WO2015195547A1 - Methods for controlling stem cell potential and for gene editing in stem cells - Google Patents

Abstract

The present invention provides methods for controlling developmental potential of a human stem cell via gene editing of a lineage-specification gene in a human stem cell genome, wherein the gene editing produces a human stem cell with limited capability of differentiating into the cell lineage for which the lineage-specification gene is specific. The invention also provides recombinant human stem cells made using the methods of the invention, and uses for such recombinant human stem cells.

Description

Methods for Controlling Stem Cell Potential and for Gene Editing in Stem Cells Cross Reference

This application claims priority to U.S. Provisional Patent Application Serial Number 62/012539 filed June 16, 2014, incorporated by reference herein in its entirety.

Federal Funding Statement:

This invention was made with government support under Grant No. R01 DK55759, awarded by the National Institutes of Health. The government has certain rights in the invention.

Background of the Invention

Human stem cells are being developed for regenerative medicine applications in which they are differentiated in vitro or in vivo into more mature cell types. In many cases, it would be desirable to begin with a stem cell that is limited in its ability to form certain cell types. For example, when using human pluripotent stem cell (hPSC)-derived hematopoietic cells, it may be desirable to prevent these cells from producing T lymphocytes and Natural Killer (NK) cells that could cause graft-vs-host- disease (GVHD), but still allow the cells to produce other types of blood cells like platelets or neutrophils.

Summary of the Invention

In a first aspect, the invention provides methods for controlling developmental potential of a human stem cell, comprising gene editing of a lineage-specification gene in a human stem cell genome, wherein the gene editing produces a human stem cell with limited capability of differentiating into the cell lineage for which the lineage-specification gene is specific. In one embodiment, the gene editing comprises knocking out the lineage- specification gene. In another embodiment, the gene editing comprises

(a) inserting a cassette comprising a negative selection marker into the lineage specification gene of the human stem cell genome, wherein the negative selection marker is under the control of the lineage-specification gene promoter and/or enhancer;

(b) isolating human stem cells that contain the negative selection marker; (c) culturing the human stem cells under conditions suitable to express the negative selection marker and exert negative selection, wherein the culturing eliminates at least 90% of cells of the lineage for which the lineage-specification gene is specific from the cultured cell population. In a further embodiment, the negative selection marker is selected from the group consisting of thymidine kinase an apoptosis inducer, and a toxic gene.

In another embodiment, the gene editing comprises:

(a) inserting a cassette comprising a selection marker into the lineage

specification gene of the human stem cell genome, wherein the selection marker is operatively linked to an exogenous promoter;

(b) culturing the human stem cells under conditions suitable to express the selection marker; and

(c) isolating positive human stem cells that stably express the selection marker; wherein positive human stem cells comprise the selection marker under the control of the exogenous promoter integrated into the lineage specification gene such that expression of the lineage specification gene is suppressed or its encoded protein is no longer functional. In a further embodiment, the selection marker is an antibiotic resistance gene, a surface marker that can be used for cell purification, a metabolic gene that confers survival in the presence of a specific medium formulation, and/or a gene that provides a growth advantage.

In another embodiment, the linage specification gene locus is silent in the human stem cell. In a further embodiment, the exogenous promoter is active at silent gene loci. In one such embodiment, the exogenous promoter is a housekeeping promoter. In a further such embodiment, the exogenous promoter comprises the nucleic acid sequence selected from the group consisting of SEQ ID NOS: l-6, or functional equivalents thereof.

In a further embodiment, the cassette further comprises a negative selectable marker under the control of the lineage-specification gene promoter and/or enhancer. In a still further embodiment, the cassette further comprises an inactivating mutation in the lineage- specification gene, and or its promoter and/or enhancer.

In one embodiment, the method further comprises excising the cassette or a portion thereof from the positive human stem cells. In a further embodiment, the edited locus retains a negative selectable marker or inactivating mutation after excising the cassette.

In another embodiment, the lineage specification gene is selected from the group consisting of a gene encoding interleukin-2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage-specification cell surface marker, a lineage- specification transcription factor, a cytokine or hormone receptor. In another embodiment, the cassette is delivered to the cell by an adenoviral or rAAV vector. In a further

embodiment,

the human stem cell is a pluripotent stem cell or an induced pluripotent stem cell.

In a further aspect, the invention provides recombinant human pluripotent stem cells knocked-out for a lineage specification gene. In one embodiment, a negative selection marker, including but not limited to thymidine kinase an apoptosis inducer, and a toxic gene, is knocked into the lineage-specification gene locus, under the control of the lineage- specification gene promoter and/or enhancer.

In another aspect, the invention provides recombinant human pluripotent stem cells comprising a selection marker and exogenous promoter, including but not limited to an antibiotic resistance gene, inserted into the lineage-specification gene, wherein the selection marker inactivates the lineage specification gene.

In one embodiment of the recombinant human pluripotent stem cells of any aspect of the invention, the linage specification gene locus is silent in the human stem cell. In a further embodiment, the exogenous promoter is active at silent gene loci. In another embodiment, the promoter is a housekeeping promoter. In a still further embodiment, the promoter comprises the nucleic acid sequence selected from the group consisting of SEQ ID NOS: l-6, of functional equivalents thereof. In another embodiment, the lineage specification gene is selected from the group consisting of a gene encoding interleukin-2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage-specification cell surface marker, a lineage-specification transcription factor, a cytokine or hormone receptor. Description of the Figures

Figure 1. Targeting a silent COLIAI-IKES-Neo cassette in human iPSCs. (a) Diagram of experimental design, (b) RT-qPCR analysis of COLIAI expression in undifferentiated iPSC clones containing COLIAI -IKES-Neo knockins. Fibro, human fibroblasts; ESC,

undifferentiated HI cells, (c) Structures of wild-type and IRES-Neo targeted COLIAI alleles in iPSC clone 1 with rAAV promoter knock-in vector overlap indicated. The targeted

COLIAI locus contains two identical IRES-Neo cassettes, each of which can be targeted with rAAVs. Black triangles, primer binding sites used for qPCR measurements of homologous recombination frequencies, (d) G418 resistance frequencies of iPSC clone 1 infected with promoter knock-in rAAVs. *, less than 4xl0^~5. (e) Homologous recombination frequencies measured by qPCR with primers shown in c. Each infected cell population was analyzed with two independent primer pairs. Data represent mean ± SEM of three (b, d) or four (e). Figure 2. Targeting a silent IL2RG gene in human ESCs. (a) Structure of the IL2RG locus and rAAV targeting vector, with the locations of restriction enzyme sites and probe used in B. (b) Southern blot performed on age-matched wild-type (W), targeted (T), and Cre-out (C) clones, digested with Hindlll and Nhel and probed with a 3' chromosomal fragment outside the homology arm, which produces fragments of 3.8, 4.7 and 2.6 kb in wild-type, targeted, and Cre-out clones respectively. The same blot was stripped and re-probed with a Neo fragment, (c) RT-qPCR analysis of gene expression of IL2RG's genomic neighbors in age- matched wild-type, targeted, and Cre-out clones, obtained from two experiments. Gene expression signals were normalized to wild-type samples. *, p < 0.05. Data represent mean ± SEM of three.

Figure 3. Epigenetic consequences of gene editing, (a) Structures of wild-type and IRES- Neo targeted COL1A1 loci shown with rAAV overlap, UCOE insertion sites, and CpG islands. DNA fragments (A to D) amplified in ChIP assays and bisulfite sequencing regions are marked, (b) Methylation status of the region spanning from exon 1 of COL1A1 (blue circles) to IRES (orange circles) in rAAV vector genomes, clone 1 genomic DNA, and clone 1 targeted at either the first or second Neo gene. Open and filled circles indicate unmethylated and methylated cytosines in CpGs respectively, (c) The relative occupancies of H3K27Ac and H3K27Me3 in regions of the COL1A1 locus before and after UCOE insertion as measured by ChIP analysis. *, p < 0.01. (d) Structure of the UCOE-Neo targeted IL2RG locus shown with rAAV overlap, loxP sites, and the locations of DNA fragments A to F amplified in ChIP assays. Histone occupancies were analyzed in age-matched wild-type, targeted, and Cre-out clones. *, p < 0.01. Data represent mean ± SEM of three (c, d).

Figure 4. NK and T cell differentiation of /L2RG-targeted human ESCs. (a, b) Flow cytometry analysis of CD2, CD7, CD 132 and CD56 expression in day 8 (a) and day 22 (b) NK cell differentiation cultures derived from UCB or wild-type, /L2 ? G-targeted or Cre-out ESC-derived EBs. (c) Flow cytometry analysis of CD34 and CD7 expression T cell differentiation cultures of wild-type, /L R G-targeted or Cre-out ESCs harvested at the indicated day. (d) Same as panel c but analyzing CD4 and CD8 expression, (e) Working model of the differentiation block of IL2RG-knockout ESCs.

Figure 5. Targeting a silent COLIAI-IKES-Neo cassette, (a) Spontaneous G418 resistance frequencies of iPSC clones 1 , 2 and 3 containing a silenced IRES-Neo cassette in COL1A1. *, < 2xl0^~5. (b) Examples of qPCR measurements of homologous recombination frequencies when inserting a REXl or UCOE promoter into the COLIAI-IKES-Neo cassette of iPSC clone 1. Red squares, standard values obtained with a cloned reconstructed targeted allele mixed with a wild-type genome; blue squares, sample measurements. Primer locations are shown in Figure lc. (c) Structure of the COLIAI-IKES-Neo cassette dimer in iPSC clone 1 with promoter insertion sites and the restriction enzyme sites used for Southern blot analysis. A Southern blot is shown that analyzes individual, G418 -resistant subclones transduced with the indicated promoter insertion vector, allowing one to identify which Neo gene was targeted. The genomic DNAs were digested with Kpnl and probed for Neo sequences. The sizes of promoter-inserted alleles are shown in the adjacent table, (d) The genomic structure of the UCOE promoter with its two chromosomal genes is shown (CBX3 and HNRNPA2B1), along with the G418-resistance frequencies produced by rAAV vectors containing the indicated full-length and truncated UCOE promoter fragments. *, < 3xl0^"5. Data represent mean ± SEM of three.

Figure 6. IL2RG targeting and UCOE-TVeo removal, (a) Identification of H2i?G-targeted clones by PCR. The structures of wild-type and targeted IL2RG loci are shown with primer binding sites indicated as triangles. An example of a multiplex PCR screening gel is shown with wild-type and targeted clones, (b) Identification of Cre-out clones. The structures of wild-type, targeted, and Cre-out IL2RG loci are shown with four primer binding sites indicated as triangles. Multiplex PCR was performed on clones after transducing with a non- integrated foamy virus that transiently expresses Cre recombinase. The efficiency of UCOE- Neo transgene removal is calculated as the number of Cre-out clones / (number of targeted clones + number of Cre-out clones) x 100.

