Chemically modified antisense oligonucleotides (ASO) and compositions for RNA editing comprising same

Abstract

The present invention relates to a chemically modified oligonucleotide comprising a sequence of 23 to 80 nucleotides in length, a target sequence capable of binding to a target RNA, a central base triad (CBT) comprising three nucleotides (I) wherein the central nucleotide (N0) is directly opposed to a target adenosine in the target RNA, and wherein the core oligonucleotide comprises the following sequence (II): wherein the core oligonucleotide comprises the following sequence (II); the sequences comprise different 2'sugar moiety modifications and bond modifications. The disclosure also provides oligonucleotides and compositions thereof for treating or preventing a genetic disorder, condition, or disease. Also provided are methods for editing a target adenosine or deaminating at least one specific adenosine in a target nucleic acid. 5 '-

Description

Technical Field

[0001] The present invention relates to the field of medicine, in particular to the field of site-directed RNA editing, in which RNA sequences are targeted by single-stranded antisense oligonucleotides (ASOs) to edit RNA for specific gene mutations ("compensatory editing") or to edit RNA from wild-type alleles ("beneficial editing"). Background Art

[0002] RNA editing is a natural process by which some cells can make discrete changes to specific nucleotide sequences within RNA molecules in a site-specific manner. Unlike DNA editing, the advantage of RNA editing is that it can modify genetic information in a more efficient manner. This is because RNA is typically degraded rapidly, and any errors introduced by off-target modifications will be eliminated rather than permanently retained in the modified DNA of the subject. RNA editing is also less likely to cause an immune response because it is an editing mechanism that exists naturally in humans. In addition, RNA editing may provide a more natural immune response than introducing external engineered genes.

[0003] Over the years, oligonucleotide therapies have been developed to specifically silence, restore, or modify the expression of pathogenic genes or disease-associated genes in, for example, cancer and (other) genetic disorders. Such therapies include, for example, antisense oligonucleotides (ASOs), small interfering RNAs (siRNAs), and microRNAs that interfere with coding and non-coding RNAs. ASO sequence customization is relatively simple and accurate, and almost any mutant gene can be targeted. Therefore, ASOs are the most developed in the clinic, and a variety of drugs have been approved by the U.S. Food and Drug Administration (FDA) and entered clinical trials (Cideciyan et al., 2019; Gagliardi and Ashizawa, 2021).

[0004] In general, a large number of proteins are involved in mediating RNA editing processes in cells (Quinones-Valdez et al., 2019). Specifically, site-directed RNA editing (SDRE) refers to the modification of RNA sequences by introducing or removing nucleotides from RNA or changing the properties of nucleobases through deamination. RNA editing enzymes are known in the art. The first RNA editing process discovered in mammals was the deamination of cytidine (C) to uridine (U) by APOBEC proteins (Zinshteyn and Nishikura, 2009). To date, the two most useful and most studied types of RNA editing are the conversion of cytidine (C) to uridine (U) ("C to U") and adenosine (A) to inosine (I) ("A to I"). Notably, for therapeutic purposes, the most prevalent type of RNA editing in higher eukaryotes is the "A to I" conversion. This conversion is catalyzed by the family of adenosine deaminases (ADARs) acting on RNA. Over the years, three vertebrate ADAR genes have been identified that produce several ADAR proteins produced from alternative promoters or representing splice variants (Wulff and Nishikura, 2010). These proteins are expressed in various types of human tissues and can alter splicing and translation mechanisms, double-stranded RNA (dsRNA) structure, and the binding affinity between RNA and RNA-binding proteins (Tomaselli et al., 2014; Zinshteyn and Nishikura, 2009). Of the three known ADAR genes, hADAR1 and hADAR2 are expressed in most tissues and encode active deaminases. Human ADAR3 (hADAR3) has been described as being expressed only in the central nervous system and reportedly lacks deaminase activity in vitro. While all ADARs are multidomain proteins comprising a targeting or dsRNA binding domain (dsRBD) and a catalytic domain, the ADAR1 protein also comprises one or more Z-binding domains, while the splice variants ADAR2R and ADAR3 comprise R domains (Zinshteyn and Nishikura, 2009; Wulff and Nishikura, 2010). Thus, in some embodiments, the ADAR is hADAR1, hADAR2, or hADAR3.

[0005] "A to I" editing was first discovered in the eggs of African clawed frogs (Bass and Weintraub, 1987; Rebagliati and Melton, 1987). Kim et al. (1994) first cloned the human cDNA encoding "double-stranded RNA adenosine deaminase" and confirmed the protein's adenosine to inosine ("A to I") conversion activity by recombinant expression in insect cells. "A to I" editing changes the information content of RNA molecules because inosine preferentially pairs with cytidine bases and is therefore interpreted as guanosine (G) by the translation and splicing mechanisms. In this enzyme-catalyzed reaction, adenosine is converted to inosine through a hydration intermediate. Guanosine can form three hydrogen bonds with the complementary base cytidine, while inosine can only form two hydrogen bonds with cytidine. The translation machinery interprets inosine as guanosine. Therefore, ADAR has the effect of introducing functional adenosine to guanosine mutations at the RNA level. The ability of ADAR to change RNA sequence has also been used to artificially target RNA in vitro for RNA editing. This approach could potentially be used to repair genetic defects and alter genetic information at the RNA level.

[0006] ASOs are typically short, ranging from 18 to 25 nucleobases in length, and are single-stranded synthetic RNA or DNA molecules that utilize Watson-Crick base pairing to bind sequence-specifically to target RNA. They can be broadly categorized as first-, second-, and third-generation ASOs. First-generation ASOs were used to inhibit translation of Rous sarcoma virus ribosomal RNA (Stephenson and Zamecnik, 1978). First-generation ASOs are characterized by modified backbones in which nucleotide bonds are modified with sulfur, methyl, or amine groups to generate phosphorothioates (PS), methylphosphonates, and phosphoramidates, respectively, while second-generation ASOs also carry alkyl modifications at the 2' position of the ribose sugar. These second-generation ASOs tend to be less toxic than PS-modified ASOs and have slightly higher affinity for their targets. In contrast, third-generation ASOs tend to be even more heterogeneous, in that they contain a large number of chemical modifications designed to improve binding affinity, stability, and pharmacokinetics (Quemener et al., 2019). This diversity of chemical modifications, along with the sequence of the ASO, provides considerable flexibility in therapeutic approaches. That is, depending on their mechanism of action, ASOs can be used to degrade target mRNA to reduce protein levels, modify or correct splicing events, regulate RNA translation, or target pathological coding or noncoding RNAs (Quemener et al., 2019).

[0007] ASOs can act through a variety of mechanisms, depending in part on the region of the RNA sequence being targeted and the design / chemistry of the ASO. To ensure the specificity of ASOs, their sequences are ideally complementary, or at least partially complementary, to the target RNA. However, in the case of site-directed mutagenesis (i.e., “A to I” RNA editing), the ASO targeting domain contains a mismatch that is opposite the target adenosine. Notably, several endogenous substrates of ADARs contain mismatches and / or bulges (Thomas and Beal, 2017), and thus, if these features are mimicked in the ASO or the resulting dsRNA, substrate recognition may be altered or even improved.

[0008] In addition, ASOs can be chemically modified to improve their properties. For example, ASOs can be modified to protect them from nucleases and improve their effectiveness. Although phosphorothioate (PS) modification appears to have a positive effect on the stability and pharmacokinetics of ASOs, the chirality of the PS bond may have a significant impact on the overall properties of the ASO. The PS bond can have two stereoisomers, Rp and Sp, and it is known in the art that the Rp bond and the Sp bond can affect the properties of ASOs, such as thermal stability, binding affinity, pharmacological properties, etc. However, the benefits of the Rp and Sp stereoisomers have been controversial (Iwamoto et al., 2017; Crooke et al., 2020).

[0009] While the use of antisense oligonucleotides in RNA editing is known in the art (Vogel et al., 2014; Merkle et al., 2019), over the past few years, ASO-based therapies have gained increasing attention for the treatment of different medical conditions and diseases, especially genetic disorders. RNA editing systems that utilize endogenous adenosine deaminases have been extensively studied, i.e., exogenous oligonucleotides are used to specifically recruit endogenous adenosine deaminases to specific target sites of target RNA, thereby providing an improved system for editing target RNA. Patent applications WO 2016 / 097212 and WO 2017 / 010556 describe oligonucleotide constructs for site-directed RNA editing that utilize endogenous cellular pathways (i.e., endogenous ADARs) to edit endogenous RNA. New designs of nucleoside analogs are being continuously investigated using the structural information available for ADAR-RNA complexes (Doherty et al., 2021). Endogenous ADAR-mediated RNA editing in non-human primates has been previously reported using stereopure, chemically modified oligonucleotides (Monian et al., 2022). These oligonucleotides typically contain a rich 2'-F-modification at the 5' terminus (these modifications are often present as a continuous stretch of 2'-F-modifications) and a uniform 2'-O-methyl modified stretch at the 3' terminus, flanking the CBT. In addition, these oligonucleotides contain a substantial stereopure PS-modified backbone and a substantial number of stereopure PS bonds, as well as an additional electroneutral PN bond (also stereopure), the latter of which has not yet been clinically applied. Precise, site-specific RNA editing can be achieved by recruiting endogenous ADARs using antisense oligonucleotides, as previously demonstrated by Merkle et al. (2019). Merkle et al. (2019) were able to demonstrate that chemically optimized ASOs can be used to recruit endogenous human ADARs to edit endogenous transcripts in a simple and programmable manner with minimal off-target editing.

[0010] In WO 2020 / 001793, the inventors of the present application provide an artificial nucleic acid for RNA site-directed editing ("A to I" editing), wherein the nucleic acid comprises a targeting sequence and a recruitment portion. Similarly, WO 2018 / 041973 relates to an ASO capable of specifically editing adenosine in a target RNA sequence, wherein the ASO does not form an intramolecular hairpin structure or a stem-loop structure. Specifically, WO2018 / 041973 relates to chemically modified oligonucleotides for single-stranded RNA editing, which are used to deaminize target adenosine by ADAR enzymes, wherein the central base triplet (CBT) of three consecutive nucleotides comprises sugar modification and / or base modification. It has been found that deoxyribose at all three positions of CBT is well tolerated and significantly improves stability against nuclease digestion.

[0011] Other prior art, including WO 2021 / 071858, relates to oligonucleotides comprising a first domain and a second domain, wherein the first domain comprises one or more 2'-F modifications, and wherein the second domain comprises one or more sugars that do not have 2'-F modifications. WO 2022 / 099159 relates to oligonucleotides comprising a first domain and a second domain, wherein the domains comprise a specific percentage of 2'-F modifications and aliphatic substitutions.

[0012] Research in the field of ASO optimization has led to the identification and further study of the CBT and the 5' and 3' regions immediately surrounding it. In addition to specifically studying CBT modifications (e.g., 2'-F and 2'-FANA), WO 2021 / 243023 also mentions the modification of the guide or targeting domain at the 3' end of the adjacent nucleobase immediately outside the CBT (at the +2 position of an oligonucleotide comprising the structure [Am]-X1-X2-X3-X4-[Bn], where X4 corresponds to the +2 position). It has been found that the choice of nucleotide at the +2 position of the triplet of the guide oligonucleotide may affect the editing rate of the target. Improved editing efficiency was observed with the 2'-F modification at the +2 position.

[0013] However, despite their promising nature, few ASOs have reached the market. This is due to difficulties associated with stability, cellular delivery and uptake, clinical efficacy, and off-target effects and / or preclinical toxicology challenges. Therefore, overcoming these various challenges is crucial to ultimately translate ASO-based therapies into widespread clinical success. Consequently, there is an unmet need for improved ASOs and effective therapies for genetic disorders associated with these improved ASOs.

[0014] The inventors of the present application have found that the artificial and chemically modified oligonucleotides of the present invention are suitable for editing a variety of endogenous RNA transcripts, such as endogenous mRNA of housekeeping genes and endogenous transcripts of disease-related genes (e.g., STAT1, SERPINA1, LRRK2, CRB1, NLRP3, CTNNB1, PEX1 and PDE6A). Surprisingly, the inventors have found that the present invention provides improved ASOs. For example, the ASOs of the present application have higher editing efficiency and effectiveness. In addition, the ASOs of the present application provide lysosomal half-life extension (i.e., lysosomal stability is improved) and are easy to produce (e.g., easy to control cost, purity and quality). The ASOs of the present invention also have the advantage of reduced RNA off-target editing. Summary of the Invention

[0015] Therefore, the problem solved by the present application is to provide improved synthetic and chemically modified antisense oligonucleotides (ASOs) that can mediate the functional conversion of adenosine (A) to guanosine (G) to correct point mutations that would otherwise have deleterious effects. The present invention addresses this problem by providing synthetic ASOs comprising specific nucleoside modifications (particularly at the 2' position of the sugar moiety), backbone bond modifications, and combinations thereof. Overall, the ASOs of the present application exhibit several differences and associated advantages compared to the ASOs disclosed in the prior art. Most importantly, the creativity of the present invention lies in the specific combinations of nucleoside and bond modifications provided herein. To date, no prior art has been found to teach or suggest the oligonucleotides, compositions, and methods disclosed in the present application, which are based on the inventors' discovery that some combinations of nucleobases, nucleosides, and bond modifications are particularly effective in providing stable and effective ASOs and compositions comprising the same.

[0016] The solution to the problem addressed by the present application is achieved by the embodiments described herein and defined by the appended claims.

[0017] The present invention generally provides oligonucleotides and compositions comprising the oligonucleotides, and their use in treating or preventing genetic disorders, conditions, or diseases. Also provided herein is an in vitro method for editing a target adenosine in a target nucleic acid, and an in vitro method for deaminating at least one specific adenosine present in a target RNA sequence in a cell. Also provided herein is a method for treating or preventing a genetic disorder, condition, or disease, wherein the method comprises administering an effective amount of an oligonucleotide of the present invention.

[0018] In a first aspect, the present invention provides a chemically modified oligonucleotide comprising a sequence of 23 to 80 nucleotides in length, capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, wherein the core oligonucleotide comprises the following sequence: 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0019] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0020] (b) the N+2 nucleotides carry a 2'-O-alkyl modification; and wherein the N+3 nucleotides carry a 2'-fluoro (2'-F)-modification;

[0021] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0022] (d) an internucleoside linkage modification content of at least 15%; and

[0023] (e) Bond h and bond i are not phosphorothioate (PS) bonds.

[0024] In a second aspect, the present invention provides a chemically modified oligonucleotide comprising a sequence of 23 to 50 nucleotides in length, capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- N - 1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, and wherein the core oligonucleotide comprises the following sequence:

[0025] 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0026] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0027] (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification;

[0028] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0029] (d) the total number of deoxyribonucleosides in the regions adjacent to the 3' and 5' ends of the CBT does not exceed 6;

[0030] (e) The content of internucleoside linkage modifications is at least 30%.

[0031] In a third aspect, the present invention provides a chemically modified oligonucleotide comprising a sequence of 40 to 80 nucleotides in length, capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- N - 1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, wherein the core oligonucleotide comprises the following sequence: 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0032] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0033] (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification;

[0034] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0035] (d) The total content of deoxyribonucleosides in the region adjacent to the 3' end and the 5' end of CBT is 5% to 50%.

[0036] In a fourth aspect, provided herein is a pharmaceutical composition comprising an oligonucleotide of the present invention or a pharmaceutically acceptable salt thereof.

[0037] In a fifth aspect, provided herein is a chemically modified oligonucleotide of the invention or a pharmaceutical composition of the invention for use in treating or preventing a genetic disorder, condition or disease.

[0038] In a sixth aspect, provided herein is an in vitro method for editing a target adenosine in a target nucleic acid, wherein the method comprises contacting the target nucleic acid with an oligonucleotide of the present invention.

[0039] In a seventh aspect, provided herein is an in vitro method for deaminating at least one specific adenosine present in a target RNA sequence in a cell, wherein the method comprises the following steps:

[0040] (a) contacting a target nucleic acid with a chemically modified oligonucleotide of the present invention;

[0041] (b) allowing cells to take up the chemically modified oligonucleotide;

[0042] (c) annealing the chemically modified oligonucleotide to the target RNA sequence; and

[0043] (d) A mammalian ADAR enzyme containing the native dsRNA binding domain found in the wild-type enzyme is caused to deaminize a target adenosine to inosine in a target RNA sequence. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings shown below are only used to illustrate the present invention and further describe the present invention. These drawings should not be understood as limiting the present invention thereto.

[0045] Figure 1 is a graph showing the editing efficiency and lysosomal stability of oligonucleotides targeting SERPINA.

[0046] Figure 2 is a graph showing the SERPINA editing efficiency of 2′-F modified oligonucleotides.

[0047] Figure 3 is a graph showing the editing efficiency and lysosomal stability of oligonucleotides targeting STAT1 Y701.

[0048] Figure 4 is a graph showing the editing efficiency and lysosomal stability of oligonucleotides targeting CRB1 C948Y.

[0049] Figure 5 is a graph showing the editing efficiency and lysosomal stability of oligonucleotides targeting LRRK2 G2019S.

[0050] Figure 6 is a graph showing the editing efficiency of oligonucleotides targeting murine PDE6A (mPDE6A) V685M.

[0051] Figure 7 Graphs showing the editing efficiency of oligonucleotides targeting NLRP3 Y166 expressed in a plasmid (A) and genomically integrated (B).

[0052] Figure 8 is a graph showing editing efficiency data for oligonucleotides targeting GAPDH 3'UTR.

[0053] Figure 9 is a graph showing editing efficiency data for truncated variants (31 nt, 40 nt, 45 nt, 50 nt, 59 nt) of oligonucleotides targeting SERPINA.

[0054] Figure 10 Data showing the editing efficiency of 5' and / or 3' truncated variants of oligonucleotides targeting SERPINA.

[0055] Figure 11 is a graph showing the editing efficiency of SERPINA-targeting oligonucleotides of 32 nt and 33 nt in length.

[0056] Figure 12 is a graph showing the editing efficiency of 3′-end truncated oligonucleotides targeting STAT1 Y701.

[0057] Figure 13 is a graph showing the editing efficiency of 3′-end truncated oligonucleotides targeting CTNNB1 T41.

[0058] Figure 14 is a graph showing the editing efficiency of oligonucleotides targeting CRB1 C948Y with 5' and / or 3' end truncations.

[0059] Figure 15 is a graph showing the editing efficiency of oligonucleotides targeting STAT1 Y701.

[0060] Figure 16 is a graph showing the editing efficiency of SERPINA-targeting oligonucleotides modified at the extended hotspot region adjacent to the 3′ end of CBT (optimal versions: +2 (2′-OMe) and +3 (2′-F)).

[0061] Figure 17 is a graph showing the editing efficiency of oligonucleotides targeting CTNNB1 T41, wherein modifications are made in an extended hotspot region adjacent to the 3' end of CBT (optimal versions: +2 (2'-OMe) and +3 (2'-F)).

[0062] Figure 18is a graph showing the editing efficiency, lysosomal stability, and relative toxicity of SERPINA-targeting oligonucleotides containing a 2'-MOE terminal segment.

[0063] Figure 19 Graphs showing the editing efficiency and lysosomal stability of a long oligonucleotide (59 nt) targeting SERPINA with reduced PS linkage using a genomic system (A) and a plasmid system (B).

[0064] Figure 20 is a graph showing editing efficiency data for short oligonucleotides (40 nt) targeting SERPINA with reduced PS bond modification.

[0065] Figure 21 is a graph showing the editing efficiency of SERPINA-targeting oligonucleotides containing contiguous PS bond segments.

[0066] Figure 22 is a graph showing the effect of LNA modification on the editing efficiency and potency of oligonucleotides targeting SERPINA.

[0067] Figure 23 is a graph showing the effect of LNA modification at the 5′ end on oligonucleotides targeting SERPINA.

[0068] Figure 24 is a graph showing the effect of 5′-terminal LNA modification on short oligonucleotides targeting SERPINA.

[0069] Figure 25 is a diagram showing the disruption of the 5' and 3' terminal segments of the SERPINA-targeting oligonucleotide.

[0070] Figure 26 is a diagram showing segment disruption by short oligonucleotides targeting STAT1.

[0071] Figure 27 is a diagram showing 2′-FANA modification of CBT of an oligonucleotide targeting CRB1 C948Y. DETAILED DESCRIPTION

[0072] the term

[0073] First, some terms are defined to make the present invention easier to understand.

[0074] As used herein, the articles "a" and "an" refer to one or more (ie, at least one) of the grammatical object of the article. For example, "an element" refers to one or more than one element, such as a plurality of elements.

[0075] The terms "about" and "approximately" should be understood to allow for standard variations as understood by one of ordinary skill in the art.

[0076] As used herein, the term "including" means and is used interchangeably with the phrase "including but not limited to." Likewise, as used herein, the term "comprising" means and is used interchangeably with the phrase "including but not limited to."

[0077] As used herein, the term "nucleic acid" is intended to include any DNA molecule (e.g., cDNA or genomic DNA) and any RNA molecule (e.g., mRNA), as well as analogs of DNA or RNA generated using nucleotide analogs. For example, in one embodiment, the oligonucleotide comprises, for example, UNA (unlocked nucleic acid), PMO (phosphodiamidate-linked morpholino), or PNA (peptide nucleic acid). Nucleic acids can be single-stranded or double-stranded. Oligonucleotides can be single-stranded (ss) or double-stranded (ds). A single-stranded oligonucleotide can have a double-stranded region (formed by two single-stranded oligonucleotides); while a double-stranded oligonucleotide comprising two oligonucleotide chains can have single-stranded regions, for example, in regions where the two oligonucleotide chains are not complementary to each other. Each component of the DNA or RNA structure can be modified and can be classified according to the following modifications: (1) internucleoside linkages, (2) deoxyribose / ribose, and / or (3) nucleobases.

[0078] The term "oligonucleotide" as used herein is defined as a molecule (e.g., a short nucleic acid polymer) comprising two or more covalently linked nucleosides. They may include DNA and / or RNA. Oligonucleotides provided herein have a backbone comprising deoxyribonucleotides and / or ribonucleotides.

[0079] The term "nucleobase" refers to the nitrogen-containing biological building blocks that make up nucleosides, which are the building blocks of nucleotides. The naturally occurring bases [guanine (G), adenine (A), cytosine (C), thymine (T), and uracil (U)] are derivatives of purines or pyrimidines, but it is understood that naturally occurring and non-naturally occurring base analogs are also included, and the term "nucleobase" also includes "modified nucleobases."

[0080] In the context of the present invention, the terms "modified nucleobase" and "modified base" can be used interchangeably with the term "nucleobase". Nucleobases can be modified or unmodified. Therefore, in some embodiments, a modified nucleobase is a nucleobase comprising a modification. In some embodiments, the modified nucleobase is capable of performing at least one function of a nucleobase, for example, forming a portion of a polymer that is capable of base pairing with a nucleic acid comprising at least a complementary base sequence. In one embodiment, the modified nucleobase is capable of increasing hydrogen bonding, base pair stacking interactions, and / or stabilizing nucleic acid complexes. In another embodiment, a modified nucleobase (e.g., a Benner base) is capable of mimicking an N3 protonated cytosine base. In some embodiments, the modified nucleobase is a substituted A, T, C, G, or U, or a tautomer of a substituted A, T, C, G, or U. In some embodiments, in the context of an oligonucleotide, a modified nucleobase is a nucleobase that is not A, T, C, G, or U. Modifications include, but are not limited to, non-standard nucleobases 5-methyl-2'-deoxycytidine (m 5 C), pseudouridine (pU), dihydrouridine, inosine (I) and 7-methylguanosine. Other modifications may include replacing the nucleobase with a (N) heterocycle (e.g., hyaluronic acid) or an aromatic ring that stacks well in the RNA duplex, for example, bennabase Z (and / or analogs) or 8-oxoadenosine (8-oxo-A). As used herein, the term "bennabase Z" refers to the pyrimidine analog 6-amino-5-nitro-3-(1'-β-D-2'-deoxyribofuranosyl)-2(1H)-pyridone (dZ). In one embodiment, the modification includes the introduction of a nucleobase analog or a simple heterocycle that enhances the editing effect. As used herein, and as generally understood by those skilled in the art, the expression "derivatives thereof" refers to derivatives of (modified) nucleobases, nucleosides or nucleotides. For example, a derivative may be a corresponding nucleobase, nucleoside or nucleotide chemically derived from the nucleobase, nucleoside or nucleotide. For example, the derivatives of deoxycytidine may include fluorine-modified deoxycytidine, 5-methyl-2'-deoxycytidine (m 5 C) or ribocytidine.

