Oligonucleotide delivery

Abstract

The present invention relates to the fields of medicine and RNA editing, wherein the target adenosine present in the target RNA molecule in a cell is deaminated to inosine by an endogenous ADAR enzyme, and this enzyme is recruited by a double-stranded complex formed between an administered antisense oligonucleotide (EON) that causes RNA editing and a region of the target RNA molecule containing the target adenosine. The present invention relates to the improvement of the delivery of EON to target cells using triterpene glycosides purified from the seeds of Agrostemma githago L.

Description

[Technical Field]

[0001] The present invention relates to the field of medicine, specifically to the delivery of oligonucleotides for the treatment of diseases. The present invention involves the use of certain triterpene glycosides (also called triterpene saponins), which are a type of secondary metabolite primarily produced by plants and consist of one or more linear and / or branched oligosaccharides and a triterpene backbone (aglycone). These triterpene glycosides improve the delivery of therapeutic oligonucleotides and oligonucleotide complexes in vitro and in vivo, preferably for RNA editing of target nucleotides in target transcription molecules. [Background technology]

[0002] Antisense oligonucleotides (AONs, often abbreviated as ASOs) have been developed for use in the treatment of a wide variety of diseases. AONs range from gapmers (generally designed to silence target transcripts), siRNAs, small activating RNAs (saRNAs), steric blockers (compounds that can modulate splicing (e.g., exon skipping-inducing AONs) or translation-blocking AONs), immunostimulatory compounds, decoy oligonucleotides, miRNA inhibitors, and ribozymes to RNA-editing oligonucleotides (often abbreviated as EONs). These can be applied, for example, to specific deamination, specific 2'-O-methylation, or uridine-to-pseudouridine isomerization of specific target nucleotides within target transcripts. AONs are generally short, single-stranded molecules containing synthetic and / or chemically modified RNA and / or DNA nucleotides or their analogs, which can hybridize to specific targets through hydrogen bonding (generally following the Watson-Crick rule). AONs can also form intramolecular structures such as stem-loops or complex with a sense strand to form (hetero)duplex oligonucleotide complexes. Depending on the application, AONs can be covalently or noncovalently linked to functional moieties such as GalNAc structures, especially when delivery to hepatocytes is desired. When AONs are intended to be delivered "naked" without the aid of a vector (such as a viral vector), they are typically chemically modified to enhance their resistance to nucleases, such as RNase H. Commonly used modifications include substituting a 2'-O-methyl group (hereafter referred to as 2'-OMe) or a 2'-O-methoxyethyl group (hereafter referred to as 2'-MOE) at the 2' position of the ribose sugar moiety and replacing the native phosphodiester bond between the two nucleosides with a more resistant phosphorothioate (PS) or methylphosphonate (MP) bond. A wide variety of chemical modifications to the nitrogenous base, ribose sugar, and / or linkages have currently been proposed and are being tested in a wide variety of preclinical and clinical settings.However, the potential toxicity of a selected chemical modification must always be weighed against the dosage required in an in vivo environment to achieve a therapeutically meaningful effect. Thus, there is a need to improve the efficiency of delivering AONs to their specific target cells in vivo and to improve intracellular trafficking so that the AONs can reach and bind to their target transcriptional molecules upon entering the cell. In other words, it is desirable to find means to aid in the delivery of AONs, which may potentially reduce the dosage required to still achieve a therapeutically significant effect, regardless of the mechanism of action of single- or double-stranded AONs.

[0003] One class of compounds that has been investigated for the delivery of AONs is the "saponin" family. These organic compounds, which have the property of foaming when stirred in water, are commonly found in plants such as Quillaja saponaria and soybeans. Saponins are used in dietary supplements, carbonated beverages, and are commonly used as adjuvants in vaccines. Their amphiphilic nature, immunological potential, and diverse biological activities make them suitable adjuvants for drug delivery, and they have also been reported to act as anti-inflammatory, antibacterial, antifungal, antiviral, insecticidal, anticancer, and molluscicidal compounds (see, e.g., Said Ashour A. et al. 2019 J. Nanomed. Res. 1:282-288). Shiri et al. (Shiri E. et al. 2022. J Manzandaran Univ Med Sci. 32:43-54) pointed out that intravenously administered saponin as a pretreatment to rats undergoing ischemia-reperfusion may be effective in reducing damage caused by cerebral ischemia. Wang et al.'s group demonstrated that the use of saponin can promote exon skipping of chemically modified AONs in vitro and in vivo (Wang M et al. 2018a. Mol Ther Nucleic Acids. 11:192-202; Wang M et al. 2018b. Drug Des Devel Ther. 12:3705-3715), indicating that this type of compound may also be applicable to the delivery of oligonucleotides.

[0004] The present invention aims to provide one or more alternative and / or improved techniques, compounds and / or compositions for use in the delivery of (antisense) oligonucleotides and oligonucleotide conjugates. Summary of the Invention

[0005] According to a first aspect, the present invention provides a composition comprising a triterpene glycoside and an antisense oligonucleotide (EON) that induces RNA editing, wherein the EON can form a double-stranded complex with a region of a target RNA molecule in a cell, the region of the target RNA molecule containing a target adenosine, the nucleotide in the EON opposite the target adenosine being a lone nucleotide, and the double-stranded complex can bind to an endogenous ADAR enzyme to deaminate the target adenosine to inosine, thereby editing the target RNA molecule. Preferably, the triterpene glycoside is conjugated to the EON. Preferably, the triterpene glycoside is AG1856. In one embodiment, the EON hybridizes with a fully or partially complementary sense oligonucleotide to form a heteroduplex RNA editing oligonucleotide (HEON) complex. In one embodiment, the target RNA molecule is a pre-mRNA or mRNA target molecule.

[0006] According to a second aspect, the present invention provides a composition according to the present invention for use in treating cardiovascular disease, a disease involving the liver, a disease involving the kidney, a disease involving the pancreas, or a disorder of the central nervous system. In one embodiment, the target RNA molecule is endogenously present in the cell, and the target RNA molecule is transcribed from a human gene selected from the group consisting of SERPINA1, IDUA, HFE, ABCA4, USH2A, PCSK9, B4GALT1, ALDH2, HTT, DMD, PNPLA3, APOC3, C9orf72, DMPK, RHO, MAPT, OTOF, SMN1, ASL, APP, PMP22, LRRK2, ASS1, GJB2, MECP2, and RS1.

[0007] In a third aspect, the present invention provides a kit of components comprising: (i) a first pharmaceutical composition comprising a triterpene glycoside; and (ii) a second pharmaceutical composition comprising an EON, wherein the EON can form a double-stranded complex with a region of a target RNA molecule in a cell, the region of the target RNA molecule comprises a target adenosine, and the double-stranded complex can recruit an endogenous ADAR enzyme to deaminate the target adenosine to inosine, thereby editing the target RNA molecule.Preferably, the triterpene glycoside is AG1856.

[0008] In a fourth aspect, the present invention provides a method for editing a target adenosine present in an endogenous target RNA molecule in a cell in a subject, said method comprising the steps of: (i) administering to the subject a triterpene glycoside; and (ii) administering EON or HEON to said subject, wherein said EON after administration can form a double-stranded complex with the region of said endogenous target RNA molecule that comprises said target adenosine in said cell, and said double-stranded complex can recruit endogenous ADAR enzyme to deaminate said target adenosine into inosine.Preferably, said EON and said triterpene glycoside are conjugated with each other, and thus are administered simultaneously.Preferably, said triterpene glycoside is AG1856.

[0009] In a fifth aspect, the present invention provides a method for deaminating a target adenosine in a target RNA molecule, preferably a pre-mRNA or mRNA molecule, in a cell, said method comprising the steps of: (i) providing to said cells a triterpene glycoside; (ii) providing to the cell an EON capable of forming a double-stranded complex with the target RNA molecule or a region thereof, wherein the region includes the target adenosine; (iii) uptake of the EON by the cells; (iv) annealing the EON to the target RNA molecule; (v) allowing an endogenous ADAR enzyme to deaminate the target adenosine to inosine in the target RNA molecule; and (vi) identifying the presence of the inosine in the target RNA molecule. Preferably, the EON and the triterpene glycoside are conjugated to each other, thereby being simultaneously provided to the cell. Preferably, the triterpene glycoside is AG1856.

[0010] In a sixth aspect, the present invention provides a method for treating a disease in a human subject in need thereof, the method comprising administering to the human subject a therapeutically effective amount of EON and a triterpene glycoside. Preferably, the EON and the triterpene glycoside are conjugated to each other and thereby administered simultaneously. Preferably, the triterpene glycoside is AG1856. Preferably, the disease is a cardiovascular disease, a disease involving the liver, a disease involving the kidney, a disease involving the pancreas, or a disorder of the central nervous system.

[0011] [Brief description of the drawing] One or more embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings in which: [Brief explanation of the drawings]

