CRISPR/Cas9-based genetic correction for recessive dystrophic epidermolysis bullosa

www.nature.com/npjregenmed ARTICLE OPEN CRISPR/Cas9-based genetic correction for recessive dystrophic epidermolysis bullosa Beau R Webber1,5, Mark J Osborn1,2,3,4,5, Amber N McElroy1, Kirk Twaroski1, Cara-lin Lonetree1, Anthony P DeFeo1, Lily Xia1, Cindy Eide1, Christopher J Lees1, Ron T McElmurry1, Megan J Riddle1, Chong Jai Kim4, Dharmeshkumar D Patel1, Bruce R Blazar1,2 and Jakub Tolar1,2,4 Recessive dystrophic epidermolysis bullosa (RDEB) is a severe disorder caused by mutations to the COL7A1 gene that deactivate production of a structural protein essential for skin integrity. Haematopoietic cell transplantation can ameliorate some of the symptoms; however, significant side effects from the allogeneic transplant procedure can occur and unresponsive areas of blistering persist. Therefore, we employed genome editing in patient-derived cells to create an autologous platform for multilineage engineering of therapeutic cell types. The clustered regularly interspaced palindromic repeats (CRISPR)/Cas9 system facilitated correction of an RDEB-causing COL7A1 mutation in primary fibroblasts that were then used to derive induced pluripotent stem cells (iPSCs). The resulting iPSCs were subsequently re-differentiated into keratinocytes, mesenchymal stem cells (MSCs) and haematopoietic progenitor cells using defined differentiation strategies. Gene-corrected keratinocytes exhibited characteristic epithelial morphology and expressed keratinocyte-specific genes and transcription factors. iPSC-derived MSCs exhibited a spindle morphology and expression of CD73, CD90 and CD105 with the ability to undergo adipogenic, chondrogenic and osteogenic differentiation in vitro in a manner indistinguishable from bone marrow-derived MSCs. Finally, we used a vascular induction strategy to generate potent definitive haematopoietic progenitors capable of multilineage differentiation in methylcellulose-based assays. In totality, we have shown that CRISPR/Cas9 is an adaptable gene-editing strategy that can be coupled with iPSC technology to produce multiple gene-corrected autologous cell types with therapeutic potential for RDEB. npj Regenerative Medicine (2016) 1, 16014; doi:10.1038/npjregenmed.2016.14; published online 8 December 2016 INTRODUCTION Recessive dystrophic epidermolysis bullosa (RDEB) is a monogenic disorder resulting from mutations in the type VII collagen gene (COL7A1) on chromosome 3. The mutational profile can be heterogeneic in regards to position and can encompass homozygous or compound heterozygous alterations.1 The resultant loss of the functional type VII collagen protein (C7) at the dermalepidermal junction compromises the integrity of the attachment of the epidermis to the dermis, resulting in severe blistering, fibrosis and a predisposition to squamous cell carcinoma. Non-cutaneous manifestations, including corneal and oesophageal lesions, further contribute to a pathogenic state leading to a multi-decade decrease in life expectancy. Treatment for RDEB includes palliative bandaging of active wounds and pain management, as well as allogeneic and autologous cellular therapy. Palliation is non-curative, and cellular therapy can include localised injection of type VII collagenexpressing cells and/or systemic infusion of haematopoietic stem/progenitor cells (HSPCs) that repopulate the host with donor-derived cells.2 Keratinocytes and fibroblasts represent the major C7 producing cells of the skin; however, their poor in vitro proliferative and expansion properties as primary cells limit their therapeutic potential and impact. Mesenchymal stromal/stem cells (MSCs) have been used as a supportive therapy and possess wound migratory potential and the ability to actively participate in, as well as to orchestrate, healing.3,4 Similar to other primary cells, primary bone marrow-derived MSCs can senesce and lose their beneficial properties with in vitro expansion. Towards mediating systemic effects, allogeneic haematopoietic cell transplant (HCT) has been employed. HCT has resulted in significant, but neither uniform nor complete, outcomes.5 For each modality, the use of allogeneic cells limits efficacy. Locally injected cells appear to persist transiently, likely due to immune clearance, necessitating repeated injections that is limiting in terms of the difficulty in long-term culture/maintenance, surface area able to be treated, and availability of allogeneic cells that can be obtained, archived and expanded for subsequent injections.6 HCT can result in graft-versus-host disease that can cause severe side effects, making the use of autologous cells highly desirous. To realise the potential of such an approach, we set out to determine whether an RDEB patient’s COL7A1 gene defect could be restored to wild-type status in a population of cells that could be utilised as a template for sustainable multilineage progeny generation. Two major platforms exist for facilitating gene correction: gene therapy and gene editing. Gene therapy for RDEB has centred primarily on lentiviral gene transfer of a copy of the COL7A1 cDNA, expression of which is governed by exogenous regulatory elements.7,8 While this strategy meets the need for autologous 1 Department of Pediatrics, Division of Blood and Marrow Transplantation, University of Minnesota, Minneapolis, MN, USA; 2Stem Cell Institute, University of Minnesota, Minneapolis, MN, USA; 3Center for Genome Engineering, University of Minnesota, Minneapolis, MN, USA and 4Asan-Minnesota Institute for Innovating Transplantation, Seoul, Republic of Korea. Correspondence: J Tolar () 5 These authors contributed equally to this work. Received 1 August 2016; revised 17 October 2016; accepted 26 October 2016 Published in partnership with the Australian Regenerative Medicine Institute CRISPR/Cas9 genetic correction for RDEB BR Webber et al 2 Figure 1. Experimental schema for gene correction and cellular engineering. (a) A punch biopsy was obtained for primary fibroblast cell derivation. (b) The CRISPR/Cas9 gene-editing platform was employed for 4317delC COL7A1 gene correction. (c) COL7A1 locus and gene repair template. The 4317delC mutation is indicated with a red box. The donor template was a plasmid containing ~ 1 kb of homology to the target sequence and flanked a floxed PGK puromycin selection cassette (yellow box). The cytosine to restore proper genotype and two silent polymorphisms were introduced into the donor arm and is indicated with a green box. (d) COL7A1 locus correction. Sanger sequence of uncorrected cells before treatment showing a deletion of a single cytosine and unmodified base sequences (top). Subsequent CRISPR/Cas9 mediated repair by the donor resulted in restoration of the deleted cytosine (shaded in blue) and incorporation of engineered marker SNPs (blue arrows). (e) The corrected fibroblasts were reprogrammed to pluripotency using Sendai virus delivery of the reprogramming factors. (f) iPSCs served as a template for directed differentiation into ker (...truncated)