[0081] The term "nucleoside" refers to a moiety in which a nucleobase or modified nucleobase is covalently bound to a sugar or modified sugar. In some embodiments, "nucleoside" refers to a nucleoside unit in an oligonucleotide or nucleic acid. The term "nucleoside" encompasses all modified forms and derivatives of "modified nucleobases."

[0082] As used herein, the term "nucleotide" refers to a polynucleotide monomer unit consisting of a core base, a sugar, and one or more bonds (e.g., phosphate bonds in natural DNA and RNA). In some cases, the bond can be a non-naturally occurring and / or modified bond. In some embodiments, the bond can be an internucleoside bond as described herein. In a specific embodiment, the modified bond is a PS bond. In some embodiments, a "nucleotide" refers to a nucleotide unit in an oligonucleotide or a nucleic acid. The term "nucleotide" encompasses all modified forms and derivatives of "nucleosides" and "modified core bases."

[0083] As used herein, the term "bond between nucleosides" refers to the bond between adjacent nucleosides. "Bond between nucleosides" and "bond" can be used interchangeably. The bond can be continuous or uninterrupted. The bond can be discontinuous or interrupted. As used herein, the term "discontinuous" or "interrupted" refers to a bond modification between continuous nucleosides of the same modification that is no more than, for example, 4, 5, 6, 7 or more. In some embodiments, the naturally occurring PO bond is replaced by a modified bond between nucleosides. Therefore, in some embodiments, the bond is a non-natural bond between nucleosides. In some embodiments, the bond between nucleosides includes but is not limited to phosphorothioate (PS) bonds, 3'-methylenephosphonate bonds, 5'-methylenephosphonate bonds, 3'-phosphoramidate bonds, 2'-5'-phosphodiester bonds, and phosphoguanidine (PN) bonds. In another embodiment, the bond between nucleosides is modified to a 3'-3' phosphate bond or a 5'-5' phosphate bond (3'-P-3' and 5'-P-5'). The bond between nucleosides can be stereopure or stereorandom. In one embodiment, the nucleoside bond of natural 3 '-5 ' phosphodiester bond is replaced by a modified nucleoside bond. In some embodiments, the nucleoside bond of naturally occurring one or more PO bonds is replaced by a modified nucleoside bond to introduce one or more PS bonds or non-phosphorus derived nucleoside bonds.

[0084] As used herein, the term "stereopure" or "stereorandom" refers to a chemically modified oligonucleotide. Specifically, the term "stereopure" refers to a chirally pure (or "stereochemically pure") oligonucleotide. The term "stereorandom" refers to a racemic (or "stereorandom," "achiral control") oligonucleotide. Therefore, the oligonucleotide of the present invention comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more stereorandom internucleoside bonds (e.g., a mixture of Rp phosphorus bonds and Sp phosphorus bonds at the internucleoside bond, which are for example synthesized by traditional achiral control oligonucleotides). In one embodiment, the internucleoside bond is a thiophosphate (PS) bond. In one embodiment, the internucleoside bond is a stereorandom PS bond. In one embodiment, the internucleoside bond is a chirally controlled PS bonding. In one embodiment, the internucleoside bond is a phosphoguanidine (PN) bond.

[0085] As used herein, the term "hydroxy" represents an -OH group.

[0086] As used herein, the term "antisense oligonucleotide" or "ASO" refers to a short chain of nucleotide analogs that can hybridize to complementary mRNA in a sequence-specific manner through Watson-Crick base pairing. ASOs can contain DNA and RNA. ASOs can be chemically modified. As used herein, the terms "antisense oligonucleotide" (ASO) and "oligonucleotide" are used interchangeably.

[0087] The term "modified sugar" refers to a portion that can replace a naturally occurring sugar. The modified sugar simulates the spatial arrangement, electronic properties, or other physicochemical properties of a sugar. Naturally occurring sugars are pentoses (five-carbon sugars), deoxyribose (for forming DNA), or ribose (for forming RNA), but it should be understood that naturally occurring and non-naturally occurring sugar analogs are also included. For example, other sugars can include, for example, C4 sugars, C5 sugars, and / or C6 sugars. In some embodiments, the modified sugar is a substituted ribose or deoxyribose. In some embodiments, the modified sugar comprises a 2'-modification. Examples of useful 2'-sugar modifications (e.g., 2'-ribose, 2'-deoxyribose, 2'-arabinose, etc.) are widely used in the art and described herein. Those skilled in the art will understand after reading this disclosure that various types of 2'-sugar modifications are known and can be utilized according to the present disclosure. In some embodiments, the modified sugar is a bicyclic sugar (e.g., a sugar for LNA, BNA, etc.). In some embodiments, the modified sugar is an LNA sugar. The term "locked nucleic acid" (LNA) is also referred to as bridged nucleic acid (BNA), which refers to a modified RNA nucleotide in which the ribose moiety is modified by an additional bridge connecting 2' oxygen and 4' carbon. In some embodiments, the sugar modification is 2'-OMe, 2'-O-methoxyethyl (2'-MOE), 2'-F, 5'-vinyl, or S-restricted ethyl (S-cEt). In one embodiment, the 2'-modification is the C2-stereoisomer of 2'-F-ribose. In one embodiment, the 2'-modification is 2'-F. In one embodiment, the 2'-modification is 2'-FANA. In one embodiment, the modified sugar is a morpholino sugar. In one embodiment, the oligonucleotide comprises, for example, UNA (non-locked nucleic acid), PMO (morpholino connected by phosphorodiamidate) or PNA (peptide nucleic acid). Therefore, in one embodiment, the nucleic acid analog is PNA (peptide nucleic acid). In one embodiment, the nucleic acid analog is PMO (morpholino connected by phosphorodiamidate). In one embodiment, the 2'-modification is 2'-O-alkyl modification. In one embodiment, the 2'-O-alkyl modification is a 2'-O-methyl-, 2'-O-ethyl-, 2'-O-propyl- or 2'-MOE modification. In a preferred embodiment, the 2'-modification is 2'-OMe. In some embodiments, the 2'-modification is 2'-MOE. In some embodiments, the 2'-modification is 2'-OR, wherein R is a substituted C1-10 aliphatic. In some embodiments, in the context of oligonucleotides, the modified sugar refers to a sugar that is not a ribose or deoxyribose sugar commonly found in natural RNA or DNA (e.g., arabinose). In some cases, the 2'-O-alkyl modification is not 2'-MOE.

[0088] The term "FANA" or "FANA-modified" refers to a 2'-fluoroarabinoside-modified nucleobase and / or an oligonucleotide comprising such a nucleobase. For example, the expression "FANA-cytidine" refers to a cytidine comprising a 2'-fluoro-β-D-arabinoside nucleoside sugar modification. In the context of the present invention, the expression "derivative thereof" refers to a corresponding nucleotide or oligonucleotide chemically derived from the nucleotide or oligonucleotide.

[0089] As used herein, the term "complementary" or "partially complementary" or "substantially complementary" refers to a nucleic acid sequence that is capable of specific intermolecular base pairing due to its complementary nucleotides. For example, an oligonucleotide may comprise a nucleic acid sequence that is complementary to a target sequence (e.g., SERPINA1 or any other target sequence). It will be understood by those skilled in the art that in many cases, complete complementarity is not necessary, and one or more wobbles (wobble base pairing), bulges, mismatches, etc. can be well tolerated. One or more wobbles, bulges, mismatches can be located inside or outside the CBT. For example, the ASO of the present invention comprises a mismatch relative to the target adenosine. Therefore, the complementarity of the ASO of the present invention can be 100%, but with the exception of the nucleoside relative to the target nucleoside to be edited. In one embodiment, the complementarity is at least 80%, 85%, 90% or 95%. In one embodiment, the complementarity is 85% to 99%. In another embodiment, when aligned with the target nucleic acid, the ASO comprises 1, 2, 3, 4 or 5 mismatches. In one embodiment, the ASO comprises a wobble base outside the CBT. In one embodiment, one or more mismatches are independently wobble base pairing. In one embodiment, the ASO comprises up to 4 mismatches or wobble bases outside the CBT. In one embodiment, the ASO comprises up to 3 mismatches or wobble bases outside the CBT.

[0090] As used herein, the term "mutation" refers to the replacement of a residue in a sequence (e.g., a nucleic acid or amino acid sequence) by another residue, or the deletion or insertion of one or more residues in the sequence. Mutations are generally described herein by identifying the original residue, subsequently identifying the position of the residue in the sequence, and identifying the identity of the newly replaced residue. It is noteworthy that the present invention is not limited to correcting mutations, as it may also be useful to convert a wild-type sequence into a mutant sequence by applying the ASO according to the present invention. Various methods for performing amino acid substitutions (mutations) provided herein are well known in the art, such as Green and Sambrook, Molecular Cloning: A Laboratory Manual (4th ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 2012).

[0091] As used herein, the term "beneficial editing" refers to editing RNA derived from a wild-type allele (rather than a mutant allele) to, for example, modulate the function of a wild-type protein in an effective manner, thereby preventing or treating a disease. For example, beneficial editing can include sites that are not the cause of a genetic disease but represent wild-type protein sites, such as STAT1 Y701, NLRP3 Y166, and CTNNB1 T41. These sites mutate (without a potential G to A mutation) and alter the function of the wild-type protein.

[0092] The term "compensatory editing" refers to modifications that alter RNA nucleotides and correct one or more deleterious or adverse changes in the RNA sequence compared to wild type. For example, a compensatory A to I change may help functionally compensate for an otherwise non-editable mutation to alleviate a disease phenotype.

[0093] As used herein, the term "off-target" refers to nonspecific and unintended genetic modifications to a target. Specifically, off-target editing can include unintended point mutations, deletions, insertions, inversions, and translocations.

[0094] As used herein, the term "adenosine deaminase" or "adenosine deaminase acting on RNA" (ADAR) refers to any (poly)peptide, protein, or protein domain or fragment thereof that can catalyze the hydrolytic deamination of adenosine to inosine. Therefore, the term refers not only to full-length and wild-type ADARs, but also to functional fragments or functional variants of ADARs. In some embodiments, ADARs are (endogenous) adenosine deaminases that catalyze the deamination of adenosine to inosine or the deamination of deoxyadenosine to deoxyinosine. In some embodiments, ADARs catalyze the deamination of adenine or adenosine in deoxyribonucleic acid (DNA) or ribonucleic acid (RNA). The ADAR can be a human ADAR. The ADAR can be an endogenous ADAR. Therefore, in some embodiments, the ADAR is endogenous human ADAR1, ADAR2, or ADAR3 (hADAR1, hADAR2, or hADAR3), or any fragment or isomer thereof (e.g., hADAR1p110 and p150).

[0095] As used herein, the term "guide RNA" (gRNA) or "guide oligonucleotide" refers to a segment of RNA or oligonucleotide (comprising RNA and / or DNA) that serves as a guide for an enzyme and forms a complex with the enzyme. The guide RNA or guide oligonucleotide may comprise an endogenous sequence and / or an exogenous sequence. The guide can be used in vitro and in vivo. For example, a guide RNA or guide oligonucleotide guides a base modification activity or base editing function (e.g., ADAR) to a target to be trans-edited.

[0096] As used herein, the term "target RNA" generally refers to the RNA that undergoes the editing reaction and is "targeted" by the corresponding ASO of the present invention.

[0097] As used herein, the terms "disease" or "disorder" are used interchangeably to refer to a condition in a subject. In some embodiments, the condition is a disease in a subject whose severity can be reduced by administering a pharmaceutical composition to induce an immune response in the subject.

[0098] As used herein, in the context of administering a therapy to a subject, the term "effective amount" refers to an amount of the therapy that has a prophylactic and / or therapeutic effect.

[0099] As used herein, the term "in combination" in the context of administering two or more therapies to a subject refers to the use of more than one therapy (e.g., more than one prophylactic and / or therapeutic agent). The use of the term "in combination" does not limit the order in which the therapies are administered to a subject.

[0100] As used herein, in the context of the present invention and when administering one or more therapies to a subject, the term "preventing" refers to inhibiting the development or onset of a disease or its symptoms. In one embodiment, it involves administering a compound to a patient known to be at high risk of developing a certain condition, disorder, or disease.

[0101] As used herein, in the context of the present invention, the term "treatment" refers to the administration of a compound to a patient who has exhibited signs and / or symptoms of a certain condition, disorder, or disease. Beneficial or desired clinical results include, but are not limited to: relief of symptoms; alleviation of the extent of the condition, disorder, or disease; stabilization (i.e., no worsening) of the condition, disorder, or disease state; delaying the onset of the condition, disorder, or disease or slowing its progression; alleviation of the condition, disorder, or disease state or relief (whether partial or complete relief), whether detectable or undetectable; improvement of at least one measurable physical parameter, but the patient may not necessarily be able to perceive it; or enhancement or improvement of the condition, disorder, or disease. Treatment includes inducing a clinically significant response without excessive side effects. Treatment also includes prolonging survival compared to the expected survival without treatment.

[0102] The terms "subject" or "patient" are used interchangeably and refer to an animal (e.g., a mammal) to whom the compounds of the invention may be administered, either in the human or veterinary setting. In a specific embodiment, the subject is a human. The oligonucleotides of the invention may be administered to the subject for beneficial editing. The oligonucleotides of the invention may be administered to the subject for compensatory editing.

[0103] As used herein, the term "pharmaceutically acceptable" refers to that approved by regulatory agencies. The term "carrier" refers to a diluent, adjuvant, excipient, or vehicle administered with a pharmaceutical composition. Saline solutions and aqueous solutions of glucose and glycerol can also be used as liquid carriers, particularly for injections. Suitable excipients include starch, glucose, lactose, sucrose, gelatin, malt, rice, flour, chalk, silica gel, sodium stearate, glyceryl monostearate, talc, sodium chloride, skimmed milk powder, glycerol, propylene glycol, water, ethanol, and the like. The formulation should be compatible with the mode of administration.

[0104] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials used in the present disclosure are described herein; other suitable methods and materials known in the art may also be used.

[0105] (Antisense) oligonucleotides

[0106] In particular, chemically modified (antisense) oligonucleotides (ASOs) are provided herein. Although not limited to any particular theory of operation, it is believed that the core base and backbone bond modifications of the ASOs contribute to stabilizing and improving the editing efficiency and lysosomal stability of the oligonucleotides of the present invention. In addition, these modifications also have the potential to reduce the off-target editing of different ASOs. Since one or more modifications can be transferred to various oligonucleotide sequences by synthesis, such modifications have the potential to improve the editing efficiency of oligonucleotides with different specificities. The ASOs of the present invention can be used for a variety of purposes. Advantageously, the oligonucleotides provided herein can contribute to editing one or more G to A mutations. It is noteworthy that the ASOs of the present invention are not limited to correcting G to A mutations, but can also be used to convert wild-type sequences into mutant sequences to regulate protein expression and / or function ("beneficial editing"), or to compensate for mutations that are not G to A mutations. Therefore, oligonucleotides and compositions comprising the oligonucleotides can be used as active agents of drugs, for the treatment of genetic disorders, conditions or diseases associated with one or more G to A mutations.

[0107] The inventors of the present invention realized that, in the context of providing oligonucleotides for RNA editing, in order to achieve a beneficial balance between high editing efficiency and lysosomal stability, it is necessary to introduce some features into the oligonucleotides. Specifically, the inventors found that the oligonucleotides should have a mixture of different modifications at the 2' position of the sugar residue, and fragments of more than 6 nucleotides with the same 2'-modification should be avoided. Avoiding uniform segments of more than 6 nucleotides with the same 2'-modification can prevent significant loss of natural ADAR editing activity. In addition to this design feature, the modified oligonucleotides of the present invention should also be such that at least two of the three nucleotides of CBT are modified at the 2' position of the sugar base, or are deoxyribonucleosides, thereby enhancing their stability against nuclease digestion. In addition, it was found that the use of phosphorothioate (PS) bonds at the h and i positions of the core sequence should be avoided. The PS bonds at these positions were found to severely impair editing. However, the oligonucleotides of the present invention do benefit from modifications to the internucleoside bonds at a basal level at other positions, as the inventors have found that modifications of at least 15% are conducive to achieving good RNA editing. Oligonucleotides can benefit from modifications at the 2' position of the nucleotide, and these modifications should be composed of different groups. Therefore, studies have found that a mixture of 2'-F- and 2'-O-alkyl modifications is beneficial, with a minimum of 10% of each modification being desirable. When these properties are combined, they confer high levels of lysosomal stability and RNA editing efficiency on the oligonucleotides. Therefore, the oligonucleotides of the present invention are preferably modified and designed accordingly.

[0108] According to the present invention, the core oligonucleotide comprises the following sequence: 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j-3', and contain specific 2'-modifications and internucleoside bond patterns that contribute particularly to the excellent properties of the oligonucleotide. The core sequence can have PS bonds at the d, e, and optionally a positions. Notably, regions that are particularly sensitive to the discovered bond patterns are CBTs and adjacent hotspots ("extended hotspots"), where very specific nucleotide modifications are also required to ideally fit into the enzymatic active site of the ADAR to achieve the editing effect. Typically, the oligonucleotide contains a mixture of 2'-F-, 2'-OMe, and 2'-H modifications, at least 15% internucleoside bond modifications, no more than 6 consecutive nucleotides of the same 2'-modification, and N +2 2'-O-alkyl modification and N +3 Oligonucleotides with a 2'-F-modification at the end of the cleavage site ("extended hotspot") may provide optimal editing and lysosomal stability (eg, "extended hotspot" in Example 15).

[0109] Therefore, the present invention provides a chemically modified oligonucleotide comprising a sequence of 23 to 80 nucleotides in length, capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, and wherein the core oligonucleotide comprises the following sequence:

[0110] 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0111] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0112] (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification;

[0113] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0114] (d) an internucleoside linkage modification content of at least 15%; and

[0115] (e) Bond h and bond i are not phosphorothioate (PS) bonds.

[0116] The inventors provide evidence in this application for the beneficial effects of having a 2'-O-alkyl modification at the +2 position and a 2'-fluoro (2'-F)-modification at the +3 position ("modification hotspot"). Specifically, the results show that at N +2 with 2'-OMe at N +3 Constructs with 2'-F at the position N exhibited high levels of RNA editing yield. +2 The nucleotides are 2'-O-alkyl modified. In one embodiment, N +3 In one embodiment, the nucleotides carry a 2'-F-modification. +2 Nucleotides are 2'-O-alkyl modified, N +3 The nucleotides are 2'-F-modified.

[0117] The inventors have also shown that uniform segments or continuous segments of nucleotides carrying the same chemical modification may interfere with the activity of ASOs. Therefore, the oligonucleotides of the present invention can be modified to not contain the same 2'-sugar modification. In one embodiment, no more than 6 consecutive nucleotides have the same 2'-modification. In one embodiment, no more than 5 consecutive nucleotides have the same modification. In one embodiment, no more than 4 consecutive nucleotides have the same modification. In one embodiment, no more than 3 consecutive nucleotides have the same modification. In one embodiment, no more than 2 consecutive nucleotides have the same modification.

[0118] In general, metabolically unstable ASOs may be beneficial for some highly transient therapeutic effects (e.g., wound healing). Reversal of classic pathogenic point mutations requires metabolically stable ASOs to reduce dosing frequency. Although embodiments with low 2'-modification content are also conceivable, the inventors aim to provide maximum stability by replacing each RNA nucleoside with a 2'-modified RNA or DNA. As shown in the application, the inventors have realized that as long as the segments with the same 2'-modification do not exceed 6 consecutive nucleotides, oligonucleotides can tolerate a high percentage of 2'-modifications without causing a deleterious loss of activity. Therefore, in one embodiment, 20%-100%, 30%-100%, 40%-100%, 50%-100%, 60%-100%, 70%-100%, 80%-100%, or 90%-100% of the nucleotides are deoxyribonucleosides (DNA) or 2'-modified nucleotides. In one embodiment, 20%-100% of the nucleotides are DNA or 2'-modified nucleotides. In one embodiment, 50%-100% of the nucleotides are DNA or 2'-modified nucleotides. In one embodiment, 100% of the nucleotides are DNA or 2'-modified nucleotides. In one embodiment, 30%-95%, 40%-95%, 40%-90%, 50%-95%, 50%-90%, 60%-95%, or 60%-90% of the nucleotides are DNA or 2'-modified nucleotides. In one embodiment, the above percentages are such that there is only 2'-modified nucleotides and no DNA.

[0119] According to the present invention, oligonucleotides of the present invention will comprise modifications at the 2' position of the nucleotides and have different modifying groups. In one embodiment, 20%-70% of the nucleotides are 2'-F modified. In one embodiment, 35%-65% of the nucleotides are 2'-F modified. In one embodiment, 20%-60% of the nucleotides are 2'-O-methyl (2'-OMe) modified. In one embodiment, 25%-55% of the nucleotides are 2'-OMe modified.

[0120] The inventor also recognizes that modified oligonucleotide of the present invention does not need all internucleosides bond all to be modified to prevent lysosomal degradation, as long as introduce minimum level internucleoside modification, and as long as the d key and the e key of core oligonucleotide sequence (as described above) are modified.In one embodiment, the content that internucleoside bond is modified is at least 15%.In one embodiment, the content that internucleoside bond is modified is at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 80% or 90%.In one embodiment, the bond that is no more than 95%, 90%, 85%, 80%, 70%, 60%, 50%, 40% or 30% is that internucleoside bond is modified.

[0121] Internucleoside linkage modifications, such as PS linkages, often have a positive impact on the pharmacokinetics of ASOs, as well as stability, protein binding, and intracellular localization. However, at the same time, it is desirable to reduce the total PS content to reduce, for example, toxicity and non-specific protein binding. In one embodiment, (a) no more than 95%, 90%, 85%, 80%, 70%, 60%, 50%, 40%, 30%, or 20% of the linkages outside the CBT are internucleoside linkage modifications; or (b) 15%-90% of the linkages are internucleoside linkage modifications, preferably 40%-80%, and most preferably 45%-60% of the linkages are internucleoside linkage modifications. In one embodiment, no more than 95%, 90%, 85%, 80%, 70%, 60%, 50%, 40%, 30%, or 20% of the linkages outside the CBT are internucleoside linkage modifications. In one embodiment, 15%-90% of the bonds are modified internucleosides, preferably wherein 40%-80%, most preferably 45%-60% of the bonds are modified internucleosides. In one embodiment, the content of the modified internucleosides bonds is at least 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80% or 90%. In one embodiment, the content of the modified internucleosides bonds is no more than 95%, 90%, 85%, 80%, 75%, 70%, 65%, 60%, 55%, 50%, 45%, 40%, 35%, 30%, 25% or 20%. In one embodiment, the content of internucleoside bond modification is 10%-90%, 15%-90%, 15%-80%, 15%-70%, 15%-60%, 20%-90%, 10%-80%, 20%-80%, 25%-80%, 30%-80%, 30%-90%, 40%-90%, 40%-80%, 40%-70%, 45%-90%, 45%-85%, 45%-75%, 45%-70%, 45%-60% or 45%-55%. In one embodiment, 15%-90% of the bond is an internucleoside bond modification. In one embodiment, 40%-80% of the bond is an internucleoside bond modification. In one embodiment, 45%-60% of the bond is an internucleoside bond modification. In one embodiment, the content of internucleoside bond modification is 20%, 30%, 40%, 45%, 50%, 60%, 70%, 80% or 90%. In one embodiment, the content of internucleoside bond modification is 30%. In one embodiment, the content of internucleoside bond modification is 15%.