[0012] [Figure 1] 1 shows the percentage editing of human APP target RNA transcripts in ARPE-19 cells after administration of the triterpene glycoside AG1856 followed by administration of an RNA-editing oligonucleotide (EON). [Figure 2] Figure 2A shows the percentage of editing in the liver, Figure 2B shows the kidney, and Figure 2C shows the percentage of editing in the pancreas of mice treated with AG1856 and then administered EON (RM3835) targeting the mouse APP transcript. The fold change in editing relative to EON administration alone is shown on the right y-axis. [Figure 3]The sequences (5' to 3') of EONs designed to edit the B4GALT1 transcript are shown, with their respective SEQ ID NOs. RM4838 / EON13 is also referred to as B4GALT1-13. RM4830 / EON05 is also referred to as B4GALT1-05. The chemical modifications of EON are as follows: m5Ue is a 2′-MOE-modified 5-methyluridine (identical to 2′-MOE-modified thymidine, i.e., Te); m5Ce is a 2′-MOE-modified 5-methylcytidine; Ae and Ge are 2′-MOE-modified adenosine and guanosine, respectively; Gm, Am, Um, and Cm are 2′-OMe-modified guanosine, adenosine, uridine, and cytidine, respectively; Af, Uf, Gf, and Cf are 2′-F-modified adenosine, uridine, guanosine, and cytosine, respectively; Zd (located at an isolated nucleotide position) is a cytidine analog and is a Benner's base (Benner's base). Also referred to as nucleosides with a 5-methyluridine base (further outlined herein), which have a deoxy moiety (=DNA) at the 2' ribose position; C2f (at the isolated nucleotide position) is a 2',2'-difluoro-modified cytidine; m5Ud (or simply Ud) is a deoxynucleotide with a 5-methyluridine base; Cd (at the isolated nucleotide position) is a deoxycytidine; an asterisk * indicates a PS bond; "!" indicates a PNdmi bond; and "^" indicates a MP bond. All other bonds are phosphodiester bonds. [Figure 4] Percent editing of endogenous B4GALT1 transcripts in human HepG2 cells after treatment with the indicated EONs and saponin (AG1856) at two different concentrations (1 and 5 μM). Negative controls were AG1856 only and untreated cells. [Figure 5] Shown is the percentage editing of endogenous B4GALT1 transcripts in hepatic spheroids generated from primary human hepatocytes after treatment with the four indicated EON+AG1856 (abbreviated here as AG). [Figure 6]A set of EONs (SEQ ID NOS: 46-90, as shown) designed with a GalNAc moiety added to the 5' end of the oligonucleotide are shown. Chemical modifications are as shown in Figure 3. L001 is a triantennary GalNAc moiety (OP-042; Hongene Biotech). L103 is a TEG linker that connects the GalNAc moiety to the first nucleotide at the 5' end. [Figure 7] Figure 6 shows the percentage editing of human B4GALT1 target transcripts in primary human hepatocytes (PHH) after treatment with 5 μM EON in the presence of 1 μM saponin (AG1856), using EON in conjunction with EON01 and EON05 (see Figure 3). Non-treated (NT) samples served as negative controls. [Figure 8] The EONs in Figure 7, along with EON01 and EON05 (see Figure 3), are used to show the percentage editing of human B4GALT1 target transcripts in PHHs after treatment with 5 μM EON in the absence of any saponin, i.e., through gymnotic uptake, as outlined in Figure 7. A non-treated (NT) sample was used as a negative control. [Figure 9] The human ANGPTL3 target RNA sequence (5'→3'; SEQ ID NO: 121) is shown at the top, with the target adenosine in bold. Below the target sequence are the 5'→3' sequences of 30 EONs designed to edit the target adenosines (RM5035 to RM5064 are SEQ ID NOs: 91 to 120, respectively). The chemical modifications in EONs are as follows: t, a, g = 2′-MOE-modified thymidine, adenosine, and guanosine, respectively; c = 2′-MOE-modified 5-methylcytosine; mU, mA, mG, and mC = 2′-OMe-modified uridine, adenosine, guanosine, and cytosine, respectively; fU, fA, fG, and fC = 2′-F-modified uridine, adenosine, guanosine, and cytosine, respectively; dA = deoxyadenosine; dZ = deoxycytidine analog with a Benner base; "!" = PNdmi bond; "*" = phosphorothioate bond; and (MeP) = methylphosphonate bond. [Figure 10]Figure 1 shows the percentage A to I editing of endogenous ANGPTL3 transcripts in human Huh-7 cells after gymnotic (no saponin) exposure to the 30 indicated EONs. Untreated incubations were performed simultaneously as negative controls (mock). [Figure 11] Figure 1 shows the percentage A to I editing of endogenous ANGPTL3 transcripts in human Huh-7 cells after exposure to the same 30 EONs as in Figure 10, but with the addition of 1 μM AG1856 to the culture medium. Untreated incubations and incubations with AG1856 were performed simultaneously as negative controls (AG1856 only and NT). [Figure 12] Figure 1 shows the percentage A to I editing of endogenous ANGPTL3 transcripts in human primary hepatocytes grown into hepatic spheroids after exposure to the same 30 EONs as in Figures 10 and 11, with 5 μM AG1856 added to the culture medium. Two negative controls were used simultaneously (mock and NT). [Figure 13] Figure 12 shows the results of the same experiment, but now spheroids were incubated with 5 μM EON in the absence of saponin using only RM5059 (EON#25), RM5060 (EON#26), RM5061 (EON#27), RM5062 (EON#28), RM5063 (EON#29), and RM5064 (EON#30). [Figure 14] The percentage of adenosine editing in mouse actin B target RNA transcripts is shown in the liver (Figure 14A), kidney (Figure 14B), and spleen (Figure 14C) of mice treated with PBS, PBS+AG1856, RM3891 EON, or RM3891+AG1856. [Figure 15] 1 shows an exemplary conjugate of AG1856 and EON, linked by an ECMH moiety and a C6S linker, where the conjugation is at the 5′ end of the EON. [Figure 16]Figure 15 shows the percentage editing of mApp transcripts in primary mouse hepatocytes after administration of EON alone (RM3835 or RM5522), a conjugate composition in which RM5522 is linked to AG1856 (RM5522@AG1856, shown in Figure 15), and combined administration of RM5522 and AG1856 (unconjugated; indicated by "+"). The purity of the RM5522@AG1856 conjugate, 53%, is shown in parentheses. A non-treated (NT) sample served as a control. Significant increases between different bars are indicated by asterisks. The left y-axis indicates the percentage editing, and the right y-axis indicates the fold change compared to treatment with RM5522 alone. DETAILED DESCRIPTION OF THE INVENTION

[0013] As mentioned above, various types of saponins have been used for many types of applications. Therapeutic uses of saponins have also been described (Weng A et al. 2009. Planta medica 75(13):1421-1422; Weng A et al. 2010. J Chromatography B 878(7):713-718; Weng A et al. 2012. Molecular Oncology 6(3):323-332; Weng A et al. J Controlled Disease 164(1):74-86;Thakur et al. 2014. J Chromatography B 955:1-9;Jia et al. 1998. J Natural Products 61(11):1368-1373;Haddad et al. 2004. Helvetica chimica acta 87(1):73-81;Fu et al. al. J Natural Products 68(5):754-758; Moniuszko-Szajwaj et al. 2016. Helvetica chimica acta 99(5):347-354; Fuchs H et al. 2017. Biomedicines 5(2):14). It has also been described that a specific saponin (SO1861) from soapwort (Saponaria officinalis) can mediate improved intracellular delivery of peptides, lipid nanoparticles, and nucleic acids (Weng A et al. 2015. J Controlled Release 206:75-90; Sama S et al. 2017. Int J Pharmaceutics 534:195-205). WO2019 / 011914 discloses a saponin (GE1741) isolated from Gypsophila elegans, which provides improved efficacy in the delivery of small molecules, such as nucleic acid molecules, to cells (see also Sama S et al. 2018. J Biotechnology 284:131-139).WO2021 / 122998 (and EP3838910B1 derived from its priority application) discloses yet another type of saponin from Agrostemma githago L., which has further improved properties compared to the previously described SO1861 and GE1741 saponins, particularly with regard to toxicity and endosomal escape (see also Clochard J et al. 2020. Int J Pharm 589:119822).

[0014] The present inventors demonstrate herein that the saponins disclosed in WO2021 / 122998 (hereinafter also referred to as triterpene glycosides or triterpene saponins) unexpectedly improve the efficiency of RNA editing caused by a particular type of nucleic acid molecule, namely, RNA-editing oligonucleotides (EONs), under various in vitro and in vivo administration conditions. Particularly surprising, it was found that when triterpene glycosides are (bio)conjugated to EONs, RNA editing is not impeded, and in fact is significantly higher than when the triterpene glycosides are administered separately from the EONs. Surprisingly, it was found that the presence of the saponin did not significantly impede the recruitment of endogenous ADAR enzymes after EONs hybridized to target molecules, and / or that the transport of the EONs through cells toward the target transcript molecules was also not impeded by the binding of the saponin to the EONs. In one embodiment, the present invention relates to the use of triterpene glycosides having a stereoisomeric form according to formula (I): JPEG2026503580000001.jpg130166 where the sugar groups are as follows: Fuc = fucose, Rha = rhamnose, Gal = galactose, Xyl = xylose, Glc = glucose, GlcA = glucuronic acid, and Qui = quinovose. This triterpene glycoside (or triterpene saponin, or simply "saponin") is generally referred to herein as AG1856. It exhibits a structural motif known, for example, from GE1741: a doubly acetylated quinovose residue is linked to a fucose residue, which is directly linked (via an ester bond) to the C-28 chain of quillac acid / gypsogenin. While GE1741 has a linear sugar chain linked to the fucose residue, AG1856 has a branched rhamnose residue with a xylose residue and an acetylated glucose residue.

[0015] RNA editing is a natural process by which eukaryotic cells alter the sequences of their RNA molecules, often in a site-specific and precise manner, thereby expanding the genome-encoded RNA repertoire by several orders of magnitude. RNA editing enzymes have been reported in eukaryotic species throughout the animal and plant kingdoms, and these processes play a critical role in managing cellular homeostasis in metazoans, from the simplest organisms (e.g., Caenorhabditis elegans) to humans. Examples of RNA editing include the conversion of adenosine (A) to inosine (I) and cytidine (C) to uridine (U), which occur through enzymes called adenosine deamidase acting on RNA (ADAR) and APOBEC / AID (cytidine deamidase acting on RNA), respectively.

[0016] ADARs are multidomain proteins containing a catalytic domain and two to three double-stranded RNA recognition domains, depending on the enzyme in question. Each recognition domain recognizes a specific double-stranded RNA (dsRNA) sequence and / or conformation. While the catalytic domain also plays a role in recognizing and binding portions of the dsRNA helix, its primary function is to convert A to I by deaminating a nucleobase at a nearby (predefined) location in the target RNA. Inosine is read as guanosine by the cellular translation machinery, meaning that the edited adenosine can recode a protein sequence if present in the coding region of an mRNA or pre-mRNA. A to I conversion can also occur in the 5' non-coding sequence of the target mRNA, generating a new translation start site upstream of the original start site, which generates an N-terminally extended protein, or it can occur in the 3' UTR or other non-coding portion of the transcript, which can affect RNA processing and / or stability. Additionally, A-to-I conversion can occur at splice elements within introns or exons of pre-mRNA, thereby altering the splicing pattern, resulting in either inclusion or skipping of the exon. Enzymes that catalyze adenosine deamination belong to the ADAR enzyme family, which includes the human deaminases hADAR1, hADAR2, and hADAR3. However, hADAR3 has not demonstrated deaminating activity.

[0017] The use of oligonucleotides to apply adenosine deaminase to edit target RNA has been described (e.g., Woolf et al. 1995. PNAS 92:8298-8302; Montiel-Gonzalez et al. PNAS 2013, 110(45):18285-18290; Vogel et al. 2014. Angewandte Chemie Int Ed 53:6267-6271). A drawback of the method described in Montiel-Gonzalez et al. (2013) is that it requires a fusion protein consisting of a genetic fusion of the boxB recognition domain of the bacteriophage lambda N protein with the adenosine deaminase domain of a truncated native ADAR protein. This method requires target cells to be transduced with the fusion protein (a major obstacle) or transfected with a nucleic acid construct encoding the modified adenosine deaminase fusion protein for expression. The system described by Vogel et al. (2014) suffers from a similar drawback: it is unclear how to apply this system without first genetically modifying ADAR and then transfecting or transforming cells harboring the target RNA to deliver the genetically modified protein. A similar system is also described in US 9,650,627. Woolf et al.'s (1995) oligonucleotides were 100% complementary to the target RNA sequence but suffered from a severe lack of specificity: nearly every adenosine in the target RNA strand complementary to the antisense oligonucleotide was edited.

[0018] ADARs are known to act on any dsRNA. Through a process sometimes referred to as "promiscuous editing," these enzymes edit multiple A residues within dsRNA. Therefore, methods and tools were needed to circumvent such promiscuous editing and target specific adenosines within target RNA molecules for therapeutic applications. Vogel et al. (2014) demonstrated that such off-target editing can be suppressed by using 2'-O-methyl (2'-OMe)-modified nucleosides in oligonucleotides opposite adenosines that should not be edited, and by using unmodified nucleosides in oligonucleotides directly opposite the specifically targeted adenosines on the target RNA. However, specific editing at the target nucleotide has not been demonstrated without the use of recombinant ADAR enzymes covalently linked to AONs. Several publications have shown that it is possible to recruit endogenous ADAR (thus not requiring exogenous and / or recombinant sources) while maintaining the specificity of targeting a single adenosine in a target RNA molecule and deaminating it to inosine. WO2016 / 097212 discloses an antisense oligonucleotide (AON) for targeted editing of RNA, which is characterized by a sequence complementary to the target RNA sequence (referred to as "targeting portion" in the publication) and the presence of a stem-loop / hairpin structure (referred to as "recruitment portion" in the publication), which is preferably non-complementary to the target RNA. Such oligonucleotides are referred to as "self-looping AONs". The recruitment portion acts to recruit the natural ADAR enzyme present in cells to the dsRNA formed by hybridization of the target sequence with the targeting portion. The presence of the recruitment moiety does not require the presence of the conjugate or modified recombinant ADAR enzyme.WO2016 / 097212 describes the recruitment moiety as a stem-loop structure that mimics a natural substrate (e.g., GluB receptor) or a Z-DNA structure (known to be recognized by the dsRNA-binding domain or Z-DNA-binding domain of ADAR enzymes). The stem-loop structure can be an intermolecular stem-loop structure formed by two separate nucleic acid strands, or an intramolecular stem-loop structure formed within a single nucleic acid strand. The stem-loop structure of the recruitment moiety is an intramolecular stem-loop structure formed within the AON itself, which is thought to attract (endogenous) ADAR. Systems containing similar stem-loop structures for RNA editing are described in WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560, and WO2022 / 078995.

