Improved methods for genome editing with and without programmable nucleases

Abstract

To provide compositions and methods for genome editing within isolated cells or within organisms.SOLUTION: Editing oligonucleotides contain an oligonucleotide strand which may contain a linker that positions an editing moiety in a proper location for modifying targeted nucleobase, a crisprRNA domain and an inactivated Cas 9 domain that cause deamination of the targeted nucleobase. The editing oligonucleotides may also contain at least one nucleotide sequence change from a targeted sequence in a genome. Certain embodiments of the method include modifying a genomic sequence within a cell utilizing the editing oligonucleotides without exogenous proteins to assist in an editing process. The editing oligonucleotides may comprise backbone modifications that increase the nuclease stability of the oligonucleotide as compared to unmodified oligonucleotides.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS The following application claims priority to U.S. Provisional Patent Application No. 62 / 333,004, filed May 6, 2016, and No. 62 / 410,487, filed October 20, 2016.

[0002] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT N / A

[0003] Names of the parties to the joint research agreement N / A

[0004] INCORPORATION-BY-REFERENCE OF MATERIAL SUBMITTED ON A COMPACT DISC N / A

[0005] The present invention relates to the use of polynucleotides, including oligonucleotides or polypeptides, including proteins, to modify genome or RNA sequences for applications in the fields of human and animal therapeutics (including in vivo and ex vivo therapeutic applications), cosmetic procedures, preclinical development, basic research, and for improving food reserves in agriculture, breeding animals (both agricultural and other domesticated animals) to provide desirable characteristics in animal husbandry, and energy production. [Background technology]

[0006] Patients born with simple genetic mutations that result in the loss of key functional proteins, such as metabolic enzymes, currently have few corrective treatment options. For a small number of metabolic birth defects, exogenous protein delivery has been used successfully to provide replacement enzymes for therapy, requiring lifelong treatment. Unfortunately, the number of diseases amenable to protein replacement is limited, as most genetic defects require that the protein be produced within the patient's specific cell types and cannot be easily treated by administration of the protein.

[0007] Gene replacement therapy ("gene therapy") has the potential to become a more generally useful method of functionally correcting genetic defects. However, gene replacement therapy is a misnamed name because in most cases, a cDNA (rather than an entire gene) is inserted into a cell and the defective gene is not replaced, but rather a wild-type cDNA is inserted into the cell extrachromosomally or into a site different from the endogenous gene.

[0008] Early enthusiasm for the tremendous potential of this technology has been tempered by decades of clinical trials, resulting in a more realistic view of gene therapy as medicine. What has become clear is the realization that gene therapy has several limitations, including: 1) the potential for adverse immune responses to viral vectors; 2) the potential for viral vectors to integrate and activate oncogenic genes, resulting in cancer; 3) epigenetic silencing of transgenes; and 4) the potential for expressed transgenes to induce cellular immune responses. Advances in oligonucleotide chemistry and in vivo nucleic acid delivery techniques over the past decade have opened up the potential for DNA and RNA modification therapies. Positive data from multiple clinical trials and the approval of the first systemic antisense drug, mipomersen (Ionis Pharmaceuticals, San Diego, CA), in the United States have further demonstrated the clinical utility of oligonucleotide drugs.

[0009] Although the clinical benefits of using therapeutic oligonucleotides to inhibit protein expression by modulating RNA levels have been demonstrated, the therapeutic potential of nucleic acid editing or repair approaches likely outweigh the benefits of these inhibitory approaches (Woolf, et al., PNAS 92:8298-8302, 1995, and Woolf, Nat. Biotech 16:341-344, 1998). A robust editing technology platform allows for site-specific correction of mutant DNA, creation of protective alleles, or otherwise creation of changes in the genome of whole organisms, cells, or tissues that are desirable for research, therapeutic, cosmetic, or agricultural purposes.

[0010] Such a platform would have broad utility as a therapeutic intervention and potential cure for a wide range of diseases caused by genetic point mutations and other genetic lesions. Unlike gene therapy, genome editing has the potential to repair the actual lesion(s), resulting in edited chromosomes with wild-type sequences, without vector sequences, integration at other sites, or random insertions or deletions. This "footprint-free" approach is highly desirable for therapeutic applications, as it avoids potential side effects from unnatural sequences. Also, even if a genome editing therapy inadvertently edits a patient's germline, the precise "footprint-free" edits to the wild type will not create unnatural sequences in the offspring.

[0011] There are two general mechanisms for sequence editing using nucleic acids: chemical modification and incorporation of a nucleic acid sequence into a target. In the chemical modification mechanism, an editing oligonucleotide causes chemical modification of a target nucleic acid base such that the coding of the target nucleic acid base is changed. The second general mechanism is by incorporation of one or more oligonucleotides into a target RNA or DNA sequence. In this mechanism, the oligonucleotide is often referred to as "donor" DNA. This mechanism is loosely referred to as homologous recombination (HR) or homology-directed repair, but may also include other mechanisms such as gene conversion, induction of mismatch repair (see FIG. 1 of PCT / US2015 / 65348), and trans-splicing, or strand invasion, followed by priming of nucleic acid synthesis.

[0012] The explosion of information on genetic and molecular pathways, driven by next-generation sequencing and SNP analysis, has provided numerous targets for therapeutic editing to treat monogenic and polygenic diseases (see PCT / US2015 / 65348). The therapeutic potential of DNA editing repair has been demonstrated by promising data. Engineered zinc finger nucleases (Sangamo Biosciences, Inc., Richmond, CA) have been used to treat HIV, repair mRNA, and exon skipping antisense morpholinos (Sarepta Therapeutics, Inc., Cambridge, MA) have been used to treat muscular dystrophy. CRISPR / Cas-9 and other gene editing approaches that use programmable nucleases to improve editing efficiency have generated several research efforts and large investments in therapeutic applications.

[0013] Therapeutic mRNA editing was first demonstrated in a vertebrate model system by Woolf et al. (PNAS 92:8298-8302, 1995). In this system, targeted stop codon mutations in Duchenne muscular dystrophy mRNA were modified by duplex formation with an editing antisense RNA that induced chemical modification of the target nucleobase by an enzyme at the target site. The enzyme was an endogenous adenosine deaminase that modified the targeted adenosine to inosine, which is translated primarily as guanine. The work by Woolf et al. induced editing with limited specificity.

[0014] Montiel et al. (PNAS 110(45):18285-90,2013) demonstrated a related mechanism of cystic fibrosis mRNA repair, achieving 20% ​​correction in mammalian cells. Although this was a successful proof of principle for therapeutic editing, Montiel's method is complicated for clinical use. The main reason for this is that Montiel et al.'s method requires the introduction of modified genes, mRNA, or proteins into cells by gene therapy, mRNA therapy, or other methods. Therefore, all of the known disadvantages of gene therapy and mRNA therapy are also relevant to Montiel's therapeutic editing method.

[0015] In another approach, Singer et al. (Nucleic Acids Research, 27(24):38-45, 1999) targeted DNA with alkylated oligomers assisted by RecA protein and hybridized to the target strand. However, crosslinking the invading oligonucleotide to the target DNA typically results in various mutations distributed over a region of DNA, which may result in inhibition of replication. Conjugation of reactive base-modifying chemical structures to oligonucleotides and sequence-specific modification of target dsDNA sequences has been achieved (Nagatsugi, et al. Nucleic Acids Research, Vol. 31(6):e31 DOI:10.1093 / nar / gng031, 2003). This study demonstrated site specificity of the target sequence with some specificity for the target base and low but significant efficiency (0.3% in one treatment). However, this method has the same disadvantage as Singer et al., since it results in crosslinking.

[0016] Sasaki et al. (J. Am. Chem. Soc., 126(29):8864-8865, 2004; see also U.S. Pat. No. 7,495,095) developed a method for delivering nitric oxide (NO) to specific cytosine sites in DNA sequences, followed by specific deamination of the cytosine base. This technique required a non-physiological pH and long incubation times to allow the reaction to occur, which may not necessarily be applicable to therapeutic intervention. In addition, chemically reactive oligonucleotide strategies, even if made efficient intracellularly, require complex chemical synthesis and may be reactive with non-target cellular components, including DNA, which is not ideal. They also require different target chemical structures for each base change and are mostly suitable for transitions rather than transversions, which limits their overall usefulness. Furthermore, this method does not repair deletions and insertions, which further limits its general applicability to correcting any mutation. Nonetheless, this chemical modification approach to editing has the advantage that it does not require the addition of exogenous proteins to cells to facilitate editing and can, in principle, be used with highly modified oligonucleotide backbones, which may allow for greater nuclease resistance, better tissue distribution and cellular uptake than oligonucleotides with one or more unmodified DNA linkages.

[0017] Editing with single-stranded editing oligonucleotides resulted in consistent and reproducible editing, but the efficiency was relatively low (approximately 0.1-1%). The most active single-stranded editing oligonucleotides had unmodified internal DNA regions, which likely led to rapid nuclease degradation in cells and to Toll-like receptor activation. Editing efficiency was increased by the following techniques: 1. Adding three phosphorothioate residues to each end of the editing oligonucleotide (however, the resulting editing oligonucleotides are still susceptible to rapid endonuclease digestion in cells, and phosphorothioates increase their toxicity); 2. Synchronizing the cell cycle so that cells are treated with editing oligonucleotides during S phase. Unfortunately, although this increased the editing efficiency to some extent, this approach is cumbersome and not always practical for in vivo therapy; 3. Treating cells with agents that slow down replication fork progression and / or induce DNA strand breaks in the cells, thereby resulting in increased DNA repair in the cells (although this increased the editing efficiency to some extent, this approach is also cumbersome and not necessarily practical for in vivo therapy; and 4. Adding a PNA clamp, a strand-invading single-stranded PNA that binds near the target edit (Bahal et al. Current Gene Therapy 14(5):331-42(2014); Chin et al. PNAS 105(36):13514-13519(2008); Rogers et al. PNAS 99(26):16695-16700(2002); U.S. Patent No. 8,309,356.

[0018] These improvements increased editing efficiency by up to approximately 8% in in vitro cell line models, but each approach had limitations as discussed above (Kmiec, Surgical Oncology 24:95-99, 2015).

[0019] Programmable nucleases, such as zinc finger nucleases, TALENs, and endonucleases based on I-CreI homing endonucleases (such as ARCUS® by Precision BioSciences, Durham, NC), have been used to improve the integration of donor DNA sequences into chromosomes by cleaving chromosomes near the editing target site. Targeted cleavage by the CRISPR-Cas9 system has been used in recent years to improve the efficiency of genome editing. However, programmable nucleases often cause off-target modifications and require single- and double-stranded breaks in chromosomes, which are potentially dangerous and undesirable. One particularly undesirable consequence of using programmable nucleases is the generation of random insertions and deletions (indels) at the cut site(s). The desired precision editing with donor DNA competes with the creation of indels, leaving a mixture of precisely edited chromosomes and chromosomes with various indels. This system also strictly requires that an exogenous modified protein is expressed or delivered to cells in a functional form. This modified protein, i.e., Cas9, is immunogenic, and therefore is less desirable for therapeutic use. In addition, the expression of protein or its delivery to cells is a key issue in clinical development. In order to make a specific change from one sequence to another defined sequence, the CRISPR-Cas9 system requires, in addition to Cas9, a gRNA of more than 70 nucleotides and one or two additional oligonucleotides for insertion in the genome. Thus, the CRISPR / Cas9 editing system is highly complex, and this complexity creates a challenge in clinical development.

[0020] As a result, there is a need in the biomedical and biotechnology industries for nucleic acid editing compounds that function more efficiently and do not strictly require crosslinking of editing agents to the nucleobases of target nucleic acids as a method of action; or introducing insertions and deletions that induce cleavage in target nucleic acids by exogenous programmable nucleases to obtain editing.In addition, it is desirable that these editing agents can repair most point mutations and possibly small insertions and deletions, contain chemical modifications that improve pharmacokinetics in the case of editing oligonucleotides, have biodistribution and intracellular nuclease stability without substantially reducing editing activity, optionally reduce the activation of Toll-like receptors, and correct underlying genetic pathology by editing target DNA sequences, and in some embodiments, RNA sequences. Summary of the Invention

[0021] One aspect of the present invention is a method that utilizes single-stranded oligonucleotides complementary to one of the DNA strands of a genome or RNA for sequence editing (Woolf, TM et al. Nature Reviews Drug Discovery 16, 296 (2017)). The method includes introducing a single-stranded oligonucleotide into a cell or organism without the strict requirement of exogenous proteins to assist in editing the target sequence. In certain embodiments, the oligonucleotide is substantially complementary to the target sequence, except for one or more mismatches, including insertions or deletions, to the target sequence. Such oligonucleotides may be referred to herein as oligonucleotides, oligonucleotides of the invention, or editing oligonucleotides.

[0022] In certain embodiments, the target recognition domain is an edited oligonucleotide that binds to a target sequence, is substantially complementary to the target sequence, and contains one or more chemical modifications.

[0023] In some embodiments, the target sequence recognition domain (see, e.g., Table I) non-covalently binds to nucleotides on the target sequence, activates a nucleobase modifying activity that reacts with nucleotides on the target sequence, or facilitates a reaction with nucleotides on the target sequence (e.g., FIG. 1). The nucleobase modifying activity can be a reactive chemical, a catalyst, or an enzyme. Examples of such reactions include alkylation, acetylation, cross-linking, amination, or deamination.

[0024] These editing oligonucleotides may each be substantially complementary to the target nucleic acid, and may be about 10 to about 50 or 10 to about 200 nucleotides, and at least one of the oligonucleotides may include a crisprRNA and a Cas9 having an inactive nuclease domain bound to a base having a modifying activity and located adjacent to a target nucleic acid base, and the base modifying activity may include a structure that causes deamination of the target nucleic acid base. The target nucleic acid may be RNA or may be DNA. When the target is RNA, the target is preferably mRNA.

[0025] Oligonucleotides may also be referred to herein as oligonucleotides, oligonucleotides of the present invention, or edited oligonucleotides. Oligonucleotides may preferably have one or more chemical modifications. The / these chemical modification(s) may include one or more backbone modification(s), sugar modification(s), nucleobase modification(s), linker, and / or conjugate.

[0026] The oligonucleotides are complementary to a target sequence in a genome and may have mismatches to the target sequence, as described below. The modifications may increase the efficiency of editing by increasing nuclease stability compared to unmodified oligonucleotides or compared to oligonucleotides with three phosphorothioates on each end.

[0027] In one embodiment, the editing oligonucleotide sequence is the sequence desired after editing is completed. The desired editing can be a transition or transversion, or a deletion or insertion. Without wishing to be bound by a particular theory or mechanism, the editing oligonucleotide binds to a partially or fully complementary target genomic DNA sequence when the target sequence separates from the opposite genomic strand during a cellular process such as transcription or replication. In some cases, hybridization of the editing oligonucleotide to the double-stranded genomic DNA target can occur during "breathing" or transient melting of the target DNA. In some embodiments, the penetration of the editing oligonucleotide into the double-stranded genomic DNA is optionally facilitated by a protein(s), such as Cas-9 (or a Cas-9 homologue) or RecA and a single-stranded DNA binding protein, or other protein that enhances strand invasion, such as those listed in Table VIII.

[0028] In one embodiment, once the heteroduplex is formed between the editing oligonucleotide and the target genomic DNA strand, the region of incomplete complementarity is corrected by cellular DNA repair.When the editing oligonucleotide is used as a "correction" template for repair, the desired edit is incorporated into the target genomic DNA strand or RNA strand.In the second mechanism, which can also occur in cells, the editing oligonucleotide is incorporated into the target nucleic acid, such as DNA, by homologous recombination (HR), or by other processes that result in the editing oligonucleotide sequence being incorporated into the target DNA or RNA.

[0029] In one embodiment, each editing oligonucleotide comprises at least one of the internucleotide linkages or sugar modifications listed in Tables II and IV, respectively. To improve editing efficiency, a protein or catalytic nucleic acid may be combined with the editing oligonucleotide (Table VIII).

[0030] Other embodiments include pharmaceutical compositions comprising a pharmaceutical carrier or delivery vehicle and one or more of the editing oligonucleotides (the carrier can be water, saline, or physiologically buffered saline), as well as cells containing one or more of the editing oligonucleotides.

[0031] Another aspect of the present invention is a method of improving the health of an individual in need of treatment for, or reducing, eliminating, or preventing, a medical condition in an individual in need of treatment for, said condition, comprising administering to the individual a composition containing at least one editing oligonucleotide. Some administration methods and target indications are listed in (PCT / US2015 / 65348).

[0032] Other aspects of the invention include methods of administering at least one editing oligonucleotide to an individual suspected of having a condition that can be treated by administration thereof, which condition can be reduced, prevented, or eliminated by reverting a mutated nucleotide in a target nucleic acid to a wild-type nucleotide; modifying a non-mutated nucleotide of a mutated codon in the target nucleic acid to generate a wild-type codon; converting a premature stop codon in the target nucleic acid to a read-through non-wild-type codon; or modifying a mutated codon in the target nucleic acid to result in a non-disease causing amino acid and to generate a non-wild-type codon that results in an edit that inserts or deletes several nucleotides (i.e., in some cases, less than about 10, less than about 5, or less than 3) (see (PCT / US2015 / 65348)).

[0033] Another aspect of the invention is a method for modifying a nucleic acid encoding a protein or functional RNA, or regulating the activity of said protein or RNA by controlling the transcription level of a gene, or modifying a mutant protein to suppress the emotion that causes the disease, comprising the step of administering to a cell or individual at least one editing oligonucleotide or protein. In these methods, the target nucleic acid for editing is DNA.

[0034] The editing oligonucleotides or proteins of the invention may perform one or more of the following functions, including: precise reversion of a mutant base to a base with the coding specificity of a wild-type DNA or RNA sequence; altering a mutant codon to code for a non-wild-type codon that results in a non-disease causing amino acid; modifying a stop codon to a non-wild-type read-through codon, i.e., a codon that still allows activity or partial activity of the target protein; altering the non-mutated base of a mutant codon to result in a wild-type codon or a non-disease amino acid codon; altering the nucleic acid sequence of a protein to increase or decrease (or eliminate) the activity of a domain of that protein; altering an RNA or DNA sequence to generate an allele known to prevent disease. altering sites other than the mutation or disease variant codon in a target mutant protein to suppress the disease-causing effect of a mutant gene; altering sites other than the mutation or disease variant codon in a gene or RNA to suppress the disease-causing effect of a mutant gene (second site suppressors); modulating expression of disease-associated genes to alter promoters, enhancers, or silencing regions of genes (up or down regulating or modulating the response of gene expression to changes in the environment) so that pathology is reduced; altering the epigenetic state of a target sequence, methylating sugars in DNA and / or altering splice site sequences at the DNA or RNA level to obtain a splicing pattern that treats a disease state.