[0122] In order to achieve optimal RNA editing, oligonucleotides of different lengths may require a mixture of different specific 2'-modifications and internucleoside bond modifications. The shorter the oligonucleotide, the greater the possibility of endosomal escape. In addition, the toxicity of a particular oligonucleotide may also depend on its length. In addition, shorter oligonucleotides can have higher specificity. On the other hand, longer oligonucleotides can bind more firmly or faster to their respective RNA targets, while enhancing the protrusions, mismatches, and wiggles of editing may also work better in long oligonucleotides. Therefore, the long oligonucleotides and short oligonucleotides of the present invention each have their pros and cons. Therefore, the oligonucleotides of the present invention can have different lengths. The length of the oligonucleotide can be about 23 to 80 nucleotides, for example, a length of about 23 to 50 nucleotides, or a length of about 40 to 80 nucleotides. In one embodiment, the oligonucleotide is 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79 or 80 nucleotides in length. In a preferred embodiment, the oligonucleotide is 59 nucleotides in length. In some embodiments, the oligonucleotide is 23 to 80 nucleotides in length. In some embodiments, the length of the oligonucleotide is 23-80, 23-70, 23-60, 23-50, 23-40, 23-33 or 23-38 nucleotides. In some embodiments, the length of the oligonucleotide is 25-80, 25-70, 25-60, 25-50, 25-40 nucleotides. In some embodiments, the length of the oligonucleotide is 30-80, 30-70, 30-60, 30-50, 30-40 nucleotides. In some embodiments, the length of the oligonucleotide is 40-80, 50-80, 60-80 or 70-80 nucleotides. 78,79, or 80 nucleotides in length.In one embodiment, the oligonucleotide is 25-60, 25-55, 25-50, 25-45, 25-40, 25-35, or 25-30 nucleotides in length. In one embodiment, the oligonucleotide is 28-70 nucleotides in length. In one embodiment, the oligonucleotide is: (i) 28-60, 28-55, or 28-45 nucleotides in length; (ii) 59 nucleotides in length; or (iii) no more than 45 nucleotides in length. Ranges and lengths between the above ranges and lengths are also considered part of the present invention. In one embodiment, the oligonucleotide is 40 nucleotides in length. In one embodiment, the oligonucleotide is 45 or fewer nucleotides in length, and wherein outside the CBT, no more than 4 nucleotides are deoxynucleotides. In one embodiment, the oligonucleotide is 40 nucleotides in length. In one embodiment, the oligonucleotide is 45 nucleotides in length. In one embodiment, the oligonucleotide is 35 nucleotides in length. In one embodiment, the oligonucleotide is 33 nucleotides in length. In one embodiment, the oligonucleotide is 32 nucleotides in length. In one embodiment, the oligonucleotide is 30 nucleotides in length. In one embodiment, the oligonucleotide is 25 nucleotides in length.

[0123] The inventors found that longer oligonucleotides (about 40 to 80 nucleotides in length) containing a mixture of the above modifications and designs can tolerate a total deoxyribonucleoside content of 5%-50%. In addition, the inventors unexpectedly found that oligonucleotides containing such modifications and designs can be further shortened to shorter sequences (≤45nt) while still providing good RNA editing (e.g., Examples 8 to 14). In addition, shorter oligonucleotides with a total number of no more than 6 deoxyribonucleosides outside the CBT and at least 30% internucleoside modification content provide good RNA editing (e.g., Example 17). Continuous nucleotide fragments and high levels of internucleoside modification content generally result in increased editing effects.

[0124] Therefore, the present invention also provides a modified oligonucleotide comprising a sequence of 23 to 50 nucleotides in length, capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, and wherein the core oligonucleotide comprises the following sequence:

[0125] 5'-N -5 a N -4 b N -3 c N -2 dA -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0126] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0127] (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification;

[0128] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0129] (d) the regions adjacent to the 3' and 5' ends of the CBT contain no more than 6 deoxyribonucleosides in total; and

[0130] (e) The content of internucleoside linkage modifications is at least 30%.

[0131] Although the oligonucleotides of the present invention can tolerate DNA outside the CBT, placing an extremely high proportion of DNA (or a high ratio of DNA to RNA) tends to interfere with editing efficiency. Nevertheless, potentially useful embodiments can be produced that contain a large number of deoxyribonucleotides that strike a reasonable balance between stability and editing efficiency. Therefore, the oligonucleotides of the present invention can contain different amounts of DNA. Specifically, the oligonucleotides can have different amounts of DNA outside the CBT (2'-H modification). Therefore, in one embodiment, the total number of deoxyribonucleosides contained in the region adjacent to the 3' end and the 5' end of the CBT does not exceed 6. In one embodiment, the total number of deoxyribonucleosides contained in the region adjacent to the 3' end and the 5' end of the CBT does not exceed 5, 4 or 3.

[0132] Outside the CBT, 8, 7, 6, 5, 4, 3, 2, 1 or 0 nucleobases may be deoxyribonucleotides. In one embodiment, outside the CBT, the oligonucleotide does not contain any deoxyribonucleosides. In one embodiment, outside the CBT, no more than 1, 2, 3 or 4 nucleobases are deoxyribonucleotides. In one embodiment, outside the CBT, no more than 3 nucleobases are deoxyribonucleotides. When compared to shorter ASOs, longer ASOs tend to tolerate a higher amount of DNA outside the CBT. Thus, further provided herein is an oligonucleotide comprising a sequence of 40 to 80 nucleotides in length capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), wherein the central nucleotide (N0) is directly opposite the target adenosine in the target RNA, and wherein the core oligonucleotide comprises the following sequence:

[0133] 5'-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3'; and wherein:

[0134] (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications;

[0135] (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification;

[0136] (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification;

[0137] (d) The total content of deoxyribonucleosides in the region adjacent to the 3' end and the 5' end of CBT is 5% to 50%.

[0138] In one embodiment, the total deoxyribonucleoside content of the region adjacent to the 3' and 5' ends of the CBT is 10% to 50%. In one embodiment, the deoxyribonucleoside content outside the CBT is 10% to 40%, more preferably 11% to 30%, even more preferably 13% to 25%.

[0139] According to the present invention, chemically modified oligonucleotides include internucleoside bond modifications. In one embodiment, oligonucleotides include at least one internucleoside bond modification selected from phosphorothioate (PS), 3'-methylenephosphonate, 5'-methylenephosphonate, 3'-phosphoramidate, 2'-5' phosphodiester and phosphoguanidine (PN). In a preferred embodiment, internucleoside bond modifications are PS bonds. In one embodiment, internucleoside bond modifications are 3'-methylenephosphonate bonds. In one embodiment, internucleoside bond modifications are 5'-methylenephosphonate bonds. In one embodiment, internucleoside bond modifications are 3'-phosphoramidate bonds. In one embodiment, internucleoside bond modifications are 2'-5' phosphodiester bonds. In one embodiment, internucleoside bond modifications are phosphoguanidine (PN) bonds. In one embodiment, nucleic acid analogs are PNA (peptide nucleic acid). In one embodiment, nucleic acid analogs are PMO (morpholino connected by phosphorodiamidate). In one embodiment, oligonucleotides include PS bonds, phosphate (PO) bonds and / or phosphorodiamidate bonds. In one embodiment, at least one internucleoside bond modification is PS. In one embodiment, the oligonucleotide has a stretch of consecutive PS bonds. In one embodiment, the stretch of consecutive PS bonds has 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or more bonds.

[0140] The positioning of the internucleoside bond within the oligonucleotide plays an important role in determining the balance between high editing yield and long half-life. At the same time, the need to carefully consider the position of the backbone modification also reflects the sensitivity of the CBT and the bases around the CBT to fine chemical modifications on the ribose, nucleobase or bond. For example, the h and i positions can be chemically modified or unmodified. However, the inventors of the present application found that the PS bond should not be placed at the h and i positions, which leads to obvious damage to the editing efficiency of the construct. Therefore, in one embodiment, bond h and bond i are not chemically modified. In one embodiment, bond h is not chemically modified. In one embodiment, bond i is not chemically modified. In some embodiments, bond h and bond i are phosphate (PO) bonds. In some embodiments, bond h and bond i are not thiophosphate (PS) bonds. In one embodiment, up to three bonds selected from bond b, bond c, bond f, bond g and bond j are PS bonds. However, the case where all bonds a to j are PS bonds is not included. In a particularly preferred embodiment, bond f is a PS bond. In a particularly preferred embodiment, bonds a, d, and e are PS bonds, while bonds h and i are PO bonds.

[0141] The stability and editing efficiency of various oligonucleotides may be affected by the number and continuous arrangement of specific 2'-modifications. That is, it has been found that repeated modifications of the same type of 2'-modification will impair the RNA editing efficiency of the oligonucleotide. Therefore, the inventors set out to study the effects of destroying continuous fragments or segments of the same 2'-modification. As shown in the present application, oligonucleotides containing no more than 6 consecutive smaller segments of nucleotides with the same 2'-modification provide the best editing efficiency. Therefore, the oligonucleotides of the present invention do not contain more than about 6 nucleotides with the same 2'-modification of uniform segments. For example, in one embodiment, the oligonucleotide contains no more than 6 consecutive nucleotides modified by 2'-F- and / or 2'-O-alkyl. In one embodiment, no more than 4, 5 or 6 consecutive nucleotides are modified by 2'-F; and / or no more than 4, 5 or 6 consecutive nucleotides are modified by 2'-O-alkyl. In one embodiment, no more than 4, 5 or 6 consecutive nucleotides are modified by 2'-F. In one embodiment, no more than 4, 5 or 6 consecutive nucleotides are modified by 2'-O-alkyl. In one embodiment, the oligonucleotide comprises four consecutive 2'-F- and / or 2'-O-alkyl modified nucleotides. In one embodiment, the oligonucleotide comprises five consecutive 2'-F- and / or 2'-O-alkyl modified nucleotides. In one embodiment, the oligonucleotide comprises six consecutive 2'-F- and / or 2'-O-alkyl modified nucleotides. The oligonucleotide may comprise less than four consecutive nucleotides having the same 2'-modification. In a preferred embodiment, the 2'-O-alkyl modification is a 2'-OMe modification.

[0142] In addition, studies have found that uniform and segmented 2'-modifications in oligonucleotides used in the prior art lead to a significant loss of editing activity. Without being bound by any theory, the inventors believe that this is consistent with the negative impact of a large number of 2'-modifications on the binding of the dsRNA binding domain to the dsRNA substrate. dsRNA recognition (e.g., by the dsRBD of deaminases and / or ADARs) and dsRNA binding typically occur through the interaction of proteins with the minor groove of the RNA helix. The 2'-ribose modification extends into the minor groove, creates a spatial demand, and changes the hydration of the RNA helix. The 2'-F modification is spatially most similar to the natural 2'-OH in ribose, but is very hydrophobic and may interfere with the hydration reaction. The spatial demand of the 2'-O-methyl modification is higher, and the 2'-MOE modification is even more so. Therefore, it makes sense that a large number of 2' modifications are not easily accepted, especially in large segments, because they hinder the binding of dsRBD. This is particularly applicable to 2'-MOE modifications, but also to larger segments modified with 2'-O-methyl. However, stretches of consecutive 2'-F residues may not be ideal due to their strong hydrophobicity, but they are more readily accepted than 2'-O-methyl residues. Therefore, a mixture of 2'-F and 2'-O-methyl modifications may provide a way to construct duplexes that facilitate ADAR binding. However, for the reasons mentioned above, 2'-F modifications are more readily accepted.

[0143] Therefore, the inventors have found that avoiding uniform segments of more than 6 nucleotides with the same 2'-modification can prevent severe loss of natural ADAR editing activity. Therefore, in one embodiment, less than 6, 5, 4 or 3 consecutive nucleotides have the same 2'-modification. In one embodiment, no more than 6 consecutive nucleotides are 2'-F modified. In one embodiment, no more than 5 consecutive nucleotides are 2'-F modified. In one embodiment, no more than 4 consecutive nucleotides are 2'-F modified. In one embodiment, no more than 6 consecutive nucleotides are 2'-O-alkyl modified. In one embodiment, no more than 5 consecutive nucleotides are 2'-O-alkyl modified. In one embodiment, no more than 4 consecutive nucleotides are 2'-O-alkyl modified, optionally wherein no more than 4 consecutive nucleotides are 2'-OMe modified. In one embodiment, the oligonucleotide comprises 2, 3, 4, 5 or 6 consecutive nucleotides with the same 2'-modification, for example, 5 consecutive nucleotides are 2'-F modified.

[0144] Different types of PS modifications can affect the efficacy of oligonucleotides. Therefore, the present application also provides chemically modified ASOs with reduced stereopure bond chemistry, that is, the ASOs of the present invention relate to stereorandom PS-modified ASOs, which are generally easier to produce and less expensive. Without being bound by any theory, the inventors believe that, in contrast to the ASOs of the prior art, high levels of stereopure bonds may be neither useful nor necessary to provide effective RNA editing. Given that the active site of the deaminase must bind the target RNA / oligonucleotide drug duplex near the N0 position, some bonds (e.g., at the position where the protein interacts with the phosphate backbone) may potentially benefit from the insertion of stereopure bond modifications, such as stereopure PS (or PN) modifications. However, the inventors believe that, if present, these beneficial effects of stereopure bond modifications may only apply to a few sites; each ASO may have no more than 10 stereopure chains, and more likely no more than 5 stereopure chains per ASO. Therefore, the inventors believe that the design of ASOs for efficient editing does not require the extremely high degree of stereopure bonds used in the prior art.

[0145] In one embodiment, the oligonucleotide comprises a stereo-random internucleoside bond. In one embodiment, the oligonucleotide comprises one or more stereo-random internucleoside bond modifications. In one embodiment, the oligonucleotide does not comprise a stereo-pure PS bond modification. In one embodiment, the oligonucleotide comprises no more than 10, preferably no more than 5 stereo-pure internucleoside bonds. In one embodiment, the oligonucleotide comprises no more than 5 stereo-pure internucleoside bonds. In one embodiment, the oligonucleotide comprises 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 stereo-pure internucleoside bonds.

[0146] Although the oligonucleotides of the present invention contain different internucleoside bond modifications, the inventors have demonstrated that optimized PS bond modifications combined with balanced 2'-modifications and reduced segment sizes are beneficial for providing oligonucleotides with good RNA editing effects. It is worth noting that, without being bound by any theory, the inventors propose that different bond modifications (or higher levels of some types of bond modifications, such as phosphoguanidine (PN) bonds) may cause the oligonucleotides to be unable to exhibit sufficient RNA editing efficiency. Therefore, in one embodiment, the stereopure bond is a PS bond. In one embodiment, the stereopure bond is a PS bond and / or a PN bond. In one embodiment, the stereopure bond is a PN bond. In one embodiment, the oligonucleotide does not comprise a stereopure PS bond and / or a stereopure PN bond. In one embodiment, the oligonucleotide does not comprise a stereopure PS bond. In one embodiment, the oligonucleotide does not comprise a stereopure PN bond. In one embodiment, the chemically modified oligonucleotide does not comprise a stereopure PS bond modification.

[0147] As previously mentioned, incorporating higher levels of RNA into oligonucleotides tends to make them metabolically unstable, which can be an attractive property. In order to achieve the metabolic stability required for oligonucleotides, ideally, the final oligonucleotide should not contain any unmodified RNA nucleobases. In one embodiment, the oligonucleotide does not contain unmodified RNA nucleobases. In one embodiment, the oligonucleotide contains more than 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% or more than 90% modified nucleotides. In one embodiment, the oligonucleotide contains less than 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20% or less than 10% unmodified RNA nucleotides. In this article, modification includes having deoxyribonucleotides.

[0148] 2'-MOE residues are used in splice-switching oligonucleotides and are generally very low in toxicity. However, due to their large size, their large-scale use is not accepted. The inventors of the present invention have found that although 2'-MOE modifications at the ends of oligonucleotides do not affect the overall editing yield, the cytotoxicity of the test constructs is unexpectedly reduced. Specifically, the inventors found that the number of 2'-MOE modifications can be limited to no more than about 6, 7 or 8 nucleotides, and good RNA editing can still be obtained. Similarly, the inventors found that oligonucleotides containing 2 to 6 LNAs provide good RNA editing effects.

[0149] Thus, within an oligonucleotide, the oligonucleotide may comprise no more than 6, 7, or 8 2'-MOE modifications. In one embodiment, the oligonucleotide comprises: (a) a terminal segment of 2'-O-(2-methoxyethyl)-oligoribonucleotide (2'-MOE) at the 3' and 5' ends, wherein each end has no more than 4 nucleotides with 2'-MOE, preferably no more than 3 nucleotides with 2'-MOE; or (b) a locked nucleic acid (LNA) at the end, wherein the oligonucleotide comprises 2 to 6 LNAs at each end or the 5' end, preferably wherein the oligonucleotide comprises 2 LNAs at each end or the 5' end. In one embodiment, each end has no more than 4 nucleotides with 2'-MOE, preferably no more than 3 nucleotides with 2'-MOE. In one embodiment, the oligonucleotide comprises a terminal segment with 2'-MOE at the 3' and 5' ends, wherein each end has no more than 4 nucleotides with 2'-MOE. In one embodiment, the oligonucleotide comprises a 2'-MOE terminal segment at the 3' end and the 5' end, wherein each end does not exceed 3 nucleotides with 2'-MOE. In one embodiment, the oligonucleotide comprises a locked nucleic acid (LNA) at the end, wherein the oligonucleotide comprises 2 to 6 LNAs at each end. In one embodiment, the oligonucleotide comprises a locked nucleic acid (LNA) at the end, wherein the oligonucleotide comprises 2 to 6 LNAs at the 5' end. In one embodiment, the oligonucleotide comprises 2 LNAs at each end. In one embodiment, the oligonucleotide comprises 2 LNAs at the 5' end. In this article, the term "end" refers to the last or terminal nucleotide at both ends of the oligonucleotide, for example, "no more than 4 nucleotides at each end" refers to the last 4 nucleotides at each end of the oligonucleotide. In one embodiment, there is no 2'-MOE modification inside the CBT. In one embodiment, there is no 2'-MOE modification at the +2 position and / or the +3 position.

[0150] In one embodiment, the g bond is not a PS bond. In one embodiment, the g bond is a phosphate (PO) bond.

[0151] In one embodiment, the 2'-O-alkyl modification is a 2'-OMe modification. Of note, in some cases, the 2'-O-alkyl modification is not a 2'-MOE modification.

[0152] CBT is very sensitive to position-specific key modifications, which is due to interference with the binding of the ADAR active site. Therefore, in order to provide efficient editing effects and stabilization of oligonucleotides, the inventors have demonstrated that specific key modifications (mixing) must be placed at specific positions in oligonucleotides. In one embodiment, d and e are PS key modifications, and optionally wherein f is an internucleoside key modification. In one embodiment, d and e are PS key modifications. In one embodiment, f is a PS key.

[0153] In one embodiment, the modification at the 2'-position of the sugar moiety is: (i) a 2'-O-alkyl modification, (ii) a 2'-F-modification, or (iii) a 2'-fluoroarabinoside (FANA)-modification.

[0154] CBT(5'-N -1 -N0-N +1 -3') can carry different modifications and arrangements of modifications. In some embodiments, CBT is chemically modified. That is, N -1 , N0 and / or N +1 The CBT may carry a modification at the 2' position. In one embodiment, only one position in the CBT is chemically modified. In one embodiment, two positions in the CBT are chemically modified. In one embodiment, all positions in the CBT are chemically modified. According to one embodiment, each of the three nucleosides of the CBT may be one of the following or a combination thereof:

[0155] (a) deoxyribonucleotides; and / or

[0156] (b) 2'-FANA-modification; and / or

[0157] (c) 2'-O-methyl-modification; and / or

[0158] (d) 2'-F-modification.

[0159] In one embodiment, at least one of the three oligonucleotides of CBT is a deoxyribonucleotide. In one embodiment, at least one of the three oligonucleotides is 2'-FANA modified. In one embodiment, at least one of the three oligonucleotides is -O-methyl modified. In one embodiment, at least one of the three oligonucleotides is 2'-F modified. In one embodiment,

[0160] (i)N -1 is 2'-F, 2'-FANA, DNA, or 2'-O-methyl; and / or

[0161] (ii) NO is 2'-FANA or DNA; and / or

[0162] (iii)N +1 It is 2'-FANA, DNA or 2'-O-methyl.

[0163] In some embodiments, position N -1 In some embodiments, position N0 is 2'-Fluoro-RNA, 2'-FANA or DNA. In some embodiments, position N+1 is 2'-FANA or DNA. CBT modifications can include any arrangement of the above modifications.

[0164] In one embodiment, NO is deoxycytidine or FANA-cytidine. In one embodiment, NO is deoxycytidine. In one embodiment, NO is FANA-cytidine. Other modifications can include replacing the nucleobase with a (N) heterocycle or aromatic ring that stacks well in the RNA duplex, for example, the nucleobase Z (dZ) (and / or analogs) or 8-oxo-adenosine (8-oxo-A). Therefore, in one embodiment, NO is a nucleobase. In one embodiment, NO is 8-oxo-adenosine.

[0165] This region as well as one or more nucleotides outside of the CBT can be modified. For example, the nucleotides at the 5' and / or 3' ends of the CBT can be chemically modified to carry a 2' modification. The oligonucleotides of the present invention can be modified within a region defining a "hotspot site" or "hotspot region". In one embodiment, the oligonucleotide is modified at the first nucleotide immediately adjacent to the 3' end of the CBT (i.e., position +2) (5'- d A -1 e N0 f A +1 g A +2 h A +3 i In one embodiment, the oligonucleotide is modified at the second nucleotide immediately following the 3' end of the CBT (ie, position +3) (5'- d A -1 e N0 f A +1 g A +2 h A +3 i In a preferred embodiment, the oligonucleotide is modified at the first two nucleotides immediately adjacent to the 3' end of the CBT (ie, positions +2 and +3) (5'- d A -1 e N0 f A +1 g A +2 h A +3 i –'3).

[0166] The nucleotides at the 5' and / or 3' ends of the CBT can be chemically modified to carry a 2'-modification. In one embodiment, positions -5, -4, and -3 are 2'-O-alkyl modified; and / or position -2 is 2'-F-modified. In one embodiment, positions -5, -4, and -3 are 2'-O-alkyl modified. In one embodiment, position -2 is 2'-F-modified. In one embodiment, positions -5, -4, and -3 are 2'-O-alkyl modified, and position -2 is 2'-F-modified. In one embodiment, positions -5, -4, and -3 are 2'-O-alkyl modified, or position -2 is 2'-F-modified. In one embodiment, the 2'-OMe modification is at position +2. In one embodiment, the 2'-F modification is at position +2. In one embodiment, the 2'-F modification is at position -2. In one embodiment, the 2'-OMe modification is at position -5, -4, and -3. In one embodiment, the 2'-OMe modification is at position -5, -4 or -3. Other modifications may include replacing the nucleobase with a (N) heterocycle or aromatic ring that stacks well in the RNA duplex, such as the bena base Z (dZ) (and / or analogs) or 8-oxoadenosine (8-oxo-A). In one embodiment, NO is a bena base. In one embodiment, NO is 8-oxoadenosine (8-oxo-A). Specifically, in a preferred embodiment, N +2 The nucleotides are 2'-O-alkyl modified. In a preferred embodiment, N +3 The nucleotides carry a 2'-fluoro (2'-F) modification. In the most preferred embodiment, N +2 Nucleotides are 2'-O-alkyl modified, N +3 The nucleotides are 2'-fluoro (2'-F) modified.

[0167] In one embodiment, 2'-OMe modification is preferred over DNA in the region near the CBT. In one embodiment, 2'-F modification is preferred over DNA in the region near the CBT. In one embodiment, there is no DNA at position +2 or +3.

[0168] As mentioned above, since internucleoside bond modifications (e.g., PS bonds) tend to have a positive impact on the stability of, for example, ASOs, placing these bonds in ASOs seems to play an important role. In addition, the inventors have found that, with regard to triton lysosome stability, the PS bond at the 3' end of DNA seems to be more important than the PS bond at the 5' end of DNA. Therefore, in some embodiments, the PS bond is located at the 3' end of DNA. In one embodiment, the PS bond is directly located at the 3' end of DNA. In one embodiment, the PS bond is located at the 3' end and the 5' end of DNA. In one embodiment, only 2' modified nucleotides are present in the ASO, and no DNA is present. In one embodiment, the stability of the ASO is improved by placing the PS bond at the 3' end of DNA.

[0169] The larger 2'-sugar modified uniform segments or fragments in the ASO tend to interfere with the binding of the dsRNA binding protein (dsRBD) of the ADAR. Therefore, the oligonucleotides of the present invention can be modified in some way to avoid such interference. For example, the oligonucleotides are modified so that they do not contain continuous segments or uniform segments of nucleotides with the same chemical modification (i.e., to avoid forming segment-shaped modified structures). Therefore, in one embodiment, the modification of the oligonucleotides is not uniform. In one embodiment, the oligonucleotides do not contain uniform segments and / or segment-shaped modified structures. In one embodiment, the oligonucleotides do not contain continuous segments or uniform segments of nucleotides with the same chemical modification (at the 2' position of the sugar moiety). In a preferred embodiment, the oligonucleotides are modified to avoid the occurrence of uniform segments modified by 2'-F- and / or 2'-OMe-. In some embodiments, the oligonucleotides do not contain any 2'-H (DNA) segments. In a preferred embodiment, the oligonucleotides are modified to avoid the occurrence of uniform segments of 2'-F-, 2'-OMe- and / or 2'-H groups. It is noteworthy that the maximum segment sizes of 2'-F- and 2'-OMe-modifications can be different. Thus, in one embodiment, the oligonucleotide comprises a larger 2'F-modified nucleotide segment. In one embodiment, the oligonucleotide comprises a larger 2'OMe-modified nucleotide segment. In one embodiment, a shorter 2'-OMe-modified segment is acceptable compared to a 2'-F-modification. In one embodiment, a longer 2'-F-modified segment is acceptable compared to a 2'-OMe-modification.

