Cell Rep. 10, 642–654 (2018).
Ye, L. et al. Proc. Natl. Acad. Sci. USA 111, 9591–9596 (2014).
Kim, S. I. et al. Nat. Commun. 9, 939 (2018).
Yagi, H. et al. Lancet 362, 1366–1373 (2003).
Nishimura, T. et al. Cell Stem Cell 12, 114–126 (2013).
Takayama, N. et al. J. Exp. Med. 207, 2817–2830 (2010).
Cradick, T. J., Qiu, P., Lee, C. M., Fine, E. J. & Bao, G. Mol. Ther. Nucleic Acids 3, e214 (2014).
This work was supported by the Amon Carter Foundation (M.H.P.), the Laurie Krauss Lacob Faculty Scholar Fund (Award in Pediatric Translational Research to M.H.P.), the California Institute for Regenerative Medicine (DISC2-08874 and LA1_C12-06917 to H.N.), the Japan Society (JSPS Postdoctoral Fellowship for Overseas Researchers to T.N.), and the Lucile Packard Foundation for Children’s Health (22q11 deletion syndrome gift). We thank K.M. Cromer for editing the manuscript.
M.H.P. is a consultant for and has equity interest in CRISPR Tx. H.N. is a member of the scientific advisory board of ReproCELL. N.U. is a current employee of ReGen Med Division, BOCO Silicon Valley. K.I. is a current employee of Daiichi-Sankyo Co., Ltd. No company, however, had input into the design, execution, interpretation, or publication of the results herein.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Integrated supplementary information
Target sites of gRNAs.
(a) Flow cytometric (FCM) analysis of transient (at day 2) and stable (at day 6) expression. (b) Before and after positive selection by magnetic-activated cell sorting (MACS) analyzed by FCM based on tCD19 expression. CD19-positive cells were enriched from 2.1% to 67.8%. (c) Fluorescent microscopy of single cell colonies after MACS positive selection. There were bright colonies (yellow arrowhead) and dim colonies (white arrowhead). Scale bar, 200 μm. (d) FCM analysis of mCherry in WT, dim, and bright clones. (e) Mean fluorescent intensity of mCherry analyzed by FCM. individual blue data points represent different clones (n = 3), and orange bars represent the mean value. (f) Schematic representation of primer design at TBX1 editing locus and PCR-based genotype analysis of single cell clones after MACS positive selection of first-round editing. FCM, fluorescent microscopy. PCR experiments were repeated three times, with similar results each time.
Scale bars, 400 μm. This experiment was repeated three times using different clones, with similar results each time.
Supplementary Figure 4 Scarless reporter gene integration at the 5′ side of the RUNX1 stop codon in a human iPSC line.
(a) Schematic representation of primer design (gray arrows) for genotyping in RUNX1 editing. The size of the PCR amplicon from each genotype is shown in the bottom table. (b) PCR-based genotyping of bright clones after first-round editing followed by MACS positive selection and single cell cloning. Clone 2 was used for the following procedure. (c) Genotyping by PCR of a clone after second-round editing followed by MACS negative selection and single cell cloning. (d) Representative image of RUNX1-mOrange iPSC sacs. Cells were observed 13 d after induction of differentiation. Scale bars, 100 μm. Fluorescent microscopy and PCR experiments were repeated three times, with similar results each time.
Supplementary Figure 5 Scarless marker integration just before the stop codon of GFI1 in a RUNX1-mOrange line.
(a) Schematic design of the donor DNAs and the guide RNAs for RUNX1 editing. Arrows indicate primers for genotyping. The size of the PCR amplicon from each genotype is shown in the table on the right. (b) Genotyping of bright clones after first-round editing followed by MACS positive selection and single cell cloning by PCR. The copy number of the UBC promoter was evaluated by ddPCR. Clone 2 was used for the following procedure. (c) Genotyping by PCR of a clone after second-round editing followed by MACS negative selection and single cell cloning. Because the second donor vector contained the same sequence (EGFP) in addition to the homology arm, undesired editing events led to contamination (1–3, 5) as depicted in (d). The PCR experiments were repeated twice, with the same results each time.
Supplementary Figure 6 Detection of random integration in RUNX1 and GFI1 editing after first-round editing.
(a) Schematic representation of primer design for detection of random integration. Primer sets 1 and 2 can detect random integration if a selection marker is integrated into the genome with the sequence at the outside of the left and right homology arms, respectively (derived from plasmid backbone). PCR results are shown in (b) for RUNX1 editing and in (c) for GFI1 editing. PCR experiments were repeated twice, with the same result each time.
Supplementary Figure 7 Improvement of first-round editing by shRNA-based removal of random integrant and PCR-based copy-number analysis using modified UBC promoter in B2M editing.