[0035] Self-delivering oligonucleotides refer to chemical structures that efficiently enter the inside of cells without a delivery vehicle, such as Gal-NAC conjugated oligonucleotides, lipophilic group-conjugated edited oligonucleotides (US Patent Application No. 20120065243 A1), or oligonucleotides having phosphorothioate tails or otherwise having about 8 or more phosphorothioate bonds (US Patent Application No. 20120065243 A1). [Brief description of the drawings]

[0036] [Figure 1] Components of an editing oligonucleotide embodiment that act by chemically modifying a target nucleobase (large triangles represent pyrimidines and smaller triangles represent purines). [Diagram 2] Exemplary Listing of Editing and Helper Oligonucleotides. Four text strings (base modification (if any), sequence, sugar, and backbone) form a summary of the modified oligonucleotide. The following abbreviations are provided for the backbone portion: o = phosphodiester; s = phosphorothioate; and m = methylphosphonate. The following abbreviations are provided for the sequence: in the case of RNA or RNA analogs, "T" is understood to be "U"; P = terminal phosphate. The following abbreviations are provided for the sugar portion: D = DNA; R = RNA; M = 2'-O-methyl; F = 2'F; L = LNA; and U = unlocked nucleic acid. The following abbreviations are provided for the base portion: A = adenine; T = thymidine; U = uracil; G = guanine; C = cytosine; and 5 = 5 methyl C. The following letters are used to indicate the linker: "-" = linker. Other abbreviations include: ND = not available; PNA begins with "N-terminus" in the sequence string; K = lysine, O = 8-amino-2,6-dioxaoctanoic acid linker; and J = pseudoisocytosine. The underlined subunits are as follows: for ETAGEN serial number 100197 = gamma miniPEG (PNA BIO, Thousand Oakes, California), for ETAGEN serial numbers 100198 and 100199 = glutamic acid (PNA BIO, Thousand Oakes, California). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] Unless otherwise defined, all terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. All patents, patent applications, website posts, and publications referenced throughout this disclosure are incorporated by reference in their entirety. In the event that there are multiple definitions for terms herein, the definitions in this section take precedence.

[0038] As used herein, the letters "G", "C", "A", "T" and "U" each generally represent a nucleotide containing guanine, cytosine, adenine, thymine and uracil as a base, respectively. However, it will be understood that the term "nucleotide" may refer to modified nucleotides, as further detailed below. In the sequence, when the chemical structure used is RNA or modified RNA, "T" is understood to refer to "U". Similarly, in the sequence, "U" is understood to be "T" in DNA or modified DNA. Those skilled in the art are familiar with the fact that guanine, cytosine, adenine, thymine and uracil can be replaced with other moieties without substantially changing the base pairing properties of the oligonucleotide containing the nucleotide carrying the replacement moiety. For example, but not limited to, a nucleotide containing inosine as its base can form base pairs with nucleotides containing adenine, cytosine or uracil. Also, for example, 5-methyl C can be present in place of C in the target site DNA or in the editing oligonucleotide.

[0039] The term "oligonucleotide" as used herein refers to a polymeric form of nucleotides, either ribonucleotides (RNA), deoxyribonucleotides (DNA), or other alternatives such as peptide nucleic acids (PNAs) (γPNA and chiral γPNA), which may be obtained commercially from a number of sources, including TriLink Biotechnologies (San Diego, CA), Exiqon (Woburn, MA), or PNA BIO (Thousand Oakes, California), or for longer oligonucleotides, Integrated DNA Technologies (Coralville, Iowa), and may be prepared according to methods known in the art (see Oligonucleotide Synthesis: Methods and Applications, In Methods in Molecular Biology Volume 288 (2005) Piet Herdewijn (Editor) ISBN: 1588292339 Springer-Verlag New York, 1999). A polymeric form of nucleic acid base incorporating natural and non-natural nucleotides of length ranging from at least 8, or generally from about 5 to about 200, or up to 500 if chemically produced, or more commonly up to about 100, may be produced using a nucleotide sequence produced by a nucleotide sequencer (such as nucleotide sequences produced by a nucleotide sequencer, such ... In addition, oligonucleotides can be nuclease resistant and include, but are not limited to, 2'-O-methyl ribonucleotides, constrained or locked nucleic acids (LNA), 2' fluoro, phosphorothioate nucleotides (including chirally enriched phosphorothioate nucleotides), phosphorodithioate nucleotides, phosphoramidate nucleotides, and methyl phosphonate nucleotides (including chirally enriched methyl phosphonates).Oligonucleotides may also contain non-natural internucleosidyl bonds, such as those found in PNA or morpholino nucleic acids (MNA). The above definition, when included in the expression "editing oligonucleotide", refers to an oligonucleotide that may further include one or more chemical modifications (e.g., nitrosamines) that react with or facilitate reaction with nucleotides on a target sequence.

[0040] The terms "5'" and "3'" in reference to PNAs are understood to mean the N-terminus or C-terminus, respectively.

[0041] The term "nucleic acid" as used herein refers to polynucleotide compounds, including oligonucleotides, that contain nucleosides or nucleoside analogs with nitrogenous heterocyclic bases or base analogs covalently linked by standard phosphodiester or other bonds. Nucleic acids include nucleic acids, DNA, RNA, chimeric DNA-RNA polymers, or analogs thereof, with 2' modified sugars. In nucleic acids, the backbone may be composed of a variety of linkages (Table II), including one or more of sugar-phosphodiester linkages, peptide-nucleic acid (PNA) linkages (PCT Application WO 95 / 32305), phosphorothioate linkages, or combinations thereof. The sugar moieties in nucleic acids may be ribose, deoxyribose, or similar compounds with substitutions, such as 2' methoxy and 2' halide (e.g., 2'-F), LNA (or other conformationally restricted modified oligonucleotides), and UNA (unlinked nucleic acid) substitutions (Table IV).

[0042] The term "2' modified sugar" as used herein with respect to nucleic acids refers to 2'F, 2' amino, 2'-OX, where X is a modification known in the art to result in hybridizable oligonucleotides that contain a group bridging the 2' ribose to the 4' ribose position (i.e., often referred to as a constrained nucleotide), including, but not limited to, LNA and cET-BNA, bridges 3'-CH2- or 5'-CH2-, bridges 3'-amide (-C(O)-NH-) or 5'-amide (-C(O)-NH-), or any combination thereof (see Table IV for additional examples).

[0043] The term "target sequence" as used herein refers to the contiguous portion of the nucleotide sequence of a DNA sequence in a cell or an RNA sequence in a cell that is modified by an editing oligonucleotide.

[0044] The term "crisprRNA" as used herein refers to crRNA or CRISPR RNA.

[0045] The term "complementary" when used to describe a first nucleotide sequence (e.g., an editing oligonucleotide and a target nucleic acid) with respect to a second nucleotide sequence refers to the ability of an oligonucleotide or polynucleotide comprising the first nucleotide sequence to hybridize with an oligonucleotide or polynucleotide comprising the second nucleotide sequence under certain conditions to form a duplex (or triplex) structure, as understood by those skilled in the art. Preferred hybridization conditions are physiologically relevant conditions that may be experienced within an organism and can be applied. Those skilled in the art will be able to determine the most appropriate set of conditions for testing the complementarity of two sequences according to the end use of the hybridized nucleotides.

[0046] Hybridization involves base pairing of an oligonucleotide or polynucleotide comprising a first nucleotide sequence with an oligonucleotide or polynucleotide comprising a second nucleotide sequence over the entire length of the first and second nucleotide sequences. Such sequences are sometimes referred to herein as "fully complementary" to each other, although in some cases of the editing oligonucleotides of the invention, one or more bases differ from the complementary bases of the target sequence.

[0047] As used herein, the term "substantially complementary" refers to a relationship between an oligonucleotide of the present invention and a target genomic sequence such that a sufficient percentage of the nucleotides of the oligonucleotide form pairs with the nucleotides of the target sequence to promote hybridization. In some embodiments, the percentage is greater than 99 percent, greater than 95 percent, or greater than 90 percent. In some embodiments, the percentage is greater than 80 percent, greater than 70 percent, or greater than 60 percent.

[0048] As used herein, the term "complementary sequence" may also include or be formed entirely from non-Watson-Crick base pairs and / or base pairs formed from non-natural and modified nucleotides, so long as the above requirements regarding their ability to hybridize are met.

[0049] The terms "hybridization", "hybridize", "anneal" or "annealing" as used herein refer to the ability of nucleic acids having substantially complementary sequences to bind to each other through Watson & Crick base pairing under suitable conditions. Nucleic acid annealing or hybridization techniques are well known in the art (see, for example, Sambrook, et al., Molecular Cloning: A Laboratory Manual, Second Edition, Cold Spring Harbor Press, Plainview, NY (1989); Ausubel, FM, et al., Current Protocols in Molecular Biology, John Wiley & Sons, Secaucus, NJ (1994); or physiological conditions within a cell).

[0050] The term "introducing into a cell" or "introducing into a cell" as used herein refers to facilitating uptake or absorption into a cell, as understood by those skilled in the art. Absorption or uptake of the editing oligonucleotide can occur through unassisted diffusive or active cellular processes, or by auxiliary agents or devices. The meaning of this term is not limited to cells in vitro, and the editing oligonucleotide can also be "introduced into a cell" when the cell is part of a living organism. In such cases, introduction into a cell includes administration to an organism. For example, for in vivo delivery, the editing oligonucleotide can be injected into a tissue site or administered systemically. Introduction into a cell in vitro includes methods known in the art, such as electroporation, microinjection, gene injection of nucleic acid, lipofection, or ballistic methods.

[0051] The term "editing," as used herein with respect to a target sequence, refers to at least partial editing of a target gene, manifested by a change in sequence within the target gene. The extent of editing can be determined by isolating RNA or DNA from a first cell or group of cells in which the target gene is transcribed and which has been treated with an editing oligonucleotide compared to a second cell or group of cells that is substantially identical to the first cell or group of cells but has not been similarly treated (control cells).

[0052] Alternatively, the extent of editing may be indicated in terms of a reduction or increase in a parameter functionally associated with transcription of the target gene, such as the amount of protein encoded by the target gene secreted by the cell, or the number of cells displaying a certain phenotype, such as apoptosis. In principle, editing can be determined in any cell expressing the target, and by any suitable assay.

[0053] For example, in certain cases, the target gene is edited in at least about 0.1%, 1%, 3%, 5%, 10%, 20%, 25%, 35%, or 50% of the target cells by administration of the editing oligonucleotide of the present invention. In specific embodiments, the target gene is edited in at least about 60%, 70%, or 80% of the target cells by administration of the editing oligonucleotide of the present invention. The target cell often contains two copies of the target gene, and one or both of these copies can be edited. In some cases, the target cell contains only one copy of the gene targeted for editing, so that only one desired edit occurs per cell.

[0054] The terms "treat", "treatment", and the like refer to the alleviation or mitigation of a condition. In the context of the present invention, insofar as it relates to any of the other conditions listed herein below, the terms "treat", "treatment", and the like refer to the alleviation or mitigation of at least one symptom associated with such a condition, or the slowing or reversing of the progression of such a condition, or the prevention of future disease formation. Treatment may include, in the case of agricultural and industrial applications, modifying the characteristics of an organism.

[0055] As used herein, the phrases "therapeutically effective amount" and "prophylactically effective amount" refer to an amount that provides a therapeutic benefit in the treatment, prevention, or management of a condition or overt symptoms of that condition. The specific amount that is therapeutically effective can be readily determined by an ordinary physician and may vary depending on factors known in the art, such as, for example, the type of disease, the patient's medical history and age, the stage of the disease, and the possibility of administration of other therapeutic agents.

[0056] As used herein, a "pharmaceutical composition" comprises a pharmacologically effective amount of an editing oligonucleotide and a pharma- ceutically acceptable carrier. As used herein, a "pharmacologically effective amount", "therapeutically effective amount", or simply an "effective amount" refers to an amount of an editing oligonucleotide that is effective to produce an intended pharmacological, therapeutic, or preventive result. For example, if a given clinical treatment is considered effective when a measurable parameter associated with a disease or disorder is reduced by at least 25%, then a therapeutically effective amount of a drug for treating that disease or disorder is the amount (including possible multiple doses) required to produce at least a 25% reduction in that parameter.

[0057] The term "pharmaceutically acceptable carrier" refers to a carrier for administration of a therapeutic agent. Such carriers include, but are not limited to, those described in (PCT / US2015 / 65348).

[0058] In order to specifically edit the target sequence site, the editing agent must have a domain that recognizes the target nucleic acid sequence (see Table I). In certain embodiments, the target recognition domain is an editing oligonucleotide that binds to the target sequence, is substantially complementary to the target sequence, and contains one or more chemical modifications. In some embodiments, the target nucleic acid is recognized by a modified sequence-specific nucleic acid binding protein (e.g., a modified transcription factor). [Table 1]

[0059] Since homologous recombination and DNA replication modes are not applicable to RNA targets, RNA editing modes are limited to chemical modification of targeted nucleobase modes or induction of exon skipping.

[0060] In one embodiment of the present invention, a single-stranded oligonucleotide complementary to one of the DNA strands of the genome is utilized for sequence editing (see FIG. 1 of PCT / US2015 / 65348). The desired edit can be a transition or transversion, or a deletion or insertion. In this embodiment of the present invention, the editing oligonucleotide sequence is the sequence desired after editing is completed. Without being bound to a particular theory or mechanism, the editing oligonucleotide binds to a partially complementary or fully complementary target genomic DNA sequence when the target sequence separates from the opposite genomic strand during a cellular process such as transcription or replication. When binding occurs during DNA replication, the editing oligonucleotide can prime DNA synthesis and thus be incorporated into the nascent genomic DNA strand. In some cases, hybridization of the editing oligonucleotide to the double-stranded genomic DNA target can occur during "breathing", transient melting, or unwinding of the target DNA. Once a heteroduplex is formed between the editing oligonucleotide and the target genomic DNA strand, the region of imperfect complementarity is corrected by cellular DNA repair, including homologous recombination (HR). When the editing oligonucleotide is used as a template for repair, the desired edit is incorporated into the target genomic DNA strand.

[0061] Self-delivering editing oligonucleotides are particularly useful because they allow for improved tissue penetration compared to that of nanoparticles (e.g., liposomes), which are much larger than non-encapsulated self-delivering editing oligonucleotides. For example, nanoparticle-incorporated editing oligonucleotides were able to edit the CFTR delta F508 mutation more efficiently in accessible cell cultures and nasal mucosal epithelium, but resulted in much lower editing efficiency in the deep lung (McNeer et al., Nature Comm. DOI:10.1038 / ncomms 7952 pgs.1-11,2015). Specifically, the lungs of CF patients contain additional mucus, which makes it difficult to penetrate the mucus with nanoparticles, but the mucus can be penetrated with molecules of editing oligonucleotides ranging in length from 20 to 70 nucleotides. Indeed, other forms of self-delivering therapeutic oligonucleotides (e.g., uniformly phosphorothioate-modified antisense oligonucleotides) have been delivered deep into the lungs by inhalation.

[0062] The editing oligonucleotide of the present invention comprises some or all of the following segments, listed in the order of 5' to 3': 5' terminal segment, 5' proximal segment, 5' editing segment, editing site, 3' editing segment, 3' proximal segment, and 3' terminal segment. These segments are distinguished by their position and / or chemical modification, but can be adjacently linked by a nucleic acid backbone, whether modified or naturally occurring DNA or RNA. The nucleotides in each of these segments may be optionally modified to improve one or more of the following properties of the editing oligonucleotide: editing efficiency; pharmacokinetic properties; biodistribution; nuclease stability in serum; nuclease stability in the endosomal / lysosomal pathway; nuclease stability in the cytoplasm and nucleoplasm; toxicity (e.g., immune stimulation of Toll-like receptors) and minimum length (e.g., shorter oligonucleotides are generally less expensive to make and easier to deliver to cells in vivo). Non-limiting examples of such modifications are provided in the table below. [Table 2] [Table 3] [Table 4] [Table 5] [Table 6]

[0063] The editing oligonucleotide may include a subset or all of the seven segments listed above, including an editing site and at least one segment 5' and 3' to the editing site. Each of these segments may optionally contain the same or different chemical modifications to improve the properties of the editing oligonucleotide, which in some cases may be uniform throughout the segment and in other cases occur at only a portion of the nucleotides within the segment.

[0064] Other embodiments include editing oligonucleotides with reversibly charge-neutralizing phosphotriester backbone modifications, as described in PCT / US2015 / 65348.

[0065] I. Editing Oligonucleotides In certain embodiments, the editing oligonucleotide of the invention has a structure according to formula (I).

number

[0066] A. 5' end segment (T5) The 5'-end segment is more suitable for many different types of modifications than the 3'-end, because the 3'-end functions in priming DNA extension, which in certain contexts may limit the 3'-end segment to modifications that generally have a free 3' hydroxyl. This priming function is not usually necessary at the 5'-end. The optional 5'-end segment, which may be 0-5 nucleotides long, functions to block 5'-exonucleases that would otherwise readily degrade the editing oligonucleotide in bodily fluids (e.g., blood or interstitial fluid), culture medium, endocytic pathways, or the cytoplasm or nucleoplasm. This segment may contain non-nucleotidic terminal blocking groups and / or modified nucleotide(s) (e.g., inverted bases such as inverted Ts) that are more nuclease-resistant than the 5'-proximal segment. The 5' phosphate can increase editing efficiency and is included in certain embodiments without (Radecke et al. The Journal of Gene Medicine 8(2):217-28, 2006) or with (to improve stability against cellular phosphatases) a thiophosphate modification. Another nuclease-stable 5' phosphate analog that can be used at the 5' end of the editing oligonucleotide is 5'-(E)-vinyl phosphonate. The 5'-(E)-vinyl phosphonate modification is particularly useful for editing oligonucleotides that are loaded into an argonaute (R. Parmar, JLS Willoughby, et al. ChemBioChem 17:85. 2016). If no 5' terminal segment is used, the 5' proximal segment is simply the 5'-most portion of the editing oligonucleotide. The non-nucleotidic terminal blocking group may include any linker known to those of skill in the art for use in carrying out this task (see Table V for these and other linkers useful at this position). Linker lengths can range from 1 carbon to about 20 carbons, or equivalent lengths of other chemical structures, but are preferably less than or equal to 10 carbons in length. Linkers can be used to attach delivery moieties such as cholesterol or 1-3 Gal-Nac moieties.