[0170] The oligonucleotides of the present invention may comprise a number of "continuous segments" or "uniform sections" of a certain length. In one embodiment, the size or length of a "continuous segment" or "uniform section" is 2, 3, 4, 5 or 6 nucleotides. In one embodiment, the size or length of a "continuous segment" or "uniform section" does not exceed 2, 3, 4, 5 or 6 nucleotides. In one embodiment, the oligonucleotide comprises no more than 2, 3, 4, 5 or 6 consecutive nucleotides comprising a 2'-F modification. In one embodiment, the oligonucleotide comprises no more than 2, 3, 4, 5 or 6 consecutive nucleotides comprising a 2'-OMe modification. In one embodiment, one or more uniform sections are interrupted. They may be interrupted by any other chemical modification (e.g., DNA, RNA, 2'-F, 2'-OMe, 2'-MOE, LNA, etc.). In one embodiment, one or more uniform sections of 2'-F modified nucleotides are interrupted, preferably by 2'-OMe modified nucleotides. In one embodiment, one or more uniform stretches of 2'-OMe modified nucleotides are interrupted, preferably by 2'-F modified nucleotides. In some embodiments, these stretches are interrupted by DNA.

[0171] According to the present invention, the oligonucleotide does not contain a segment of more than 6 consecutive 2'-OMe modified nucleotides. According to the present invention, the oligonucleotide does not contain a segment of more than 6 consecutive 2'-F modified nucleotides. In one embodiment, the oligonucleotide does not contain a segment of more than 5, 4 or 3 consecutive 2'-OMe modified nucleotides. In one embodiment, the oligonucleotide does not contain a segment of more than 4 consecutive 2'-OMe modified nucleotides.

[0172] Bond g can be unmodified or modified. In one embodiment, bond g is a phosphate (PO) bond. In one embodiment, bond g is a 3',5'-phosphodiester bond. In one embodiment, bond g is a PS bond. In one embodiment, the oligonucleotide comprises: (a) at least 10 consecutive internucleoside bond modifications; and / or (b) 3 consecutive internucleoside bond modifications at each terminus. In one embodiment, the oligonucleotide comprises at least 10 consecutive internucleoside bond modifications. In one embodiment, the oligonucleotide comprises 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or more consecutive PS bonds. In some embodiments, each terminus comprises 3 consecutive internucleoside bond modifications. In some embodiments, each terminus comprises 4, 5 or 6 consecutive internucleoside bond modifications. In some embodiments, each terminus comprises no more than 8, 7, 6, 5, 4 or 3 consecutive internucleoside bond modifications. In a preferred embodiment, each end comprises 3 continuous internucleoside key modifications.In some embodiments, be modified to 3 '-methylene phosphonate, 5 '-methylene phosphonate, 3 '-phosphoramidate, 2 '-5 ' phosphodiester or phosphoguanidine (PN) and modify.In one embodiment, internucleoside key is modified to PS key and modify.In another embodiment, internucleoside key is modified to 3 '-3 ' phosphate bond or 5 '-5 ' phosphate bond (3 '-P-3 ' and 5 '-P-5 ').

[0173] Without being bound by any theory, the inventors believe that the ideal asymmetry for each target may depend on the length of the particular oligonucleotide and its specific potential sequence. ADARs are known to function in the form of asymmetric dimers with footprints of up to 50 bp. Although some substrates are more efficiently edited by using the deaminase domain alone than by using the full-length protein, the opposite is true for other substrates. This suggests that, depending on the size of the target / drug RNA helix, ADARs may bind in different ways (monomers compared to dimers) and register conformations (with no, one, two or up to six dsRBDs). This leads to a situation in which, depending on the length of the ASO, a specific symmetry and a specific modification pattern (e.g., ribose and bond) on the target adenosine are preferred. In order for the deaminase to achieve optimal binding, a shorter 3' end seems to be sufficient (at least 4 nt next to the CBT). Extension of the 3' end may even lead to the loss of target editing. On the other hand, the 5' end can provide binding space for the dsRBD, so more nucleotides (at least 16 nt) are generally required. Table A lists some effective symmetry embodiments identified by the inventors of the present application.

[0174] Therefore, in some embodiments, the ASO can be asymmetric. That is, the 3' end and the 5' end of the oligonucleotide can have different numbers of nucleotides (nt). For example, the 5' end can contain 20 nt to 40 nt, while the 3' end can contain 5 nt to 15 nt. In one embodiment, a) the 3' direction of the CBT has at least 4 nucleotides; or b) the 5' direction of the CBT has at least 16 nucleotides.

[0175] In some embodiments, the 3' direction of CBT has 4nt-30nt. In some embodiments, the 3' direction of CBT has no more than 10nt. In some embodiments, in the 3' direction of CBT, the length of the 3' end is shortened to 5nt. In some embodiments, in the 3' direction of CBT, the length of the 3' end is shortened to 4nt. In one embodiment, the region adjacent to the 3' end of CBT contains 4nt, 5nt or 6nt. In some embodiments, the 5' direction of CBT contains 4nt-30nt. In one embodiment, the 5' direction of CBT does not exceed 35nt. In one embodiment, in the 5' direction of CBT, the length of the 5' end is shortened to 25nt or 26nt. In one embodiment, the region adjacent to the 5' end of CBT contains 24nt, 25nt or 26nt.

[0176] According to the present invention, oligonucleotides can have the symmetry and length shown in Table A below. Oligonucleotides have the following scheme: (5' end length)-(CBT length)-(3' end length). For example, the ASO of the present invention is 32nt long, wherein the 5' end is 24nt long, the CBT is 3nt long, and the 3' end is 5nt long (scheme "24-3-5"). Therefore, in one embodiment, the oligonucleotide has any of the symmetries listed in Table A.

[0177] Table A: Preferred asymmetries for some ASO designs according to the present invention.

[0178]

[0179] According to the length and / or minimum bond modification degree of a specific oligonucleotide, it can be modified differentially, as described herein. It is known that DNA prefers a sugar folding mode different from RNA, and 2'-modified RNA forms a preferred B-type helix. In addition, DNA has a fairly strong hydrophobicity, which can change the hydrazine effect of the duplex. Therefore, DNA is only accepted in some positions and is less susceptible to being accepted in larger sections. Therefore, the oligonucleotide provided herein can have different ratios and contents of DNA and / or RNA. In some embodiments, the oligonucleotide comprises a combination of RNA and DNA. In some embodiments, the oligonucleotide comprises a combination of RNA and DNA outside a CBT.

[0180] In addition, the oligonucleotides of the present invention may contain a limited amount of DNA outside the CBT. In one embodiment, the DNA is located outside the CBT. In one embodiment, the DNA is located in the 3' direction and / or 5' direction of the CBT. In one embodiment, the DNA is located in the 3' direction of the CBT. In one embodiment, the DNA is located in the 5' direction of the CBT. The DNA content of shorter oligonucleotides (≤45nt) may be lower than the DNA content of longer oligonucleotides (≥50nt). In one embodiment, the length of the oligonucleotide is 45nt. In one embodiment, the length of the oligonucleotide is 45nt or shorter nucleotides, and no more than 3 nucleotides outside the CBT are deoxynucleotides. In one embodiment, no more than 1, 2, 3 or 4 nucleotides outside the CBT are deoxyribonucleotides. In one embodiment, the length of the nucleotides is 28-60, 28-55 or 28-45 nucleotides. In one embodiment, the length of the nucleotides is 28-60, and the deoxyribonucleoside content outside the CBT is 10% to 40%, more preferably 11% to 30%, and even more preferably 13% to 25%. In one embodiment, the oligonucleotide does not comprise any unmodified RNA nucleotides.

[0181] In one embodiment, at least one of the three nucleotides of the CBT is chemically modified at the 2' position of the sugar moiety, wherein the modification is a 2'-F-modification.

[0182] Loop-hairpin structure oligonucleotides have been previously described (WO2020 / 001793) and have been successfully used to control ADAR using chemically modified oligonucleotides. However, they are relatively large and are not bound by any theory. The inventors believe that a more rational ASO design can form a substrate duplex that can also be quickly and well recognized by endogenous ADAR, so that the larger recruitment motif can be omitted. This is an obvious advantage for delivery and manufacturing because shorter ASOs can be designed. Therefore, the oligonucleotide may or may not contain a recruitment motif for the deaminase. In contrast, the chemically modified nucleic acids of the present invention form RNA duplexes, to which the ADAR enzyme can bind, thereby improving editing efficiency. In one embodiment, the oligonucleotide does not contain an ADAR recruitment motif of a loop-hairpin structure.

[0183] In some cases, the ASO targeting domain or the core base relative to the target core base to be edited includes one or more wobble bases to compensate for the variability of the target sequence. That is, the wobble base has lower requirements for base pairing (for example, GU, IA, GA, IU, IC, etc.), which enables ASO to be paired with more than one target nucleic acid. Therefore, in some embodiments, mispairing and / or wobble enable targeting different target nucleic acids. In one embodiment, the oligonucleotide includes one or more additional mispairing, wobble bases and / or projections. In some embodiments, the oligonucleotide of the present invention can include projections of 1,2,3 or more nucleotides. In one embodiment, the oligonucleotide includes one or more mispairing, wobble bases and / or projections relative to its target, and / or the mispairing at NO. In one embodiment, the oligonucleotide includes one or more mispairing, wobble bases and / or projections relative to its target. In one embodiment, the oligonucleotide includes mispairing at NO.

[0184] The targeting sequence of the artificial nucleic acid typically comprises a nucleic acid sequence that is complementary or at least partially complementary to a nucleic acid sequence in the target RNA. In some embodiments, the targeting sequence comprises a nucleic acid sequence that is complementary or at least 60%, 70%, 80%, 90%, 95%, or 99% complementary to a nucleic acid sequence in the target RNA.

[0185] Although oligonucleotides can contain DNA and / or RNA, they can also contain other modifications. LNA improves the binding ability of ASOs by keeping nucleosides in a preferred sugar configuration (entropic favor). However, this pre-organization of sugars by additional bridges also reduces flexibility. Double-stranded RNA (dsRNA) structures are strongly interfered with in the ADAR active site (flip-out mechanism). LNA may interfere with this process, so it is desirable not to place any LNA inside or too close to the CBT. In one embodiment, the oligonucleotide comprises one or more LNAs. In one embodiment, the oligonucleotide comprises DNA and / or RNA and / or LNA. In one embodiment, the oligonucleotide comprises DNA, RNA and LNA. In one embodiment, the oligonucleotide comprises DNA. In one embodiment, the oligonucleotide comprises RNA. In one embodiment, the oligonucleotide comprises one or more LNAs.

[0186] In addition to specific backbone bond modification patterns and modifications at the 2' position of the sugar moiety, purines and / or pyrimidines of the oligonucleotide can also be specifically targeted. Purines and / or pyrimidines can be modified or unmodified. In one embodiment, purines and / or pyrimidines are modified. In some embodiments, the nucleobase is a substituted purine base residue. In some embodiments, the nucleobase is a substituted pyrimidine base residue. In one embodiment, the purine is modified with 2'-OMe, 2'-F, or 2'-deoxy. In one embodiment, the pyrimidine is modified with 2'-OMe, 2'-F, or 2'-deoxy. In some embodiments, the nucleobase is a substituted heterocyclic base analog. In some embodiments, the heterocyclic base analog is a nitrogen (N), oxygen (O), sulfur (S), or boron (B) heterocyclic base analog. In some embodiments, the modification includes base Z and / or its analog.

[0187] The oligonucleotide of the present invention can be modified at its 5' end and / or 3' end. The oligonucleotide of the present invention can include one or more different joints, labels or coupling agents at one end or both ends. For example, the oligonucleotide can include an amino joint, preferably a C6 amino joint. Therefore, in some embodiments, the oligonucleotide of the present invention includes a C6 amino joint at the 5' end. In some embodiments, the oligonucleotide includes a C6 amino joint at the 3' end. The oligonucleotide can include a portion that enhances cell uptake of the oligonucleotide, such as N-acetylgalactosamine (GalNAc). Therefore, in some embodiments, the chemically modified oligonucleotide includes a portion that enhances cell uptake of the oligonucleotide or is conjugated to a portion that enhances cell uptake of the oligonucleotide. Preferably, the portion that enhances cellular uptake is triantennary N-acetylgalactosamine (GalNAc3), which is preferably conjugated to the 3' end or 5' end of the oligonucleotide.

[0188] Composition

[0189] The nucleic acids or oligonucleotides (or ASOs) provided herein can be incorporated into compositions. For example, the use of bi- or tri-antennary N-acetylgalactosamine (GalNAc) conjugates for targeted delivery of oligonucleotides to liver hepatocytes has been previously described for, for example, the treatment of liver diseases including hepatitis B virus (HBV), non-alcoholic fatty liver disease, and genetic diseases (Debacker et al., 2020).

[0190] Therefore, a composition comprising the oligonucleotide of the present invention is provided herein. In some embodiments, the disclosure provides an oligonucleotide composition comprising the oligonucleotide described herein. In some embodiments, the composition is a pharmaceutical composition. As used herein, pharmaceutical composition refers to a mixture of materials suitable for use to an individual. For example, the pharmaceutical composition can comprise one or more active agents (e.g., oligonucleotide) and a sterile aqueous solution. In one embodiment, the composition comprises one or more oligonucleotides of the present invention.

[0191] The pharmaceutical compositions provided herein can be in any form that can be administered to a subject. The compositions can be used in methods for treating and / or preventing genetic disorders, conditions, or diseases. In specific embodiments, the pharmaceutical compositions are suitable for veterinary and / or human administration.

[0192] Provided herein is a pharmaceutical composition comprising an oligonucleotide of the present invention or a pharmaceutically acceptable salt thereof.

[0193] In one embodiment, the composition comprises a mixture of an oligonucleotide of the present invention and a pharmaceutically acceptable carrier. In some embodiments, the pharmaceutically acceptable carrier may be simply a saline solution. It may be isotonic or hypotonic.

[0194] In some embodiments, the pharmaceutical composition may further comprise one or more other therapeutics in addition to the oligonucleotides of the invention.

[0195] In certain embodiments, the compositions of the present invention further comprise diluents of varying buffer content (e.g., Tris-HCl, acetate, phosphate), pH, and ionic strength, as well as additives such as detergents and solubilizers (e.g., Tween 80, polysorbate 80), antioxidants (e.g., ascorbic acid, sodium metabisulfite), preservatives (e.g., thimerosal, benzyl alcohol), and fillers (e.g., lactose, mannitol). In some embodiments, the material can be incorporated into a granular formulation of a polymer compound (e.g., polylactic acid, polyglycolic acid, etc.) or into liposomes. In some embodiments, hyaluronic acid can also be used. Such compositions can affect the physical state, stability, in vivo release rate, and / or in vivo clearance rate of the ASO and its derivatives of the present invention. In some embodiments, the composition can be prepared in liquid form, or can be prepared in a dry powder, such as a lyophilized form.

[0196] In certain embodiments, the pharmaceutical compositions described herein further comprise one or more salts, such as sodium chloride, calcium chloride, sodium phosphate, sodium glutamate, and aluminum salts (e.g., aluminum hydroxide, aluminum phosphate, alum (potassium aluminum sulfate), or mixtures of these aluminum salts). In other embodiments, the pharmaceutical compositions described herein do not comprise a salt.

[0197] The pharmaceutical compositions described herein can be included in a container, pack, or dispenser together with instructions for administration.

[0198] Preventive and therapeutic uses

[0199] The present invention generally describes the use of chemically modified oligonucleotides and / or compositions comprising the same in a medical setting. Specifically, they are used to edit a target RNA at a fixed site, for example, by binding to the target RNA via a targeting sequence and recruiting a deaminase to the target site. The present invention describes chemically modified oligonucleotides and / or compositions for treating or preventing a genetic disorder, condition, or disease, as well as methods for treating or preventing a genetic disorder, condition, or disease. Fixed-site editing can be performed in vitro, in vivo, or ex vivo.

[0200] Chemically modified oligonucleotides or compositions comprising the same can be used to treat and / or prevent medical conditions. In one aspect, provided herein are uses of oligonucleotides of the invention and / or compositions comprising the same for treating or preventing genetic disorders, conditions or diseases. In one embodiment, the genetic disorder, condition or disease is selected from the group consisting of retinitis pigmentosa (RP), Stargardt macular degeneration, age-related macular degeneration (AMD), choroideremia, cone-rod dystrophy, cystic fibrosis (CF), Hurler syndrome, alpha-1-antitrypsin (A1AT) deficiency, Parkinson's disease, Alzheimer's disease, albinism, amyotrophic lateral sclerosis, asthma, beta-thalassemia, Cadasil syndrome, Charcot-Marie Tooth disease, chronic obstructive pulmonary disease (COPD), distal spinal muscular atrophy (DSMA), Duchenne / Becker muscular dystrophy, dystrophic epidermolysis bullosa, epidermolysis bullosa, Fabry disease, Factor V Leiden mutation (Factor V Leiden mutation), and fibrosis. VLeiden-related diseases, familial adenomatous polyposis, galactosemia, Gaucher disease, glucose-6-phosphate dehydrogenase, hemophilia, hereditary hemochromatosis, Hunter syndrome, Huntington disease, inflammatory bowel disease (IBD), hereditary polyagglutination syndromes, Leber congenital amaurosis (LCA), Lesch-Nyhan syndrome, Lynch syndrome, Marfan syndrome, mucopolysaccharidoses, muscular dystrophy, myotonic dystrophy types I and II, neurofibromatosis, Niemann-Pick disease types A, B, and C, NY-eso1-related cancers, Peutz-Jeghers syndrome, phenylketonuria, Pompe disease, primary ciliary disorders, and prothrombin mutation-related diseases.

[0201] In one embodiment, the hereditary disorder, condition or disease is associated with a point mutation. For example, the SERPINA1 gene encodes a serine protease inhibitor α-I antitrypsin (A1AT). A1AT protects tissues from certain inflammatory enzymes (including neutrophil elastase). A1AT deficiency (α-I antitrypsin deficiency, A1AD) can cause neutrophil elastase to excessively decompose elastin in the lungs. This may lead to reduced lung elasticity and trigger subsequent respiratory complications, including emphysema and chronic obstructive pulmonary disease (COPD). Mutated A1AT may also accumulate in the liver, leading to cirrhosis and liver failure. Therefore, in one embodiment, the hereditary disorder, condition or disease is associated with a G to A mutation of a gene selected from the following list: SERPINA1, PDE6A, LRRK2, NLRP3 and CRB1. In one embodiment, the mutation is selected from the following list: SERPINA1 E342K, PDE6A V685M, NLRP3 Y166 and CRB1 C948. In one embodiment, the mutation is a PiZZ mutation (alpha 1 -antitrypsin deficiency).

[0202] Chemically modified oligonucleotides of the present invention or compositions comprising the same can be used to edit adenosine bases in wild-type alleles (beneficial editing). In one embodiment, such editing modulates signal transduction, such as JAK / STAT signal transduction. In one embodiment, editing introduces a STAT1 Y701C change. In one embodiment, editing modulates inflammasome signaling by introducing a mutation from NLRP3 Y166 to C.

[0203] Chemically modified oligonucleotides or (drug) compositions of the present invention can be orally administered in any oral acceptable dosage form (including but not limited to capsules, tablets, aqueous suspensions or solutions), or administered parenterally (e.g., by parenteral injection). In some embodiments, preparations suitable for parenteral administration comprise the sterile aqueous formulations of at least one embodiment of the present disclosure, which are roughly isotonic with the blood of the intended recipient. The content, dosage, and dosing regimen of the oligonucleotides to be administered or compositions can vary according to cell type, disease to be treated, target population, mode of administration (e.g., systemic administration compared to local administration), severity of the disease, and acceptable side effect levels. In some embodiments, the content of the oligonucleotides administered in the pharmaceutical composition depends on the object being treated, the weight of the object, and the mode of administration.

[0204] Various delivery systems can be used to deliver the oligonucleotide of the present invention. Oligonucleotide according to the present invention can be delivered to individuality, organ (eyes) in original state (i.e. naked and / or isolated form), or specifically delivered to cells. When using the oligonucleotide according to the present invention, the oligonucleotide is preferably dissolved in a solution compatible with the delivery method. This type of delivery can be in vivo, in vitro or ex vivo. Nanoparticles and microparticles that can be used for delivering ASO in vivo are well known in the art. Alternatively, known transfection reagents can be used to provide plasmids by transfection.

[0205] In a preferred embodiment, the oligonucleotides of the present invention are administered and delivered in an "as is" manner (also referred to as "naked"). However, the art includes a variety of methods for delivering oligonucleotides to cells, including in vitro, ex vivo, or in vivo. That is, depending on the disease, disorder, or infection to be treated, or the cell, tissue, or body part to which the oligonucleotides of the present invention need to be delivered (e.g., in the case of beneficial editing), a route of administration or method of delivery can be selected. When the oligonucleotide is not naked for delivery, examples of delivery include delivery agents or carriers, such as nanoparticles (e.g., polymer nanoparticles), liposomes, antibody-conjugated liposomes, cationic lipids, polymers, or cell-penetrating peptides.

[0206] The use of excipients or transfection reagents can help deliver each oligonucleotide or composition defined herein to cells and / or inside cells (preferably cells affected by G to A mutations or cells that will achieve "beneficial editing" as described herein). Excipients or transfection reagents that can form complexes, nanoparticles, micelles, vesicles and / or liposomes are preferred, which can deliver each oligonucleotide or composition defined herein (which is complexed or embedded in vesicles or liposomes) through the cell membrane. Many such excipients are known in the art. Suitable excipients or transfection reagents include polyethyleneimine (PEI; ExGen500 (MBIFermentas)), LipofectAMINE ... TM 2000 (Invitrogen), lipofectin TM , or derivatives thereof, and / or viral capsid proteins capable of self-assembly into particles that can deliver each component defined herein to target cells. Such excipients have been shown to effectively deliver oligonucleotides to a variety of cultured cells. They have high transfection potential and low to moderate toxicity in terms of overall cell viability.

[0207] The ASO of the present invention can be connected to a part that enhances the cell to take in the ASO. Examples of such parts include cholesterol, carbohydrates, vitamins, biotin, lipids, phospholipids, cell penetrating peptides (including but not limited to antennapedia mutations (antennapedia), TAT, transporters and positively charged amino acids, such as oligoarginine, polyarginine, oligolysine or polylysine), antigen binding domains (for example, provided by antibodies, Fab fragments or single-chain antigen binding domains (such as camelid single domain antigen binding domains)). Therefore, in some embodiments, ASO is delivered using drugs conjugated to antibodies, nanobodies, cell penetrating peptides and aptamers. In one embodiment, oligonucleotides are conjugated to antibodies (preferably Fab fragments).

[0208] In some embodiments, toxicity and therapeutic efficacy can be determined by standard pharmaceutical procedures in cell culture or experimental animals, for example, determining the LD 50 (the dose that is lethal to 50% of the population) and LD50 50 (The dose that is therapeutically effective in 50% of the population.) In some embodiments, data obtained from cell culture assays or animal studies can be used in formulating a range of dosage for use in humans.

