EA045868B1 - REGULATION OF GENE EXPRESSION USING CONSTRUCTED NUCLEASES - Google Patents

Description

Cross references to related applications

This application claims priority to US Provisional Application No. 62/378978, filed August 24, 2016; US Provisional Application No. 62/443981, filed January 9, 2017; and US Provisional Application No. 62/545,778, filed August 15, 2017, the disclosures of which are incorporated herein by reference in their entirety.

Technical field

This invention relates to the field of genomic engineering, in particular to targeted modification of the genome of a hematopoietic cell.

State of the art

When one considers that genome sequencing efforts have revealed that the human genome contains between 20,000 and 25,000 genes but fewer than 2,000 transcriptional regulators, it is clear that many factors must interact to control gene expression in all its various manifestations over time, development and tissue specificity. manifestations. Gene expression is controlled by a very complex mixture of general and specific transcriptional regulators that interact with DNA elements. These DNA elements include both local DNA elements, such as the core promoter and its associated transcription factor binding sites, and distal elements, such as enhancers, silencers, insulators, and locus control regions (LCRs) (see Matson et al. (2006) Ann Rev Genome Hum Genet 7: 29-50).

Enhancer elements were first identified in the SV40 viral genome and then found in the human immunoglobulin heavy chain locus. They are now known to play a regulatory role in the expression of many genes, and enhancers appear to primarily influence the temporal and spatial patterns of gene expression. It has also been found that enhancers can function to regulate expression over large distances from the core promoter of the target gene and are not dependent on any particular sequence orientation with respect to the promoter. Enhancers can be located several hundred kilobases upstream or downstream of the core promoter region, where they can be located in the intronic sequence or even beyond the 3' end of the gene.

Various methods and compositions for targeted cutting of genomic DNA have been described. Such targeted cuts can be used, for example, to induce targeted mutagenesis, induce targeted deletions of cellular DNA sequences, and facilitate targeted recombination at a predetermined chromosomal locus. See, for example, US Pat. Nos. 9,255,250; 9200266; 9045763; 9005973; 9150847; 8956828; 8945868; 8703489; 8586526; 6534261; 6599692; 6503717; 6689558; 7067317; 7262054; 7888121; 7972854; 7914796; 7951925; 8110379; 8409861; US Patent Publications No. 2003/0232410; 2005/0208489; 2005/0026157; 2005/0064474; 2006/0063231; 2008/0159996; 2010/00218264; 2012/0017290; 2011/02 65198; 2013/0137104; 2013/0122591; 2013/0177983; 2013/0196373; 2015/0056705 and 2015/0335708, the disclosure of which is incorporated by reference in its entirety.

These methods often involve the use of engineered cutting systems to induce a double strand break (DSB) or cut in the target DNA sequence in such a way that the breaks caused by an error such as non-homologous end joining (NHEJ), non-homologous end-hijacking, or repair using repair template (homology-directed repair or HDR) can result in gene knockout or insertion of a sequence of interest (targeted integration). See, for example, US Pat. Nos. 9,045,763; 9200266; 9005973 and 8703489. These methods can also be used to introduce site-specific changes to the genome sequence through the use of a donor oligonucleotide, including the introduction of specific deletions of genomic regions or specific point mutations or localized changes (also known as gene correction). Cutting can occur through the use of specific nucleases such as engineered zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), or using the CRISPR/Cas system with an engineered crRNA/tracrRNA (single guide RNA) to direct the specific cut. In addition, target nucleases are being developed based on the Argonaute system (for example, from T. thermophilus, known as TtAgo, see Swarts et al. (2014) Nature 507 (7491): 258-261), which may also have the potential for use in genome editing and gene therapy.

Red blood cells (KKK, RBC), or erythrocytes, are the main cellular component of blood and make up a quarter of all human cells. Mature red blood cells lack a nucleus and many other organelles and are full of hemoglobin, a metalloprotein that has the function of carrying oxygen from the lungs to the tissues and carrying carbon dioxide out of the tissues and back to the lungs for removal from the body. This protein makes up approximately 97% of the dry mass of red blood cells and increases the blood's ability to carry oxygen by approximately seventy times. Hemoglobin is a heterotetramer consisting of two alpha (a)-like globin chains and two beta (b)-like globin chains

- 1 045868 chains and 4 heme groups. In adults, the α2β2 tetramer is called hemoglobin A (HbA) or adult hemoglobin. Typically, alpha and beta globin chains are synthesized in an approximate ratio of 1:1, and this ratio appears to be critical for the stabilization of hemoglobin and red blood cells. The developing fetus produces another form of hemoglobin, fetal hemoglobin (HbF), which has a higher binding affinity for oxygen than hemoglobin A so that oxygen can be delivered to the baby's system through the mother's bloodstream. There are two genes that encode embryonic globin that are very similar in sequence and are called HBG1 (also called Ggamma) and HBG2 (Agamma), based on their order in the beta globin gene locus.

Like adult hemoglobin, fetal hemoglobin protein contains two α-globin chains, but instead of the adult β-globin chains, it has two fetal gamma^)-globin chains (eg, fetal hemoglobin is α2γ2). After about 30 weeks of pregnancy, fetal gammaglobin synthesis begins to decrease, and beta-globin production increases. By approximately 10 months of age, the newborn's hemoglobin is almost all α2β2, although some HbF persists into adulthood (approximately 1-3% of total hemoglobin). The regulation of the transition from gamma to beta globin production is quite complex and mainly involves suppression of gamma globin transcription while simultaneously increasing beta globin transcription.

Genetic defects in the sequences coding for hemoglobin chains may be responsible for a group of diseases known as hemoglobinopathies, which include sickle cell anemia and alpha and beta thalassemias. In most patients with hemoglobinopathies, the genes encoding gamma globin remain, but expression is relatively low due to the normal gene repression that occurs during childbirth, as described above.

An estimated 1 in 5,000 people in the United States have sickle cell disease (SCD), mostly seen in people of sub-Saharan African descent (Roseff (2009) Immunohematology 25(2):67). There is a benefit to heterozygous carriers of the sickle cell mutation in protection against malaria, so this trait may have been selected through positive selection over time, resulting in the sickle cell mutation being estimated to be present in sub-Saharan Africa up to 28% of the population (Elguero et al., (2015) PNAS USA 112(22):7051). Sickle cell disease is caused by a mutation in the β-globin gene due to the substitution of valine for glutamic acid at amino acid position No. 6 (GAG-GTG mutation at the DNA level), and the resulting hemoglobin is called hemoglobin S or HbS. Under low oxygen conditions, a conformational shift in the deoxy form of HbS exposes a hydrophobic region on the protein between helices E and F. Hydrophobic valine residues at position 6 of the beta chain in hemoglobin are able to bind to the hydrophobic region, causing aggregation of HbS molecules and the formation of fibrous precipitates. These aggregates, in turn, cause abnormal or sickled red blood cells, which results in loss of cell flexibility. Sickle red blood cells can no longer squeeze into the capillary beds and can lead to vaso-occlusive crisis in patients with sickle cell disease. In addition, sickle red blood cells are more fragile than normal red blood cells and tend to undergo hemolysis, which ultimately leads to anemia in the patient.

Treatment and care for patients with sickle cell disease is lifelong and includes antibiotic treatment, pain relief, and blood transfusions during acute episodes. One approach is the use of hydroxyurea, which works in part by increasing gamma globin production. However, the long-term side effects of long-term hydroxyurea therapy are still unknown, and the treatment produces unwanted side effects that lead to low compliance rates, and has varying effectiveness from patient to patient (Brandow and Panepinto (2011) Am J Hematol 86(9) :804-806). Despite improvements in the effectiveness of treatments for sickle cell disease, life expectancy for patients is still only in the mid to late 50s, and the disease-related conditions have a profound impact on patients' quality of life.

Thalassemias are also hemoglobin-related diseases and usually involve decreased expression of globin chains. This may occur through mutations in the regulatory regions of genes or due to a mutation in the globin coding sequence that results in decreased expression or reduced levels of functional globin protein. Alpha thalassemias, caused by mutations in the alpha globin locus, are mainly associated with people of West African and South Asian descent and can cause resistance to malaria. Beta thalassemia, caused by mutations in the beta globin locus, is mainly associated with people of Mediterranean origin, typically from Greece and coastal areas of Turkey and Italy. In thalassemia minor, only one of the βglobin alleles carries a mutation. These individuals suffer from microcytic anemia, and its finding is usually associated with a smaller than normal mean corpuscular volume (<80 fL). The alleles of subjects with thalassemia minor are β+/β or β0/β3 (where β+ refers to alleles that allow the formation of some β chain, β refers to wild-type β-globin alleles, and β0 refers

- 2 045868 to β-globin mutations associated with a complete lack of beta-globin expression). Subjects with thallasemia intermedia can often lead a normal life, but may require blood transfusions from time to time, especially during illness or pregnancy, depending on the severity of their anemia. The alleles of these patients may be β+/β+ or β0/β+. Thalassemia major occurs when both alleles have thalassemia mutations. This is a highly microcytic and hypochromic anemia. Without treatment, it causes anemia, splenomegaly and severe bone deformities and leads to death before the age of 20 years. Treatment consists of periodic blood transfusions; splenectomy to treat splenomegaly and chelate transfusion-induced iron overload. Bone marrow transplantation is also used to treat people with thalassemia major if a suitable donor can be identified, but the procedure can have significant risks.

One approach that has been proposed for the treatment of both SCD and beta thalassemia is to increase gamma globin expression to functionally replace aberrant adult hemoglobin with HbF. As mentioned above, treating patients with SCD with hydroxyurea is believed to be successful in part due to its effect on increasing gamma globin expression. The first group of compounds discovered to affect gamma-globin expression were cytotoxic drugs. The ability to induce de novo gamma-globin synthesis by pharmacological manipulation was first demonstrated using 5-azacytidine in experimental animals (DeSimone (1982) Proc Nat'l Acad Sci USA 79(14):4428-31). Subsequent studies confirmed the ability of 5-azacytidine to increase HbF in patients with β-thalassemia and sickle cell disease (Ley, et al., (1982) N. Engl. J. Medicine, 307:1469-1475, and Ley, et al., ( 1983) Blood 62: 370-380). In addition, short-chain fatty acids (eg, butyrate and derivatives) have been shown to increase HbF levels in experimental systems (Constantoulakis et al., (1988) Blood 72(6):1961-1967). Additionally, there is a portion of the human population with a condition known as hereditary persistence of fetal fetal disease (HPFH), in which elevated amounts of HbF persist into adulthood (10-40% in HPFH heterozygotes (see Thein et al. (2009) Hum. Mol Genet 18 (R2): R216-R223) This is a rare physiological condition, but in the absence of any associated beta globin abnormalities, it is not associated with any significant clinical manifestations, even if 100% of an individual's hemoglobin is HbF. When In people with beta thalassemia also have concomitant HPFH, expression of HbF may reduce the severity of the disease (Potoka and Gladwin (2015) A MJ Physiol Lung Cell Mol Physiol. 308(4): L314-L324). In addition, the severity of the natural history Sickle cell disease can vary significantly from patient to patient, and this variability, in part, may be due to the fact that some people with milder forms of the disease express higher levels of HbF.

One approach to increasing HbF expression involves identifying genes whose products play a role in regulating gamma-globin expression. One such gene is BCL11A, first identified for its role in lymphocyte development. BCL11A encodes a zinc finger protein that is believed to be involved in the specific regulation of gamma globin expression during development. BCL11A is expressed in adult erythroid progenitor cells, and downregulation of its expression results in increased gamma-globin expression. In addition, BCL11A mRNA splicing is regulated depending on the developmental stage of the organism. Fetal cells predominantly express shorter variants of BCL11A mRNA, known as BCL11A-S and BCL11A-XS, whereas adult cells predominantly express longer variants of BCL11A-L and BCL11A-XL mRNA. See Sankaran et al. (2008) Science 322 p. 1839. The BCL11A protein appears to interact with the beta-globin locus, changing its conformation and, consequently, its expression at different stages of development. The use of inhibitory RNA targeting the BCL11A gene has been proposed (see, for example, US Patent Publication No. 2011/0182867), but this technology has several potential disadvantages in that complete knockdown cannot be achieved; delivery of such RNAs may be problematic, and the RNA must be constantly present, requiring multiple treatments throughout life.

Targeting BCL11A enhancer sequences provides a mechanism for increasing HbF abundance. See, for example, US Patent Publication No. 2015/0132269 and PCT Publication No. WO 2016/183298. Genomic association studies have identified a number of genetic variations at the BCL11A gene locus that are associated with elevated HbF levels. These variations are a collection of small (single) nucleotide polymorphisms (SNPs) found in the noncoding regions of BCL11A that function as a stage-specific lineage-restricted enhancer region. Further studies showed that this BCL11A enhancer is required in erythroid cells for BCL11A expression, but is not required for its expression in B cells (see Bauer et al. (2013) Science 343:253-257). An enhancer region was found in intron 2 of the BCL11A gene, and three DNase I hypersensitive regions (often indicative of a chromatin state that is associated with regulatory potential) were identified in intron 2. These three regions were identified as +62, +58, and +55 according to the distance in kilobases from the trans start site

- 3 045868 BCL11A transcriptions.

These enhancer regions are approximately 350(+55); 550(+58) and 350(+62) nucleotides in length (Bauer 2013, ibid).

When developing a nuclease for use in human therapeutics, it is important that the nuclease has maximum safety characteristics for patients. In particular, nucleases must have very low levels of off-target cutting. Significant numbers of double-stranded cuts at locations other than the user-specified target can lead to repression of off-target genes and, in rare cases, to the occurrence of chromosomal translocations (see Hoban and Bauer (2016) Blood, 127(21):2525-2535 and Tsang et al (2017) Nature Methods, in press). Improvements in specificity can be achieved by eliminating nonspecific interactions between the engineered nuclease and genomic DNA (see US Provisional Patent Application Nos. 62/378978 and 62/443981).

Thus, there remains a need for additional highly specific methods and compositions for altering the expression of the BCL11A gene, for example, for the treatment of hemoglobinopathies such as sickle cell anemia and beta thalassemia.

The essence of the invention

This invention describes highly specific compositions and methods for use in gene therapy and genome engineering. In particular, the described methods and compositions relate to the inactivation (eg, by completely or partially abrogating its expression) of the BCL11A gene, for example, a gene that acts as a regulator of one or more additional genes. In particular, the invention describes methods and compositions for influencing the function of an enhancer in the BCL11A gene to reduce or suppress its activity in specific cell lines (eg, erythroid). In addition, the invention provides methods and compositions for interfering with the functions of the BCL11A enhancer, wherein the enhancer sequences are not located in the coding sequences of the BCL11A gene, and wherein the proposed reagents exhibit highly specific activity. The resulting suppression of the BCL11A gene under these circumstances results in increased gammaglobin expression and the incidence of off-target cutting events is reduced.

In some aspects, the invention includes a non-naturally occurring zinc finger protein comprising a zinc finger protein (ZFP) comprising 4, 5 or 6 fingers, wherein each finger contains a recognition helix region that recognizes a minor target site in DNA, wherein the recognition helix regions contain sequences in the order specified in one row of the table. 1. Within each zinc finger helix recognition region of 7 amino acids is numbered -1 to +6 in the zinc finger backbone (of approximately 30 residues, including the coordinating residues for zinc). In certain embodiments, the 1, 2, 3, or more zinc finger components of the zinc finger proteins described herein further comprise mutations at one or more residues outside the recognition helix region, including, but not limited to, amino acid mutations at the -5 position -14 or mutations at both positions -5 and -14 (numbering continues from the -1 to +6 numbering used for the recognition helix region). See, for example, the Qm4 and Qm14 mutations described in US Provisional Patent Applications 62/378978 and 62/443981. The component zinc fingers of a zinc finger protein may be linked by any linkers, for example, as described in US Pat. No. 8,772,453. In some embodiments, ZFP (zinc finger protein) comprises the recognition helices shown in Table 1. 1 for proteins designated as follows: 63014 (which binds to the target site shown in SEQ ID NO: 1) and 65722 (which binds to the target site shown in SEQ ID NO: 2).

In certain embodiments, zinc finger proteins as described herein are fused to a functional domain (eg, a transcriptional activation domain, a transcriptional repression domain, a cutting domain (to form a zinc finger nuclease), etc.). Any linker can be used to operably link the cutting domain and the zinc finger protein, including, but not limited to, the linkers described in US Pat. Nos. 9,394,531 and 9,567,609. Additionally, when a FokI cutting domain is used, further mutations in the catalytic protein may be present. domain, dimerization domain at phosphate contact residues (not present in the dimerization or catalytic domain), and combinations of mutations in any one of the catalytic domain, dimerization domain at phosphate contact residues, including, but not limited to, ELD or KKR mutations in the domain dimerization, mutations in residues 525 (K to S) of the FokI domain, and combinations of ELD or KKR mutations in the dimerization domain and mutations in residues 525 (K to S) of the FokI domain, numbered relative to the wild type. See US Patent No. 7888121; 7914796; 8034598; 8623618 and US Patent Publications No. 2011/0201055 and US Provisional Patent Applications Nos. 62/378978 and 62/443981.

In some embodiments, zinc finger nucleases (ZFNs) can be used in dimerizing pairs to cut at one or both target sites for the ZFN pair, for example, the left partner in Table. 1 (for example, 63014) can form a dimer with the right partner from the table. 1 (e.g. 65722) to cut BCL11A enhancer sequences. In certain embodiments, the ZFN pair contains the following amino acid sequences:

- 4 045868

63014:

MDYKDHDGDYKDHDIDYKDDDDKMAPKKKRKVGIHGVPAAMAERPFQCRICMQNFSDQSNLRAH IRTHTGEKPFACDICGRKFARNFSLTMHTKIHTGSQKPFQCRICMQNFSSTGNLTNHIRTHTGE KPFACDICGRKFATSGSLTRHTKIHTHPRAPIPKPFQCRICMQNFSDQSNLRAHIRTHTGEKPF ACDIC GRKFAAQCCLFHHTKIH- Linker ELEEKKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKHLGGSR

KPDGAIYTVGS PIDYGVIVDTKAYS GGYNLPIGQADEMERYVEENQTRDKHLNPNEWWKVYPS S VTEFKFLFVSGHFKGNYKAQLTRLNHITNCNGAVLSVEELLIGGEMIKAGTLTLEEVRRKFNNG EINFRS (SEQ ID NO:3) ; and

MDYKDHDGDYKDHDIDYKDDDDKMAPKKKRKVGIHGVPAAMAERP FQCRICMQKFARND HRTTHTKIHTGEKPFQCRICMQNFSQKAHLIRHIRTHTGEKPFACDICGRKFAQKGTLGEHTKI HTGSQKPFQCRICMQNFSRGRDLSRHIRTHTGEKPFACDICGRKFARRDNLHSHTKIHLinkerELEE KKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKHLGGSR

KPDGAIYTVGS PIDYGVIVDTKAYS GGYNLPIGQADEMQRYVKENQTRNKHINPNEWWKVYPS S

VTEFKFLFVSGHFSGNYKAQLTRLNRKTNCNGAVLSVEELLIGGEMIKAGTLTLEEVRRKFNNG

EINF (SEQ ID NO: 4), wherein the linker sequence may be any linker sequence known in the art, for example, as described in US Pat. Nos. 9,394,531 and 9,567,609. In some embodiments, the linker for 63014 comprises or consists of a L7c5 linker ( LRGSISRARPLNPHP (SEQ ID NO: 5)), and the linker used in 65722 contains or consists of a LO linker (LRGSQLVKS (SEQ ID NO: 6), see US Patent No. 9567609). The FokI cutting domain sequence that is C-terminal to the linker sequences shown above may also contain alternative FokI domains operably linked to the zinc finger protein. In some embodiments, the FokI cutting domain may include alternative or additional mutations in the catalytic domain, dimerization domain, phosphate contact residues, and combinations of mutations in any one of the catalytic domain, dimerization domain, and phosphate contact residues.

In another aspect, the invention includes delivering at least one nuclease (eg, a nuclease that binds to a BCL11A enhancer sequence) into a human stem or progenitor cell (HSC/PC) for the purpose of genome engineering. In some embodiments, the nuclease comprises a zinc finger protein (ZFP) containing 4, 5, or 6 fingers, each finger containing a recognition helix region that recognizes a target site, the recognition helix regions containing sequences in the order shown in one row of the table. 1. In other embodiments, the ZFN nuclease contains a pair of nucleases designated 63014/65722. The nuclease(s) as described herein may further comprise a linker (e.g., between a DNA binding domain and a cleavage domain), such as a linker as shown in 9567609, including, but not limited to, LRGSISRARPLNPHP (SEQ ID NO: 5) or (LRGSQLVKS (SEQ ID NO: 6)).

In some embodiments, the nuclease is delivered as a peptide, while in others it is delivered as a nucleic acid encoding at least one nuclease. In some embodiments, more than one nuclease is used. In some preferred embodiments, the nucleic acid encoding the nuclease is mRNA, and in some cases the mRNA is protected. In some aspects, the mRNA can be chemically modified (see, for example, Kormann et al. (2011) Nature Biotechnology 29(2):154-157). In other aspects, the mRNA may contain an ARCA cap (see US Pat. Nos. 7,074,596 and 8,153,773). In other embodiments, the mRNA may contain a mixture of unmodified and modified nucleotides (see US Patent Publication No. 2012/0195936). In a preferred embodiment, the nucleic acid encoding the nuclease(s) is delivered to the HSC/PC via electroporation. In some embodiments, the nuclease cuts at or near a transcription factor binding site. In some aspects, the transcription factor is GATA-1.