(a, b) Schematic representation of donor DNA design and genotyping by PCR in B2M editing of the H9 hESC line. (a) First-round editing without shGFP expression cassette. (b) First-round editing with shGFP expression cassette. All clones shown in a and b expressed bright GFP fluorescence. (c) GG>AA mutation at 314th and 315th upstream from ATG start codon. In B2M editing, modified UBC promoter (GG>AA) was used for selection marker expression. (d) Flow cytometry analysis showing WT and GG>AA mutant of UBC promoter expressed at almost the same intensity of GFP in K562 cells. (e) Schematic representation of primer design and size of PCR amplicon after EcoRI digestion. (f) Agarose gel electrophoresis of PCR products digested with EcoRI-HF enzyme. DNA was stained with MidoriGreen dye. (g) Representative image from image analysis by ImageJ software. (h) Quantification of fluorescence intensity (FI) in samples from regular PCR combined with EcoRI digestion. All samples from bright clones show the same intensity between WT and GG>AA, meaning two copies of exogenous UBC promoter were integrated, and FIs of GG>AA were half of those of the WT in samples from dim clones, meaning one copy of exogenous UBC promoter was integrated. (i) Copy-number analysis of UBC promoter via a ddPCR-based method. The result from ddPCR was consistent with that of the regular PCR-EcoRI digestion–based experiment. In panels h and i, individual blue points represent technical replicates (n = 3) and orange bars represent the mean value. FCM and PCR shown here were repeated twice, with similar results each time.
Supplementary Figure 8 Second step of scarless HLA-A*24 cDNA (1.1-kb) integration just before the stop codon of B2M in the H9 hESC line.
(a) Editing design of second-round editing using clone 7 from Supplementary Fig. 7 and PCR-based genotyping of obtained GFP-negative clones after MACS negative enrichment and single-cell cloning. (b) Representative FCM profile of the cells. Class I HLA type of the H9 hESC line is (A*02, A*03, B*35, B*44, C*04 and C*07). iAM9 lines are hiPSCs with HLA-A*24, used as a positive control of natural HLA-A24 expression. Cells were treated with 50 ng/ml INF-γ for 3 d prior to FCM analysis. FCM and PCR experiments were repeated twice, with similar results each time.
Copy-number variation (CNV) across the genome analyzed with cnvPartition plugin v. 3.2.0 for GenomeStudio (Illumina) in TBX1 edited lines. (a) Information of the edited lines. (b) Detected CNV. Arrowhead indicates chromosomal change caused by the editing process. (c) B allele frequency and log R ratio at chromosome with CNV. 2 Copy CNV is highlighted in green. Copy-neutral loss of heterozygosity (LOH) was seen from the targeted site to chromosome end in line F.
CNV across the genome analyzed with cnvPartition plugin v. 3.2.0 for GenomeStudio (Illumina) in RUNX1 and GFI1 edited lines. (a) Information of the edited lines. (b) Detected CNV. Arrowhead indicates chromosomal change caused by the editing process. (c) B allele frequency and log R ratio at chromosome with CNV. 2 Copy CNV is highlighted in green. Duplication of the X chromosome was seen in all lines, indicating that this chromosomal abnormality was derived from the original line, and not the editing process. Copy-neutral LOH was seen from targeted site to chromosome end in lines H–J.
CNV across the genome analyzed with cnvPartition plugin v. 3.2.0 for GenomeStudio (Illumina) in B2M edited lines. (a) Information of the edited lines. (b) Detected CNV. Arrowhead indicates chromosomal change caused by the editing process. (c) B allele frequency and log R ratio at chromosome with CNV. 0 Copy CNV is highlighted in red. Small loss of SNP copy was detected at the editing site in lines K and L, indicating that this CNV was detected owing to overlap of the SNP probe with the editing site.
The editable range in scarless editing in terms of limitation of cut-to-edit distance (Paquet et al.5) and Cas9 tolerance of mismatch (Fu, Y. et al., High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat. Biotechnol. 31, 822–826, 2013).
Supplementary Figure 13 Potential donor DNA design for various patterns of scarless editing based on two-step HR.
We demonstrated in this paper that HR-based editing is available to integrate and delete more than 3 kb in TBX1 editing and modify about 3 kb to 1 kb sequence in B2M editing in hPSCs if it is combined with appropriate marker selection. Combining them will theoretically enable additional application of scarless edit as follows: (a) Large insertion for, for example, introduction of a reporter gene, which should be useful for optimization of differentiation in pluripotent stem cells or drug screening. (b) Large deletion for making disease models or knockout animals. (c) Large substitution for making knock-in animals including humanized models. (d) Editing in case there is no good target site of site-specific nucleases (SSNs) including zinc finger nuclease, near intended mutation site. With this technique, almost all limitations to the editing site should be eliminated.
All other FCM analysis was performed after gating of living cells according to FSC-A and SSC-A.
GeneRuler 1 kb Plus DNA Ladder (Thermo Fisher, SM1331) was used as a DNA size marker (loaded in each first left lane) for all gels. This size marker indicates 20,000, 10,000, 7,000, 5,000, 4,000, 3,000, 2,000, 1,500, 1,000, 700, 500, 400, 300, 200, and 75 bp from top to bottom.
About this article
Cite this article
Ikeda, K., Uchida, N., Nishimura, T. et al. Efficient scarless genome editing in human pluripotent stem cells. Nat Methods 15, 1045–1047 (2018). https://doi.org/10.1038/s41592-018-0212-y
Nature Protocols (2021)
Nature Genetics (2021)
Generation of human striatal organoids and cortico-striatal assembloids from human pluripotent stem cells
Nature Biotechnology (2020)