[0209] The oligonucleotides or compositions can be used as monotherapy or in combination with other different drugs, in particular drugs suitable for the treatment or prevention of the following: retinitis pigmentosa (RP), Stargardt macular degeneration, age-related macular degeneration (AMD), cystic fibrosis (CF), Hurler syndrome, alpha-1-antitrypsin (A1AT) deficiency, Parkinson's disease, Alzheimer's disease, albinism, amyotrophic lateral sclerosis, asthma, beta-thalassemia, Cadasil syndrome, Charcot-Marie Tooth disease, chronic obstructive pulmonary disease (COPD), distal spinal muscular atrophy (DSMA), Duchenne / Becker muscular dystrophy, dystrophic epidermolysis bullosa, epidermolysis bullosa, Fabry disease, Factor V Leiden mutation (Factor V Leiden-related diseases, familial adenomatous polyposis, galactosemia, Gaucher disease, glucose-6-phosphate dehydrogenase, hemophilia, hereditary hemochromatosis, Hunter syndrome, Huntington disease, inflammatory bowel disease (IBD), hereditary polyagglutination syndromes, Leber congenital amaurosis (LCA), Lesch-Nyhan syndrome, Lynch syndrome, Marfan syndrome, mucopolysaccharidoses, muscular dystrophy, myotonic dystrophy types I and II, neurofibromatosis, Niemann-Pick disease types A, B, and C, NY-eso1-related cancers, Peutz-Jeghers syndrome, phenylketonuria, Pompe disease, primary ciliary disorders, prothrombin mutation-related diseases, and / or symptoms associated therewith. In preferred embodiments, the oligonucleotides or compositions can be used as monotherapy or in combination with other different drugs for the treatment of any retinal disease, including, for example, inherited retinal diseases such as retinitis pigmentosa (RP), choroideremia, macular degeneration, cone-rod dystrophy and / or Leber congenital amaurosis (LCA).

[0210] In some embodiments, provided oligonucleotides or compositions have surprising efficacy compared to a reference oligonucleotide or composition. In some embodiments, the change is measured by an increase in the desired mRNA and / or protein levels compared to a reference condition. In some embodiments, the change is measured by an increase or decrease in the editing efficiency mediated by the oligonucleotide or a composition comprising the oligonucleotide. In some embodiments, the change is measured by an increase in the stability of the oligonucleotide or a composition comprising the oligonucleotide.

[0211] Further provided herein is a method for targeting adenosine. Specifically, provided herein is a method for targeting wild-type adenosine for beneficial RNA editing and / or compensatory RNA editing. Provided herein is a method for targeting wild-type adenosine for beneficial editing. Provided herein is a method for targeting wild-type adenosine for compensatory editing.

[0212] Also provided herein is a method of treating a subject having a genetic disorder, condition, or disease, wherein the method comprises administering to a subject in need thereof an effective amount of a chemically modified oligonucleotide of the invention or a composition of the invention. In one embodiment, the genetic disorder, condition, or disease is associated with a G to A mutation. Treating a disorder associated with a G to A mutation can result in a reduced mortality rate in a population of treated subjects compared to a population not receiving treatment.

[0213] Provided herein are uses of the oligonucleotides of the present invention for therapy. Also provided herein are uses of the oligonucleotides of the present invention in the manufacture of a medicament for treating a condition, disorder, or disease associated with a G to A mutation. Also provided herein are uses of the oligonucleotides of the present invention in the manufacture of a medicament for treating a genetic disorder, condition, or disease associated with a G to A mutation. Also provided herein are uses of the oligonucleotides of the present invention in the manufacture of a medicament for treating a genetic disorder, condition, or disease associated with a G to A mutation.

[0214] Compositions disclosed herein can be used in a variety of ways, depending on the needs of local treatment or systemic treatment and area to be treated. Use can be by suction (for example, by atomization), intranasal, oral, by injection or infusion, intravenous, subcutaneous, intradermal, intracranial, intramuscular, intratracheal, intraperitoneal, intrarectal, by direct injection to tumor etc. carry out. Application form can be solid, powder, pill or any other form that is suitable for human medicine use. In some embodiments, oligonucleotide construct can be systemically delivered.

[0215] Patient population

[0216] The oligonucleotides of the present invention or compositions comprising the oligonucleotides can be administered to different groups of subjects or patients. In certain embodiments, the patient is in need of treatment. In other embodiments, the patient does not need treatment ("beneficial editing"), i.e., the subject receives the oligonucleotide or composition to edit RNA derived from the wild-type allele (rather than the mutant allele) to, for example, modulate the function of the wild-type protein in a useful manner, thereby preventing or treating the disease.

[0217] In certain embodiments, an oligonucleotide as described herein or a composition comprising the oligonucleotide is administered to a naive subject, i.e., a subject not suffering from a disease or condition. In one embodiment, an oligonucleotide as described herein or a composition comprising the oligonucleotide is administered to a subject. In one embodiment, an oligonucleotide as described herein or a composition comprising the oligonucleotide is administered to a naive subject at risk of developing a disease or condition.

[0218] In certain embodiments, the oligonucleotides described herein or compositions comprising the oligonucleotides are administered to patients who have been diagnosed with a disease or condition. In some embodiments, the oligonucleotides described herein or compositions comprising the oligonucleotides are administered to patients before symptoms appear or worsen.

[0219] In some embodiments, the oligonucleotides described herein or compositions comprising the oligonucleotides are administered to humans. In some embodiments, the human subject to whom the oligonucleotides described herein or compositions comprising the oligonucleotides are administered is any individual at risk for developing a disease or condition associated with a G to A mutation in a gene. In one embodiment, the patient suffers from a disease or condition associated with a G to A mutation in a gene.

[0220] In some embodiments, a subject or patient suitable for treatment of a condition, disorder, or disease associated with a G to A mutation can be identified or diagnosed by a health care professional. In some embodiments, the symptoms of a condition, disorder, or disease associated with a G to A mutation can be any condition, disorder, or disease that would benefit from an A to I transition.

[0221] Also provided herein is a method of treating a condition, disorder, or disease associated with a G to A mutation in a subject. In some embodiments, the methods of the present disclosure can be used to treat a condition, disorder, or disease associated with a G to A mutation in a subject, wherein the method comprises administering to the subject a therapeutically effective amount of an oligonucleotide or a pharmaceutical composition thereof.

[0222] Also provided herein are uses of the oligonucleotides of the present invention for therapy. Furthermore, provided herein are uses of the oligonucleotides of the present invention in the manufacture of a medicament for treating a condition, disease, and / or condition associated with a G to A mutation in a subject. Also provided herein are uses of the oligonucleotides of the present invention in the manufacture of a medicament for treating a condition, disease, and / or condition associated with a G to A mutation. In certain embodiments, the oligonucleotides of the present invention are used in the manufacture of a medicament for treating a condition, disease, and / or condition associated with a G to A mutation.

[0223] The composition of the present invention comprises the oligonucleotide of the present invention. According to another aspect, the present invention relates to a kit or a multi-component kit comprising the oligonucleotide of the present invention and / or the (pharmaceutical) composition of the present invention. The kit further comprises instructions for use.

[0224] Editing Method

[0225] The present invention also relates to methods for editing a target adenosine in a target nucleic acid. For example, the present invention provides methods for editing a SERPINA1 polynucleotide (e.g., a SERPINA1 polynucleotide comprising a single nucleotide polymorphism (SNP) associated with α-I-antitrypsin deficiency). In addition, the present invention relates to in vitro methods for editing a target adenosine in a target nucleic acid, as well as in vitro methods for deaminating at least one specific adenosine present in a target RNA sequence in a cell.

[0226] In one aspect, provided herein is an in vitro method for editing a target adenosine in a target nucleic acid, wherein the method comprises contacting the target nucleic acid with an oligonucleotide of the present invention.

[0227] In another aspect, provided herein is an in vitro method for deaminating at least one specific adenosine present in a target RNA sequence in a cell, wherein the method comprises the following steps:

[0228] (a) contacting a target nucleic acid with a chemically modified oligonucleotide of the present invention;

[0229] (b) allowing cells to take up the chemically modified oligonucleotide;

[0230] (c) annealing the chemically modified oligonucleotide to the target RNA sequence; and

[0231] (d) A mammalian ADAR enzyme containing the native dsRNA binding domain found in the wild-type enzyme is caused to deaminize a target adenosine to inosine in a target RNA sequence.

[0232] In one embodiment, the method comprises the step of identifying the presence of inosine in the RNA sequence after step (d).

[0233] The editing reaction is best monitored or controlled by sequence analysis of the target RNA.

[0234] Furthermore, the chemically modified oligonucleotides or (pharmaceutical) compositions of the present invention can be used to diagnose genetic disorders, diseases or conditions. The diseases or conditions are preferably selected from infectious diseases, neoplastic diseases, cardiovascular diseases, autoimmune diseases, allergies, and neurological diseases or conditions. In one embodiment, the genetic disorder, condition or disease is associated with a G to A mutation.

[0235] The present invention is used to make desired changes to a target sequence in a cell or subject by performing site-directed editing of nucleotides using oligonucleotides that are capable of affecting adenosine deaminase acting on RNA (ADAR)-mediated adenosine to inosine. As a result, the target sequence is edited through an adenosine deamination reaction mediated by ADAR, thereby converting adenosine to inosine. In some embodiments, since I is recognized as G, correcting the deamination of the pathogenic mutation in the SERPINA1 gene can reverse the E342K mutation back to the wild type, thereby reversing or alleviating the A1AD-related symptoms experienced by the patient.

[0236] The method of the present invention can be used for the cell from any organ (such as skin, lung, heart, kidney, liver, pancreas, intestinal tract, muscle, gland, eye, brain, blood etc.).The present invention is particularly suitable for modifying the sequence in the cell, tissue or organ related to the disease state of (human) object.For example, such cell can include but not limited to hepatocyte, hepatocyte-like cell and / or alveolar type II cell, neuron (PNS, CNS), retina, photoreceptor cell, Muller glial cell, RPE, immune cell, B cell, T cell, dendritic cell, macrophage.

[0237] The present invention will be described in more detail with reference to the following figures and examples.

[0238] Example

[0239] The embodiments shown below are merely illustrative and will describe the present invention in another way.These embodiments should not be construed as limiting the present invention thereto.Sequences disclosed herein are also shown in the accompanying sequence table.Yet the sequence table only shows the nucleotide sequence, and the modification of the key between Nucleotide and Nucleotide is not shown in the sequence table.The relevant modifications associated with the sequence are disclosed in the following table and are disclosed in the accompanying drawings of the application to a certain extent.

[0240] For all experiments, editing efficiency was expressed as the percentage [%] of edited target sites found among all detected target sites in the target transcript.

[0241] Example 1. Editing efficiency and lysosomal stability of SERPINA1 E342K-targeting oligonucleotides with increasing DNA to RNA (DNA:RNA) ratio.

[0242] To determine the effects of DNA and RNA on the lysosomal stability of various oligonucleotides, different versions of the SERPINA1 E342K targeting construct (v117.26 to v117.58) were generated and tested in vitro. The DNA and RNA content of each ASO varied and are listed in Table 1. The results of Example 1 are shown in Table 1. Figure 1 v117.26 was used as a control.

[0243] To evaluate cell efficiency, 2.5 × 10 4 HeLa cells (Catalog No.: ATCC CCL-2) were seeded in 24-well plates. After 24 h, the cells were forward transfected with a plasmid containing the SERPINA1 E342K mutant cDNA. 300 ng of plasmid and 0.9 μl 6 (Promega (Promega)) were each diluted in 50μl Opti-MEM and incubated for 5 minutes, then combined and incubated for another 20 minutes. The culture medium was replaced, and the transfection mixture was evenly distributed into one well. 24h after plasmid transfection, the cells were forward transfected with 5pmol construct / well and 1.5μl / well Lipofectamine RNAiMAX reagent (ThermoFisher Scientific)). After 24h, the culture medium was replaced. 48h after transfection, the cells were harvested for RNA isolation and sequencing. Lysosomal degradation assay was performed in triton lysosomes of mixed-sex rat liver (tebu-bio, cat. no. 098R0610.LT). The triton lysosome solution was diluted to an acid phosphatase concentration of 0.1115U / ml with 20mM sodium citrate solution (pH=5.0). In the experiment, 15 pmol of the corresponding ASO was diluted in 0.1115 U / ml acid phosphatase Triton lysosome solution. The mock sample contained only 15 pmol of ASO diluted in PBS. All samples were incubated at 37 ° C for a given time point, then frozen with liquid nitrogen and immediately stored at -80 ° C. The degradation of oligonucleotides was observed by denaturing urea PAGE. By adding 7 μl of RNA loading dye ( Sequencing gel buffer concentrate The samples were denatured by diluting 1:10 in sequencing gel diluent, Carl Roth) and then incubated at 70°C for 2 minutes. The denatured samples were then loaded onto urea (7M) polyacrylamide (15%) electrophoresis (PAGE) gels and run at 1200V for 4 to 6 hours in 1× TBE (Tris-borate-EDTA) buffer. SYBR TM Gold nucleic acid gel stain (Thermo Fisher Scientific) was used to visualize the bands, and the fluorescence images were analyzed using a Fujifilm FLA-5100 fluorescence image analyzer at an excitation wavelength of λ ex = Scanning at 473 nm. Half-life was quantified by combining the grayscale value of each individual full-length ASO band with ImageJ. Construct half-life was calculated relative to mock samples.

[0244] Table 1: Sequence of the SERPINA1 E342K targeting construct containing the nucleobase and backbone modifications used in Example 1. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond; GalNAc = triantennary N-acetylgalactosamine.

[0245]

[0246]

[0247]

[0248] like Figure 1 As shown in A, incorporation of increasing amounts of DNA into the oligonucleotide backbone, thereby increasing the DNA to RNA (DNA:RNA) ratio, resulted in an overall decrease in the editing yield of SERPINA1 E342K. Two control constructs were used: v117.26 and v117.39 (5.1% total DNA). These ASOs differed from each other, with v117.39 differing from v117.26 in that v117.39 contained additional phosphorothioate bonds at positions d, e, and f (Table 1). The inclusion of additional PS bonds in CBT alone was able to increase the lysosomal half-life (t 50 ) increased from 2h (v117.26) to 5h (v117.39). However, from a therapeutic perspective, an even longer lysosomal half-life would be more preferred. Compared to the v117.26 control construct that does not carry CBT external DNA, the ASO v117.42 (39% DNA), in which all remaining RNA nucleosides are replaced by DNA, showed a lower editing yield. Interestingly, when compared to constructs with smaller DNA: RNA ratios (e.g., v117.43 (23.7% DNA) vs. v117.46 (22% DNA)), constructs containing higher DNA: RNA ratios in their backbones (e.g., v117.53 (33.9% DNA) vs. v117.58 (30.5%)) showed lower RNA editing levels.

[0249] Furthermore, incorporating increased amounts of DNA into the ASO backbone, thereby increasing the DNA:RNA ratio of the ASO, leads to greater lysosomal stability ( Figure 1 B). This stability peaked when all remaining RNA nucleosides were replaced with DNA, with ASO v117.42 (control DNA backbone; 39% DNA) exhibiting the highest degree of stability with a half-life exceeding 31 days (t 50>d31). Constructs v117.49 and v117.53 were able to tolerate 10% fetal bovine serum (FBS) and showed a high level of lysosomal stability until approximately day 7 (7d), with a half-life (t 50 On the other hand, v117.39, which lacks chemical modifications on many 2'-hydroxyl groups of the nucleoside sugar moiety, was rapidly degraded at 6 h (t 50 =5h). Although compared with v117.39, the two ASOs have a better stability against degradation (t 50 ) were improved, but the editing efficiency decreased to 39.6% and 30.9% respectively (see Figure 1 A and B).

[0250] In general, these data show that increasing the amount of DNA outside CBT (DNA: RNA ratio) in oligonucleotides carrying 2'-OMe- and 2'-F- modifications at +2 and +3 positions, respectively (wherein h and i are not PS bonds, and wherein no more than 6 consecutive nucleotides have the same 2'-modification) tends to reduce editing efficiency. At the same time, increasing DNA: RNA ratio improves lysosomal stability, and thereby improves the persistence of serum-resistant ribonuclease degradation, thereby extending the lifespan of different constructs in serum. However, very high degrees of DNA substitution interfere with editing efficiency. Nevertheless, potentially useful embodiments can be generated that contain a significant number of 2'-unmodified ribonucleotides that have a reasonable balance of stability and editing efficiency and also have relatively low levels of (potentially toxic) 2'-F (e.g., 20% in v117.53) and 2'-O-methyl (e.g., about 35% in v117.49), although 2'-F and 2'-O-methyl are generally required to some extent to stabilize the ASO (e.g., at least 10% of all nucleosides are 2'-F modified and at least 10% are 2'-O-methyl modified). However, for many applications, further stabilization is required. Therefore, without being bound by any particular theory, the inventors believe that there must be an optimal balance between the mixture of 2'-F, 2'-OMe-nucleoside modifications, PS bond modifications and the target editing efficiency and lysosomal stability of the oligonucleotide, which can be achieved by replacing RNA nucleosides with DNA nucleosides.

[0251] Example 2. Editing efficiency of SERPINA1 E342K targeting oligonucleotides obtained by mixing 2'-F-, 2'-OMe-modifications and DNA

[0252] To more thoroughly investigate the stabilizing effects of 2'-F- and 2'-OMe backbone modifications relative to and in combination with DNA content, different oligonucleotides were generated and tested for their in vitro RNA editing efficiency according to the protocol described in Example 1. The different ASO constructs tested are listed in Table 2. The results are presented in Figure 2 As described in Example 1, v117.26 and v117.39 were used as controls.

[0253] like Figure 2 As shown in Figure A, high levels of 2'-F modifications are associated with reduced editing efficiency. Oligonucleotides containing high levels of DNA content (v117.42; 29.3%, containing 39% DNA) or 2'-F-modified nucleosides (v117.62; 7.2%, containing 58% 2'-F) showed a sharp drop in editing efficiency. On the other hand, constructs carrying more balanced amounts of 2'-F-modifications and DNA showed RNA editing efficiencies between approximately 37.5% (v117.60) and 58.9% (v117.59), comparable to the control. These data confirm that a balanced mix of 2'-F-, 2'-OMe-, and 2'-H-modifications can be used to compensate for substitutions with 2'-OH modifications, and that excess amounts of a single type of modification, i.e., 2'-F (v117.62) or 2'-H (v117.42), have an adverse effect on overall RNA editing. A preference for more 2'-F than DNA can also be observed. Embodiments containing higher amounts of 2'-F generally performed better than those containing more DNA, such as v117.65 (55.4% edited, containing 42% 2'-F and 20% DNA) versus v117.60 (37.5% edited, containing 32% 2'-F and 31% DNA). In addition, despite containing 39% DNA and only 24% 2'-F, v117.42 still achieved an editing efficiency of 29.3%, which may be sufficient when combined with its desired high lysosomal stability (see Example 1). The amount of 2'-OMe for all embodiments remained between 32% and 37% (e.g., v117.99 and v117.59, respectively).

[0254] To further determine whether the 2'-sugar modifications at the ends of the CBT and the terminal segments are position-dependent or position-independent of the nature of the nucleotide (purines vs. pyrimidines), the RNA editing efficiency of ASOs v117.99 and v117.100, both of which contain the same amount of total 2'-F-, 2'-OMe-, and 2'-H-modifications (25.4% DNA, 42.4% 2'-F, and 32.2% 2'-OMe), was evaluated. +2 and N +3The positions of all 2'-sugar modifications (2'-F, 2'-OMe and DNA) were randomized, except for the three modifications at the positions of 2'-F, 2'-OMe and DNA. Compared with all other embodiments, the linker backbone modifications were not changed. The results were then compared with the parent reference sequence v117.59. Figure 2 As shown in A, v117.99 and 177.100 showed good editing (39.2% and 48.1%, respectively), indicating that RNA editing efficiency is not significantly affected by the relative position of the 2'-modification.

[0255] Table 2: SERPINA1 E342K targeting construct sequences and modifications used in Example 2. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond; GalNAc = triantennary N-acetylgalactosamine.

[0256]

[0257]

[0258]

[0259] In summary, the data show that high levels of only one type of 2'-modification (e.g., high levels of 2'-F- or 2'-H modifications) have a deleterious effect on the overall editing efficiency of oligonucleotides (see v117.42 and v117.62). In general, the best performing (achieving> 45% editing) longer embodiments (≥40nt, here: 59nt) contain more than 35% 2'-F modifications and no more than 25% DNA content, indicating that these may be preferred 2' modification amounts. However, it can be observed that there is a certain tolerance for more DNA (up to about 40%) and less 2'-F modifications (down to about 25%), and may depend on other factors, such as the nucleobase sequence in the embodiment. In addition, the data show that the total amount of certain 2'-modifications has a stronger effect on the corresponding editing yield than the precise position of each 2'-modification within the oligonucleotide.

[0260] Example 3. Editing efficiency and lysosomal stability of STAT1 Y701-targeting oligonucleotides

[0261] Similar to the evaluation of RNA editing efficiency and lysosomal stability of SERPINA1 E342K-targeting ASOs, the editing efficiency and lysosomal stability of various human STAT1Y701-targeting ASOs were determined. The sequences of the different ASO constructs and the corresponding modifications are listed in Table 3. The results are shown in Figure 3Constructs v117.28 and v117.29 (with 2'-F- and 2'-OMe-modifications but without 2'-H outside of CBT) were used as controls.

[0262] To evaluate cell efficiency, 1 × 10 cells were forward transfected with 25 pmol of construct and 1.5 μl of Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific) per 24 wells according to the Lipofectamine RNAiMAX reagent protocol. 5 HeLa cells were cultured for 24 h. RNA was isolated and used for Sanger sequencing. Lysosomal degradation assay was performed as described in Example 1.

[0263] The data showed a similar trend to that of the SERPINA1 E342K-specific oligonucleotides. Replacing the 2'-OH group (RNA) with a 2'-H group (DNA) to increase the DNA:RNA ratio resulted in a general decrease in editing efficiency to 2% and 2.3% for v117.30 and v117.31, respectively (containing 44% and 22% DNA, respectively) ( Figure 3 A). Similarly, containing high levels of 2'-F-modifications also resulted in low levels of RNA editing (v117.36 and v117.41, containing 41% and 63% 2'-F modifications, respectively). Similarly, v117.39 contained a relatively high 2'-H content compared to, for example, v117.83, showing an editing yield of only 11%. On the other hand, ASOs comprising a combination of 2'-F, 2'-OMe, and 2'-H-modifications showed enhanced editing efficiency. However, the performance of these embodiments was similar to that of the control only when the amount of 2'-F modification was about twice that of DNA, which was seen in construct v117.40 (comprising 49.2% 2'-F, 18.6% DNA, and 32.2% 2'-OMe). Compared to v117.37, v117.40 showed enhanced editing efficiency (35.8%) and lysosomal stability (t 50 >7d compared to t 50 =7d)( Figure 3 A and 3B).

[0264] Table 3: STAT1 Y701 targeting construct sequences and modifications used in Example 3. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond; GalNAc = triantennary N-acetylgalactosamine.

[0265]

[0266]

[0267] As previously discussed, the data indicate that optimal editing efficiency is achieved when 2'-F and 2'-OMe-modifications are combined with a 2'-H group in the ASO sugar moiety, and where the 2'-O-alkyl modification is at +2(N +2 ) position, 2'-F-modification at +3(N +3 ) position, and the ASO contained no more than 6 consecutive nucleotides with the same 2' modification. Although specific 2'-modification combinations provided the highest editing, there was a trend to prioritize nucleosides containing more 2'-F-modifications rather than DNA (v117.40 had an editing efficiency of 35.8%, while v117.39 had an editing efficiency of 11%). Compared with the SERPINA1 E342K targeting construct, there was also greater tolerance for high 2'-F content (v117.40 up to approximately 50%) and less DNA (v117.40 contained approximately 19%) in the STAT1 Y701 targeting construct.

[0268] In general, the data confirm the results observed with SERPINA1 E342K targeting ASO, which should include a certain level of 2'-F and 2'-OMe modifications, and should avoid the occurrence of continuous homogeneous sugar-modified segments. In all embodiments that no longer contain natural RNA, there are no segments of more than six continuous 2'-F, 2'-O-Me or DNA nucleosides. More importantly, these data show that specific modification patterns can be transferred to different targets (for example, STAT1 Y701 v117.37 pattern is taken from SERPINA1 E342K v117.59). That is, although general modification patterns are largely transferable, they may still require some additional target-specific adjustments to provide optimal editing efficiency for ASO.

[0269] Example 4. Editing efficiency and lysosomal stability of CRB1 C948Y targeting oligonucleotides

[0270] To further determine the combined effects of 2'-F-, 2'-OMe-, and 2'-H-modifications on editing efficiency and stability of other gene targets, different human CRB1 C948Y targeting oligonucleotides were generated and tested. The different construct sequences and their modifications are listed in Table 4. The results of Example 4 are shown in Table 4. Figure 4 The construct v117.20 without 2'-H(DNA) outside of CBT was used as a control.