In other aspects, the invention includes a cell or cell line in which the endogenous BCL11A enhancer sequence is genetically modified by a nuclease as described herein (eg, shown in Table 1), for example, compared to the sequence of a wild-type cell. Genetic modification of the BCL11A enhancer results in modification of globin gene expression (beta and gamma). Nuclease-modified cells or cell lines, as described herein, differ in structure, function, and combinations of both structure and function from the wild type. Genetically modified cells or cell lines can be heterozygous or homozygous for the modification. Modifications may include insertions (eg, transgene insertions) and combinations of insertions and deletions; such insertions, deletions, and combinations of insertions and deletions are commonly referred to as indels. In some preferred embodiments, indels result in disruption of the transcription factor binding site. In some va

- 5 045868 embodiments, the modification occurs at or near the nuclease(s) binding site(s), cutting site(s), and combinations of binding and cutting sites, for example, within 1-300 (or any value in between) base pairs above- or downstream of the cleavage site(s), more preferably within 1-100 base pairs (or any value in between) on either side of the binding site(s), cleavage site(s), and combinations of binding and cleavage sites as shown in Table . 1, even more preferably ranging from 1 to 50 base pairs (or any value in between) on either side of the binding site(s), cutting site(s), and combinations of binding and cutting sites. In certain embodiments, the genetic modification of the BCL11A enhancer sequence is within and/or between the sequences shown in table. 1 (target sites). The modification may also include modifications to one or more nucleotides at the cutting sites. The modification may also include modifications to one or more nucleotides at the binding sites. The modification may further include modifications to one or more nucleotides at the cutting sites and at one or more binding sites. In certain embodiments, one or more nuclease target sites are unmodified. In other embodiments, at least one of the target sites for the nuclease(s) is modified. In some embodiments, the modification is at or near the +58 region of the BCL11A enhancer, for example, at or near the nuclease binding site shown in either SEQ ID NO: 1 and SEQ ID NO: 2. Any cell or cell line can be modified with nucleases, as described herein, for example, a stem cell (hematopoietic stem cell, such as a CD34+ hematopoietic stem cell) or a red blood cell progenitor cell (RBC).

Also described are cells or cell lines obtained after modification with a nuclease as described herein, for example cells or cell lines derived from a nuclease modified cell or cell line as described herein. Also provided are partially or fully differentiated cells derived from modified stem cells as described herein (eg, red blood cells or red blood cell progenitor cells). Cells derived from nuclease(s) modified cells may be expanded, differentiated, and combinations from both those expanded and differentiated in in vitro culture, or may be differentiated in a living subject, for example, following ex vivo administration of the nuclease(s) modified stem cell. Any of the genetically modified cells or cell lines disclosed herein may exhibit increased expression of gamma globin. Compositions, such as pharmaceutical compositions, containing genetically modified cells, as described herein, are also provided.

In other aspects, the invention includes delivering a donor nucleic acid to a target cell to provide a genetically modified cell in which the donor is integrated into the cell. The donor may be delivered before, after, or together with the nucleic acid encoding the nuclease(s) from Table. 1. The donor nucleic acid may contain an exogenous sequence (transgene) for integration into the cell genome, for example, an endogenous locus. In some embodiments, the donor may contain a full-length gene or fragment thereof flanked by regions of homology to the target cutting site. In some embodiments, the donor has no homologous regions and is integrated into the target locus through a homology-independent mechanism (ie, NHEJ). The donor may contain any nucleic acid sequence, for example a nucleic acid that, when used as a substrate for homology-directed nuclease-induced double-strand break repair, results in the generation of a donor-specified deletion at an endogenous chromosomal locus (e.g., the BCL11A enhancer region) or, alternatively (or in addition to), new allelic forms (eg, point mutations that inactivate the transcription factor binding site) of the endogenous locus created. In some aspects, the donor nucleic acid is an oligonucleotide, wherein integration results in the phenomenon of gene correction or targeted deletion.

In other aspects, the nuclease, donor, and combinations of both nuclease and donor are delivered by viral, nonviral, and combinations of viral and nonviral gene transfer methods. In preferred embodiments, the donor is delivered into the cell via an adeno-associated virus (AAV). In some cases, AAV contains LTRs that are of a heterologous serotype compared to the capsid serotype.

In some aspects, deletion-containing regions within DNase I hypersensitive enhancer regions (eg, the +58 region of the BCL11A enhancer) are generated using one or more nucleases, as shown in Table 1. 1. These deletions can range from about 1 nucleotide to about 551 nucleotides. Thus, deletions may contain 1, 5, 10, 15, 20, 25, 30, 40, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550 nucleotides, or any value in between. In some embodiments, the deletions contain binding regions for one or more transcription factors. In some preferred embodiments, the deletions contain a GATA-1 binding site or a binding site for GATA-1 in combination with other factors.

In some embodiments, the DNA binding domains from Table. 1 merged with function

- 6 045868 local domain. Some aspects include fusion of DNA-binding domains with domains capable of regulating gene expression. In some embodiments, the fusion proteins comprise the DNA binding domain from Table 1. 1 fused to a gene expression modulation domain, wherein the modulator suppresses gene expression.

In some embodiments, HSC/PC cells are contacted with nucleases, DNA-binding proteins, and combinations of nucleases and DNA-binding proteins of the invention (ie, ZFPs, as shown in Table 1). In some embodiments, nucleases, DNA-binding proteins, and combinations of nucleases and DNA-binding proteins are delivered as nucleic acids, and in other embodiments, they are delivered as proteins. In some embodiments, the nucleic acids are mRNAs encoding DNA-binding proteins and combinations of nucleases and DNA-binding proteins, and in additional embodiments, the mRNAs may be protected. In some embodiments, the mRNA may be chemically modified and may include an ARCA cap, a mixture of unmodified and modified nucleotides, and combinations of an ARCA cap and a mixture of unmodified and modified nucleotides. Also provided are cells or cell lines derived from these cells, including partially or fully differentiated cells.

In some aspects, HSC/PCs are contacted with nucleases, DNA-binding proteins, and combinations of nucleases and DNA-binding proteins of the invention ex vivo, following apheresis of the HSC/PC from the subject or purification from the resulting bone marrow. In some embodiments, the nucleases described herein cause modifications in BCL11A enhancer regions, for example, resulting in a genetically modified cell that is structurally, functionally, and combined structurally and functionally different from wild-type cells, otherwise modified (e.g., nuclease-modified (s)) cells and combinations of wild type and other modified cells. In other embodiments, HSC/PC containing modifications of the BCL11A enhancer region are administered back to the subject. In some cases, HSC/PCs containing modifications of the BCL11A enhancer region are expanded before injection. In other aspects, genetically modified HSC/PCs are administered to a subject as part of a bone marrow transplant, wherein the HSC/PCs engraft, differentiate and mature in vivo. In some embodiments, HSC/PCs are isolated from a subject following G-CSF-induced mobilization, plerixafor-induced mobilization, and combinations of G-CSF and plerixafor-induced mobilization, and in others, the cells are isolated from human bone marrow or human umbilical cords . In some aspects, the subject is subjected to a mild myeloablative procedure prior to receiving a graft containing the modified HSC/PC, while in other aspects, the subject is subjected to an intense myeloablative regimen. In some embodiments, the methods and compositions of this invention are used to treat or prevent hemoglobinopathies. In some aspects, the hemoglobinopathy is thalassemia. In some aspects, the hemoglobinopathy is beta thalassemia, while in other aspects, the hemoglobinopathy is sickle cell anemia.

In some embodiments, the HSC/PC is further contacted with a donor molecule. In some embodiments, the donor molecule is delivered by a viral vector. The donor molecule may contain one or more sequences encoding a functional polypeptide (eg, cDNA or fragment thereof), with or without a promoter. Additional sequences (coding or non-coding sequences) may be included when the donor molecule is used for inactivation, including, but not limited to, sequences encoding the 2A peptide, SA site, IRES, etc.

In one aspect, the methods and compositions of this invention include methods of contacting HSC/PC in vivo. Nucleases, DNA-binding proteins, or a combination of nucleases and DNA-binding proteins are delivered to HSC/PC in situ by methods known in the art. In some embodiments, the nucleases and/or DNA-binding proteins of the present invention comprise a viral particle that is administered to a subject in need, while in others, the nucleases, DNA-binding proteins, or a combination of nucleases and DNA-binding proteins comprise a nanoparticle (eg, a liposome). In some embodiments, viral particles, nanoparticles, or a combination of viral particles and nanoparticles are delivered to the organ (eg, bone marrow) in which the HSC/PC resides.

In another aspect described herein, methods are described for integrating a donor nucleic acid into the genome of a cell through homology-independent mechanisms. These methods involve creating a double-strand break (DSB) in the genome of a cell and cutting the donor molecule using a nuclease as described herein so that the donor nucleic acid is integrated into the DSB site. In some embodiments, the donor nucleic acid is integrated through homology-independent methods (eg, NHEJ). As noted above, after cutting in vivo, donor sequences can be specifically integrated into the cell genome at the location of the DSB. The donor sequence may include one or more

- 7 045868 unique target sites for one or more nucleases used to create DSBs. Thus, the donor sequence can be cut by one or more of the same nucleases used to cut the endogenous gene into which integration is desired. In some embodiments, the donor sequence includes nuclease target sites other than the nucleases used to induce DSBs. DSBs in the target cell genome can be created by any mechanism. In some embodiments, the DSB is created by one or more zinc finger nucleases (ZFNs), fusion proteins containing a zinc finger binding domain that is designed to bind a sequence within the region of interest, and a cutting domain or cutting half-domain.

In one aspect, the donor may encode a regulatory protein of interest (eg, a ZFP TF, a TALE TF, or a CRISPR/Cas TF) that binds to, modulates the expression of, or both binds to and modulates the expression of a gene of interest. In one embodiment, said regulatory proteins bind to a DNA sequence and prevent the binding of other regulatory factors. In another embodiment, binding of a regulatory protein may modulate (ie, induce or repress) the expression of a target DNA.

In some embodiments, the transgenic HSC/PC cell, transgenic animal, or combination of transgenic HSC/PC cell and animal includes a transgene that encodes a human gene. In some cases, the transgenic animal contains a knockout at an endogenous locus and replacement of the endogenous gene with its human counterpart, allowing the development of an in vivo system where the human protein can be studied in isolation. Such transgenic models can be used for screening purposes to identify small molecules or large biomolecules or other entities that may interact with or modify a human protein of interest. In some aspects, the transgene is integrated at a selected locus (e.g., genomic safe-harbor) into a stem cell (e.g., embryonic stem cell, induced pluripotent stem cell, hematopoietic stem cell, etc.) or an animal embryo obtained by any of the methods described herein, and then the embryo is implanted so that a live animal is born. The animal is then brought to sexual maturity and allowed to produce offspring, at least some of the offspring containing an edited endogenous gene sequence or an integrated transgene.

In another aspect, provided herein is a method of altering the expression of a gene (e.g., BCL11A, a globin gene, and combinations of BCL11A and a globin gene) in a cell, comprising: introducing into the cell one or more of the nucleases described herein (shown in Table 1), in conditions where one or more proteins are expressed and gene expression is altered. In some embodiments, the expression of a globin gene (eg, gamma globin or beta globin) is altered (eg, increased). Any of the methods described herein may further include integrating a donor sequence (e.g., a transgene or fragment thereof under the control of an exogenous or endogenous promoter) into the genome of the cell, for example, integrating the donor at or near a nuclease cleavage site in the BCL11A gene. The donor sequence is introduced into the cell using a viral vector, as an oligonucleotide, on a plasmid, and a combination of one or more methods selected from using a viral vector, as an oligonucleotide, or on a plasmid. The cell in which gene expression is altered may be, for example, a red blood cell progenitor (RBC), a hematopoietic stem cell (eg, a CD34+ cell), or a combination of red blood cell progenitor cells and hematopoietic stem cells.

In other embodiments, provided herein is a method of producing a genetically modified cell comprising a genomic modification in an endogenous BCL11A enhancer sequence (BCL11A enhancer nucleotide sequence modification), the method comprising the steps of: a) contacting the cell with a polynucleotide (e.g., DNA or mRNA) , encoding a zinc finger nuclease containing 4, 5 or 6 zinc finger domains, in which each of the zinc finger domains contains a recognition helix region in the order shown in one row of the table. 1; b) exposing the cell to conditions conducive to expression of the zinc finger protein from the polynucleotide; and c) modifying the endogenous BCL11A enhancer sequence with the expressed zinc finger protein sufficient to produce a genetically modified cell. In some embodiments, the cells are stimulated with at least one cytokine (eg, prior to step (a)). The polynucleotide can be brought into contact with a cell using any suitable method, including, but not limited to, through transfection, using a non-viral vector, using a viral vector, using chemical means, or by applying an electric field (eg, electroporation).

Also provided are cells containing one or a combination of the genomic modifications described herein, including cells derived from cells obtained by the methods described herein.

Also proposed is a method of treating a patient in need of increased expression of the globin gene, comprising administering to the patient a pharmaceutical drug, wherein the pharmaceutical drug with

- 8 045868 contains genetically modified cells, proteins, polynucleotides and combinations of one or more selected from genetically modified cells, proteins and polynucleotides, as described herein, in an amount sufficient to increase globin gene expression in a patient. In certain embodiments, the patient is known to have been diagnosed with, suspected of having, or at risk of developing thalassemia or sickle cell disease.

Also provided is a kit containing nucleic acids, proteins, genetically modified cells, and combinations of one or more selected from the nucleic acids, proteins, and genetically modified cells of this invention. The kit may include nucleic acid encoding nucleases (for example, RNA molecules or genes encoding ZFN, TALEN or the CRISPR/Cas system contained in a suitable expression vector), aliquots of nuclease proteins, donor molecules, suitable modifiers of stem cell self-renewal (stemness), cells, buffers, instructions (eg, for carrying out the methods of this invention), and the like, including various combinations of these kit components. The invention includes, but is not limited to, a genetically modified cell (e.g., a stem cell such as a hematopoietic (CD34+) stem cell or red blood cell progenitor cell) containing at least one genomic modification made by a nuclease (e.g., as shown in one row Table 1), wherein the genomic modification is located in the endogenous BCL11A enhancer sequence, and wherein, in addition, the genomic modification is selected from the group consisting of insertions, deletions, and combinations thereof, and contains a modification at, near, or between any of SEQ ID NO: 1 and SEQ ID NO: 2. In some embodiments, the cell is a genetically modified differentiated cell derived from a stem cell as described herein (eg, an RBC derived from a hematopoietic stem cell or an RBC progenitor cell).

The nuclease may comprise at least one zinc finger nuclease (ZFN) (eg, as shown in Table 1), at least one TALEN, and combinations of at least one ZFN and at least one TALEN. Said nuclease(s) may be introduced into the cell as a protein, as a polynucleotide encoding the nuclease(s), or as a combination of a protein form and a polynucleotide encoding the nuclease(s). In certain embodiments, the genomic modification includes an insertion that includes the integration of a donor polynucleotide encoding the transgene. Also provided are pharmaceutical compositions containing one or more genetically modified cells, as described herein.

Also provided is a DNA binding protein comprising a zinc finger protein comprising 4, 5 or 6 zinc finger domains containing a recognition helix region, the zinc finger protein containing recognition helix regions in the order shown in one row of the table. 1. Also provided is a TALE protein containing multiple repeats that bind to a sequence comprising a portion (eg, at least 4, 5, 6 or more) of the target site base pairs shown in Table 1. 1. A fusion protein comprising a zinc finger protein or a TALE protein as described herein and a wild-type or engineered or half-cutting domain is also provided, as are the protein-encoding polynucleotides (ZFP, TALE, ZFN, TALEN) described in this document. Also provided are cells (eg, isolated stem cells, such as hematopoietic (CD34+) stem cells) containing one or more polynucleotides, proteins, and combinations of polynucleotides and proteins, as described herein. Also provided are kits containing one or more proteins, polynucleotides, cells, or combinations thereof, as described herein.

Also described is a method for altering the expression of a globin gene in a cell (e.g., an erythrocyte progenitor cell, a hematopoietic stem cell, and combinations of an erythrocyte progenitor cell and a hematopoietic stem cell), comprising: introducing into the cell one or more polynucleotides encoding one or more nucleases as described herein, under conditions in which one or more proteins are expressed and the expression of a globin gene (eg, gamma globin, beta globin, and combinations of gamma and beta globin) is altered (eg, increased). In some embodiments, the methods further include integrating the donor sequence into the genome of the cell, for example, using a viral vector, as an oligonucleotide, or on a plasmid. Said donor sequence may contain a transgene under the control of an endogenous or exogenous promoter.

Also proposed is a method for producing a genetically modified cell containing a genomic modification in the endogenous enhancer sequence of BCL11A (for example, the target site, as shown in Table 1), the method comprising the steps of: (a) bringing the cell into contact with a polynucleotide encoding a fusion protein containing a zinc finger nuclease containing 4, 5 or 6 zinc finger domains, in which each of the zinc finger domains contains a recognition helix region in the order shown in one row of the table. 1; (b) exposing the cell to conditions conducive to expression of the fusion protein from the polynucleotide; and (c) modification of the endogenous BCL11A enhancer sequence by the expressed fusion protein sufficient to produce a genetic

- 9 045868 ski modified cell. In certain embodiments, the method further includes stimulating the cells with at least one cytokine. The polynucleotide(s) may be delivered into a cell, for example, using a non-viral delivery system, a viral delivery system, a delivery vehicle, and combinations selected from a non-viral delivery system, a viral delivery system, and a delivery vehicle, and may involve exposing the cells to an electric field or using disruption cell membrane as a delivery mechanism (so-called compression technology, see e.g. Sharei et al. (2015) PLOS ONE doi: 10.1371/joumal/pone. 0118803).

Methods of treating a patient in need of increasing globin gene expression (e.g., the patient is known to have, is suspected of having, or is at risk for developing a hemoglobinopathy such as thalassemia (e.g., β-thalassemia) or sickle cell disease are also provided, the method comprising administering to the patient a pharmaceutical composition as described herein (eg, proteins, polynucleotides, cells, or a combination selected from proteins, polynucleotides and cells) in an amount sufficient to increase globin gene expression in the patient.

These and other aspects will be apparent to one of ordinary skill in the art in light of the disclosure as a whole.

Brief description of graphic materials

Fig. 1A and 1B are diagrams depicting an overview of the analysis of the integration site of an oligonucleotide duplex. In fig. 1 depicts the first step of the assay, in which cells are treated with ZFN in the presence of additional oligonucleotide duplex DNA, which is captured in the resulting cutting event fraction. In fig. 1B shows the next step in which cells are cultured for seven days, genomic DNA is isolated, and genomic segments flanking the donor integration sites are amplified by adapter-mediated PCR using an integrated oligonucleotide duplex primer. The amplicons are then sequenced to identify potential cutting sites.

In fig. Figure 2 depicts the location of 455 potential off-target loci identified by oligonucleotide duplex integration site analysis using the parent 51857/51949 or parent ZFN pair. The top 63 loci are highlighted in gray and were analyzed in subsequent analyses. Each locus can be identified by a chromosome and base number indicating the most likely location of the cut and also indicating the number of integrants found at that locus.

In fig. 3A-3C show exemplary phosphate contact moieties within a zinc finger framework. In fig. Figure 3A depicts the interaction of a zinc finger with a DNA molecule and depicts the location of the wild-type arginine side chain and how it interacts with the phosphate backbone of the DNA molecule. In fig. 3B (SEQ ID NO: 7-17) shows representative ZFN sequences showing where this arginine is located in the primary sequence of each zinc finger (vertical row of 'R' residues indicated by a thick arrow), and also highlights those arginines that have been replaced by glutamine in the ZFP backbone to eliminate the corresponding phosphate contact (individually labeled 'R' residues). The sequences also show the recognition helix regions (where residues −1 to +3, +5, and +6 are boxed and shaded), as well as part of the linker between the C-terminal zinc finger domain and the cutting domain (the cutting domain is not shown). In fig. 3C further depicts the spatial arrangement of another potentially backbone-contacting lysine side chain present in the FokI cutting domain, which may be replaced by a serine during specificity optimization.

Fig. 4 is a graph comparing the levels of modification by treating CD34 cells with either the original 51857/51949 pair or the optimized 63014/65722 pair at the various loci tested. The bottom or x-axis shows the loci identified as potential cutting targets, with the percentage of indels for each region (site) shown on the vertical or y-axis. Note that the y-axis is shown on a logarithmic scale. Dark gray bars represent loci cleaved by the 51857/51949 pair and the amount of cutting detected, while light gray bars represent loci cleaved by the optimized pair, where almost all of the measured cutting is in the BCL11a target sequence.

Fig. 5A and 5B are graphs depicting the relative ratios of globin mRNA produced in hCD34+ cells after treatment with BCL11A-specific ZFN and erythroid differentiation. CD34+ cells obtained from two healthy human donors (PB-MR-003 or PBMR-004) were treated or not with the ZFN pair and then analyzed for α, β and γ globin expression. The best method for determining the amount of γ-globin mRNA detected after ZFN treatment is to express the change in expression as either the ratio of γ-globin to β-globin (Fig. 5A) or γ-globin to α-globin mRNA (Fig. 5B).

Fig. 6 is a graph depicting the relative amounts of γ-globin protein produced in treated CD34+ cells. As above, two sets of CD34+ cells obtained from healthy human donors (PB-MR-003 and PB-MR-004) were used. In this experiment approximately 3

- 10 045868

A 4-fold increase in the percentage of fetal globin protein to levels of about 15-20% was observed in HSPC erythroid progeny upon 63014/65722-mediated disruption of the BCL11A enhancer in both donor batches.

Fig. 7 is a graph showing relative human chimerism in mice transplanted with lots of 63014/65722 treated donors (+ZFN) as described above. Human chimerism was measured by detecting cells carrying the hCD45 marker on their surface using FACS. The percentage of human hCD4 5+ cells in peripheral blood collected 8 or 12 weeks after transplantation is shown. The data showed good engraftment levels in this study, with comparable human chimerism following engraftment of untransfected control ((-)) and ZFN-transfected HSPCs (+ZFN). Open circles and triangles represent individual animals.

In fig. Figure 8 depicts the percentage of chimera found in the bone marrow of engrafted mice in which human cells were identified by the presence of hCD45 on their cell surface. Samples were analyzed 12 weeks after engraftment.

Fig. 9A-9D are graphs depicting the recovery of various hematopoietic cell lines tested by FACS analysis of bone marrow cells in mice inoculated at week 12 with antibodies recognizing cell surface markers specific to certain populations. Data showed comparable presence of all human hematopoietic lineages analyzed in the bone marrow at week 12 post-injection in progeny CD34+ cells (14/22) treated with BCLHA-specific ZFN mRNA and untransfected cells ((-)). Lymphoid, myeloid, erythroid and HSPC data are shown (Figures 9A-9D, respectively) for cells obtained from both donors (003 and 004). Data showed comparable representation of all human hematopoietic lineages analyzed in the bone marrow at week 12 postinjection in progeny of CD34+ cells treated with ZFN BCL11A mRNA and progeny of untransfected cells.

Fig. 10 is a graph showing the level of gene modification at the BCL11A target in DNA isolated from the peripheral blood of modified mice analyzed by deep sequencing. Data are shown for introduced cells (2 days after ZFN(+) transfection) and then for blood cells 8 or 12 weeks after engraftment and demonstrated good retention of the gene modification. Untransfected cells are indicated (-) in the Treatment line.