Figure 7. CpG methylation at the IL2RG locus, (a) Structure of the UCOE-Neo targeted IL2RG locus shown with rAAV overlap, loxP sites, and the locations of bisulfite sequencing fragments. The IL2RG locus contains no CpG islands, (b) Open and filled circles indicate unmethylated and methylated cytosines respectively in the CpGs assayed in exons 1 and 5.

Figure 8. Characterization of T and NK cells derived from ESCs. (a) Total number of CD14-, CD56+ NK lineage cells produced during NK differentiation of UCB, or wild-type, H2i?G-targeted and Cre-out EBs. (b) Chromium release assays showing lysis of MHC- negative K562 cells by peripheral blood (PB)-, UCB-, or wild-type ESC-derived NK cells. NK:K562 cell ratios are indicated, (c) Flow cytometry analysis showing surface expression of CD45 (hematopoietic), CD 14 (monocyte) and CD 15 (granulocyte) markers in cells obtained after four weeks of NK differentiation from H2i?G-targeted and Cre-out EBs. (d) Flow cytometry analysis of surface expression of lymphoid markers CD5 and CD7 on different days of T cell differentiation in wild-type, H2i?G-targeted and Cre-out EBs.

Detailed Description of the Invention

All references cited are herein incorporated by reference in their entirety. Within this application, unless otherwise stated, the techniques utilized may be found in any of several well-known references such as: Molecular Cloning: A Laboratory Manual (Sambrook, et al,

1989, Cold Spring Harbor Laboratory Press), Gene Expression Technology (Methods in Enzymology, Vol. 185, edited by D. Goeddel, 1991. Academic Press, San Diego, CA),

"Guide to Protein Purification" in Methods in Enzymology (M.P. Deutshcer, ed., (1990) Academic Press, Inc.); PCR Protocols: A Guide to Methods and Applications (Innis, et al.

1990. Academic Press, San Diego, CA), Culture of Animal Cells: A Manual of Basic Technique, 2^nd Ed. (R.I. Freshney. 1987. Liss, Inc. New York, NY), Gene Transfer and Expression Protocols, pp. 109-128, ed. E.J. Murray, The Humana Press Inc., Clifton, N.J.), and the Ambion 1998 Catalog (Ambion, Austin, TX).

As used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. "And" as used herein is interchangeably used with "or" unless expressly stated otherwise.

All embodiments of any aspect of the invention can be used in combination, unless the context clearly dictates otherwise.

Unless the context clearly requires otherwise, throughout the description and the claims, the words 'comprise', 'comprising', and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of

"including, but not limited to". Words using the singular or plural number also include the plural and singular number, respectively. Additionally, the words "herein," "above," and "below" and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of the application.

The description of embodiments of the disclosure is not intended to be exhaustive or to limit the disclosure to the precise form disclosed. While the specific embodiments of, and examples for, the disclosure are described herein for illustrative purposes, various equivalent modifications are possible within the scope of the disclosure, as those skilled in the relevant art will recognize.

In a first aspect, the invention provides methods for controlling developmental potential of a human stem cell, comprising gene editing of a lineage-specification gene in a human stem cell genome, wherein the gene editing produces a human stem cell with limited capability of differentiating into the cell lineage for which the lineage-specification gene is specific.

As shown in the examples that follow, the inventors have surprisingly demonstrated methods for editing of lineage-specification genes in pluripotent human stem cells that limit the ability of the resulting stem cells to differentiate into the cell lineage for which the lineage-specification gene is specific. Thus, the methods of the invention (and recombinant cells derived using the methods) can be used, for example, to generate cellular disease models to study the function of lineage-specification genes, to correct lineage-specification gene mutations that lead to disease, to prevent graft versus host disease in regenerative medicine applications,

The human stem cell may be any suitable stem cell, including but not limited to embryonic stem cells and somatic stem cells. The stem cell may be a pluripotent stem cell (PSC) or an induced pluripotent stem cell (iPSC). In one embodiment, the human stem cells may be somatic stem cells (PSCs or iPSCs) including, but not limited to, hematopoietic stem cells, mesenchymal stem cells, intestinal stem cells, endothelial stem cells, neural stem cells, neural crest stem cells, olfactory adult stem cells, and mammary stem cells. In another embodiment, the human stem cells are embryonic stem cells.

As used herein, a "lineage-specification gene" is a gene that is required for a stem cell to differentiate into a specific cell lineage, or specifically expressed in that lineage. For example, the interleukin 2 receptor gamma (IL2RG) gene is required for pluripotent stem cells to differentiate into T-cells. Thousands of such lineage-specification genes are known to those of skill in the art, and genes can be identified as lineage-specification genes using standard techniques in the art. In various non-limiting embodiments, the lineage- specification genes may be selected from the group consisting of genes encoding interleukin- 2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage- specific cell surface marker, a lineage-specific transcription factor, a cytokine or hormone receptor.

The methods of the invention result in a human stem cell with limited capability of developing into the cell lineage for which the lineage-specification gene is specific. As used herein, "limited capability" means at least a 90% decrease in the cell lineage for which the lineage-specification gene is specific; in various embodiments, at least a 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or a complete elimination in the human stem cell's capability of developing into the cell lineage for which the lineage-specification gene is specific.

The methods comprise use of gene editing of a lineage-specification gene in a human stem cell genome. Any suitable gene editing technique can be used, including gene editing to result in knock-in mutations, knock-out mutations, homologous recombination, site specific nucleases, zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), CRISPR-mediated editing, disruption of the lineage-specification gene open reading frame, alteration in splicing, inactivation of the poly-adenylation site, inactivation of the promoter and/or enhancer, etc. In one embodiment, the gene editing comprises knocking out the lineage-specification gene. Knock-out of the lineage-specification gene directly results in a human stem cell with limited capability of developing into the cell lineage for which the lineage-specification gene is specific. In one exemplary embodiment, the Brachury gene is knocked out in human pluripotent stem cells (PSCs), which limits the PSCs capability to differentiate into mesoderm, and promotes differentiation of the PSCs into ectoderm (such as neurons) or endoderm (such as hepatocytes). In another exemplary embodiment, the IL2RG gene is knocked out in human PSCs, which limits the PSCs capability to differentiate into NK or T cells, and which can also be used prevent graft versus host disease (GVHD) by transplanting the resulting human PSCs for therapeutic purposes, by substantially reducing/eliminating PSC-derived T cells that may react against HLA- mismatched host cells. In another embodiment, the methods can be used to eliminate dangerous cells. For example, when transplanting stem cell-derived cardiomyocytes, it has been shown that tachycardias can occur with potentially dangerous consequences. Pacemaker cells responsible for this tachycardia can be eliminated by knocking out a gene required for pacemaker formation such as HCN4 (hyperpolarization activated cyclic nucleotide-gated potassium channel 4) or TBX3 (T-box transcription factor 3). This would allow for the preparation of cardiomyocyte cultures that do not cause arrhythmias after transplantation. In another example, cells capable of producing teratomas present in a stem cell population could be eliminated by knocking out a pluripotency gene such as OCT4.

In another embodiment, the gene editing comprises

(a) inserting a cassette comprising a negative selection marker into the lineage specification gene of the human stem cell genome, wherein the negative selection marker is under the control of the lineage-specification gene promoter and/or enhancer;

(b) isolating human stem cells that contain the negative selection marker;

(c) culturing the human stem cells under conditions suitable to express the negative selection marker and exert negative selection, wherein the culturing results in elimination at least 90% of cells of the lineage for which the lineage-specification gene is specific from the cultured cell population.

This embodiment is a targeted "knock-in" of a negative selection marker to be under the control of the lineage-specification gene's promoter, wherein application of selection results in specific elimination of cells that the lineage-specification gene is specific for. Knock-in gene editing targeting the lineage-specification gene can be carried out using standard techniques, including but not limited to via homologous recombination. Any suitable negative selection marker can be used, including but not limited to thymidine kinase apoptosis inducers, and toxic genes. Isolating the human stem cells containing the negative selection marker can be done by standard techniques in the art, including but not limited to polymerase chain reaction to identify the insertion. Conditions for culturing of the human stem cells will depend on the negative selection marker used, but can be determined by those of skill in the art based on the teachings herein. In one embodiment, such knock-in gene editing of human PSCs can be used to eliminate highly specific subtypes of terminally differentiated cells. For example, when differentiating stem cells into pancreatic islet β cells that secrete insulin for use in treating diabetes, the cultures can also contain a cells that secrete glucagon and delta cells that secrete somatostatin. Negative selection markers can be knock-into the Glucagon and/or Somatostatin genes, and then these cells can be eliminated by applying negative selection. In another embodiment, the methods can be used to eliminate dangerous cells.

For example, when transplanting stem cell-derived cardiomyocytes, it has been shown that tachycardias can occur with potentially dangerous consequences. Pacemaker cells responsible for this tachycardia can be eliminated by knocking in a negative selection marker into a pacemaker-specific gene such as HCN4 or TBX3. This would allow for the preparation of cardiomyocyte cultures that do not cause arrhythmias after transplantation.

In another embodiment, the gene editing comprises:

(a) inserting a cassette comprising a selection marker into the lineage

specification gene of the human stem cell genome, wherein the selection marker is operatively linked to an exogenous promoter;

(b) culturing the human stem cells under conditions suitable to express the selection marker; and

(c) isolating positive human stem cells that stably express the selection marker; wherein positive human stem cells comprise the selection marker under the control of the exogenous promoter integrated into the lineage specification gene such that expression of the lineage specification gene is suppressed or its encoded protein is no longer functional.

In this embodiment, an exogenous promoter (i.e.: not the promoter of the lineage- specification gene) controlling expression of a selection marker is inserted into the lineage- specification gene, at any suitable location such that expression of the lineage-specific gene is suppressed. Any suitable selection marker can be used, including but not limited to an antibiotic resistance gene, a surface marker that can be used for cell purification, a metabolic gene that confers survival in the presence of a specific medium formulation, and/or a gene that provides a growth advantage. Conditions for culturing of the human stem cells to express the selection marker will depend on the stem cells, the promoter, and the selection marker, and can be determined by those of skill in the art based on the teachings herein.

