Efficient and accurate homologous recombination in hESCs and hiPSCs using helper-dependent adenoviral vectors

Abstract

Low efficiencies of gene targeting via homologous recombination (HR) have limited basic research and applications using human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs). Here, we show highly and equally efficient gene knockout and knock-in at both transcriptionally active (HPRT1, KU80, LIG1, LIG3) and inactive (HB9) loci in these cells using high-capacity helper-dependent adenoviral vectors (HDAdVs). Without the necessity of introducing artificial DNA double-strand breaks, 7-81% of drug-resistant colonies were gene-targeted by accurate HR, which were not accompanied with additional ectopic integrations. Even at the motor neuron-specific HB9 locus, the enhanced green fluorescent protein (EGFP) gene was accurately knocked in in 23-57% of drug-resistant colonies. In these clones, induced differentiation into the HB9-positive motor neuron correlated with EGFP expression. Furthermore, HDAdV infection had no detectable adverse effects on the undifferentiated state and pluripotency of hESCs and hiPSCs. These results suggest that HDAdV is one of the best methods for efficient and accurate gene targeting in hESCs and hiPSCs and might be especially useful for therapeutic applications.

Figure 1

Gene targeting at HPRT1. (a) Schematic illustration of HPRT1 knockout with HDAdV. The probes for Southern analyses are shown as black bars. a, 5′ probe; b, neo probe; c, 3′ probe. HSVtk, the herpes simplex virus thymidine kinase gene cassette; Neo, the neomycin-resistant gene cassette; Venus, an expression cassette for the Venus (F46L mutant yellow fluorescent protein gene ref. 41); A, AhdI sites; Sb, SbfI sites. (b) Southern analyses of wild type (WT) and the HPRT1-knockout human induced pluripotent stem clones. Genomic DNA was digested with AhdI and SbfI. HDAdV, helper-dependent adenoviral vector.

Figure 2

Gene targeting at KU80. (a) Schematic illustration of KU80 heterozygous knockout with HDAdV. The probes for Southern analyses are shown as black bars; a, neo probe; b, 5′ probe; c, 3′ probe; d, deleted-region probe; e, internal probe. HSVtk, the herpes simplex virus thymidine kinase gene cassette; Blue box, the neomycin-resistant gene cassette; Venus, the Venus gene cassette; Triangle, loxP site; B, BglII sites; K, KpnI sites. (b) Southern analyses of WT (246H1) and heterozygous knockout clones (#61 and #111). #201 is an inaccurate recombinant. Genomic DNA was digested with KpnI. An arrow indicates a nonspecific band. (c) Southern hybridization analyses of the KU80 heterozygous knockout clones obtained from 246H1, and the clones from which the PGKneo cassette was removed by transient Cre expression. The genomic DNA was digested with Bg II and hybridized to the ³²P-labeled internal probe or neo probe. WT, wild-type parental cells. (d) Quantitative Southern hybridization. Genomic DNA was digested with BglII and hybridized with probe “d.” A DNA fragment from intron 3 of HPRT1 locus was used as a control probe. The copy number of wild-type KU80 allele determined by a densitometric analysis is indicated. (e) Reduction of KU80 gene expression in the heterozygous mutants. The upper panel indicates the mRNA level determined by quantitative RT-PCR. The average of three independent experiments is shown with the standard deviation. The lower panel indicates the protein level determined by Western blot. An anti-α-tubulin antibody was used as a loading control. HDAdV, helper-dependent adenoviral vector. RT-PCR, reverse transcription-PCR.