[0019] WO2017 / 220751 and WO2018 / 041973 describe next-generation AONs, which do not contain such stem-loop structures but are (almost perfectly) complementary to target regions.In one embodiment, one or more mismatched nucleotides, wobbles, or bulges exist between oligonucleotides and target sequences.Sometimes, mismatches exist only at the nucleoside site opposite the target adenosine, while in other embodiments, AONs (or RNA editing oligonucleotides, abbreviated as "EONs") have multiple bulges and / or wobbles when added to target sequence regions.If the sequence of EONs is carefully selected to attract / recruit ADAR, it appears that it is possible to achieve RNA editing in vitro, ex vivo, and in vivo using EONs that lack stem-loop structures that are not complementary to target sequences and endogenous ADAR enzymes. An "orphan nucleoside" or "orphan nucleotide" is defined as a nucleoside / nucleotide within an EON that is located directly opposite a target adenosine in a target RNA molecule but does not have a 2'-OMe modification. The orphan nucleoside may be a deoxyribonucleoside (DNA), and the remainder of the EON may still have a 2'-O-alkyl modification (e.g., 2'-OMe) in the sugar moiety, or the nucleotides immediately surrounding the orphan nucleoside may contain chemical modifications (e.g., DNA compared to RNA), resulting in further improved RNA editing efficiency and / or increased resistance to nucleases. Such effects may be further improved by the use of sense oligonucleotides (SONs) that "protect" the EON from degradation (as described in WO 2018 / 134301).The use of chemical modifications and specific structures in oligonucleotides that can be used for ADAR-mediated editing of specific adenosines in target RNA has been the subject of numerous publications in the field, for example, WO2019 / 111957, WO2019 / 158475, WO2020 / 165077, WO2020 / 201406, WO2020 / 211780, WO2021 / 008447, WO2 021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, WO2022 / 007803, WO2022 / 018207, WO2022 / 026928, and WO2022 / 124345. The use of certain sugar moieties is disclosed, for example, in WO2020 / 154342, WO2020 / 154343, WO2020 / 154344, WO2022 / 103839, and WO2022 / 103852.On the other hand, the use of sterically defined linker moieties (generally for oligonucleotides that can be used, for example, for exon skipping, gapmers, siRNA, or specifically for RNA editing oligonucleotides related to a wide variety of target sequences) has been described in WO2011 / 005761, WO2014 / 010250, WO2014 / 012081, WO2015 / 107425, WO2017 / 015575(HTT), WO2017 / 062862, WO2017 / 160741, WO2017 / 192664, WO2017 / 192679(DMD), WO2017 / 198775 , WO2017 / 210647, WO2018 / 067973, WO2018 / 098264, WO2018 / 223056(PNPLA3), WO2018 / 223073(APOC3), WO2018 / 223081(PNPLA3), WO2018 / 237194, WO2019 / 032607(C9orf72), WO2019 / 055951, WO2019 / 075357(SMA / ALS), WO20 19 / 200185(DM1), WO2019 / 217784(DM1), WO2019 / 219581, WO2020 / 118246(DM1), WO2020 / 160336(HTT), WO2020 / 191252, WO2020 / 196662, WO2020 / 219981 (USH2A), WO2020 / 219983 (RHO), WO2020 / 227691 (C9orf72), WO2021 / 071788 (C9orf72), WO2021 / 071858, WO2021 / 178237 (MAPT), WO2021 / 234459, WO2021 / 237223, and WO2022 / 099159. In addition to these disclosures, there are numerous publications relating to targeting specific RNA target molecules, or specific adenosines within such RNA target molecules, to repair mutations that result in premature stop codons or other disease-causing mutations.Examples of such disclosures targeting adenosines within specific RNA target molecules include: WO2020 / 157008 and WO2021 / 136404 (USH2A); WO2021 / 113270 (APP); WO2021 / 113390 (CMT1A); WO2021 / 209010 (IDUA, Hurler syndrome); WO2021 / 231673 and WO2021 / 242903 (LRRK2); WO2021 / 231675 (ASS1); WO2021 / 23 1679 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 and WO2021 / 231692 (OTOF, autosomal recessive non-syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); WO2021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1).

[0020] In a first aspect, the present invention provides a composition comprising a triterpene glycoside and an EON that causes RNA editing, wherein the EON can form a double-stranded complex with a region of a target RNA molecule in a cell, the region of the target RNA molecule comprises a target adenosine, and the double-stranded complex can recruit an endogenous ADAR enzyme to deaminate the target adenosine into inosine, thereby editing the target RNA molecule.Thus, although the EON itself may be referred to as an RNA editing oligonucleotide, it does not deaminate the target adenosine by itself, but can cause (produce, induce, produce) the deamination of the target adenosine by hybridizing and binding to the target RNA molecule that comprises the target adenosine or its complementary region, and by generating this double-stranded structure, the deamination enzyme that can then deaminate the target adenosine into inosine can be recruited. A preferred triterpene glycoside for use in all aspects of the present invention is AG1856, which can be purified from seeds of Agrostemma githago L. (Agrostemma githago L.). A preferred triterpene glycoside for use in all aspects of the present invention is shown in Formula (I). To improve the intracellular entry efficiency and / or stability of EON, it is preferred that the composition hybridize with a fully or partially complementary sense oligonucleotide to form an oligonucleotide (HEON) complex that generates heteroduplex RNA editing. A preferred HEON complex is described in GB2215614.5 (unpublished). To further improve editing efficiency, the triterpene glycoside is preferably attached to the EON via conjugation. Preferably, the triterpene glycoside is attached to the 5'-end of the EON, although it is not excluded that the triterpene glycoside can also be attached to the 3'-end of the EON. Those skilled in the art are aware of the manufacturing possibilities for conjugating saponins to either the 5'- or 3'-end, and the preferred linkers required for appropriate conjugation.In one embodiment, the nucleotide in the EON opposite the target adenosine is cytidine, a cytidine analog, a cytidine derivative, a uridine, a uridine analog, or a uridine derivative. In one embodiment, at least one nucleotide in the EON contains one or more non-natural chemical modifications in the ribose moiety, the linker moiety, or the base moiety, provided that the nucleotide in the EON opposite the target adenosine is not a cytidine containing a 2'-OMe ribose substitution. In all aspects of the present invention, the target RNA molecule is preferably pre-mRNA or mRNA. However, RNA editing according to any aspect of the present invention can also be applied to tRNA, rRNA, and vRNA, as needed. In one embodiment, the endogenous ADAR enzyme is human ADAR1, ADAR2, or ADAT. In one embodiment, the composition according to the present invention is for use in treating cardiovascular disease, liver-related diseases, kidney-related diseases, pancreatic diseases, or central nervous system disorders. There are a wide variety of diseases, which can be caused by a variety of genetic alterations that do not necessarily affect only specific types of tissues. This means that the present invention also relates to targeting target RNA molecules involved in diseases that can have adverse effects on multiple tissues. In another embodiment, the present invention also relates to the deamination of target adenosines present in wild-type transcript molecules, where the deamination of the target adenosines results in a translation product that alleviates, improves, or prevents a specific disease or disorder. These include the introduction of loss-of-function alterations and gain-of-function alterations, for example, by removing or introducing phosphorylation or glycosylation sites.In one embodiment, the target RNA molecule is endogenously present in the cell, and the target RNA molecule is transcribed from a human gene selected from the group consisting of: SERPINA1, IDUA, HFE, ABCA4, USH2A, PCSK9, B4GALT1, NTCP, ALDH2, HTT, DMD, PNPLA3, APOC3, C9orf72, DMPK, RHO, MAPT, OTOF, SMN1, ASL, APP, PMP22, LRRK2, ASS1, GJB2, MECP2, and RS1.

[0021] In a second aspect, the present invention provides a kit of components comprising: (i) a first pharmaceutical composition comprising a triterpene glycoside; and (ii) a second pharmaceutical composition comprising an EON, wherein the EON is capable of forming a double-stranded complex with a region of a target RNA molecule in a cell, the region of the target RNA molecule comprising a target adenosine, and the double-stranded complex is capable of recruiting an endogenous ADAR enzyme to deaminate the target adenosine to inosine, thereby editing the target RNA molecule. The triterpene glycoside and EON may be administered separately in the dosing regimen, for example, the triterpene glycoside is administered 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, or 31 days before or after the EON is administered, or the triterpene glycoside is administered 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 weeks before or after the EON is administered, or even several months apart. Of note, the triterpene glycoside and EON may be administered separately on the same day but in separate administrations, with no time between each other, or the administrations may be separated by, for example, 1, 2, 3, 4, 5, 6, 7, or 8 hours. In a preferred embodiment, the triterpene glycoside and EON are administered simultaneously, and preferably conjugated to each other, meaning that the EON-saponin forms a single component in the kit of components. In one embodiment, the triterpene glycoside present in the kit of components is AG1856. In one embodiment, the EON in the second pharmaceutical composition hybridizes with a fully or partially complementary sense oligonucleotide to form a HEON. The HEON may also be present separately in the kit of components and hybridize with the EON (or EON-saponin conjugate) immediately before administration. In one embodiment, the nucleotide in the EON opposite the target adenosine is cytidine, a cytidine analog, a cytidine derivative, a uridine, a uridine analog, or a uridine derivative.In one embodiment, at least one nucleotide in the EON contains one or more non-natural chemical modifications in the ribose moiety, linkage moiety, or base moiety, provided that the nucleotide in the EON opposite the target adenosine is not a cytidine containing a 2'-OMe ribose substitution. In one embodiment, the target RNA molecule is pre-mRNA or mRNA. In one embodiment, the endogenous ADAR enzyme is human ADAR1, ADAR2, or ADAT. In one embodiment, the kit of components is for use in treating cardiovascular disease, liver-related disease, kidney-related disease, pancreatic disease, or central nervous system disorders. In one embodiment, the target RNA molecule is endogenously present in the cell, and the target RNA molecule is transcribed from a human gene selected from the group consisting of: SERPINA1, IDUA, HFE, ABCA4, USH2A, PCSK9, B4GALT1, NTCP, ALDH2, HTT, DMD, PNPLA3, APOC3, C9orf72, DMPK, RHO, MAPT, OTOF, SMN1, ASL, APP, PMP22, LRRK2, ASS1, GJB2, MECP2, and RS1.

[0022] In a third aspect, the present invention provides a method for editing a target adenosine present in an endogenous target RNA molecule in a cell in a subject, said method comprising the steps of: (i) administering to said subject a triterpene glycoside, preferably AG1856; and Administer EON or HEON to the subject, wherein the EON after administration can form a double-stranded complex with the region of the endogenous target RNA molecule containing the target adenosine in the cell, and the double-stranded complex can recruit endogenous ADAR enzyme to deaminate the target adenosine into inosine.Preferably, EON, triterpene glycoside and EON are administered simultaneously, preferably as an EON-triterpene glycoside conjugate composition.In one embodiment, the cell is hepatocyte (preferably hepatocyte), kidney cell, or neuron (preferably CNS neuron).In one embodiment, the EON-triterpene glycoside conjugate is for delivery to hepatocyte, and EON also has a GalNAc moiety conjugated to the opposite side of triterpene glycoside.