In one embodiment of any of the above embodiments, the lineage specification gene locus is silent in the human stem cell. As used herein "silent" means that transcripts produced from the linage specification gene are present at a level in the lowest 10% of all genes in the human stem cell. Many applications of pluripotent stem cells (PSCs) require efficient editing of silent chromosomal genes. Transcription is known to increase

homologous recombination and gene targeting by trans fection-based methods. As shown in the examples that follow, the lower targeting frequencies observed at silent loci are due to inadequate expression of the selectable marker gene after it integrates, rather than a decrease in homologous recombination.

In various embodiments, transcripts produced from the linage specification gene are present at a level in the lowest 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, or 0.1% of all genes in the human stem cell. In another embodiment, no transcripts are produced from the lineage specification gene in the human stem cell.

In a further embodiment, the exogenous promoter is active at silent gene loci. An assay for identifying such promoters is detailed in the examples that follow, and thus in light of the present teachings those of skill in the art can readily identify promoters that are active at silent gene loci. In one embodiment, the exogenous promoter is a housekeeping promoter, defined herein as a promoter that promotes gene expression in all proliferating cells types. Exemplary such housekeeping promoters include, but are not limited to human elongation factor 1 a (EFl ), human phosphoglycerate kinase (PGK) promoter, and ubiquitous chromatin opening element (UCOE) promoter (which is a bidirectional promoter), or Ubiquitin C (UBC) promoter, or functional equivalents thereof, including but not limited to point mutation(s), deletions, truncations, etc. (which can be identified, for example, using the assays disclosed in the examples that follow). As shown in the examples that follow, ubiquitous promoters, exemplified by the UCOE promoter can be used to select for human stem cells that undergo silent gene editing.

Human UCOE promoter sequence (sense)

GCAGCGTCTCCGGCCCTCCGCGCCTACAGCTCAAGCCACATCCGAAGGGGGAGG GAGCCGGGAGCTGCGCGCGGGGCCGCCGGGGGGAGGGGTGGCACCGCCCACGC CGGGCGGCCACGAAGGGCGGGGCAGCGGGCGCGCGCCCGGCGGGGGGAGGGGC CGCGCGCCGCGCCCGCTGGGAATTGGGGCCCTAGGGGGAGGGCGGAGGCGCCGA CGACCGCGGCACTTACCGTTCGCGGCGTGGCGCCCGGTGGTCCCCAAGGGGAGG GAAGGGGGAGGCGGGGCGAGGACAGTGACCGGAGTCTCCTCAGCGGTGGCTTTT CTGCTTGGCAGCCTCAGCGGCTGGCGCCAAAACCGGACTCCGCCCACTTCCTCGC CCCTGCGGTGCGAGGGTGTGGAATCCTCCAGACGCTGGGGGAGGGGGAGTTGGG AGCTTAAAAACTAGTACCCCTTTGGGACCACTTTCAGCAGCGAACTCTCCTGTAC ACCAGGGGTCAGTTCCACAGACGCGGGCCAGGGGTGGGTCATTGCGGCGTGAAC AATAATTTGACTAGAAGTTGATTCGGGTGTTTCCGGAAGGGGCCGAGTCAATCCG CCGAGTTGGGGCACGGAAAACAAAAAGGGAAGGCTACTAAGATTTTTCTGGCGG GGGTTATCATTGGCGTAACTGCAGGGACCACCTCCCGGGTTGAGGGGGCTGGAT CTCCAGGCTGCGGATTAAGCCCCTCCCGTCGGCGTTAATTTCAAACTGCGCGACC GTTTCTCACCTGCCTTGCGCCAAGGCAGGGGGCGGGACCCTATTCCAAGAGGTAG TAACTAGCAGGACTCTAGCCTTCCGCAATTCATTGAGCGCATTTACGGAAGTAAC GTCGGGTACTGTCTCTGGCCGCAAGGGTGGGAGGAGTACGCATTTGGCGTAAGG TGGGGCGTAGAGCCTTCCCGCCATTGGCGGCGGATAGGGCGTTTACGCGACGGC CTGACGTAGCGGAAGACGCGTTAGTGGGGGGGAAGGTTCTAGAAAAGCGGCGGC AGCGGCTCTAGCGGCAGTAGCAGCAGCGCCGGGTCCCGTGCGGAGGTGCTCCTC GCAGAGTTGTTTCTCGAGCAGCGGCAGTTCTCACTACAGCGCCAGGACGAGTCCG GTTCGTGTTCGTCCGCGGAGATCTCTCTCATCTCGCTCGGCTGCGGGAAATCGGG CTGAAGCGACTGAGTCCGC (SEQ ID NO: 1) human UCOE promoter (antisense) GCGGACTCAGTCGCTTCAGCCCGATTTCCCGCAGCCGAGCGAGATGAGAGAGAT CTCCGCGGACGAACACGAACCGGACTCGTCCTGGCGCTGTAGTGAGAACTGCCG CTGCTCGAGAAACAACTCTGCGAGGAGCACCTCCGCACGGGACCCGGCGCTGCT GCTACTGCCGCTAGAGCCGCTGCCGCCGCTTTTCTAGAACCTTCCCCCCCACTAA CGCGTCTTCCGCTACGTCAGGCCGTCGCGTAAACGCCCTATCCGCCGCCAATGGC GGGAAGGCTCTACGCCCCACCTTACGCCAAATGCGTACTCCTCCCACCCTTGCGG CCAGAGACAGTACCCGACGTTACTTCCGTAAATGCGCTCAATGAATTGCGGAAG GCTAGAGTCCTGCTAGTTACTACCTCTTGGAATAGGGTCCCGCCCCCTGCCTTGG CGCAAGGCAGGTGAGAAACGGTCGCGCAGTTTGAAATTAACGCCGACGGGAGGG GCTTAATCCGCAGCCTGGAGATCCAGCCCCCTCAACCCGGGAGGTGGTCCCTGCA GTTACGCCAATGATAACCCCCGCCAGAAAAATCTTAGTAGCCTTCCCTTTTTGTTT TCCGTGCCCCAACTCGGCGGATTGACTCGGCCCCTTCCGGAAACACCCGAATCAA CTTCTAGTCAAATTATTGTTCACGCCGCAATGACCCACCCCTGGCCCGCGTCTGT GGAACTGACCCCTGGTGTACAGGAGAGTTCGCTGCTGAAAGTGGTCCCAAAGGG GTACTAGTTTTTAAGCTCCCAACTCCCCCTCCCCCAGCGTCTGGAGGATTCCACA CCCTCGCACCGCAGGGGCGAGGAAGTGGGCGGAGTCCGGTTTTGGCGCCAGCCG CTGAGGCTGCCAAGCAGAAAAGCCACCGCTGAGGAGACTCCGGTCACTGTCCTC GCCCCGCCTCCCCCTTCCCTCCCCTTGGGGACCACCGGGCGCCACGCCGCGAACG GTAAGTGCCGCGGTCGTCGGCGCCTCCGCCCTCCCCCTAGGGCCCCAATTCCCAG CGGGCGCGGCGCGCGGCCCCTCCCCCCGCCGGGCGCGCGCCCGCTGCCCCGCCCT TCGTGGCCGCCCGGCGTGGGCGGTGCCACCCCTCCCCCCGGCGGCCCCGCGCGCA GCTCCCGGCTCCCTCCCCCTTCGGATGTGGCTTGAGCTGTAGGCGCGGAGGGCCG GAGACGCTGC (SEQ ID NO: 2) human EFla promoter

TCGATTCTGCGATCGCTCCGGTGCCCGTCAGTGGGCAGAGCGCACATCGCCCACA GTCCCCGAGAAGTTGGGGGGAGGGGTCGGCAATTGAACCGGTGCCTAGAGAAGG TGGCGCGGGGTAAACTGGGAAAGTGATGTCGTGTACTGGCTCCGCCTTTTTCCCG AGGGTGGGGGAGAACCGTATATAAGTGCAGTAGTCGCCGTGAACGTTCTTTTTCG CAACGGGTTTGCCGCCAGAACACAGCTGAA (SEQ ID NO: 3) human PGK promoter

CGGGGTTGGGGTTGCGCCTTTTCCAAGGCAGCCCTGGGTTTGCGCAGGGACGCGG CTGCTCTGGGCGTGGTTCCGGGAAACGCAGCGGCGCCGACCCTGGGTCTCGCAC ATTCTTCACGTCCGTTCGCAGCGTCACCCGGATCTTCGCCGCTACCCTTGTGGGCC CCCCGGCGACGCTTCCTGCTCCGCCCCTAAGTCGGGAAGGTTCCTTGCGGTTCGC GGCGTGCCGGACGTGACAAACGGAAGCCGCACGTCTCACTAGTACCCTCGCAGA CGGACAGCGCCAGGGAGCAATGGCAGCGCGCCGACCGCGATGGGCTGTGGCCAA TAGCGGCTGCTCAGCAGGGCGCGCCGAGAGCAGCGGCCGGGAAGGGGCGGTGC GGGAGGCGGGGTGTGGGGCGGTAGTGTGGGCCCTGTTCCTGCCCGCGCGGTGTT CCGCATTCTGCAAGCCTCCGGAGCGCACGTCGGCAGTCGGCTCCCTCGTTGACCG AATCACCGACCTCTCTCCCCAG (SEQ ID NO: 4) mouse PGK promoter

ATTCTACCGGGTAGGGGAGGCGCTTTTCCCAAGGCAGTCTGGAGCATGCGCTTTA GCAGCCCCGCTGGCACTTGGCGCTACACAAGTGGCCTCTGGCCTCGCACACATTC CACATCCACCGGTAGCGCCAACCGGCTCCGTTCTTTGGTGGCCCCTTCGCGCCAC TTCTACTCCTCCCCTAGTCAGGAAGTTTCCCCCAGCAAGCTCGCGTCGTGCAGGA CGTGACAAATGGAAGTAGCACGTCTCACTAGTCTCGTGCAGATGGACAGCACCG CTGAGCAATGGAAGCGGGTAGGCCTTTGGGGCAGCGGCCAATAGCAGCTTTGTT CCTTCGCTTTCTGGGCTCAGAGGCTGGGAAGGGGTGGGTCCGGGGGCGGGCTCA GGGGCGGGCTCAGGGGCGGGCGGGCGCCCGAAGGTCCTCCCGAGGCCCGGCATT CTGCACGCTTCAAAAGCGCACGTCTGCCGCGCTGTTCTCCTCTTCCTCATCTCCGG GCCTTTCGACC (SEQ ID NO: 5)