Fig. 11 is a graph showing the amount of gene modification on the BCL11A target for bone marrow cell samples after engraftment of ZFN-treated cells (+). Untreated cells are indicated (-) in the Process line. Comparable modification was observed in both BCL11A-dependent lineages (B cells expressing the marker CD19; primitive progenitors expressing CD45 and high levels of CD38) and BCLHA-independent (myeloid) lineages. Although the levels of gene modification upon input were higher in the donor sample PB-MR-003 than in the donor sample PB-MR-004, cells derived from PB-MR-004 repeatedly showed higher levels of modification, i.e. better retention of the modification in mice than those obtained from PB-MR-003.

Fig. 12 is a graph showing the amount of gene modification in erythroid cells obtained from bone marrow cells at 12 weeks after in vitro differentiation for 14 days. Data were obtained from mice initially inoculated with the two different donors described above and demonstrated that BCL11A modification mediated by ZFN (+ ZFN) treatment is not markedly altered during erythroid differentiation. Cells that were not treated with ZFN are designated as (−) ZFN.

Fig. 13A and 13B are graphs depicting the relative amount of mRNA encoding γ-globin, wherein the concentration of γ-globin mRNA is plotted as either the γ-globin mRNA/β-globin ratio (Fig. 9A) or the γ-globin mRNA/α-globin ratio (Fig. 9B). In both untransfected ((-)ZFN) and ZFN-treated (+ZFN) samples, the ratios of γ-globin to β-globin or γ-globin to α-globin mRNA varied greatly between the erythroid cell progeny of individual mice from the same same groups. However, despite this variability and the variability caused by the use of two different human donors, the average values for the 63014/65722-treated samples demonstrate an ~1.5- to 2-fold increase in γ-globin mRNA levels compared to their corresponding untransfected counterparts.

Fig. 14 is a graph depicting the difference in the amount of γ-globin protein (expressed as the ratio γ-globin/α-globin or γ-globin/total β-like protein) in cells obtained from bone marrow from engineered mice, wherein the bone marrow cells brains were subjected to an in vitro differentiation protocol. Protein levels were measured 16 days after the onset of differentiation. The ratios of gamma (γ) globin (sum of Agamma and Ggamma peaks) to alpha (α) globin, as well as gamma globin (sum of Agamma and Ggamma peaks)/beta-like globin ratios (sum of Agamma, Ggamma, beta and delta globin peaks) were determined ). Consistent with the weak

- 11 045868 erythroid differentiation of samples derived from PB-MR-003, gamma globin levels in untransfected cells obtained from this donor were very high (~30%), and therefore ZFN (+ ZFN) treatment resulted in an increase in gamma levels -globin only 1.2 times compared to untreated cells ((-) ZFN). PB-MR-004 exhibited more typical untransfected levels (~9%) and showed a ~2-fold increase in gamma globin protein levels after 12 weeks of circulation through the mouse body.

Detailed description of the invention

Disclosed herein are compositions and methods for genome engineering to modulate the expression of BCL11A, gamma globin, and combinations of BLC11A and gamma globin, and for the treatment, prevention, or treatment and prevention of hemoglobinopathies. In particular, through exposure using nucleases containing ZFP having recognition helix regions, as shown in one row of the table. 1, disruption of the BCL11A enhancer is efficiently achieved in HSC/PC and results in changes in the relative expression of gamma globin during subsequent erythropoiesis. This modulation of BCL11A and gamma globin expression is particularly useful for the treatment of hemoglobinopathies (eg, beta thalassemia, sickle cell disease) in which there is insufficient expression of beta globin or expression of a mutated form of beta globin. When using the methods and compositions of this invention, complications and disease-related consequences caused by aberrant amounts of beta globin can be overcome by altering the expression of gamma globin in red blood cell progenitor cells.

general description

The practice of the methods and the preparation and use of the compositions disclosed herein utilize, unless otherwise indicated, conventional techniques of molecular biology, biochemistry, chromatin structure and analysis, computational chemistry, cell culture, recombinant DNA and related fields that are within skills in this cloud of technology. These methods are fully explained in the literature, see for example Sambrook et alMOLECULAR CLONING: A LABORATORY MANUAL, Second edition, Cold Spring Harbor Laboratory Press, 1989 and third edition, 2001; Ausubel et al., CURRENT PROTOCOLS IN MOLECULAR BIOLOGY, John Wiley & Sons, New York, 1987 and periodic reprints; METHODS IN ENZYMOLOGY series, Academic Press, San Diego; Wolffe, CHROMATIC STRUCTURE AND FUNCTION, third edition, Academic Press, San Diego, 1998; METHODS IN ENZYMOLOGY, Vol. 304, Chromatin (P. M. Wassarman and A. P. Wolffe, eds.), Academic Press, San Diego, 1999; and METHODS IN MOLECULAR BIOLOGY, Vol. 119, Chromatin Protocols (P. B. Becker, ed.) Humana Press, Totowa, 1999.

Definitions

The terms nucleic acid, polynucleotide and oligonucleotide are used interchangeably and refer to a deoxyribonucleotide or ribonucleotide polymer in a linear or circular conformation and in single-stranded or double-stranded form. For the purposes of this disclosure, these terms should not be construed as limiting with respect to the length of the polymer. These terms may include known analogues of natural nucleotides, as well as nucleotides that are modified with a base moiety, sugar moiety, phosphate moiety (eg, phosphorothioate backbone), and combinations selected from base moiety, sugar moiety, and phosphate moiety. Typically, an analogue of a particular nucleotide has the same pairing specificity with another base; those. the analogue of A will be the base pair of T.

The terms polypeptide, peptide, and protein are used interchangeably to refer to a polymer of amino acid residues. The term also applies to amino acid polymers in which one or more amino acids are chemical analogues or modified derivatives of the corresponding naturally occurring amino acids.

Binding refers to the sequence-specific, non-covalent relationship between macromolecules (eg, between a protein and a nucleic acid). Not all components of a binding interaction need to be sequence specific (eg contacts with phosphate residues on the DNA backbone) if the interaction as a whole is sequence specific. Such interactions are typically characterized by a dissociation constant (K _d ) of 10 ^-6 M ^-1 or lower. Affinity refers to the strength of binding: increased binding affinity correlates with lower _Kd .

A binding protein is a protein that is capable of binding to another molecule. The binding protein may bind, for example, a DNA molecule (DNA-binding protein), an RNA molecule (RNA-binding protein), a protein molecule (protein-binding protein), or may bind to a combination of molecules selected from a DNA molecule, an RNA molecule, or a protein. . In the case of a protein-binding protein, it may bind to itself (to form homodimers, homotrimers, etc.), it may bind to one or more molecules of another protein or proteins, or it may bind to both itself and one or more more molecules of another protein or proteins. A binding protein may have more than one type of binding activity. For example, zinc finger proteins have DNA-binding, RNA-binding, and protein-binding activities.

- 12 045868

A zinc finger DNA-binding protein (or binding domain) is a protein or domain within a larger protein that binds DNA in a sequence-specific manner through one or more zinc fingers, which are regions of amino acid sequence within the binding domain whose structure is stabilized by coordination of zinc ion. The term zinc finger DNA binding protein is often abbreviated as zinc finger protein or ZFP.

A TALE DNA-binding domain or TALE is a polypeptide containing one or more TALE repeat domains/units. Repeat domains are involved in the binding of TALE to its recognized target DNA sequence. A single repeat unit (also called a repeat) is typically 33-35 amino acids in length and exhibits at least some sequence homology to other TALE repeat sequences within a naturally occurring TALE protein.

The zinc finger and TALE binding domains can be designed to bind to a predetermined nucleotide sequence, for example, by engineering (altering one or more amino acids) the recognition helix region of a naturally occurring zinc finger or TALE protein. Therefore, engineered DNA-binding proteins (zinc fingers or TALEs) are proteins that do not occur in nature. Non-limiting examples of methods for constructing DNA-binding proteins include design and selection. An engineered DNA binding protein is a protein not found in nature, the design/composition of which is primarily derived from rational criteria. Rational criteria for design include the application of substitution rules and computerized algorithms to process information in a database storing information about existing ZFP and/or TALE designs and binding data. See, for example, US Pat. No. 6,140,081; 6453242; 6534261 and 8585526; see also PCT publication No. WO 98/53058; WO 98/53059; WO 98/53060; WO 02/016536 and WO 03/016496.

A selected zinc finger protein, or TALE, is a non-naturally occurring protein whose production occurs as a consequence of an empirical process such as phage display, trap interaction, or hybrid selection. See, for example, US Pat. No. 5,789,538; 5925523; 6007988; 6013453; 6200759; 8586526; PCT Publication No. WO 95/19431; WO 96/06166; WO 98/53057; WO 98/54311; WO 00/27878; WO 01/60970, WO 01/88197, WO 02/099084.

TtAgo is a prokaryotic Argonaute protein that is believed to be involved in gene silencing. TtAgo comes from the bacterium Thermus thermophilus. See, for example, Swarts et al. ibid., G. Sheng et al. (2013) Proc. Natl. Acad. Sci. USA 111, 652). And the TtAgo system contains all the necessary components, including, for example, DNA guides for cutting using the TtAgo enzyme.

Recombination refers to the process of exchange of genetic information between two polynucleotides, including, but not limited to, non-homologous end joining (NHEJ) capture and homologous recombination. For purposes of this disclosure, the term homologous recombination (HR) refers to a specialized form of such exchange that occurs, for example, during the repair of double-strand breaks in cells using homology-driven repair mechanisms. This process requires nucleotide sequence homology, uses a donor molecule to template repair the target molecule (i.e., the one that has undergone a double-strand break), and is variously known as non-crossover gene conversion or short tract gene conversion) because it results in the transfer of genetic information from the donor to the target. Without wishing to be bound by any particular theory, such transfer may involve correction of heteroduplex DNA mismatch that is formed between the damaged target and the donor, synthesis-dependent strand annealing in which the donor is used to resynthesize genetic information that will become part of the target, related processes or a combination of the above. Such specialized HR often results in a change in the sequence of the target molecule such that part or all of the sequence of the donor polynucleotide is included in the target polynucleotide.

In the methods of the present invention, one or more targeting nucleases, as described herein, create a double-strand break (DSB) in a target sequence (eg, cellular chromatin) at a predetermined site. DSBs can lead to indels through either homologous repair or nonhomologous repair mechanisms. Deletions can include any number of base pairs. Likewise, insertions may include any number of base pairs, including, for example, the integration of a donor polynucleotide, optionally having homology to the nucleotide sequence in the region of the break. The donor sequence may be physically integrated or, alternatively, the donor polynucleotide is used as a template to repair the break through homologous recombination, resulting in the introduction of all or part of the nucleotide sequence as that of the donor into cellular chromatin. Thus, the first sequence in the cellular chromatin can be altered and, in certain embodiments, can be converted to a sequence present in the donor polynucleotide. Thus, the use of the terms replace or replacement can be understood as the replacement of one nucleotide

- 13 045868 sequence of another (i.e., replacement of a sequence in an informational sense) and does not necessarily require physical or chemical replacement of one polynucleotide by another.

In any of the methods described herein, additional pairs of zinc finger proteins or TALENs can be used for additional double-stranded cutting of additional target sites in the cell.

Any of the methods described herein can be used to introduce a donor of any size or to partially or completely inactivate one or more target sequences in a cell by targeted integration of a donor sequence that disrupts the expression of the gene(s) of interest. Cell lines with partially or completely inactivated genes are also proposed.

In any of the methods described herein, the exogenous nucleotide sequence (donor sequence or transgene) may contain sequences that are homologous but not identical to genomic sequences in the region of interest, thereby promoting homologous recombination to insert a non-identical sequence into the region of interest. Thus, in certain embodiments, portions of the donor sequence that are homologous to sequences in the region of interest exhibit from about 80 to 99% (or any integer therebetween) sequence identity with the genomic sequence being replaced. In other embodiments, the homology between the donor and genomic sequences is greater than 99%, for example, if only 1 nucleotide differs between the donor and genomic sequences per 100 contiguous base pairs. In some cases, a non-homologous portion of the donor sequence may contain sequences not present in the region of interest, so that new sequences are introduced into the region of interest. In these cases, the nonhomologous sequence is typically flanked by sequences of 50-1000 base pairs (or any integral value in between) or any number of base pairs in excess of 1000 that are homologous or identical to sequences in the region of interest. In other embodiments, the donor sequence is non-homologous to the first sequence and is inserted into the genome through non-homologous recombination mechanisms.

Cutting refers to the destruction of the covalent backbone of the DNA molecule. Cutting can be initiated in a variety of ways, including, but not limited to, enzymatic or chemical hydrolysis of the phosphodiester bond. Both single-strand cutting and double-strand cutting are possible, and double-strand cutting can occur as a result of two different single-strand cutting phenomena. Cutting DNA can result in either blunt ends or sticky ends. In certain embodiments, fusion polypeptides are used to specifically cut double-stranded DNA.

A cutting half-domain is a polypeptide sequence that, when combined with a second polypeptide (either the same or different), forms a complex having cutting activity (preferably double-strand cutting activity). The terms first and second cutting half-domains, + and - cutting half-domains, and right and left cutting half-domains are used interchangeably to refer to pairs of cutting half-domains that dimerize.

An engineered cutting half-domain is a cutting half-domain that has been modified to form obligate heterodimers with another cutting half-domain (eg, another engineered cutting half-domain). See also US Pat. Nos. 7,888,121; 7914796; 8034598; 8623618 and US Patent Publication No. 2011/0201055, incorporated herein by reference in their entirety.

The term sequence refers to a nucleotide sequence of any length, which may be DNA or RNA; can be linear, circular or branched and can be either single-stranded or double-stranded. The term donor sequence refers to a nucleotide sequence that is inserted into the genome. The donor sequence may be of any length, for example, from 2 to 100,000,000 nucleotides in length (or any integer value in between or greater), preferably from about 100 to 100,000 nucleotides in length (or any integer value in between), more preferably between about 2000 and 20,000 nucleotides in length (or anything in between) and even more preferably between about 5 and 15 kb. (or any value in between).

Chromatin is a nucleoprotein structure that makes up the cell genome. Cellular chromatin contains nucleic acid, mainly DNA, and protein, including histones and non-histone chromosomal proteins. Most eukaryotic cellular chromatin exists in the form of nucleosomes, where the nucleosome core contains approximately 150 base pairs of DNA associated with an octamer containing two each of histones H2A, H2B, H3 and H4; and linker DNA (of variable length depending on the organism) runs between the nucleosome cores. The histone H1 molecule is usually associated with linker DNA. For purposes of this disclosure, the term chromatin is intended to cover all types of cellular nucleoproteins, both prokaryotic and eukaryotic. Cell chromatin turn on

- 14 045868 looks at both chromosomal and episomal chromatin.

A chromosome is a complex of chromatin containing all or part of a cell's genome. A cell's genome is often characterized by its karyotype, which is the collection of all the chromosomes that make up the cell's genome. The genome of a cell may contain one or more chromosomes.

An episome is a replicating nucleic acid, nucleoprotein complex, or other structure comprising a nucleic acid that is not part of the cell's chromosomal karyotype. Examples of episomes include plasmids and some viral genomes.

An accessible region is a site in cellular chromatin at which a target site present in a nucleic acid can be bound by an exogenous molecule that recognizes the target site. Without wishing to be bound by any particular theory, it is believed that an accessible region is one that is not packaged into a nucleosomal structure. The distinctive structure of an accessible region can often be detected by its sensitivity to chemical and enzymatic probes, such as nucleases.

A target site or target sequence is a nucleic acid sequence that specifies the portion of the nucleic acid to which a molecule binds, given sufficient conditions for binding.

An exogenous molecule is a molecule that is not normally present in a cell, but can be introduced into the cell by one or more genetic, biochemical or other methods. Presence in a cell is normally determined taking into account the specific developmental stage and environmental conditions of the cell. Thus, for example, a molecule that is present only during embryonic muscle development is an exogenous molecule to the adult muscle cell. Likewise, a heat shock-induced molecule is an exogenous molecule to a cell that has not undergone heat shock. An exogenous molecule may contain, for example, a malfunctioning version of a malfunctioning endogenous molecule or a malfunctioning version of a normally functioning endogenous molecule.

An exogenous molecule may be, but is not limited to, a small molecule, such as one generated by a combinatorial chemical process, or a macromolecule, such as a protein, nucleic acid, carbohydrate, lipid, glycoprotein, lipoprotein, polysaccharide, any modified derivative of the foregoing molecules, or any complex comprising one or more of the above molecules. Nucleic acids include DNA and RNA and can be single-stranded or double-stranded; can be linear, branched or circular; and can be of any length. Nucleic acids include those capable of forming duplexes as well as triplex-forming nucleic acids. See, for example, US Patent Nos. 5,176,996 and 5,422,251. Proteins include, but are not limited to, DNA binding proteins, transcription factors, chromatin remodeling factors, methylated DNA binding proteins, polymerases, methylases, demethylases, acetylases, deacetylases, kinases, phosphatases, integrases, recombinases, ligases, topoisomerases, gyrases and helicases.

The exogenous molecule may be the same type of molecule as the endogenous molecule, such as an exogenous protein or nucleic acid. For example, the exogenous nucleic acid may contain an infecting viral genome, a plasmid or episome introduced into a cell, or a chromosome that is not normally present in the cell. Methods for introducing exogenous molecules into cells are known to those skilled in the art and include, but are not limited to, lipid-mediated transfer (i.e., liposomes, including neutral and cationic lipids), electroporation, direct injection, cell fusion, particle bombardment, phosphate coprecipitation calcium, DEAE-dextran-mediated transfer and viral vector-mediated transfer. An exogenous molecule may also be the same type of molecule as an endogenous molecule, but be derived from a species different from the species to which the cell belongs. For example, a human nucleic acid sequence may be introduced into a cell line originally derived from a mouse or hamster.

In contrast, an endogenous molecule is a molecule that is normally present in a specific cell at a specific stage of development under specific environmental conditions.

For example, the endogenous nucleic acid may comprise a chromosome, the genome of a mitochondria, chloroplast or other organelle, or a naturally occurring episomal nucleic acid. Additional endogenous molecules may include proteins such as transcription factors and enzymes.

As used herein, the term exogenous nucleic acid product includes both polynucleotide and polypeptide products, for example, transcription products (polynucleotides such as RNA) and translation products (polypeptides).

A fusion molecule is a molecule in which two or more subunit molecules are linked, preferably covalently. The subunit molecules may be molecules of the same chemical type or may be molecules of different chemical types. Examples of the first type of fusion molecule include, but are not limited to, fusion proteins (e.g., a fusion between a ZFP or TALE DNA binding domain and one or more activation domains) and nucleic acid fusions (e.g., nucleic acids encoding a fusion protein described above) . Examples of the second type of fusion molecule include, but are not limited to, a fusion between

- 15 045868 triplex-forming nucleic acid and polypeptide, and fusion of the minor groove binder and the nucleic acid.

Expression of the fusion protein in a cell may result from delivery of the fusion protein into the cell or delivery of a polynucleotide encoding the fusion protein into the cell, wherein the polynucleotide is transcribed and the transcript is translated to generate the fusion protein. Transsplicing, polypeptide cutting, and polypeptide ligation may also be involved in protein expression in the cell. Methods for delivering polynucleotides and polypeptides into cells are presented elsewhere in this specification.

A gene, for purposes of this disclosure, includes the region of DNA encoding the gene product (see below), as well as all regions of DNA that regulate the production of the gene product, whether such regulatory sequences are adjacent to coding sequences, transcribed sequences, and combinations of coding and transcribed sequences. Accordingly, a gene includes, but is not limited to, promoter sequences, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators, boundary elements, origins of replication, matrix attachment sites and regions locus control.

Gene expression refers to the conversion of information contained in a gene into a gene product. The gene product may be a direct product of gene transcription (e.g., mRNA, tRNA, rRNA, antisense RNA, microRNA, ribozyme, structural RNA, or any other type of RNA) or a protein produced by translation of the mRNA. Gene products also include RNAs, which are modified by processes such as capping, polyadenylation, methylation and editing, and proteins modified, for example, by methylation, acetylation, phosphorylation, ubiquitination, DCP-ribosylation, myristylation and glycosylation.

Modulation of gene expression refers to changes in the activity of a gene. Modulation of expression may include, but is not limited to, gene activation and gene repression. Genome editing (eg, cutting, alteration, inactivation, random mutation) can be used to modulate expression. Gene inactivation refers to any decrease in gene expression compared to a cell that does not include ZFP, TALE, or the CRISPR/Cas system as described herein. Thus, gene inactivation can be partial or complete.

Protected mRNA is one in which the mRNA has been altered in some way to improve the stability or translation of the mRNA. Examples of defenses include the use of replacement of up to 25% of cytidine and uridine residues with 2-thiouridine (s2U) and 5-methylcytidine (m5C). The resulting mRNA exhibits less immunogenicity and greater stability compared to the unmodified analog (see Karik0 et al. ((2012), Molecular Therapy, Vol. 16, No. 11, pages 1833-1844). Other changes include the addition of a so-called ARCA cap , which increases the translation of mRNA produced in vitro (see US patent No. 7074596).

A region of interest is any region of cellular chromatin, such as, for example, a gene or non-coding sequence within or adjacent to a gene, at which it is desired to bind an exogenous molecule. Linking can be for targeted DNA cutting, targeted recombination, and combinations of targeted DNA cutting and targeted recombination. The region of interest may be present, for example, in a chromosome, an episome, an organelle genome (eg, mitochondrial, chloroplast), or in an infecting viral genome. The region of interest may be within the coding region of a gene, within transcribed non-coding regions, such as, for example, leader sequences, tail sequences or introns, or within non-transcribed regions, either upstream or downstream of the coding region. The region of interest can be as small as one nucleotide pair, or up to 2000 nucleotide pairs long, or any integer value of nucleotide pairs.

Eukaryotic cells include, but are not limited to, fungal cells (such as yeast), plant cells, animal cells, mammalian cells, and human cells (such as T cells).

The terms functionally related and functionally related (or operatively related) are used interchangeably to refer to the mapping of two or more components (such as sequence elements) in which the components are arranged such that both components function normally and allow for the possibility that at least one of the components may mediate a function that affects at least one of the other components. By way of illustration, a transcriptional regulatory sequence, such as a promoter, is operably linked to a coding sequence if the transcriptional regulatory sequence controls the level of transcription of the coding sequence in response to the presence or absence of one or more transcriptional regulatory factors. A transcriptional control sequence is usually operably linked in cis to the coding sequence, but need not be directly adjacent to it. For example, an enhancer is a transcriptional regulatory sequence that is operably linked to a coding sequence even though they are not adjacent.

