A ZFN/piggyBac step closer to autologous liver cell therapy


  • Aras N. Mattis M.D., Ph.D.,

    1. Eli and Edythe Broad Center of Regeneration Medicine and Stem Cell Research University of California, San Francisco, San Francisco, CA
    2. Department of Pathology University of California, San Francisco, San Francisco, CA
    Search for more papers by this author
  • Holger Willenbring M.D., Ph.D.

    1. Eli and Edythe Broad Center of Regeneration Medicine and Stem Cell Research University of California, San Francisco, San Francisco, CA
    2. Department of Surgery, Division of Transplantation University of California, San Francisco, San Francisco, CA
    3. Liver Center University of California, San Francisco San Francisco, CA
    Search for more papers by this author

Yusa K, Rashid ST, Strick-Marchand H, Varela I, Liu PQ, Paschon DE, et al. Targeted gene correction of α1-antitrypsin deficiency in induced pluripotent stem cells. Nature 2011;478:391-394. www.nature.com (Reprinted with permission)


Human induced pluripotent stem cells (iPSCs) represent a unique opportunity for regenerative medicine because they offer the prospect of generating unlimited quantities of cells for autologous transplantation, with potential application in treatments for a broad range of disorders. However, the use of human iPSCs in the context of genetically inherited human disease will require the correction of disease-causing mutations in a manner that is fully compatible with clinical applications. The methods currently available, such as homologous recombination, lack the necessary efficiency and also leave residual sequences in the targeted genome. Therefore, the development of new approaches to edit the mammalian genome is a prerequisite to delivering the clinical promise of human iPSCs. Here we show that a combination of zinc finger nucleases (ZFNs) and piggyBac technology in human iPSCs can achieve biallelic correction of a point mutation (Glu342Lys) in the α1-antitrypsin (A1AT, also known as SERPINA1) gene that is responsible for α1-antitrypsin deficiency. Genetic correction of human iPSCs restored the structure and function of A1AT in subsequently derived liver cells in vitro and in vivo. This approach is significantly more efficient than any other gene-targeting technology that is currently available and crucially prevents contamination of the host genome with residual non-human sequences. Our results provide the first proof of principle, to our knowledge, for the potential of combining human iPSCs with genetic correction to generate clinically relevant cells for autologous cell-based therapies.


Clinical evidence suggests that hepatocyte replacement therapy has potential as a less invasive alternative to liver transplantation.1 To render hepatocyte replacement therapy independent of scarce donor livers, much effort is currently being devoted to establishing embryonic stem cells or induced pluripotent stem cells (iPSCs) as a source of therapeutically effective and safe hepatocytes. iPSCs can be generated from readily accessible somatic cells, which facilitates autologous liver cell therapy.2 By bypassing the need for chronic immune suppression, autologous iPSC-based liver cell therapy may avoid not only drug side effects, but also progressive loss of therapeutic efficacy observed after allogeneic hepatocyte transplantation.

Genetically encoded liver diseases with little or no fibrosis are considered to be the most promising targets for hepatocyte replacement therapy. Therefore, development of autologous iPSC-based liver cell therapy will require effective and safe ways to restore gene function. Introducing a wild-type copy of the mutated gene into a safe but ectopic locus may be sufficient in some liver diseases. Ideally, however, the mutated sequence is corrected to maintain physiological gene regulation and prevent accidental disruption or activation of other genes. Moreover, in the most common genetically encoded liver disease, α1-antitrypsin (A1AT) deficiency,3 gene correction is necessary to prevent hepatocyte damage due to intracellular accumulation of misfolded mutant A1AT protein. Yusa et al.4 developed a strategy that combines the power of zinc finger nucleases (ZFNs) and piggyBac transposase to genetically correct iPSCs derived from patients with A1AT deficiency.

A1AT, a serpin superfamily protease inhibitor (Pi), is produced and secreted by hepatocytes to protect the lungs from neutrophil elastase.3 A range of mutant forms of A1AT exist. Mild mutants show slightly decreased serum levels of A1AT (60% of normal), whereas the homozygous Z genotype (PiZZ) causes a severe reduction to 10%-15% of normal serum A1AT levels. Individuals with the PiZZ genotype often show accumulation of the misfolded protein in hepatocytes.5 Over time, lack of A1AT in the blood leads to emphysema, whereas accumulation of misfolded A1AT in hepatocytes leads to liver fibrosis and cancer. To reduce progression of emphysema, patients can receive recombinant A1AT protein. Strategies to reduce the accumulation of misfolded A1AT protein in hepatocytes, such as the autophagy-promoting drug carbamazepine,6 are in development, but no definitive treatment is currently available. Therefore, A1AT deficiency is a promising target for hepatocyte replacement therapy with cells derived from gene-corrected autologous iPSCs.

To develop a gene-correction strategy that would be safe enough for clinical application, Yusa et al. relied on homologous recombination. Because spontaneous homologous recombination is inefficient in iPSCs,7 they used ZFNs to stimulate the process. ZFNs create double-stranded DNA breaks in a sequence-specific fashion.8 They are designed around two components, the zinc finger DNA binding motif and the FokI endonuclease. Recent insights into zinc finger DNA recognition have enabled targeting the activity of FokI to specific nucleotide sequences. Each zinc finger array recognizes approximately three base pairs but can be linked to additional arrays to recognize nine basepairs or more, thereby increasing sequence specificity. Because FokI is only active when dimerized, pairing ZFNs that recognize distinct, but adjacent sequences is typically used to further minimize off-target cleavage. ZFNs have been used to generate double-stranded DNA breaks to stimulate nonhomologous end-joining, or to induce homologous recombination with a donor sequence in a specific genomic locus.