[0067] The 5'-terminal nucleotide exonuclease-resistant segment may contain one, two, three, or four phosphorothioate modifications. In addition to or in place of one or more of these phosphorothioate modifications, the 5'-terminal exonuclease-resistant nucleotide may contain a 2' sugar modification known to improve exonuclease stability. Furthermore, neutral nucleotide analogs such as methylphosphonates, morpholinos, or PNAs are highly resistant to exonucleases, and one, two, three, four, or five of such modifications at the 5'-terminus may be utilized as exonuclease end-blocking groups. In a particularly useful embodiment, the 5'-terminus is two methylphosphonates (see FIG. 2 of PCT / US2015 / 65348). These end groups are not necessarily complementary to the target.

[0068] B. 5' proximal segment (P5) The 5' proximal segment is more suitable for certain types of modifications than the 3' end for some of the same reasons as described above for the 5' end segment. It can be 1 to 150 nucleotides long, or up to 1500 nucleotides long, preferably about 5 to about 20 nucleotides long. The main function of the 5' proximal segment is to improve the affinity and ability of the editing oligonucleotide to hybridize to the target sequence. Therefore, this segment may optionally be more substantially modified than the editing segment. The 5' proximal segment may contain any of the oligonucleotide modifications mentioned herein. This segment may be composed of DNA or RNA (optionally 2' modified RNA, broadly defined to include LNA and other constrained backbones). Alternative phosphorothioates (e.g., diphosphorothioates and phosphorothioates with improved chiral purity) in this region are not strictly required for nuclease stability, but these alternative phosphorothioates are useful for improving the nuclease stability of RNA and DNA binding. Also, even if the phosphorothioates in this segment are not necessary for nuclease stability, they can increase the overall phosphorothioate content, which increases serum protein binding and cell binding due to the chemically "sticky" nature of phosphorothioates, leading to increased serum half-life in animals and humans, and improving cytoplasmic uptake. For these reasons, in particularly useful embodiments, the total phosphorothioate content of the editing oligonucleotide can be more than 5, 10, 15, or 20. A content of about 20 phosphorothioates often results in excellent serum protein binding and cell binding / uptake. However, because a large number (e.g., more than 6) of phosphorothioate bonds in the complementary region of the editing oligonucleotide can inhibit editing efficiency, phosphorothioate tails can be added to the 5' or 3' end of the region complementary to the target DNA. The tail can be from 1 to about 4, from about 5 to about 9, or from about 10 to about 25 nucleotides in length, and is preferably non-complementary to the target.Other modifications described herein, such as ANA or FANA modified nucleotides, can be incorporated into the 5'-proximal segment to improve nuclease stability, reduce immune stimulation, and / or increase target DNA binding affinity. When editing oligonucleotides are used in a "self-delivery" format (non-encapsulated), the oligonucleotide must survive passage through the endolysosomal pathway, which has high endonuclease activity. Therefore, in a "self-delivery" format, editing oligonucleotides with endonuclease-resistant modifications at most or all nucleotides of the editing oligonucleotide are particularly useful. This can be achieved with a variety of modifications, but phosphorothioate and / or 2'-O-methyl modifications are particularly useful in the 5'-proximal segment. When it is desired to space the 5'-terminal segment intracellularly, for example to expose a free 5'OH or 5'phosphate, one to several unmodified DNA or RNA linkages, or other intracellularly labile linkages, can be optionally placed within the 5'-most region of this segment.

[0069] C. 5' editing segment (E5) The 5' editing segment is 1 to about 10 nucleotides, or 1 to about 100 nucleotides, or 1 to about 200 nucleotides in length, and is located at the 5' end of the editing site, close enough to the editing site to affect the cellular machinery to effect editing of the opposing genomic target DNA strand. Without being bound by any theory, the DNA mismatch repair system may use the editing segments (which are the 5' editing segment, the editing site, and the 3' editing segment) as template strands for editing of the target strand or for subsequent chromosomal DNA replication after incorporation of the editing oligonucleotide into the replication fork. Thus, the majority of the nucleotides in the 5' and 3' edited segments are preferably substantially similar to natural DNA or natural DNA chemical structures (e.g., editing oligonucleotide 100013 has approximately eight unmodified nucleotides 5' to the editing site (see Figure 2), which did not suppress overall editing efficiency compared to the parent compound) and may be unmodified or may include one or more modifications, such as phosphorothioate (Radecke et al. The Journal of Gene Medicine 8:217-28, 2006), 5'S, 2'F, 2'amino, or 3'S, reversibly charge-neutralizing phosphotriester, and nucleobase modifications.

[0070] In one particularly useful embodiment of the present invention, one, some, or all of the deoxy-cytosines in the editing oligonucleotide are modified to 5-methylcytosine, particularly cytosine nucleosides within about 5 to about 10 bases 5' or 3' of the editing site(s). One reason for incorporating 5-methylcytosine into the editing oligonucleotide is that during replication and subsequent mismatch repair, the mismatch repair mechanism recognizes the unmethylated cytosine as the nascent strand and preferentially uses the DNA strand containing 5-methylcytosine as the template strand for repair. In addition, if the editing oligonucleotide contains few or no 5-methylcytosines, the repair mechanism is unlikely to select this strand as a template during the DNA repair reaction that causes editing. The fact that editing oligonucleotides in the art do not contain multiple 5-methylcytosines is one of the reasons for their relatively low efficiency in editing.

[0071] D. Editing site (S E ) The editing site contains a nucleotide(s) that is not complementary to the target genomic DNA and can be 1-6 nucleotides long, but can be longer if necessary. In the case of transition / transversion modifications, the editing site is equal to the number of mismatched bases (e.g., 1 to about 6, especially 1 nucleotide). In the case of editing to create a deletion, the editing site is the junction between the two 5' and 3' nucleotides that base pair with the target genomic DNA strand just opposite the non-base pairing nucleotide in the genomic DNA. In the case of editing to create an insertion, the editing site will be the nucleotide that contains the insertion. In some cases, one editing oligonucleotide can be used to treat different mutations at neighboring sites within the region of complementarity to the target DNA that may occur in different patients in a population. In this case, there will be different editing sites depending on the patient's mutant genotype. In these cases, the editing site will include the 5'-most and 3'-most mutations, as well as the area between these mutations. In a particularly useful embodiment, the editing site spans an entire exon in the target gene, or an exon including an intro / exon boundary region, which is often the site of mutation. Alternatively, when using chemical modification methods, the base-modifying activity is located in close proximity to the nucleobase targeted for editing. In this case, the base-modifying activity can be covalently attached to the editing oligonucleotide at any position along the oligonucleotide, or can be non-covalently associated with the editing oligonucleotide (Woolf et al. PNAS USA 92(18):8298-302, 1995, Woolf Nature biotechnology. 16(4):341-4, 1998, Montiel-Gonzalez et al. PNAS 110(45):18285-90, 2013, and Komor et al. Nature, 2016; advance online publication (Komor et al. Nature. April 20 2016; advance online publication doi:10.1038 / nature17946).Phosphorothioate modifications at editing sites are consistent with significant editing activity (Radecke et al. The Journal of Gene Medicine 8:217-28, 2006 and Papaioannou et al. The Journal of Gene Medicine 11:267-74, 2009).

[0072] In a particularly useful embodiment, the editing oligonucleotide has a 5-methylcytosine in the CpG sequence of the editing site. In a more useful embodiment, this CpG of the editing site is mispaired with a TpG in the target sequence. In this case, the 5-methyl binding protein of the cell binds to the mismatched methylated CpG and induces the cell mismatch repair system to convert the mismatched T to a matched C. The 5' editing segment may contain no modified nucleotides or may contain one or more modified nucleotides up to the number of nucleotides in the segment.

[0073] When an editing oligonucleotide is bound to a target DNA strand and a mismatch occurs between the target sequence and the editing oligonucleotide, the cellular mismatch repair mechanism can repair the target DNA sequence to match the editing oligonucleotide sequence ("productive repair"), or the cellular mismatch repair mechanism can "repair" the editing oligonucleotide ("non-productive repair"). Non-productive repair alters the editing oligonucleotide to a mutant or otherwise undesirable target DNA sequence. Without wishing to be bound by a particular theory or mechanism, one way to avoid "non-productive repair" is to chemically modify the "editing site" of the editing oligonucleotide and / or the nucleotides flanking the "editing site" in the 5' editing segment and / or 3' editing segment with a modification that inhibits "non-productive repair" (see PCT / US2015 / 65348). Modified nucleotides that have been shown to suppress "non-productive" repair by the cellular mismatch repair machinery include 2'F (Rios, X. et al. PLoS ONE 7, e36697, 2012 doi:10.1371 / journal.pone.0036697) and LNA (van Ravesteyn, TW et al. PNAS USA 113, 4122-4127, 2016 doi:10.1073 / pnas.1513315113). Other useful modifications 3' and / or 5' to the "editing site" include the oligonucleotide chemical modifications described herein, in particular the nuclease resistant chemical modifications described herein, in particular ANA, FANA, more particularly the 2' modifications described herein, in particular the 2'-O-alkyl modifications, which are chemically "between" 2'F and LNA, and are easier to manufacture and less toxic than 2'F and LNA, more particularly the 2'-O-methyl modifications described herein (see example in Figure 2). In addition to standard LNA, such as cET chemistry, constrained nucleic acids can also be used at this position(s).

[0074] In the case of a mismatch(es) that has been edited, the mismatched base can be modified (the "editing site"), or the nucleotide(s) 3' and / or 5 next to the mismatched base(s) and the "editing site" can be modified (PCT / US2015 / 65348, Rios, X. et al. PLoS ONE 7, e36697, 2012 doi:10.1371 / journal.pone.0036697 and van Ravesteyn, TW et al., Proc Natl Acad. Sci. USA 113, 4122-4127, 2016 doi:10.1073 / pnas.1513315113). In the case of editing oligonucleotides inserting a nucleotide(s) (e.g., repair oligonucleotides for cystic fibrosis delta F508 (see Example 4 herein and the cystic fibrosis delta F508 oligonucleotide sequence), the sequence inserted may be modified, or the 3' and / or 5 next nucleotide(s) to the "editing site" may be modified (van Ravesteyn, TW et al. PNAS USA 113, 4122-4127, 2016 doi:10.1073 / pnas.1513315113). These modification patterns (PCT / US2015 / 65348) can lead to significantly improved editing, especially when editing is targeted to cells competent for mismatch repair (van Ravesteyn, TW et al. PNAS USA 113, 4122-4127, 2016 doi:10.1073 / pnas.1513315113).

[0075] E. 3' editing segment (E3) The 3' editing segment may have the same range of characteristics, properties, chemical modifications and parameters as the 3' editing segment of E5, except for its location 3' of the editing site.

[0076] F. 3' proximal segment (P3) The 3'-proximal segment has the same range of characteristics and parameters as the 5'-proximal segment, except for its position relative to the other segments. In a particularly useful embodiment, the 3'-proximal segment is composed of 2'-modified nucleotides, and in a more preferred embodiment, 2'F-modified nucleotides. In a particularly useful embodiment, it comprises eight 2'F-modified nucleotides (Figure 2 of PCT / US2015 / 65348 and Figure 2 herein). Other chemical modifications described herein, such as ANA or FANA-modified nucleotides, can be incorporated into the 3'-proximal segment to improve nuclease stability, reduce immune stimulation, and / or increase target DNA binding affinity.

[0077] G. 3' end segment (T3) The 3'-end segment may include the same range of characteristics and properties as the 5'-end segment, except as detailed below. The 3'-end segment may function as a primer for DNA synthesis during DNA replication and repair, thus allowing the editing oligonucleotide to be incorporated adjacently into the genomic DNA. As a result, in one embodiment, this segment will have a free 3' hydroxyl and may be composed of naturally-like modified or unmodified DNA or RNA.

[0078] Non-nucleotidic terminal blocking groups at the 3' end may in some cases reduce or eliminate editing activity, but if there is a region of RNA between the editing site and the 3' end of the editing oligonucleotide that is cleaved by RNase H upon hybridization of the editing oligonucleotide to the target DNA, a free 3' hydroxyl suitable as a primer for chain extension is created. As Monia et al. clearly demonstrated exonuclease degradation of internal phosphodiester DNA bonds in oligonucleotides in cells (Monia et al. J. Biol. Chem. 271(24):14533-40, 1996), other designs that liberate a free 3' hydroxyl include a region of unmodified DNA (e.g., 1-10 nucleotides in length) that is degraded by endonucleases in the nucleus. Another method to liberate a free 3' hydroxyl is the use of a thermolabile bond or terminal blocking group that can be tailored to slowly decompose at various temperatures, preferably physiological temperatures, to liberate a 3' hydroxyl group (Lebedev et al. Nucleic Acids Res. 36(20):e131, 2008). Also, another editing mechanism known as mismatch repair may not require a free 3' hydroxyl or a region of the editing oligonucleotide that base pairs with the target at the 3' end (PCT / US2015 / 65348).

[0079] The oligonucleotide of formula (I) may contain 20-2000 nucleotides or up to about 3000 nucleotides. In one embodiment, the oligonucleotide may contain 100-250 nucleotides. In another embodiment, the oligonucleotide may contain 250-2000 nucleotides. In a specific embodiment, the oligonucleotide contains 20-100 nucleotides. In a specific embodiment, the oligonucleotide contains 25-90 nucleotides. In a particularly useful embodiment, the oligonucleotide contains 19-50 nucleotides. Edited oligonucleotides as short as 12 nucleotides are useful when using modified chemistries that improve the affinity of the edited oligonucleotide for the target DNA.

[0080] Current designs of single stranded editing oligonucleotides using the method of Brachman and Kmiec use either unmodified DNA, which is less efficient for editing, or three phosphorothioates on each end, with the remainder of the editing oligonucleotide containing unmodified DNA, the latter being more efficient for editing, with an optimal length of approximately 72 nucleotides in both cases. Brachman and Kmiec found that synchronized cells treated in S phase were most efficiently edited (Engstrom and Kmiec, Cell Cycle 7(10):1402-1414, 2008), but it is now known that the editing oligonucleotide designs of Kmiec described above are highly susceptible to cellular endonucleases. To increase editing efficiency and eliminate the need for cell synchronization (e.g., because the stable editing oligonucleotides persist in each cell until the cells naturally enter S phase), the present invention provides embodiments of editing oligonucleotides with improved nuclease resistance. Figure 2 provides examples of these editing oligonucleotides. Increased resistance to cellular endonucleases can be achieved by modifying one or more of the nucleotide linkages to phosphorothioate linkages located near the 5' and / or 3' ends. In one embodiment, four or more phosphorothioate linkages at one or both ends are particularly useful. Resistance can also be achieved by replacing all of the nucleotide linkages with phosphorothioate linkages, except for the "editing segment". In some embodiments, the phosphorothioate modification can include one to up to seven nucleotide linkages surrounding the edited base (e.g., a base that differs from the target sequence). In one embodiment, all of the linkages of the editing oligonucleotide are phosphorothioate DNA. In other embodiments, the editing oligonucleotide has the phosphorothioate pattern and length configuration shown in Table 2 of (PCT / US2015 / 65348). The phosphorothioate linkages may also be fully or partially alternating, such as every other linkage being a phosphorothioate linkage, or every third linkage being a phosphorothioate linkage, or two phosphorothioate linkages alternating with one or two phosphodiester linkages.The editing oligonucleotides may contain about 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% phosphorothioate linkages. The editing oligonucleotides (ETAGEN100007, 100010, and 100022) tested in cell cultures containing about 50% or more phosphorothioate DNA produced low or no GFP cell numbers above background in our cell culture GFP targeting system, but these chemistries also cleave the active antisense of the target RNA. The active antisense chemistries will transiently (days) suppress GFP by antisense, thus confounding the editing efficiency analysis of these compounds in this assay. In other cell targeting systems using editing oligonucleotides in a sense orientation, sense DNA editing oligonucleotides with approximately half of the phosphorothioate substitutions on the 3' half of the oligonucleotide had high genome editing efficiency, and even 100% phosphorothioate DNA editing oligonucleotides retained low but significant editing efficiency (Radecke, et al. The journal of gene medicine 8,217-28,2006). Although 100% phosphorothioate substitutions reduced the editing efficiency, the molecules were sufficiently nuclease stable to distribute in tissues and efficiently "self-deliver" to the interior of cells, which is a major advantage when using naked (non-encapsulated) editing oligonucleotides in animals and humans. These phosphorothioate configurations can be combined with other modified or natural sugars as described herein. Specifically, some of the DNA sugars can be replaced with RNA sugars. Preferably, the RNA replacement comprises an RNA binding block beginning at the 3' end and extending in the 5' direction for 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 bases. This is to make the 3' end more like a natural Okazaki fragment or DNA primer, which results in more natural priming of synthesis and removal by RNase H. However, a stretch of three or more unmodified RNA nucleotides is not preferred in the "edited segment" since it is undesirable to remove the edited segment by RNase H.

[0081] Notwithstanding the above description of editing oligonucleotides that act by homologous recombination or primer extension, if the editing oligonucleotide is designed to edit only by virtue of containing a reactive chemical group or enzymatic activity that modifies the target nucleic acid base, the requirements for the editing oligonucleotide modification pattern are simpler. This class of editing oligonucleotides can be made with one or two uniform chemical modifications, such as all PNA, or all phosphorothioates, as well as all 2' modifications (i.e., all phosphorothioates and all 2'-O-methyl, or all phosphorothioates and all LNA).