- 16 045868

With regard to fusion polypeptides, the term operably linked may refer to the fact that each of the components performs the same function in connection with the other component as if they were not so linked. For example, with respect to a fusion polypeptide in which a ZFP, TALE, or Cas DNA-binding domain is fused to an activation domain, the ZFP, TALE, or Cas DNA-binding domain and the activation domain are in a functional relationship if, in the fusion polypeptide, the ZFP, TALE, or Cas DNA-binding domain is capable of bind its target site, its binding site and combinations of the target site and binding site, and at the same time the activation domain is able to enhance gene expression. When a fusion polypeptide in which a ZFP, TALE, or Cas DNA-binding domain is fused to a cleavage domain, the ZFP, TALE, or Cas DNA-binding domain and the cleavage domain are in a functional relationship if the fusion polypeptide has a ZFP, TALE, or Cas DNA-binding domain is able to bind its target site, its binding site, and combinations of the target site and binding site, and at the same time the cutting domain is able to cleave DNA in close proximity to the target sites.

A functional fragment of a protein, polypeptide or nucleic acid is a protein, polypeptide or nucleic acid whose sequence is not identical to the full-length protein, polypeptide or nucleic acid, but retains the same function as the full-length protein, polypeptide or nucleic acid. A functional fragment may contain more, fewer, or the same number of residues as the corresponding native molecule, may contain one or more amino acid or nucleotide substitutions, and may be combinations containing more, fewer, or the same number of residues as the corresponding native molecule, and containing one or more amino acid or nucleotide substitutions. Methods for determining the function of a nucleic acid (eg, coding function, ability to hybridize with another nucleic acid) are well known in the art. Likewise, methods for determining protein function are well known. For example, the DNA-binding function of a polypeptide can be determined, for example, using filter binding, electrophoretic mobility, or immunoprecipitation assays.

DNA cutting can be analyzed using gel electrophoresis. See Ausubel et al., supra. The ability of a protein to interact with another protein can be determined, for example, by coimmunoprecipitation, two-hybrid assays, or complementation, either genetic or biochemical.

The vector is capable of transferring gene sequences into target cells. In general, vector construct, expression vector, and gene transfer vector mean any nucleic acid construct capable of directing the expression of a gene of interest and which can transfer gene sequences into target cells. Thus, the term includes cloning and expression vehicles as well as integration vectors.

The terms subject and patient are used interchangeably and refer to mammals such as human patients and non-human primates, as well as experimental animals such as pigs, cows, rabbits, dogs, cats, rats, mice and other animals. Accordingly, the term subject or patient as used herein means any patient or mammalian subject to whom the stem cells of this invention can be administered.

Subjects of the present invention include those who have been exposed to one or more chemical toxins, including, for example, a nerve toxin.

Stemness refers to the relative ability of any cell to act in a stem cell-like manner, i.e. to the degree of toti-, pluri-, multi- or oligopotency and enhanced or infinite capacity for self-renewal that any particular stem cell may have.

Nucleases

Described herein are compositions, in particular nucleases, that are useful for cutting in vivo a donor molecule carrying a transgene, and nucleases for cutting the genome of a cell so that the transgene is specifically integrated into the genome. In certain embodiments, one or more nucleases occur naturally. In other embodiments, one or more nucleases are not naturally occurring, i.e. modified in the DNA binding domain, the cutting domain and the combination of the DNA binding domain and the cutting domain. For example, the DNA binding domain of a naturally occurring nuclease can be modified to bind to a selected target site (eg, a meganuclease that has been engineered to bind to a site other than its natural binding site). In other embodiments, the nuclease comprises heterologous DNA binding and cutting domains (eg, zinc finger nucleases; TAL effector domain DNA binding proteins; meganuclease DNA binding domains with heterologous cutting domains).

A. DNA-binding domains

In some embodiments, the DNA binding domain of one or more nucleases used for in vivo cutting, targeted cell genome cutting, and combinations of in vivo cutting and targeted cell genome cutting comprises a zinc finger protein. One zinc protein

- 17 045868 zinc finger consists of several/many zinc finger domains (eg, 3, 4, 5, 6 or more zinc finger domains). Each zinc finger domain is approximately 30 amino acids long and contains a beta turn (containing two zinc coordination residues) and an alpha helix (containing two invariant zinc coordination residues) that are held in a specific conformation that allows the protein to bind the target sequence. Canonical (C2H2) zinc finger domains having two zinc coordinating cysteine residues (Cys) in the beta turn and two zinc coordinating histidine residues (His) in the alpha helix or non-canonical (CH3) can be used (see, for example, US Pat. No. 9234187). The 7-amino acid recognition helix is contained between the zinc-coordinating residues of the beta-helix and the zinc-coordinating residues of the alpha-helix. The recognition helix region is numbered -1 to +6 within the zinc finger domain, and amino acids outside this recognition region (and excluding zinc coordinating residues) are called backbone residues.

Preferably, the zinc finger protein is non-naturally occurring in that the recognition helix is designed to bind to a selected target site. See, for example, Beerli et al. (2002) Nature Biotechnol. 20:135-141; Pabo et al. (2001) Ann. Rev. Biochem. 70:313340; Isalan et al. (2001) Nature Biotechnol. 19:656-660; Segal et al. (2001) Curr. Opin. Biotechnol. 12:632-637; Choo et al. (2000) Curr. Opin. Struct. Biol. 10: 411-416; US Patent Nos. 6,453,242; 6534261; 6599692; 6503717; 6689558; 7030215; 6794136; 7067317; 7262054; 7070934; 7361635; 7253273; and US Patent Publications No. 2005/0064474; 2007/0218528; 2005/0267061, all of which are incorporated herein by reference in their entirety.

The engineered zinc finger binding domain may have a novel binding specificity compared to native zinc finger protein. Design methods include, but are not limited to, rational design and various types of selection. Rational design includes, for example, the use of databases containing triplet (or quadruplet) nucleotide sequences and individual zinc finger amino acid sequences, in which each triplet or quadruplet nucleotide sequence is associated with one or more zinc finger amino acid sequences that link the particular triplet sequence or quadruplet. See, for example, US Patent Nos. 6,453,242 and 6,534,261, incorporated herein by reference in their entirety.

Additionally, in certain embodiments, the ZFP DNA binding domains further comprise one or more backbone modifications of one or more component zinc finger domains. The specificity of ZFP to the target DNA sequence depends on sequence-specific contacts between zinc finger domains and specific DNA bases, in particular between the recognition helix region and the target site (usually each recognition helix binds to a 3-nucleotide daughter target site) . In addition, zinc finger domains also contain amino acid residues that participate in nonspecific interactions with DNA backbone phosphates. Elrod-Erickson et al. ((1996) Structure 4:1171), demonstrated by cocrystallizing the zinc finger protein and its cognate DNA target that there are specific amino acids that can interact with phosphates on the DNA backbone through hydrogen bonding. Zinc finger proteins that use the well-known Zif268 backbone typically have an arginine as the terminal amino acid residue in their second beta sheet chain, which is also the second carboxy terminal position to the second invariant cysteine. This position may be referred to as (−5) in each zinc finger domain because it is the 5th residue preceding the start of the α-helix (and position −5 relative to the recognition helix numbered −1 to +6). Arginine at this position can interact with a phosphate on the DNA backbone by forming a charged hydrogen bond with its guanidine group on the side chain. Zinc finger proteins in the Zif268 backbone also often contain a lysine at a position that is shifted 4 residues in the amino-terminal direction relative to the first invariant cysteine. This position can be designated as (-14) within each finger, since it is the 14th residue preceding the start of the α-helix for zinc fingers with two residues between the zinc-coordinating cysteine residues (and position -14 relative to the recognition helix region numbered -1 to + 6). Lysine can interact with a phosphate on the DNA backbone through the formation of a water-mediated charged hydrogen bond with its amino side group. Because phosphate groups are found along the entire length of the DNA backbone, this type of interaction between the zinc finger and the DNA molecule is generally considered to be non-sequence specific (J. Miller, Massachusetts Institute of Technology Ph. D. Thesis, 2002).

Recent studies have hypothesized that nonspecific phosphate contact side chains in some nucleases may also account for some of the nonspecificity of these nucleases (Kleintiver et al. (2016) Nature 529(7587):490-5; Guilinger et al. (2014) Nat Meth: 429-435). The researchers hypothesized that these nucleases may have excess DNA binding energy, meaning that these nucleases may have greater affinity for their target DNA than is required to substantially bind and cut the target site. Thus, attempts have been made

- 18 045868 reduce the cationic charges in the DNA-binding domain of TALE (Guilinger, ibid.) or in the DNA-binding domain of Cas9 (Kleinstiver, ibid.) to reduce the DNA binding energy of these nucleases, which led to increased cutting specificity in vitro. However, additional research (Sternberg et al. (2015) Nature 527 (7576): 110-113) also suggests a role in the proper folding and activation of the Cas9 nuclease domain for some cationic amino acids that were mutated in Kleinstiver's study of the Cas9 DNA-binding domain. Thus, the exact role of these amino acids in Cas9 activity is unknown.

The methods and compositions of the present invention thus involve amino acid mutations within the ZFP DNA binding domain (ZFP backbone) that may interact nonspecifically with phosphates in the DNA backbone, but they do not involve changes in DNA recognition helices. Thus, the invention includes mutations of cationic amino acid residues in the ZFP backbone that are not required for nucleotide target specificity. In some embodiments, these mutations in the ZFP backbone include mutation of a cationic amino acid residue to a neutral or anionic amino acid residue. In some embodiments, these mutations in the ZFP backbone include mutation of a polar amino acid residue to a neutral or non-polar amino acid residue. In preferred embodiments, mutations are made at position (-5), position (-9), position (-14) and a combination of mutations selected from mutations at position (-5), position (-9) and position (-14) relative to DNA binding helix. In some embodiments, the zinc finger may contain one or more mutations in (-5), (-9), (-14) and combinations of mutations selected from mutations in (-5), (-9) and (-14). In other embodiments, one or more zinc fingers in a multi-finger zinc finger protein may contain mutations in (-5), (-9), (-14), and combinations selected from (-5), (-9), and (-14). In some embodiments, amino acids at (-5), (-9), (-14) and combinations selected from (-5), (-9) and (-14) (for example, arginine (R) or lysine (K )) are mutated to alanine (A), leucine (L), Ser (S), Asp (D), Glu (E), Tyr (Y) and/or glutamine (Q).

In any of these fusion polypeptides described herein, the ZFP partners may further contain mutations in the zinc finger DNA binding domain at positions (-5), (-9), (14) and combinations of mutations selected from mutations at (- 5), (-9) and (-14). In some embodiments, the Arg (R) at position -5 is changed to Tyr (Y), Asp (D), Glu (E), Leu (L), Gln (Q), or Ala (A). In other embodiments, the Arg (R) at position (-9) is changed to Ser (S), Asp (D), or Glu (E). In other embodiments, Arg (R) at position (-14) is replaced by Ser (S) or Gln (Q). In other embodiments, the fusion polypeptides may contain mutations in the zinc finger DNA binding domain, where the amino acids are at positions (-5), (-9), (-14), and combinations of mutations selected from mutations at (-5), ( -9) and (-14) are replaced by any of the amino acids listed above in any combination.

Exemplary selection methods, including phage display and two-hybrid systems, are disclosed in US Pat. No. 5,789,538; 5925523; 6007988; 6013453; 6410248; 6140466; 6200759 and 6242568; and also WO 98/37186; WO 98/53057; WO 00/27878; WO 01/88197 and GB 2338237. In addition, enhancing binding specificity for zinc finger binding domains has been described, for example, in joint application WO 02/011221.

Selection of target sites; ZFPs and methods for designing and constructing fusion proteins (and polynucleotides encoding them) are known to those skilled in the art and are described in detail in US Pat. Nos. 6,140,081; 5789538; 6453242; 6534261; 5925523; 6007988; 6013453; 6200759; PCT publications No. WO 95/19431; WO 96/06166; WO 98/53057; WO 98/54311; WO 00/27878; WO 01/60970; WO 01/88197; WO 02/099084; WO 98/53058; WO 98/53059; WO 98/53060; WO 02/016536 and WO 03/016496.

Almost any linker (spacer) can be used between one or more components of a DNA-binding domain (eg, zinc finger), between one or more DNA-binding domains, between a DNA-binding domain and a functional domain (eg, nuclease), and between one or more DNA -binding domains and between the DNA-binding domain and the functional domain. Non-limiting examples of suitable linker sequences include US Pat. Nos. 8,772,453; 7888121; 6479626; 6903185 and 7153949; and US Patent Publications No. 2009/0305419; 2015/0064789 and 2015/0132269. Thus, the proteins described herein may include any combination of suitable linkers between individual DNA-binding components, between a DNA-binding domain and a functional domain, or between one or more DNA-binding domains, and between a DNA-binding domain and a functional domain of the compositions. described in this document.

B. Cutting domains

Any suitable cutting domain can be operably linked to DNA binding domains as described herein to form a nuclease. The cutting domain may be heterologous to the DNA binding domain, for example a zinc finger DNA binding domain and a nuclease cutting domain. Heterologous cutting domains can be derived from any endonuclease or exonuclease. Exemplary endonucleases from which a cutting domain can be derived include, but are not limited to, restriction endonucleases and homing endonucleases. See, for example, Catalog 2002-2003, New England Biolabs, Beverly, Massachusetts; And

- 19 045868

Belfort et al. (1997) Nucleic Acids Res. 25:3379–3388. Additional enzymes are known that cleave DNA (e.g., S1 nuclease; mung bean nuclease; pancreatic DNase I; micrococcal nuclease; yeast HO endonuclease; see also Linn et al. (Eds.)Nucleases, Cold Spring Harbor Laboratory Press, 1993) . One or more of these enzymes (or functional fragments thereof) can be used as a source of cutting domains and half-cutting domains.

Likewise, a cleavage half-domain can be derived from any nuclease or portion thereof, as defined above, that requires dimerization for cleavage activity. Typically, two fusion proteins are required for cutting if the fusion proteins contain cutting half-domains. Alternatively, a single protein containing two cutting half-domains can be used. The two cleavage half-domains may be derived from the same endonuclease (or functional fragments thereof), or each cleavage half-domain may be derived from a different endonuclease (or functional fragments thereof). In addition, the target sites for the two fusion proteins are preferably positioned relative to each other such that binding of the two fusion proteins to their respective target sites positions the cutting half-domains in spatial orientation to each other in a manner that allows the cutting half-domains to form a functional cutting domain. for example, by dimerization. Thus, in certain embodiments, the proximal edges of the target sites are separated by 5-8 nucleotides or 15-18 nucleotides. However, any integer number of nucleotides or nucleotide pairs may lie between two target sites (eg, 2 to 50 nucleotide pairs or more). In general, the cutting site lies between the target sites.

Restriction endonucleases (restriction enzymes) are present in many species and are capable of specifically binding to a DNA sequence (at the recognition site) and cleaving DNA at or near the binding site. Some restriction enzymes (for example, type IIS) cleave DNA at sites distant from the recognition site and have binding and cutting domains suitable for separation. For example, the Type IIS enzyme FokI catalyzes double-stranded DNA cleavage 9 nucleotides from its recognition site on one strand and 13 nucleotides from its recognition site on the other strand. See, for example, US Pat. No. 5,356,802; 5436150 and 5487994; and Li et al. (1992) Proc. Natl. Acad. Sci. USA 89:4275-4279; Li et al. (1993) Proc. Natl. Acad. Sci. USA 90:2764-2768; Kim et al. (1994a) Proc. Natl. Acad. Sci. USA 91:883-887; Kim et al. (1994b) J Biol. Chem. 269:31,978-31,982. Thus, in one embodiment, the fusion proteins comprise a cutting domain (or half-cutting domain) from at least one type IIS restriction enzyme and one or more zinc finger binding domains, which may or may not be modified.

For optimal cutting specificity with a sequence-selective (artificial) nuclease, it is desirable to ensure that binding and activity on the target are not saturable. Under saturation conditions, by definition, an excess of nuclease is used compared to the amount required to achieve full target activity. This excess does not provide an advantage when impacting the target, but may still result in increased cutting in non-target areas. For monomeric nucleases, saturation conditions can be easily avoided by performing a simple dose response study to identify and avoid a saturating plateau in the titration curve. However, for a dimeric nuclease such as ZFN, TALEN or dCas-Fok, identifying and avoiding saturating conditions can be more challenging if the binding affinities of the individual monomers are different. In such cases, a dose-response study using a simple 1:1 nuclease ratio will only show the saturation point of the weaker binding monomer. In this scenario, if, for example, the affinity for a monomer differs by a factor of 10, then at the saturation point determined in a 1:1 titration study, the monomer with the higher affinity will be present at a concentration that is 10 times higher than it should be present. The resulting excess of higher affinity monomer may in turn lead to increased off-target activity without providing any beneficial increase in cutting of the intended target, potentially leading to decreased overall specificity for any given pair of nucleases.

A typical type IIS restriction enzyme whose cutting domain is separable from its binding domain is FokI. This particular enzyme is active as a dimer. Bitinaite et al. (1998) Proc. Natl. Acad. Sci. USA 95:10, 570-10,575. Accordingly, for purposes of this disclosure, the portion of the FokI enzyme used in the described fusion proteins is considered to be a cutting half-domain. Thus, for targeted double-strand cutting, targeted replacement of cellular sequences using zinc finger-FokI fusions, and combinations of targeted double-strand cutting and targeted replacement of cellular sequences using zinc finger-FokI fusions, two fusion proteins each containing a half-FokI cutting domain can be used to repair the catalytically active cutting domain. Alternatively, one polypeptide molecule containing a zinc finger binding domain and two FokI cutting half-domains can also be used. Parameters for targeted cutting and targeted sequence alteration using zinc finger-FokI fusions are provided elsewhere in this disclosure.

A cutting domain or cutting half-domain can be any portion of a protein that retains cutting activity or that retains the ability to multimerize (e.g.

- 20 045868 dimerization) to form a functional cutting domain.

Exemplary Type IIS restriction enzymes are described in US Pat. No. 7,888,121, which is incorporated herein in its entirety. Additional restriction enzymes also contain separable binding and cutting domains and are provided herein (See, for example, Roberts et al. (2003) Nucleic Acids Res. 31:418-420).

In some embodiments, the cutting domain comprises one or more engineered cutting half-domains (also called dimerization domain mutants) that minimize or prevent homodimerization, as described, for example, in see, for example, US Pat. Nos. 7,914,796; 8034598 and 8623618, the disclosures of which are incorporated herein by reference in their entirety. Amino acid residues at positions 446, 447, 479, 483, 484, 486, 487, 490, 491, 496, 498, 499, 500, 531, 534, 537, and 538 of FokI are all targets for influencing the dimerization of FokI cutting hemidomains, with numbering refers to 1FOK crystal structures. pdb and 2FOK. pdb (see Wah et al. (1997) Nature 388:97-100) having the sequence shown below.

Wild-type FokI cutting half-domain (SEQ ID NO: 18)

QLVKSELEEKKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFMKVYGYRGKH

LGGSRKPDGAIYTVGS РIDYGVIVDTKAYS GGYNLPIGQADEMQRYVEENQTRNKHINPNEWWK

VYPSSVTEFKFLFVSGHFKGNYKAQLTRLNHITNCNGAVLSVEELLIGGEMIKAGTLTLEEVRR KFNNGEINF

Exemplary engineered FokI cutting half-domains that form obligate heterodimers include a pair in which the first cutting half-domain includes mutations at amino acid residues at positions 490 and 538 of FokI, and the second cutting half-domain includes mutations at amino acid residues 486 and 499.

Thus, in one embodiment, the mutation at 490 replaces Glu (E) with Lys (K); mutation in 538 replaces Ile (I) with Lys (K); mutation in 486 replaced Gln (Q) with Glu (E); and a mutation at position 499 replaces Ile (I) with Lys (K). Specifically, the engineered cleavage half-domains described herein were generated by mutating positions 490 (E^K) and 538 (I^K) in one cleavage half-domain to produce the engineered cleavage half-domain designated E490K:I538K, and by mutating positions 486 (Q^E) and 499 (I^L) in another cutting half-domain to obtain a constructed cutting half-domain designated Q486E:I499L.

The engineered cutting half-domains described herein are obligate heterodimeric mutants in which aberrant cutting by ZFN homodimers is minimized or eliminated. See, for example, US Patent Publication No. 2008/0131962, the disclosure of which is incorporated by reference in its entirety for all purposes. In certain embodiments, the engineered cleavage half-domain contains mutations at positions 486, 499, and 496 (numbered relative to wild-type FokI), such as mutations that replace the wild-type Gln(Q) residue at position 486 with a Glu(E) ), Ile(I) residue ) wild type at position 499 by the Leu (L) residue and the wild type Asn (N) residue at position 496 by the Asp (D) or Glu (E) residue (also referred to as the ELD and ELE domains, respectively). In other embodiments, the engineered cleavage half-domain contains mutations at positions 490, 538, and 537 (numbered relative to wild-type FokI), such as mutations that replace the wild-type Glu(E) residue at position 490 with a Lys(K)), an Ile(I) residue ) wild-type at position 538 by a Lys (K) residue and a wild-type His (H) residue at position 537 by a Lys (K) residue or an Arg (R) residue (also called KKK and KKR domains, respectively). In other embodiments, the engineered cleavage half-domain contains mutations at positions 490 and 537 (numbered relative to wild-type FokI), such as mutations that replace the wild-type Glu (E) residue at position 490 with a Lys (K) residue and a wild-type His (H) residue at position 537 by a Lys (K) residue or an Arg (R) residue (also called KIK and KIR domains, respectively). See, for example, US Pat. Nos. 7,914,796; 8034598 and 8623618. In other embodiments, the engineered cutting half-domain contains Sharkey, Sharkey mutations, and combinations of Sharkey and Sharkey mutations (see Guo et al. (2010) J. Mol. Biol. 400(1):96-107).