[0023] In a fourth aspect, the present invention provides a method for deaminating a target adenosine in a target RNA molecule, preferably a pre-mRNA or mRNA molecule, in a cell, said method comprising the steps of: (i) providing to said cells a triterpene glycoside, preferably AG1856; (ii) providing to the cell an EON capable of forming a double-stranded complex with the target RNA molecule or a region thereof, wherein the region includes the target adenosine; (iii) uptake of the EON by the cells; (iv) annealing the EON to the target RNA molecule; (v) allowing an endogenous ADAR enzyme to deaminate the target adenosine to inosine in the target RNA molecule; and (vi) identifying the presence of the inosine in the target RNA molecule. The triterpene glycoside does not necessarily have to be provided before the EON is provided; these steps (i) and (ii) can be performed in reverse or simultaneously, for example, when the EON and triterpene glycoside are present in a single composition, or when they are conjugated to each other. In one embodiment, the EON is hybridized with a fully or partially complementary sense oligonucleotide to form a HEON before forming a double-stranded complex with the target RNA molecule. The HEON may also be present in the same composition as the triterpene glycoside. In yet another embodiment, the saponin is conjugated to a sense strand, and the sense strand is hybridized with the EON through Watson-Crick base pairing.

[0024] In a fifth aspect, the present invention provides a method of treating a disease in a human subject in need thereof, the method comprising administering to the human subject a therapeutically effective amount of EON and a triterpene glycoside, the triterpene glycoside preferably being AG1856. Preferred diseases to be treated with the compositions of the present invention (or kits of components, if EON and triterpene glycoside are administered separately) are cardiovascular diseases, diseases involving the liver, diseases involving the kidneys, diseases involving the pancreas, or disorders of the central nervous system. However, other disorders are not expressly excluded, and those skilled in the art will understand that in cases where a particular disease is caused by a genetic abnormality affecting multiple organs and / or tissues, various organs and / or tissues may be treated by administration of EON and a triterpene glycoside according to the present invention.

[0025] In a preferred embodiment, the EON of the composition of the present invention is administered as a single-stranded oligonucleotide comprising an isolated nucleotide opposite the target adenosine, the isolated nucleotide being chemically modified as disclosed herein, and the remainder of the oligonucleotide being chemically modified as disclosed herein to prevent degradation by nucleases; however, in another embodiment, the present invention relates to any kind of oligonucleotide or heteroduplex oligonucleotide complex, which may or may not be linked to a hairpin structure (internal or at terminal(s)), may be linked to ADAR or its catalytic domain, or the oligonucleotide may be cyclic. It should be understood that any kind of oligonucleotide-based RNA editing is encompassed by the present invention when it is related to nucleotide deamination and uses triterpene glycosides (or triterpene saponins) (preferably AG1856) for delivery. In preferred embodiments, the EONs of the present invention are "naked" oligonucleotides (meaning that they are not delivered by a vector "means," such as a virus or plasmid encoding the EON), contain various chemical modifications to the ribose sugar, base, and / or internucleoside linkage of one or more nucleotides within the sequence, and are capable of recruiting endogenous ADARs for deamination of target adenosines.

[0026] In another preferred embodiment, the EON of the composition of the present invention is hybridized with a sense oligonucleotide to form a heteroduplex RNA-editing oligonucleotide complex, or HEON for short, where the sense oligonucleotide may be the same length, shorter, or longer than the antisense oligonucleotide, and the sense strand may be fully or partially complementary to the antisense oligonucleotide. In such a complex, the partially complementary antisense and sense oligonucleotides may also be referred to as a partial HEON, or pHEON (see GB2215614.5 (unpublished)). In one embodiment, the triterpene glycoside administered separately or together (preferably conjugated) with the antisense oligonucleotide may therefore also be combined with HEON or pHEON in another embodiment.

[0027] In one embodiment, the triterpene glycoside is attached to the sense strand in the EON or HEON at either the 5' or 3' end, i.e., conjugated. Preferably, the saponin is conjugated to the 5' end of the EON, preferably by using a linker. Conjugation can be achieved in a variety of ways using conjugation methods known to those skilled in the art and various linker moieties well known to those skilled in the art. Conjugation (also referred to as "bioconjugation" because of its biomedical relevance) means that two moieties (EON and saponin) are bound to each other before and during administration, whether in vitro in cell culture, in vivo during administration to mice or non-human primates, or in human clinical trials. Conjugation between EON and saponin is generally performed in a laboratory environment or in a manufacturing facility, since EONs are also manufactured in a laboratory environment, while saponins are generally purified from natural sources. The binding may be irreversible or reversible. It should be noted that when a targeting moiety such as GalNAc is used, GalNAc and triterpene glycosides (as well as tocopherol or cholesterol, or their analogs, as disclosed in GB 2215614.5) can all be attached to the EON, to the sense strand, at the 5'-end, and / or at the 3'-end of each oligonucleotide strand. In a preferred embodiment, when delivery to liver cells (particularly hepatocytes) is desired, the saponin is conjugated to the 5'-end of the EON, while the GalNAc moiety is attached to the 3'-end of the EON. Those skilled in the art can attach each of these additional groups to any position within the EON and / or sense strand (if it is an HEON conjugate), depending on the route of administration, the disease to be treated, the target sequence, etc.

[0028] The term "nucleoside" refers to a nucleic acid base linked to a (deoxy)ribose without a phosphate group. A "nucleotide" is composed of a nucleoside and one or more phosphate groups. The term "nucleotide" therefore refers to the respective nucleic acid base-(deoxy)ribose-phospholinker and any chemical modification of the ribose moiety or phospho group. The term therefore includes: nucleotides containing locked ribose moieties (including 2'-4' bridges containing methylene groups or any other groups), unlocked nucleic acids (UNA), threose nucleic acids (TNA), nucleotides containing linkers containing phosphodiesters, phosphonoacetates, phosphotriesters, phosphorothioates (PS), phosphoro(di)thioates, methylphosphonates (MP), methylthiophosphonates, phosphoramidates, phosphorylguanidine bonds, etc. The terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, and inosine and hypoxanthine are sometimes used interchangeably to refer to the corresponding nucleobases on the one hand and the nucleosides or nucleotides on the other hand. Thymine (T) is a nucleotide that is substituted with 5-methyluracil (m 5 Also known as 5-methyluridine, thymine is a derivative of uridine; thymine, 5-methyluridine, and uridine are used interchangeably throughout the document. The terms nucleobase, nucleoside, and nucleotide are sometimes used interchangeably unless the context clearly requires otherwise (e.g., when a nucleoside is linked to adjacent nucleosides and the bond between these nucleosides is modified). As used herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms "ribonucleoside" and "deoxyribonucleoside," or "ribose" and "deoxyribose," are used as they are used in the art.

[0029] The term "saponin" has its ordinary scientific meaning and is used herein to refer to a group of amphipathic glycosides that contain one or more hydrophilic glycone moieties in combination with a hydrophobic aglycone core that is a sapogenin. Saponins may be naturally occurring, derived from natural sources, synthetically produced, or non-naturally occurring. The term "saponin" includes naturally occurring saponins, derivatives of natural saponins, and saponins synthesized de novo through chemical and / or biotechnological synthetic routes.

[0030] The term "saponin derivative" has its ordinary scientific meaning and, as used herein, refers to a modified saponin that has a chemical modification at the position where an aldehyde group previously existed in the saponin, i.e., underivatized saponin, prior to chemical modification to provide the saponin derivative. For example, a saponin derivative is provided by chemically modifying an aldehyde group in the saponin on which the saponin derivative is based. That is, a saponin is provided and its aldehyde group is chemically modified to provide the saponin derivative. For example, the saponin that is derivatized to provide the saponin derivative is a naturally occurring saponin. Typically, a saponin derivative is a synthetic saponin, and typically, a saponin derivative is a derivatization of a natural saponin and is thus derived from a natural saponin, although a saponin derivative may also be derived from a synthetic saponin that may or may not have a natural counterpart. Typically, a saponin derivative does not have a natural counterpart; that is, a saponin derivative is not naturally produced, for example, by a plant or tree. Optionally, the saponin derivative may further comprise one or more chemical modifications at positions where any of the carboxyl, carboxylic acid, acetate, and / or acetyl groups previously existed in the underivatized or derivatized saponin prior to chemical modification to provide the saponin derivative. For example, a saponin derivative may be provided in which the aldehyde, carboxyl, carboxylic acid, acetate, and / or acetyl groups are chemically modified to provide the saponin derivative.

[0031] The term "mono-desmosidic saponin" has its ordinary scientific meaning and, as used herein, refers to a triterpenoid, steroid, or alkaloid saponin that contains one sugar chain attached to an aglycone core, said sugar chain consisting of one or more sugar moieties.

[0032] The term "bis-desmosidic saponin" has its ordinary scientific meaning and, as used herein, refers to a triterpenoid, steroid, or alkaloid saponin that contains two sugar chains attached to an aglycone core, each of which consists of one or more sugar moieties.

[0033] The term "triterpenoid saponin" has its ordinary scientific meaning and is used herein to refer to saponins having a triterpenoid-type aglycone core structure. Triterpenoid saponins differ from steroidal glycoside-based saponins, such as sapogenol, in that steroidal glycoside-containing saponins have a steroidal core structure, and triterpenoid saponins differ from alkaloidal glycoside-based saponins, such as tomatidine, in that alkaloidal glycoside-containing saponins have an alkaloidal core structure.

[0034] In certain embodiments, the saponin derivative comprises an aglycone core structure selected from the group consisting of: - 2α-hydroxyoleanolic acid; - 16α-hydroxyoleanolic acid; - Hederagenin (23-hydroxyoleanolic acid); - 16α,23-dihydroxy(digudroxy)oleanolic acid; - gypsogenin; - Quillaic acid; - Protoaescigenin-21(2-methylbut-2-enoate)-22-acetate; - 23-oxo-barringtogenol C-21,22-bis(2-methylbut-2-enoate); - 23-oxo-vallingtogenol C-21(2-methylbut-2-enoate)-16,22-diacetate; - digitogenin; - 3,16,28-trihydroxyoleanan-12-ene; - gypsogenic acid; and - their derivatives.

[0035] Preferably, the saponin derivative comprises an aglycone core structure selected from quillaric acid and gypsogenin, or derivatives thereof, and more preferably, the aglycone core structure of the saponin derivative is quillaric acid or a derivative thereof.

[0036] Conjugates of EONs and saponins can take many forms and structures, the main feature being that at least one saponin and at least one EON (or sense strand of an EON) are covalently linked to each other. Thus, the general structure of the conjugate compositions of the present disclosure can be depicted as follows: XYZ, X-(Y) n -Z, XY(-Z) n , X-(YZ) n , (X) n -YZ, (XY) n -Z, (X) n -Y-(Z) n , (XY) n -(Z) n , (X) n -(YZ) n ,or (X) n -(Y) n -(Z) n where X refers to an EON linked to one or more linkers at the same or multiple sites; Y is one or more linkers, which may be the same type or different types as appropriate; Z is a saponin or a cluster of saponins (e.g., a dendrimer structure containing multiple saponins and then linked to a linker); and n=2-10. Thus, a conjugate composition containing any of such structures may consist of a single EON and a single saponin, or multiple EONs and a single saponin, or multiple saponins and a single EON, or multiple EONs and multiple saponins. The EON and saponin are linked to each other by one or more linkers; a single saponin is understood to be linked to a single EON through one or more linkers, while multiple saponins may be linked to each other first as a cluster (e.g., a dendrimer) and then linked to one or more linkers, or may be linked to the same linker; the linker essentially functions as part of the cluster. Thus, the isolated linker has at least two reactive sites for linking at least one EON (or another linker that will ultimately link to at least one EON) to at least one saponin (or another linker that will ultimately link to at least one saponin).

[0037] A single linker can be classified as a stable or unstable linker. A stable linker has much greater metabolic, chemical, or biological stability than an unstable linker, and is understood to be introduced without the intention of metabolically, chemically, or biologically cleaving the linker. Examples of stable linkers include alkyl chains, amides, thioethers (e.g., thiol / ene), and "click" bonds introduced by, for example, azide / (constrained)alkyne or triazine / (constrained)alkyne reactions. Examples of unstable linkers include those that are cleaved under certain pH (e.g., oximes, hydrazones, acetals, carbonates, silyl ethers, semicarbazones), in reducing environments (e.g., disulfides), or under enzymatic conditions (e.g., phosphate ester hydrolysis, cathepsin cleavage, endonuclease cleavage). An unstable linker may also further include self-immolative moieties. It is understood that the classification of a linker as stable or unstable may vary for different purposes and associated time scales: for example, a linker half-life of 40 hours may be classified as stable for applications requiring linker cleavage in the minute time range, while it may be classified as unstable for applications requiring linker cleavage in the week time range.