[0082] H. Other Modifications and Modification Patterns An approach to making editing oligonucleotides more efficient is to increase affinity by chemical modification. The modifications described herein may be combined within an editing oligonucleotide, including constrained nucleic acids including 2'-O-methyl RNA, 2'F RNA, and LNA. These modifications have the added advantage that they may also reduce immune stimulation. Modifications may be grouped to form a high affinity "seed" region for hybridization. In a particularly useful embodiment, this seed region will be located within the 5' proximal segment with about 2 to about 12 consecutive modifications. In other embodiments, the seed region may be located within the 3' proximal segment. Modifications may be alternating, i.e., every third or fourth bond, in the 5' segment, editing segment, and / or 3' segment. The total percentage of chemically modified nucleobases within the 5' segment, 3' segment, or editing segment may independently range from about 20%, 30%, 50%, 60%, 70%, 80%, 90%, or 100%. In one embodiment, the editing oligonucleotide contains a stretch of eight or more alternating 2'F and 2'-O-methyl linkages, as these are known to interact well with Argonaute proteins, which can improve the rate of hybridization to the target, thereby increasing potency.

[0083] Various embodiments of oligonucleotides of formula (I) are provided herein. For example, formula (I) can be RNA, DNA, single-stranded, unmodified, and / or chemically modified, and the chemical modification can include a sugar modification (e.g., 2'-O-methyl and 2'-fluoro). Formula (I) can include one or more backbone modifications and / or linkers as described herein. Formula (I) can further include a conjugation molecule (e.g., gal-nac or lipophilic modification). The aforementioned modifications can be present in various combinations. For example, one, two, or three backbone modifications can be present with one, two, or three sugar modifications, and / or linkers, and / or conjugation molecules.

[0084] Certain oligonucleotides of the present invention contain one or more chemical modifications that react with or facilitate reaction with nucleotides on target sequences. Examples of such reactions include alkylation, acetylation, crosslinking, amination, deamination, and generation of free (non-covalent) reactive compounds. Various classes of sequence modifying moieties useful for editing oligonucleotides are listed in Table VII, and the structure, synthesis, and conjugation of some of these moieties are detailed in PCT / US2015 / 65348. [Table 7]

[0085] As an alternative to using oligonucleotides to effect editing, nucleobase modifying activities (from Table VII) can be linked, conjugated, or fused to protein-only target sequence recognition domains (see examples of such proteins in Table VIII) to achieve genome editing. [Table 8]

[0086] In certain embodiments, chemical modification methods can be combined with "HR" methods of editing to activate endogenous DNA mismatch repair mechanisms to repair target genomic DNA strands. In this embodiment, chemical modification of the target DNA strand can be targeted to the nucleobase targeted for editing, or to the nucleobase or nucleotide adjacent to the nucleobase to be edited (e.g., within 1-10 or 1-50 bases), simply by causing a lesion (adduction) in the target strand, endogenous repair mechanisms can be activated to repair DNA damage and mismatches adjacent to the lesion using the editing oligonucleotide as a template (productive editing).

[0087] The oligonucleotides of the invention may contain protecting groups. Suitable protecting groups are known to those skilled in the art to protect chemically reactive groups during synthesis, purification, storage, and use (e.g., to protect the oligonucleotide from conditions including acidic, intracellular esterase, or reducing conditions).

[0088] While it is convenient to use only a single oligonucleotide to achieve editing, enhancing embodiments herein include the use of additional oligonucleotides. Use of "helper" oligonucleotides or "helper" oligonucleotides, where "helper" oligonucleotides refer to oligonucleotides that bind to an editing oligonucleotide (e.g., at the 3' end, 5' end, or both ends in the case of two helper oligonucleotides, or at a further location within 200 nucleotides of the editing oligonucleotide binding site, e.g., 5' or 3' of the editing site). The helper oligonucleotide helps open up the structure of the target site or otherwise improves the binding efficiency of the editing oligonucleotide. For helper oligonucleotides that bind in whole or in part by strand invasion via Watson / Crick binding, high affinity chemistries are particularly useful. In particular, chemistries that have a higher affinity for the target DNA than the affinity for the natural opposing DNA strand of the target DNA are particularly useful, so that the displacement loop that is formed is energetically favored. LNA (Geny et al. Nucl. Acids Res. (2016) doi:10.1093 / nar / gkw021 First published online: February 8, 2016) or R Even higher affinities, such as those obtainable with PNAs, including mini-PEGγPNAs, can allow for more efficient strand invasion (Sahu et al. Journal of Organic Chemistry 76(14):5614-27, 2011; Bahal et al. Current Gene Therapy 14(5):331-42, 2014) and Bahal, R. et al. Nat. Commun. 7, 13304 (2016). Another target for the helper oligonucleotide can be the strand opposite the DNA strand targeted by the editing sequence. In this case, it is preferred that the helper oligonucleotide(s) bind just 5' and / or 3' to the binding site of the editing oligonucleotide so as not to hybridize strongly to the editing oligonucleotide itself.In other embodiments, the 5' and / or 3' helper oligonucleotides overlap the editing oligonucleotide binding site by about 1-5, about 5-10, or about 1-15 bases. In this way, the helper oligonucleotides will not bind too closely to the editing oligonucleotides and adversely affect the editing oligonucleotide binding to the target. These helper oligonucleotides may optionally be covalently linked to the editing oligonucleotides by phosphodiester or modified phosphodiester bonds or other covalent linkers (see Table V, Useful Linkers for Editing and Helper Oligonucleotides, for examples of linkers). In another embodiment, triplex-forming oligonucleotides or oligonucleotide analogs bind to target DNA within about 200 nucleotides of the editing site, resulting in increased editing efficiency (McNeer, NA et al., Nature Comm. DOI: 10.1038 / ncomms 7952 pgs.1-11, 2015); Bahal et al. Current Gene Therapy 14(5) pp331-42 (2014); Chin et al. PNAS 105(36): 13514-13519 (2008); Rogers et al. PNAS 99(26): 16695-16700 (2002); U.S. Patent No. 8,309,356; and Bahal, R. et al. Nat. Commun. 7, 13304 (2016).

[0089] Some helper oligonucleotides protect oligonucleotides complementary to the editing oligonucleotides and block nuclease degradation by single-strand specific nucleases (PCT / US2015 / 65348).

[0090] Helper oligonucleotides can be made that contain the self-delivering chemical structure described herein.For neutral backbone helper oligonucleotides such as PNA, a group or sequence of self-delivering charged amino acids known in the art can be used (Natee Jearawiriyapaisarn et al.Molecular Therapy 16:1624-1629,2008; Sazani et al.Nature Biotechnology,20:1228-1233,2002).For example, see in Figure 2 the helper oligonucleotide in which lysine is placed at the end of the PNA oligonucleotide to enable naked oligonucleotide delivery in cell culture and in vivo.

[0091] Although not cleaving the target DNA is useful from the perspective of clinical safety, it is known that cleaving the target DNA can improve editing efficiency. If higher editing efficiency is desired, chemical DNA cleavage moieties (e.g., chelators, Simon et al. Nucleic Acids Research 36(11):3531-3538, 2008. doi:10.1093 / nar / gkn231.) can be attached to the editing oligonucleotide or helper oligonucleotide to cleave one or both strands of the target DNA to further improve editing activity. This approach has advantages over programmable nuclease methods known in the art because this method does not require delivery or expression of modified immunogenic proteins, and because the cleavage activity is covalently linked to the editing oligonucleotide, placing the editing oligonucleotide in close proximity to the cleavage site.

[0092] I. Enhanced editing by exogenous and endogenous proteins While it is advantageous not to strictly require exogenous proteins in the editing composition, certain exogenous proteins can improve the above-mentioned embodiments by protecting the editing oligonucleotide from nuclease degradation and by enhancing the binding of the editing oligonucleotide to the target genomic DNA. The chemically modified editing oligonucleotides described herein can be used as donor oligonucleotides for homologous recombination editing together with programmable nucleases to improve editing efficiency and accuracy (Renaud et al., 2016, Cell Reports 14, 2263-2272 March 8, 2016). One or more of the following proteins or ribonucleoproteins can be added in conjunction with the editing oligonucleotide (see also Table VIII and Table IX):Programmable nucleases, including zinc finger nucleases (Carroll D. Genetics 188:773e82. (2011)), TALENs, megaTALENs, other homing endonucleases, CRISPR-Cas9 (Jinek et al. Elife 2013;2:e00471), or homologous or similarly acting ribonucleoproteins (i.e., Cpf1, C2C1, or C2C3), including mutant forms thereof that have been selected to increase target specificity by reducing DNA binding affinity (eSpCas9, Slaymaker et al. Rationally Engineered Cas9 Nucleases with Improved Targeting Specificity, Science (2015)), and a mutant form that has been selected to tolerate more DNA replacements in the crRNA region so that the "crRNA" can act as donor DNA for editing, and a mutant form in which the Cas9 nuclease has been inactivated), Argonaut (Natronobacterium gregorii Argonaute, also spelled Argonaute (Gao et al. Nat Biotech. 2016; advance online publication doi:10.1038 / nbt.3547) and related Argonautes using DNA guides, as well as mutant forms of such Argonautes that lack DNA deletion activity ("dead Argonautes"). When using dead Argonautes, editing will be achieved by using guide DNA that is an editing oligonucleotide with the desired editing sequence or linked to a nucleobase modifying activity for editing. The advantage of using Argonautes guided by dead DNA is that DNA guides can be made that correspond to the desired editing sequence, and thus the DNA guide can achieve editing without the need for cleavage of the target DNA and the need for a separate editing oligonucleotide ("donor DNA").Alternatively, when Argonaute is used with an RNA or DNA guide, the guide can be an editing oligonucleotide that perfectly matches the target DNA, and the editing oligonucleotide is coupled to a target nucleobase-modifying activity to achieve editing (see Table VII). [Table 9-1] [Table 9-2]

[0093] The protein can be produced separately from the editing oligonucleotide, purified, and then pre-complexed with the editing oligonucleotide(s) (Kim et al. Genome Res. 24:1012e9 (2014)), or the protein can be expressed in the target cell / tissue. Expression of exogenous proteins in cells or tissues can be performed by methods known in the art, including gene therapy vectors, naked DNA transfection, or mRNA transfection.

[0094] When using Cas9, a particularly useful embodiment uses Cas9 in which both nuclease domains have been inactivated by mutations known in the art (known as deadCa9 or dCas9). The crRNA is preferably separate from the tracRNA and carries the desired edited sequence. The crRNA may also have from one up to about 15 replaced DNA bonds within and optionally around the editing site as defined herein (the following references show that DNA replacements can be made within the guide region of the tracrRNA: Zachary Kartje et al. Abstract 12 thAnnual Meeting of the Oligonucleotide Therapeutics Society. September, 2016, https: / / custom.cvent.com / F89D960A94384DDB8049882DD4DFBD4E / files / cdb733770e9e4c14ac0a51b7386a9462.pdf). In this way, the specificity and non-chromosomal cleavage advantages of Brachman Kmiec-type genome editing, or editing using oligonucleotides containing chemically reactive groups that modify target nucleobases to change their coding as described herein, are improved in efficacy due to improved target hybridization driven by nuclease-inactivated Cas9 (dCas9). In another embodiment, the approximately 18 nucleotide guide RNA portion of the crRNA or tracRNA is extended in the 5' or 3' direction by an editing oligonucleotide. The editing oligonucleotide may be unmodified or have various modifications as described herein. The editing oligonucleotide will be covalently linked to the crRNA or tracRNA guide by a phosphodiester (or phosphodiester analog) bond or chemical linker, or non-covalently linked by base pairing with a portion of the CRISPR guide RNA or by extension of the CRISPR guide RNA. In a particularly useful embodiment, the editing oligonucleotide portion hybridizes adjacent to a sequence complimentary to the guide portion of the tracRNA or crRNA, and extends the duplex with the target DNA into the region of the target mutation. Because Cas-9 / CRISPR improves the efficiency of strand invasion, this approach will be more efficient than oligonucleotide-directed genomic approaches that do not use CRISPR-Cas9. This approach will be more selective and simpler than the common Cas-9 / CRISPR approach, which cuts the target chromosome and requires a separate donor oligo.A crucial distinction between our editing approach described above and the common method of using CRISPR / Cas9 to obtain precise editing is that the common method of using CRISPR / Cas9 includes a guide portion of the gRNA or crRNA (strand invasion portion) that is perfectly complementary to the target DNA strand, while our approach herein includes a guide portion of the crRNA or gRNA that has the desired editing sequence (e.g., the wild-type sequence when the desired edit is a change from a point mutation to wild-type). The current editing approach also has a region within the crRNA or gRNA guide segment that binds to the mutation and halts DNA replacement within and optionally around the editing site that can serve as donor DNA for editing (by serving as a template for repair or by HR).

[0095] In another embodiment, the crRNA (either separate from the tracrRNA or as a segment of the gRNA) is associated with a nucleobase-modifying moiety, such that the crRNA functions as an editing oligonucleotide in accordance with the present invention as well as in our previous application PCT / US15 / 65348 and recent work demonstrating the present invention (Komor et al. Nature, 2016 Apr 20. doi:10.1038 / nature17946, Epub ahead of print).

[0096] J. Small molecules that improve editing Small molecules can also improve editing frequency. Adding small molecules is much easier to handle than adding programmable nucleases (see Table X, Non-catalytic agents that can be combined with editing oligonucleotides to improve editing efficiency). In each case, editing efficiency can be improved by optionally synchronizing cells in S phase (such as aphidicolin), slowing down replication forks (Erin E. Brachman and Eric B. Kmiec DNA Repair 4:445-457, 2005) or treating the target cell or organism with agents that otherwise increase the expression and / or activity of homologous DNA repair machinery, such as hydroxyl urea, HDAC inhibitors, or camptothecin (Ferrara and Kmiec Nucleic Acids Research, 32(17):5239-5248, 2004) during or before exposure to editing oligonucleotides (see Table X, Non-catalytic agents that can be combined with editing oligonucleotides to improve editing efficiency). [Table 10]

[0097] Another method to improve the efficiency of homologous recombination of chemically modified "donor" editing oligonucleotides is the addition of a PNA clamp near the target mutation (Schleifman et al., Chem. Biol. 18(9):1189-1198, 2011). Glazer used this technique with second generation editing chemistries (e.g., three phosphorothioate modifications at one of the ends of the donor DNA oligonucleotide), but PNA clamps have not been used with the third generation, more heavily modified donor DNAs described and referenced herein. (Bahal et al. Current Gene Therapy 14(5):331-42, 2014; Chin et al. PNAS 105(36):13514-13519, 2008; Rogers et al. PNAS 99(26):16695-16700, 2002; and U.S. Patent No. 8,309,356). Another embodiment of the invention combines the internally modified editing oligonucleotides described herein (not just terminal or near-terminal modifications) with PNA helper oligonucleotides (including PNA clamps, tail clamps, and strand-invading PNAs).

[0098] K. Synthesis Instructions regarding the synthesis of specific oligonucleotides to be utilized as editing oligonucleotides of the present invention can be found in the art, including the references cited in PCT / US2015 / 65348 and (PCT / US2015 / 65348).

[0099] Phosphodiester or phosphodiester analog editing oligonucleotides can be conjugated to PNAs (e.g., PNA helper oligonucleotides, or PNAs that form segments of editing oligonucleotides) by methods known in the art (e.g., Rogers et al. PNAS 99(26):16695-700, 2002).

[0100] L. Internucleoside bond Particularly useful modified internucleoside linkages or backbones are described herein (see Table II and (PCT / US2015 / 65348). Various salts, mixed salts, and free acid forms are also included.

[0101] M. Nucleoside Mimetics In other particularly useful oligonucleotide mimetics, both the sugar and the internucleoside linkage, i.e., the backbone, of the nucleoside unit are replaced with novel groups. The nucleobase unit is maintained for hybridization with the appropriate nucleic acid target compound. One such oligonucleotide, i.e., oligonucleotide mimic, that has been shown to have excellent hybridization properties is called peptide nucleic acid (PNA). In PNA compounds, the sugar backbone of the oligonucleotide is replaced with an amide-containing backbone, specifically an aminoethylglycine backbone. The nucleobase is retained and is directly or indirectly bound to the atom of the amide portion of the backbone. Representative U.S. patents that teach the preparation of PNA compounds include, but are not limited to, U.S. Patent Nos. 5,539,082, 5,714,331, and 5,719,262. Further teaching of PNA compounds can be found in Nielsen et al., Science, 254:1497, 1991.

[0102] Some particularly useful embodiments of the invention employ oligonucleotides having phosphorothioate linkages, as well as oligonucleosides having heteroatom backbones, specifically -CH2-NH-O-CH2-, -CH2-N(CH3)-O-CH2- (known as the methylene (methylimino) or MMI backbone), -CH2-ON(CH3)-CH2-, -CH2-N(CH3)-N(CH3)-CH2-, and -ON(CH3)-CH2-CH2- of the above-referenced U.S. Patent No. 5,489,677 (where the natural phosphodiester backbone is represented as -OPO-CH2-), and the amide backbone of the above-referenced U.S. Patent No. 5,602,240. Oligonucleotides having morpholino backbone structures of the above-referenced U.S. Patent No. 5,034,506 are also particularly useful.

[0103] N. Nucleobase Modifications The oligonucleotides used in the editing oligonucleotides of the present invention may additionally or alternatively contain nucleobase modifications or substitutions. As used herein, "unmodified" or "natural" nucleobases include the purine bases adenine (A) and guanine (G), and the pyrimidine bases thymine (T), cytosine (C), and uracil (U). Modified nucleobases include other synthetic and natural nucleobases (for nucleobase modifications and synthesis, see PCT / US2015 / 65348) (see also Table III, Nucleobase Modifications Useful for Oligonucleotides).

[0104] O. Complementarity An editing oligonucleotide and a target nucleic acid are complementary to each other when a sufficient number of nucleobases of the oligonucleotide can hydrogen bond with corresponding nucleobases of the target nucleic acid so that a desired function occurs (e.g., to allow a desired base modification to occur after hybridization).

[0105] Non-complementary nucleobases between the editing oligonucleotide and the target nucleic acid can be tolerated, provided that the editing oligonucleotide remains capable of specifically hybridizing to the target nucleic acid. In certain embodiments, the oligonucleotides provided herein, or specific portions thereof, are complementary to, or at least 70%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% complementary to, a target nucleic acid, target region, target segment, or specific portion thereof (see Table III, Nucleobase Modifications Useful for Oligonucleotides, for a list of possible nucleobase modifications. In certain embodiments, the editing oligonucleotides provided herein, or specific portions thereof, are complementary to, or at least 70%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% complementary to, a target nucleic acid, target region, target segment, or specific portion thereof. The percent complementarity of an editing oligonucleotide with a target nucleic acid can be determined using routine methods. For example, an editing oligonucleotide in which 16 of 20 nucleobases are complementary to a target nucleic acid and will therefore hybridize specifically will have 80% complementarity. In this example, the remaining non-complementary nucleobases may be clustered or interspersed with complementary nucleobases and do not have to be adjacent to each other or to complementary nucleobases. They may be at the 5' end, 3' end, or in an internal position of the editing oligonucleotide. In another example, an editing oligonucleotide that is 18 nucleobases long with one non-complementary nucleobase flanked by two oligonucleotides with full complementarity to the target nucleic acid will have 94.4% overall complementarity with the target nucleic acid and is therefore within the scope of the present invention.