] human UBC promoter

GGCCTCCGCGCCGGGTTTTGGCGCCTCCCGCGGGCGCCCCCCTCCTCACGGCGAG CGCTGCCACGTCAGACGAAGGGCGCAGCGAGCGTCCTGATCCTTCCGCCCGGAC GCTCAGGACAGCGGCCCGCTGCTCATAAGACTCGGCCTTAGAACCCCAGTATCA GCAGAAGGACATTTTAGGACGGGACTTGGGTGACTCTAGGGCACTGGTTTTCTTT CCAGAGAGCGGAACAGGCGAGGAAAAGTAGTCCCTTCTCGGCGATTCTGCGGAG GGATCTCCGTGGGGCGGTGAACGCCGATGATTATATAAGGACGCGCCGGGTGTG GCACAGCTAGTTCCGTCGCAGCCGGGATTTGGGTCGCGGTTCTTGTTTGTGGATC GCTGTGATCGTCACTTGGTGAGTAGCGGGCTGCTGGGCTGGCCGGGGCTTTCGTG GCCGCCGGGCCGCTCGGTGGGACGGAAGCGTGTGGAGAGACCGCCAAGGGCTGT AGTCTGGGTCCGCGAGCAAGGTTGCCCTGAACTGGGGGTTGGGGGGAGCGCAGC AAAATGGCGGCTGTTCCCGAGTCTTGAATGGAAGACGCTTGTGAGGCGGGCTGT GAGGTCGTTGAAACAAGGTGGGGGGCATGGTGGGCGGCAAGAACCCAAGGTCTT GAGGCCTTCGCTAATGCGGGAAAGCTCTTATTCGGGTGAGATGGGCTGGGGCAC CATCTGGGGACCCTGACGTGAAGTTTGTCACTGACTGGAGAACTCGGTTTGTCGT CTGTTGCGGGGGCGGCAGTTATGGCGGTGCCGTTGGGCAGTGCACCCGTACCTTT GGGAGCGCGCGCCCTCGTCGTGTCGTGACGTCACCCGTTCTGTTGGCTTATAATG CAGGGTGGGGCCACCTGCCGGTAGGTGTGCGGTAGGCTTTTCTCCGTCGCAGGAC GCAGGGTTCGGGCCTAGGGTAGGCTCTCCTGAATCGACAGGCGCCGGACCTCTG GTGAGGGGAGGGATAAGTGAGGCGTCAGTTTCTTTGGTCGGTTTTATGTACCTAT CTTCTTAAGTAGCTGAAGCTCCGGTTTTGAACTATGCGCTCGGGGTTGGCGAGTG TGTTTTGTGAAGTTTTTTAGGCACCTTTTGAAATGTAATCATTTGGGTCAATATGT AATTTTCAGTGTTAGACTAGTAAATTGTCCGCTAAATTCTGGCCGTTTTTGGCTTT TTTGTTAGAC (SEQ ID NO: 6)

In one exemplary embodiment discussed in detail in the examples that follow, IL2RG encodes a common subunit for several cytokine receptors expressed in hematopoietic cells, and is mutated in individuals with X-linked Severe Combined Immunodeficiency (X-SCID). IL2RG is silent in human PSCs, is present as a single copy in male cells, and represents a promising target for developing PSC-based therapies. A recombinant adeno-associate virus (rAAV) editing vector was used to insert a UCOE-Neomycin cassette (further comprising a polyA encoding region) into exon 2 of IL2RG. HI human ESCs (a male cell line) were infected with the vector and 3 of 18 G418 -resistant colonies screened by PCR were targeted at the IL2RG gene. This represented 17% of G418-resistant colonies and 0.14% of the unselected cell population, which was similar to what we observed when targeting the COL1A1 locus, confirming that the UCOE promoter could be used to select for PSC clones with edited IL2RG genes.

Many gene editing applications also require the removal of the selectable marker cassette, so that only a linked, subtle editing change remains. Thus the methods may further comprise excising the cassette or a portion thereof from the positive human stem cells. In one embodiment, the edited locus retains a negative selectable marker or inactivating mutation after excision. In one non-limiting example, the cassette may be designed so that Cre- mediated recombination removes the selection marker and promoter, and leaves behind a polyadenylation signal and three stop codons to inactivate the lineage-specification gene. As shown in the examples that follow, two different /L2 ? G-targeted clones were infected with a non-integrating foamy virus vector that transiently expressed Cre, and efficiently removed the Neo transgene cassette from 6-28% of cells. Southern blots confirmed the structures of the targeted and Cre-out alleles, as well as the lack of random integrants in targeted clones.

As will be understood by those of skill in the art, the nucleic acid cassette may include any other components as deemed suitable for an intended use. In one embodiment, the cassette further comprises a negative selectable marker under the control of the lineage- specification gene promoter and/or enhancer. In another embodiment, the cassette further comprises an inactivating mutation in the lineage-specification gene, and or its promoter and/or enhancer. In these embodiments, a positive selection marker can be used to isolate gene-edited cells, while a negative selectable marker or point mutation is used to inactivate the lineage-specification gene. In another embodiment, the cassette may further comprise a nuclease capable of generating a DNA break in the lineage-specification gene locus. The methods of the invention can be carried out in the absence of any nuclease or integrase proteins that might lead to unwanted on- or off-target mutations. However, in settings where nuclease-induced genotoxicity can be tolerated or lowered to an acceptable level, sequence- specific nucleases can be included in the nucleic acid cassette for efficient delivery of nuclease genes to the human stem cells. Such site-specific nucleases may include, but are not limited to, zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), meganucleases, and CRISPR nucleases. The nucleic acid cassette may be present on any suitable vector that can be delivered to the human stem cells. In one non-limiting embodiment, the cassette is delivered by an adenoviral vector, a recombinant adeno-associated viral (rAAV) vector, or a parvoviral vector. In the examples that follow, rAAV vectors were used to edit silent genes in human stem cells. rAAV vectors deliver single-stranded linear DNA genomes that efficiently recombine with homologous chromosomal sequences in human cells, including PSC.

In another aspect, the invention provides recombinant human stem cells, such as those that can be generated using the methods of the invention. The recombinant human stem cells can be used both therapeutically, diagnostically/prognostically, and for research purposes (including but not limited to drug screening). The human stem cell may be any suitable stem cell, including but not limited to embryonic stem cells and somatic stem cells. The stem cell may be a pluripotent stem cell (PSC) or an induced pluripotent stem cell (iPSC). In one embodiment, the human stem cells may be somatic stem cells (PSCs or iPSCs) including, but not limited to, hematopoietic stem cells, mesenchymal stem cells, intestinal stem cells, endothelial stem cells, neural stem cells, neural crest stem cells, olfactory adult stem cells, and mammary stem cells. In another embodiment, the human stem cells are embryonic stem cells.

In one embodiment, the human stem cells are knocked-out for a lineage-specification gene. In one non- limiting example, a negative selection marker is inserted into the lineage- specification gene locus, under the control of the lineage-specification gene promoter and/or enhancer, as described above. In various non-limiting embodiments, the negative selection marker is selected from the group consisting of thymidine kinase, an apoptosis inducer, and a toxic gene (i.e., a gene whose expressed product is toxic to the cell, including but not limited to the diphtheria toxin gene).

In another embodiment, a negative selection marker can be used that is not under control of the lineage-specification gene promoter and/or enhancer, including but not limited to thymidine kinase, an apoptosis inducer, and a toxic gene.

In another embodiment, the human stem cells comprise a selection marker and exogenous promoter inserted into the lineage-specification gene, wherein the selection marker inactivates the lineage specification gene. As will be understood by those of skill in the art, the selectable marker does not need to be expressed to inhibit the lineage-specific gene. Any suitable selection marker can be used, including but not limited to an antibiotic resistance gene, a surface marker that can be used for cell purification, a metabolic gene that confers survival in the presence of a specific medium formulation, and/or a gene that provides a growth advantage. In one embodiment, the selection marker is an antibiotic resistance gene.

In one embodiment of any of the above embodiments, the linage specification gene locus is silent in the human stem cell. In various embodiments, transcripts produced from the linage specification gene are present at a level in the lowest 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, or 0.1 % of all genes in the human stem cell. In another embodiment, no transcripts are produced from the lineage specification gene in the human stem cell. In a further embodiment, the exogenous promoter is active at silent gene loci, as described herein. In one embodiment, the exogenous promoter is a housekeeping promoter, defined herein as a promoter that promotes gene expression in all proliferating cells types. Exemplary such housekeeping promoters include, but are not limited to human elongation factor 1 a (EFl ), human phosphoglycerate kinase (PGK) promoter, and ubiquitous chromatin opening element (UCOE) promoter, and Ubiquitin C (UBC) promoter.

In various non-limiting embodiments, the lineage-specification genes may be selected from the group consisting of genes encoding interleukin-2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage-specific cell surface marker, a lineage- specific transcription factor, a cytokine or hormone receptor.

The recombinant human stem cells can be used therapeutically, diagnostically, and for research purposes (including but not limited to drug screening). In one embodiment, the recombinant human stem cells of the invention can be transplanted into a subject (such as a human) in need thereof. In one exemplary embodiment, the IL2RG gene is knocked out in human PSCs, which limits the PSCs capability to differentiate into NK or T cells, which can be used to prevent graft versus host disease (GVHD) by transplanting the resulting human PSCs into a subject in need thereof, and thus substantially reducing/eliminating PSC-derived T cells that may react against HLA-mismatched host cells.

Edited PSCs can be used as cellular disease models to study the function of lineage- restricted genes. As shown in the examples that follow, H2i?G-knockout PSCs were unable to differentiate into NK or T cells, confirming a central role for IL2RG-dependent signaling in the developing immune system, and demonstrating that the lack of NK and T cells observed in X-SCID patients is due to a differentiation block at the Pro T and Pro NK stage of lymphopoiesis.

IL2RG knockouts could also be used to prevent graft versus host disease (GVHD) when transplanting PSCs for therapeutic purposes, by eliminating PSC-derived T cells that may react against HLA-mismatched host cells. This would be especially valuable when using allogeneic PSC-derived cells to produce non-lymphoid hematopoietic cell types, such as macrophages and neutrophils to treat chronic granulomatous disease, or erythrocytes to treat hemoglobinopathies. In these settings, transplanted, PSC-derived hematopoietic

stem/progenitor cells capable of long-term engraftment would continuously produce terminally differentiated therapeutic cell types in the absence of host-reactive allogeneic lymphoid cells. /L2i?G-knockout cells may even be an advantage when transplanting autologous cells, because PSC-derived T and NK cells do not develop in a normal embryo and may not be educated appropriately to tolerate autologous host cells.

In another embodiment, human stem cells with a knockout of a gene required to produce pacemaker cells (or a knockin of a negative selection marker into such a gene) are used to prepare cardiomyocyte cultures that do not cause arrhythmias after transplantation.