Thus, FokI-derived cutting half-domains may contain a mutation at one or more amino acid residues as shown in SEQ ID NO: 18, including mutations in the dimerization domain as described above; mutations in the catalytic domain, mutations in other amino acid residues, such as phosphate contact residues, and any combination of mutations selected from mutations in the dimerization domain, mutations in the catalytic domain, and mutations in other amino acid residues, such as phosphate contact residues. Mutations include substitutions (of a wild-type amino acid residue with another residue), insertions (of one or more amino acid residues), deletions (of one or more amino acid residues), and any combination of mutations selected from substitutions, insertions, and deletions. In some embodiments, one or more of residues 414-426, 443-450, 467-488, 501-502, 521531 (numbered relative to SEQ ID NO: 18) and any combination of such residues are mutated because those residues are located close to the DNA backbone in molecular model of ZFN associated with its

- 21 045868 target site described in Miller et al. ((2007) Nat Biotechnol 25:778–784). In certain embodiments, one or more residues at positions 416, 422, 447, 448, and 525 are mutated. In some embodiments, the mutation involves replacing a wild-type residue with another residue, such as a serine (S) residue. In certain embodiments, the FokI cleavage domain of the nucleases described herein comprises an ELD dimerization domain mutation, a KKR dimerization domain mutation, a K52 5S mutation, and any combination selected from an ELD dimerization domain mutation or a KKR dimerization domain mutation and a K525S mutation.

The engineered cleavage domains described herein can be produced using any suitable method, for example, by site-directed mutagenesis of wild-type (FokI) cleavage half-domains as described in US Pat. Nos. 7,888,121; 7914796; 8,034,598 and 8,623,618. Additionally, the cleavage domains described herein can be fused to a DNA binding domain (e.g., ZFP) using any suitable linker, including, but not limited to, the linkers described in US Pat. Nos. 9,394,531 and 9,567,609.

Alternatively, nucleases can be assembled in vivo at the target nucleic acid site using so-called split enzyme technology (see, for example, US Patent Publication No. 2009/0068164). Components of such split enzymes can be expressed either on separate expression constructs or can be linked in a single open reading frame where the individual components are separated, for example, by a 2A self-cutting peptide or an IRES sequence. The components may be individual zinc finger binding domains, or domains of a meganuclease nucleic acid binding domain.

Nucleases can be screened for activity before use, for example, in a yeast-based chromosomal system, as described in WO 2009/042163 and 20090068164. Nuclease expression can be under the control of a constitutive promoter or an inducible promoter, for example the galactokinase promoter, which is activated (derepressed) in the presence of raffinose, galactose, and a combination of raffinose and galactose and is repressed in the presence of glucose.

The nuclease(s) as described herein can produce one or more double-stranded, one or more single-stranded cuts, and combinations of one or more double-stranded and one or more single-stranded cuts at the target site. In some embodiments, the nuclease contains a catalytically inactive cutting domain (eg, FokI, Cas protein, and combinations of FokI and Cas protein). See, for example, US Pat. Nos. 9,200,266; 8703489 and Guillinger et al. (2014) Nature Biotech. 32(6):577-582. A catalytically inactive cutting domain can, in combination with a catalytically active domain, act as a nickase to form a single-stranded cut. Therefore, two nickases can be used in combination to make a double-stranded cut in a specific area.

Additional nickases are also known in the art, for example, McCaffery et al. (2016) Nucleic Acids Res. 44(2):e11. doi: 10.1093/nar/gkv878. Epub 2015 Oct 19.

Target sites

As described in detail above, DNA domains can be designed to bind to any selected sequence. The engineered DNA binding domain may have a new binding specificity compared to the natural DNA binding domain. In some embodiments, the DNA binding domains bind to a sequence in the BCL11A enhancer sequence, such as a target site (typically 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or even more base pairs) is located between exon 2 and exon 3 of BCL11A, including DNA-binding domains that bind to a sequence at the DNaseI hypersensitive site in the BCL11A enhancer sequence (eg, +58), as shown in Table. 1. Design methods include, but are not limited to, rational design and various types of selection. Rational design includes, for example, the use of databases containing triplet (or quadruplet) nucleotide sequences and individual zinc finger amino acid sequences, in which each triplet or quadruplet nucleotide sequence is associated with one or more zinc finger amino acid sequences that link the particular triplet sequence or quadruplet. See, for example, US Patent Nos. 6,453,242 and 6,534,261, incorporated herein by reference in their entirety. Rational design of TAL effector domains can also be achieved. See, for example, US Patent Publication No. 2011/0301073.

Exemplary selection methods applicable to DNA-binding domains, including phage display and two-hybrid systems, are disclosed in US Pat. No. 5,789,538; 5925523; 6007988; 6013453; 6410248; 6140466; 6200759 and 6242568; as well as PCT publication No. WO 98/37186; WO 98/53057; WO 00/27878; WO 01/88197 and UK Patent No. GB 2338237. In addition, enhancing binding specificity for zinc finger binding domains has been described, for example, in joint application WO 02/077227.

Selection of target sites; nucleases and methods for designing and constructing fusion proteins (and polynucleotides encoding them) are known to those skilled in the art and are described in detail in US Patent Publications Nos. 2005/0064474 and 2006/0188987, which are incorporated herein in their entirety.

- 22 045868 information by reference.

In addition, as disclosed in these and other references, DNA binding domains (eg, multifinger zinc finger proteins) and fusions of DNA binding domain(s) and functional domain(s) can be linked together using any suitable linker sequences, including, for example, linkers of 5 or more amino acids. US Patent Nos. 8,772,453; 7888121 (eg ZC linker); 6479626; 6903185 and 7153949; US Publication No. 2009/0305419) and 2015/0064789. The proteins described herein may include any combination of suitable linkers between individual DNA binding domains of the protein (see also US Pat. No. 8,586,526).

Donors

In certain embodiments, this disclosure relates to nuclease-mediated targeted integration of an exogenous sequence into the genome of a cell using the BCL11A enhancer region binding molecules described herein. As noted above, the insertion of an exogenous sequence (also called a donor sequence or donor or transgene), for example, to delete a specified region, to correct a mutant gene, to combine deletion of a specified region and correction of a mutant gene, or to increase expression of a wild-type gene. Obviously, the donor sequence is usually not identical to the genomic sequence into which it is inserted. The donor sequence may contain a nonhomologous sequence flanked by two regions of homology to provide efficient HDR at the location of interest, or may be integrated by nonhomologously directed repair mechanisms such as NHEJ. In addition, the donor sequences may contain a vector molecule containing sequences that are not homologous to the region of interest in the cellular chromatin. The donor molecule may contain multiple discontinuous regions of homology with cellular chromatin and, for example, result in deletion of the BCL11A enhancer region (or fragment thereof) when used as a substrate for DSB repair induced by one of the nucleases described herein. In addition, for the targeted insertion of sequences that are not normally present in the region of interest, the sequences may be present in the donor nucleic acid molecule and flanked by regions of homology to the sequence in the region of interest.

Insertion polynucleotides may also be referred to as exogenous polynucleotides, donor polynucleotides, or molecules or transgenes. The donor polynucleotide can be DNA or RNA, single-stranded or double-stranded, and can be introduced into the cell in linear or circular form. See, for example, US Patent Application Publications No. 2010/0047805 and 2011/0207221. The donor sequence(s) are preferably contained in the DNA-MS, which can be introduced into the cell in a circular or linear form. When introduced in linear form, the ends of the donor sequence can be protected (eg, from exonucleolytic degradation) by methods known to those skilled in the art. For example, one or more dideoxynucleotide residues are added to the 3' end of a linear molecule, and self-complementary oligonucleotides are optionally ligated to one or both ends. See, for example, Chang et al. (1987) Proc. Natl. Acad. Sci. USA84:4959-4963; Nehls et al. (1996) Science 272:886–889. Additional methods for protecting exogenous polynucleotides from degradation include, but are not limited to, the addition of terminal amino group(s) and the use of modified internucleotide compounds, such as, for example, phosphorothioates, phosphoramidates, and O-methyl ribose or deoxyribose residues. When administered in double-stranded form, the donor may include one or more nuclease target sites, for example, nuclease target sites flanking the transgene to be integrated into the genome of the cell. See, for example, US Patent Publication No. 2013/0326645.

The polynucleotide can be introduced into a cell as part of a vector molecule having additional sequences such as, for example, origins of replication, promoters and genes encoding antibiotic resistance. In addition, donor polynucleotides may be administered as naked nucleic acid, as nucleic acid complexed with an agent such as a liposome or poloxamer, or may be delivered by viruses (eg, adenovirus, AAV, herpesvirus, retrovirus, lentivirus, and integrase-deficient virus). lentivirus (IDLV)).

In some embodiments, the double-stranded donor includes sequences (e.g., coding sequences, also called transgenes) greater than 1 kb in length, e.g., between 2 and 200 kb, between 2 and 10 kb .O. (or any value in between). The double-stranded donor also includes, for example, at least one nuclease target site. In certain embodiments, the donor includes at least 2 target sites, for example for a ZFN or TALEN pair. Typically, nuclease target sites are located outside the transgene sequences, for example, 5' and/or 3' relative to the transgene sequences for cutting the transgene. The nuclease cleavage site(s) can be for any nuclease(s). In some embodiments, the target site(s) of the nuclease(s) contained in the double-stranded donor are targeted to the same nuclease(s) that are used to cut the endogenous target into which the cut donor is integrated in homology-independent ways.

- 23 045868

The donor is typically inserted such that its expression is driven by an endogenous promoter at the integration site, namely a promoter that drives the expression of the endogenous gene into which the donor is inserted (eg, globin, AAVS1, etc.). However, it will be appreciated that the donor may contain a promoter, an enhancer, and combinations of both promoter and enhancer, for example, a constitutive promoter or an inducible or tissue-specific promoter.

The donor molecule may be inserted into an endogenous gene such that all or part of the endogenous gene is not expressed. In other embodiments, the transgene (e.g., with or without globin coding sequences) is integrated into any endogenous locus, such as a safe harbor locus (see, for example, US Patent Publications Nos. 2008/0299580; 2008/0159996 and 2010/00218264) .

In addition, although not required for expression, exogenous sequences may also include transcriptional or translational regulatory sequences, such as promoters, enhancers, insulators, internal ribosomal entry sites, sequences encoding 2A peptides, polyadenylation signals, and combinations thereof.

Transgenes carried on the donor sequences described herein can be isolated from plasmids, cells, or other sources using standard techniques known in the art, such as PCR. Donors for use may include various types of topologies, including circular supercoiled, circular relaxed, linear, and the like. Alternatively, they can be chemically synthesized using standard oligonucleotide synthesis methods. Additionally, donors may or may not be methylated. Donors can be in the form of bacterial or yeast artificial chromosomes (BAC or YAC).

The double-stranded donor polynucleotides described herein may include one or more non-natural bases, one or more backbones, and combinations of one or more non-natural bases and one or more backbones. In particular, the insertion of a donor molecule with methylated cytosines can be accomplished using the methods described herein to achieve transcriptional quiescence in the region of interest.

The exogenous (donor) polynucleotide may contain any sequence of interest (exogenous sequence). Exemplary exogenous sequences include, but are not limited to, any polypeptide coding sequence (eg, cDNA), promoter sequences, enhancer sequences, epitope tags, marker genes, cutting enzyme recognition sites, and various types of expression constructs. Marker genes include, but are not limited to, sequences encoding proteins that confer antibiotic resistance (eg, ampicillin resistance, kanamycin resistance, neomycin resistance, G418 resistance, puromycin resistance, hygromycin resistance, blastidin resistance) , sequences encoding colored or fluorescent or luminescent proteins (eg, green fluorescent protein, enhanced green fluorescent protein, red fluorescent protein, luciferase) and proteins that promote enhanced cell growth and gene amplification (eg, dihydrofolate reductase) and combinations of enhanced cell growth and gene amplification. Epitope tags include, for example, one or more copies of FLAG, His, Myc, a tandem affinity purification (TAP) tag, HA, a biotinylatable peptide, or any detectable amino acid sequence.

In a preferred embodiment, the exogenous sequence (transgene) contains a polynucleotide encoding any polypeptide whose expression in a cell is desired, including, but not limited to, antibodies, antigens, enzymes, receptors (cellular or nuclear), hormones, lymphokines, cytokines, reporter polypeptides , growth factors and functional fragments of any of the above. The coding sequences may be, for example, cDNAs.

For example, the exogenous sequence may contain a sequence encoding a polypeptide that is absent or nonfunctional in a subject having a genetic disease, including, but not limited to, any of the following genetic diseases: achondroplasia, achromatopsia, acid maltase deficiency, adenosine deaminase deficiency (OMIM No. 102700 ), adrenoleukodystrophy, Ecardi syndrome, alpha-1-antitrypsin deficiency, alpha thalassemia, androgen insensitivity syndrome, Apert syndrome, arrhythmogenic right ventricular, dysplasia, ataxia telangictasia, Barth syndrome, beta thalassemia, blue vesicular nevus syndrome, Canavan disease, chronic granulomatous diseases (CGD), Cri de Cat syndrome, cystic fibrosis, Dercum's disease, ectodermal dysplasia, Fanconi anemia, progressive fibrodysplasia, Martin syndrome, galactosemia, Gaucher disease, generalized gangliosidosis, e.g. GM1), hemoglobin C mutation at codon ⁶ beta-globin (HbC), hemophilia, Huntington's disease, Hurler's syndrome, hypophosphatasia, Klineflether syndrome, Krabbe disease, Langer-Giedion syndrome, leukocyte adhesion deficiency (LAD, OMIM no. 116920), leukodystrophy, long QT syndrome, Marfan syndrome, syndrome Moebius, mucopolysaccharidosis (MPS), patella nail syndrome, nephrogenic diabetes insipidus, neurofibromatosis, Neumann-Pick disease, brittle bone syndrome, porphyria, Prader-Willi syndrome, progeria,

- 24 045868 Proteus syndrome, retinoblastoma, Rett syndrome, Rubinstein-Taybi syndrome, Sanfilippo syndrome, severe combined immunodeficiency syndrome (SCID), Shwachman syndrome, sickle cell disease (sickle cell anemia), Smith-Magenis syndrome, Stickler syndrome, Tay-Sachs disease, thrombocytopenia syndrome with absent radius (TAR), Trisher-Collins syndrome, trisomy, tuberous sclerosis, Turner syndrome, urea cycle disorder, Hippel-Lindau disease, Wardenburg syndrome, Williams syndrome, Wilson disease, Wiskott-Aldrich syndrome, X-linked lymphoproliferative syndrome ( XLP, OMIM No. 308240).

Additional illustrative diseases that may be treated by targeted integration include acquired immunodeficiencies, lysosomal storage diseases (eg, Gaucher disease, GM1, Fabry disease, and Tay-Sachs disease), mucopolysacchidosis (eg, Hunter disease, Hurler disease), hemoglobinopathies (eg, sickle cell anemia, HLS, α-thalassemia, β-thalassemia) and hemophilia.

In some embodiments, the exogenous sequences may comprise a marker gene (described above) allowing selection of cells that have undergone targeted integration, and an associated sequence encoding additional functionality. Non-limiting examples of marker genes include GFP, antibiotic selection marker(s), and the like.

Additional gene sequences that may be inserted may include, for example, wild-type genes to replace mutated sequences. For example, a wild-type factor IX gene sequence can be inserted into the genome of a stem cell in which the endogenous copy of the gene is mutated. The wild-type copy may be inserted at an endogenous locus or, alternatively, may be targeted to a safe harbor locus.

The construction of such expression cassettes, following the instructions of this description, uses methodologies well known in the field of molecular biology (see, for example, Ausubel or Maniatis). Before using an expression cassette to create a transgenic animal, the sensitivity of the expression cassette to the stress inducer associated with selected control elements can be tested by introducing the expression cassette into a suitable cell line (eg, primary cells, transformed cells, or immortalized cell lines).

In addition, although not required for expression, exogenous sequences may also contain transcriptional or translational regulatory sequences, such as promoters, enhancers, insulators, internal ribosome entry sites, sequences encoding 2A peptides, polyadenylation signals, and combinations of 2A polypeptides and polyadenylation signals . In addition, control elements of genes of interest can be operably linked to reporter genes to create chimeric genes (eg, reporter expression cassettes).

Targeted insertion of a non-coding nucleic acid sequence can also be accomplished.

Sequences encoding antisense RNAs, RNAi, shRNAs (shRNAs) and microRNAs (miRNAs) can also be used for targeted insertions.

In additional embodiments, the donor nucleic acid may contain non-coding sequences that are specific target sites for additional nuclease constructs. Subsequently, additional nucleases can be expressed in cells such that the original donor molecule is cleaved and modified by insertion of another donor molecule of interest. In this way, iterative integrations of donor molecules can be generated, allowing traits to be stacked at a specific locus of interest or a safe harbor locus.

Delivery

Nucleases as described herein (Table 1), polynucleotides encoding these nucleases, donor polynucleotides and compositions containing proteins, polynucleotides and combinations of proteins and polynucleotides described herein can be delivered in vivo or ex vivo by any suitable method to any cell type.

Suitable cells include eukaryotic (eg animal) and prokaryotic cells and eukaryotic and prokaryotic cell lines. Non-limiting examples of such cells or cell lines derived from such cells include COS, CHO (eg, CHO-S, CHO-K1, CHODG44, CHO-DUXBll, CHO-DUKX, CHOK1SV), VERO, MDCK, WI38, V79, B14AF28 -G3, BHK, HaK, NSO, SP2/0-Ag14, HeLa, HEK293 (e.g. HEK293-F, HEK293-H, HEK293-T) and perC6 cells, as well as insect cells such as Spodoptera fugiperda (Sf) or fungal cells such as Saccharomyces, Pichia and Schizosaccharomyces cells. In some embodiments, the cell line is a CHO, MDCK, or HEK293 cell line. Suitable cells also include stem cells such as, for example, embryonic stem cells, induced pluripotent stem cells, hematopoietic stem cells, neuronal stem cells and mesenchymal stem cells.

- 25 045868

Methods for delivering nucleases as described herein are described, for example, in US Pat. Nos. 6,453,242; 6503717; 6534261; 6599692; 6607882; 6689558; 6824978; 6933113; 6979539; 7013219 and 7163824, all of which are incorporated herein by reference in their entirety.

Nucleases, donor constructs, and combinations of nucleases and donor constructs, as described herein, can also be delivered using vectors containing sequences encoding one or more ZFNs described herein. Any vector systems can be used, including, but not limited to, plasmid vectors, retroviral vectors, lentiviral vectors, adenoviral vectors, poxviral vectors; herpesvirus vectors and adeno-associated viral vectors, etc. See also US patents No. 6534261; 6607882; 6824978; 6933113; 6979539; 7013219 and 7163824, which are incorporated herein by reference in their entirety. In addition, it will be appreciated that any of these vectors may contain one or more sequences necessary for treatment. Thus, when one or more nucleases and a donor construct are introduced into a cell, the nucleases, donor polynucleotide, and combinations of nucleases and donor polynucleotide can be carried on the same vector or on different vectors (DNA MS). When multiple vectors are used, each vector may contain sequence encoding one or more nucleases, one or more donor constructs, and combinations of one or more nucleases and one or more donor constructs. Conventional viral and non-viral gene transfer methods can be used to introduce nucleic acids encoding nucleases, donor constructs, and combinations of nucleases and donor constructs into cells (eg, mammalian cells) and target tissues. Delivery systems not based on viral vectors include DNA or RNA plasmids, DNA-MS, naked nucleic acid, and nucleic acid complexed with a delivery vehicle such as a liposome or poloxamer. Suitable non-viral vectors include nanotaxis vectors, including those commercially available from InCellArt (France). Viral vector delivery systems include DNA and RNA viruses that have either episomal or integrated genomes after delivery into the cell. For in vivo delivery of engineered DNA-binding proteins and fusion proteins containing these binding proteins, see, for example, Rebar (2004) Expert Opinion Invest. Drugs 13(7):829–839; Rossi et al. (2007) Nature Biotech. 25(12):1444-1454 as well as general articles on gene delivery Anderson, Science 256:SOS-SIS (1992); Nabel & Feigner, TIBTECH 11:211-217 (1993); Mitani & Caskey, TIBTECH 11:162-166 (1993); Dillon, TIBTECH 11:167-175 (1993); Miller, Nature 357:455-460 (1992); Van Brunt, Biotechnology 6(10):11491154 (1988); Vigne, Restorative Neurology and Neuroscience 8:35-36 (1995); Kremer & Perricaudet, British Medical Bulletin 51(1):31-44 (1995); Haddada et al., in Current Topics in Microbiology and Immunology Doerfler and Bohm (eds.) (1995); and Yu et al., Gene Therapy 1:13-26 (1994).

Methods for non-viral delivery of nucleic acids include electroporation, lipofection, microinjection, biolistics, virosomes, liposomes, immunoliposomes, polycation or lipid:nucleic acid conjugates, naked DNA, artificial virions, membrane deformation and agent-enhanced DNA uptake. Sonoporation using, for example, the Sonitron 2000 system (Rich-Mar) can also be used to deliver nucleic acids.

Additional exemplary nucleic acid delivery systems include those available from Amaxa Biosystems (Cologne, Germany), Maxcyte, Inc. (Rockville, MD), BTX Molecular Delivery Systems (Houston, MA) and Copernicus Therapeutics Inc. (see, for example, US 6008336). Lipofection is described, for example, in US patent No. 5049386; 4946787 and 4897355 and lipofection reagents are commercially available (eg, Transfectam™ and Lipofectin™). Cationic and neutral lipids that are suitable for effective receptor-mediated lipofection of polynucleotides include Feigner lipids, WO 91/17424, WO 91/16024.

The preparation of lipid:nucleic acid complexes, including target liposomes such as immunolipid complexes, is well known to one skilled in the art (see, for example, Crystal, Science 270: 404-410 (1995); Blaese et al., Cancer Gene Ther. 2:291-297 (1995); Behr et al., Bioconjugate Chem. 5:382-389 (1994); Remy et al., Bioconjugate Chem. 5:647-654 (1994); Gao et al., Gene Therapy 2:710722 (1995), Ahmad et al., Cancer Res. 946787). Other lipid:nucleic acid complexes include complexes comprising novel cationic lipids, novel PEGylated lipids, and combinations of novel cationic lipids and novel PEGylated lipids (see, for example, US Provisional Application Nos. 62/432,042 and 62/458,373).

Additional delivery methods include the use of nucleic acid packaging for delivery into EnGeneIC delivery vehicles (EDVs). These EDVs are specifically delivered to target tissues using bispecific antibodies, with one arm of the antibody being specific for the target tissue and the other being specific for the EDV. The antibody delivers EDV to the surface of the target cell, and then EDV is introduced into the cell by endocytosis. Once inside the cell, the contents are released (see MacDiarmid et al. (2009) Nature Biotechnology 27(7):643).