[0106] The percent complementarity of the editing oligonucleotide with a region of the target nucleic acid can be routinely determined using BLAST (base local sequence search tool) and PowerBLAST programs known in the art (Altschul et al., J. Mol. Biol., 215:403 410, 1990; Zhang and Madden, Genome Res., 7:649-656, 1997). The Gap program (Wisconsin Sequence Analysis Package, Version 8 for Unix, Genetics Computer Group, University Research Park, Madison WI), etc., using the Smith and Waterman algorithm (Adv. Appl. Math., 2:482-489, 1981) and default settings, can also be used.

[0107] II. Mode of Action A. Hybridization Hybridization of the editing oligonucleotide with the target nucleic acid can occur under different stringency conditions, is sequence-dependent, and is determined by the nature and composition of the nucleic acid molecule to be hybridized. The most common hybridization mechanism involves hydrogen bonding (e.g., Watson-Crick, Hoogsteen, or reversed Hoogsteen hydrogen bonding) between complementary nucleic acid bases of the nucleic acid molecule.

[0108] Methods for determining whether a sequence is specifically hybridizable to a target nucleic acid are well known in the art. In certain embodiments, the editing oligonucleotides provided herein are specifically hybridizable to a target nucleic acid.

[0109] B. Target binding The editing oligonucleotides of the present invention are designed to target DNA or RNA. The editing oligonucleotide(s) may be flanked on either or both sides by oligonucleotides that are fully complementary or substantially complementary to the target nucleic acid.

[0110] Particularly useful methods of attachment are by strand displacement, resulting in hybridization to either the Watson or Crick strand of DNA, or, if RNA is the target, by hybridization of the antisense editing oligonucleotide to the sense RNA strand.

[0111] C. Editing Oligonucleotides The editing oligonucleotide may optionally contain a "linker" that covalently attaches the delivery moiety to the oligonucleotide. The linker attachment may be to any nucleobase in the editing oligonucleotide, to the 5' end, to the 3' end, to the sugar residue, or to the backbone. The linker may be any linker known to those skilled in the art for use in carrying out this task. (See Table V, Linkers Useful for Editing and Helper Oligonucleotides, for these and additional linkers). The linker length may range from 1 carbon to about 20 carbons, or equivalent lengths of other chemical structures, but is preferably less than or equal to 10 carbons.

[0112] D. Editing by chemical modification of targeted nucleobases Nucleobase 1 shows a mechanism of editing using the editing oligonucleotides of the present invention for editing chemical modification patterns. The editing oligonucleotides can be of various lengths, as described herein.

[0113] The editing oligonucleotide used in the chemical modification method of editing comprises at least three components, including a "guide oligonucleotide", a "linker" or non-covalent bond that connects the "sequence modifying moiety" to the guide oligonucleotide. Methods of synthesis, use, examples, and compositions of editing oligonucleotides that operate by the nucleobase chemical modification mode are described in PCT / US2015 / 65348. In FIG. 1, the linker is shown attached to the nucleobase of the editing oligonucleotide. However, this attachment may be to any nucleobase in the editing oligonucleotide, to the 5' end, to the 3' end, to the sugar residue, or to the backbone. The linker may be any linker known to those skilled in the art for use in carrying out this task. The linkers described herein may include non-covalent bonds to the "sequence-modifying moiety" (Montiel et al., PNAS 110(45):18285-90, 2013. Woolf, et al., PNAS 92:8298-8302, 1995.) Alternatively, the linker may be utilized and tested to determine its performance in an editing oligonucleotide. For example, linkers that may be utilized in the present invention include those in Table V. Linker lengths may range from 1 carbon to about 20 carbons, or equivalent lengths of other chemical structures, but are preferably less than or equal to 10 carbons in length). As an alternative to chemical covalent linkers, non-covalent connections between the editing oligonucleotide and the sequence-modifying moiety can be created (for examples, see (Montiel et al., PNAS 110(45):18285-90, 2013; Woolf, et al., PNAS 92:8298-8302, 1995; and Woolf, Nat. Biotech 16:341-344, 1998).

[0114] Successful treatment with an editing oligonucleotide in a chemically modified form results in some percentage of the "target nucleic acid" being modified. In Figure 1, the "chemical modification" (triangle) represents the addition of a chemical moiety (e.g., a methyl group), but the modification described herein may be one of a variety of additions or removals (e.g., deamination) of chemical groups from a target nucleobase of the target nucleic acid sequence.

[0115] For chemical modifications that can result in editing, see Table VII, sequence modifying moieties for nucleobase chemical modifications, and PCT / US2015 / 65348.

[0116] E. Chemical structure For chemical reactions that can lead to editing by nucleobase modification, see Table VII and PCT / US2015 / 65348.

[0117] F. Editing The present invention provides editing oligonucleotides that can reduce or eliminate the effects resulting from various mutations. If desired, editing can be reversed by administering editing oligonucleotides that return the editing to the original sequence using the methods and compositions described herein. The ability to quickly reverse editing is an important option that improves the safety of genome editing for therapeutic applications.

[0118] In one embodiment of the present invention, delta F508, a common mutation sequence that causes cystic fibrosis in Western populations, can be corrected. Repair of deletion mutations such as delta F508 can be achieved by using an editing oligonucleotide to reinsert the three deleted nucleotides. McNeer et al. (Nature Comm. DOI: 10.1038 / ncomms 7952 pgs.1-11, 2015) provides an example using an editing oligonucleotide with three phosphorothioate modifications at each end. The oligonucleotide with improved chemical modification pattern and the composition of the editing oligonucleotide of the present invention that targets the same region can be replaced with an editing oligonucleotide with three phosphorothioate modifications at each end and similar to that used by McNeer et al. However, single base transitions or transversions can be achieved more efficiently by editing compared to insertions, and thus a change of R553 to M (R553M) in the CF protein coding sequence that suppresses the deleterious effect of the delta F508 mutation is an alternative approach to correcting the phenotypic effect of this mutation (Liu et al. Biochemistry 51(25):5113-5124, 2012. doi:10.1021 / bi300018e. Another change of R555 to K (R555K) in the CF protein coding sequence suppresses the deleterious effect of the delta F508 mutation (Liu et al. supra).

[0119] Other non-limiting examples of common CFTR mutations that can be corrected by the methods and compositions herein include M470V, W1282X, G542X, Y122X, and 3849+10Kb C→T.

[0120] Another aspect of the invention includes administering editing oligonucleotides to an individual to create allelic sequences in the individual's DNA or RNA that are protective against one or more diseases (for examples, see and PCT / US2015 / 65348).

[0121] III. Treatment A. Disease See Figure 22 of (PCT / US2015 / 65348) for some target diseases, indications, and editing classes for treating such diseases and indications by the methods and compositions of the present invention.

[0122] Unless listed in Figure 22 of (PCT / US2015 / 65348), the target indications, genes, and edited oligonucleotide sequences include those described in U.S. Patent Nos. 7,258,854, 7,226,785, and U.S. Patent Application Nos. 20150118311 and 20150232881.

[0123] Non-limiting examples of editing oligonucleotides targeting genes associated with representative diseases and disorders are provided in the figures.

[0124] An exemplary (non-limiting) list of editing oligonucleotides targeting genes associated with representative diseases and disorders is shown in Figure 24 of (PCT / US2015 / 65348).

[0125] B. Pharmaceutical Compositions The pharmaceutical composition of the present invention is administered in a dosage sufficient to result in expression of the target gene. In general, a suitable dosage of the editing oligonucleotide is in the range of 0.01 to 5.0 milligrams per kilogram of the recipient's body weight per day, or up to 50 milligrams per kilogram if necessary, preferably in the range of 0.1 to 200 micrograms per kilogram of body weight per day, more preferably in the range of 0.1 to 100 micrograms per kilogram of body weight per day, even more preferably in the range of 1.0 to 50 micrograms per kilogram of body weight per day, and most preferably in the range of 1.0 to 25 micrograms per kilogram of body weight per day. The pharmaceutical composition may be administered once a day, or the editing oligonucleotide may be administered in two, three, four, five, six or more partial doses at appropriate intervals throughout the day, or even using continuous infusion. In that case, the editing oligonucleotide contained in each partial dose must be correspondingly smaller to achieve the total daily dosage. Dosage unit can also be formulated for delivery over several days, for example, by using conventional sustained release formulations that provide sustained release of edited oligonucleotide over a period of several days.Sustained release formulations are well known in the art.In this embodiment, dosage unit contains a corresponding multiple of the daily dose.

[0126] i. Dosage Certain factors may affect the dosage and timing required to effectively treat a subject, including but not limited to the severity of the disease or disorder, previous treatments, the overall health and / or age of the subject, and other diseases present. Furthermore, treatment of a subject with a therapeutically effective amount of the composition may include a single treatment or a series of treatments. Editing with programmable nucleases has been designed for one or several treatments so far because of immune responses to the programmable nucleases and vector proteins, and because cleavage by programmable nucleases destroys most target sequences by causing random insertions and deletions. The reduced immunogenicity and precision of editing (less or no random insertions and deletions) of the editing oligonucleotides described herein allows for multiple administrations, up to 3, up to 20, up to 50, or up to 100 or more administrations. Multiple administrations have a safety advantage since patients can be monitored as editing progresses over time. Multiple administrations also allow for longer treatment periods for editing cells when they are dividing, since replicating cells are more amenable to genome editing. Estimation of effective dosages and in vivo half-lives of individual editing oligonucleotides encompassed by the present invention can be made using conventional methodologies or based on in vivo testing using appropriate animal models as described elsewhere herein.

[0127] ii. Route of administration Pharmaceutical compositions encompassed by the present invention may be administered by any means known in the art (for non-limiting examples, see and PCT / US2015 / 65348).

[0128] Useful pharmaceutical compositions according to the present invention also include capsule formulations that protect the editing oligonucleotide from rapid elimination from the body, such as controlled release formulations including implants and microencapsulated delivery systems (for non-limiting examples, see PCT / US2015 / 65348). These can be prepared according to methods known to those skilled in the art, for example, as described in U.S. Pat. No. 4,522,811, PCT Application Publication No. WO 91 / 06309, and European Patent Publication No. EP-A-43075, or commercially available from Northern Lipids (Burnaby, British Columbia), Avanti Polar Lipids (Alabaster, Alabama), or Arbutus BioPharma (Burnaby, British Columbia). Nanoparticle delivery may also be used, examples of which are described in Zhou et al. Pharmaceuticals, 6:85-107, 2013, doi:10.3390 / ph6010085, McNeer et al., Gene Ther. 20(6):658-669, 2013, doi:10.1038 / gt.2012.82, McNeer et al., Nature Comm. DOI:10.1038 / ncomms 7952 pgs.1-11, 2015, and Yuen YC et al. Pharmaceuticals, 5:498-507, 2013, doi:10.3390 / pharmaceutics5030498. The oligonucleotides of the present invention may also be encapsulated in Invivofectamine 3.0 as described by the manufacturer (Thermo Fisher, Waltham, MA) or in LUNAR nanoparticles as described by the manufacturer (Arcturus, San Diego, California, US Patent Application No. 20150141678).For use in cell culture, Lipofectamine 2000 is particularly useful for delivering editing oligonucleotides, as it works in the presence of serum, allowing for delivery of oligonucleotides over a period of hours or days while cells are dividing.(Thermo Fisher, Waltham, MA).

[0129] iii. Toxicity and Efficacy Toxicity and therapeutic efficacy of such compounds can be determined by standard pharmaceutical procedures in cell cultures or experimental animals (see also PCT / US2015 / 65348).

[0130] iv. Compositions and Methods The oligonucleotides of the invention are provided in the following compositions and methods.

[0131] In one aspect, provided herein is a method of modifying a nucleic acid sequence in an isolated cell(s) within an organism, comprising introducing an oligonucleotide into the cell such that a modification(s) of complementary cellular nucleic acid occurs, said modification creating an allele that prevents disease, repairs a mutation, or inactivates a gene.

[0132] In another embodiment, the disease-preventing allele does not inactivate the function of the target gene, but rather modulates the function of the target gene.

[0133] In yet another embodiment, the method results in modulation of target gene function. Modulation of target gene function can increase activity or expression of the gene product. Modulation of target gene function can partially decrease activity or expression of the gene product.

[0134] Modulation of target gene function may partially reduce the activity or expression of the gene product by up to 50 percent in the modified cell. Modulation of target gene function may partially reduce the activity or expression of the gene product by up to 75 percent in the modified cell. Modulation of target gene function may partially reduce the activity or expression of the gene product by up to 90 percent in the modified cell.

[0135] In certain embodiments of this method, target gene product is a protein that is post-translationally modified by protease cleavage.In a specific embodiment, target gene protein is APP, and gene modification changes the sequence of APP to make it less susceptible to cleavage by beta-secretase.The sequence that codes for position 673 in APP can be changed from alanine to threonine.

[0136] Another aspect provided herein is a composition comprising the oligonucleotide of the present invention. The composition can be used in the methods described herein. In some embodiments, the oligonucleotide is present in a formulation (e.g., an editing oligonucleotide formulation). In one embodiment, the editing oligonucleotide formulation comprises an exogenous protein or ribonucleoprotein (or a nucleic acid expressing the protein or ribonucleoprotein) that increases the editing efficiency. The exogenous protein or ribonucleoprotein that increases the editing efficiency can be a programmable nuclease. The exogenous protein or ribonucleoprotein that increases the editing efficiency can be a CRISPR-Cas9, zinc finger, or Talen programmable nuclease. The editing oligonucleotide can be a single-stranded unmodified DNA. The editing oligonucleotide can be single-stranded and contain at least 10 deoxyribose sugars. The editing oligonucleotide can be chemically modified. In one embodiment, the editing oligonucleotide is attached (non-covalently or covalently) to a programmable nuclease (e.g., Talen or Cas-9) by methods known in the art to increase the local concentration of the editing oligonucleotide near the cleavage site, thus increasing the frequency of HR relative to the frequency of indels. Since indels often disrupt the function of the target gene for repair, reducing the rate of indels is useful when HR and precise editing are the desired outcome.

[0137] The chemical modification of the editing oligonucleotide may include phosphorothioate. The chemical modification of the editing oligonucleotide may include three phosphorothioate internucleotide bonds at each end. The chemical modification of the editing oligonucleotide may include a total of 1-5 phosphorothioate internucleotide bonds at the ends. The chemical modification of the editing oligonucleotide may include a total of 7 or more phosphorothioate internucleotide bonds. In one embodiment, the chemical modification of the editing oligonucleotide includes a total of 7 or more phosphorothioate internucleotide bonds, but at least 10 internucleotide bonds remain that are not phosphorothioate modified. In one embodiment, the modification does not contain any phosphorothioate modifications to reduce toxicity, which is particularly useful when using encapsulated editing oligonucleotides, since encapsulation protects the editing oligonucleotide from serum and endolysosomal nucleases. In one embodiment, the modification includes an exonuclease terminal blocking group that is not phosphorothioate.

[0138] In certain embodiments, the composition comprises an oligonucleotide having a chemically modified nucleobase. The chemically modified nucleobase(s) can be 5 methyldeoxycytidine. In some embodiments, there are 1 to about 500 5 methyldeoxycytidines. In other embodiments, there are 1 to about 50 5 methyldeoxycytidines. In other embodiments, there are 1 to about 10 5 methyldeoxycytidines. In other embodiments, there are 1 to about 5 5 methyldeoxycytidines. In other embodiments, there are 1 to about 5 5 methyldeoxycytidines. In other embodiments, there is 1 5 methyldeoxycytidine. In other embodiments, one of the 5 methyldeoxycytidines is within a 5'CpG sequence hybridized to a mismatched 5'TG, and this modification directs editing to the target strand. In a specific embodiment, the 5'TG target site is a TG at the methionine start codon, and the editing reduces or eliminates the production of a functional target protein. In other embodiments, the modification from the target nucleotide can direct editing to the target strand with little or no sequence restriction, the chemical structure being, in one embodiment, 2'F, 2'-O-alkyl, or LNA. In a more specific embodiment, to minimize toxicity, the modification at or near the editing site of the editing oligonucleotide that directs editing to the target strand is on the editing oligonucleotide without any phosphorothioate (see FIG. 2 for an example). In other embodiments, to further improve nuclease stability and increase in vivo biodistribution, the modification chemical structure that directs editing to the target strand is combined with phosphorothioate linkages (see FIG. 2 for an example). In some embodiments, the editing oligonucleotide contains >15 phosphorothioates and a modification to direct editing to the target strand at or near the editing site (see FIG. 2 for an example).In some embodiments, the editing oligonucleotide is designed to repair the deletion, and the chemical modification that directs the edit to the target strand is placed at the nucleotides immediately 5' and 3' of the editing oligonucleotide sequence that is inserted to correct the deletion (see ETAGEN serial numbers 100243-100245 in Figure 2 for an example of this modification pattern using LNA, 2'-O-methyl, and 2'F, respectively, which in this case targets the cystic fibrosis delta F508 mutation). In some embodiments, the editing oligonucleotide with the chemical modification from the target nucleobase that directs the edit to the target strand is a complex 5' phosphate modification that also improves editing efficiency.

[0139] In some embodiments, where low toxicity and high nuclease stability are desired, the editing oligonucleotide has a majority of linkages modified at the 2' modification, does not have phosphorothioates, and has chemical modifications such as 2'-O-alkyl, 2'F, or LNA from the editing site that directs editing to the target strand (see FIG. 2 for examples). In some embodiments, the editing oligonucleotides described above also have a self-delivering conjugate, which in one particularly useful embodiment includes Gal-NAc.

[0140] In some particularly useful embodiments, the editing oligonucleotides of this Compositions and Methods section are delivered to cells in vitro or in vivo in combination with a programmable nuclease of Table VIII.