In another embodiment, human PSCs with negative selection markers knocked into the Glucagon and/or Somatostatin genes are used to differentiate stem cells into pancreatic islet β cells that secrete insulin, and which can be administered to a subject with diabetes.

Examples

ABSTRACT

Many applications of pluripotent stem cells (PSCs) require efficient editing of silent chromosomal genes. Here we show that a major limitation in isolating edited clones is silencing of the selectable marker cassette after homologous recombination, and that this can be overcome by using a UCOE promoter-driven transgene. We use this strategy to edit the silent IL2RG locus in human PSCs with a recombinant adeno-associated virus (rAAV) targeting vector in the absence of potentially genotoxic, site-specific nucleases, and show that IL2RG is required for NK and T cell differentiation of human PSCs. Insertion of an active UCOE promoter into a silent locus altered the histone modification and cytosine methylation pattern of surrounding chromatin, but these changes resolved when the UCOE promoter was removed. This same approach could be used to correct IL2RG mutations in X-SCID patient- derived induced PSCs, to prevent graft versus host disease in regenerative medicine applications, or to edit other silent genes. INTRODUCTION

Many applications require that silent genes be edited. This is especially true for pluripotent stem cells (PSCs), which may not express the tissue-specific genes responsible for diseases. Transcription has long been known to increase homologous recombination ¹ and gene targeting by transfection-based methods ². One way to overcome this limitation is the introduction of sequence-specific double strand breaks (DSBs) by engineered nucleases in order to enhance silent gene targeting. However, the use of engineered nucleases may also lead to a variety of genotoxic effects, including unwanted sequence changes at on- and off-target sites ⁶' ⁸, which could be a disadvantage in some settings.

Alternatively, the lower targeting frequencies observed at silent loci could be due to inadequate expression of the selectable marker gene after it integrates, rather than a decrease in homologous recombination. The epigenetic changes that presumably occur during silent gene editing are poorly understood, which include potential alterations induced by the recombination and repair enzymes acting on the locus, the effects of introducing an expressed selectable marker into silent chromatin, and in many cases the subsequent removal of that same expressed marker after isolating an edited clone. In general, the epigenetic consequences of gene editing remain an important but largely unexplored area of research. The epigenetic effects of gene editing in human cells have not yet been described.

In this study, we use recombinant adeno-associated virus (rAAV) vectors to edit silent genes in human PSCs. Here we evaluate different selectable marker cassettes to develop a robust, silent gene editing method for human PSCs that does not require a site-specific nuclease, we examine the epigenetic consequences of targeting silent loci, and we determine the developmental effects of IL2RG gene editing. RESULTS

Transgene promoter type determines targeted clone survival. In order to optimize vector designs, we developed an assay to detect gene editing events at a non-transcribed locus, in which only gene-targeted cells survive selection (Fig. la). The assay uses induced PSCs containing a silenced Neo gene that can be activated by upstream promoter insertion. We first infected human mesenchymal stem/stromal cells (MSCs) with a rAAV knock-in vector designed to insert a Neo gene at the endogenous COLIAI locus encoding Type I collagen, which is highly expressed in MSCs. A polyclonal population of G418 -resistant MSCs was then converted to induced pluripotent stem cells (iPSCs) by expressing OCT4, SOX2, NANOG, and LIN28 transgenes ²⁰. Three of these iPSC clones were analyzed further, and clone 1 had the lowest level of COLIAI expression after reprogramming (Fig. lb). Southern blot analysis showed that this clone also had a duplication of the Neo transgene (Fig. 5c), which happens in a small percentage of targeted clones when vector genomes form dimers before recombination ¹⁶. Although this complicated our analysis, we confirmed that clone 1 was completely sensitive to G418 (Fig. 5a), so both Neo transgenes had been silenced and could therefore be activated by promoter insertion. The residual COLIAI transcription detected in clone 1 cells may have been derived from the subpopulation of differentiating cells present in PSC cultures, which do not contribute to the PSC clones isolated by selection.

A series of gene editing vectors were designed to insert different promoters upstream of either silenced Neo transgene cassette, each of which contained a truncated Neo gene fragment in the right homology arm so that random integration could not confer G418-resistance, and only gene-edited clones would survive selection (Fig. lc). Two types of promoters were incorporated into the rAAV gene editing vectors: developmentally regulated promoters that are expressed in human PSCs (REX1, murine Sox2, and EPCAM), and ubiquitously expressed promoters (EFla, murine Pgk, human PGK, and UCOE). When clone 1 PSCs with silenced Neo genes were infected with each of these vectors at the same multiplicity of infection, there were dramatic differences in the number of G418 -resistant colonies obtained (Fig. Id), with the UCOE promoter producing the highest number. Southern blots showed that the G418-resistant colonies had been targeted at one of the two silent Neo gene targets (examples in Fig. 5c).

For a subset of vectors, gene editing frequencies were also measured in unselected cells directly by qPCR, using one primer within the inserted promoter and one in chromosomal DNA outside of the homology arm (Figs, le and 5b). This showed that despite a more than 100-fold difference in their ability to produce G418 -resistant colonies, the REX1, EPCAM and UCOE promoter vectors all produced homologous recombinants at similar frequencies. Thus rAAV vectors can edit a silent COLIAI gene in human PSCs regardless of the transgene promoter they contain, but transgene selection requires a promoter that can drive expression at a silent locus. In the case of the UCOE promoter, the G418 -resistance and homologous recombination frequencies were very similar, suggesting that almost every recombination event produced a G418 -resistant cell. One potential drawback of the UCOE promoter is its relatively large size. Unfortunately, smaller promoter fragments did not produce as many G418 -resistant colonies (Fig. 5d). These experiments demonstrate that the 1.2 kb UCOE promoter can be used to select for PSCs that undergo silent gene editing. Editing of the silent IL2RG gene. IL2RG encodes a common subunit for several cytokine receptors expressed in hematopoietic cells, and is mutated in individuals with X-linked Severe Combined Immunodeficiency (X-SCID). IL2RG is not expressed in human PSCs, is present as a single copy in male cells, and represents a promising target for developing PSC- based therapies. We therefore tested if a rAAV editing vector could be used to insert a UCOE- Neo-pA cassette into exon 2 of IL2RG (Fig. 2a). HI human ESCs (a male cell line) were infected with vector AAV-IL2RGe2UNA and 3 of 18 G418 -resistant colonies screened by PCR were targeted at the IL2RG gene (Fig. 6a). This represented 17% of G418-resistant colonies and 0.14% of the unselected cell population, which was similar to what we observed when targeting the COL1A1 locus, confirming that the UCOE promoter could be used to select for PSC clones with edited IL2RG genes.

Many gene editing applications also require the removal of the selectable marker cassette, so that only a linked, subtle editing change remains. Vector AAV-IL2RGe2UNA was designed so that Cre-mediated recombination would remove the floxed Neo cassette and leave behind a polyadenylation signal and three stop codons to inactivate IL2RG. We infected two different H2i?G-targeted clones with a non-integrating foamy virus vector that transiently expressed Cre ²¹ , and efficiently removed the Neo transgene cassette from 6-28% of cells (Fig. 6b). Southern blots confirmed the structures of the targeted and Cre-out alleles, as well as the lack of random integrants in targeted clones (Fig. 2b).

Ideally, the editing of a silent locus would not impact the expression of other genes.

IL2RG gene targeting and subsequent Cre-out had no measurable impact on neighboring gene expression, as shown for the four genes spanning a 110 kb window surrounding IL2RG (Fig. 2c). One of these genes (SNX12) had increased expression in targeted cells that still contained the UCOE-Neo cassette, reflecting long distance effects of the UCOE promoter. Interestingly, expression of the FOX04 gene located between IL2RG and SNX12 was unchanged in cells containing UCOE-Neo, demonstrating that the closest neighboring gene may not be the gene most affected by promoter insertion. Epigenetic consequences of gene editing. Insertion of an active promoter could change the epigenetic status of the surrounding chromatin, as could the recombination and repair proteins that carry out homologous recombination, yet the epigenetic consequences of gene editing remain a largely unexplored area of research. One possible effect is removal of 5-methylcytosine (5mC) residues in DNA, in particular at CpG islands, which are typically methylated in silent loci ²². The wild-type COL1A1 gene contains a CpG island that was duplicated in the PSC clone we used for promoter insertion studies (Fig. 3a). When the locus is silent, these CpG islands are highly methylated, but after UCOE promoter insertion at either duplicated site, one of the islands became mostly unmethylated (Fig. 3b). Although the bisulfite sequencing reaction does not distinguish between the two islands, these results are consistent with a localized region of hypomethylation at the CpG island nearest to the UCOE insertion. Since this region is included within the 5' homology arm of the UCOE insertion vector, incorporation of the unmethylated vector genome could have led directly to the loss of 5mC residues. In support of this hypothesis, we confirmed that the packaged rAAV vector genome was unmethylated (Fig. 3b), and we showed previously that the entire homology arm sequence extending to the terminal repeats are typically incorporated into a targeted locus ²³. Unfortunately, a similar analysis could not be performed at the IL2RG locus, which does not contain a CpG island within the homology region. We found no consistent difference in the methylation status of eight non-island CpGs present in the IL2RG gene that were assayed in wild-type cells and knockout cells, suggesting that UCOE promoter insertion may not lead to demethylation of non-island CpGs (Fig. 7).

Histone modifications can also vary depending on the transcriptional activity of a locus and other factors. Although many such modifications have been described, here we studied acetylation and methylation at lysine 27 of histone H3 (H3K27Ac and H3K27Me3), which are

24 25 26

associated with active and silent chromatin respectively ' ' . In the case of COL1A1, the 1 kb region immediately surrounding the UCOE insertion site that includes the IRES element and Neo gene contained H3K27Me3 markers before UCOE insertion, indicative of silent chromatin (Fig. 3 c). After UCOE insertion, this pattern changed to one of active chromatin, with an increase in H3K27Ac levels that extended throughout the duplicated locus to sites 2.8 and 8.4 kb distal to the insertion site, as well as lower levels of H3K27Me3 in the more localized IRES and Neo regions. Interestingly, the change in H3K27Ac markers extended to regions outside of the homology arms, demonstrating that epigenetic changes had been propagated beyond the recombination site.

