Assistant Professor New York University New York, New York, United States
Introduction:: Human induced pluripotent stem cells (iPSCs) could provide a renewable source for cell therapies and regenerative medicine, but in the clinic, currently must be sourced directly from each patient to avoid immune rejection. Alternatively, a donor-derived “Universal” iPSC line that could be used “off-the-shelf” without immune rejection, might serve as a go-to platform cell for scalable biomanufacturing. This would enable safer, centralized, and reproducible Design-Build-Test cycles for genetic programming of advanced cellular functions in differentiated cell types, like T-cells, Dendritic Cells, or neurons. Banks of iPSCs pre-matched to ethnic populations at the polymorphic class-I HLA genes (human leukocyte antigen) could perhaps provide this “off-the-shelf” solution, but finding donors with rare HLA haplotypes for underserved populations is statistically challenging.
Synthetic Genomics advances in DNA assembly and genome engineering might provide an on-demand bespoke solution, inclusive of minorities with rare HLA types. However, the HLA locus, composed of multiple polymorphic gene homologs, is one of the most challenging genomic loci to manipulate due to its size ( > megabase) and sequence repetition, especially in human iPSCs. Moreover, there are limited options for genetically modifying human iPSCs at the locus (100+ kilobase, or “bigDNA”) scale. Our goal is to develop an approach for site-specifically “rewriting” sections of the human genome, to first build this Universal iPSC platform cell line. The eventual goal is to engineer large-scale synthetic genetic circuits able to control cellular differentiation (forward direct reprogramming), cell-to-cell communication (exosome therapies), and theranostics (cell therapies) in one scalable cellular platform.
Materials and Methods:: Here, we developed a Genome Writing approach called REcombinase WRiting of Iterative DNA and Trap Excision (REWRITE) for genetically engineering any large desired DNA sequences (bigDNA) into human iPSCs. REWRITE combines CRISPR, recombinases, and synthetic genome assembly in a two-step process to enable nearly scarless site-specific genome engineering at the locus scale (100+ kilobases) in hiPSCs. The ability to do sequential integrations also presents opportunities to reach the megabase scale. In addition to CRISPR/Cas9, we developed the use of CRISPR/erCas12a (Mad7) for use in human iPSCs.
REWRITE first begins with bi-allelically removing a desired section of the human genome of up to 100 kilobases. (1) High-specificity and high-activity Cas9 and/or Mad7 gRNAs are designed for the genomic region to generate double-strand breaks. These site-specific breaks are then used to stimulate, through homology-directed repair, the simultaneous removal of the genomic region and integration of the REWRITE “landing pad” construct on one allele, and a second “deleter” construct on the other allele. (2) Designer bigDNAs are then assembled from multiple linear DNA fragments of ~3-5 kilobases into one larger bigDNA of up to 100 kilobases using yeast recombineering. (3) The bigDNA is then nucleofected into the REWRITE cell line, and recombinases are used to integrate the bigDNA into the landing pad. (4) Finally, an inducible “self-deletion” cassette carried in the bigDNA called “FlpOUT” is activated to remove selectable markers, thereby making the genetic modification nearly scarless.
Results, Conclusions, and Discussions:: Initial REWRITE “landing pad” designs were genetically silenced in mouse ESCs. We eliminated DNA silencing by including a UCOE insulator upstream of our promoter, and generated new constructs for routine REWRITE use. We integrated a 10-kilobase human DNA sequence into 106 nucleofected mESCs, finding >270 clones. We then activated the FlpOUT cassette, and found that it efficiently removed the selectable markers in ~46% of clones. Finally, we showed bi-allelic removal of as much as 1.7 megabases in mESCs.
We placed REWRITE at 5 genomic locations in human iPSCs, both transcriptionally active and silenced. In these locations, the landing pad was never silenced after months of culture. We verified the pluripotency of these iPSC lines using embryoid body assays and qPCR-based scores. We then integrated a series of < 10-kb constructs, observing robust integration.
To generate our HLA-matching universal iPSC, we first needed to generate an “HLA blank” iPSC devoid of class-I HLAs. There are three class-I genes, HLA-A, HLA-B, and HLA-C, all on chromosome 6. HLA-B and -C are 93-kilobases apart, whereas HLA-A is 1.3-megabases further. Using MAD7, we deleted both alleles of the HLA-A locus (8-kb). We attempted bi-allelic removal of the 93-kb HLA-B/C locus. Bi-allelic deletions in human iPSCs appear far more challenging than in mESCs. We readily made many mono-allelic versions. After optimizing conditions, we successfully obtained >2 distinct human iPSC lines with bi-allelic removal of the 93-kb HLA-B/C locus, thus making a blank line.
We determined that 12 HLA haplotypes would match >50% of europeans/chinese. Using yeast recombineering, we constructed a “Designer” HLA locus (“synHLA”) relocating HLA-A inbetween HLA-B/C. This refactored 100-kb locus retains native transcriptional regulation, linking all three HLAs. We built three ethnically rare synHLA haplotypes. We also constructed a “mini-synHLA”, lacking intergenic DNA, of 35-kilobases, and are constructing the 12 HLA haplotypes.
We tested integration of mini-synHLA in monoallelic deletion iPSCs, observing >80 integrants of 35-kb mini-synHLA. Finally, we integrated the 100-kb synHLA, fully “rewriting” HLAs on chromosome 6, although the number of clones was < 10.
We demonstrate the first “restructuring” of gene location in the human genome. More generally, REWRITE will enable >100-kb locus rewriting in human iPSCs. Finally, synthetic HLA-tailored iPSCs will enable a rapid on-demand solution for off-the-shelf cell therapies and tissues.
Acknowledgements (Optional): : We would like to acknowledge members of the Truong lab and the Neochromosome Inc team. David Truong is a former employee of Neochromosome. We would like to acknowledge generous funding from NIAID grants R43AI148008 and DP2AI154417.