[0141] In a specific embodiment of the methods and compositions described herein, the editing oligonucleotide comprises a 2' sugar modification. In another specific embodiment of the methods and compositions described herein, the editing oligonucleotide comprises only 2' modifications. In a specific embodiment, the editing oligonucleotide comprises a 2'F on the 5' side of the editing site and a 2'-O-mt on the 3' side of the editing site, or both. In another specific embodiment, the editing oligonucleotide comprises a modified base that increases affinity near the editing site, and the modified base is not a 5 methyl C.

[0142] In other specific embodiments of the methods and compositions described herein, the oligonucleotides are encapsulated within a delivery vehicle (for a description of delivery vehicles for nucleic acids, see Yin et al. Nature Reviews, Genetics. 15:541-555, 2014) and the oligonucleotides used for editing include a helper oligonucleotide listed in Table XIV or include the sequence of one of the helper oligonucleotides listed in Table XIV.

[0143] The oligonucleotide contains four or more of the optional segments, or five of these optional segments, or six of these optional segments, or all seven of these optional segments, and the oligonucleotide acts primarily in one of the manners described in Table XI below. [Table 11]

[0144] In specific embodiments of the methods and compositions described herein, the oligonucleotide further comprises a conjugation molecule that confers enhanced cellular uptake.

[0145] In specific embodiments of the methods and compositions described herein, the methods and compositions further comprise a helper oligonucleotide(s).

[0146] In certain embodiments of the methods and compositions described herein, the oligonucleotides of the invention have one or more of the improved properties and advantages listed herein and / or recited in Table XII. [Table 12]

[0147] In certain embodiments of the methods and compositions described herein, the oligonucleotides of the invention are delivered using a delivery vehicle listed herein and / or listed in Table XIII. [Table 13]

[0148] In one aspect, provided herein is an editing oligonucleotide capable of editing a complementary target sequence in a cell, the editing oligonucleotide comprising one backbone modification selected from the modifications listed in Table II, or two modifications listed in Table II, or three or more modifications listed in Table II.

[0149] In one embodiment of the editing oligonucleotide, the backbone modification is neutral. The backbone modification can include from 1 to about 20 neutral modifications. In a specific embodiment, the backbone modification includes from 2 to about 4 neutral modifications.

[0150] In another embodiment, the backbone modification is a methylphosphonate. The backbone modification can include from 1 to about 20 methylphosphonates. In a specific embodiment, the backbone modification includes from 2 to 4 methylphosphonates. In a specific embodiment, the backbone modification includes two methylphosphonates. In a more specific embodiment, the backbone modification includes two methylphosphonates on the 5' end.

[0151] In another embodiment, the editing oligonucleotide may contain from 1 to about 20 backbone modifications in a single modified backbone editing oligonucleotide. In one embodiment, at least two of the modifications are in the terminal segment. In a specific embodiment, the editing oligonucleotide contains two modifications at the 5' end.

[0152] In another aspect, provided herein is a method of editing a gene in a cell or organism using the editing oligonucleotides described herein. In one embodiment, the cell is an isolated human cell. In one embodiment, the organism is a human. In certain embodiments, the method is used to treat an indication selected from the indications listed in Figure 23 of (PCT / US2015 / 65348). In a specific embodiment, the indication is selected from the indications listed in Figure 24 of (PCT / US2015 / 65348).

[0153] In one embodiment, the gene is a target gene listed in Figure 23 of (PCT / US2015 / 65348). In a specific embodiment, the editing oligonucleotide comprises at least 25 percent of the sequence listed in Figure 24 of (PCT / US2015 / 65348). In another specific embodiment, the editing oligonucleotide comprises at least 51 percent of the sequence from Figure 24 of (PCT / US2015 / 65348).

[0154] In another aspect, an editing oligonucleotide is provided herein, the editing oligonucleotide being capable of editing a complementary target sequence in a cell, the editing oligonucleotide comprising one or more nucleobase modifications listed in Table III, Nucleobase Modifications Useful for Oligonucleotides. In one embodiment, the editing oligonucleotide comprises 1 to about 100 modified nucleobases from Table III. In one embodiment, the editing oligonucleotide comprises 1 to about 30 modified nucleobases from Table III. In one embodiment, the editing oligonucleotide comprises 1 to about 10 modified nucleobases from Table III. In a specific embodiment, the editing oligonucleotide comprises one or more modified nucleobases according to the modification pattern types in Table III.

[0155] In another embodiment of the editing oligonucleotide, the modified nucleobase reduces immune stimulation by the editing oligonucleotide in a mammal. In a specific embodiment, the modified nucleobase comprises a 5' methyl C chemical modification. In another specific embodiment, the nucleobase modification increases the affinity of the editing oligonucleotide for its complementary target.

[0156] In another aspect, provided herein is an editing oligonucleotide capable of editing a complementary target sequence in a cell, the editing oligonucleotide comprising one or more sugar modifications listed in Table IV, Useful Sugars for Oligonucleotides. In one embodiment, the sugar modification is selected from a 2' sugar modification. The 2' sugar modification can be 2'F. The 2' sugar modification can be 2'O-methyl. The 2' sugar modification can be a combination of 2'F and 2'O-methyl modifications.

[0157] In one embodiment of the editing oligonucleotide, the majority (e.g., greater than 50%) of the 2'F modifications are 3' to the editing site. In another embodiment of the editing oligonucleotide, the majority (e.g., greater than 50%) of the 2'O-methyl modifications are 5' to the editing site.

[0158] The 2' sugar modification can increase the affinity of the oligonucleotide for its target nucleic acid. In one embodiment, the edited oligonucleotide comprises 1-75 sugar modifications. In another embodiment, the edited oligonucleotide comprises 2-30 sugar modifications. In another embodiment, the edited oligonucleotide comprises 2-16 sugar modifications.

[0159] In one embodiment, the editing oligonucleotide comprises about 5-100% chemically modified bases. In another embodiment, the editing oligonucleotide comprises about 25-75% chemically modified bases. In another embodiment, the editing oligonucleotide comprises about 40-60% chemically modified bases. In a specific embodiment, the editing oligonucleotide comprises two modifications. In a specific embodiment, the editing oligonucleotide contains 2'F and 2'O-methyl modifications.

[0160] In one embodiment, the editing oligonucleotide targets a gene listed in Figure 23 of (PCT / US2015 / 65348).

[0161] In another aspect, provided herein is a method of treating a human disease by genome editing, the method comprising administering to an individual in need of such treatment an editing oligonucleotide described herein.

[0162] In yet another aspect, an editing oligonucleotide comprising one or more delivery conjugates is provided herein. In a specific embodiment, the editing oligonucleotide comprises one delivery conjugate. The delivery conjugate can enhance cellular uptake of the oligonucleotide. The delivery conjugate can enhance intracellular uptake of the oligonucleotide in an organism. The delivery conjugate can be a chemical moiety that is directly or indirectly covalently attached to the editing oligonucleotide. Direct covalent attachment includes, for example, covalent attachment of the chemical moiety to the oligonucleotide. Indirect covalent attachment includes, for example, the use of a linker that is covalently attached to both the oligonucleotide and the chemical moiety. In one embodiment, the editing oligonucleotide is not encapsulated by a delivery vehicle that enhances intracellular uptake in an organism.

[0163] In one embodiment, the delivery conjugate is a ligand for a receptor. In a particularly useful embodiment, the ligand is 1-10 Gal-Nac. In another specific embodiment, the ligand is 3 Gal-Nac. In one embodiment, the delivery conjugate is a lipophilic group. The lipophilic group can have from about 10 to about 50 carbons. The lipophilic group can be a form of cholesterol.

[0164] Editing oligonucleotides comprising one or more delivery conjugates are useful for the treatment of disease. In one aspect, provided herein is a method of treating or preventing a human disease by administering to a patient in need of such treatment an editing oligonucleotide comprising one or more delivery conjugates described herein. In one embodiment, the method targets a gene for editing, and the target gene is one listed in Figure 23 of (PCT / US2015 / 65348). In another embodiment, the target gene for treatment is one listed in Figure 24 of (PCT / US2015 / 65348). In one embodiment of the method, the editing oligonucleotide sequence comprises one of the sequences in Figure 24 of (PCT / US2015 / 65348).

[0165] Compositions and methods using encapsulated editing oligonucleotides. One embodiment is an editing oligonucleotide with 8 or less sequence differences compared to a genomic DNA target sequence. Each of the sequence differences is a mismatch that will result in a transition or transversion, insertion and / or deletion of a nucleotide(s) compared to the target sequence, and the editing oligonucleotide edits the target genomic DNA sequence to the sequence of the editing oligonucleotide, and the resulting edit has a therapeutic effect on the organism (or ex vivo treated cells) treated with the editing oligonucleotide, or a desired change useful for studying the target cell or organism. The editing oligonucleotide may further have 1-4 exonuclease blocking group(s) at its 5'-end and / or 3'-end, and / or; have less than 8 phosphorothioate modifications; be encapsulated in a nanoparticle; have one or more nucleobase chemical modifications that reduce immune stimulation; and / or have chemical modifications that prevent endogenous mismatch repair mechanisms from editing the editing oligonucleotide at or near the sequence difference with the target genomic DNA.

[0166] In certain embodiments, the editing oligonucleotide has 6 or less sequence differences, 4 or less sequence differences, 2 or less sequence differences, or 1 sequence difference compared to the genomic DNA target sequence. In any of the above embodiments, the editing oligonucleotide whose sequence difference compared to the genomic DNA target sequence is a mismatch is an editing oligonucleotide when interacting with the DNA target sequence resulting in transition(s) and / or transversion(s). Alternatively, the editing oligonucleotide whose sequence difference compared to the genomic DNA target sequence is an editing oligonucleotide when interacting with the DNA target resulting in the insertion of nucleotide(s) into the target genomic DNA or the deletion of nucleotide(s) in the target genomic DNA is an editing oligonucleotide.

[0167] The editing oligonucleotide may have one to four exonuclease blocking group(s) at its 5' end and zero exonuclease blocking group(s) at its 3' end; three exonuclease blocking group(s) at its 5' end and zero exonuclease blocking group(s) at its 3' end; two exonuclease blocking group(s) at its 5' end and zero exonuclease blocking group(s) at its 3' end; one exonuclease blocking group(s) at its 5' end and zero exonuclease blocking group(s) at its 3' end. exonuclease blocking group; 1 to 4 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group at its 5' end; 3 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group at its 5' end; 2 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group at its 5' end; or 1 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group at its 5' end.

[0168] Further, the exonuclease blocking group(s) is a phosphorothioate linkage, a non-phosphorothioate linkage, a non-nucleotidic linker, an amino linker, a C2-C9 amino linker, a C3 amino linker, or a constrained nucleic acid. The exonuclease blocking group(s) may have a 2' sugar modification (including a constrained nucleic acid). When the nuclease blocking group is a constrained nucleic acid, it is a 2',4'-BNA, cET, or non-LNA, or an LNA.

[0169] The editing oligonucleotide chemical modification(s) that prevent the endogenous mismatch repair machinery from repairing the editing oligonucleotide are 2' sugar modification(s) that include constrained nucleic acid, 2' sugar modification(s) that include constrained nucleic acid but excluding LNA(s), 2' sugar modification(s) that exclude 2'F, LNA, or 2'F sugar modification.

[0170] The encapsulated editing oligonucleotide may be administered intravenously to an organism; 10 or more times to achieve a therapeutically significant editing; or 20 or more times to achieve a therapeutically significant editing.

[0171] The edited oligonucleotides can be used to provide a therapeutic effect to an organism treated with the edited oligonucleotide (or ex vivo treated cells) or to a human treated with the edited oligonucleotide (or ex vivo treated human cells).

[0172] The editing oligonucleotide or the method of using the editing oligonucleotide additionally has a 2' sugar modification between the terminal or terminal exonuclease blocking group and the modification proximal to the sequence difference with the target genomic DNA that prevents the endogenous mismatch repair mechanism from repairing the editing oligonucleotide. The additional 2' sugar modification is both 5' and 3' to the editing site, only 3' to the editing site, or only 5' to the editing site. In certain embodiments, the 2' sugar modification 3' to the editing site is 2'F or the 2' sugar modification 5' to the editing site is 2'-O-methyl.

[0173] The editing oligonucleotide may further be combined with an additional oligonucleotide having a 5' phosphate or a nuclease-stable analogue thereof, which binds genomic DNA within 200 nucleotides of the editing site to improve the editing efficiency of the editing oligonucleotide, or may be combined with a PNA oligonucleotide which binds genomic DNA within 200 nucleotides of the editing site to improve the editing efficiency of the editing oligonucleotide.

[0174] The editing efficiency of the editing oligonucleotides utilized in the methods of the present invention has an improved editing efficiency by cleaving within 200 nucleotides of the target site with a programmable nuclease.

[0175] The edited oligonucleotides and methods of using the edited oligonucleotides of the present invention can be used to treat diseases including β-thalassemia, cystic fibrosis or Duchenne muscular dystrophy, Alzheimer's disease, type 2 diabetes, sickle cell disease, and β-thalassemia.

[0176] The editing oligonucleotide may be targeted to the sense strand of genomic DNA or the antisense strand of genomic DNA.

[0177] Compositions and methods using non-encapsulated editing oligonucleotides include an editing oligonucleotide having eight or fewer sequence differences compared to a genomic DNA target sequence, each of the sequence differences being a mismatch that will result in a transition or transversion, an insertion and / or deletion of a nucleotide compared to the target sequence, and the editing oligonucleotide edits the target genomic DNA sequence to the sequence of the editing oligonucleotide, and the resulting edit has a desired change that has a therapeutic effect on an organism (or ex vivo treated cells) treated with the editing oligonucleotide or is useful for studying the target cell or organism. Additionally, the editing oligonucleotide has one to four exonuclease blocking group(s) at its 5' and / or 3' end, is not encapsulated within a nanoparticle or other delivery vehicle, and has one or more nucleobase chemical modifications that reduce immune stimulation and chemical modifications at or proximal to the sequence difference with the target genomic DNA that prevent endogenous mismatch repair mechanisms from editing the editing oligonucleotide.

[0178] The editing oligonucleotide of the above embodiment has 6 or less sequence differences compared to the genomic DNA target sequence, 4 or less sequence differences compared to the genomic DNA target sequence, 2 or less sequence differences compared to the genomic DNA target sequence, or 1 sequence difference compared to the genomic DNA target sequence. The sequence difference of the editing oligonucleotide compared to the genomic DNA target sequence: by having a mismatch that will result in transition(s) and / or transversion(s); will result in the insertion of nucleotide(s) into the target genomic DNA; will result in the deletion of nucleotide(s) in the target genomic DNA.

[0179] The editing oligonucleotide may have 1 to 4 exonuclease blocking group(s) at its 5' end and 0 exonuclease blocking group at its 3' end, 3 exonuclease blocking group(s) at its 5' end and 0 exonuclease blocking group at its 3' end, 2 exonuclease blocking group(s) at its 5' end and 0 exonuclease blocking group at its 3' end, 1 exonuclease blocking group(s) at its 5' end and 0 exonuclease blocking group at its 3' end, exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group(s) at its 5' end; 3 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group(s) at its 5' end; 2 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group(s) at its 5' end; or 1 exonuclease blocking group(s) at its 3' end and 0 exonuclease blocking group(s) at its 5' end.

[0180] The exonuclease blocking group(s) can be phosphorothioate linkages, non-phosphorothioate linkages, non-nucleotidic linkers, amino linkers, C2-C9 amino linkers, or C3 amino linkers. The exonuclease blocking group(s) can have a 2' sugar modification (including constrained nucleic acids) or constrained nucleic acids. The constrained nucleic acids can be 2',4'-BNA, LNA, cET, or non-LNA constrained nucleic acids.

[0181] An editing oligonucleotide in which the chemical modification(s) that prevent the endogenous mismatch repair mechanism from repairing the editing oligonucleotide include a 2' sugar modification(s) and may further contain a constrained nucleic acid but may exclude LNA(s) and / or 2' F. Alternatively, the chemical modification(s) that prevent the endogenous mismatch repair mechanism from repairing the editing oligonucleotide may be an LNA or 2' F sugar modification.

[0182] The editing oligonucleotide of the present invention may be administered to an organism by subcutaneous injection, intravenous injection or infusion, intravitreal injection, or inhalation. The editing oligonucleotide may be administered 10 or more times to achieve therapeutically significant editing, or 20 or more times to achieve therapeutically significant editing. Editing resulting from the editing oligonucleotide may have a therapeutic effect on the organism (or ex vivo treated cells) treated with the editing oligonucleotide or on the human (or ex vivo treated human cells) treated with the editing oligonucleotide.

[0183] The methods and compositions of the invention include editing oligonucleotides that have an additional 2' sugar modification between the terminal or terminal exonuclease blocking group and a modification that prevents endogenous mismatch repair mechanisms from repairing the editing oligonucleotide proximal to the sequence difference with the target genomic DNA. The additional 2' sugar modification can be both 5' and 3' to the editing site, only 3' to the editing site, or only 5' to the editing site. In other embodiments, the 2' sugar modification 3' to the editing site is 2'F or 5' to the editing site is 2'-O-methyl. The editing oligonucleotide can contain a 5' phosphate or nuclease stable analogue thereof, a phosphorothioate modification at every internucleotide linkage, a phosphorothioate modification at >90% of the internucleotide linkages, a phosphorothioate modification at >40% of the internucleotide linkages, a phosphorothioate modification at 8 or more internucleotide linkages, or another endonuclease resistant modification at 40% or more of the non-phosphorothioate linkages. The editing oligonucleotide may have a 2' sugar modified endonuclease resistant modification at 40% or more of the non-phosphorothioate linkages; a 2'-O-methyl modified endonuclease resistant modification at 40% or more of the non-phosphorothioate linkages; or a BNA modified endonuclease resistant modification at 20% or more of the non-phosphorothioate linkages. In one embodiment, the BNA modification is LNA or cET.

[0184] In other embodiments, the editing oligonucleotide may be conjugated to a ligand that enhances delivery to a target cell and / or tissue, the ligand comprising one or more Gal-Nac residues, the ligand that enhances delivery to a target cell and / or tissue, the ligand comprising a lipophilic group, or the ligand that enhances delivery to a target cell and / or tissue, the ligand comprising cholesterol or a cholesterol analog. Additionally, the editing oligonucleotide may be combined with an additional oligonucleotide that binds genomic DNA within 200 nucleotides of the editing site to improve the editing efficiency of the editing oligonucleotide, or with a PNA oligonucleotide that binds genomic DNA within 200 nucleotides of the editing site to improve the editing efficiency of the editing oligonucleotide.