We studied the same histone markers at the IL2RG locus, only in this case the analysis was more relevant because the target was single copy, the UCOE promoter was inserted at a wild-type gene that had not been previously targeted, and we could assay after both UCOE insertion and subsequent Cre-mediated UCOE removal. The wild-type locus had low levels of H3K27Ac and higher levels of H3K27Me3 throughout a 3.3 kb region surrounding the exon 2 insertion site, consistent with silent chromatin (Fig. 3d). Insertion of the UCOE-Neo cassette activated this entire locus, including regions beyond the vector homology arms, as evidenced by increased H3K27Ac levels. However, unlike the COL1A1 locus, we did not observe a corresponding reduction in H3K27Me3 throughout this region. Instead, the decrease in

H3K27Me3 was only observed downstream of the UCOE-Neo cassette. The basis for this asymmetry is unclear, and was not shared by COL1A1, which had decreased H3K27Me3 levels on both sides of the UCOE insertion site. Once the UCOE-Neo cassette was removed by Cre- mediated recombination, the epigenetic status of the entire locus reverted back to that of the wild- type locus. /L2RG-knockout human ESCs do not produce NK or T cells. While IL2RG mutation correction could be used to treat X-SCID, inactivation of IL2RG by gene editing could also have applications in regenerative medicine. For example, when differentiating pluripotent cells into hematopoietic progeny to be used for transplantation, it may be desirable to prevent the formation of NK or T cells that could react against host cells in a form of graft vs host disease (GVHD). Based on the phenotype of X-SCID patients ²¹ ' ²⁸, and the known roles of the cytokine receptors that include IL2RG subunits ²⁹ , we predicted that IL2RG knockout stem cells should not produce NK or T cells.

We first differentiated wild-type and /L2i?G-edited ESC lines into embryoid bodies (EBs) containing hematopoietic progenitors, which were then selectively differentiated into NK or T cells as described ³⁰' ³¹. After eight days of NK differentiation, all the ESC lines produced similar numbers of cells expressing the CD56 NK marker, including CD2+ and CD7+ early lymphoid subsets (Fig. 4a). In comparison, NK differentiation of CD34+ cells isolated from umbilical cord blood (UCB) cells produced fewer CD56+ cells and lower expression levels of CD7. However, after 22 days of differentiation, the UCB and wild-type EB cultures both contained nearly 80% CD56+ NK cells with substantial CD2+ and CD7+ subpopulations that were largely absent from both /L2i?G-edited cultures (targeted and Cre-out) (Fig. 4b). The wild-type and UCB cultures also expressed IL2RG (CD132), and supported the expansion of CD14-, CD56+ NK cells that could lyse MHC class I-negative target cells as expected, while the gene-edited cells did not (Figs. 4a, b), confirming that both the targeted and Cre-out alleles were functional knockouts. Importantly, the IL2RG-Qdited lines were still able to produce CD14+ and CD15+ progeny (Fig. 8c), demonstrating that IL2RG is not necessary for monocyte and granulocyte differentiation. When EBs were cultured under T cell differentiation conditions, the /L2i?G-knockout cultures produced slightly fewer CD34+, CD7+ progenitors than wild-type cells at day 10, and these progenitors did not mature further and down-regulate CD34 at later time points (Fig. 4c). The H2i?G-knockout cells were capable of producing CD5+, CD7+ pro-T cell progeny (Fig. 8d) but failed to produce more mature CD4+, CD8+, and double -positive (CD4⁺CD8⁺) T cells (Fig. 4d). Both /L2 ? G-edited lines generated normal numbers of primitive and definitive erythroid/myeloid progenitors indicating that the development of these lineages was not affected by the deletion of the IL2RG gene (data not shown). These combined data suggest that IL2RG expression is required for progression from a CD7+ Pro-NK cell or Pro-T cell into more mature cell types (Fig. 4e). DISCUSSION

Here we have described a robust method for editing silent loci in human PSCs without employing a nuclease. Our approach involves the use of a UCOE promoter-driven selectable marker gene that resists silencing, rAAV vectors that efficiently deliver the targeting

construct to PSCs, and subsequent selectable marker removal if desired. With this approach, 0.1-1% of the entire cell population undergoes gene editing, and these clones can be isolated by antibiotic selection. The Neo selection cassette used for IL2RG targeting could also function after random integration, and 1/6 of G418-selected clones were accurately edited in those experiments, which is comparable to results obtained when using rAAV vectors to edit active human genes ³². Cre-mediated recombination can be used to efficiently excise the transgene and produce a minimally altered locus that reverts to silent chromatin.

The raw (unselected) editing frequencies we obtained were similar to what has been reported for nuclease-based editing of silent genes in PSCs. For example, both our approach (Fig. 6a) and TALEN-based targeting of IL2RG ⁶ led to gene editing in 0.14% of unselected PSCs. After antibiotic selection, 0.3-60% of PSC clones were edited by ZFN- or TALEN- based targeting of silent PITX3 or HBB genes ³' ⁵, which demonstrates the variability

observed in these types of experiments, but still encompasses the 17% editing frequency we observed in G418-selected clones. An important advantage of our approach is that the rAAV vector does not include any nuclease or integrase proteins that might lead to unwanted on or off-target mutations ⁶' ⁸. And while rAAV can integrate randomly at spontaneously occurring chromosomal DSBs, infection with rAAV does not increase background mutation rates in cellular genes ³³. Random rAAV integrants are rarely found in edited PSC clones and can be easily ruled out by PCR or Southern blots for vector sequences ¹³. In contrast, the small in- del mutations produced by non-homologous end joining at off-target, nuclease-induced DSBs can only be identified in an unbiased manner by full genome sequencing. This reduced genotoxicity of rAAV -mediated gene editing may be an advantage when preparing cells for clinical applications.

In settings where nuclease-induced genotoxicity can be tolerated or lowered to an acceptable level, our findings may also lead to further improvements in these gene editing methods. Prior studies of silent gene editing in PSCs with ZFNs or TALENs used a PGK promoter to express the selectable marker ³' ⁴' ⁵, and our results suggest that using the UCOE promoter instead would have increased the number of edited clones that survived selection ~4-fold (Fig. Id). Sequence-specific nucleases could also be combined with rAAV vectors for efficient delivery of both nuclease genes and UCOE-based targeting constructs to PSCs. Target site DSBs can increase rAAV-mediated gene editing significantly ³⁴' ³⁵, and rAAV- encoded ZFNs have been combined with rAAV targeting vectors for efficient in vivo gene editing ³⁶ , demonstrating the potential of this approach. Finally, the recently developed Clustered, Regularly Interspaced Short Palindromic Repeat (CRISPR) systems that employ guide RNAs to induce sequence-specific DSBs could also be combined with rAAV and provide further enhancements in silent gene editing, given the promising results obtained so

37 38 39

far in CRIPSR-based editing of expressed genes in human PSCs ' ' .

Our study begins to describe the epigenetic changes that can occur at an edited locus. We found that insertion of a UCOE promoter into silent chromatin can lead to a loss of CpG island 5mC residues and convert histone modifications to a more active signature, but the details of these changes can be complex. For example, UCOE insertion increased H3K27Ac levels throughout the transcribed Neo cassette and into both upstream and downstream regions, but only reduced H3K27Me3 levels over a more localized region in the case of COLIAI, and only in downstream sequences in the case of IL2RG. The gene editing process itself could have played a role in some of these changes, for example by incorporating unmethylated vector DNA into the chromosome. Or alternatively, UCOE-dependent transcription could have indirectly altered the epigenetic signature, which may explain why some changes in histone modifications extended beyond the region of vector homology. Removal of the UCOE-Neo cassette caused the edited locus to return to an inactive epigenetic signature indistinguishable from the unedited, parental locus, based on the limited analysis we performed. However, a more detailed examination of the many other histone

modifications that have been described ²⁶ , as well as the identification of DNA-binding proteins ⁴⁰' ⁴¹ , DNase hypersensitive sites ⁴² and long-range chromatin interactions ⁴³' ⁴⁴ found at the locus would presumably reveal additional epigenetic changes associated with gene editing, and it remains to be seen if all these changes convert back to a wild-type signature after removing the UCOE-Neo cassette.

Our choice of the IL2RG gene illustrates some of the potential applications of silent gene editing in PSCs. Edited PSCs can be used as cellular disease models to study the function of lineage-restricted genes. H2i?G-knockout PSCs were unable to differentiate into NK or T cells, confirming a central role for IL2RG-dependent signaling in the developing immune system ²⁹ , and demonstrating that the lack of NK and T cells observed in X-SCID patients ²¹ ' ²⁸ is due to a differentiation block at the Pro T and Pro NK stage of lymphopoiesis. A similar rAAV editing strategy could be used to correct the IL2RG point mutations that typically cause X-SCID ⁴⁵ so that patient-derived, gene-edited induced PSCs could in principle be differentiated ex vivo into hematopoietic cells and transplanted into autologous recipients. In vivo selection should enrich for edited cells, and only a few cells would be required to correct the disease based on the mild phenotype of patients with spontaneous reversion mutations ⁴⁶' ⁴⁷ and the success of IL2RG gene therapy ⁴⁸. TALEN-mediated gene editing was recently used to correct an IL2RG mutation in X-SCID iPSCs ¹.

IL2RG knockouts could also be used to prevent graft versus host disease (GVHD) when transplanting PSCs for therapeutic purposes, by eliminating PSC-derived T cells that may react against HLA-mismatched host cells. This would be especially valuable when using allogeneic PSC-derived cells to produce non-lymphoid hematopoietic cell types, such as macrophages and neutrophils to treat chronic granulomatous disease, or erythrocytes to treat hemoglobinopathies. In these settings, transplanted, PSC-derived hematopoietic

stem/progenitor cells capable of long-term engraftment would continuously produce terminally differentiated therapeutic cell types in the absence of host-reactive allogeneic lymphoid cells. /L2i?G-knockout cells may even be an advantage when transplanting autologous cells, because PSC-derived T and NK cells do not develop in a normal embryo and may not be educated appropriately to tolerate autologous host cells. Other scenarios can also be envisioned where preventing the expression of a lineage-specification gene could produce a therapeutic advantage, such as PSCs with edited glucagon or somatostatin genes that can differentiate into insulin-secreting beta cells for the treatment of diabetes without producing the alpha or delta cells that frequently contaminate PSC-derived pancreatic islet

· 49 50

cell preparations ' . MATERIALS AND METHODS

Cell culture. HI human ESCs ⁵¹ and human iPSC lines were cultured on mouse embryo fibroblasts as described ²⁰' ⁵². COL1A1 -targeted G418-sensitive iPSCs were derived by reprogramming of (COZJ^i-IRES-Neo)-targeted MSCs with lentiviral vectors as described ⁵². 50 μg/mL active G418 was used for selection.