[0271] 300 ng of plasmid (containing CRB1 C948Y cDNA) and 0.9 μl 6 forward transfection to 5x10 4Cellular efficiency was assessed on HeLa cells seeded in 24-well plates 24 hours prior to transfection. The culture medium was changed every 24 hours. Twenty-four hours after plasmid transfection, 25 pmol of the construct was forward transfected using 1.5 μl of Lipofectamine RNAiMAX (Thermo Fisher Scientific) according to the Lipofectamine RNAiMAX protocol. After 24 hours of culture, cells were harvested for RNA isolation and Sanger sequencing. Lysosomal degradation assays were performed as described in Example 1.

[0272] Table 4: CRB1 C948Y targeting construct sequences and modifications used in Example 4. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0273]

[0274] like Figure 4 As shown in A, although replacing some of the 2'OH groups in v117.20 with 2'H groups only resulted in a slight decrease in ASO editing efficiency from 50.8% (v117.20, containing 5.1% DNA and 22% 2'-F) to 32.5% (v117.22, containing 25.4% DNA), good editing was still maintained. Similar observations were made by replacing the 2'OH group in v117.20 with a 2'-F- modification (V117.23; 39.5%, 2'-F levels increased to 42.4%). It is worth noting that the introduction of a mixture of 2'-F-, 2'-OMe- and 2'-H modifications resulted in similar editing efficiencies (v117.24 containing 40.7% 2'-F, 22% DNA, 37.3% 2'-Ome, and v117.25 containing 44.1% 2'-F, 25.4% DNA, 30.5% 2'-OMe). However, regarding lysosomal stabilization, it was observed that the introduction of a mixture of 2'-F-, 2'-OMe-, and 2'-H modifications resulted in an overall improvement. 50 5h, while v117.24 achieves t 50 >7d, indicating that 2'-F / 2'-OMe / DNA mixtures replacing all natural RNA nucleosides promote ASO stability ( Figure 4 A).

[0275] These data again demonstrate that a well-balanced mix of 2'-F and 2'-OMe modifications (approximately 40%-45% 2'-F, 22-25% DNA, and 31% 2'-OMe) that avoids the appearance of continuous, uniformly sugar-modified segments results in efficient editing yields and lysosomal stabilization. Furthermore, as already observed in Example 3, the length of the uniform 2'-modified segments does not exceed 6 nt, or even 3 nt. Furthermore, these data again demonstrate that this combination of 2'-modifications can be applied to different oligonucleotide sequences and is therefore independent of the actual target used (e.g., the CRB1 C948Y v117.24 pattern was taken from SERPINA1 E342K v117.59).

[0276] Example 5. Editing efficiency and lysosomal stability of LRRK2 G2019S targeting oligonucleotides

[0277] Several mutations in leucine-rich repeat kinase-2 (LRRK2) are associated with Parkinson's disease (PD), among which G2019S is a very prominent mutation. To determine the effects of 2'-F, 2'-OMe-, and 2'-H-backbone modifications on LRRK2 G2019S targeting oligonucleotides, differently modified constructs were generated and tested for their in vitro editing efficiency and lysosomal stability. The different human LRRK2 G2019S targeting constructs used in Example 5 are listed in Table 5. The results are shown in Table 5. Figure 5 Shown in.

[0278] To evaluate cell efficiency, 2.5 × 10 4 HeLa cells were seeded in 24-well plates. After 24 h, 300 ng of cDNA containing human LRRK2G2019S mutation and 0.9 μl Cells were forward transfected with plasmids containing 6 constructs. Twenty-four hours after plasmid transfection, 25 pmol of each construct was forward transfected with 1.5 μl of Lipofectamine RNAiMAX reagent. After 24 hours of culture, cells were harvested for RNA isolation and Sanger sequencing. Lysosomal degradation assays were performed as described in Example 1.

[0279] Table 5: LRRK2 G2019S targeting construct sequences and modifications. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0280]

[0281] like Figure 5As shown in A, in v117.42, the backbone modification only consists of replacing the 2'-OH group in v117.20 with a 2'-H group (49.2% DNA, 16.9% 2'-F, 33.9% 2'-OMe), resulting in a significant drop in editing efficiency (less than 10% editing yield). However, oligonucleotides carrying a combination of 2'-F-, 2'-OMe- and 2'-H modifications are able to restore their editing efficiency to that of the v117.20 construct. For example, v117.59, which carries a mixture of 42.4% 2'-F, 30.5% 2-OMe and 27.1% 2'-H modifications, has an editing efficiency of about 40%, which is similar to v117.20. In fact, in addition to showing normal editing efficiency, v117.59 also shows enhanced lysosomal stability (t 50 >7 days)( Figure 5 B). Similarly, v117.60, which contains a mixture of 40.7% 2'-F, 37.3% 2-OMe, and 22% 2'-H modifications outside of CBT, not only showed stable editing efficiency but also experienced enhanced lysosomal stability when compared to v117.20 (t 50 >7 days)( Figure 5 B).

[0282] These data confirm the observations made for the SERPINA1 E342K (Example 2), STAT1 Y701 (Example 3), and CRB1 C948Y (Example 4) targeting constructs, that a certain level of 2'-F, 2'-OMe, and 2'-H modifications should be included in the oligonucleotide backbone to provide good editing efficiency and lysosomal stability. As already observed in Example 3, the length of the uniform 2'-modified segment does not exceed 6nt, or even 4nt. As previously observed in Example 2, embodiments containing a combination of high levels of DNA (nearly 50% for LRRK2 G2019S v117.42) with lower levels of 2'-F (only more than 15% for LRRK2 G2019S v117.42) significantly enhanced lysosomal stability, but showed a compromise in editing efficiency. Preferably, the LRRK2 G2019S targeting embodiment contains about 40% 2'-F, about 25% DNA, and about 35% 2'-OMe. Furthermore, these data indicate that this combination of 2′-modifications can be applied to different oligonucleotide sequences and thus used independently of the actual target (see LRRK2 G2019S v117.60 and SERPINA1 E342K v117.59).

[0283] This suggests that modification patterns can be transferred in a position-specific manner independent of the target sequence. Although the transferability between the SERPINA1E342K and LRRK2G2019S embodiments appears high, the respective embodiments may still require some additional target-specific adjustments to provide optimal editing efficiency for ASOs.

[0284] Example 6. Editing efficiency of murine PDE6A (mPDE6A) V685M targeting oligonucleotides

[0285] Mutations in the PDE6A gene can cause rod photoreceptor degeneration and are associated with the blinding disease retinitis pigmentosa (RP) (Sothilingham et al., 2015). Animal models of RP include the Pde6a V685M mutant mouse model. Therefore, to determine the effects of 2'-F, 2'-OMe, and 2'-H modifications on murine PDE6A (mPDE6A)-targeted ASOs, constructs containing such modifications were prepared and tested for in vitro editing efficiency. The different constructs used in Example 6 are listed in Table 6. The results are shown in Table 6. Figure 6 Shown in A.

[0286] Table 6: mPDE6A targeting construct sequences and modifications used in Example 6. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond.

[0287]

[0288] Construct v117.21 contained a mixture of 2'-F-, 2'-OMe and 2'-OH (RNA) (28.8% 2'F, 39% 2'OMe and 27.1% 2'OH nucleoside modifications. The results showed a stable editing efficiency of approximately 19%. Replacing the 2'-OH group with a 2-H group and / or introducing 2'-F- and 2'-OMe-modifications produced v117.27 (40.7% 2'F, 37.3% 2'OMe and 22% 2'H), which showed similar editing efficiency.

[0289] These data confirm previous observations that the combination of 2'-F and 2'-OMe modifications with 2'H groups can provide stable ASOs without negatively affecting their editing efficiency. As already observed in previous examples, the length of uniform 2'-OH modified segments does not exceed 6 nt, or even 3 nt.

[0290] Example 7. Editing efficiency of NLRP3 Y166-targeting oligonucleotides

[0291] It has been reported that the NLRP3 (nucleotide-binding domain, leucine-rich repeat family, pyrin domain-3) inflammasome plays a key role in retinal neurodegeneration and many other diseases. NLRP3 is responsible for the formation of inflammasomes, which are associated with a variety of diseases, such as inflammation, aging, cardiac and vascular diseases, metabolic syndrome, gout, autoimmune diseases, etc. Once activated, it is often a factor that prolongs the duration of the disease. The target site Y166 is a phosphorylation site required for NLRP3 activation (Bittner et al., 2021). Through beneficial RNA editing, the phosphorylation site can be blocked so that NLRP3 is no longer activated or is activated inefficiently. Therefore, the editing efficiency of human NLRP3 Y166 targeting oligonucleotides was tested using a plasmid expression system (A) and a genome-integrated oligonucleotide sequence (B). The different NLRP3 Y166 targeting construct sequences and modifications used in Example 7 are shown in Table 7. The results are shown in Figure 7 "No ASO" was used as a negative control.

[0292] Plasmid-based methods (A): 5×10 4 HeLa cells (catalog number: ATCC CCL-2) were seeded into 24-well plates and then incubated with 300 ng of wild-type NLRP3 cDNA containing plasmid and 0.9 μl 6 (Promega). 24 h after transfection, 25 pmol of the construct was transfected in the forward direction using 1.5 μl of Lipofectamine RNAiMAX reagent. After 24 h of incubation at 37°C, cells were harvested for RNA isolation and Sanger sequencing.

[0293] Genome integration method (B): 1×10 cells containing human wild-type NLRP3-mNeonGreen cDNA stably integrated into its genome via the piggyBac transposase system were 5 HeLa cells were seeded in 24-well plates. 24 hours after seeding, each well was forward transfected with 25 pmol of construct and 1.5 μl of RNAiMAX. After incubation at 37°C for 24 hours, cells were harvested for RNA isolation and Sanger sequencing.

[0294] Table 7: NLRP3 Y166 targeting construct sequences and modifications used in Example 7. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0295]

[0296] like Figure 7 As shown in A and 7B, no significant differences were observed between the plasmid-based method (A) and the genomic integration method (B). Overall, the data show that the combination of 40.7% 2'-F, 37.3% 2'-OMe, and 22% 2'-H modifications can stabilize and maintain an efficient editing yield (v117.20), and this stability and maintenance is achieved in the absence of a continuous (≤6nt) 2'-modified segment (neither 2'-F, nor 2'-O-methyl, nor DNA).

[0297] Example 8. Adjustment of the 5' and 3' end lengths of GAPDH 3'UTR targeting oligonucleotides

[0298] Depending on the type and / or sequence of a specific ASO, and without being bound by any theory, the inventors believe that there is an optimal ASO length, wherein the hybridization strength is optimal and the ADAR driving ability is at least sufficient for editing. In addition, it is assumed that the shorter the ASO, the easier it is to deliver (e.g., endosomal escape), the lower the risk of aggregation, toxicity and / or immunogenicity, and the easier it is to mass-produce and screen such ASOs. Using the artificial SNAP-ADAR method, the inventors have previously shown that guide RNA can be easily shortened to 14nt while still maintaining recognition of the substrate duplex and editing by the ADAR deaminase domain (data not shown). The latter discovery is different from the prior art, in which the recruitment of endogenous ADAR to endogenous substrates is mediated by a segment design ASO with a stereo-pure PS / PN modified skeleton and a 30-33nt length (Monian et al., 2022).

[0299] To determine the possible length range of ADAR recruitment oligonucleotides, the in vitro editing efficiency of four embodiments targeting the 3'UTR of human GAPDH was evaluated (see v121.10 to v121.13). Figure 8 The different ASO constructs and their modifications are listed in Table 8.

[0300] To evaluate cell efficiency, 1 × 10 cells were forward transfected per well of 24 cells using 25 pmol of construct and 1.5 μl of Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific) according to the Lipofectamine RNAiMAX reagent protocol. 5 HeLa cells were isolated 24 hours later for RNA sequencing.

[0301] Table 8: GAPDH 3'UTR targeting construct sequences and modifications used in Example 8. mN = 2'-O-methyl, fN = 2'-fluoro, N = 2'OH (ribose; RNA), dN = 2'H (deoxyribose; DNA), * = phosphorothioate bond.

[0302]

[0303] Overall, the dataset shows that even very short embodiments (e.g., down to at least 25 nt) can achieve efficient editing.

[0304] Example 9. Adaptation of the 5' and 3' end lengths of SERPINA1 E342K targeting oligonucleotides

[0305] To determine the importance and role of the 5'- and 3'-ends in RNA editing, several truncated forms of human SERPINA1 E342K targeting oligonucleotides were generated and their in vitro editing efficiencies were assessed (see v117.80 to v117.83). Figure 9 Various 5' and / or 3' end truncated ASO constructs and their modifications are listed in Table 9. 31 nt long oligonucleotide [Segment Design_31 nt] and 40 nt long oligonucleotide [Segment Design_40 nt] were used as negative controls.

[0306] Table 9: SERPINA1 E342K targeting construct sequences and modifications used in Example 9. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0307]

[0308]

[0309] Initially, ASO constructs with different 3' end truncations were tested for their ability to maintain RNA editing efficiency ( Figure 9 A symmetrical 59nt long ASO (v117.59) was used as the starting point and sequentially shortened from the 3' end (while maintaining the 3' end 1-3 × 2'-OMe modification), generating 3' end truncated ASOs of different lengths (v117.80 to v117.83) ( Figure 9 B). Figure 9As shown in Figure 3A), shortening the 3'-end of v117.59 to 50nt (v117.80) or 45nt (v117.81) did not change the editing yield. Although it experienced a slight decrease, construct v117.82 (40nt) maintained its ability to effectively mediate SERPINA target editing. Most importantly, when compared to v117.59, constructs v117.81 and v117.80 even showed a small increase in editing efficiency ( Figure 9 A).

[0310] However, simultaneous truncation of both the 5' and 3' ends resulted in a significant decrease in editing efficiency, approaching the detection limit of Sanger sequencing (v117.83; 31 nt) ( Figure 9 That is, while shortening the 3' end to 40 nt resulted in only a slight decrease in editing efficiency, shortening the ASO from the 3' end and especially the 5' end to 31 nt (v117.83) resulted in almost complete loss of editing ( Figure 9 A). Without being bound by any theory, the inventors believe that truncation from the 3' end may be better tolerated than truncation from the 5' end.

[0311] Next, to further determine the importance of the 5' and 3' ends in regulating editing efficiency, additional constructs carrying different combinations of 5' and 3' end truncations were tested. Specifically, to evaluate the editing efficiency of ASOs with double 5' and 3' end truncations, different ASOs were generated that lacked 1, 2, 3, 4, or 5 nucleobases at one or both ends. "No ASO" served as a negative control. The SERPINA1E342K targeting constructs and their modifications are also listed in Table 9. The results are shown in Table 9. Figure 10 Shown in.

[0312] ASOs carrying 5' and 3' end truncations (v117.83 and Segment Design_31 nt) showed an overall decrease in editing efficiency when compared to the full-length construct (v117.82 and Segment Design_40 nt). Figure 10 A). Interestingly, the two constructs with 3' end truncations (v117.85 and v117.86) showed only a small reduction when compared to the full-length construct (v117.82), suggesting that 3' end truncations are generally tolerated, particularly to a length of 5nt outside the CBT. On the other hand, 5' end truncations had a greater effect on the editing efficiency of specific oligonucleotides (v117.87 and v117.88). Surprisingly, truncating the 5' end to only 25 or 24 nucleotides outside the CBT was particularly effective, especially when combined with a 3' end length of 5 nucleotides, resulting in ASO v117.141 and v117.142 (33nt and 32nt in length, respectively). Figure 10As shown, the 5'- and 3'-terminally truncated versions maintained similar editing efficiencies as the 3'-truncated ASO.

[0313] Overall, these data confirm the results presented above, that is, 2'-F and 2'-OMe modifications combined with the 2'H group can provide stable ASOs without negatively affecting editing efficiency. The observed 2'-F modifications ranged from 37.5% to 42.4% (v117.141 and v117.82, respectively), 2'-H ranged from 20% to 25% (v117.82 and v117.142, respectively), and 2'-OMe ranged from 32.4% to 40% (v117.91 and v117.88, respectively). Importantly, the data confirm that continuous sugar-modified segments should be avoided and a certain level of 2'-F and 2'-OMe modifications should be included to obtain effective editing yields, as all embodiments contain uniform 2'-modified segments of no more than 6nt or even 3nt. In addition, the data from Example 9 show that truncation at the 3' end is generally better tolerated than truncation at the 5' end. In summary, the truncation of oligonucleotides results in the asymmetric positioning of oligonucleotides around CBT, for which there are preferred specific 5' ends and 3' end lengths. That is, the shortening of the 3' end can tolerate the 3' direction of CBT to be about 5nt in length (e.g., v117.86, v117.141, v117.142), while the shortening of the 5' end significantly reduces the overall editing efficiency of oligonucleotides (e.g., v117.88). Surprisingly, the data from Example 9 show that when optimal 2'-modifications and ideal internucleoside modifications / PS patterns and correct truncation are applied to the 5' end and 3' end, very short (33nt and 32nt) and highly effective oligonucleotides can be obtained. This is especially true for the symmetry 5'-24-3-5 (scheme: 5' end-CBT-3' end) of the 32nt embodiment (v117.142) and the symmetry 5'-25-3-5 of the 33nt embodiment (v117.141). Finally, our data suggest that a high content of stereopure bond modifications is not necessary to obtain high editing yields of short, stable ASOs.

[0314] Example 10. Fine screening of truncated SERPINA (33nt / 32nt) targeting oligonucleotides

[0315] To determine the combined effects of continuous and / or discontinuous 2'-F- and 2'-OMe-modifications in combination with DNA on the editing efficiency of short (32nt or 33nt), asymmetric ASOs, several SERPINA targeting constructs were prepared and used for additional screening experiments. Table 10 lists different 32nt and 33nt ASO designs and the corresponding backbone modifications. The corresponding results are shown in Figure 11"No ASO" was used as a negative control.

[0316] Table 10: 32nt and 33nt long SERPINA1 targeting construct sequences and modifications used in Example 10. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0317]

[0318] like Figure 11 As shown in Figure A, a 33 nt long construct (v117.141) containing a mix of 42.4% 2'-F- and 33.3% 2'-OMe-modifications and 24.2% DNA (5 nt outside of the CBT) showed good editing (47.3%) when compared to v117.167, a prior art-based construct containing a 2'-F-modification at the 5' end and a continuous segment (segment ≥ 5 nt) of 2'-OMe-modifications adjacent to the 5' and 3' ends of the CBT ( Figure 11 B) Interestingly, in the 33 nt construct, 2'-F- and 2'-OMe modified stretches of limited length (eg ≤ 5 nt) were well tolerated (v117.168).

[0319] Similarly, for the 32 nt long construct, the continuous 2'-F-strands at the 5' end of the DNA interruption by 2'-OMe-modification resulted in an increase in the overall editing efficiency (see v117.169 compared to v117.142 and v117.170) ( Figure 11 A). Both versions v117.168 and v117.170 show that exchanging DNA with 2'F in the short version (<40nt) can increase the editing yield. Specifically, the increase in editing efficiency in v117.170 indicates that continuous 2'-F- and 2'-OMe modifications of limited length (e.g., ≤6nt) may be beneficial for 33nt and 32nt short ASOs. In addition, it is shown that short ASOs may benefit from a reduced amount of DNA nucleosides (≤6nt) outside the CBT (v117.141 versus v117.168; v117.142 versus v117.170). In summary, this suggests that higher levels of 2'-F (40.6%-57.6% in v117.142 and v117.168, respectively) are well tolerated and are preferred even for shorter embodiments (≤50nt).

[0320] Overall, these findings are similar to the editing efficiencies of previously described oligonucleotides (Monian et al., 2022). Without being bound by any theory, the inventors believe that for efficient editing efficiency, a length of 30 nt (or a "sweet spot" of approximately 33 nt / 32 nt) is desirable (see Figure 11 The ASOs (results in ) rely on certain modification rules, namely lacking uniform stretches of 2'-F and 2'-OMe in both the 5'-half (5' end to CBT) and the 3'-half (3' end to CBT) of the ASO. These findings are also shown in Example 9 for the 40nt leader sequence, where the uniform scheme (i.e., continuous 2'-F-modification) is relatively inefficient in target editing.

[0321] Therefore, the inventors believe that the editing efficiency and stability of oligonucleotides can be optimized by using a combination of chemical modifications. For example, when using optimized 2'- and stereo-random bond modification patterns in oligonucleotides with optimized length and asymmetry, a large number of stereo-pure bond modifications (or simply introducing stereo-pure bonds) are possible. However, the data show that when the optimal modification pattern (maximum segment size, e.g., ≤6nt) and the amount of 2'-F and 2'-O-alkyl (approximately 35%-65% 2'-F and 30%-35% 2'-O-alkyl) and natural and stereo-random internucleoside modification patterns (e.g., bond d, bond e are modified; bond h, bond i are not thiophosphates) are obtained, efficient RNA editing can also be achieved with chemically modified oligonucleotides that do not contain a large number of stereo-pure internucleoside bonds. Although this does not exclude that a small amount of stereo-pure bond modifications may further improve editing efficiency (e.g., no more than 10, more preferably no more than 5), it clearly shows that high levels of stereo-pure bond modifications as seen in the prior art are not absolutely necessary. Furthermore, the data show that the positioning of the ASO, particularly the asymmetric end lengths comprising at least 4 nt in the 3' direction of the CBT and / or at least 16 nt in the 5' direction of the CBT, is very important in combination with other modification rules to obtain very efficient editing oligonucleotides of short length (e.g., less than 40 nt).

[0322] Example 11. Adjustment of the 3' end length of STAT1 Y701 targeting oligonucleotides

[0323] To determine the combined effects of 2'-F- and 2'-OMe-modifications, bond modifications, and DNA on editing efficiency, similar truncation experiments were performed on STAT1Y701-targeting oligonucleotides. The sequences of the specific oligonucleotide constructs used in Example 11 and their modifications are listed in Table 12, and the corresponding results are shown in Table 13. Figure 12 Shown in.

[0324] Table 12: STAT1 Y701 targeting construct sequences and modifications used in Example 11. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond; GalNAc = triantennary N-acetylgalactosamine.

[0325]

[0326] As a basis for introducing a mixture of 2'-F, 2'-OMe and 2'-H backbone modifications and subsequently generating a 3' end truncated ASO (v117.42), a 59nt long ASO construct (v117.29) was used (see Example 3). Construct v117.29 had a lysosomal stability of 2h (data not shown) and an editing efficiency of approximately 38%, with most of the bonds being phosphorothioate (PS) bonds and not containing any 2'-H modifications outside of the CBT. This editing efficiency was slightly reduced to 37.4% in construct v117.40 (59nt), which contained a mixture of 49.2% 2'-F, 32.2% 2'-OMe and 13.5% CBT external DNA (8nt outside of the CBT) modifications ( Figure 12 A). In addition, shortening the 3' end to achieve a total oligonucleotide length of 40nt (v117.42 and v117.44) improved the editing yield compared to the efficiency of the parental construct v117.29, thereby increasing the 2'-OH modification range to 62.5% 2'F (v117.44), reducing it to 30% 2'O-Me and reducing it to 0% CBT external DNA. When the oligonucleotide was further shortened to 33nt (v117.56 and v117.57), the achieved editing yield (approximately 50%) was even maintained. These embodiments contained 60.6%-69.7% 2'-F and 9.1%-0% DNA (up to 3nt CBT external DNA, v117.56 and v117.57, respectively). Of particular interest is that these short embodiments (≤50nt) also tolerate relatively low amounts of 2'-OMe modifications (21.2% in v117.56 and v117.57). This suggests that sufficient 2' ribose modification through the appropriate pattern of 2'-F- and 2'-OMe-modifications, PS bonds, and DNA should be observed when reasonable editing efficiency should be achieved.

[0327] In general, these findings indicate that a mixture of specific oligonucleotide modifications is required to maintain, improve, or restore ASO editing efficiency. Specifically, the data not only show that a mixture of 2'-F, 2'-OMe, and 2'-H modifications can be used to stabilize or maintain the editing efficiency of ASOs, but also show that an embodiment in which these oligonucleotides are subsequently shortened to at least 33nt (which contains such a mixture of 2'-F, 2'-OMe, and 2'-H modifications and PS bonds) does not negatively impact overall editing efficiency, and therefore, when reasonable editing efficiency should be achieved, a mixture of 2'-F, 2'-OMe, and 2'-H- and PS bond modifications is required. In addition, the embodiments presented in this data set show that a 2'-F segment size of up to 6nt is acceptable, and embodiments of short ASO lengths (e.g., ≤ less than 45nt) give preference to slightly higher levels of 2'-F (e.g., up to 70%), and benefit from reduced CBT external DNA content (as low as 0nt).

