Microhomology-assisted scarless genome editing in human iPSCs

Gene-edited induced pluripotent stem cells (iPSCs) provide relevant isogenic human disease models in patient-specific or healthy genetic backgrounds. Towards this end, gene targeting using antibiotic selection along with engineered point mutations remains a reliable method to enrich edited cells. Nevertheless, integrated selection markers obstruct scarless transgene-free gene editing. Here, we present a method for scarless selection marker excision using engineered microhomology-mediated end joining (MMEJ). By overlapping the homology arms of standard donor vectors, short tandem microhomologies are generated flanking the selection marker. Unique CRISPR-Cas9 protospacer sequences nested between the selection marker and engineered microhomologies are cleaved after gene targeting, engaging MMEJ and scarless excision. Moreover, when point mutations are positioned unilaterally within engineered microhomologies, both mutant and normal isogenic clones are derived simultaneously. The utility and fidelity of our method is demonstrated in human iPSCs by editing the X-linked HPRT1 locus and biallelic modification of the autosomal APRT locus, eliciting disease-relevant metabolic phenotypes.

Sequence of HPRT1 alleles from 409B2 (female) iPSC clones transfected with HPRT1_B NC-TALENs and enriched by 6-TG selection on SNL feeders. PCR amplicons of the target site were TA-cloned and at least 8 bacterial colonies from each transformation were PCR-amplified to determine individual alleles by Sanger sequencing. Clones are labeled numerically and alleles alphabetically. iPSC clones with more than two alleles likely represent mosaic populations. Upper case letters represent TALEN binding sites (Fig. 1a). Inserted bases are in italics. Deletion or insertion sizes are indicated on the right. REF, parental 409B2 iPSC reference genomic sequence; NORM, non-mutant allele for the region examined by sequencing. a. SSA assay comparing the relative activities of HPRT1_B TALENs assembled using a PthXo1-based TALE scaffold (NC-TALEN) to an AvrBs3-based +136/+63 scaffold (Avr-TALEN). Error bars show s.e.m. (n = 3). b. TALEN activity in 1383D6 human male iPSCs as measured by 6-TG R colony formation, indicating HPRT1 disruption. Spontaneous colony formation in the absence of nuclease was not noted. For the assay, 3 µg of each nuclease plasmid was transfected into 1 x 10 6 cells by electroporation, followed by plating at a density of 4.5 x 10 5 cells per 60 mm dish. iPSCs were selected and stained as described in the Methods. Figure 4. TIDE analysis of indel formation at the HPRT1_B TALEN target site. a. Schematic of the genomic PCR assay used to analyze the locus targeted by HPRT1_B TALENs. For TIDE analysis, the breakpoint was arbitrarily positioned at the beginning of the spacer as indicated (black arrow).

Supplementary
b. Sequence trace files of the original 1383D6 iPSCs, and 6-TG R population following transfection with TALENs. The position of the breakpoint used for TIDE analysis is shown (black arrow). An ambiguous A/T base is noted upstream of the predicted breakpoint (red arrow). c. Aberrant sequence plot determined by the online TIDE software. Arrows are as in Panel b. d. Spectrum of indels in the 6-TG R iPSC population as predicted by TIDE.
Deletions are more common than insertions, with a clear bias towards 17 bp deletions. The data in Panel c and d was reproduced across independent experiments (n = 3). e. Sequence trace files of the original H1 ESCs, and 6-TG R population following transfection with TALENs. The position of the breakpoint used for TIDE analysis is shown (black arrow). An ambiguous A/T base is noted upstream of the predicted breakpoint (red arrow). f. Aberrant sequence plot determined by the online TIDE software. Arrows are as in Panel e. g. Spectrum of indels in the 6-TG R ESC population as predicted by TIDE. As with 1383D6 iPSCs, deletions are more common than insertions, with a clear bias towards 17 bp deletions (n = 1). aaatagtgatagatCCATTCCTATGACTG

Supplementary Figure 5. Spectrum of Avr-TALEN-induced mutations in human male iPSCs clones.
Sequence of HPRT1 alleles types detected in a series of individual clones derived from 1383D6 iPSC clones transfected with HPRT1_B Avr-TALENs and enriched by 6-TG selection under feeder-free conditions. PCR amplicons of the target site were directly Sanger sequenced. Mixed sequences were not included in the analysis. Clones are labeled numerically. Upper case letters represent HPRT1_B Avr-TALEN binding sites. Inserted bases are in italics. Modified bases are underlined. Deletion or insertion sizes are indicated on the right. Apart from Δ17, the most common deletion was Δ46 (3/31 deletions), where the deletion boundaries were positioned within T-rich sequences following a predicted 'GATT' microhomology. The Δ77 mutation occurred at another short tandem repeat 'CTGA', again indicative of MMEJ. REF, parental 1383D6 iPSC reference genomic sequence.