[0038] A conjugate may include more than a single linker connecting the EON(s) and the saponin(s), and both linkers may be of the same type or different types. For example, a conjugate may have the following structure: EON-Linker1-Linker2-Saponin, where Linker1 is a C6 thio linker and Linker2 is an EMCH linker, as shown in formula (II): JPEG2026503580000002.jpg30166

[0039] A linker has at least two reactive groups to connect two chemical moieties to each other, thus allowing covalent attachment to these moieties. The moiety linking these two or more reactive groups can be any chemical moiety known to those skilled in the art. For example, such a linking moiety can be a (branched) alkyl chain, or a diol, diamine, or disulfide thereof, polyethylene glycol, or triethylene glycol, or hexaethylene glycol.

[0040] Many linkers and combinations thereof are known to those skilled in the art and can be applied to produce conjugates of (at least one) EON and (at least one) saponin.

[0041] In the conjugates of (at least one) EON and (at least one) saponin disclosed herein, the linker connecting the saponin can be covalently attached to the aldehyde moiety, carboxylic acid moiety, or primary or secondary alcohol of the saponin; or a mixture thereof. Alternatively, the saponin can be chemically modified at another position on either the carbohydrate moiety (including the carbon atom and acetyl group) or the aglycone (including the carbon atom and, if present, the double bond) to allow attachment to the linker. Thus, the (at least one) saponin linked to the (at least one) linker is linked to the (at least one) EON, for example, at its 3' end, 5' end, 1' position, 2' position, 3' position, 4' position, 5' position, a secondary amine in amino-LNA, at a nucleobase (e.g., at the 5 position of a pyrimidine base or the N7 position of a purine base), by a bond (e.g., via a phosphoramidate bond, or a phosphotriester), or through a non-natural nucleotide analog, such as an unlocked nucleic acid (UNA) monomer, or other branching element in the EON, to allow addition.

[0042] The EON and saponin conjugate compositions disclosed herein may further comprise other moieties covalently attached via one or more of the stable or unstable linkers described above. Examples of such moieties include carbohydrates (e.g., GalNAc, Glc, GalN, GlcN, or clusters thereof), lipids, vitamins, small molecules, drugs, peptides, and antibodies, which may confer additional tissue or cell targeting, general or specific cellular uptake, endosomal escape, protein binding, and / or intracellular trafficking properties to the conjugate.

[0043] In one embodiment, a kit of components is provided, in which a first formulation contains EON conjugated with a GalNAc moiety (for delivery to hepatocytes), and a second formulation contains a triterpene glycoside also conjugated with GalNAc (also for delivery to hepatocytes). Both conjugates then translocate to the liver, where the triterpene glycoside can subsequently contribute to the entry and endosomal release of EON, which has also translocated to hepatocytes by its GalNAc conjugation.

[0044] The conjugates of EONs and triterpene glycosides disclosed herein can be further formulated as nanoparticles, for example, lipid, metal, carbon, ceramic, or polymer-based nanoparticles.

[0045] The conjugates of EONs and triterpene glycosides disclosed herein may be used as the sole active compound or, if desired, may be combined with unconjugated EONs in a single composition to reduce total saponin exposure.

[0046] The term "linker" has its ordinary scientific meaning, and linkers are well known in the art of bioconjugation. As used herein, the term linker refers to a chemical moiety or a linear, contiguous region of amino acid residues conjugated through peptide bonds, which is suitable for covalently attaching (binding) a first molecule (e.g., a triterpene saponin disclosed herein) to another molecule (e.g., an EON or (p)HEON oligonucleotide disclosed herein) or to a scaffold (e.g., composed of or including amino acid residues, nucleic acids, etc.). Typically, a linker comprises a chain of atoms linked by chemical bonds. Any linker molecule or linker technology known in the art can be used in the compositions of the present disclosure. The linker is preferably one for covalently linking molecules through a chemical group on the molecule suitable for forming a covalent bond or bond with the linker. Examples of linkers are disclosed in WO2022 / 164316, a publication that also discloses several saponins and their derivatives that can be used in combination with the EONs or (p)HEONs disclosed herein. Preferred linkers are KMUH, EMCH, BMPH, and maleimide-PEG2-hydrazide.

[0047] Whenever oligonucleotides, oligos, ONs, ASOs, oligonucleotide compositions, antisense oligonucleotides, AONs, (RNA) editing oligonucleotides, EONs, and RNA (antisense) oligonucleotides are mentioned, unless the context dictates otherwise, to refer to both oligoribonucleotides and deoxyoligoribonucleotides. In some cases, oligonucleotides may be completely devoid of RNA or DNA nucleotides (as they naturally occur) or may be composed entirely of modified nucleotides. Whenever "oligoribonucleotides" are mentioned, they may contain the bases A, G, C, U, or I. Whenever "deoxyoligoribonucleotides" are mentioned, they may contain the bases A, G, C, T, or I. However, oligonucleotides of the present invention may contain a mixture of ribonucleosides and deoxyribonucleosides. When deoxyribonucleotides are used, i.e., when there is no modification at the 2' position of the sugar, the nucleotide is often abbreviated as dA, dC, dG, or T, where "d" denotes the deoxy nature of the nucleoside. On the other hand, ribonucleosides that are normal RNA or modified at the 2' position are often abbreviated without the "d" and often abbreviated with the respective modification as described herein.

[0048] Whenever nucleotide is mentioned in oligonucleotide, for example, cytosine includes 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5-hydroxycytosine, and β-D-glucosyl-5-hydroxymethylcytosine.Whenever adenine is mentioned, it includes N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine, and 7-methyladenine.Whenever uracil is mentioned, it includes dihydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1-methylpseudouracil, 4-thiouracil, and 5-hydroxymethyluracil.Whenever guanine is mentioned, it includes 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7-trimethylguanosine, and N2,7-dimethylguanosine. Whenever a nucleoside or nucleotide is referred to, ribofuranose derivatives such as 2'-deoxy, 2'-hydroxy, and 2'-O-substituted variants (e.g., 2'-OMe) are included, as well as other modifications, including 2'-4' bridged variants. Whenever an oligonucleotide is referred to, the linkage between two mononucleotides may be a phosphodiester bond, and modifications thereof include phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP, phosphoramidate linker, phosphorylguanidine, thiophosphorylguanidine, sulfonophosphoamidate, and the like.

[0049] The term "comprising" encompasses "including" and "consisting of," e.g., a composition "comprising X" may consist solely of X or may include something additional (e.g., X+Y). The term "about" in reference to a numerical value x is optional, and means, for example, x±10%.

[0050] The word "substantially" does not exclude "completely", for example a composition that is "substantially free of Y" may be completely free of Y. Where relevant, the word "substantially" may be omitted from the definition of the invention.

[0051] The term "complementary" as used herein refers to the fact that an EON hybridizes with a second nucleic acid strand under physiological conditions (e.g., when an oligonucleotide, as a first nucleic acid strand (=guide oligonucleotide), forms a heteroduplex RNA-editing oligonucleotide complex (i.e., HEON) with another complementary nucleic acid strand, or when it forms a double-stranded complex with a target RNA sequence). This term does not necessarily mean that each nucleotide in a nucleic acid strand perfectly pairs with the opposite nucleotide in the opposite sequence. In other words, an EON can be complementary to a target sequence, but there may be mismatches, wobble, and / or bulges between the oligonucleotide and the target sequence, while the EON still hybridizes with the target sequence under physiological conditions, allowing RNA-editing enzymes in cells to edit the target adenosine. Thus, the term "substantially complementary" also means that the EON has sufficient nucleotide matches between the EON and the target sequence such that the EON can hybridize to the target RNA under physiological conditions, even if mismatches, wobble, and / or bulges are present. As provided herein, an EON can be complementary but contain one or more mismatches, wobble, and / or bulge with the target sequence, provided that the EON can hybridize to its target under physiological conditions.

[0052] The term "downstream," with respect to a nucleic acid sequence, means further along the sequence in the 3' direction; the term "upstream" means the opposite. Thus, in any sequence that encodes a polypeptide, the start codon is upstream of the stop codon on the sense strand, but downstream of the stop codon on the antisense strand.

[0053] References to "hybridization" typically refer to specific hybridization and exclude non-specific hybridization, which may occur under selected experimental conditions using techniques well known in the art to ensure that the most stable interaction between probe and target is when the probe and target have at least 70%, preferably at least 80%, and more preferably at least 90% sequence identity.

[0054] As used herein, the term "mismatch" refers to opposing nucleotides in a double-stranded RNA complex that do not form a perfect base pair according to the Watson-Crick base pairing rules. In the traditional sense, mismatched nucleotides are GA, CA, UC, AA, GG, CC, and UU pairs. In some embodiments, the first nucleic acid strand of the present invention contains fewer than four mismatches with the target sequence, e.g., 0, 1, or 2 mismatches. "Wobble" base pairs are GU, IU, IA, and IC base pairs. Although G:G pairing is considered a mismatch, this does not necessarily mean that the interaction is unstable; the term "mismatch" may be somewhat outdated based on the current invention, where Hoogsteen base pairings are still relatively stable even when considered mismatches based on the origin of the nucleotides. An isolated G:G pair within a duplex RNA, for example, may be very stable but is still defined as a mismatch.

[0055] The term "splice mutation" refers to a mutation in a gene encoding a pre-mRNA in which the splicing machinery is dysfunctional in the sense that splicing of an intron from an exon is disrupted and this aberrant splicing causes subsequent translation to frame out, resulting in premature termination of the encoded protein. Such truncated proteins are often rapidly degraded and lack any functional activity.

[0056] In one embodiment, an EON capable of forming a double-stranded nucleic acid complex with a target RNA molecule, wherein the double-stranded nucleic acid complex is capable of recruiting adenosine deaminase to deaminate at least one target adenosine in the target RNA molecule, comprises a cytidine analog directly opposite the at least one target adenosine, wherein the cytidine analog functions as a hydrogen bond donor at the N3 position. Preferably, the cytidine analog is pseudoisocytidine (piC) or a base Z of Benar. These cytidine analog nucleotides may exist in RNA or DNA form, or may be modified at the 2' position. Other cytidine analogs that may be used in oligonucleotides according to the invention include 5-hydroxyC-H+, 5-aminoC-H+, and 8-oxoA (syn), variants of Benner's base Z with substituents other than nitro groups (e.g., alkyl groups, F, Cl, Br, CN, etc.) such as cytidine C5 methyl, ethyl, propyl, and variants of 8-oxoA substituted at C2 (e.g., methyl, ethyl, propyl, halogen, etc.). In one embodiment, the cytidine or cytidine analog does not have a 2'-O-Me or 2'-MOE ribose modification.

[0057] In one embodiment, an EON capable of forming a double-stranded nucleic acid complex with a target RNA molecule, wherein the double-stranded nucleic acid complex is capable of recruiting adenosine deaminase to deaminate at least one target adenosine in the target RNA molecule, comprises a uridine analog or derivative directly opposite the target adenosine, wherein the uridine analog or derivative functions as a hydrogen bond donor at the N3 position. Examples of preferred uridine analogs and uridine derivatives include isouridine, pseudouridine, 4-thiouridine, thienouridine, 5-methoxyuridine, dihydrouridine, 5-methyluridine, N3-glycosylated uridine, dihydroisouridine, N3-uracil, and N3-glycosylated uracil. These uridine analogs / derivatives may exist in RNA or DNA form, or may be modified at the 2' position. Other uridine analogs that may be used in the oligonucleotides according to the present invention further include derivatives of isouridine, such as substituted isouridine variants (eg, bearing nitro groups, alkyl groups, F, Cl, Br, CN, etc.).