[0328] Example 12. Adjustment of the 3' end length of the mCTNNB1 T41 targeting oligonucleotide

[0329] To further test the hypothesis that an optimal mix of 2'-F-, 2'-OMe-, and 2'-H- and PS-linkage modifications is required to provide stable and effective ASOs, and to further validate the results obtained with 3'-terminally truncated STAT1-targeting ASOs (Example 10), similar truncation tests were performed using the mCTNNB1T41 targeting oligonucleotide. The specific constructed sequences and their modifications are listed in Table 12. The corresponding results are given in Table 12. Figure 13 "No ASO" served as a negative control.

[0330] Table 12: mCTNNB1 T41 targeting construct sequences and modifications used in Example 12. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0331]

[0332] The different ASOs were based on v117.22, a 59 nt long construct containing a mixture of 2'-F- and 2'-OMe-modifications, RNA / DNA and PS linkage modifications. Based on this construct, three 3' end truncated versions were generated containing different mixtures of 2'-F- and 2'-OMe-modifications, RNA / DNA and PS linkage modifications ( Figure 13). Replacing all natural RNA nucleosides with DNA and additional 2'-F- and 2'-OMe-modifications initially resulted in decreased editing efficiency (v117.24, 15.2% and v117.25, 12.6%). Interestingly, a 37nt-long construct (v117.29) containing up to 21.6% total DNA content (outside CBT, 5nt long) and 2'-F- and 2'-OMe-modifications showed an editing efficiency of 32.7%.

[0333] These data suggest that for 40 nt long ASOs, a mix of 2'-F- and 2'-OMe-modifications, RNA / DNA content, and PS bond modifications are important for editing. In particular, for these embodiments, a maximum of 6 DNA nucleosides were tolerated outside the CBT for short (≤50 nt, here 37-40 nt) oligonucleotides.

[0334] Example 13. Adjustment of the 5' and 3' end lengths of CRB1 C948Y targeting oligonucleotides

[0335] Similar to the generation of 5' and 3' truncated constructs specific for the SERPINA1 target (Example 8), 5' and 3' truncated versions of CRB1 C948Y specific oligonucleotides were generated and tested for their in vitro RNA editing efficiency. The different constructs tested in Example 13 are listed in Table 13. The results are shown in Table 13. Figure 14 "No ASO" served as a negative control.

[0336] Table 13: CRB1 C948Y targeting construct sequences and modifications used in Example 13. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0337]

[0338]

[0339] like Figure 14 As shown in A and 14C, the 45nt long version (v120.2) containing 2'-F, 2'-OMe and CBT external 2'-H modifications showed good RNA editing efficiency (50.9% and 49.9%). It is worth noting that truncation from the 5' end resulted in a gradual decrease in editing (v117.26 (15%) and v117.28 (18.7%)). It is worth noting that extension of the 3' end resulted in an overall increase in editing efficiency (v117.24 (37.2%) and v117.25 (38.2%)) ( Figure 14A). These data suggest that truncation of the 5' end of the oligonucleotide may have a significant impact on the editing efficiency of the ASO. Furthermore, the data show that higher 2'-F content (>50%) in 40nt long ASOs is associated with higher editing values ​​(v117.27, containing 52.5% 2'-F). Notably, all embodiments do not contain uniformly modified segments greater than 6nt.

[0340] The inventors continued to test shorter versions of the CRB1 C948Y targeting construct to assess whether there was a correlation between ASO length and overall 2' modification pattern and editing efficiency. To this end, a set of shorter CRB1 C948Y targeting constructs (v117.39 to v117.44) were generated (Table 13) and tested for their in vitro editing efficiency. Figure 14 As shown in C, approximately 20% of the DNA content (5 DNA nucleosides outside the CBT) in a 40nt long ASO (i.e., v117.26) resulted in reduced editing. However, further shortening to a length of 38nt (v117.44), the inventors discovered an "optimal length of 38nt." For example, for v117.44 (38nt long), the inventors found that the editing yield increased by approximately 33% (27.2% editing yield) compared to the similar 40nt embodiment v117.27. As seen in Example 12, these shorter embodiments (≤50nt) also tolerate very low amounts of 2'-OMe (21.2% in v117.40 and v117.41). In addition, the inventors discovered certain short (e.g., ≤50 nt) embodiments in which low levels of DNA (e.g., 3 nt outside of CBT) can improve editing yields compared to pure 2'-F and 2'-OMe versions, such as v117.27 (15% DNA / three DNA nucleosides outside of CBT) compared to v117.39 (7.5% DNA / no DNA nucleosides outside of CBT) and v117.40 (9.1% DNA / no DNA nucleosides outside of CBT) compared to v117.41 (18.2% DNA / three DNA nucleosides outside of CBT).

[0341] Overall, while these data demonstrate that ASOs containing, for example, no more than six consecutive nucleotides with the same modification and a mixture of 2'-F- and 2'-OMe-modifications have good editing efficiencies, the data further suggest that certain levels of DNA may perform better than pure 2'-F- and 2'-OMe-modified versions.

[0342] Example 14. Modification Preference of Short (<45nt) STAT1 Y701-Targeting ASOs after Abbreviation

[0343] To determine whether 2'-OMe modification of the substituted DNA affects RNA editing in a position-dependent manner, various ASOs were generated and tested for their STAT1 Y701 RNA editing efficiency. The different construct sequences and their modifications are listed in Table 14. The results are shown in Table 14. Figure 15 "No ASO" served as a negative control.

[0344] Table 14: STAT1 Y701 targeting construct sequences and modifications used in Example 14. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0345]

[0346]

[0347] The data suggest that substitution of DNA with 2'-OMe modifications affects editing in a position-dependent manner. Figure 16 As shown in A, in the region near CBT, 2'-OMe is preferentially modified instead of DNA (from N -9 to N +9 , v117.48 compared to v117.47). It was also shown that all four DNA nucleotides outside the CBT (see v117.42) could be replaced by 2'-F-modifications (in v117.44), keeping the maximum segment of 2'-F-modification ≤ 6 nt to increase the editing yield to a level of 64%, similar to the level of construct v117.48 (65.2%) ( Figure 16 A), showing that 2'-F modifications are generally well tolerated, even better than 59nt ASOs (see Figure 3 , for v117.44, the editing efficiency was 17.3%). However, in short ASOs (≤50nt), 2'-F-modified nucleosides outside the CBT can also be replaced to some extent by 2'-OMe and / or DNA nucleosides (e.g., 2nt DNA without loss of editing efficiency, v117.48), for example to reduce the overall 2'-F / PS content, which can cause toxicity in certain sequence contexts. However, the effect of replacing 2'-F-modifications with other 2'-modifications on editing efficiency may be very position-specific. Similarly, avoiding uniform 2'-modified segments (≤6nt) seems to be preferred. In addition, the data also show that higher levels of 2'-OMe (up to 40% in v117.43) are generally well tolerated and may have only a small effect on editing yield when properly placed.

[0348] Example 15. Identification of the 3' hotspot site of the central base triplet (CBT) of the SERPINA1 E342K targeting ASO

[0349] As previously described by the inventors of the present application, the first position in the 3' direction of conventional CBT (between [Am]-N -1 -N0-N +1 -N +2 The choice of nucleotide at the +2 position in the -[Bn] structure, where N0 is the editing site, can significantly affect the editing rate of the target. To further examine the effect of nucleotide modifications adjacent to the 3' end of the CBT, different SERPINA1 targeting constructs were generated with nucleotide modifications at the first two nucleotides immediately adjacent to the 3' end of the CBT (i.e., positions +2 and +3) (5'- d A -1 e N0 f A +1 g A + 2 h N +3 i In general, different SERPINA1 targeting constructs carry the same base sequence, differing only at positions +2 and +3 (Tables 15 and Figure 16 B) The results are Figure 16 Various SERPINA1 targeting ASOs and their modifications are listed in Table 15.

[0350] like Figure 16 As shown in A, construct v117.82 showed the highest level of editing efficiency, with a value of 42.8%. This construct v117.82 had a cleavage site at position +2 (N +2 ) contains a 2'-O-methyl modification at the +3 position (N +3 ) contains a 2'-fluoro modification. On the other hand, replacing the 2'-O-methyl modification at the +2 position with 2'-F, 2'-H or 2'-MOE modification results in a gradual decrease in SERPINA1 editing efficiency. Similarly, ASO constructs carrying 2'-OMe, 2'-H or 2'-MOE modifications instead of 2'-fluoro modifications at the +3 position show that SERPINA1 editing efficiency is reduced, with editing levels falling to as low as 0.4% (see v117.101, v117.102, v117.104, v117.106, v117.119, v117.120 and v117.124). Interestingly, carrying 2'-MOE modifications at the +2 position (v117.118 and v117.119) or at both +2 and +3 positions (v117.120) results in the greatest reduction in editing efficiency ( Figure 16 A). These data suggest that a 2-O-methyl modification at position +2 and a 2'-F modification at position +3 are required to provide optimal editing. Position +2 is particularly important, followed by position +3.

[0351] Collectively, these data suggest that in the immediate vicinity of CBT (5'–d A -1 e N0 f A +1 g –3') at the first two nucleosides at the 3' end of the 2'-modification type (i.e., N +2 and N +3 ) plays an important role in determining the efficiency of ASO editing. Interestingly, the inventors discovered a modification-sensitive "hotspot" region, which is adjacent to the 3' end of CBT, at the +2 and +3 positions. Therefore, the inventors believe that CBT and the first two nucleosides adjacent to the 3' end of CBT together form an "extended CBT" (5'- d A -1 e N0 f A +1 g A +2 h A +3 i –3'), where the optimal arrangement and / or modification to achieve efficient target editing is

[0352] 5'–CBT-mN-fN–3',

[0353] Where CBT refers to the central base triplet, mN is a 2'-O-methyl-modification, and fN is a 2'-fluoro-modification. The organization of this extended modification-sensitive "hotspot" is also Figure 16 Shown in B.

[0354] Interestingly, this extended CBT is also sensitive to internucleoside bond modifications, accepting PS modifications well at bonds d and e, and also well at f and / or g, but not at positions h and i. Thus, in one embodiment, d and e are PS bond modifications. In one embodiment, f is a PS bond. In one embodiment, g is a PS bond. In some cases, h and i are not PS bonds.

[0355] Table 15: SERPINA1 E342K targeting construct sequences and modifications used to evaluate hotspot sites in the 3' direction of CBT in Example 15. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), oN = 2'-MOE, N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond.

[0356]

[0357] To further investigate the expansion hotspot region, the in vitro editing efficiency of CTNNB1 T41A targeting ASOs carrying 2'-O-methyl or 2'-F-modification at the +2 or +3 position was evaluated. The sequence modifications of the different ASOs tested are shown in Table 16 and the results are presented in Figure 17 "No ASO" served as a negative control.

[0358] Table 16: Murine CTNNB1 T41A targeting construct sequences and modifications used in Example 15. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), oN = 2'-MOE, N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0359]

[0360] like Figure 17 As shown in A, construct v117.22 carrying a 2'-O-methyl modification at the +2 position and a 2'-F-modification at the +3 position showed an editing efficiency of 33.9%, which was higher than the editing efficiency of the construct with 2'-OH at the +2 position and 2'-O-methyl at the +3 position (v117.20 (18.9%)) or the construct with a 2'-F-modification at the +3 position (v117.21 (27.5%)).

[0361] Thus, these data suggest and confirm that the optimal mix of combinations and arrangements for well-maintained RNA editing is defined by an extended hotspot with a 5'-CBT-mN-fN-3' structure.

[0362] Example 16. Substitution of 2'-MOE for the 2'-OMe-modified terminal segment in SERPINA1 E342K-targeting ASO

[0363] It is known that 2'-MOE residues are used in large numbers in splice-switching oligonucleotides and generally have very low toxicity. Due to their bulk and location in the minor groove, they are not easily accepted in large numbers and are completely unacceptable at certain positions in ASOs that recruit ADARs. To determine the effect of 2'-MOE modification on stability, toxicity, and editing efficiency, different SEPINA1 targeting constructs were generated and tested for their editing efficiency and lysosomal stability. The different ASOs and their sequence modifications are listed in Table 17. The results are shown in Table 17. Figure 18 As shown in the figure, “no ASO” was used as a negative control. CellTox TM Toxicity was quantified using the Green Cytotoxicity Assay (Promega) and normalized to negative ("no ASO") and positive controls (v117.59).

[0364] Initial testing involved generating a 59nt long ASO containing a total of six 2'-MOE nucleotides, three of which were placed consecutively at the end of each terminus (v117.68). This ASO variant was based on a 59nt lead ASO and an 85% PS modified backbone (v117.59). Notably, the introduction of the 2'-MOE modification at the end of the lead ASO did not affect the overall editing yield ( Figure 18 A). Similarly, the lysosomal half-life of the v117.68 variant was not affected when compared to the control, with both constructs showing t 50 >7 days( Figure 18 B) However, if Figure 18 As shown in Figure C, the cytotoxicity of the test construct was surprisingly reduced to only 26.6% when compared to the v117.59 control.

[0365] Similarly, the 2'-MOE modification was introduced into a 40nt long ASO in the same manner as v117.68. This ASO variant is based on the 40nt lead ASO and also contains an 85% PS modified backbone (v117.137). Figure 18 E. The construct containing the 2'-MOE end had a slightly lower editing yield than the construct without the 2'-MOE end. The lower editing yield of v117.172 may be due to the 2'-MOE modification being positioned closer to the CBT, which is a well-known contact site for ADAR enzymes. Without being bound by any theory, the inventors suggest that the 2'-MOE modification should be placed away from known enzyme contact sites to prevent interference with protein binding by the bulky 2'-MOE modified side chain.

[0366] Table 17: SERPINA1 E342K targeting construct sequences and modifications used in Example 16. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), oN = 2'-MOE, N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) bond; oN ​​= 2'-MOE.

[0367]

[0368] These data indicate that the presence of 2'-MOE terminal segments at the 5' and 3' ends of the ASO affects overall ASO activity. For example, these findings suggest that 2'MOE modifications at the 5' and 3' ends of the ASO may contribute to enhanced lysosomal stability. In addition, the data indicate that for long (>40nt) embodiments, 2'-MOE terminal segments at the 5' and 3' ends can be incorporated without affecting editing efficiency while significantly reducing cytotoxicity. Relative to embodiments containing smaller 2'-modifications (e.g., 2'-OMe), the editing efficiency of shorter embodiments containing terminal 2'-MOE modifications is slightly reduced.

[0369] Example 17. Reduction of PS bond backbone modifications in SERPINA1 E342K-targeted ASOs

[0370] Phosphorothioate (PS) bonds can add beneficial properties to ASOs by improving albumin binding, cellular uptake, endosomal escape, and protein binding. Furthermore, PS bonds have been reported to transport ASOs from the cytosol to the nucleoplasm (Crooke et al., 2020). On the downside, PS bonds tend to make ASOs sticky, which can lead to protein and / or ASO aggregation and toxicity. For example, an increase in PS-related toxicity has been observed in ASOs enriched in 2'-F modifications. Therefore, there is growing interest in understanding how to modify ASOs with PS bonds to ultimately reduce the amount of PS bonds and / or 2'-F content per ASO.

[0371] To test four different ASO constructs with reduced PS content and to determine the impact of reduced PS bond modification on editing efficiency, two different model systems [a plasmid-based approach (A) and a genomic integration approach (B)] were used.

[0372] Plasmid transfection method (A): 2.5×10 4 HeLa cells were seeded in 24-well plates. After 24 h, cells were forward transfected with a plasmid containing the SERPINA1E342K (PiZZ) mutant cDNA. Forward transfection was performed by mixing 300 ng of plasmid with 0.9 μl 6 (Promega) were diluted in 50 μ l Opti-MEM and cultivated for 5 minutes, then the two mixtures were combined and cultivated for another 20 minutes. The culture medium was changed, and the transfection mixture was evenly distributed in one hole. 24 hours after the plasmid transfection, the cells were forward transfected with 5 pmol construct / well and 1.5 μ l / well Lipofectamine RNAiMAX reagent (ThermoFisher Scientific). The forward transfection of the ASO construct was carried out by mixing the construct and liposome transfection reagent in 50 μ L OptiMEM. After cultivating for 5 minutes, the two solutions were combined and cultivated for another 20 minutes. The culture medium was changed, and the transfection mixture was evenly distributed in one hole. After 24 hours, the culture medium was changed again. 48 hours after transfection, the cells were harvested for RNA separation and prepared for Sanger sequencing.

[0373] Genome integration method (B): 1×10 cells containing the cDNA gene of human SERPINA1 E342K mutation stably integrated into its genome via the piggyBac transposase system were 5 HeLa cells were seeded in 24-well plates. After 24 hours, cells were forward transfected using the Lipofectamine RNAiMAX protocol with 25 pmol of construct / well and 1.5 μl / well of Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific) (procedure described above). After 24 hours, cells were harvested for RNA isolation and Sanger sequencing.

[0374] The different test constructs were compared to a PS-rich leader construct (v117.59, 59 nt long) with approximately 90% PS content. In the most extreme construct, PS bonds were only contained at the 5' and 3' ends (2 × 3 PS bonds per end) and around the CBT (4 PS), providing a total PS content of only approximately 17% (v117.71). Although this PS-poor ASO still has significant lysosomal stability ( Figure 19 C, t 50 >7 days) and underwent significant editing, but the editing effect was greatly reduced compared to the PS-rich leader (v117.59) ( Figure 19 A and Figure 19 B).

[0375] To further evaluate the effects of different levels of PS content on ASO activity and stability, additional PS bonds were introduced in the 5' or 3' direction of each DNA base of the ASO (see v117.72 and v117.73) (Tables 18 and Figure 19D). This resulted in a total PS content of approximately 30.5% (v117.72) and 32.2% (v117.73). Figure 19 As shown in A and 19B, regardless of the method used, increasing total PS content mediated a clear improvement in editing yield when compared to v117.71.

[0376] Furthermore, reducing the total PS content in each ASO did not significantly reduce triton lysosomal stability (t 50 >7 days). Notably, the PS bond at the 3' end of DNA was found to be more important than the PS bond at the 5' end of DNA for triton lysosomal stability ( Figure 19 C, v117.73, t 50 =96h). Introducing PS linkages 5' and 3' to each DNA base generated an ASO with only 49.2% PS content (v117.74), however, the triton lysosomal stability (t 50 >7 days) and editing efficiency, this v117.74 was similar to the PS-rich leader construct v117.59 ( Figure 19 ).

[0377] Table 18: SERPINA1 E342K targeting construct sequences and modifications used in Example 17. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0378]

[0379]

[0380] These findings suggest that the PS content in fully modified ASOs can be significantly reduced without reducing editing efficiency. However, a decrease in editing efficiency may be encountered. A bond modification degree of at least 15% is required. For long embodiments (e.g., ≥40nt), the total internucleoside modification content should be in the range of 15%-90%, and editing efficacy and lysosomal stability are maintained when the internucleoside modification content is as low as 30% (v117.72 and v117.73).

[0381] The inventors continued to test the effect of PS content on shorter ASOs (40nt). The various test constructs are listed in Table 19. The results are shown in Figure 20 Shown in.

[0382] Table 19: SERPINA1 E342K targeting construct sequences and modifications used in Example 17. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0383]

[0384] The 40nt short leader ASO contains 85% PS bond content (v117.82). As with longer ASOs, PS bonds are only added directly at the 3', 5', or 3' and 5' positions of each DNA nucleotide, thereby reducing the total PS bond content in the 40nt short ASO to about 30% or 50%. Construct v117.96 is identical in sequence to v117.109, differing only in the last two nucleotides at the 3' end (v117.96 contains two 2'-OMe modifications, which are not present in v117.109) and the PS content, and serves as a control oligonucleotide for 50% PS content because v117.82 also has a terminal segment at the 3' end (which is not present in the other tested versions). Constructs v117.107, v117.108, and v117.109 exhibited a total PS content of approximately 30%, 30%, and 47.5%, respectively. Figure 20 As shown, v117.109 failed to fully restore the editing yield of PS-rich leader ASOs, suggesting that short ASO leaders (40 nt) may be less tolerant to reduced PS modification content than long ASO leaders (59 nt).

[0385] In addition, the inventors found that for short ASOs (≤50nt, for example 40nt), fragments with consecutive linkage modifications (at least 10) outperformed random distribution of linkage modifications when reducing the total PS content. Different SERPINA1 E342K targeting constructs and their modifications are listed in Table 20. "No ASO" served as a negative control.

[0386] Table 20: SERPINA1 E342K targeting construct sequences and modifications used in Example 17. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0387]

[0388]

[0389] An ASO with no more than 3 consecutive PS bonds (v117.107, 30% PS content) was tested and compared to an ASO with the same sequence and 10 consecutive PS bonds (v117.132; 47.5% PS) or 16 consecutive PS bonds (v117.133; 60% PS). Figure 21 As shown in A, 10 consecutive PS bonds increased the editing yield, but did not rescue the editing yield of the high PS ASO (v117.82). However, 16 consecutive PS bonds achieved similar editing levels as the PS-rich leader ASO (v117.82; with 25 consecutive PS bonds, 85% PS). This highlights that for short embodiments (≤50nt), the bond modification content can be reduced (e.g., to at least 30%), but continuous stretches of modified bonds (e.g., PS) are more advantageous than random distribution of modified bonds (e.g., PS) throughout the ASO. In addition, the data suggest that the PS content in these fully modified ASOs can be greatly reduced, which may be important when toxicity needs to be reduced.

[0390] It is worth noting that many ASOs known in the prior art (e.g., WO 2021 / 071858 and WO 2022 / 099159) contain very high degrees of (stereopure) phosphorothioate bond modification to achieve editing yields in a similar range. More importantly, their data are based on editing results in primary mouse hepatocytes, which generally provide high editing yields and may therefore provide a limited perspective.

[0391] It is worth noting that bond modifications often have a certain threshold. Therefore, according to the present invention, in one embodiment, the bond modification content (e.g., PS) is at least 30% to obtain the best editing yield. In one embodiment, the PS bond modification content (e.g., PS) is at least 50%. In one embodiment, the PS content is at least 60%. Without being bound by any particular theory, shorter ASOs appear to be more likely to use more than 10 consecutive bond modifications (e.g., PS) rather than a dispersed modification pattern.

[0392] Example 18. Improved Editing Efficiency and Potency of LNA-Modified SERPINA1 E342K-Targeting ASO

[0393] When shortening the ASO, it was found that the 40nt lead ASO can produce slightly reduced editing yield and efficacy. It is known in the art that LNA provides enhanced stability against enzymatic degradation and provides improved specificity and binding affinity in base pairing. Therefore, a 40nt long ASO with 85% PS content (v117.82; "no LNA") was modified by introducing LNA building blocks at the 5' and 3' ends or within the oligonucleotide sequence. The different ASOs used in Example 17 are listed in Table 21, and the corresponding results are in Table 22. Figure 22 Shown in.

[0394] Table 21: SERPINA1 E342K targeting construct sequences and modifications used in Example 18. lN = LNA, mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0395]

[0396]

[0397] like Figure 22 As shown in A, for construct v117.97, which contains a total of four terminal LNAs (two LNAs at each end separated by 2'-OMe), LNA modification had little effect on the overall editing yield at high ASO doses compared to v117.82. However, v117.97 had a slightly improved potency, with significantly higher editing yields at 2.5 pmol and 1.25 pmol ASO doses ( Figure 22 B) Placing the LNA modification closer to the CBT in the ASO appears to perturb the editing efficiency and potency of the ASO when compared to ASOs carrying LNA modifications only at the 5' and 3' ends (see v117.97 vs. v117.98).

[0398] To determine the combined effect of terminal LNA modification and ASO length on editing efficiency, LNA-modified versions of shorter ASOs (36 nt) with 85% PS content (v117.129 to v117.131) were generated to contain 2, 4, or 6 LNA-modified nucleotides at their 5' end (v117.86; "no LNA"), as shown in Tables 22 and Figure 23 The results are depicted in B. Figure 23 Shown in A.

[0399] While the two 5'-terminal LNAs were well-accepted and increased the overall editing of the ASO from 29% (v117.86) to 49.4% (v117.129), the additional 5'-terminal LNA interfered with the editing yield (v117.130 and v117.131). Therefore, without being bound by any theory, the inventors believe that short ASOs may benefit from a small amount of 5'-terminal LNA.

