Generation of transgene-free induced pluripotent mouse stem cells by the piggyBac transposon

Abstract

Induced pluripotent stem cells (iPSCs) have been generated from somatic cells by transgenic expression of Oct4 (Pou5f1), Sox2, Klf4 and Myc. A major difficulty in the application of this technology for regenerative medicine, however, is the delivery of reprogramming factors. Whereas retroviral transduction increases the risk of tumorigenicity, transient expression methods have considerably lower reprogramming efficiencies. Here we describe an efficient piggyBac transposon-based approach to generate integration-free iPSCs. Transposons carrying 2A peptide-linked reprogramming factors induced reprogramming of mouse embryonic fibroblasts with equivalent efficiencies to retroviral transduction. We removed transposons from these primary iPSCs by re-expressing transposase. Transgene-free iPSCs could be identified by negative selection. piggyBac excised without a footprint, leaving the iPSC genome without any genetic alteration. iPSCs fulfilled all criteria of pluripotency, such as pluripotency gene expression, teratoma formation and contribution to chimeras. piggyBac transposon-based reprogramming may be used to generate therapeutically applicable iPSCs.

Figure 1. Construction of piggyBac transposon-mediated reprogramming vectors

(a) Schematic representation of piggyBac transposons. Four or five factors were linked through 2A peptides. Expression of the factors was driven by the constitutively active CAG promoter. bpA, bovine growth hormone polyadenylation signal; puΔtk, PGK promoter-driven puΔtk expression cassette; PB5′ and PB3′, terminal repeats of the piggyBac transposon. (b) Western blot analyses of 2A peptide-mediated expression of the reprogramming factors. The left most lanes of each panel (individual) are positive control transfections using unlinked factor expression vectors. ‘−’, a negative control transfection using an eGFP expression vector.

Figure 2. Generation of primary iPSCs using piggyBac transposons

(a) Time schedule of transposon-mediated reprogramming. EFM, embryonic fibroblast medium; F15L, serum-based ESC medium; K15L, serum-free ESC medium. Triangles, medium change. (b) Alkaline phosphatase staining of day 14 colonies. ‘−’, a negative control transfection using an eGFP expression vector. (c) Enlarged view of AP-stained colonies with (bottom) and without (top) VPA treatment. (d) Immunostaining of Nanog at day 14 with (bottom) and without (top) VPA treatment. Left, phase contrast; right, anti-Nanog immunostaining. (e) The numbers of colonies obtained by transposon-based reprogramming. OSKM* is identical to OSKM except that Sox2 and Klf4 are linked by F2A instead of T2A. (f) Transposon copy number distribution depending on the transposon plasmid amount. The OSKM* vector was used for the 2.0 μg and the 0.50 μg transfection, and the OSKML vector for the 0.25 μg transfection. *** Only one single-copy transposon iPS colony was identified in the experiments. Data are shown as mean ± s.d. Experiments were performed at least 3 times per each condition. **, p<0.05 (Student's t test). Scale bars, 5 mm (b), 100 μm (c, d)

Figure 3. Transposon removal from the established iPSC lines

(a) Schematic representation of transposon removal. After piggyBac transposase transfection, cells in which transposons either do not transpose or mobilize into other loci are FIAU sensitive, whereas cells in which transposons do not re-integrate elsewhere become FIAU resistant. (b-e) Evidence of transposon-free iPSC lines shown by Southern blot analysis (b) and genomic PCR (c-e). (b) The piggyBac transposon 5′ repeat was used as a probe. Parental iPSC lines have 2 copies of transposon integration (iPS25, 28, and 216). Note that in the HindIII digest of iPS28, two fragments have similar sizes and migrate at the same position (see also Supplemental Fig. 3d). FIAU-resistant sub-clones (designated Δ) did not possess any transposons. The blot was re-hybridized with a control probe to show equal sample loading. (c-e) Specific primers were designed to detect individual transposon integration sites. MEFs and each primary iPSC clone serve as negative and positive controls, respectively. (f, g) Evidence of perfect restoration of the original sequences after transposon excision in iPS28-derived clones. The top panels are the original genomic sequences. The middle panels show transposon integration sites where TTAA sequences were duplicated at both ends of the transposon. The electrophoregrams in the bottom panels show that sequences in iPS28-derived clones after excision are identical to the original sequences in MEFs. Integration site 1 on chr.6 (f) and site 2 on chr.15 (g) are shown.

Figure 4. Characterization of integration-free iPSCs

(a) Normal karyotype of the integration-free iPSC line iPS25Δ1. (b) RT-PCR analyses of integration-free iPSC lines, compared to expression in ESCs. (c) Immunofluorescent analysis of Oct4 and Nanog expression in integration-free iPSCs. (d) Bisulphite sequencing of the promoter region of Oct4 and Nanog. Open and closed circles indicate unmethylated and methylated CpG dinucleotides, respectively. (e) Teratomas generated from integration-free iPSCs, showing differentiation to all three germlayers. Scale bar, 100 μm

Figure 5. Contribution of integration-free iPSCs to somatic tissue and germ line in chimeras

(a, b) Chimeric embryos at 10.5 d.p.c. (a) and postnatal day 5 pups (b) derived from integration-free iPSCs with constitutive eGFP expression. GFP images are shown at the bottom. (c) Developing gonad of an e12.5 embryo derived from Nanog-GFP knock-in iPSCs. GFP image is shown below. The broken line outlines the gonad. Scale bars, 1 mm (a), 5 mm (b), 200 μm (c)

Figure 6. Schemes to generate transgene-free iPSCs

Scheme A (black arrows): After transfection of transposon and transposase DNA into MEFs to generate primary iPSCs, individual iPSC colonies are picked to identify colonies with 2 copy transposon integrations, by Southern blot analysis. These clones are then expanded and the transposase is re-expressed to remove the transposons. HSVtk-FIAU negative selection is used to identify integration-free iPSCs. Scheme B (red arrows): Using an optimized protocol, analysis of individual clones can be bypassed. Primary iPSCs can be pooled and directly subjected to transposon removal and HSVtk-FIAU selection.