Figure 3

Generation of LIG1 and LIG3 heterozygous mutants in human induced pluripotent stem cells. (a) Schematic illustration of the LIG1 heterozygous knockout with HDAdV. The probes for Southern hybridization analyses are shown as black bars; a, neo probe; b, 5′ probe; c, 3′ probe; d, deleted-region probe; e, internal probe. HSVtk, the herpes simplex virus thymidine kinase gene cassette; Blue box, the neomycin-resistant gene cassette; Venus, the Venus gene cassette; Triangle, loxP site; B, BglII; P, PacI. (b) Schematic illustration of the LIG3 heterozygous knockout in human induced pluripotent stem cells. The probes for Southern hybridization analyses are shown as black bars; a, neo probe; b, 5′ probe; c, 3′ probe; d, deleted-region probe. HSVtk, the herpes simplex virus thymidine kinase gene cassette; Neo, the neomycin-resistant gene cassette; Venus, the Venus gene cassette; Red triangle, loxP site; B, BglII; Sw, SwaI. (c) Southern hybridization analyses of WT, LIG1, and LIG3 heterozygous knockout clones (LIG1: #4, #16, #45, and #69; LIG3: #3, #72, and #180). Genomic DNA was digested with PacI or SwaI. Asterisks indicate inaccurate homologous recombinants. (d) Southern hybridization analyses of the LIG1 and LIG3 heterozygous knockout clones obtained from 246H1 cells and clones in which the PGKEM7neo cassette was removed by transient expression of Cre recombinase to confirm the insertion and removal of the PGKneo cassette. The genomic DNA was digested with BglII and hybridized with the ³²P-labeled internal probe or neo probe. #4-12 is an inaccurate recombinant. WT, wild-type parental cells. (e) Quantitative Southern hybridization. Genomic DNA was digested with BglII and hybridized with probe “d.” A DNA fragment from intron 3 of the HPRT1 locus was used as a control probe to normalize the amount of loaded DNA. The copy number of wild-type LIG1 or LIG3 allele determined by a densitometric analysis is indicated. (f) Reduction of LIG1 (left panel) or LIG3 (right panel) gene expression at mRNA and protein levels in the heterozygous LIG1 or LIG3 mutants. The upper panel indicates the mRNA level determined by quantitative RT-PCR. The average of three independent experiments is shown with the standard deviation. The lower panel indicates the protein level determined by Western blotting. An anti-α-tubulin antibody was used as a loading control. HDAdV, helper-dependent adenoviral vector. RT-PCR, reverse transcription-PCR.

Figure 4

Generation of HB9-EGFP knock-in reporter cell lines from hESCs and hiPSCs. (a) Schematic illustration of HB9-EGFP knock-in with HDAdV. The probes for Southern analyses are shown as black bars. a, 5′ probe; b, EGFP probe; c, 3′ probe. Green box, the EGFP gene. Blue box, the neomycin-resistant gene. Red triangle, loxP site. White bar, stuffer DNA for adjusting the vector size. CMV-βgal, the βgal expression cassette; M, MfeI sites. (b) Southern analyses of wild-type parental and the HB9-EGFP knock-in clones (hESC: #47, #G1, and #G10; hiPSC: #24 and #72). Genomic DNA was digested with MfeI. An arrow indicates a signal from the “endogenous” GFP gene, which was encoded by a retroviral vector used for induction of iPSCs. (c) Immunofluorescence analysis of motor neurons differentiated from the HB9-EGFP knock-in hiPSC clone (#24). Bar, 100 µm. EGFP, enhanced green fluorescent protein; HDAdV, helper-dependent adenoviral vector; hESCs, human embryonic stem cells; hiPSCs, human induced pluripotent stem cells.

Figure 5

Characterization of HB9-EGFP knock-in hESC and hiPSC clones. (a) Expression of stem cell markers in the HB9-EGFP knock-in hESC clones (#47, #G1, and #G10). ALP, alkaline phosphatase. Bars, 200 mm. (b) Multipotency of the HB9-EGFP knock-in hESC clones. Embryoid bodies (EBs) derived from the hESCs were analyzed by RT-PCR for expression of the lineage-specific markers. GAPDH was used as a control. RT (–), without reverse transcriptase. (c) In vivo differentiation of HB9-EGFP knock-in hESC (#47) and hiPSC (#24) clones. Tissues derived from three germ layers are indicated. Bars, 50 mm. EGFP, enhanced green fluorescent protein; hESC, human embryonic stem cell; hiPSC, human induced pluripotent stem cell; RT-PCR, reverse transcription.