[0185] The editing efficiency of the editing oligonucleotide can be improved by cleaving within 200 nucleotides of the target site with a programmable nuclease. Diseases that can be treated with the editing oligonucleotides and methods of the invention include β-thalassemia, cystic fibrosis or Duchenne muscular dystrophy, Alzheimer's disease, type 2 diabetes, sickle cell disease, or β-thalassemia.

[0186] The editing oligonucleotide may be targeted to the sense strand of genomic DNA or the antisense strand of genomic DNA.

[0187] V. Results Even though their activity in cell culture may be less than that of the parent compound, the oligonucleotide structures and modification patterns presented in FIG. 2 are useful for the research, therapeutic, and other applications described herein because each of these oligonucleotides contributes predicted therapeutic benefits, such as reduced immune stimulation, increased nuclease stability, increased target specificity, reduced chemical toxicity, and / or increased affinity, compared to unmodified DNA or DNA with three phosphorothioates on each end.

[0188] The examples of editing oligonucleotides listed in FIG. 2 have various features that improve their usefulness in genome editing. Phosphorothioate bonds allow serum protein binding to improve serum half-life and tissue distribution, increased nuclease stability compared to terminal blocks alone, and cellular uptake into the cytoplasm without a delivery vehicle. More phosphorothioates result in increased nuclease stability. Uniform (all) phosphorothioate-substituted oligonucleotides are stable enough to be efficiently taken up without a delivery vehicle and effectively survive passage through the high nuclease endolysosomal compartment. However, uniform phosphorothioate substitutions also reduce the editing efficiency of one delivered to the inside of the cell, so there is a trade-off between the number of phosphorothioate salts and phosphodiesters. Extensive phosphorothioate substitutions are more tolerant of editing on the 3' half of the editing oligonucleotide up to the editing site, which is the rationale for the exemplary oligonucleotides with phosphorothioate bonds on the 3' approximately half of the editing oligonucleotide. Another design to reduce phosphorothioate linkages places other modifications at the phosphodiester linkage that are tolerant to editing, as exemplified by designs with 2'F phosphodiester segments toward the three parts of the editing oligonucleotide. Phosphorothioate substitutions at or near the editing site will also suppress non-productive mismatch repair.

[0189] The five key (5') phosphates, like all phosphorothioates, are designed to improve editing efficiency, especially for short (<30mer) and stable sequences (Radecke et al. 2006). The 5' thiophosphate is designed to further stabilize against removal by cellular phosphatases. Shorter oligonucleotides have improved free uptake into cells, simplifying synthesis and reducing the cost of synthesis, which is balanced against the improved editing efficiency of some longer oligonucleotides. The 5-methyl C modification improves affinity for the target, reduces immune stimulation, and directs editing to the target strand. 5' non-phosphorothioate end blocks such as linkers or linkers containing fluorescent conjugates (e.g., 6FAM) are less toxic and more tolerant to editing than phosphorothioate end blocks. The four 3' phosphorothioate end blocks, which are not complimentary to the target DNA, are pruned off by the cellular machinery, allowing for natural 3' OH ends. Two to five non-complementary 3' end blocks are particularly useful.

[0190] Cholesterol conjugates improve cellular uptake, especially when used for highly nuclease-stable oligonucleotides that can survive endolysosomal nucleases.For liver uptake, Gal-NAc conjugates, known in the art to function in vivo with siRNA and antisense, can replace cholesterol as delivery conjugates.A stretch of 3' unmodified nucleobases can be inserted at the end next to the conjugate, allowing cellular nucleases to cut the conjugate, liberating the free 3'OH required for most efficient editing.

[0191] Two major (2') modifications at or proximal to the "editing site" (i.e., 2'F, LNA, 2'-O-methyl) suppress non-productive mismatch repair of the editing oligonucleotide, improving editing efficiency, or increasing hybrid affinity (improving efficacy), increasing nuclease stability, and reducing immune stimulation.

[0192] Three main (3') F segments improve editing efficiency (increase affinity) and reduce immune stimulation. LNA end modification improves nuclease stability, increases hybrid affinity, resulting in higher editing efficiency and reduced immune stimulation, and reduces toxicity. However, two or more LNA bonds at the end may reduce editing efficiency.

[0193] Examples of helper PNA editing oligonucleotides listed in Figure 2 herein have various features that improve their usefulness in genome editing. PNA binding confers nuclease stability and high affinity to the target. High target affinity promotes strand invasion. Lysine at the end improves solubility, improves strand invasion, and allows "self-delivery" to cells in culture and in vivo without a delivery vehicle. Substitution of gamma miniPEG or gamma glutamate PNA improves solubility and target affinity. Glutamate confers a negative charge that allows efficient encapsulation when used with a positively charged delivery vehicle (see Figure 2 for an example).

[0194] Editing oligonucleotide 100034 has 5' and 3' 2'F arms (5' and 3' proximal segments) and shows low but significant editing. This was unexpected because 2'F is sterically more similar to DNA than 2'-O-methyl, which was highly active in editing when incorporated into the 3' arm. In the 5' arm, the 2'-O-methyl modification was better tolerated than the 2'F modification, which was also unexpected. This suggests that constructs such as editing oligonucleotide 100058 are particularly useful compared to constructs with the same modifications in each arm.

[0195] A wide range of modifications, such as those found in the editing oligonucleotide 100047, are compatible with editing, which is useful because each of the modifications reduces the predicted toxicity compared to the parent oligonucleotides often used in the art (5' and 3' phosphorothioate DNA exonuclease blocking end segments are commonly used in the art). Part of this reduced toxicity is believed to be due to the reduced activation of Toll-like receptors by the 2' modified linkage compared to 2'H in DNA or 2'OH in RNA. Higher target specificity is achieved because the arms do not function as efficient editing sites, and therefore the potential for off-target editing is reduced.

[0196] While the 5' methylphosphonate exonuclease blocking end segment was quite active, the use of both 5' and 3' methylphosphonate end segments was useful but less active. This construct removed all phosphorothioates, which were associated with blocking cell proliferation in many in vitro assays.

[0197] A single 5-methyl C modification near the editing site was consistent with relatively high editing efficiency, as were multiple 5-methyl C modifications.

[0198] Extending the span of 3' proximal segment modifications toward the 3' edited segment may be less favorable due to interference with the editing reaction, but these additional modifications are predicted to further increase nuclease stability and reduce immune stimulation (e.g., edited oligonucleotide 100062). This is also true for 5' modifications (e.g., edited oligonucleotide 100066).

[0199] Longer editing oligonucleotides have more linkages that can be modified in the 5' or 3' editing segment or at positions away from the editing site (e.g., editing oligonucleotide 100072).

[0200] Although methylphosphonates provided an excellent 5' end block, editing oligonucleotide 100074 has a 5' end block of CY3 and three complimentary phosphorothioate DNAs on the 3' end, providing another useful modification combination.

[0201] Locked nucleic acids (LNAs) can also be used as end-blocking groups, but they may increase in vivo toxicity. For this reason, embodiments using unlocked nucleic acids (UNAs) (e.g., editing oligonucleotide 100078) or simple linkers at one or both ends as nuclease end blocks are particularly useful (see Table V). Exonucleases can jump over single modifications of DNA, but in combination with 2' modified end residues (e.g., editing oligonucleotide 100080), this problem can be mitigated. End-blocking linkers have the added advantage that they can also be used to link conjugates to editing oligonucleotides. These conjugates (e.g., conjugates with cholesterol (U.S. Patent Application No. 20130131142 A1) and Gal-Nac (U.S. Patent No. 8,106,022) have been shown to increase cellular uptake of oligonucleotides in culture and in vivo in animals. Conjugates of these moieties with editing oligonucleotides can be prepared using methods known in the art and would eliminate the need for delivery vehicles (e.g., liposomes), which add cost and / or toxicity.

[0202] Editing oligonucleotide 100082 contains a complimentary end block to editing oligonucleotide 100005, 100031, and other oligonucleotides in this modified series. Editing oligonucleotides may be added to cells separately from the complimentary oligonucleotides, or may be prehybridized with the complimentary editing oligonucleotides of any of the modified chemical structures described herein to form a duplex. The advantage of preformed duplexes is that double-stranded DNA is resistant to single-stranded nucleases. However, a disadvantage of duplexes is that the bases may not be free to hybridize with the target DNA unless some cellular repair / recombination mechanism promotes target binding. Depending on the target gene, cell type, and route of administration, single-stranded or double-stranded editing oligonucleotides may be more suitable for editing.

[0203] Editing oligonucleotide 100083 is an RNA protector oligonucleotide with complimentary end blocks to editing oligonucleotide 100005, 100031, and others in the series of chemically modified editing oligonucleotide targeting GFP disclosed herein. This oligonucleotide protects the complimentary editing guide oligonucleotide from nucleases in serum, endolysosomal pathways, and cytoplasm. When in the cytoplasm or nucleoplasm, the RNA strand is eventually degraded by endogenous RNase H, and the single-stranded editing oligonucleotide is released for hybridization to target DNA. This is an improvement over 2'-O-methyl protected oligonucleotides, which likely reduced the activity of editing oligonucleotides by interfering with hybridization to target DNA.

[0204] Editing oligonucleotide 100085 is designed such that when the 5'-proximal region is hybridized to protector oligonucleotide 10086, it forms a duplex that can be packaged into an RNA-induced silencing complex (RISC). The hybridization rate of the guide strand to the complementary target nucleic acid (both RNA and DNA targets; Saloman, et al. Cell 162:84-96, 2015) is dramatically increased as a result of being packaged into the Argonaute. This increase in hybridization on-rate by RISC makes the siRNA approximately 10-100 times more potent than the corresponding antisense (e.g., no enhancement by RISC is observed with antisense). Thus, the 5' end of the editing oligonucleotide packaged into the Argonaute hybridizes more rapidly to the target chromosomal DNA, increasing the potency and / or efficiency of genome editing. This mechanism uses endogenous cellular machinery to enhance target binding and therefore does not require the addition of exogenous proteins such as Cas9 or Natronobacterium gregorii Argonaute to enhance target binding. The advantage of this embodiment of the invention is that it avoids the challenges of delivering exogenous proteins to cells. Once binding to the target DNA is seeded by the 5'-proximal region of the editing oligonucleotide complexed with Argonaute, the remaining duplex is rapidly generated.

[0205] Based on the data herein for the editing oligonucleotide 100037, the 5'-terminal region of the editing oligonucleotide can be modified with 2'-O-methyl RNA while maintaining the editing efficiency, and partial modification of the oligonucleotide with 2'-O-methyl modification is compatible with the requirements of RISC loading (US Patent Application Nos. 20130317080, 20150267200, 20150105545, 20110039914, and US Patent Application No. 12 / 824,011). The protector oligonucleotide (passenger strand) is preferably 10-50 nucleotides, more preferably 12-30 nucleotides, and most preferably 19-27 nucleotides, and is fully or substantially complementary to the target. In this structure, the editing oligonucleotide is designed according to some generally accepted design rules for the preparation of RNAi or microRNA (miRNA). For example, a 2 base pair 3' overhang on the passenger strand is particularly useful, but blunt ends and other terminal structures compatible with RNAi are also useful. A free 5' hydroxyl or phosphorylated 5' hydroxyl on the guide (edited strand) is also provided. A range of chemical modifications and structures compatible with RNAi can be used in this editing strategy. It is particularly useful that the 5' end of the editing oligonucleotide that forms a duplex with the passenger RNA does not have more than about 4 DNA bonds attached to the RNA in the passenger strand, which may activate RNase H cleavage of the passenger strand and reduce RISC loading. In particularly useful embodiments, when using a RISC-filled double-stranded region within an editing oligonucleotide that is long enough to function as a Dicer substrate, chemical modifications or mismatches may be inserted in a manner known in the art to reduce or eliminate Dicer cleavage, such as incorporation of 2'-O-methyl modification(s) at the Dicer cleavage site(s) (Salomon et al. Nucleic Acids Research, 38(11):3771-9 Feb. 2010).Reduced Dicer cleavage is beneficial because Dicer cleavage of the editing oligonucleotide at the side of the duplex closest to the editing site separates the RISC filling region from the remainder of the editing oligonucleotide, eliminating the advantage of RISC filling if this occurs prior to strand invasion into the target DNA duplex.

[0206] Double-stranded structures that can be packed into RISC are known in the art and include STEALTH RNAi compounds (Life Technologies, San Diego CA and U.S. Pat. No. 8,815,821), Dicer substrates (U.S. Pat. Nos. 8,349,809, 8,513,207, and 8,927,705), rxRNA ori (RXi These include RNAi triggers with truncated duplexes (U.S. Patent Application No. 20120065243, filed in 2009), and siRNAs (U.S. Patent Nos. 7,923,547, 7,956,176, 7,989,612, 8,202,979, 8,232,383, 8,236,944, 8,242,257, 8,268,986, 8,273,866, and U.S. Patent Application No. 13 / 693,478). These RNAi trigger constructs can be incorporated into editing oligonucleotides, as is done with the siRNAs in the RISC-loading editing oligonucleotides described herein (ETAGEN accession number 100085 hybridized to 100086), so long as a free 5' hydroxyl or phosphorylated 5' hydroxyl on the edited strand is maintained or made free within the cell by various chemical modification patterns known to support RISC loading and, in some cases, enhance tissue and cellular uptake.

[0207] Additional examples of editing oligonucleotides are listed in Figure 2 and helper oligonucleotides in Table XIV, which include various features described above. The chemical modification patterns, lengths, and configurations of the editing oligonucleotides and optional helper oligonucleotides described in Figure 2 can be applied to a variety of editing targets, including those listed herein, using the methods described herein. The number and location of each chemical modification can be varied as described herein.

[0208] VI. Advantages The embodiment of the present invention using the Kmiec method has several advantages over chemical modification methods, because the Kmiec method does not require a chemical reactive group to be attached to the editing oligonucleotide, and allows editing of bases to any other natural base to be achieved, and insertions or deletions to be created. These "footprint-free" edits can be more easily reversed, if necessary, by performing additional "footprint-free" edits back to the original sequence, an important safety feature for genome editing therapy.

[0209] Embodiments of the present invention that employ chemical modification methods have several advantages over the Kmiec method because chemical modification methods involve the addition or removal of specific groups (i.e., methylation, ethylation, or deamination) to alter the base-pairing specificity of target nucleobases and therefore do not require active cellular recombination machinery.

[0210] The editing oligonucleotide of the present invention can be used without CRISPR or proteins such as zinc finger or modified programmable nuclease.The method of using CRISPR and / or zinc finger uses a single-stranded oligonucleotide that is not the guide RNA in CRISPR but is a separate single-stranded oligonucleotide as a donor to repair the site.However, the method and composition of the present invention do not strictly require these other exogenous protein components and provide precise editing with the same or substantially similar efficiency as current methods.

[0211] The present invention is a nucleic acid repair approach that differs from the approaches that strictly require CRISPR / Cas9, zinc finger, and Talen DNA nucleases, because it directly and precisely repairs mutant sequences without the need to create potentially dangerous breaks in DNA. In addition, some embodiments of the present invention may be administered, optionally, without the use of delivery particles or immunogenic proteins.

[0212] The present invention can be used to permanently inactivate any gene by creating a site-specific mutation, e.g., a stop codon, at a desired location that prevents translation. One unique application of the present invention is to target point mutations that modulate or correct gene function (e.g., gain-of-function mutations caused by dominant mutations) that cannot be addressed by other known silencing methods.

[0213] Other approaches, such as competitive gene therapy and mRNA replacement strategies, can replace mutant gene products. However, certain embodiments of the invention have the advantage of achieving completely normal gene regulation and expression levels without causing integration of vector sequences or chromosomal damage at the vector insertion site.

[0214] It is understood that any of the above methods may or may not be used in combination with certain other methods herein. Furthermore, any of the above compositions may be optionally used with certain methods described herein and / or combined with other compositions herein. Furthermore, the improved features of the editing oligonucleotide and optional helper oligonucleotide compositions described herein (i.e., chemical modifications, structures such as hairpins and delivery vehicles) may be used in combination with other improved features of the editing oligonucleotide and optional helper oligonucleotide compositions described herein.

[0215] VII. Working Examples FIG. 2 describes examples of editing oligonucleotides of the invention. Some of the editing oligonucleotide sequences in FIG. 2 target null mutations in green fluorescent protein and correct the mutations to functional sequences that can be easily monitored by fluorescent assays (Erin E. Brachman and Eric B. Kmiec, supra). Other oligonucleotides in FIG. 2 target other genes. However, the chemical modification patterns in FIG. 2 can be applied to editing oligonucleotides that target mutations, create protective alleles, or create other desired genomic changes in other target genes. In each case of these examples, unless already determined by the disclosed sequences, and in other embodiments of the invention, the editing site may be in the central region of the editing oligonucleotide, or may be offset toward the 5' or 3' end. In particularly useful embodiments, the editing site is more than 5 nucleotides away from either end. It is also preferred to have editing sites within regions of DNA that are unmodified or, if modified, have modifications (i.e., phosphorothioate, 2'F, LNA, 2'-O-methyl, or 5' methyl C) that still are recognized as template DNA by the cellular machinery that repairs the target DNA and / or the replication machinery.

[0216] Editing oligonucleotides can be designed to be complementary to either strand of genomic DNA. It is particularly useful to design them to bind to the template strand for lagging strand synthesis, as this tends to result in more efficient editing. However, each strand can be targeted, and it can be easily determined which strand results in more efficient editing. Phosphorothioate backbone modification patterns and lengths of editing oligonucleotides useful in the present invention can be found in PCT / US2015 / 65348. [Table 14]

[0217] The general methodology employed in the experiments used to generate the editing efficiency data in Figure 2. A. Cell lines and culture conditions For GFP targeting, genetically modified HCT116 cells were used. HCT116 cells were obtained from ATCC (American Type Cell Culture, Manassas, VA). HCT116-19 was created by incorporating pEGFP-N3 vector (Clontech, Palo Alto, CA) containing a mutant eGFP gene. The mutant eGFP gene has a nonsense mutation at position 167 that results in a nonfunctional eGFP protein. In these experiments, HCT116-19 cells were cultured in McCoy's 5A modified medium (Thermo Scientific, Pittsburgh, PA) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, and 1% penicillin / streptomycin. Cells were maintained at 37°C and 5% carbon dioxide.