Use of RNA or DNA virus-based systems for delivery of nucleic acids

- 26 045868 engineered systems ZFP, TALE and CRISPR/Cas take advantage of well-developed processes for targeting the virus to specific cells in the body and transferring the cargo the virus carries into the nucleus. Viral vectors can be administered directly to patients (in vivo) or they can be used to treat cells in vitro and the modified cells are administered to patients (ex vivo). Common viral systems for delivering ZFP include, but are not limited to, retroviral, lentiviral, adenoviral, adeno-associated viral vectors, vaccinia virus vectors, and herpes simplex virus vectors for gene transfer. Integration into the host genome is possible using gene transfer methods using retroviruses, lentiviruses and adeno-associated viruses, which often leads to long-term expression of the inserted transgene. In addition, high transduction efficiency was observed in many different cell types and target tissues.

Retrovirus tropism can be altered by the incorporation of foreign envelope proteins, thereby expanding the potential target cell population. Lentiviral vectors are retroviral vectors that are capable of transducing or infecting nondividing cells and typically produce high viral titers. The choice of retroviral gene transfer system depends on the target tissue. Retroviral vectors consist of cis-acting long terminal repeats with packaging capacity of up to 6-10 kb. foreign sequence. Minimal cis-acting LTRs are sufficient to replicate and package vectors, which are then used to integrate the therapeutic gene into the target cell to ensure persistent transgene expression. Commonly used retroviral vectors include those based on murine leukemia virus (MuLV), gibbon leukemia virus (GaLV), simian immunodeficiency virus (SIV), human immunodeficiency virus (HIV), and combinations thereof (see, for example, Buchscher et al. , J. Virol. 66:2731-2739 (1992); Johann et al., J. Virol. 66:1635-1640 (1992); Sommerfelt et al., Virol. 176:58-59 (1990); Wilson et al., J. Virol. 63:2374-2378 (1989); Miller et al., J. Virol. 65:2220-2224 (1991); PCT/US 94/05700).

For applications where transient expression is preferred, adenovirus-based systems can be used. Adenovirus-based vectors are capable of very high transduction efficiency in many cell types and do not require cell division. With such vectors, high titers and high expression levels were obtained. This vector can be produced in large quantities in a relatively simple system. Adeno-associated virus (AAV) vectors are also used to transduce cells with target nucleic acids, such as in vitro production of nucleic acids and peptides, and for in vivo and ex vivo gene therapy procedures (see, e.g., West et al., Virology 160 : 38-47 (1987); US Patent No. 4797368; WO 93/24641; Kotin Human Therapy 5: 793-801 (1994); Muzyczka, J. Clin. Invest. 94:1351 (1994) Construction of recombinant AAV vectors described in a number of publications, including US Pat. No. 5,173,414; Tratschin, et al., Mol. Cell. Biol. 5:3251-3260 (1985); Tratschin, et al., Mol. Cell. Biol. 4:2072-2081 (1984); Hermonat & Muzyczka, PNAS 81:6466-6470 (1984); and Samulski et al., J. Virol. 63:03822-3828 (1989).

At least six viral vector approaches are currently available for gene transfer in clinical trials, which use approaches that involve complementation of defective vectors with genes inserted into helper cell lines to generate the transducing agent.

pLASN and MFG-S are examples of retroviral vectors that have been used in clinical trials (Dunbar et al., Blood 85: 3048-305 (1995); Kohn et al., Nat. Med. 1: 1017-102 (1995); Malech et al., PNAS 94:22 12133-12138 (1997)). PA317/pLASN was the first therapeutic vector used in a gene therapy trial. (Blaese et al., Science 270:475-480 (1995)). Transduction efficiencies of 50% or greater have been observed for packaged MFG-S vectors (Ellem et al., Immunol Immunother. 44(1):10-20 (1997); Dranoff et al., Hum. Gene Ther. 1:111-2 ( 1997).

Recombinant adeno-associated virus vectors (rAAV) are promising alternative gene delivery systems based on the defective and non-pathogenic adeno-associated virus type 2, a member of the parvovirus family. All vectors are derived from a plasmid that contains only 145-bp AAV inverted terminal repeats flanking the transgene expression cassette. Efficient gene transfer and stable delivery of transgenes through integration into the genomes of the transduced cell are key characteristics of this vector system. (Wagner et al., Lancet 351:9117 1702-3 (1998), Kearns et al., Gene Ther. 9:748-55 (1996)). Other AAV serotypes including AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9 and AAVrh. 10 and any new AAV serotype can also be used in accordance with this invention.

Replication-incompetent recombinant adenovirus (Ad) vectors can be produced at high titers and readily infect a number of different cell types. Most adenoviral vectors are designed so that the transgene replaces the Ad E1a, E1b, or E3 genes; the replication-defective vector subsequently replicates in 293 human cells that provide the function of the deleted genes in trans. Ad vectors can transduce various tissue types in vivo, including nondividing, differentiated cells such as liver, kidney, and muscle cells. Conventional Ad vectors have a high load-bearing capacity. An example of the use of an Ad vector in clinical trials included polynucleotide therapy for antitumor immunization using

- 27 045868 intramuscular injection (Sterman et al., Hum. Gene Ther. 7:1083-9 (1998)). Additional examples of the use of adenoviral vectors for gene transfer in clinical trials include Rosenecker et al., Infection 24:1 5-10 (1996); Sterman et al., Hum. Gene Ther. 9:7 1083-1089 (1998); Welsh et al., Hum. Gene Ther. 2:205-18 (1995); Alvarez et al., Hum. Gene Ther. 5:597-613 (1997); Topf et al., Gene Ther. 5:507-513 (1998); Sterman et al., Hum. Gene Ther. 7:1083-1089 (1998).

Packaging cells are used to form viral particles that are capable of infecting a host cell. Such cells include HEK293 and Sf9 cells, which can be used to package AAV and adenovirus, and ψ2 cells or PA317 cells, which package retrovirus. Viral vectors used in gene therapy are typically generated by a producer cell line that packages a nucleic acid vector into a viral particle. Vectors typically contain the minimal viral sequences necessary for packaging and subsequent integration into the host (if applicable), with other viral sequences being replaced by an expression cassette encoding the protein to be expressed. The missing viral functions are provided in trans by the packaging cell line. For example, AAV vectors used in gene therapy typically contain only inverted terminal repeat (ITR) sequences from the AAV genome, which are required for packaging and integration into the host genome. The viral DNA is packaged into a cell line that contains a helper plasmid encoding other AAV genes, namely rep and cap, but without the ITR sequences. The cell line is also infected with adenovirus as a helper. The helper virus promotes replication of the AAV vector and expression of AAV genes from the helper plasmid. The helper plasmid is not packaged in significant quantities due to the lack of ITR sequences.

Adenovirus contamination can be reduced, for example, by heat treatment, to which adenovirus is more sensitive than AAV. In some embodiments, AAV is produced using a baculovirus expression system (see, for example, US Pat. Nos. 6,723,551 and 7,271,002).

Purification of AAV particles from the 293 or baculovirus system typically involves growing the cells that produce the virus, followed by collection of the virus particles from the cell supernatant, or lysing the cells and collecting the virus from the crude lysate. AAV is then purified using methods known in the art, including ion exchange chromatography (for example, see US patent Nos. 7419817 and 6989264), ion exchange chromatography and CsCl density gradient centrifugation (for example, PCT publication WO 2011094198A10), immunoaffinity chromatography ( eg WO 2016128408) or purification using AVB Sepharose (eg GE Healthcare Life Sciences).

In many gene therapy applications, it is desirable that the gene therapy vector be delivered with a high degree of specificity to a particular tissue type. Accordingly, the viral vector can be modified to provide specificity for a given cell type by expressing a ligand as a fusion protein with the viral envelope protein on the outer surface of the virus. A ligand is selected that has an affinity for a receptor known to be present in the cell type of interest. For example, Han et al., Proc. Natl. Acad. Sci. USA 92: 9747-9751 (1995), reported that the Moloney murine leukemia virus can be modified to express human heregulin fused to gp70, and this recombinant virus infects certain human breast cancer cells expressing human epidermal growth factor receptor. This principle can be extended to other virus-target cell pairs, in which the target cell expresses the receptor and the virus expresses a fusion protein containing a ligand for the receptor on the cell surface. For example, a filamentous phage can be engineered to display antibody fragments (eg, FAB or Fv) that have a specific binding affinity for virtually any chosen cellular receptor. Although the above description applies primarily to viral vectors, the same principles can be applied to non-viral vectors. Such vectors can be designed to contain specific uptake sequences that promote uptake into specific target cells.

Gene therapy vectors can be delivered in vivo by application to an individual patient, typically by systemic application (eg, intravenous, intraperitoneal, intramuscular, subcutaneous or intracranial infusion) or topical application, as described below. Alternatively, vectors can be delivered to cells ex vivo, such as cells explanted from an individual patient (eg, lymphocytes, bone marrow aspirates, tissue biopsy) or universal donor hematopoietic stem cells, followed by reimplantation of the cells into the patient, typically after selection for cells that turned on the vector.

Vectors (eg, retroviral, adenoviral, liposomal, etc.) containing nucleases, donor constructs, and combinations of nucleases and donor constructs can also be administered directly into the body to transduce cells in vivo. Alternatively, naked DNA can be introduced. Administration is by any route commonly used to bring the molecule into final contact with blood or tissue cells, including, but not limited to, injection, infusion, topical application

- 28 045868 tion and electroporation. Suitable routes for administering such nucleic acids are available and well known to those skilled in the art, and although more than one route may be used to administer a particular composition, a particular route can often provide a more immediate and more efficient response than another route.

Vectors suitable for introducing polynucleotides (eg, nuclease-encoding, double-stranded donors, and combinations of nuclease-encoding and double-stranded donors) described herein include non-integrating lentiviral vectors (IDLVs). See, for example, Ory et al. (1996) Proc. Natl. Acad. Sci. USA 93:11382-11388; Dull et al. (1998) J. Virol. 72:8463-8471; Zuffery et al. (1998) J. Virol. 72:9873-9880; Follenzi et al. (2000) Nature Genetics 25: 217-222; US Patent Publication No. 2009/0117617.

Pharmaceutically acceptable carriers are determined in part by the particular composition administered as well as the particular method used to administer the composition. Accordingly, there is a wide variety of suitable pharmaceutical compositions available, as described below (see, for example, Remington's Pharmaceutical Sciences, ^17th ed., 1989).

It will be appreciated that nuclease coding sequences and donor constructs can be delivered using the same or different systems. For example, nucleases and donors can be transferred by the same MS DNA. Alternatively, the donor polynucleotide can be transferred using MS, while one or more nucleases can be transferred using a standard AAV plasmid or vector. In addition, different vectors can be administered by the same or different routes (intramuscular injection, tail vein injection, other intravenous injection, intraperitoneal injection, or intramuscular injection). Vectors can be delivered simultaneously or in any sequential order.

Thus, this disclosure includes the in vivo or ex vivo treatment of diseases and conditions that are amenable to the insertion of transgenes encoding a therapeutic protein. The compositions are administered to a human patient in an amount effective to obtain the desired concentration of the therapeutic polypeptide in the serum or target organ or cells. Administration can be accomplished by any means in which the polynucleotides are delivered to the desired target cells. For example, both in vivo and ex vivo methods are contemplated. Intravenous injection into the portal vein is the preferred route of administration. Other routes of in vivo administration include, for example, direct injection into the liver lobes or bile duct and intravenous injection distal to the liver, including through the hepatic artery, direct injection into the liver parenchyma, injection through the hepatic artery, and retrograde injection through the biliary ducts Methods of ex vivo administration include in vitro transduction of excised hepatocytes or other liver cells followed by infusion of the transduced excised hepatocytes back into the portal vasculature, liver parenchyma or biliary tree of a human patient, see, for example, Grossman et al. (1994) Nature Genetics, 6: 335-341.

The effective amount of nuclease(s) and donor administered will vary from patient to patient and depending on the therapeutic polypeptide of interest.

Accordingly, effective amounts are best determined by the physician administering the compositions, and suitable dosages can be readily determined by one skilled in the art. After offering sufficient time for integration and expression (usually, for example, 4-15 days), analysis of therapeutic polypeptide levels in serum or other tissues and comparison with the pre-administration baseline will determine whether the amount administered is too low, in the correct range, or too high. Suitable schedules for initial and subsequent administrations are also variable, but are categorized by initial administration followed by subsequent administrations if necessary. Subsequent administrations may occur at varying intervals, from daily to annual to every few years. One skilled in the art will appreciate that appropriate immunosuppressive techniques may be recommended to avoid inhibition or blocking of transduction with immunosuppressive delivery vectors, see, for example, Vilquin et al., (1995) Human Gene Ther., 6:1391- 1401.

Compositions for administration both ex vivo and in vivo include suspensions in liquids or emulsified liquids. Active ingredients are often mixed with excipients that are pharmaceutically acceptable and compatible with the active ingredient. Suitable excipients include, for example, water, saline, dextrose, glycerol, ethanol or the like, and combinations thereof. In addition, the composition may contain minor amounts of auxiliary substances, such as wetting or emulsifying agents, pH-buffering agents, stabilizing agents or other reagents that increase the effectiveness of the pharmaceutical composition.

Cells

Also described herein are cells and cell lines in which the endogenous BCL11A enhancer sequence is modified by the nucleases described herein (Table 1). The modification may be, for example, compared to the sequence of a wild-type cell. The cell or cell lines may be heterozygous or homozygous for modification. Modifications to the BCL11A sequence may include indels.

- 29 045868

The modification preferably occurs at or near the nuclease(s) binding site(s), cleavage site(s), and combinations of binding site(s) and cleavage site(s), e.g., within 1300 (or any value in between) base pairs upstream or downstream of the cleavage site(s), more preferably within 1-100 base pairs (or any value therebetween) on either side of the binding site(s), cleavage site(s), or binding site(s) and site cutting (s), even more preferably within 1 to 50 base pairs (or any value in between) on either side of the binding site(s), cutting site(s), or binding site(s) and cutting site(s) . In some embodiments, the modification is at or near the +58 region of the BCL11A enhancer, for example, at or near the nuclease binding site shown in any cell of the first column of the table. 1.

Any cell or cell line can be modified, for example, a stem cell, such as an embryonic stem cell, an induced pluripotent stem cell, a hematopoietic stem cell, a neuronal stem cell, and a mesenchymal stem cell. Other non-limiting examples of cells as described herein include T cells (eg, CD4+, CD3+, CD8+, etc.); dendritic cells; B cells. A descendant of a stem cell, including a partially or fully differentiated cell (eg, an RBC progenitor cell or RBC) is also provided. Non-limiting examples of other cell lines including the modified BCL11A sequence include COS, CHO (eg, CHO-S, CHO-K1, CHO-DG44, CHO-DUXB11, CHO-DUKX, CHOK1SV), VERO, MDCK, WI38, V79, B14AF28 -G3, BHK, HaK, NSO, SP2/0-Agl4, HeLa, HEK293 (e.g. HEK293-F, HEK293-H, HEK293-r) and perC6 cells, as well as insect cells such as Spodoptera fugiperda (Sf), or fungal cells such as Saccharomyces, Pichia and Schizosaccharomyces.

The cells described herein are useful for treating or preventing a disorder, for example, through ex vivo therapy. Nuclease-modified cells can be expanded and then reintroduced to the patient using standard techniques (see, for example, Tebas et al. (2014) New Eng J Med 370(10):901). In the case of stem cells, after infusion into the subject, the in vivo differentiation of these progenitors into cells expressing a functional transgene also occurs. Pharmaceutical compositions containing cells as described herein are also provided. In addition, the cells can be cryopreserved prior to administration to the patient.

Any of the modified cells or cell lines disclosed herein may exhibit increased gamma globin expression. Compositions, such as pharmaceutical compositions, containing genetically modified cells, as described herein, are also provided

Applications

The methods and compositions disclosed herein are intended to modify the expression of a protein or to correct an aberrant gene sequence that encodes a protein expressed in a genetic disease such as sickle cell disease or thalassemia. Thus, the methods and compositions provide treatment or prevention of such genetic diseases. Genome editing, such as in stem cells, can be used to correct an aberrant gene, insert a wild-type gene, or alter the expression of an endogenous gene. As a non-limiting example, a wild-type gene, e.g., encoding at least one globin (e.g., α-globin, γ-globin, β-globin, and combinations thereof), can be inserted into cells (e.g., into the endogenous BCL11A enhancer sequence with using one or more nucleases as described herein) to ensure that globin proteins are present in reduced amounts or absent in the cell and thereby treat a genetic disease, such as hemoglobinopathy, caused by abnormal expression of globin. Alternatively or in addition, genome editing, with or without the introduction of a matching donor, can correct a defective endogenous gene, for example, correct a point mutation in α- or β-hemoglobin, restore gene expression, or cure a genetic disease, such as sickle cell disease, knockout or change (overexpression or suppression) of any direct or indirect globin regulation gene (for example, inactivation of the γ-globin regulator gene BCL11A or the BCL11A regulator KLF1). In particular, the methods and compositions of this invention find use in the treatment or prevention of hemoglobinopathy.

The nucleases of the present invention target the BCL11A enhancer region, which is known to be required for the expression of BCL11A during erythropoiesis and, therefore, for the suppression of gamma globin expression. Modification of this enhancer region may result in red blood cells with increased expression of gamma globin and thus may be useful for the treatment or prevention of sickle cell anemia or beta thalassemia.

The following examples relate to exemplary embodiments of this disclosure in which the nuclease comprises a zinc finger nuclease (ZFN). It is understood that this is for illustrative purposes only and that other nucleases, such as the TtAgo and

- 30 045868

CRISPR/Cas, homing endonucleases (meganucleases) with engineered DNA-binding domains, fusions of naturally occurring DNA-binding domains of engineered homing endonucleases (meganucleases), including combinations of homing endonucleases (meganucleases) with engineered DNA-binding domains and fusions of naturally occurring the nature of the DNA-binding domains of engineered homing endonucleases (meganucleases) and heterologous cleavage-fusion domains of meganucleases and TALE proteins, including combinations of heterologous cleavage-fusion domains of meganucleases and TALE proteins.

Examples

Example 1: Assembly of Zinc Finger Nucleases

ZFN was assembled against the human BCL11A gene, and activity was verified by deep sequencing of DNA isolated from transfected cells as described below. ZFNs specific for the +58 region of the enhancer region were prepared as described. The ZFN 51857/51949 pair has been described previously (see WO 2016/183298).

Example 2: Off-target analysis.

To analyze off-target cutting by ZFN pairs, a two-step unbiased specificity analysis was performed. In the first step (Fig. 1), candidate nontarget sites for each ZFN were identified by oligonucleotide duplex integration site analysis using a procedure similar to that described by Tsai et al. ((2015), Nat Biotechnol 33(2):187–197. doi:10.1038/nbt3117).

Analysis of the integration site of an oligonucleotide duplex is based on the observation that co-introduction of a nuclease and a short segment of duplex DNA into a target cell results in integration of the duplex during the repair of a portion of genome cutting events through the NHEJ DNA repair pathway (Orlando et al. (2010), Nucleic Acids Res , 38 (15) el52. doi: 10. 1093/nar/gkq512; Gabriel et al. (2011), Nat Biotechnol. 2011 Aug 7; 29(9):816-23. doi: 10. 1038/nbt. 1948 ; Tsai et al., ibid). Once integrated, the duplex provides a permanent tag for that particular cutting phenomenon. Integration sites are then identified by ligating the oligonucleotide adapter to the sheared genomic DNA, followed by 2 rounds of 25 cycles of nested PCR and deep sequencing of the resulting donor-genomic junctions. This analysis evaluates all potential integration sites in the genome.

Integration site analysis was performed in K562 cells to maximize donor delivery, ZFN expression, and donor integration. Moreover, because K562 cells divide rapidly (doubling time approximately 24 h), they are expected to impose minimal epigenetic restrictions on the ability of ZFN to degrade cellular targets. Cells (2x10 ⁵ ) were electroporated with 0.47 μg of oligonucleotide duplex donor and 400 ng of each ZFN-encoding mRNA using an Amaxa shuttle and settings optimized for maximum ZFN target activity. Four replicate samples were prepared for each oligo and mRNA combination. On day 7 post-transfection, genomic DNA for each sample was isolated (Qiagen DNeasy Blood and Tissue Kit) and 400 ng (133,000 haploid genomes) was used as input for the amplification protocol depicted in FIG. 1. Samples were then processed essentially as previously described (Tsai et al. Ibid). The final products were pooled, quantified, and sequenced on a MiSeq instrument (Illumina) using a v2 300-cycle sequencing kit with 150-bp paired-end reads. and reads with double indexing sequences (dual index reads) of 8 bp. /16 bp to detect sample barcodes at each end of the amplicon.

To generate a list of possible off-target sites, sequencing data were filtered for the correct seed sequence, followed by trimming of adapter sequences and mapping into the genome. The coordinates of the junctions were then mapped, and the duplex-genome junction as well as the position of the break caused by the DNA shear were used to identify various integration events. Integration events were then processed to identify integration clusters in close proximity in the genome (a minimum of 4 different integration events within 100 bp of each other, summed across all replicates). Clusters located on contigs that could not be mapped to the hg38 assembly (i.e., chrUn in hg38) were removed from further analysis. Mapping of clusters at repetitive loci (median of three or more hits per genome across all sequences in a cluster) was also removed as previous experience indicated that these were amplification artifacts. The remaining clusters were scored as candidate sites for ZFN cutting if they were derived from at least 2 replicates of ZFN-treated samples (4 total) and showed >5-fold excess integration events in ZFN-treated samples compared to controls. Candidate cutting sites were classified by the total number of unique integrations in ZFN-treated samples. The candidate loci identified by this analysis are presented in Fig. 2 for the ZFN 51857/51949 pair, classified by the number of integrators.

Example 3: ZFN Optimization

- 31 045868

To reduce off-target cutting, a nuclease optimization strategy was adapted in which non-specific phosphate contacts are selectively removed to cause global suppression of off-target cutting (Guilinger et al. (2014) Nat Methods. 11(4):429-35. doi:10.1038/ nmeth. 2845; Kleinstiver et al. (2016) Nature 529(7587):490-5. doi:10.1038/naturel6526; Slaymaker et al. (2016) Science) 351 (6268): 84-8. doi:10.1126/science.aad5227) (see US Provisional Application Nos. 62/443981 and 62/378978). Amino acid substitutions were made at a key position within the zinc finger framework that interacts with the phosphate backbone of DNA (Pavletich and Pabo, (1991) Science 252 (5007):809-17; Elrod-Erickson et al. (1996) Structure 4 (10) :1171-80) (Fig. 3A-3B), as well as at one position in the right FokI domain of ZFN which is also predicted to create a phosphate contact (Fig. 3C).