Viral vectors. AAV vector plasmids were assembled from PCR products by standard methods and confirmed by DNA sequencing. Homology arm fragments were amplified from the target cell type, and promoter fragments were amplified from HI human ESCs and CF1 mice respectively. Plasmid sequences are available upon request. AAV vectors were packaged in serotype 3b capsids by co-transfection of vector plasmids and packaging plasmid pDGM3B into 293T cells, purified by iodixanol step gradients, and their titers were determined by Southern blots as described ⁵³.

Gene targeting. In promoter comparison experiments, 4xl0⁵ iPSCs were seeded in triplicate in 35 mm wells and transduced with rAAVs at a multiplicity of infection of 10⁴ genome- containing particles/cell the next day. G418 selection was initiated 2 days after infection and the surviving colonies were counted, picked and expanded for further analysis. The total number of colony- forming units (CFU) was calculated by culturing 4xl0³ cells in a 6 cm dish without selection. In IL2RG targeting experiments, 5xl0⁵ wild-type HI ESCs were seeded in a 10 cm dish and transduced with AAV at a multiplicity of infection of 3xl0³ genome- containing particles/cell the next day. Four days later, transduced ESCs were disaggregated into single cells using Accutase® (Stemgent, Cambridge, MA), and plated in serial dilutions in 10 cm dishes for G418 selection. 5 xlO³ transduced HI ESCs were also plated in a 10 cm dish without selection to determine the total number of CFUs. G418 -resistant colonies were counted, picked and screened initially by PCR to identify targeted clones.

Cre-mediated transgene removal. A polyclonal population of wild-type and IL2RG- targeted HI ESCs was transduced as described with the non-integrating foamy vector, NIFV- EokCreW that expresses Cre recombinase ⁵². Four days later, infected ESCs were

disaggregated into single cells with Accutase® and serial dilutions were plated in 10 cm dishes. The surviving colonies were randomly picked and screened by PCR to identify age- matched clones with wild-type, UCOE-Neo-containing, or Cre-out IL2RG alleles for subsequent experiments.

DNA and RNA isolation. Genomic DNA was prepared from PSCs as described ⁵³' ⁵⁴. Total cellular RNA was extracted by the Trizol method (Life Technologies, Grand Island, NY) and used to generate cDNA with M-MLV reverse transcriptase and oligo-dT primers according to the manufacturer's protocol (Life Technologies).

Quantitative PCR and RT-PCR. cDNA qRT-PCR reactions were performed in triplicate with SYBR® Select Master Mix (Life Technologies) on a StepOnePlus® Real-Time PCR System (Life Technologies) and the relative gene expression levels were calculated by the delta-delta CT method. Homologous recombination frequencies were measured by infecting iPSCs with rAAV vectors, culturing for 5 days without selection, and determining the number of promoter-targeted alleles in 1 μg of genomic DNA by Taqman® qPCR (Life Technologies). Plasmids containing promoter-targeted COLIAI sequences were constructed by conventional cloning methods and used in qPCR reactions containing 0 to 10⁴ plasmid molecules and 1 μg of wild-type genomic DNA (1.5xl0⁵ diploid genome equivalents) to generate standard curves. Bisulfite sequencing. Genomic DNA was treated as described in the EZ® DNA

Methylation-Gold Kit (ZYMO Research, Irvine, CA) and used to PCR-amplify COLIAI CpG island fragments. These PCR products were cloned into the pGEM-Teasy® vector (Promega, Madison, WI) and the recombinant plasmids were sequenced. Chromatin immunoprecipitation (ChIP). The ChIP protocol was adapted from Abeam' s 'X-ChIP® protocol' as follows. PSCs were dissociated with Accutase® and fixed with PBS containing 1% formaldehyde at room temperature for 10 minutes with constant mixing. Fixation was stopped by adding glycine to a final concentration of 125 mM. Samples were sonicated until chromatin was sheared to 500 to 1000 bp. For binding, chromatin from 10⁷ cells, 10 μg antibodies, and 30 μΐ, solid protein- A/G beads (Santa Cruz Biotechnology, Dallas, Texas) were combined and incubated at 4°C overnight. After washing the beads, chromatin was eluted and incubated at 65°C overnight to reverse cross-linking. The eluted chromatin was purified using a PCR purification kit (Qiagen, Valencia, CA) and used in qPCR reactions performed in triplicate with SYBR® Select Master Mix (Life Technologies) on a StepOnePlus® Real-Time PCR System (Life Technologies). Relative occupancy was calculated by the delta-delta CT method. Antibodies used were against H3K27me3 (EMD Millipore, Temecula, CA), H3K27Ac (Abeam, Cambridge, MA), and normal rabbit IgG (Santa Cruz Biotechnology) as a control.

NK cell differentiation. NK differentiation was performed as described ³⁰' ⁵⁵. Briefly, day 13 EBs were co-cultured with EL08-1D2 stromal cells in media supplemented with IL-3, IL7, IL-15, SCF, and FLT3L (Peprotech, Rocky Hill, NJ) and cells were harvested at appropriate time points for analysis. In chromium release assays, derived NK cells were stimulated by Clone 9.mbIL-21 aAPCs as described ⁵⁶ , and incubated with radioactive 51Cr-labeled K562 target cells, and lysis was measured by scintillation counter using the equation: % specific lysis = 100 x (test release - spontaneous release) / (maximal release - spontaneous release) as described ⁵⁷. Flow cytometry was performed with a BD LSRII (BD Biosciences, San Jose, CA) flow cytometer and the data was analyzed by Flow Jo® software version 10.0 (Tree Star). Antibodies, which were used according to the manufacturers' recommendations, were from BD Biosciences unless otherwise indicated. Live cells were distinguished from dead cells by CYTOX® blue dead cell stain (Life Technologies). Antibodies used for NK cell phenotype analysis were: CD56 (PE-Cy7-clone B159); CD7 (Alexa Fluor 700-clone M- T701); CD2 (PE-CF594-clone RPA-2.10); CD132 (PE-clone TUGh4, eBioscience, San Diego, CA).

T cell differentiation. T cell differentiation and analysis were performed as described previously ³¹. Briefly, at day 8 of EB differentiation, 2xl0⁴ CD34+ CD43- CD73- CXCR4- cells isolated by FACS were plated onto individual wells of a 6-well plate containing OP9- DL4 stromal cells in the presence of rhFLT3L and rhIL-7. rhSCF was added for the first 7 days only (R&D Systems). Every 5 days, co-cultures were passaged onto fresh OP9-DL4 stromal cells. Cells were harvested and assayed at various time points. Cell suspensions were stained and analyzed on a BD LSR II flow cytometer. Data analysis was performed using Flow Jo Software by gating on live cells followed by lack of DAPI uptake. Fluorophore- conjugated antibodies against CD4, CD5, CD7, CD8, CD34 and CD45 were purchased from BD Biosciences and eBioscience. Statistical analysis. Statistical significance was assessed using the two-tailed Student's t- test. P values less than 0.05 were considered statistically significant.

REFERENCES

1. Nickoloff JA, Reynolds RJ. Transcription stimulates homologous recombination in mammalian cells. Mol Cell Biol 10, 4837-4845 (1990).

2. Thyagarajan B, Johnson BL, Campbell C. The effect of target site transcription on gene targeting in human cells in vitro. Nucleic Acids Res 23, 2784-2790 (1995).

3. Hockemeyer D, et al. Efficient targeting of expressed and silent genes in human ESCs and iPSCs using zinc-finger nucleases. Nat Biotechnol 27, 851-857 (2009).

4. Zou J, Mali P, Huang X, Dowey SN, Cheng L. Site-specific gene correction of a point mutation in human iPS cells derived from an adult patient with sickle cell disease.

Blood 118, 4599-4608 (2011).

5. Sun N, Zhao H. Seamless correction of the sickle cell disease mutation of the HBB gene in human induced pluripotent stem cells using TALENs. Biotechnol Bioeng 111,

1048-1053 (2014).

6. Hendel A, et al. Quantifying Genome-Editing Outcomes at Endogenous Loci with SMRT Sequencing. Cell Rep 7, 293-305 (2014).

7. Menon T, et al. Lymphoid Regeneration from Gene-Corrected SCID-X1 Subject- Derived iPSCs. Cell Stem Cell, (2015).

8. Fu Y, et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol 31, 822-826 (2013).

9. Ripoche MA, Kress C, Poirier F, Dandolo L. Deletion of the H19 transcription unit reveals the existence of a putative imprinting control element. Genes Dev 11, 1596-

1604 (1997).

10. Tsai TF, Bressler J, Jiang YH, Beaudet AL. Disruption of the genomic imprint in trans with homologous recombination at Snrpn in ES cells. Genesis 37, 151-161 (2003).

11. Lieberman-Lazarovich M, Melamed-Bessudo C, de Pater S, Levy AA. Epigenetic alterations at genomic loci modified by gene targeting in Arabidopsis thaliana. PLoS ONE S, e85383 (2013). 12. Russell DW, Hirata R . Human gene targeting by viral vectors. Nat Genet 18, 325- 330 (1998).

13. Khan IF, et al. Engineering of human pluripotent stem cells by AAV-mediated gene targeting. Mol Ther 18, 1192-1199 (2010).

14. Li LB, Chang KH, Wang PR, Hirata RK, Papayannopoulou T, Russell DW. Trisomy correction in Down syndrome induced pluripotent stem cells. Cell Stem Cell 11, 615- 619 (2012).

15. Mitsui K, et al. Gene targeting in human pluripotent stem cells with adeno-associated virus vectors. Biochem Biophys Res Commun 388, 711-717 (2009).

16. Chamberlain JR, et al. Gene targeting in stem cells from individuals with osteogenesis imperfecta. Science 303, 1198-1201 (2004).

17. Wang PR, Xu M, Toffanin S, Li Y, Llovet JM, Russell DW. Induction of

hepatocellular carcinoma by in vivo gene targeting. Proc Natl Acad Sci USA 109,

11264-11269 (2012).

18. Rogers CS, et al. Production of CFTR-null and CFTR-DeltaF508 heterozygous pigs by adeno-associated virus-mediated gene targeting and somatic cell nuclear transfer. J

Clin Invest 118, 1571-1577 (2008).

19. Sun X, et al. Adeno-associated virus-targeted disruption of the CFTR gene in cloned ferrets. J Clin Invest 118, 1578-1583 (2008).

20. Yu X et al. Induced pluripotent stem cell lines derived from human somatic cells.

Science 318, 1917-1920 (2007).

21. Deyle DR, Li Y, Olson EM, Russell DW. Nonintegrating foamy virus vectors. J Virol

84, 9341-9349 (2010).