[0058] The EON (and the complementary nucleic acid strand when two oligonucleotides form an HEON) used in the present invention can be chemically modified throughout, for example, by providing ribose sugar moieties with 2'-OMe, 2'-F, or 2'-MOE substitutions at the nucleotide. The isolated nucleotide in the EON can contain a diF modification at the 2' position of the sugar, or contain deoxyribose (2'-H, DNA), and in yet another embodiment, at least one, and in another embodiment, both of the two adjacent nucleotides flanking the isolated nucleotide do not contain a 2'-OMe modification. Full modification in which all nucleotides of an oligonucleotide have natural bases but retain 2'-OMe modifications results in an oligonucleotide that does not function in RNA editing (as is known in the art), likely because it prevents ADAR activity at the target position. Generally, an adenosine in a target RNA can be protected from editing by providing a 2'-OMe group at the opposing nucleotide (at least in the absence of other chemical substitutions or modifications within the nucleotide) or by providing guanine or adenine as the opposing base, as these two nucleobases can also reduce editing of the opposing adenosine.

[0059] A variety of chemical techniques and modifications that can be readily employed are known in the field of oligonucleotides. The normal internucleoside linkage between nucleotides can be altered by mono- or dithiolation of the phosphodiester bond to produce PS esters or phosphorodithioate esters, respectively. Other modifications of the internucleoside linkage are possible, including amidation and peptide linkers.

[0060] In certain embodiments, an EON of the invention comprises 15, 16, 17, 18, 19, 20, 21, 22, 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, or 60 nucleotides.

[0061] It is known in the art that RNA editors (e.g., human ADAR enzymes) edit dsRNA structures with varying specificity, depending on multiple factors. One important factor is the degree of complementarity between the two strands that make up the dsRNA sequence. When the two strands are completely complementary, the catalytic domain of human ADAR usually reacts with all adenosines it encounters, causing indiscriminate adenosine deamination. The specificity of hADAR1 and 2 can be improved by introducing chemical modifications and / or ensuring multiple mismatches within dsRNA, which is presumed to help position the dsRNA binding domain in a yet-to-be-defined manner. Furthermore, the deamination reaction itself can be promoted by providing an oligonucleotide containing a mismatch opposite the adenosine to be edited. Those skilled in the art can design the complementary portion of the oligonucleotide according to their needs.

[0062] Those skilled in the art will understand that the degree to which editing bodies (for example, ADAR1 or ADAR2 enzymes) in cells are redirected to other target sites can be adjusted by changing the affinity of the first nucleic acid strand for the recognition domain of editing molecules.The modification itself can be determined by trial and error and / or by calculation methods based on the structural interaction between EON and the recognition domain of editing molecules.In addition, or alternatively, the degree of recruitment and redirection of editing bodies present in cells can be adjusted by the dosage and administration regimen of EON.This is determined by experimenters (in vitro) or clinicians, usually in phase I and / or phase II clinical trials.

[0063] The RNA editing molecule present in cells is usually proteinaceous in nature, for example, the ADAR enzyme found in metazoans, including mammals.Preferably, the editing entity in cells is an enzyme, more preferably adenosine deaminase or cytidine deaminase, even more preferably adenosine deaminase.These are enzymes with ADAR activity.One of the most interesting is human ADAR, namely, hADAR1 and hADAR2, including any isoforms thereof.The RNA editing enzymes known in the art that can be conveniently designed for the oligonucleotide construct of the present invention include adenosine deaminase (ADAR) acting on RNA, such as hADAR1 and hADAR2 in humans or human cells, and cytidine deaminase.It is known that there are two isoforms of hADAR1: a long interferon-inducible form of 150 kDa, and a shorter form of 110 kDa (which is produced through alternative splicing from a common pre-mRNA). As a result, the level of the 150 kDa isoform available in cells can be affected by interferon, particularly interferon-γ (IFN-γ). hADAR1 is also induced by TNF-α. This provides an opportunity to develop combination therapies, in which IFN-γ or TNF-α and the EON of the present invention are administered to patients simultaneously or sequentially (in any order) as a combined product or as separate products. In some disease states, elevated levels of IFN-γ or TNF-α may already exist in certain tissues of patients, which creates additional opportunities for more specific editing of diseased tissues. Those skilled in the art will understand that the degree to which the editing product inside the cell is redirected to other target sites can be adjusted by changing the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0064] The present invention relates to the modification of target RNA sequences in eukaryotic, preferably metazoan, more preferably mammalian, and most preferably human cells.

[0065] Target cells can be located in vitro, ex vivo, or in vivo. One advantage of the present invention is that it can be used for cells in situ in vivo, but can also be used for cultured cells. In some embodiments, cells are treated ex vivo and then introduced into a living body (e.g., reintroduced into the living body from which they originally originated). The present invention can also be used to edit target RNA sequences in cells derived from transplants or cells in so-called organoids (e.g., liver tissue organoids). Organoids are considered to be three-dimensional in vitro-derived tissues, but are driven to generate individual, isolated tissues using specific conditions. In therapeutic settings, organoids are useful because they can be generated in vitro from patient cells and then reintroduced into patients as autologous material, which is less likely to be rejected than normal transplants.

[0066] Without wishing to be bound by theory, it is believed that RNA editing via hADAR2 occurs on the primary transcript either in the nucleus during transcription or splicing, or in the cytoplasm, where, for example, mature mRNA, miRNA, or ncRNA may be edited.

[0067] Generally speaking, RNA editing can be used to create RNA sequences with different properties. Such properties can be coding properties (creating proteins with different sequences or lengths, altering the protein's properties or function) or binding properties (causing inhibition or overexpression of the RNA itself or its target or binding partner; entire expression pathways can be altered by recoding their corresponding sequences on the miRNA or target RNA). Protein function or localization can be freely modified by functional domains or recognition motifs (including, but not limited to, signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co- or post-translational modification, catalytic sites for enzymes, binding sites for binding partners, signals for degradation or activation, etc.). These and other forms of RNA and protein "engineering" are encompassed by the present invention, whether for the purposes of disease prevention, delay, or treatment, or for other purposes in medicine or biotechnology, as diagnostics, prophylactics, therapeutics, research tools, or otherwise.

[0068] The amount of triterpene saponin administered and the amount, dosage, and administration regimen of EON may vary depending on the cell type, the disease being treated, the target population, the mode of administration (e.g., systemic or local), the severity of the disease, and the acceptable level of side effects, but these may and should be evaluated by trial and error during in vitro studies, preclinical, and clinical trials. Testing is particularly easy when the modified sequence causes an easily detectable phenotypic change or a change in a specific biomarker (level or activity). Higher doses of EON may compete for binding to ADAR in cells, thereby depleting the amount of entities free to participate in RNA editing, but regular administration tests will reveal any such effect for a given EON and a given target. The same applies to the amount of triterpene saponin administered; the amount may vary depending on the amount of EON, the tissue being treated, or the (human) subject requiring treatment, such as weight, age, sex, etc.

[0069] One suitable testing technique involves delivering EONs to a cell line or test organism and then taking biopsies at various time points. The sequence of the target RNA in the biopsy samples can be assessed, allowing for easy tracking of the percentage of cells with the modification. Once this testing is performed, knowledge is retained, allowing subsequent deliveries to be performed without the need for biopsies. Thus, the methods of the present invention can include identifying the presence of a desired change in the target RNA sequence in a cell, thereby verifying that the target RNA sequence has been modified. This step typically involves sequencing the relevant portion of the target RNA, or a cDNA copy thereof (or, if the target RNA is pre-mRNA, a cDNA copy of its splicing product), as described above, so that the sequence change can be easily verified. Alternatively, the change can be assessed based on protein function or other potential markers before and after treatment. These measurements are preferably performed in vitro on samples obtained from the treated subject.

[0070] After RNA editing occurs in a cell, the modified RNA may be diluted over time, e.g., due to cell division, the finite half-life of the edited RNA, etc. Thus, in terms of practical treatment, the methods of the invention may involve repeated delivery of EONs until enough of the target RNA is modified to provide a tangible benefit to the patient and / or to maintain that benefit over time.

[0071] The compositions and kits of components of the present invention are particularly suitable for therapeutic applications. Accordingly, the present invention also relates to pharmaceutical compositions comprising EON and a pharmaceutically acceptable carrier. Alternatively, triterpene glycosides may be combined in the same pharmaceutical composition, or may be maintained separately in the kit of components if the triterpene glycoside and EON are to be administered at different times. Thus, the kit of components may contain a triterpene glycoside in a suitable administrable composition and an EON in a suitable administrable composition, each containing a suitable solvent, carrier, diluent, or the like. When EON and triterpene glycoside are combined in a single pharmaceutical composition, i.e., when EON and triterpene glycoside are administered simultaneously, the specific solvent, carrier, or diluent may be the same or different. In some embodiments of the present invention, the pharmaceutically acceptable carrier may simply be saline solution, which may be isotonic or hypotonic, particularly for pulmonary delivery. The present invention also provides a delivery device (e.g., a syringe, inhaler, nebulizer) containing the pharmaceutical composition of the present invention. EONs are suitably administered in aqueous solution (e.g., saline) or suspension, which may contain additives, excipients, and other ingredients suitable for pharmaceutical use, at concentrations ranging from 1 ng / ml to 1 g / ml, preferably from 10 ng / ml to 500 mg / ml, and more preferably from 100 ng / ml to 100 mg / ml. Suitable dosages may range from about 1 μg / kg to about 100 mg / kg, preferably from about 10 μg / kg to about 10 mg / kg, and more preferably from about 100 μg / kg to about 1 mg / kg. Administration may be intranasal, oral, by injection or infusion, intravenous, subcutaneous, intradermal, intramuscular, intratracheal, intraperitoneal, rectal, intrathecal, intracisternal, parenteral, or the like. Administration may be in the form of a solid, powder, tablet, gel, solution, sustained-release formulation, or any other form suitable for human pharmaceutical use.

[0072] The present invention provides EON and a triterpene glycoside (preferably AG1856) for use in treating any disorder that can be treated by deaminating a specific adenosine in a specific target transcript molecule. This treatment can be achieved by, but is not necessarily limited to, altering a target RNA sequence in a mammal (e.g., human hepatocytes). Similarly, the present invention provides the use of EON plus a triterpene glycoside (preferably AG1856) in the manufacture of a medicament for altering a target RNA sequence in a mammal (preferably human hepatocytes), as described herein, thereby treating, preventing, or ameliorating the disease.

[0073] The present invention also relates to a method for deaminating at least one specific target adenosine present in a target RNA molecule (mRNA or pre-mRNA) in a cell, the method comprising the following steps: providing a triterpene glycoside, preferably AG1856, to the cell, and providing an EON to the cell; allowing the EON to be taken up by the cell; allowing endosomal release of the EON; annealing the EON to the target RNA molecule; allowing a mammalian ADAR enzyme containing a native dsRNA-binding domain present in a wild-type enzyme to deaminate the target adenosine in the target RNA molecule to inosine; and identifying the presence of the inosine in an RNA sequence.

[0074] In a preferred embodiment, depending on the net effect of the A to I conversion, the identifying step comprises the following steps: sequencing the target RNA; assessing the presence or absence of an A to G conversion in cDNA derived from the target RNA; assessing the presence or absence of a functional protein; assessing whether the deamination has altered splicing of the pre-mRNA; or using a functional readout. Of course, highly suitable methods for identifying the presence of inosine after deamination of a target adenosine are dPCR and even sequencing, using methods well known to those skilled in the art.