Specificity was further improved by allowing each ZFN to be independently expressed from two separated mRNAs, allowing optimization of delivery ratios. These efforts have produced optimized ZFN pairs that are highly related to the parent pair, differing by substitutions that reduce the energetics of interaction with the DNA phosphate backbone but that have minimal or no effect on sequence-specific base recognition. Consistent with this, integration site analysis yielded 455 loci for potential ZFN cutting targets for the seed pair 51857/51949. For the optimized pair, this analysis identified a much smaller number of loci for further validation as potential ZFN cutting targets (72 in total). For both pairs, the putative target in the BCL11A enhancer was the top-ranking locus. Moreover, at the BCL11A enhancer, a much higher proportion of integration events was observed for the optimized pair, consistent with its greater specificity.

It is important to note that in defining the sequence data processing pipeline, key parameters were chosen conservatively to include as many off-target loci as possible rather than filter them out. This was done to ensure that every locus that might represent a good cutting site for the optimized ZFNs would be identified and tested in subsequent indel studies, even at the expense of accepting many more loci that would become false positives. It was expected that the first stage of analysis would produce a large set of candidate loci for each ZFN pair, of which the vast majority (especially for optimized ZFNs) would not represent true off-target cutting sites, but rather background events that might be negative for cutting at subsequent studies of indels.

In the second step of the analysis, candidate off-target loci identified by integration site analysis were screened for evidence of modification (e.g., presence of indels) in ZFN-treated CD34+ HSPCs.

Specifically, human CD34+ HSPCs derived from mobilized peripheral blood were treated with original and optimized ZFN pairs using a clinical scale and clinical conditions for RNA transfection (120 μg/ml mRNA for the original ZFN pair and 100 μg/ml mRNA for the optimized pair). Genomic DNA was isolated 2 days after transfection, followed by PCR amplification of candidate off-target loci and deep sequencing to quantify indel levels. On both the original and optimized ZFN pairs, this step screened the same set of 137 potential off-target loci, as well as a smaller number of potential off-target sites that had been identified by other methods in earlier studies with original ZFN.

The results showed that the optimized ZFNs were noticeably more specific than the original pair. This is evident not only from the number of loci that were scored positive for evidence of ZFN cutting (52 for the original pair vs. 3 for the optimized pair), but also from the observed indel levels, which were much lower for the optimized pair. In fig. Figure 4 shows plots of indel values at each locus showing evidence of ZFN cutting in this study (note log y-axis scale).

The summation of off-target indels for all such loci indicates a 300-fold decrease in off-target activity (cumulative off-target indels for the original pair were 46.5%, versus off-target indels for the optimized pair - 0.15%). This reduction in off-target activity was achieved without loss of activity at the putative target site (72.5% indels for the original pair vs. 81.9% for the optimized ZFNs). In these studies, the original pair (or parent pair) was 51857/51949, whereas the optimized ZFN pair was 63014/65722 (see below). The nuclease constructs are shown in the table below. 1.

- 32 045868

Table 1. ZFN pairs specific to the +58 enhancer region of BCL11A

Design SBS No. Linker [helix sequence, SEQ ID] (target site,__ Mutants 5'-3') [Mutations in the core of the finger] Fok ^F1 _F2 Left RNFSL TM (SEQ ID NO: 20) no RNFSL TM (SE Q ID NO: 20) _F3 partner STG NLT N (SE Q ID NO: 21) no STG NLT N (SE Q ID NO: 21) _F 4 ep TSGSL TR (SEQ ID NO: 22) no TSGSL TR (SE Q ID NO:22) _F 5 DQSNL RA (SEQ ID NO: 19 ) no DQSNL RA (SE Q ID NO: 19 ) F6 51857 aaAGCAACtGTTA GCTTGCACtagacta (SEQ ID NO: 1) 63014 aaAGCAACtGTTA GCTTGCACtagacta (SEQ ID NO:1) DQSNLRA (SEQ ID NO:19) No DQSNLRA (SEQ ID NO:19) AQ CC LF H (S EQ ID NO :2 3) not T AQ CC LF H( SE Q ID NO :2 3) L7c5 ELD L7c5 Qm5 No Qm5 No Qm5 Not T ELD 65459 aaAGCAACtGTTA GCTTGCACtagac ta (SEQ ID NO:1) DQSNLRA (SEQ ID NO:19) RNFSL TM (SE QID NO: 20 ) STG NLT N (SE Q ID NO: 21) TSGSL TR (SE QID NO: 22 ) DQSNL RA(SE QID NO: 19 ) AQ CC LF H( S.E. Q ID NO :2 3) L7c5 Qml4Qm5 No Qm5 No Qm5 T ELD Right partner 51949 caCAGGCTCCAGG AAGGgtttggcct ct (SEQ ID NO: 2) RNDHRTT (SEQ ID NO: 24) QKAHL IR (SEQ ID NO: 25) QKG TLG E (SE Q ID NO: 26) RGRDL SR (SEQ ID NO: 26) RRDNL H.S. (SEQ ID NO: 27) N/ D L0 No No No No No N/ D KKR 65722 caCAGGCTCCAGG RNDHRTT (SEQ ID QKAHL IR QKG TLG RGRDL S.R. RRDNL H.S. N/ D L0

- 33 045868

AAGGgtttggcct ct (SEQ ID NO: 2) NO: 24) (SEQ ID NO: 25) E (SE Q ID NO: 26) (SEQ ID NO: 26) (SEQ ID NO: 27) Qm5 Qm5 No Qm5 No N/ D KKR K525S 65526 caCAGGCTCCAGG AAGGgtttggcctct (SEQ ID NO: 2) RNDHRTT (SEQ ID NO: 24) QKAHL IR (SEQ ID NO: 25) QKG TLG E (SE Q ID NO: 26) RGRDL SR (SEQ ID NO: 26) RRDNL H.S. (SEQ ID NO: 27) N/ D L0 Qm5 Qm5 No Qm5 Qm5 N/ D KKR R416S 65549 caCAGGCTCCAGG AAGGgtttggcctct (SEQ ID NO: 2) RNDHRTT (SEQ ID NO: 24) QKAHL IR (SEQ ID NO: 25) QKG TLG E (SE Q ID NO: 26) RGRDL SR (SEQ ID NO: 26) RRDNL H.S. (SEQ ID NO: 27) N/ D L0 Qm5 Qm5 Qm5 Qm5 No N/ D KKR K525S 65550 caCAGGCTCCAGG AAGGgtttggcct ct (SEQ ID NO: 2) RNDHRTT (SEQ ID NO: 24) QKAHL IR (SEQ ID NO: 25) QKG TLG E SEQ ID NO: 26 RGRDL SR (SEQ ID NO: 26) RRDNL H.S. (SEQ ID NO: 27) N/ D L0 Qm5 Qm5 No Qm5 Qm5 N/ D KKR K525S

Table 1 shows characterization information related to each ZFN. Starting at the left, the SBS number (eg, 51857) is displayed with the target DNA to which ZFN binds shown under the SBS number. The following shows the amino acid recognition helix constructs for fingers 1–6 or 1–5 (divided column 2 of Table 1). In table 1 also shows below the corresponding helix constructs mutations made in the ZFP backbone sequences of the finger, as described in US Provisional Patent Application Nos. 62/378978 and 62/443981. In the notation used in table. 1, Qm5 means that at position minus 5 (relative to the helix, which is numbered -1 to +6) of the indicated finger, arginine at that position has been replaced by glutamine (Q), while Qm14 means arginine (R), typically present at position minus 14 was replaced by glutamine (Q). No means no change outside the recognition spiral area. Thus, for example, SBS #63014 includes a Qm5 mutation in fingers 1, 3, and 5, while fingers 2, 4, and 6 have no mutations in the zinc finger backbone (e.g., a zinc finger sequence outside the recognition helix region).

Finally, in the rightmost column of the table. 1 shows the linker used to link the DNA binding domain to the FokI cutting domain (e.g. L7c5 (LRGSISRARPLNPHP (SEQ ID NO: 5), as described, for example, in US patent No. 9567609 is displayed in the top row of the column, and the FokI mutation sites in contact with phosphates and dimerization mutations are shown in a box below the linker designation. Specifically, the top line of the Fok mutant box indicates the type of mutation found in the dimerization domain (e.g., ELD or KKR, as described, for example, in US Pat. No. 8,962,281). Below the dimerization designations mutants are shown any mutations present in the FokI domain made to remove the nonspecific phosphate contact shown at the bottom (e.g., K525S or R416S, where the serine residues at amino acid positions 525 or 416 were replaced by either lysine or arginine, respectively, as described in the provisional application for US patent No. 62/378978 and

- 34 045868

62/443981). Thus, for example, in SBS #63014, the linker is the L7c5 linker, and the FokI cutting domain includes ELD dimerization mutants and does not contain phosphate contact mutations. In addition, for SBS No. 65722, the linker is the L0 linker (LRGSQLVKS (SEQ ID NO: 6), also called the standard linker, see US Pat. No. 9567609), and the FokI cutting domain includes the dimerization mutations KKR, and the phosphate contact mutation K525S FokI.

All ZFNs were tested for functionality (cutting activity as determined by the indel assay as described in Example 4 below) and were determined to be active.

In addition, to determine which ZFN constructs were most specific, indel analyzes of known off-target cleavage sites by the parent ZFN pair were performed in ZFN-treated CD34+ HSPCs. To do this, human CD34+ HSPCs derived from mobilized peripheral blood were treated with the original and optimized ZFN pairs using clinical conditions and mRNA concentrations (120 μg/ml for the original ZFN pair and 100 μg/ml for the optimized pair). Genomic DNA was isolated 2 days posttransfection from these cells and untreated controls, followed by PCR amplification of each candidate locus and deep sequencing to quantify indel levels.

Levels of modification at each locus were determined by paired-end deep sequencing on an Illumina MiSeq using a 300-cycle cartridge. Pairwise sequences were concatenated, adapters trimmed through SeqPrep, filtered for quality scores >15 across all bases, and then mapped to the human genome (assembly hg38). Sequences mapped to the wrong locus were discarded. The sequences are shorter than the wild type amplicon by >70 bp. or >70% were removed to minimize primer-dimer products. A Needleman-Wunsch alignment (Needleman and Wunsch, (1970) J Mol Biol 48(3):443-53) was performed between the target amplicon and each MiSeq read for indel mapping. Indels in the aligned sequences were defined as described in Gabriel et al. 2011 (ibid.), except that 1 bp indices were adopted to avoid underestimating the real phenomena. Note that a subset of loci were either not amplified, not sequenced, or were rejected from analysis due to high background (>1% modification in control samples) or insufficient sequencing depth (<10,000 reads). The results of this analysis and comparison with the parent pair 51857/51949 ZFN are presented in table below. 2.

________Table 2. Off-target cutting analysis________

Dimer Z FN Off-target: number of loci number of indel-positive VM loci (off-target) Left Right mcg RNA L:P BCL11A % indels target PCR Drive but for analysis P<0.05 manual In parental language? Capture/confirmed 51857 51949 60: 60 73.0 23 21 17 15 17 - 63014 65722 60: 15 82.2 31 24 14 0 4 4/3 63014 65526 60: 15 81.2 23 22 9 3 4 4/4 63014 65527 60: 60 81.4 thirty 24 10 4 4 4/2 63014 65549 60: 60 80.0 thirty 24 13 0 2 2/1 63014 65550 60: 60 79.8 thirty 24 9 0 1 1/1 65459 65526 60: 15 76, 9 23 19 14 0 2 2/2

Example 4: ZFN activity in human CD34+ cells

For in vitro testing, nucleases were tested in CD34+ cells. ZFN was supplied as mRNA, where the mRNA was prepared in vitro as follows: plasmids containing genes encoding ZFN were linearized and used to transcribe the mRNA in vitro using the mMessage mMachine® T7 Ultra kit (Ambion/Applied Biosystems). The mRNA was then purified using the RNeasy® mini kit (Qiagen). CD34+ cells were isolated from mobilized peripheral blood and maintained in XVIVO 10 medium supplemented with penicillin, streptomycin and glutamine, as well as StemSpan CC110, and incubated at 37°C and 5% CO ₂ . Cells were transfected 48 h after isolation or after thawing. A small aliquot was mixed 1:1 with trypan blue 0.4% (wt/vol) in PBS (Corning), and cell numbers were determined on a TC20 automatic cell counter (Bio-Rad).

- 35 045868

For large-scale transfections, cells were washed with MaxCyte electroporation buffer (Maxcyte) and resuspended at 3-5e7 cells per ml in 100 μl electroporation buffer. Typically, mRNA concentrations between 60 and 120 μg/mL have been used to screen suitable ZFN arrays. Cells were then grown in growth medium at 3e6 cells per ml for 18 h at 30°C and then diluted to 1e6 cells per ml for an additional 24 h at 37°C. To determine cutting activity, genomic DNA was isolated 2–3 days after transfections, and the level of gene modification at the BCL11A enhancer locus was measured by deep sequencing on a MiSeq sequencer (Illumina).

ZFN pairs from table. 1 were tested in CD34+ cells and the activity results are shown in Table 1 below. 3.

Table 3. Activity of ZFN pairs against the target BCL11A

Right ZFN Left ZFN P ZFN conc. (mcg) L ZFN conc. (mcg) Indels (%) 51857 51949 60 60 72.98 63014 65722 15 60 80, 62 63014 65722 60 15 82.19 63014 65526 60 15 81.22 63014 65527 60 60 81.43 63014 65549 60 60 79, 82 63014 65550 60 60 79, 96 GFP control 0.07

In addition to analyzing nuclease activity in CD34+ cells prior to erythroid differentiation, the edited cells were also differentiated in vitro into erythroid cells. The protocol was based on the publication of Giarratana et al. ((2011) Blood 120 (15):294 5-53). The protocol given below was briefly followed.

Day 0 - Day 7: ^4x104 CD34+ cells were cultured at a density of 2x104/ml in differentiation medium (EDM) (Iscove's modified Dulbecco's medium [IMDM], 330 μg/ml transferrin, 10 μg/ml human insulin, 2 U/ml heparin sodium, 5% human AB + plasma) in the presence of 10 ^-6 M hydrocortisone, 100 ng/ml stem cell factor (GFR), 5 ng/ml IL 3 and 3 IU/ml erythropoietin (EPO).

Day 4: Cells are resuspended in fresh EDM containing SCF, IL-3, EPO and hydrocortisone.

Day 7 - Day 11: Cells were resuspended at a density of 1.5 x 105 cells per ml of fresh EDM supplemented with SCF and EPO.

Day 11 - Day 21: On Day 11, cells were transplanted at a concentration of 1x10 ⁶ /ml in fresh EDM supplemented with EPO. The cells were subsequently transplanted into the same medium at a concentration of 5x10 ⁶ /ml on day 14. Growth plateaued during this time period between days 14 and 18, when cell viability began to decline until cultures ceased at 21 days.

Cells were counted before plating and throughout differentiation by measuring Acridine Orange positivity and propidium iodide exclusion (AOPI) using a Nexcelom Bioscience K2 cell meter with AOPI Erythroid Assay mode with fluorescence channel 1 (AO) set to 700 ms and fluorescence channel 2 (PI) set to 5000 ms.

The percentage of enucleated cells (cells without a nucleus) was determined at day 21 of differentiation using the following protocol. Enucleation rates were comparable among untransfected controls and ZFN-transfected samples, with percentages of 59–63% from the two groups.

1. Number of cells.

2. 100,000 cells, centrifugation at 450xg, 5 min, RT.

3. Resuspension in 50 µl PBS-BSA + 1 µl GlyA-FITC (DAKO).

4. Coloring for 15 minutes in the refrigerator.

5. Add 1 ml PBS-BSA, vortex, centrifuge.

6. Resuspend in 250 µl PBS-BSA-NucRed (2 drops NucRed per ml).

7. Receive data on FACS Canto using the APC channel for NucRed.

8. Nucleated erythroid cells will be in the GlyA-positive NucRed-negative/low fraction, and erythroblasts will be in the double GlyA-NucRed-positive fraction.

BCL11A gene modification was measured using MiSeq deep sequencing in DNA samples collected: a) 48 h after electroporation, b) on the day of cell thawing, at the time in vitro differentiation began, and c) on day 14 of erythroid differentiation in vitro. Although differentiation was carried out for 21 days, the time point of day 14 was chosen for DNA analysis because it precedes the enucleation of a large proportion of erythroid cells, which results in losses during DNA extraction. The observed percentages of modification at the BCL11A enhancer are given in Table. 4 along with details of transfection conditions.

- 36 045868

Table 4. BCL11A gene modification levels determined using MiSeq analysis

A drug CD34+ Transfection BCL11A gene modification (%) Day 2 After 14th day cells Drug No. 1 80 µg/ml 63014+20 µg/ml 65722 untransfected after TF 78, 6 0, 1 defrosting 81.4 0.2 differentiation. 72.0 0.2 A drug No. 2 80 µg/ml 63014+20 µg/ml 65722 75, 4 77.3 72.3 untransfected 0, 1 0, 1 0, 0

These data show that transfection of CD34+ cells with the optimized mRNA pair 63014 and 65722 results in very efficient gene modification at the BCL11A enhancer target site (>75% of alleles modified) and that the modification is retained very well (>90% modification retention) after freezing and thawing cells and after erythroid differentiation. ZFN pair 63014/65722 was selected for further analysis. The amino acid sequences of these ZFNs are shown below, each containing a nuclear localization signal (NLS -nuclear localization signal, Kalderon et al. (1984) Cell 39 (3 Pt 2):499-509) and a hydrophilic peptide (Hopp et al. (1988) Nat Biotechnol 6:1204-10), which enhances the activity of ZFN on the target, both fused to the N-terminal coding sequence. Thus, the mRNA sequence and amino acid sequence of ZFN are as follows:

63014 mRNA (1725 nt)

5'ggggagacaagcuuugaauuacaagcuugcuuguucuuuuugcagaagcucagaauaa acgcucaacuuuggcagaucgaauucgccauggacuacaaagaccaugacggugauuauaaaga ucaugacaucgauuacaaggaugacgaugacaagauggcccccaagaagaagaggaaggucggc auccacgggguacccgccgcuauggcugagaggcccuuccagugucga aucugcaugcagaacu ucagugaccaguccaaccugcgcgcccacauccgcacccacaccggcgagaagccuuuugccug ugacauuugugggaggaaauuugcccgcaacuucucccugaccaugcauaccaagauacacacg ggcagccaaaagcccuuccagugucgaaucugcaugcagaacuucaguuccaccggcaaccuga ccaaccacauccgcaccca caccggcgagaagccuuuugccugugacauuugugggaggaaauu ugccaccuccggcucccugacccgccauaccaagauacacacgcacccgcgcgccccgaucccg aagcccuuccagugucgaaucugcaugcagaacuucagugaccaguccaaccugcgcgcccaca uccgcaccacacaccggcgagaagccuuuugccugugacauuuguggga ggaaauuugccgccca guguugucuguuccaccauaccaagauacaccugcggggauccaucagcagagccagaccacug aacccgcacccggagcuggaggagaagaaguccgagcugcggcacaagcugaaguacgugcccc acgaguacaucgagcugaucgagaucgccaggaacagcacccaggaccgcauccuggagaugaa ggugauggaguucuucaugaagguguacggcu acaggggaaagcaccugggcggaagcagaaag ccugacggcgccaucuauacagugggcagccccaucgauuacggcgugaucguggacacaaagg ccuacagcggcggcuacaaucugccuaucggccaggccgacgagauggagagauacguggagga gaaccagacccgggauaagcaccucaaccccaacgagugguggaagguguacccuagcagcgug accga guucaaguuccuguucgugagcggccacuucaagggcaacuacaaggcccagcugacca ggcugaaccacaucaccaacugcaauggcgccgugcugagcguggaggagcugcugaucggcgg cgagaugaucaaagccggcacccugacacuggaggaggugcggcgcaaguucaacaacggcgag aucaacuucagaucuugauaacucgagucuagaagcuc agg gccuugagcaucuggauucugccuaauaaaaaacauuuauuuucauugcugcgggacauuc uuauauuaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa (SEQ ID NO:28).

Amino acid sequence 63014 (recognition helix regions are underlined; linker is shown in capital letters in italics; mutations in backbone residues 1, 3 and 5 are shown in double underlines; mutations in the dimerization domain (ELD) are shown in bold and italics); the hydrophilic peptide is shown in small text; and the nuclear localization signal (NLS) is shown in small italic letters):

- 37 045868

MdykdhdgdykdhdidykddddkMApkkkrkvGIHGVPAAMAERPFQCRICMQNFSDQSNLRAH IRTHTGEKPFACDICGRKFARNFSLTMHTKIHTGSQKPFQCRICMQNFSSTGNLTNHIRTHTGE KPFACDICGRKFATSGSLTRHTKIHTHPRAPIPKPFQCRICMQNFSDQSNLRAHIRTHTGEKPF ACDICGRKFA AVLSVEELLIGGE MIKAGTLTLEEVRRKFNNGEINFRS (SEQ ID NO:29).

mRNA 65722 (1680 nucleotides):

5'gggagacaagcuugaauacaagcuugcuuguucuuuuugcagaagcucagaauaaac gcucaacuuuggcagaucgaauucgccuagagaucuggcggcggagagggcagaggaagucuuc uaaccugcggugacguggaggagaaucccggcccuaggaccauggacuacaaagaccaugacgg ugauuauaaagaucaugacaucgauuacaaggaugacgaugaca agauggcccccaagaagaag aggaaggucggcauucaugggguacccgccgcuauggcugagaggcccuuccagugucgaaucu gcaugcagaaguuugcccgcaacgaccaccgcaccacccauaccaagauacacgggcgagaa gcccuuccagugucgaaucugcaugcagaacuucagucagaaggcccaccugauccgccacauc cgcacccacaccggcgagaag ccuuuugccugugacauuugugggaggaaauuugcccagaagg gcacccugggcgagcauaccaagauacacacgggaucucagaagcccuuccagugucgaaucug caugcagaacuucagucgcggccgcgaccugucccgccacauccgcacccacaccggcgagaag ccuuuugccugugacauuugugggaggaauuugcccgccgcgacaaccug cacucccauacca agauacaccugcggggaucccagcuggugaagagcgagcuggaggagaagaaguccgagcugcg gcacaagcugaaguacgugccccacgaguacaucgagcugaucgagaucgccaggaacagcacc caggaccgcauccuggagaugaaggugauggaguucuucaugaagguguacggcuacaggggaa agcaccugggcggaagcagaaagccugac ggcgccaucuauacagugggcagccccaucgauua cggcgugaucguggacacaaaggccuacagcggcggcuacaaucugccuaucggccaggccgac gagaugcagagauacgugaaggagaaccagacccggaauaagcacaucaaccccaacgaguggu ggaagguguacccuagcagcgugaccgaguucaaguuccuguucgugagcggccacuucagcgg ca acuacaaggcccagcugaccaggcugaaccgcaaaaccaacugcaauggcgccgugcugagc guggaggagcugcugaucggcggcgagaugaucaaagccggcacccugacuggaggaggugc ggcgcaaguucaacaacggcgagaucaacuucugauaacucgagucuagaagcucgcuuucuug cuguccaauuucuauuaaagguuccuuuguuc ccuaaguccaacuacuaaacugggggauauua ugaagggccuugagcaucuggauucugccuaauaaaaaacauuuuauuuucauugcugcgcuaga agcucgcuuucuugcuguccaauuucuauuaaagguuccuuuguucccuaaguccaacuacuaa acugggggauauuaugaagggccuugagcaucuggauucugccuaauaaaaacauuuau (SEQ ID NO:30).