[0218] B. Transfection of HCT116-19 cells In the experiments using synchronized cells, HCT116-19 cells were plated at 2.5 × 10 6 Cells were seeded and synchronized with 6 mM aphidicolin for 24 h prior to targeting. Cells were released for 4 h (or as indicated) and then trypsinized and transfected by washing with PBS (2 / 2) and adding complete growth medium. Synchronized and unsynchronized HCT116-19 cells were cultured at 5 × 10 in 4 mm gap cuvettes (BioExpress, Kaysville, UT). 5Transfection was performed using approximately 1ug of edited oligonucleotide at a concentration of cells / 100ul. Single-stranded oligonucleotides were electroporated (250V, LV, 13ms pulse length, 2 pulses, 1s interval) using a Bio-Rad Gene Pulser XCell™ electroporation system (Bio-Rad Laboratories, Hercules, CA). Cells were then allowed to recover for the indicated times at 37°C in 6-well plates containing complete growth medium and subsequently analyzed. Analysis of gene-edited cells. Fluorescence (eGFP) was measured by a Guava EasyCyte 5 HT™ flow cytometer (Millipore, Temecula, CA). Cells were harvested by trypsinization, washed once with PBS, and resuspended in buffer (PBS containing 0.5% BSA, 2mM EDTA, 2mg / mL propidium iodide (PI)). Cell viability was measured as such using propidium iodide, and viable cells stain negative for PI (uptake). Correction efficiency was calculated as the percentage of total eGFP-positive live cells compared to total live cells in each sample. Error bars are generated from triplicate data points using standard error calculations (for methods, see: Bialk P, Rivera-Torres N, Strouse B, Kmiec EB. PLoS One. 2015;10(6):e0129308).

[0219] Examples of methods for optimizing the length and position of editing oligonucleotides can be found in and PCT / US2015 / 65348.

[0220] Example 1 Methods for preparing editing oligonucleotides with optimized editing activity in cells The editing oligonucleotides are easily tested for editing in cells to obtain an optimized editing oligonucleotide sequence using the following steps. A. Define the editing oligonucleotide 3 nucleotides 5' to the nucleobase of the editing target (or the 3'-most target nucleotide, if more than one target is present in this region) as the 5'-terminal complementary nucleotide of the editing oligonucleotide sequence, move in one nucleotide increments to add a single 5' nucleotide extension of the editing oligonucleotide, and repeat the above process 30 times, or up to 50 times, or up to 200 times. B. Define the 3' end of the editing oligonucleotide and perform the same process as above except to the 3' of the editing target nucleobase (or the 5'-most target nucleotide if multiple targets are present in this region), move in increments of 1, 2, 5, or 10 nucleotides, add a 1, 2, 5, or 10 3' nucleotide extension of the editing oligonucleotide, and repeat the above process 30 times, or up to 50 times, or up to 200 times. C. A matrix of all the resulting 5' and 3' terminal nucleotides of the editing oligonucleotides is generated and editing oligonucleotide sequences less than 12 nucleotides are eliminated as they are unlikely to be unique within the genome. All remaining sequences less than 30, 50, or 200 nucleotides are then selected and their editing activity in cells is tested to determine the most efficient editing oligonucleotide sequences.

[0221] Example 3 Design of helper oligonucleotide sequences Designing the helper oligonucleotide sequence is carried out starting with a helper oligonucleotide that is at least 8 bases long. For helper oligonucleotides, the particularly useful length of the W / C binding region is less than 25 nucleotides or less than 50 nucleotides. For helper oligonucleotides, the sequence can be shifted 5' or 3' (N-terminus or C-terminus for PNA) in increments of 1 nucleotide until the binding site is more than 100 nucleotides, more than 200 nucleotides, or more than 400 nucleotides away from the editing site. For triplex forming regions on helper oligonucleotides, they generally target all purine sites or sites that are 75% or more purine, or all pyrimidine sites or sites that are 75% or more pyrimidine. For triplex forming regions on helper oligonucleotides, the site can be shifted or extended only as long as the site maintains the desired base composition. The triplex region may match the antisense region of a clamp (in the case of PNAs, sometimes referred to as bis-PNAs), or the antisense sequence may be longer than the triplex region by extension of complementarity with the target in the 5' or 3' direction (for examples of clamps, see McNeer, 2013 supra; McNeer 2015 supra; Bahal et al. Current Gene Therapy 14(5):331-42, 2014; Nielsen et al. Current Issues in Molecular Biology 1(1-2):89-104, 1999; and Gaddis et al. Oligonucleotides, 2006 Summer;16(2):196-201. See http: / / spi.mdanderson.org / tfo / , which provides a search engine for identifying triplex binding sites, and Table XIV).

[0222] Example 4 Selecting optimized editing oligonucleotides targeting CFTR A. Optimizing editing oligonucleotides in cell culture 1. Initial Cell Culture Screening for Optimal Editing Oligonucleotide Sequences Twenty editing oligonucleotide sequences are selected for initial cell culture efficacy screening, along with a control editing oligonucleotide of the mutant sequence. The editing oligonucleotides are designed to correct the delta F508 mutation that causes cystic fibrosis. An existing active editing oligonucleotide targeting CFTR is used as a positive control and reference sequence (see McNeer et al. 2015 supra). These screens are performed in human cell lines. Screens start with an editing oligonucleotide with a length of about 40 nucleotides centered on the delta F508 mutation site, which generally provides a good balance between efficacy and ease of manufacture. In most examples in the literature, the editing site is approximately in the center of the editing oligonucleotide. This results in a level of editing efficiency, but it is not necessarily optimal. Therefore, the length, target strand (antisense or sense), and 5'-3' position of the editing nucleotide are varied. The editing oligonucleotides are tested for editing activity in cells by lipofection or electroporation to determine the optimal editing oligonucleotide sequence, as assayed by PCR / next-generation sequencing (see methods below). Starting with an editing oligonucleotide sequence in the desired editing efficiency range from the first round of optimization, the editing oligonucleotide sequence is further modified in length by adding or removing bases complementary to the target at the 5' end, 3' end, or both ends. Additional editing efficacy testing in cells will result in a lead editing oligonucleotide sequence and a reserve lead editing oligonucleotide sequence. Similarly, helper oligonucleotide PNA clamps are varied as described herein and tested with the best editing oligonucleotide to identify the most active helper oligonucleotide. A baseline editing efficiency of 2% or higher can be achieved at the DNA level for each in vitro treatment, as assayed by sequencing.

[0223] 2. Optimization of chemical structure and composition, and fine-tuning of sequence Despite the successful application of chemically modified end-blocked donors for editing in cells and animals, the predicted half-life of end-blocked editing oligonucleotides is only 10-30 minutes in cells. They therefore require high doses, resulting in moderate editing frequencies. High DNA endonuclease activity in mammalian cells has been demonstrated in antisense gapmers, where even one or a few unmodified DNA linkages resulted in lower intracellular half-lives and reduced efficacy (Monia et al., 1995 supra). In the case of antisense, endonuclease stability has been addressed by using internal modifications throughout the oligonucleotide, in addition to optional end-blocks. Although a number of endonuclease-resistant modified nucleic acid chemical structures exist, the majority are not recognized by cellular recombination / repair machinery and therefore do not support editing when placed internally near the editing site. The present invention modifies editing oligonucleotides at several positions along their length to improve the therapeutic properties of the nucleic acid, including reducing inflammation; increasing nuclease stability; increasing target binding - improving efficacy; increasing free uptake (self-delivery without encapsulation) by conjugating delivery ligands; and stabilizing against nucleases during transit through the endolysosomal pathway.

[0224] It has been discovered that certain exonuclease stabilizing modifications can be used within the editing oligonucleotides (herein referred to as third generation editing oligonucleotides). A toolbox of third generation chemical modifications is utilized to optimize editing oligonucleotides for use in cells and in vivo. Examples of advanced chemical modification patterns and configurations that are deployed are as follows: 1. Generations (Gens) 3A, 3B, and 3C of Figure 2 of PCT / US2015 / 65348. 2. Combining helper oligonucleotides PNAs of various compositions and chemistries with the editing oligonucleotides, which have been found to improve editing efficiency (initial studies have used PNA-clamps, specifically hCFPNA-2, McNeer et al., 2015 supra, which have already been shown to have baseline activity). 3. Implementing edited oligonucleotide structural chemistries that interact with the endogenous RNAi cellular machinery that have been shown to enhance hybridization rates to DNA; and 4. Utilizing self-delivering conjugates of editing oligonucleotides in combination with more heavily modified editing oligonucleotides that survive passage through the endolysosomal pathway, allowing delivery without encapsulation. 5. To inhibit mismatch repair of editing oligonucleotides, chemical modifications such as 2'F, LNA, or 2'-O-methyl are used at or near the editing site (see CFTR delta F508 targeting editing oligonucleotides in Figure 2). 6. Inclusion of a 5' phosphate or 5' phosphate analog as described herein and illustrated in FIG.

[0225] This will achieve 2-8% or more editing at the genomic DNA level per treatment, as assayed by PCR and confirmed by targeted protein assays. Editing oligonucleotides are used with multiple doses in animals and eventual clinical trials to obtain the desired cumulative level of editing (see McNeer, 2015 supra for formulation, transfection, and DNA and functional assays). Editing and helper oligonucleotides are formulated for nebulization and patients are treated by inhalation. Systematic formulation is used to target secondary tissues affected by CF.

[0226] 3. Edited Oligonucleotide Synthesis Better editing efficacy was observed for higher quality compounds. Therefore, high quality control gel-isolated edited oligonucleotides will be synthesized at 1 umole scale, and their identity will be confirmed using mass spectrometry, and their purity will be confirmed using analytical HPLC. High quality edited oligonucleotides will be >80% pure (>1 mg amount) in a scalable and reproducible manner, so that future batches of lead edited oligonucleotides can be prepared with similar activity and at a larger scale for animal testing and eventual human clinical trials.

[0227] 4. Assay for Editing Efficiency i. Target nucleic acid Assays for nucleic acid target sequences will be performed using genomic PCR and RT-PCR gels to assay splice skipping. For sequencing, primers for genomic DNA PCR will be designed and synthesized to assay editing efficiency. To distinguish between true editing and potential contamination with environmental human DNA, established protocols will be used to control for potential PCR artifacts, including zero-time renaturation controls and editing that adds silent mutations in addition to the desired correction. Next-generation sequencing will be performed as previously established to assay genome editing.

[0228] ii. Assay for target protein Assays for protein targets are performed using Western blot analysis, immunohistochemistry, and functional NPD (see McNeer, 2015 supra).

[0229] 5. Assay for Off-target Editing in Cultured Cells Measuring specificity is a critical step for therapeutic genome editing. When using donor DNA, the most dramatic consequence of off-target editing is potential insertion of donor DNA into non-target sites. Full or partial insertion of donor DNA is easily assayed by FISH, PCR, or genome-wide sequencing. Test the 10 most homologous genomic sites for off-target editing to obtain less than 0.1% off-target editing at each site per treatment. Survey genomic DNA for off-target integration of editing oligonucleotide sequences to achieve less than 1 off-target integration event per 1000 cells. Confirmatory testing is then performed in primary target cells. These studies provide the basis for animal studies.

[0230] 6. Optimizing editing in vivo tissue penetration Animal efficacy in mice and formulation studies were performed with the lead edited oligonucleotides developed in cell culture screening. Preliminary PK ADME and toxicity studies were then performed.

[0231] These studies will optimize formulations or self-delivery chemistry to maximize in vivo editing efficiency, identify lead editing oligonucleotides that edit 2-5% of the target DNA per in vivo treatment, evaluate off-target effects by targeted genome sequencing, and perform preliminary toxicity assessments. One consideration is that editing oligonucleotide sequences are generally species-specific, and thus if one editing oligonucleotide sequence is to be used for human cell culture and mouse animal testing, modified knock-in transgenic animals will be used.

[0232] Two delivery modes will be tested: self-delivery (also known as free or naked oligomers) will be achieved using delivery conjugates in combination with nuclease-stabilizing internal chemical modifications to obtain free uptake (self-delivery without encapsulation). Uptake through the endolysosomal pathway exposes the editing oligonucleotide to potent nucleases, and thus nuclease-resistant modifications will be utilized along the editing oligonucleotide, as long as the modification pattern is consistent with a retained interaction with the endogenous repair system. Chemical structures that have been shown to achieve self-delivery with oligomers of similar size and charge to the editing oligonucleotide (i.e., Byrne et al. 2013 supra and Alterman et al. 2016 supra) will be used. Potential lung-specific conjugates for nanoparticle or direct editing oligonucleotide conjugation include receptor ligands, binding targets to increase retention time, and / or diffusion enhancers to better penetrate glycolax.

[0233] In vivo delivery by nanoparticles is challenging because the nanoparticles must not be destabilized by serum and the nanoparticles must diffuse into the target tissue. Fortunately, the encapsulated nucleic acid is completely protected from nucleases on the way to the target cell. Initial results from the Saltzman, Glazer, and Egan groups at Yale University have identified a suitable nanoparticle formulation for delivering editing oligonucleotides to ciliated epithelial cells of the nose and lungs by intranasal instillation with moderate editing efficiency in mice (McNeer et al., 2015 supra). Specific delivery vehicles useful for delivering genome editing oligonucleotides in vitro and in vivo include PLGA and 15% (by weight) poly(β-amino ester) (PBAE). These polymers are particularly useful when using neutral helper oligonucleotides such as PNA clamps, because neutral or positively charged helper oligonucleotides can be easily and efficiently loaded into these particles along with negatively charged editing oligonucleotides. PLGA / PBAE nanoparticles derivatized with MPG via a PEGylated phospholipid linker (DSPE-PEG2000) have also been shown to improve uptake in vivo, particularly in the lungs. (McNeer, et al., Gene Ther. 20(6):658-669, 2013. doi:10.1038 / gt.2012.82; McNeer, et al., Nature Comm. DOI:10.1038 / ncomms 7952:1-11, 2015)

[0234] Initially, the animal model and administration protocol previously used by McNeer et al., 2015 supra may be utilized, with editing oligonucleotides and delivery polymers as positive controls. Variations in the composition of delivery particles and alternative delivery particles (including alternative ligand conjugates listed in the self-delivery section above) will be evaluated with the goal of increasing delivery of editing oligonucleotides to target epithelial cells and epithelial cell progenitor cells, thereby increasing editing efficiency. See also Bahal, R. et al. Nat. Commun. 7, 13304 (2016) for examples of in vivo delivery of editing oligonucleotides and successful editing in vivo.

[0235] Example 5 Oligonucleotide-free editing of engineered transcription factors by targeted nucleobase chemical modification Another specific illustration of our nucleobase modification genome editing method is from Yang et al. 2016 (Yang, L. et al. bioRxiv, doi:10.1101 / 066597, 2016. Also published as Yang, L. et al. Nat. Commun. 7, 13330 (2016) doi:10.1038 / ncomms13330), in which a deaminase nucleobase modifying activity (also known as a sequence modification or editing moiety) is fused to a modified transcription factor (e.g., zinc finger or TALEN) and used to obtain highly accurate point editing in mammalian cells in culture. This method can be used for various classes of editing, sequence, target indications and corresponding target genes as well as nucleobase chemical modifications that result in altered target editing within genes as described herein. In the case of enzymatic alterations, the relevant nucleobase modifying enzyme can be tethered or fused to the modified transcription factor. In the case of reactive chemicals, reactive chemicals can be conjugated to the modified transcription factor. In addition to deamination, other chemical modifications described herein can be used (see Table VII).

Claims

1. 1. A composition comprising a single stranded editing oligonucleotide for modifying a genomic sequence in a cell of a subject, the editing oligonucleotide comprising, in 5' to 3' order: 5' end segment; 5' proximal segment; 5' editing segment; Editing site; 3' editing segment; a 3' proximal segment; and 3' end segment, wherein the edited oligonucleotide comprises the chemical modification shown below: a) the editing oligonucleotide further comprises a delivery conjugate; b) said 5' terminal segment comprises a backbone modification and optionally a nucleobase modification; c) the 5' proximal segment and / or the 5' edited segment optionally comprise a backbone modification, a nucleobase modification and / or a sugar modification; d) the editing site comprises a mismatch for modifying a genomic sequence, wherein the editing site comprises a sugar modification selected from a 2' sugar modification and a locked nucleic acid (LNA); and e) the 3' proximal segment and / or the 3' edited segment optionally comprise a backbone modification, a nucleobase modification and / or a sugar modification; and f) said 3' terminal segment comprises a backbone modification and optionally said 3' terminal segment comprises a nucleobase modification.

2. The composition of claim 1 , wherein not all nucleotides in the edited oligonucleotide are chemically modified.

3. The composition of claim 1 or 2, wherein the backbone modification is a modification other than a phosphorothioate modification.

4. The composition of any one of claims 1 to 3, wherein the sugar modification in the 5' proximal segment and / or the 5' edited segment is a 2' sugar modification or a locked nucleic acid (LNA).

5. The composition of any one of claims 1 to 4, wherein the sugar modification in the 3' proximal segment and / or the 3' edited segment is a 2' sugar modification or a locked nucleic acid (LNA).

6. The composition of any one of claims 1 to 5, wherein the 5' edited segment further comprises an LNA.

7. The composition of any of claims 1 to 6, wherein the 5' proximal segment and / or the 5' edited segment comprises one or more of 5-methylcytosine, LNA, 2'-F sugars and 2'-OMe sugars.

8. The composition of any one of claims 1 to 7, wherein the edited segment comprises a 2'-F sugar or a 2'-OMe sugar.

9. The composition of any one of claims 1 to 8, wherein the editing segment comprises an LNA.

10. The composition of any of claims 1 to 9, wherein the 3' proximal segment and / or the 3' edited segment comprises one or more of 5-methylcytosine and 2'-F sugars.

11. The composition of any one of claims 1 to 10, wherein the 3' terminal segment further comprises an LNA.

12. The composition of any one of claims 1 to 11, wherein the 3'-terminal segment comprises 5-methylcytosine.

13. The composition of any of claims 1 to 12, wherein the delivery conjugate is in the form of a cholesterol, a lipophilic conjugate, or a negatively charged moiety.

14. The composition of claim 13 , wherein the negatively charged moiety is sulfate, fucose, or sialic acid.

15. The composition of any of claims 1 to 12, wherein the delivery conjugate is Gal-Nac, glp-1, polyethylene glycol (PEG), hemagglutinin, neuraminidase, or endothelin.

Images