[0075] In one embodiment, the method of the present invention comprises the following steps: administering a triterpene glycoside and an EON to a subject; forming a double-stranded nucleic acid complex between the EON and its specific, complementary target nucleic acid molecule in cells within the subject; engaging an existing endogenous adenosine deaminase enzyme, such as ADAR1 and / or ADAR2; and allowing the enzyme to deaminate a target adenosine in the target nucleic acid molecule to inosine, thereby alleviating, preventing, or ameliorating the disease.

[0076] The composition according to the present invention or the kit of components according to the present invention can be applied to any kind of disease that can be beneficially treated by ADAR-mediated deamination of the target adenosine present in the target transcript molecule involved in the disease. Non-limiting examples are as follows: - hypercholesterolemia, where the target adenosine is present in the codon encoding position 152 of the human PCSK9 proprotein (see PCT / EP2023 / 053503); - Hurler syndrome, in which the target adenosine is the c.1205G>A mutation in the human IDUA gene (see, for example, WO2021 / 209010); - HFE hemochromatosis, in which the target adenosine is the c.845G>A mutation in the human HFE gene (see PCT / EP2023 / 082797, unpublished); and - Cardiovascular disease (CVD), the target adenosine is present at position c.1055A in the human B4GALT1 transcript (see PCT / EP2023 / 084865, unpublished).

[0077] Other examples of target transcripts and associated diseases include USH2A (Usher syndrome), APP, NTCP, CMT1A, LRRK2, ASS1, GJB2, MECP2, OTOF (autosomal recessive nonsyndromic hearing loss), XLRS, argininosuccinate lyase deficiency, ABCA4 (Stargardt disease), and SERPINA1 (A1AT deficiency).

[0078] [Chemical modification] All of the chemical modifications listed below that can be used for the EON of the present invention can also be used for the sense strand complementary to the EON when the EON and the complementary strand form a so-called heteroduplex RNA-editing oligonucleotide (HEON) complex, as described in GB2215614.5 (unpublished). The exception is that the opposite sense strand does not have an orphan nucleotide. Note that the definition of "orphan nucleotide" refers to the nucleotide directly opposite the target adenosine in the target RNA molecule (when the EON binds to a target sequence in a cell). The orphan nucleotide does not necessarily mismatch with the target adenosine, for example, when the nucleotide opposite the target adenosine is uridine. In nature, when ADAR is active, the orphan nucleotide opposite the target adenosine is generally cytidine. Therefore, this definition applies only to that position, not to any chemical modification or properties of that nucleotide. The isolated nucleotide is present only in the EON, which is the guide oligonucleotide, and is not present in the sense strand when the EON is combined with a complementary or partially complementary sense strand.Therefore, the modifications related to the isolated nucleotide only relate to the EON of the present invention, and all other modifications relate to the EON of the present invention and any (protected) sense oligonucleotide that can be used with the EON in pharmaceuticals.This includes the use of hydrophobic moieties (e.g., tocopherol and cholesterol) and cell-specific ligands (e.g., GalNAc moieties), which are also described in detail herein and in GB2215614.5 (unpublished), and can be attached to the EON or its opposite strand, or both, as single or multiple copies, at various positions.

[0079] The internucleoside bond in the oligonucleotide of the present invention may comprise one or more natural internucleoside bond(s) and / or modified internucleoside bond(s).Without limitation, at least one, at least two, or at least three internucleoside bond(s) from the 5'-end and / or 3'-end of the EON are preferably modified internucleoside bond(s).Preferably, the modified internucleoside bond(s) is / are PS bond(s).In one embodiment, all internucleoside bond(s) of the EON are modified internucleoside bond(s).In one embodiment, the EON comprises a phosphorylguanidine bond (e.g., PNdmi bond) connecting the most terminal nucleoside at the 5'-end and / or 3'-end and the second nucleoside at each of these ends, respectively.The PNdmi bond preferably used in the EON of the present invention has the structure of formula (III): JPEG2026503580000003.jpg49166

[0080] Common limiting factors in oligonucleotide-based therapeutics include the ability of the oligonucleotide to be taken up by cells (when delivered "naked" without the application of a delivery vehicle), biodistribution, and resistance to nuclease-mediated degradation. Those skilled in the art recognize that various chemical modifications can help overcome such limitations, and these have been extensively reported in the art. Examples of such currently commonly used chemical modifications include 2'-O-methyl (often abbreviated as 2'-OMe or 2'-O-Me), 2'-F, and 2'-O-methoxyethyl (often referred to as 2'-methoxyethoxy or 2'-MOE) sugar modifications, and the use of PS linkages between nucleosides. WO2020 / 201406 discloses the use of MP linkage modifications at specific positions around isolated nucleotides within a first nucleic acid strand. The ribose 2' group in all nucleotides of an EON, except for the ribose sugar moiety of an isolated nucleotide, which has certain limitations with respect to compatibility with RNA editing, can be independently selected from the following: 2'-H (i.e., DNA), 2'-OH (i.e., RNA), 2'-OMe, 2'-MOE, 2'-F, or a 2'-4' linkage (e.g., locked nucleic acid (LNA)), or other ribose 1'-, 2'-, 3'-, 4'-, or 5'-substitutions. Isolated nucleotides in an EON that do not contain other chemical modifications to the ribose sugar, base, or linkage preferably do not have 2'-OMe or 2'-MOE substitutions, but may have 2'-F, 2',2'-disubstituted sugars (e.g., 2',2'-difluoro (diF), or 2'-fluoro-2'-methyl), or 2'-ara-F (FANA) substitutions, or may be DNA. GB2214347.3 (unpublished) describes modifying the 2' position of the ribose sugar moiety of isolated nucleotides with a 2',2'-disubstituted modification such as diF, which is also applicable to the invention described herein. The 2'-4' bond can be selected from many linkers known in the art, such as a methylene linker, an amide linker, or a constrained ethyl linker (cEt).

[0081] The present invention relates to an EON for use in deaminating a target nucleotide (preferably adenosine) in a target RNA, wherein the EON is complementary to a contiguous region of nucleotides in the target RNA that includes the target adenosine, and the nucleotide directly opposite the target nucleotide in a first nucleic acid strand is an isolated nucleotide. When the target nucleotide is adenosine, the isolated nucleotide preferably contains a base with an NH moiety in the same position as the ring nitrogen, or a modified base, or a base analog (e.g., Benner's base Z). Nucleotides in the EON are numbered such that the isolated nucleotide is numbered 0 and the nucleotide 5' to the isolated nucleotide is numbered +1. Numbers increase positively (+) toward the 5' end and negatively (-) toward the 3' end, with the first nucleotide 3' to the isolated nucleotide being numbered -1. The numbering of internucleoside linkages in EON is such that linkage 0 is the 5' linkage of the lone nucleotide, and the linkage positions within the oligonucleotide increase in positive (+) order toward the 5' end and decrease in negative (-) order toward the 3' end.

[0082] Preferably, the EON comprises one or more (chirally pure or chirally mixed) PS bonds. In one embodiment, the PS bonds link the terminal 3, 4, 5, 6, 7, or 8 nucleotides at each end of the first nucleic acid strand. In one embodiment, the EON comprises one or more phosphoramidate (PN) bonds. In one embodiment, the PN bonds link the terminal 2 nucleotides at each end of the EON.

[0083] Nucleosides in EONs may be natural nucleosides (deoxyribonucleosides or ribonucleosides) or unnatural nucleosides. In RNA editing, double-stranded RNA generally serves as a substrate for enzymes with deaminase activity (e.g., ADAR), so ribonucleosides are considered "natural," while deoxyribonucleosides may be considered unnatural or modified for the sake of discussion, since DNA does not exist in an RNA-RNA double-stranded substrate structure. Those skilled in the art will understand that even if a nucleotide has a natural ribose moiety, the base and / or linkage may be unnaturally modified.

[0084] In addition to the specific preferred chemical modifications at certain positions in the compounds of the present invention, the compounds of the present invention may also contain one or more (additional) modifications to the nucleobase, scaffold, and / or backbone linkage. These may or may not be present in the same monomer, for example, at the 3' and / or 5' positions. Scaffold modification refers to the presence of modified forms of the ribose moiety (i.e., pentose moiety) naturally occurring in RNA, such as bicyclic sugars, tetrahydropyrans, hexoses, morpholinos, 2'-modified sugars, 4'-modified sugars, 5'-modified sugars, and 4'-substituted sugars.Examples of suitable modifications include, but are not limited to, 2'-O-modified RNA monomers, such as 2'-O-alkyl or 2'-O-(substituted) alkyl (e.g., 2'-OMe, 2'-O-(2-cyanoethyl), 2'-MOE, 2'-O-(2-thiomethyl)ethyl, 2'-O-butyryl, 2'-O-propargyl, 2'-O-allyl, 2'-O-(2-aminopropyl), 2'-O-(2-(dimethylamino)propyl)). pyr), 2'-O-(2-amino)ethyl, 2'-O-(2-(dimethylamino)ethyl)); 2'-deoxy (DNA); 2'-O-(haloalkyl)methyl (e.g., 2'-O-(2-chloroethoxy)methyl (MCEM), 2'-O-(2,2-dichloroethoxy)methyl (DCEM)); 2'-O-alkoxycarbonyl (e.g., 2'-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2'-O-[2- 2'-O-[2-(N,N-methylcarbamoyl)ethyl] (MCE), 2'-O-[2-(N,N-dimethylcarbamoyl)ethyl] (DCME)); 2'-halo (e.g., 2'-F, FANA); 2'-O-[2-(methylamino)-2-oxoethyl] (NMA); bicyclic or bridged nucleic acid (BNA) scaffold modifications, e.g., conformationally restricted nucleotide (CRN) monomers, locked nucleic acid (LNA) monomers, xylo-LNA monomers, α-LNA mono mer, α-l-LNA monomer, β-d-LNA monomer, 2′-amino-LNA monomer, 2′-(alkylamino)-LNA monomer, 2′-(acylamino)-LNA monomer, 2′-N-substituted 2′-amino-LNA monomer, 2′-thio-LNA monomer, (2′-O,4′-C) constrained ethyl (cEt) BNA monomer, (2′-O,4′-C) constrained methoxyethyl (cMOE) BNA monomer, 2′,4′-BNA. NC (NH) monomer, 2′,4′-BNA NC (NMe) monomer, 2′,4′-BNA NC(NBn) monomers, ethylene-bridged nucleic acid (ENA) monomers, carba-LNA (cLNA) monomers, 3,4-dihydro-2H-pyran nucleic acid (DpNA) monomers, 2'-C-bridged bicyclic nucleotide (CBBN) monomers, oxo-CBBN monomers, heterocyclic bridged BNA monomers (e.g., triazolyl- or tetrazolyl-linked), amide-bridged BNA monomers (e.g., AmNA), urea-bridged BNA monomers, sulfonamide-bridged BNA monomers, bicyclic carbocyclic nucleotide monomers, TriNA monomers, α-l-TriNA monomers, bicyclic DNA (bcDNA) monomers, F-bcDNA monomers, tricyclic DNA (tcDNA) monomers, F-tc DNA monomers, α-anomeric bicyclic DNA (abcDNA) monomers, oxetane nucleotide monomers, locked PMO monomers derived from 2'-amino-LNA, guanidine-bridged nucleic acid (GuNA) monomers, spirocyclopropylene-bridged nucleic acid (scpBNA) monomers, and their derivatives; cyclohexenyl nucleic acid (CeNA) monomers, altritol nucleic acid (ANA) monomers, hexitol nucleic acid (HNA) monomers, fluorinated HNA (F-HNA) monomers, pyranosyl-RNA (p-RNA) monomers, 3'-deoxypyranosyl DNA (p-DNA), unlocked nucleic acid (UNA); and inverted versions of any of the above monomers. All of these modifications are known to those skilled in the art.

[0085] The base sequence of the EON herein is complementary to a portion of the base sequence of the target transcript, including at least the target adenosine to be deaminated to inosine, and therefore can anneal (or hybridize) with the target transcript. The complementarity of the base sequence can be determined using a BLAST program or the like. Those skilled in the art can easily determine the conditions (temperature, salt concentration, etc.) under which the double strands can hybridize, taking into account the complementarity between the strands.