Supplementary Figure 6. Drug sensitivities of 1383D6 parental and HPRT1 knockout iPSC clones.
Crystal violet staining of representative HPRT1 knockout clonal iPSC lines following treatment with 6-TG or HAT media for 3 days. Resistance and sensitivity correlates with the status of the HPRT1 locus, as determined by PCR genotyping and sequencing ( Supplementary Fig. 5). Parental 1383D6 iPSCs are included as a control.  Table 1 Fig. 2a. b. Detailed schematic of HPRT1 gene targeting and MMEJ resolution. Labelling is consistent with Fig. 2b. Southern blot verification of targeted clones using the mCherry probe (bottom right), where an asterisk (*) denotes clones used for subsequent assays (Fig. 3 and Supplementary Fig. 12) while "x" indicates clones with random integration. a. Schematic overview of gene targeting to generate clones for the HPRT1 chromosomal excision assay. Left and right donor vector homology arms overlap, generating a 29 bp tandem µH (blue) flanking the positive/negative selection marker (red). Synonymous mutations disrupting the endogenous µ5A3 sequence are shown in red. Gene targeting was stimulated with AvrHPRT1_B TALENs (yellow bolt). The remaining elements are as described in Fig. 2a. b. Detailed schematic of HPRT1 gene targeting and MMEJ resolution. Labelling is consistent with Fig. 2b. Southern blot verification of targeted clones using the mCherry and HPRT-B probes (bottom right), where an asterisk (*) denotes clones used for subsequent assays (Fig. 3, Table 2 and Supplementary Fig. 12) while "x" indicates clones with random integration.

Supplementary Figure 12. Effect of protospacer inversion on MMEJ repair.
a. FACS for mCh neg cells following transfection of targeted iPSC clones (differing in µH length) with pX-ps1 to stimulate cassette excision. µ29 excision data is representative of three independent clones. b. FACS analysis for mCh neg cells following transfection of targeted iPSC clones (inverted protospacers) with pX-ps1. Parental 1383D6 iPSCs are included as a control. Clones for this assay were generated using gene targeting as outlined in Supplementary Fig. 11, except with inverted ps1 protospacers in the case of ps1-rev. c. Sanger sequencing of excised populations shown in Panel b with and without HAT selection. With HAT selection, the predominance of indel-free sequences bearing engineered synonymous mutations indicates that the population is biased towards MMEJ repair, irrespective of the ps1 protospacer orientation. µH regions (blue) and synonymous mutations (red) are indicated.

Supplementary Figure 13. Validation of APRT sgRNAs.
a. Schematic of the human APRT locus and strategy for engineering the APRT*J mutation. Detail is shown for exon 5 (orange) including the splice junction, CRISPR-Cas9 target sites 1 through 4 (green), and selected µ32 microhomology (blue). APRT codons are numbered above. Chromosome positions refer to H. sapiens GRCh38. Bases targeted for MhAX editing are shown in blue (silent) or red (APRT*J). SA, splice acceptor. b. T7EI assay results revealing the activity of sgRNAs 1 through 4 in HEK293T cells. n.c., negative control without nuclease transfection. c. Puro R iPSC colony numbers resulting from APRT gene targeting stimulated with sgRNAs 1 through 4. One million 1383D6 iPSCs were electroporated with 3 µg of APRT-2A-puroΔTK donor vector only (n.c.), or the donor plus 1 µg of the appropriate sgRNA expression vector and plated on two 60 mm dishes (5 x 10 5 cells each). Colony numbers are the total from two dishes.

Supplementary Figure 14. Flow cytometry analysis of APRT gene targeting and excision.
FACS for mCh neg cells following transfection with pX-ps1 to stimulate cassette excision. As expected, excision rates are lower for homozygously targeted clones. a. Schematic of the parental and edited APRT alleles, and the resulting RFLP generated by the Silent mutation. b. Gel electrophoresis following Acc65I digestion of PCR amplicons from excised hetero-or homozygously targeted iPSC clones, indicating the presence of the engineered Silent mutation. 1383D6 iPSCs are included as a negative control for cleavage.

Supplementary Figure 18. FACS-based isolation of edited HPRT Munich iPSCs.
Representative FACS plots for the isolation of iPSCs edited at the HPRT1 locus. The donor vector, allele, and additional features are as described in Fig. 2a and b.

Supplementary Figure 19. Uncropped Southern blot images.
a. Complete images for Southern blot genotyping data shown in Fig. 2d. b. Complete images for Southern blot genotyping data shown in Fig. 4c and f.