To allow specific expansion of iPSCs that underwent homologous recombination, Yusa et al. delivered a homologous donor sequence in tandem with a drug selection cassette. Because their goal was to generate gene-corrected iPSCs with no or little additional genomic modification, they designed the selection cassette so that it could eventually be excised. For this purpose, they used piggyBac transposase. In contrast to genome editing systems based on Cre recombinase or sleeping beauty transposase,9piggyBac affords site-specific excision without leaving behind a large footprint.10 Furthermore, piggyBac-mediated transposition is not associated with a high frequency of reintegration events.9

Yusa et al. started out with iPSC lines carrying the PiZZ genotype that were generated from patient fibroblasts by transduction with retroviruses expressing the four Yamanaka factors.11 They transfected the cells with plasmids expressing two ZFNs that targeted sequences immediately left and right of the Z mutation, respectively, and another plasmid encoding wild-type A1AT as donor sequence for homologous recombination (Fig. 1, step 1). Cleavage by the ZFNs stimulated homologous recombination and led to correction of one allele in 54% of the iPSC colonies emerging after drug selection; both alleles were corrected in 4% of the colonies (Fig. 1, step 2). Because the selection cassette was flanked by piggyBac TTAA recognition sequences, most of the foreign DNA was simply removed by additional transfection of a plasmid expressing piggyBac transposase (Fig. 1, step 3). piggyBac-mediated loop out of the selection cassette left behind only a single foreign TTAA sequence in the iPSC genome (Fig. 1, step 4). Importantly, Yusa et al. had positioned this TTAA sequence in the donor plasmid so that the resulting conversion from CTG to TTA in the iPSC genome maintained the wild-type A1AT amino acid code. After cells that had retained the drug selection cassette were eliminated, biallelic correction was detected in 11% of the iPSC colonies (Fig. 1, step 5).

Figure 1.

Strategy of ZFN/piggyBac-mediated correction of PiZZ genotype in human iPSCs. (1) Transfection of plasmids expressing ZFNs and wild-type A1AT and drug selection cassette. (2) Homologous recombination replacing PiZZ with wild-type sequence and selection cassette allowing isolation of gene-corrected iPSC colonies. (3) Transfection of plasmid encoding piggyBac transposase (T) targeting TTAA sequences flanking the selection cassette. (4) Loop out of selection cassette. (5) Final isolation of iPSC colonies that have a wild-type (WT) A1AT genotype and no selection cassette.

To establish that this strategy of PiZZ correction facilitated normal cell function, Yusa et al. differentiated the repaired iPSC lines into hepatocytes using a previously reported protocol.11 Indeed, the cells efficiently acquired characteristic functions of primary hepatocytes and, importantly, secreted normal A1AT while lacking signs of accumulation of the mutant protein. In addition, after transplantation into immunodeficient mice with hepatocyte injury due to overexpression of urokinase plasminogen activator, hepatocytes derived from gene-corrected iPSCs formed clusters and secreted albumin into the mice's serum.

Finally, Yusa et al. investigated the genomic integrity and thus the safety profile of corrected iPSC lines. Most of the amplifications, deletions, and mutations they detected had occurred in the process of reprogramming to pluripotency or subsequent cell culture, which is in accordance with previous reports.12-14 However, a few mutations manifested during the process of gene correction. The nature of these mutations suggested that they were not the result of off-target cleavage, a known complication of ZFN-mediated gene correction,8 nor of piggyBac-mediated excision of the selection cassette. Furthermore, although these mutations occurred in protein-coding genes, they did not appear to affect the function of hepatocytes derived from the corrected iPSC lines. iPSC-derived hepatocytes also did not form tumors after transplantation, but larger numbers of recipient mice and longer observation periods are needed to conclude that these mutations do not impair safety.

As a further step toward clinical application, Yusa et al. successfully used their method to correct the PiZZ genotype of iPSCs generated with Sendai viruses. In contrast to retroviruses, Sendai viruses do not integrate into the genome and are therefore considered a safer method of iPSC generation.15 Taken together, these findings suggest that combining an integration-free method of reprogramming with this new ZFN/piggyBac-based strategy of gene correction produces iPSC lines that would be safe enough for clinical application. However, the findings also underscore the need for careful genomic analysis of iPSC lines, whether naïve or gene-corrected, if they are intended for cell therapy. Furthermore, the study highlights some of the roadblocks that remain to be overcome before therapy of A1AT deficiency with hepatocytes derived from autologous iPSCs can be attempted. Foremost, protocols need to be developed that produce iPSC-derived hepatocytes that more closely resemble primary hepatocytes in both function and ability to proliferate. Then, hepatocyte replacement therapy will likely have to be combined with strategies that reduce accumulation of mutant A1AT protein in residual mutant hepatocytes. Although A1AT deficiency may promote the expansion of transplanted iPSC-derived hepatocytes to an extent that would be sufficient for preventing emphysema,16 replacing all mutant hepatocytes will not be possible. Therefore, without additional application of drugs like carbamazepine to reduce the mutant protein load,6 chronic injury of residual mutant hepatocytes may lead to liver fibrosis and cancer. Although several major tasks remain to be accomplished, the study by Yusa et al. has moved the field an important step closer to the realization of autologous liver cell therapy of A1AT deficiency and potentially other genetically encoded liver diseases.