[0400] Finally, the inventors evaluated whether 40nt ASOs with reduced PS content of approximately 30% (v117.107 and v117.126) and 48% (v117.109 and v117.127) would benefit from terminal LNA. Different SERPINA1E342K targeting constructs were generated to contain a total of 4 LNAs (2 LNA modifications at each end interrupted by 2'-OMe). The different constructs and their modifications are listed in Table 22, and the results are shown in Table 23. Figure 24 Shown in A.

[0401] Table 22: SERPINA1 E342K targeting construct sequences and modifications used in Example 18. lN = LNA, mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), N = 2'-OH (ribose; RNA), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0402]

[0403]

[0404] Interestingly, it was found that the 40nt short SERPINA1 E342K targeting ASO with reduced PS strongly benefited from 3' and 5' terminal LNAs (v117.127), achieving an editing yield higher than that obtained with a 40nt lead ASO (v117.82) with 85% PS and lacking terminal LNA modification. These data suggest that the overall editing efficiency and efficacy of ASOs can be improved by LNA modification, and more specifically, by placing a certain number of LNAs (e.g., up to 6, more preferably less than 4) at the end (e.g., preferably at the 5' end) of the ASO (preferably in a short embodiment of length ≤50nt). More importantly, the data show that despite the reduction in PS content, the performance of the ASO is improved by LNA base modification. Without being bound by any particular theory, the inventors believe that LNA can help compensate for the low PS content in short embodiments (≤50nt, e.g., ≤40nt).

[0405] Example 19. Interruption of 2′-modified continuous segments in SERPINA1 E342K and STAT1 targeting constructs

[0406] ASOs carrying long stretches (or "continuous stretches") of the same sugar modification at the 2' position (e.g., long stretches of 2'-O-methyl interrupted only by CBT, or a combination of long 2'-F stretches and long 2'-O-methyl stretches interrupted only by CBT) are known in the art. Specifically, Monian et al. (2022) previously showed that chemically modified oligonucleotides ("AIMers") with chimeric backbones containing continuous 2'-Fluoro- and / or 2'-OMe-modified backbones and highly (stereopure) PS and PN bonds can mediate efficient in vitro target editing.

[0407] To determine the effect of interrupting such "continuous stretches" (e.g., continuous stretches of 2'-fluoro- and / or 2'-OMe-backbone modifications), different SERPINA 1E342K and STAT1 Y701 targeting constructs were generated and tested for their in vitro editing efficiency. The constructs used in Example 18 are listed in Tables 23 and 24, and the results are in Tables 23 and 24, respectively. Figure 25 and 26 "No ASO" served as a negative control.

[0408] Table 23: SERPINA1 E342K targeting construct sequences and modifications used in Example 19. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0409]

[0410]

[0411] like Figure 25 As shown in Figure 3, for the SERPINA1 E342K targeting construct, interrupting the continuous segment of 2'-modifications (e.g., 2'-F and / or 2'-O-methyl) increased the overall editing efficiency of the ASO. The data showed a negative correlation between the size of the continuous 2'-modification segment and the respective editing yield. Figure 25As shown in A, the 40nt control ASO (segment design_40nt) contains a 2'-F segment (e.g., 20nt) and a 2'-OMe segment (e.g., 8nt) adjacent to the 5' end of CBT and a 2'-OMe segment (e.g., 9nt) adjacent to the 3' end of CBT, which shows a low editing efficiency of only about 20%. In contrast, v117.123 and v117.158 obtained the highest editing yields (achieving an editing yield of about 60%), both of which contain a 2'-F segment interrupted by a single 2'-O-methyl modification (e.g., a maximum 2'-F segment size of 5nt) and two 2'-O-methyl segments (e.g., a maximum 2'-O-methyl segment size of 4nt) interrupted by a single 2'-F modification or a small segment (e.g., 3nt) modified by 2'-F. This shows that the interrupted segment is very important for optimizing editing efficiency.

[0412] The data further illustrate that DNA nucleosides can be used to interrupt 2'-F segments or 2'-O-methyl segments (v117.82). Although in this case, DNA may not be ideally placed, resulting in some losses in editing efficiency, it still performs better than the segment design-40nt control. In addition, the data illustrate that relatively large 2'-F segments and 2'-O-methyl segments can be accepted in the 5'-half of ASO (v117.155). However, the editing yield is significantly lower than those constructs with smaller 2'-F and / or 2'-O-methyl segment sizes (e.g., v117.121, v117.123 and v117.158 compared to v117.153, v117.154, v117.158). The low performance of constructs v117.152, v117.153, and v117.154 (e.g., editing yields of approximately 25%-40%) when compared to the two best embodiments in this dataset (e.g., v117.158 and v117.123, which had editing yields of approximately 60%) indicates the importance of combining segmental disruptions in the large 2'-F segment with disruptions in the two large 2'-OMe segments surrounding the CBT. Furthermore, these embodiments show that a large amount of 2'-OMe (55% in v117.121) can be well tolerated and even improves editing yields when replacing DNA nucleosides (approximately 60% editing yield for v117.121 compared to 40% editing yield for v117.82).

[0413] Similarly, STAT1 Y701 targeting ASOs were generated that interrupted the continuous modified segment carried and tested for their RNA editing efficiency. The STAT1 Y701 targeting constructs are listed in Table 24 and the results are in Figure 26 "No ASO" served as a negative control.

[0414] Table 24: STAT1 Y701 targeting construct sequences and modifications used in Example 19. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage.

[0415]

[0416] The STAT1 Y701 targeting ASO is based on a short embodiment (≤50nt, e.g., 40nt) in which the ASO contains less than or equal to three (≤3nt) consecutive 2'-OMe modifications and less than or equal to six (≤6nt) consecutive 2'-F-modifications (v117.44). The editing results of this embodiment (42%) were the best in the data set. This was compared to a control ASO with a large 5'-terminal 2'-F segment (20nt) and two large 2'-O-methyl segments (8nt) adjacent to the 3' end of the CBT and a 2'-O-methyl segment (9nt) adjacent to the 5' end of the CBT (see v117.53). This version performed very poorly, achieving a low editing yield of only 7%, close to the detection limit of Sanger sequencing. Through domain swapping experiments, the inventors showed that for this target sequence, segment breaks in the 2'-O-methyl segment are particularly important. This was particularly demonstrated in v117.54, which still contained a large 2′-F segment and still performed well compared to v117.55, which contained a segment interruption in both the 2′-F segment and two large contiguous 2′-O-methyl segments.

[0417] This suggests that large 2'-F segments may be more readily tolerated than large 2'-O-methyl segments. However, the best performing implementation featured a maximum contiguous segment size of 6 nt (v117.44).

[0418] Example 20. 2'-FANA Modification within the CBT of the CRB1 C948Y Targeting Construct

[0419] 2'-FANA modification within CBT has been shown previously (WO 2021 / 243023), and in particular in N +1 To determine the effect of 2'-FANA modification within and adjacent to the 5' end of the CBT, the N of the CRB1 C948Y targeting construct was -2 、N -1 , N0 and / or N +1 Two different patterns of 2'-FANA modifications were introduced at the positions. The various constructs and their modifications used in Example 19 are listed in Table 25. The corresponding results are shown in Table 25. Figure 27 "No ASO" was used as a negative control.

[0420] Table 25: CRB1 C948Y targeting construct sequences and modifications used in Example 20. mN = 2'-O-methyl (2'-OMe), fN = 2'-fluoro (2'-F), dN = 2'-H (deoxyribose; DNA), * = phosphorothioate (PS) linkage; aN = 2'-FANA.

[0421]

[0422] like Figure 27 As shown in A, the combination of one or more 2'-FANA modifications within the CBT and / or adjacent to the 5' end of the CBT, along with 2'-F and / or 2'-OMe modifications outside the CBT, significantly improved overall editing efficiency when compared to the control, highlighting that in this embodiment FANA is more than just a replacement for DNA, 2'-F, and 2'-O-methyl groups within the CBT.

[0423] Relative to the control containing 17 natural RNA nucleosides, the introduction of 2'-F, 2'-OMe and 2'-H modifications alone resulted in a slight decrease in editing efficiency (see v117.29 compared to v120.2). +1 The introduction of a 2'-FANA modification at the N position (v117.30; two 2'-FANAs inside the CBT) resulted in an increase in editing efficiency to the level of an RNA-enriched control ASO (v120.2), significantly outperforming a fully 2'-modified embodiment lacking FANA (v117.29; no FANA). -2 and N -1The addition of an additional 2'-FANA modification has been well accepted but did not further enhance the editing efficiency of ASO (v117.31). Interestingly, the control ASO (v117.33), which contains two additional nucleosides at the 5' end and three additional nucleotides at the 3' end relative to v117.30, showed a sharp drop in editing efficiency, below the Sanger sequencing detection limit (less than 5%). This control construct is based on an embodiment previously disclosed in the prior art. That is, the 2'-sugar and bond modification pattern is the same as the sequence KB-018-698 listed in patent WO 2021 / 243023. In this case, the nucleobase sequence is changed to match the CRB1 C948Y site by copying the modification framework and sequence symmetry of KB-018-698 and transferring it to another target transcript / site. As expected, uniform 2'-O-methyl modifications on the outside of the CBT (a hallmark of v117.33) strongly interfered with RNA editing efficiency and could not be rescued by the presence of FANA of the CBT or by increasing the length of the ASO (50 nt), again highlighting the importance of combining optimal bond and 2' modification patterns (2'-O-methyl, 2'-F, and / or DNA), correct positioning of the ASO (asymmetric for short embodiments ≤50 nt), and limited segment size (e.g., ≤6 nt) of uniform 2'-modifications (e.g., 2'-O-methyl, but also 2'-F and DNA).

[0424] These data suggest that a combination of 2'-FANA modifications within and adjacent to the 5' end of the CBT, along with 2'-F and 2'-OMe modifications, can be used to stabilize ASOs and maintain or improve their editing efficiency. In general, ASOs containing a combination of 2'-F and 2'-OMe modifications and DNA were able to tolerate a total of at least four 2'-FANA modifications.

[0425] In summary, the inventors have shown that a balanced mixture of modifications at the 2' position of the sugar moiety of an oligonucleotide (e.g., at least 10% (preferably 20%-70%) 2'-F-, at least 10% (preferably 20%-60%) 2'-OMe-, sometimes containing 2'-H (no more than 50% in long ASOs (e.g., ≥40nt) and no more than 6 DNA in short ASOs (e.g., ≤50nt)), 2'-OH, etc.) can be stable in the lysosomal environment, thereby protecting them from nuclease digestion, such as during uptake, while retaining a high level of editing efficiency efficacy. In particular, the inventors have successfully shown that a mixture of such modifications with specifically placed internucleoside linkage modifications (e.g., linkage modifications at linkages d and e, but not PS modifications at linkages h and i) provides a method for generating efficient and stable ASOs. In particular, the inventors have shown that, depending on the length of the ASO, placing these specific 2'-modifications at specific sites within the ASO can have a significant impact on the overall editing efficiency of the ASO. In addition, the inventors have identified extended hotspot regions (5'-CBT-mN-fN-3'). In cases where the positioning of the 2'-modification is less important, the combination of chemical modifications is crucial. In addition, while the inventors have shown that a certain level of 2'-F and 2'-OMe modifications should be included ("extended hotspots"), continuous stretches of homogeneous sugar modifications should be explicitly avoided (e.g., ≤6nt for 2'-F, 2'-O-methyl and / or DNA, with 2'-F being the most readily accepted for larger homogeneous stretches). Furthermore, it has been shown that the placement of these modifications is position-independent to some extent, and that specific modifications and / or modification patterns can be transferred to different ASOs with different target specificities in a position-specific manner to some extent. The inventors have shown that, at least for embodiments of the present invention, FANA modifications inside CBTs are well accepted at specific positions and can increase editing yields compared to CBTs composed solely of DNA. Furthermore, the inventors have successfully shown that the PS content can be significantly reduced (down to 15%), especially for long embodiments (≥40 nt); and short embodiments (≤50 nt) benefit from longer consecutive stretches of bond modifications (e.g., PS) (e.g., stretches of ≥10 bond modifications) and / or in combination with (terminal) LNA modifications, or generally slightly higher levels of bond modifications (e.g., at least 30%, such as PS).

[0426] Those skilled in the art will appreciate that the present disclosure may be modified in ways not specifically described herein.

Claims

1. A chemically modified oligonucleotide comprising a sequence of 23 to 80 nucleotides in length capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), where The central nucleotide (N0) is directly opposite the target adenosine in the target RNA, wherein the core oligonucleotide comprises the following sequence: 5’-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3’; and in: (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications; (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification; (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification; (d) an internucleoside linkage modification content of at least 15%; and (e) Bond h and bond i are not phosphorothioate (PS) bonds.

2. The chemically modified oligonucleotide according to claim 1, wherein 20% to 100% of the nucleotides are deoxyribonucleosides or 2'-modified nucleotides, preferably wherein 50% to 100% of the nucleotides are 2'-modified nucleotides.

3. The chemically modified oligonucleotide according to claim 1 or 2, wherein 20% to 70% of the nucleotides are 2'-F-modified, preferably wherein 35% to 65% of the nucleotides are 2'-F-modified; and / or wherein 20% to 60% of the nucleotides are 2'-O-methyl (2'-OMe) modified, preferably wherein 25% to 55% of the nucleotides are 2'-OMe modified.

4. The chemically modified oligonucleotide according to any one of claims 1 to 3, wherein (i) no more than 95%, 90%, 85%, 80%, 70%, 60%, 50%, 40%, 30%, or 20% of the linkages outside of the CBT are internucleoside linkage modifications; or (ii) 15% to 90% of the linkages are internucleoside linkage modifications, preferably wherein 40% to 80%, most preferably 45% to 60% of the linkages are internucleoside linkage modifications.

5. The chemically modified oligonucleotide according to any one of claims 1 to 4, wherein the oligonucleotide is 28 to 70 nucleotides in length.

6. The chemically modified oligonucleotide according to claim 5, wherein the length of the oligonucleotide is: (i) 28 to 60, 28 to 55, or 28 to 45 nucleotides; (ii) 59 nucleotides; or (iii) not more than 45 nucleotides.

7. The chemically modified oligonucleotide of claim 6, wherein the oligonucleotide is 45 or fewer nucleotides in length and wherein no more than 4 nucleotides outside the CBT are deoxyribonucleotides.

8. A chemically modified oligonucleotide comprising a sequence of 23 to 50 nucleotides in length capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), where The central nucleotide (N0) is directly opposite the target adenosine in the target RNA, wherein the core oligonucleotide comprises the following sequence: 5’-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3’; and in: (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications; (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification; (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification; (d) the regions adjacent to the 3' and 5' ends of the CBT contain no more than 6 deoxyribonucleosides in total; and (e) The content of internucleoside linkage modifications is at least 30%.

9. The chemically modified oligonucleotide according to claim 8, wherein outside the CBT, (i) the oligonucleotide does not contain any deoxyribonucleosides, or (ii) no more than 1, 2, 3 or 4 nucleotides are deoxyribonucleotides.

10. The chemically modified oligonucleotide according to claim 8 or 9, wherein the content of the internucleoside linkage modification is between 30% and 90%.

11. A chemically modified oligonucleotide comprising a sequence of 40 to 80 nucleotides in length capable of binding to a target sequence in a target RNA, comprising a central base triplet (CBT) of three nucleotides (5'- A -1 e N0 f A +1 g -3'), where The central nucleotide (N0) is directly opposite the target adenosine in the target RNA, wherein the core oligonucleotide comprises the following sequence: 5’-N -5 a N -4 b N -3 c N -2 d A -1 e N0 f A +1 g N +2 h N +3 i N +4 j -3’; and in: (a) at least two of the three nucleotides of the CBT are chemically modified at the 2' position of the sugar moiety or are deoxyribonucleosides, or a combination thereof, and wherein d and e are internucleoside linkage modifications; (b)N +2 The nucleotide carries a 2'-O-alkyl modification; and wherein N +3 Nucleotides carry 2'-fluoro (2'-F)-modification; (c) at least 10% of the nucleotides are 2'-F-modified and at least 10% of the nucleotides are 2'-O-alkyl modified, wherein no more than 6 consecutive nucleotides have the same 2'-modification; (d) The total content of deoxyribonucleosides in the region adjacent to the 3' end and the 5' end of CBT is 5% to 50%.

12. The chemically modified oligonucleotide according to claim 11, wherein the content of deoxyribonucleosides outside the CBT is 10% to 40%, more preferably 11% to 30%, even more preferably 13% to 25%.

13. The chemically modified oligonucleotide of any one of claims 1 to 12, comprising at least one internucleoside linkage modification selected from the group consisting of phosphorothioate (PS), 3'-methylenephosphonate, 5'-methylenephosphonate, 3'-phosphoramidate, 2'-5' phosphodiester, and phosphoguanidine (PN).

14. The chemically modified oligonucleotide of claim 13, wherein at least one internucleoside linkage modification is PS.

15. A chemically modified oligonucleotide according to claim 8 or 11 or any claim dependent therefrom, wherein bond h and bond i are not phosphorothioate (PS) bonds.

16. The chemically modified oligonucleotide according to any one of claims 1 to 15, wherein bond h and bond i are phosphate (PO) bonds.

17. The chemically modified oligonucleotide according to any one of claims 1 to 15, wherein (i) no more than 4, 5 or 6 consecutive nucleotides are 2'-F-modified; and / or (ii) no more than 4, 5 or 6 consecutive nucleotides are 2'-O-alkyl modified.

18. The chemically modified oligonucleotide according to any one of claims 1 to 17, wherein fewer than 6, 5, 4 or 3 consecutive nucleotides have the same 2'-modification.

19. The chemically modified oligonucleotide according to any one of claims 1 to 18, wherein the chemically modified oligonucleotide comprises one or more stereorandom internucleoside linkage modifications.

20. The chemically modified oligonucleotide according to any one of claims 1 to 19, wherein the oligonucleotide comprises no more than 10, preferably no more than 5, stereopure internucleoside linkages.

21. The chemically modified oligonucleotide according to claim 20, wherein the stereopure bond is a PS bond and / or a PN bond.

22. The chemically modified oligonucleotide according to any one of claims 1 to 21, wherein the oligonucleotide does not comprise a stereopure PS bond and / or does not comprise a stereopure PN bond.

23. The chemically modified oligonucleotide according to any one of claims 1 to 22, wherein the chemically modified oligonucleotide does not comprise a stereopure PS bond modification.

24. The chemically modified oligonucleotide of any one of claims 1 to 23, wherein the oligonucleotide comprises: (i) 2'-O-(2-methoxyethyl)-oligoribonucleotide (2'-MOE) terminal stretches at the 3' and 5' termini, wherein at each terminus, no more than four nucleotides have 2'-MOE, preferably no more than three nucleotides have 2'-MOE; and / or (ii) locked nucleic acids (LNAs) at the terminals, wherein the oligonucleotide comprises 2 to 6 LNAs at each terminal or 5' terminus; preferably wherein the oligonucleotide comprises 2 LNAs at each terminal or 5' terminus.

25. The chemically modified oligonucleotide according to any one of claims 1 to 24, wherein bond g is not a PS bond, preferably wherein bond g is a phosphate (PO) bond.

26. The chemically modified oligonucleotide of any one of claims 1 to 25, wherein d and e are PS bond modifications, optionally wherein f is an internucleoside bond modification.

27. according to any one of claims 1 to 26 chemically modified oligonucleotide, wherein the modification at the 2'-position of (a) is (i) 2'-O-alkyl modification, (ii) 2'-F-modification, or (iii) 2'-fluoroarabinoside (FANA) modification.

28. The chemically modified oligonucleotide according to any one of claims 1 to 27, wherein the 2'-O-alkyl modification is a 2'-OMe-modification.

29. The chemically modified oligonucleotide according to any one of claims 1 to 28, wherein each of the three nucleosides of CBT is one or a combination thereof: (i) deoxyribonucleotides; and / or (ii) 2'-FANA-modification; and / or (iii) 2'-O-methyl-modification; and / or (iv) 2'-F-modification.

30. The chemically modified oligonucleotide according to any one of claims 1 to 29, wherein (i)N -1 is 2'-F, 2'-FANA, DNA or 2'-O-methyl; and / or (ii) NO is 2'-FANA or DNA; and / or (iii)N +1 It is 2'-FANA, DNA or 2'-O-methyl.

31. A chemically modified oligonucleotide according to any one of claims 1 to 30, wherein NO is deoxycytidine or FANA-cytidine.

32. The chemically modified oligonucleotide of any one of claims 1 to 31, wherein positions -5, -4 and -3 are 2'-O-alkyl modified; and / or wherein position -2 is 2'-F-modified.

33. The chemically modified oligonucleotide of any one of claims 1 to 32, wherein: (i) the 3' direction of the CBT has at least 4 nucleotides; and / or (ii) The 5′ direction of the CBT has at least 16 nucleotides.

34. A chemically modified oligonucleotide according to any one of claims 1 to 33, comprising (i) at least 10 consecutive internucleoside linkage modifications; and / or (ii) Three consecutive internucleoside linkage modifications at each terminus.

35. A chemically modified oligonucleotide according to any one of claims 1 to 34, wherein the oligonucleotide does not comprise an ADAR recruitment motif of a loop-hairpin structure.

36. A chemically modified oligonucleotide according to any one of claims 1 to 35, wherein the oligonucleotide comprises one or more mismatches, wobble bases and / or bulges, and / or a mismatch at NO relative to its target.

37. A pharmaceutical composition comprising the oligonucleotide according to any one of claims 1 to 36 or a pharmaceutically acceptable salt thereof.

38. The chemically modified oligonucleotide of any one of claims 1 to 36 or the pharmaceutical composition of claim 37 for use in treating or preventing a genetic disorder, condition or disease.

39. The chemically modified oligonucleotide or pharmaceutical composition for use according to claim 38, wherein the genetic disorder, condition or disease is selected from the group consisting of retinitis pigmentosa (RP), Stargardt macular degeneration, age-related macular degeneration (AMD), cystic fibrosis (CF), Hurler syndrome, alpha-1-antitrypsin (A1AT) deficiency, Parkinson's disease, Alzheimer's disease, albinism, amyotrophic lateral sclerosis, asthma, beta-thalassemia, Cadasil syndrome, Charcot-Marie Tooth disease, chronic obstructive pulmonary disease (COPD), distal spinal muscular atrophy (DSMA), Duchenne / Becker muscular dystrophy, dystrophic epidermolysis bullosa, epidermolysis bullosa, Fabry disease, Factor V Leiden mutation (Factor V Leiden mutation), VLeiden-related diseases, familial adenomatous polyposis, galactosemia, Gaucher disease, glucose-6-phosphate dehydrogenase, hemophilia, hereditary hemochromatosis, Hunter syndrome, Huntington disease, inflammatory bowel disease (IBD), hereditary polyagglutination syndromes, Leber congenital amaurosis (LCA), Lesch-Nyhan syndrome, Lynch syndrome, Marfan syndrome, mucopolysaccharidoses, muscular dystrophy, myotonic dystrophy types I and II, neurofibromatosis, Niemann-Pick disease types A, B, and C, NY-eso1-related cancers, Peutz-Jeghers syndrome, phenylketonuria, Pompe disease, primary ciliary disorders, and prothrombin mutation-related diseases.

40. The chemically modified oligonucleotide or the pharmaceutical composition for use according to claim 38 or 39, wherein the genetic disorder, condition or disease is associated with a G to A mutation in a gene selected from the following list: SERPINA1, PDE6A, LRRK2, NLRP3 and CRB1.

41. The chemically modified oligonucleotide or the pharmaceutical composition for use according to claim 40, wherein the mutation is selected from the following list: SERPINA1 E342K, PDE6AV685M, LRRK2 G2019S and CRB1 C948Y.

42. An in vitro method for editing a target adenosine in a target nucleic acid, wherein the method comprises contacting the target nucleic acid with the oligonucleotide of any one of claims 1 to 36.

43. An in vitro method for deaminating at least one specific adenosine present in a target RNA sequence in a cell, wherein the method comprises the following steps: (a) contacting a target nucleic acid with a chemically modified oligonucleotide according to any one of claims 1 to 36; (b) allowing cells to take up the chemically modified oligonucleotide; (c) annealing the chemically modified oligonucleotide to a target RNA sequence; and (d) causing a mammalian ADAR enzyme comprising the native dsRNA binding domain found in the wild-type enzyme to deaminize a target adenosine in the target RNA sequence to inosine.

44. An in vitro method for deaminating at least one specific adenosine present in a target RNA sequence in a cell according to claim 43, wherein the method comprises the step of identifying the presence of inosine in the RNA sequence after step (d).

Images