22. Deaton AM, Bird A. CpG islands and the regulation of transcription. Genes Dev 25,

1010-1022 (2011).

23. Deyle DR, Li LB, Ren G, Russell DW. The effects of polymorphisms on human gene targeting. Nucleic Acids Res 42, 3119-3124 (2014).

24. Creyghton MP, et al. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci USA 107, 21931-21936 (2010). 25. Rada-Iglesias A, Bajpai R, Swigut T, Brugmann SA, Flynn RA, Wysocka J. A unique chromatin signature uncovers early developmental enhancers in humans. Nature 470, 279-283 (2011). Ernst J, et al. Mapping and analysis of chromatin state dynamics in nine human cell types. Nature 473, 43-49 (2011).

Noguchi M, et al. Interleukin-2 receptor gamma chain mutation results in X-linked severe combined immunodeficiency in humans. Cell 73, 147-157 (1993).

Buckley RH, et al. Human severe combined immunodeficiency: genetic, phenotypic, and functional diversity in one hundred eight infants. J Pediatr 130, 378-387 (1997). Leonard WJ. Cytokines and immunodeficiency diseases. Nature reviews 1, 200-208

(2001) .

Knorr DA, et al. Clinical-scale derivation of natural killer cells from human pluripotent stem cells for cancer therapy. Stem cells translational medicine 2, 274-283 (2013).

Kennedy M, et al. T Lymphocyte Potential Marks the Emergence of Definitive Hematopoietic Progenitors in Human Pluripotent Stem Cell Differentiation Cultures. Cell Rep, (2012).

Hirata R, Chamberlain J, Dong R, Russell DW. Targeted transgene insertion into human chromosomes by adeno-associated virus vectors. Nat Biotechnol 20, 735-738

(2002) .

Miller DG, Petek LM, Russell DW. Adeno-associated virus vectors integrate at chromosome breakage sites. Nat Genet 36, 767-773 (2004).

Porteus MH, Cathomen T, Weitzman MD, Baltimore D. Efficient gene targeting mediated by adeno-associated virus and DNA double-strand breaks. Mol Cell Biol 23, 3558-3565 (2003).

Miller DG, Petek LM, Russell DW. Human gene targeting by adeno-associated virus vectors is enhanced by DNA double-strand breaks. Mol Cell Biol 23, 3550-3557

(2003) .

Li H, et al. In vivo genome editing restores haemostasis in a mouse model of haemophilia. Nature 475, 217-221 (2011).

Mali P, et al. RNA-guided human genome engineering via Cas9. Science 339, 823- 826 (2013).

Hou Z, et al. Efficient genome engineering in human pluripotent stem cells using Cas9 from Neisseria meningitidis. Proc Natl Acad Sci USA 110, 15644-15649 (2013). Ding Q, Regan SN, Xia Y, Oostrom LA, Cowan CA, Musunuru K. Enhanced efficiency of human pluripotent stem cell genome editing through replacing TALENs with CRISPRs. Cell Stem Cell 12, 393-394 (2013).

Consortium EP. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57-74 (2012).

Kharchenko PV, Tolstorukov MY, Park PJ. Design and analysis of ChlP-seq experiments for DNA-binding proteins. Nat Biotechnol 26, 1351-1359 (2008).

Thurman RE, et al. The accessible chromatin landscape of the human genome. Nature 489, 75-82 (2012).

Sanyal A, Lajoie BR, Jain G, Dekker J. The long-range interaction landscape of gene promoters. Nature 489, 109-1 13 (2012).

de Wit E, et al. The pluripotent genome in three dimensions is shaped around pluripotency factors. Nature 501, 227-231 (2013).

Puck JM, et al. Mutation analysis of IL2RG in human X- linked severe combined immunodeficiency. Blood 89, 1968-1977 (1997).

Kuijpers TW, et al. A reversion of an IL2RG mutation in combined

immunodeficiency providing competitive advantage to the majority of CD8+ T cells. Haematologica 98, 1030-1038 (2013).

Speckmann C, et al. Clinical and immunologic consequences of a somatic reversion in a patient with X- linked severe combined immunodeficiency. Blood 112, 4090-4097 (2008).

Cavazzana-Calvo M, et al. Gene therapy of human severe combined

immunodeficiency (SCID)-Xl disease. Science 288, 669-672 (2000).

D Amour KA, et al. Production of pancreatic hormone-expressing endocrine cells from human embryonic stem cells. Nat Biotechnol 24, 1392-1401 (2006).

Jiang J, et al. Generation of insulin-producing islet- like clusters from human embryonic stem cells. Stem Cells 25, 1940-1953 (2007).

Thomson JA, et al. Embryonic stem cell lines derived from human blastocysts.

Science 282, 1 145-1 147 (1998).

Deyle DR, et al. Normal Collagen and Bone Production by Gene-targeted Human Osteogenesis Imperfecta iPSCs. Mol Ther 20, 204-213 (2012).

Khan IF, Hirata RK, Russell DW. AAV-mediated gene targeting methods for human cells. Nature protocols 6, 482-501 (201 1). Papapetrou EP, Sadelain M. Derivation of genetically modified human pluripotent stem cells with integrated transgenes at unique mapped genomic sites. Nature protocols 6, 1274-1289 (2011).

McCullar V, et al. Mouse fetal and embryonic liver cells differentiate human umbilical cord blood progenitors into CD56-negative natural killer cell precursors in the absence of interleukin-15. Exp Hematol 36, 598-608 (2008).

Denman CI et al. Membrane-bound IL-21 promotes sustained ex vivo proliferation of human natural killer cells. PLoS ONE 7, e30264 (2012).

Miller JS, Oelkers S, Verfaillie C, McGlave P. Role of monocytes in the expansion of human activated natural killer cells. Blood 80, 2221-2229 (1992).

Claims

We claim

1. A method for controlling developmental potential of a human stem cell, comprising gene editing of a lineage-specification gene in a human stem cell genome, wherein the gene editing produces a human stem cell with limited capability of differentiating into the cell lineage for which the lineage-specification gene is specific.

2. The method of claim 1, wherein the gene editing comprises knocking out the lineage- specification gene.

3. The method of claim 1, wherein the gene editing comprises

(a) inserting a cassette comprising a negative selection marker into the lineage specification gene of the human stem cell genome, wherein the negative selection marker is under the control of the lineage-specification gene promoter and/or enhancer;

(b) isolating human stem cells that contain the negative selection marker;

(c) culturing the human stem cells under conditions suitable to express the negative selection marker and exert negative selection, wherein the culturing eliminates at least 90% of cells of the lineage for which the lineage-specification gene is specific from the cultured cell population.

4. The method of claim 3, wherein the negative selection marker is selected from the group consisting of thymidine kinase an apoptosis inducer, and a toxic gene.

5. The method of claim 1, wherein the gene editing comprises:

(a) inserting a cassette comprising a selection marker into the lineage

specification gene of the human stem cell genome, wherein the selection marker is operatively linked to an exogenous promoter;

(b) culturing the human stem cells under conditions suitable to express the selection marker; and

(c) isolating positive human stem cells that stably express the selection marker; wherein positive human stem cells comprise the selection marker under the control of the exogenous promoter integrated into the lineage specification gene such that expression of the lineage specification gene is suppressed or its encoded protein is no longer functional.

6. The method of claim 5, wherein the selection marker is an antibiotic resistance gene, a surface marker that can be used for cell purification, a metabolic gene that confers survival in the presence of a specific medium formulation, and/or a gene that provides a growth advantage.

7. The method of any one of claims 1-6, wherein the linage specification gene locus is silent in the human stem cell.

8. The method of any one of claims 5-7, wherein the exogenous promoter is active at silent gene loci.

9. The method of claim 8, wherein the exogenous promoter is a housekeeping promoter.

10. The method of claim 8 or 9, wherein the exogenous promoter comprises the nucleic acid sequence selected from the group consisting of SEQ ID NOS: 1-6, or functional equivalents thereof.

11. The method of any one of claims 5-10, wherein the cassette further comprises a negative selectable marker under the control of the lineage-specification gene promoter and/or enhancer.

12. The method of any one of claims 5-10, wherein the cassette further comprises an inactivating mutation in the lineage-specification gene, and or it's promoter and/or enhancer.

13. The method of any one of claims 3-12, wherein the method further comprises excising the cassette or a portion thereof from the positive human stem cells.

14. The method of any one of claim 13, wherein the edited locus retains a negative selectable marker or inactivating mutation after excising the cassette.

15. The method of any one of claims 1-14, wherein the lineage specification gene is selected from the group consisting of a gene encoding interleukin-2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage-specification cell surface marker, a lineage-specification transcription factor, a cytokine or hormone receptor.

16. The method of any one of claims 1-15, wherein the lineage specification gene is IL2RG.

17. The method of any one of claims 1-16, wherein the cassette is delivered by an adenoviral or rAAV vector.

18. The method of any one of claims 1-17, wherein the human stem cell is a pluripotent stem cell or an induced pluripotent stem cell.

19. A recombinant human pluripotent stem cell knocked-out for a lineage specification gene.

20. The recombinant human pluripotent stem cell of claim 19, wherein a negative selection marker is knocked into the lineage-specification gene locus, under the control of the lineage-specification gene promoter and/or enhancer.

21. The recombinant human pluripotent stem cell of claim 20, wherein the negative selection marker is selected from the group consisting of thymidine kinase an apoptosis inducer, and a toxic gene

22. A recombinant human pluripotent stem cell comprising a selection marker and exogenous promoter inserted into the lineage-specification gene, wherein the selection marker inactivates the lineage specification gene.

23. The recombinant human pluripotent stem cell of claim 22, wherein the selection marker is an antibiotic resistance gene.

24. The recombinant human pluripotent stem cell of any one of claims 19-23, wherein the linage specification gene locus is silent in the human stem cell.

25. The recombinant human pluripotent stem cell of any one of claims 22-25, wherein the exogenous promoter is active at silent gene loci.

26. The recombinant human pluripotent stem cell of claim 27, wherein the promoter is a housekeeping promoter.

27. The recombinant human pluripotent stem cell of claim 25 or 26, wherein the promoter comprises the nucleic acid sequence selected from the group consisting of SEQ ID NOS: l-6, of functional equivalents thereof.

28. The recombinant human pluripotent stem cell of any one of claims 19-27, wherein the lineage specification gene is selected from the group consisting of a gene encoding interleukin-2 receptor subunit gamma (IL2RG), brachyury, glucagon, insulin, somatostatin, a lineage-specification cell surface marker, a lineage-specification transcription factor, a cytokine or hormone receptor.