Amino acid sequence 65722 (recognition helix regions are underlined; linker is shown in large italic letters; hydrophilic peptide is shown in small letters; nuclease localization signal is shown in small italic letters; mutations in backbone residues 1, 2 and 4 are shown in double underline; dimerization domain (ELD) mutations are shown in bold type and italics; and FokI phosphate contact mutation(s) are shown with wavy underlining):

MdvkdhdgdvkdhdidvkddddkMApkkkrkvGIHGVPAAMAERPFOCRICMOKFARND

HRTTΗТКIΗТGEКРFQCRICMQNFSQKAHLIRHIRTΗТGEКРFACDИСGRKFAQKGTLGEΗТКI HTGSQKPFQCRICMQNFSRGRDLSRHIRTHTGEKPFACDICGRKFARRDNLHSHTKIHLRGSQL VKSELEEKKSELRHKLKYVPHEYIELIEIARNSTQDRILEMKVMEFFTIKVYGYRGK HLGGSRKP DGAIYTVGS РIDYGVIVDTKAYS GGYNLPIGQADEMQRYVKENQTRNKHINPNEWWKVYPS SVT EFKFLFVSGHFSGNYKAQLTRLNPKTNCNGAVLSVEELLIGGEMIKAGTLTLEEVRRKFNNGEI NF (SEQ ID NO:31).

Example 5: Assessment of globin levels in erythroid progeny

α-, β-, and γ-globin mRNA levels in cellular mRNA isolated at day 14 of differentiation (before total mRNA levels decline sharply during enucleation and erythroid maturation) were determined by RT-qPCR for the two cell preparations shown above in the table. 4. γglobin mRNA values are shown normalized to β-globin mRNA (Fig. 5A) or αglobin mRNA values (Fig. 5B) from the same samples (using arbitrary units based on the ratio in the untransfected RT-PCR standard defined as 1).

In reverse phase HPLC of protein samples isolated at day 21, a measured parameter of erythroid differentiation was used to determine whether ZFN-mediated modi

- 38 045868 fication of the BCL11A enhancer of erythrocytes fetal hemoglobin at the protein level. The ratio of gamma globin (sum of Agamma and Ggamma peaks) to alpha globin, as well as the ratio of gamma globin (sum of Agamma and Ggamma peaks) to beta-like globins (sum of Agamma, Ggamma, beta and delta globin peaks) were determined and they are shown in Fig. 6.

In this experiment, an approximately 3-4-fold increase in the percentage of fetal globin protein to levels of about 15-20% was observed in erythroid HSPC progeny upon 63014/65722-mediated disruption of the BCL11A enhancer.

Example 6 Engraftment of Edited Cells in NSG Mice

The edited human CD34+ cells were then injected into NSG mice to assess engraftment. The degree of human chimera (ie, the percentage of human CD45+ cells) was measured using flow cytometry (FACS fluorescence activated cell sorting) in peripheral blood collected at 8, and 12, 16 and 20 weeks after transplantation, and in bone marrow collected at 12 weeks and 20 weeks. In addition, to test the level of engraftment of ZFN-modified cells, the level of gene disruption at the BCL11A enhancer locus was assessed by direct high-throughput sequencing of the ZFN target locus and compared with the levels of target gene modification measured in the starting material.

HSPCs from two healthy donors (termed PB-MR-003 and PB-MR-004) were mobilized with G-SCF and Plerixafor and purified as described in Yannaki et al. ((2012) Mol Ther 20(1):230-8. doi:10.1038/mt.2011). Platelet depletion was performed on the leukapheresis product using the Fresenius-Kabi Lovo device before enrichment with CD34+ cells using the Miltenyi Biotech CliniMACS Plus instrument. The purified cells were then seeded into culture for transfection.

Two days after cell purification, CD34+ cells were electroporated using the Maxcyte instrument in the presence of 120 μg/ml single mRNA encoding the parental ZFN pair, 63014/65722, or optimized amounts of two separate mRNAs encoding the optimized ZFN pair 80 μg/ml 63014 and 20 μg /ml 65722. Before transfection, an aliquot of cells was set aside as an untransfected control. 95 million cells were transfected from PB-MR-003, and 120 million cells were transfected from PB-MR-004.

After electroporation, temporary culture was performed overnight at 30°C, and then the cells were cultured for an additional 24 hours at 37°C. Two days after electroporation, aliquots of cells were collected for DNA analysis, and the remaining cells were collected, cryopreserved, and stored in liquid nitrogen.

Conditioning:

Mice were treated with 10 mg/kg/day Baytril water 1–2 days before irradiation and sublethally irradiated with 300 RAD 16–24 h before transplantation. The transplant was performed by injection into the tail vein (see below). The mice then received fresh water containing Baytril. The Baytril water was changed after a week, and the addition of Baytril to the water was stopped 14 days after transplantation.

Transplantation:

On the day of transplantation, pre-warm X-Vivo 10/1% PSG + 3-cytokine cocktail (recombinant human stem cell factor (SCF), recombinant human thrombopoietin (TPO) and recombinant human Flt-3 ligand (Flt-3L)) at 37°C, prepare fresh PBS/0.1% BSA at ambient temperature (sterile/filtered). Cryopreserved cells were thawed at 37°C, pelleted, resuspended in prewarmed X-Vivo medium, pelleted again, resuspended in PBS/0.1% BSA, and counted. After another pelleting, the cell pellet was resuspended in 550 μl per mouse PBS/0.1% BSA (2χ10 ^Λ 6 cells/ml based on cell number). The cells were then injected at room temperature into the tail vein of the mouse using a 25-gauge needle. The study groups are shown in the table below. 5.

Table 5. Dosing groups for engraftment of edited hCD34+ cells

Species: Mouse Gender, Age, Number: 60 female NSG mice Group No. (Quantity/group ppu) Researchable a drug % Indian fishing (Day 2) Viability via Id after defrosting Dose (cells/mouse) Qty/ Killing I'm 12 weeks 20th week

- 39 045868

Animals were observed daily to assess general health and weighed daily for the first 2 weeks and then weighed every two weeks. Peripheral blood was collected from the submandibular vein (100 μl) at 8, 12, 16, and 20 weeks after transplantation or through cardiac puncture (1 ml) for sacrificed animals at 12 and 20 weeks after transplantation. Half of the animals in each group (5 mice per group) were euthanized 12 weeks after transplantation, and bone marrow and terminal blood were collected for analysis. The remaining animals in each group (5 mice per group) were sacrificed 20 weeks after transplantation.

Blood collection, cell collection and processing: Peripheral blood was collected through the submandibular vein or cardiac puncture into EDTA tubes and centrifuged at 500xg for 5 min to remove plasma. After washing with phosphate-buffered saline (PBS) with bovine serum albumin (BSA) and centrifugation, a 10-fold volume of hemolytic buffer was added to the pellet, and the mixture was incubated at 37°C for 15 min, centrifuged and washed again. The precipitated fraction was reconstituted in 1 ml PBS BSA; the aliquot was removed and centrifuged at 1000xg for 5 min, and the resulting pellet was saved for genotyping. The supernatant fraction was used for FACS analyses.

Bone marrow, femur, tibia, and pelvic bones were collected in Iscove's modified Dulbecco's medium (IMDM) containing fetal calf serum (FCS); the whole bone marrow was washed in PBS BSA solution and filtered using a 70 μm nylon sieve. The volume was adjusted to 10 ml with PBS BSA and an aliquot was used for cell counting (Cellometer).

ZFN activity was analyzed using MiSeq deep sequencing. Briefly, genomic DNA from mice injected with either untransfected control CD34+ HSPCs or CD34+ HSPCs transfected with enhancer-targeted ZFN mRNA was isolated from blood samples obtained at 8 weeks and 12 weeks, or from bone marrow at 12 weeks after injection. The region of interest (containing the ZFN binding site in the BCL11A locus) was amplified by PCR and the level of modification was determined by paired-end deep sequencing on the Illumina (MiSeq) platform.

To generate libraries compatible with the Illumina MiSeq sequencing platform, adapters, barcodes, and a flow cell binder (short DNA sequence) were attached to target specific amplicons using two sets of fusion primers in sequential PCRs. For MiSeq assessment of human BCL11A enhancer modification in mouse blood and bone marrow samples, the protocol must be adjusted due to the low amounts of target DNA in these samples.

For MiSeq adapter PCR, the following primers were used: PRJIYLFN-f2:

ASA CGA CGC TCT TCC GAT CTN NNN AGT CCT CTT STA CCC CAC CCA (SEQ ID NO:32) and

PRJIYLFN-r4:

GAC GTG TGC TST TCC GAT STS TAS TCT TAG ACA TAA CAC ACC AGG G (SEQ ID NO:33).

For DNA analysis, mouse bone marrow samples were isolated using DNeasy and approximately 100 ng of DNA was used in each PCR reaction. DNA from mouse blood samples was isolated using Tissue XS, and 10 μl of 15 μl of extracted DNA was used in each reaction. In addition to DNA, the following was added to each MiSeq PCR reaction: 25 μl of HotStar Taq mix (Qiagen), 0.5 μl of each

- 40 045868 BCL11A enhancer primers listed above (at a concentration of 100 nM) and water up to 50 μl of the total reaction volume. Typical conditions for MiSeq PCR were denaturation at 95°C for 15 min and 30 cycles of 94°C for 30 s, 62°C for 30 s and 72°C for 40 s followed by extension for 10 min. at 72°C. After MiSeq PCR, the PCR product was diluted with water between 1:50 and 1:200 or left undiluted for samples with very low initial cell counts. Barcode PCR was performed with 1 µl MiSeq PCR product diluted as described above, 25 µl HotStar Taq mix, 1 µl forward barcode primer, 1 µl reverse barcode primer (both at a concentration of 10 nM) and water to 50 µl total reaction volume. Barcode PCR conditions were as follows: denaturation at 95°C for 15 min and 18 cycles of 94°C for 30 s, 60°C for 30 s and 72°C for 30 s followed by extension for 10 min at 72°C. Barcode PCR products were pooled and sequenced on an Illumina MiSeq sequencer. The results are shown in table. 5 above.

FACS analysis for chimerism and cell lineage determination. To assess the extent of human chimerism, a fraction of peripheral blood cells (8, 12, 16, and 20 weeks postengraftment) and bone marrow (12 and 20 weeks postengraftment) were stained with hCD45-APC Cy7 (Biolegend) and anti-hCD45-BV510 antibodies (BD Biosciences) respectively and performed FACS analysis. In addition, hematopoietic lines were analyzed by staining bone marrow cells with specific antibodies described in Table. 6 below:

Table 6. Sources of antibodies for cellular markers

Antibody Links CD3-FITC: BD 561807 (UCHT1 clone) BD 561807 CD19-PE: BD 340364 (clone SJ25C1) BD 340364 CD45-BV510 BD 563204 Lin-APC (CD3/UCHT1, CD14/HCD14, CD16/3G8, CD19/HIB19, CD20/2H7, CD56/HCD56) BIOLEGEND 348803 CD33-PE-CF594 BD 562492 GlyA-FITC DAKO 0870 CD38-PerCP Cy5. 5 BD 551400 CD14-PE BD 555398 CD34-PE Cy7 BD 560710 CD71-APC Cy7/H7 BD 563671 CD15-BV650 BD 564232 CD8-PerCP Cy5. 5 BD 341051 CD4-PE Cy7 BD 344612 CD56-APC BD 318310 1dM—ARS Su7 BD 314520 CD20-BV650 BD 563780

Additionally, we used an enrichment/depletion strategy using magnetic cell separation (MACS) to purify and sort HSPC populations. Bone marrow cells were first stained with CD19-6uotuhom, CD3-6uotuhom, B220-biotin, TER119-6uotuhom, and m-ckit-biotin (BD Biosciences) and then incubated with anti-biotin beads (Miltenyi Biotec).

The positive fraction and the depleted fraction were separated using LS columns (Miltenyi Biotec) placed in a MACS magnetic field. After separation, the positive fraction was stained with streptavidin-APC, CD3-FITC, CD19-PE, CD45-BV510 (BD Biosciences) and the debit fraction with CD34-FITC (BD Biosciences), Gly-A-PE (DAKO), CD19-APC ( BD), Lin-APC (Biolegend), streptavidin-APC, CD45-BV510, CD33-PE-CF594 (BD), and CD38-PECy-7 (Biolegend).

Untransfected HSPCs and 63014/65722 transfected HSPCs were engrafted into NSG mice using standard procedures as described above. The extent of human chimerism in these mice after engraftment was assessed by measuring the proportion of hCD45-positive cells using FACS.

In fig. 7 depicts the percentage of human CD45+ cells in peripheral blood collected at 8 and 12 weeks post-transplant, and FIG. Figure 8 depicts the percentages of bone marrow cells collected at week 12. As shown, engraftment levels in this study were comparable to human chimerism after engraftment of untransfected control and 63014/65722 transfected HSPCs. Only 3 mice out of 60 allocated to groups had no CD45+ cells, indicating lack of engraftment.

Recovery of various hematopoietic cell lines was tested by FACS analysis of bone marrow cells obtained at week 12 using antibodies recognizing lineage-specific cell surface markers using standard procedures. As shown in FIG. 9 observed comparable representation of all analyzed human hematopoietic lineages in the bone marrow at 12 weeks post-injection between CD34+ cell-treated progeny

- 41 045868

BCLllA-specific mRNA encoding ZFN and the progeny of untransfected cells. Bone marrow from mice sacrificed at 12 weeks postengraftment was isolated, and the distribution of different hematopoietic lineages was analyzed by FACS using antibodies recognizing the indicated lineage markers. All numbers are given as the ratio of cells staining positive for the indicated lineage marker to the percentage of human CD45-positive cells, with the exception of cells expressing the erythroid marker Cd71+ (Ter119) in Fig. 9C, which are presented as the percentage of positively stained cells in the entire population, since erythroid cells are not CD45 positive.

Levels of gene modification at the BCL11A erythroid enhancer (% insertion and deletion alleles [indels]) were assessed by deep sequencing of the ZFN target region using the MiSeq sequencing platform as described above. The data is shown in Fig. 10 for the 8th week and 12th week blood samples, and FIG. 11 for week 12 bone marrow samples and sorted lines derived from week 12 bone marrow samples of cells treated with 63014/65722. For comparison, the percentages of indels measured 2 days after transfection (as listed in Table 5) are also shown in the plots of Fig. 10 and 11.

In addition, both sets of HSPC donors treated with 63014/65722 were found to maintain a gene modification in the BCL11A enhancer at different time points and in different lines. Comparable modification was observed in both BCL11A-dependent (B cells, 'CD19'; primitive progenitors, 'CD38H') and BCLHA-independent (myeloid 'CD33') lineages. Although the levels of introduced gene modification were higher in the donor sample PB-MR-003 than in the donor sample PB-MR004, cells derived from PB-MR-004 consistently exhibited higher levels of modification, i.e. better retention of the modification in mice than those derived from PB-MR-003.

Overall, the observed conservation of gene modification at the BCL11A enhancer in mice was consistent with what was observed in previous mouse experiments using a range of ZFNs targeting different gene targets.

Additionally, since human erythroid progenitors are unable to differentiate in mice, to determine the number of BCL11A-targeting gene modifications that occurred in these cells, bone marrow cells were removed from mice and differentiated in vitro. In these experiments, human bone marrow-derived cells were removed from sacrificed mice at 12 weeks postengraftment and differentiated in vitro as described above. BCL11A target gene modification was measured using Miseq high-throughput sequencing of DNA isolated from cells at day 14 of differentiation.

These modifications (indels) are presented in Fig. 12, which shows modification levels at day 14 of erythroid differentiation. The percentages of indels at day 14 of in vitro differentiation varied markedly for each culture obtained from cells isolated from a single mouse, reflecting the oligocyte nature of the expansion obtained under these conditions. These data indicate that ZFN 63014/65722-mediated modification of the BCL11A enhancer was not markedly altered during erythroid differentiation. As observed in blood and bone marrow samples, samples of erythroid progeny cells derived from PB-MR-004 showed higher average levels of modification than erythroid progeny cells derived from PB-MR-003.

The relative levels of various globin mRNAs were determined by RT-PCR analysis of RNA isolated from cells on day 14 of erythroid differentiation in vitro, and the data are presented in FIG. 13A, in which the relative ratios of γ-globin mRNA to β-globin mRNA and γ-globin mRNA to α-globin mRNA (FIG. 13B) are averaged for 5 erythroid cultures from each group. In both untransfected and 63014/65722-treated samples, the ratios of γ-globin mRNA to β-globin mRNA and γ-globin mRNA to α-globin mRNA varied greatly between the erythroid offspring of individual mice from the same group. Cultures derived from donor PB-MR-004 exhibited lower γ-globin ratios on average than cultures derived from donor PB-MR-003, which is consistent with the better maturation observed for samples derived from PB-MR- 004. However, despite this variability, average values for ZFN-treated samples show ∼1.5- to 2-fold increases in γ-globin mRNA levels compared to their corresponding untransfected counterparts.

Globin protein levels were assessed by HPLC analysis. Fig. 14 shows analysis of globin proteins in samples collected on day 16 of differentiation. The ratio of gamma globin (sum of Agamma and Ggamma peaks) to alpha globin, as well as the ratio of gamma globin (sum of Agamma and Ggamma peaks) to beta-like globins (sum of Agamma, Ggamma, beta and delta globin peaks) and the average for each group is shown above each column. Consistent with the poor erythroid differentiation of samples derived from PB-MR-003, gamma globin levels in untransfected cells obtained from this donor were very high (~30%), and therefore ZFN treatment resulted in only a 1.2-fold increase in gamma levels. -globin. PB-MR-004 exhibited more typical untransfected levels (~9%) and showed a ~2-fold increase in gamma globin protein levels after 12

-

Claims (23)

weekly circulation through the mouse body. Patients who have >8.6% naturally occurring γ-globin are considered to have an advantage over patients with <8.6% γ-globin (Platt et al. (1994) N Engl J Med, 330:163944 ). In fact, achieving a chimeric percentage of non-sickle red blood cells of 10-20% through engraftment of edited cells may lead to clinical improvement (Chang et al. (2017) Mol Ther Methods Clin Dev 4:137-148. doi:10.1016/j.omtm. 2016.12.009). Thus, despite the need to undergo the process of erythroid differentiation in vitro, the percentage of chimeric cells and the level of detectable γ-globin protein are indicators of therapeutic efficacy. All patents, patent applications and publications referenced herein are incorporated herein by reference in their entirety. Although the disclosure has been offered in some detail by way of illustration and example for purposes of clarity of understanding, it will be apparent to those skilled in the art that various changes and modifications may be practiced without departing from the spirit or scope of the disclosure. Accordingly, the above descriptions and examples should not be construed as limiting. CLAIM 1. A pair of zinc finger nucleases (ZFNs) comprising a left and a right ZFN that dimerize as a pair, where the left ZFN contains the amino acid sequence shown in SEQ ID NO: 29 and the right ZFN contains the amino acid sequence shown in SEQ ID NO: 29 31. 2. A polynucleotide encoding a pair of zinc finger nucleases according to claim 1. 3. The polynucleotide according to claim 2, wherein said polynucleotide is mRNA. 4. The polynucleotide according to claim 3, containing SEQ ID NO: 28 and SEQ ID NO: 30. 5. A cell containing zinc finger nucleases according to claim 1 or a polynucleotide according to any of claims 2-4. 6. The cell of claim 5, wherein said cell is a stem cell or a progenitor cell. 7. The cell of claim 6, wherein said cell is a human cell. 8. The cell according to any one of claims 5-7, where the genome of said cell is modified by zinc finger nucleases in the +58 region of the BCL11A enhancer sequence. 9. The cell of claim 8, wherein said genomic modification is selected from the group consisting of insertions, deletions, and combinations thereof. 10. A cell derived from a cell according to any one of claims 5 to 9, wherein said cell is obtained by culturing a cell according to any one of claims 5 to 9. 11. A cell line derived from a cell according to any one of claims 5 to 9, wherein said cell line is obtained by culturing a cell according to any one of claims 5 to 9. 12. A partially or fully differentiated cell derived from a cell according to any one of claims 5 to 10 or a cell line according to claim 11. 13. A cell according to any one of claims 5 to 10, wherein said cell exhibits increased expression of gamma and/or beta globin compared to a cell without said genomic modification. 14. A pharmaceutical composition containing a pair of zinc finger nucleases according to claim 1, a polynucleotide according to any of claims 2-4 or a cell according to any of claims 5-10. 15. A method of modifying an endogenous BCL11 enhancer sequence in a cell, the method comprising introducing a zinc finger nuclease according to claim 1 or a polynucleotide according to any one of claims 2 to 4 into said cell such that the endogenous BCL11A enhancer sequence is modified. 16. The method of claim 15, further comprising introducing an exogenous sequence into said cell such that said exogenous sequence is inserted into an endogenous BCL11A enhancer sequence. 17. The method according to claim 15, wherein said modification comprises a deletion. 18. A method of increasing globin production in a subject, comprising administering to said subject a cell according to any one of claims 5 to 10. 19. The method of claim 18, wherein the subject is a human and the cell is a human stem cell or a human progenitor cell. 20. The method of claim 19, wherein said cell is administered to a patient and the cell engrafts, differentiates and matures in said subject. 21. The method according to any one of claims 18-20, wherein said subject has hemoglobinopathy. 22. The method according to claim 21, wherein said hemoglobinopathy is beta thalassemia or sickle cell anemia. 23. A method for producing a genetically modified cell containing a genomic modification in the endogenous enhancer sequence BCL11A, wherein said method includes one hundred -