[0086] In contrast to what has been described with respect to gapmers and their relationship to ribonuclease degradation, and the use of such gapmers in double-stranded complexes (see, for example, EP 3954395 A1), the EONs of the present invention do not contain consecutive stretches of DNA nucleotides that would render the target sequence (or sense nucleic acid strand) subject to ribonuclease-mediated degradation. In one embodiment, the EON does not contain four or more consecutive DNA nucleotides at any position within its sequence. In certain embodiments, the EON is composed of as many (chemically) modified nucleotides as possible to increase resistance to ribonuclease-mediated degradation, while at the same time being as efficient as possible in producing an RNA editing effect. This means that isolated nucleotides and other multiple nucleotides within the EON may be DNA, but also means that there are no consecutive stretches of four or more consecutive DNA nucleotides within the EON. Thus, the EONs of the present invention are not gapmers. A gapmer is essentially a single-stranded nucleic acid, consisting of a central region (a DNA gap region having at least four consecutive deoxyribonucleotides) and wing regions located directly at its 5'-end (5' wing region) and 3'-end (3' wing region). In contrast, the EON according to the present invention may be any oligonucleotide that produces an RNA editing effect by deaminating target adenosines to inosines in target RNA molecules, and therefore is as resistant as possible to RNase-mediated degradation to achieve this effect.

[0087] In one embodiment, the EON or the sense strand that can anneal to it before entering the target cell is conjugated to a hydrophobic moiety, such as palmityl or its analog, cholesterol or its analog, or tocopherol or its analog. This is preferably conjugated to the 5'-end. When the hydrophobic moieties are conjugated to the 5'-end and the 3'-end, the hydrophobic moieties may be the same or different. The hydrophobic moiety conjugated to the oligonucleotide may be directly conjugated or indirectly conjugated via another substance. When the hydrophobic moiety is directly conjugated, it is sufficient that the moiety is conjugated via a covalent bond, ionic bond, hydrogen bond, or the like. When the hydrophobic moiety is indirectly conjugated, it may be conjugated via a linker. The linker may be cleavable or non-cleavable. A cleavable linker refers to a linker that can be cleaved under physiological conditions, for example, within a cell or an animal body (e.g., the human body). Cleavable linkers are selectively cleaved by endogenous enzymes such as nucleases or by physiological conditions specific to a portion of the body or cell, such as pH or a reducing environment (e.g., glutathione concentration). Examples of cleavable linkers include, but are not limited to, amides, esters, phosphodiesters (either esters or both), phosphoesters, carbamates, and disulfide bonds, as well as natural DNA linkers. Cleavable linkers also include self-immolative linkers. A non-cleavable linker refers to a linker that is not cleaved under physiological conditions or that cleaves much more slowly than a cleavable linker, such as a linker consisting of a PS bond, a modified or unmodified deoxyribonucleoside linked by a PS bond, a spacer linked through a PS bond, and a modified or unmodified ribonucleoside. When the linker is a nucleic acid such as DNA or an oligonucleotide, there is no limit to its chain length. However, it can typically be 2 to 20 bases long, 3 to 10 bases long, or 4 to 6 bases long. The length or structure of the spacer linking the ligand and the oligonucleotide is not limited, and may include, for example, ethylene glycol, TEG, HEG, an alkyl chain, propyl, 6-aminohexyl, or dodecyl.

[0088] In one embodiment, the EON for RNA editing is administered separately from the triterpene glycoside (preferably AG1856). In this case, the triterpene glycoside may be administered first, followed by the EON, or the EON may be administered first, followed by the triterpene glycoside. Alternatively, they may be administered simultaneously, in which case the EON and the triterpene glycoside are preferably administered in the same composition. In another embodiment, the triterpene glycoside is covalently or non-covalently bound to the EON.

[0089] The present invention also relates to a pharmaceutical composition comprising the EON of the present invention, further comprising a pharmaceutically acceptable carrier and / or other additives, and optionally dissolved in a pharmaceutically acceptable organic solvent or the like. The dosage form in which the EON or the pharmaceutical composition is administered may depend on the disorder to be treated and the tissue that needs to be targeted, and may be selected according to general procedures in the art. The pharmaceutical composition may be administered in a single dose or multiple doses. It may be administered daily or at appropriate time intervals determined using general knowledge in the art, and may be adjusted based on the disorder and the effectiveness of the active ingredient. Therefore, the order in which the triterpene glycosides are administered may also be changed.

[0090] In one embodiment, an EON comprises at least one nucleotide having a sugar moiety comprising a 2'-OMe modification. In one embodiment, an EON comprises at least one nucleotide having a sugar moiety comprising a 2'-MOE modification. In one embodiment, an EON comprises at least one nucleotide having a sugar moiety comprising a 2'-F modification. In one embodiment, an isolated nucleotide has a 2'-H in the sugar moiety, and is therefore referred to as a DNA nucleotide, although additional modifications may be present in its base and / or its linkage to the adjacent nucleoside. In one embodiment, an isolated nucleotide has a 2'-F in the sugar moiety. In one embodiment, an isolated nucleotide has a diF substitution in the sugar moiety. In one embodiment, an isolated nucleotide has 2'-F and 2'-C-methyl in the sugar moiety. In one embodiment, an isolated nucleotide comprises a 2'-F in the arabinose configuration (FANA) in the sugar moiety. In one embodiment, the EON is an antisense oligonucleotide capable of forming a double-stranded nucleic acid complex with a target RNA molecule, wherein the double-stranded nucleic acid complex is capable of recruiting adenosine deaminase to deaminate a target adenosine in the target RNA molecule, wherein the nucleotide opposite the target adenosine in the EON is an isolated nucleotide, and the isolated nucleotide is according to formula (IV): JPEG2026503580000004.jpg38166 where: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from the group consisting of cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxo-adenine, and 6-amino-5-nitro-2(1H)-pyridone; R1 and R2 are each independently selected from H, OH, F, or CH3; R3 is a portion of the EON located 5' to the isolated nucleotide and consisting of 7 to 30 nucleotides; and R4 is a portion of the EON located 3' to the isolated nucleotide and consisting of 4 to 25 nucleotides. The nucleotides on the 3' and / or 5' sides of the isolated nucleotide may be DNA, more preferably the nucleotide on the 3' side (-1 position).

[0091] In one embodiment, the first nucleic acid strand comprises at least one MP internucleoside linkage according to formula (V): JPEG2026503580000005.jpg38166

[0092] A preferred position for MP attachment in an EON according to the present invention is the -1 attachment position, thereby linking the nucleoside at the -1 position with the nucleoside at the -2 position, although MP attachment at other positions is not expressly excluded.

[0093] In one embodiment, the EON comprises at least one nucleotide having a sugar moiety containing a 2'-fluoro (2'-F) modification. A preferred position for the nucleotide having the 2'-F modification is position -3 in the EON, which may occur in conjunction with the same 2' modification in an isolated nucleotide, as described above.

[0094] In one embodiment, the EON comprises at least one phosphonoacetate or phosphonoacetamide internucleoside linkage.

[0095] In one embodiment, the EON comprises at least one nucleotide that comprises a locked nucleic acid (LNA) ribose modification or an unlocked nucleic acid (UNA) ribose modification. In certain embodiments, the EON comprises at least one nucleotide that comprises a threose nucleic acid (TNA) ribose modification.

[0096] Those skilled in the art know that oligonucleotides, such as the EONs outlined herein, are generally composed of repeating monomers. These monomers are most often nucleotides or chemically modified nucleotides. The most common naturally occurring nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (U). They consist of a pentose sugar (ribose), a 5'-linked phosphate group (which is linked via a phosphate ester), and a 1'-linked base. The sugar is often referred to as the "scaffold" of the nucleotide because it connects the base and the phosphate group.

[0097] Therefore, modifications to the pentose sugar are often referred to as "scaffold modifications." The original pentose sugar may be completely replaced by another moiety that similarly links the base and phosphate group. Thus, while a pentose sugar is often the scaffold, it is understood that a scaffold does not necessarily have to be a pentose sugar. Examples of scaffold modifications that can be applied to the monomers of the EONs of the present invention are disclosed in WO2020 / 154342, WO2020 / 154343, and WO2020 / 154344.

[0098] In one embodiment, an EON of the present invention may contain one or more nucleotides with a 2'-MOE ribose modification. In another embodiment, an EON contains one or more nucleotides without a 2'-MOE ribose modification, and the 2'-MOE ribose modification is located at a position where an enzyme with adenosine deaminase activity does not prevent deamination of the target adenosine. In another embodiment, an EON contains a 2'-OMe ribose modification at a position where it does not contain a 2'-MOE ribose modification, and / or an oligonucleotide contains a deoxynucleotide at a position where it does not contain a 2'-MOE ribose modification. In one embodiment, the EON comprises one or more nucleotides containing a 2'-position containing 2'-MOE, 2'-OMe, 2'-OH, 2'-deoxy, TNA, 2'-fluoro (2'-F), 2',2'-disubstituted modifications (e.g., 2',2'-difluoro (diF) modifications, 2'-fluoro-2'-C-methyl modifications, or other modifications, including 2'-spirocyclic modifications, e.g., as described in Grosse et al. (ACS Med Chem Lett 2022 DOI: 10.1021 / acsmedchemlett.2c00372)), or a 2'-4' linkage (i.e., a bridged nucleic acid, such as a locked nucleic acid (LNA), e.g., as described in WO2018 / 007475). In another embodiment, other nucleic acid monomers that can be used include arabinonucleic acid and 2'-deoxy-2'-fluoroarabinonucleic acid (FANA), e.g., for improved affinity. The 2'-4' linkage can be selected from linkers known in the art, such as methylene linkers or constrained ethyl linkers. A wide variety of 2' modifications are known in the art. Further examples are disclosed in further detail in, for example, WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, WO2018 / 134301, WO2019 / 219581, WO2019 / 158475, and WO2022 / 099159.In all cases, the modification must be compatible with editing, so that the EON serves as an editing-initiating oligonucleotide that forms a double-stranded complex with the target RNA and recruits a deaminating enzyme, which then deaminates the target adenosine. If the monomer contains an unlocked nucleic acid (UNA) ribose modification, the monomer can have a 2' position containing the same modifications as those described above, such as 2'-MOE, 2'-OMe, 2'-OH, 2'-deoxy, 2'-F, 2',2'-diF, 2'-fluoro-2'-C-methyl, arabinonucleic acid, FANA, or a 2'-4' linkage (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA)).

[0099] A base (sometimes called a nucleobase) is generally adenine, cytosine, guanine, thymine, or uracil, or a derivative thereof. A base (sometimes called a nucleobase) is defined as a moiety that can bind to another nucleobase through hydrogen bonds, polar bonds (e.g., through a CF moiety), or aromatic electron interactions. Cytosine, thymine, and uracil are pyrimidine bases, generally linked to the scaffold through their 1-nitrogen. Adenine and guanine are purine bases, generally linked to the scaffold through their 9-nitrogen. As used herein, the terms "adenine," "guanine," "cytosine," "thymine," "uracil," and "hypoxanthine" refer to the nucleobase itself. The terms "adenosine," "guanosine," "cytidine," "thymidine," "uridine," and "inosine" refer to a nucleobase linked to (deoxy)ribose.

[0100] The nucleobases in the EONs of the present invention can be adenine, cytosine, guanine, thymine, or uracil, or any other moiety capable of interacting with another nucleobase through hydrogen bonding, polar bonding (e.g., C—F), or aromatic electronic interactions. The nucleobases at any position within the nucleic acid strand can be modified forms of adenine, cytosine, guanine, or uracil, such as, for example, hypoxanthine (nucleobase at inosine), pseudouracil, pseudocytosine, isouracil, N3-glycosylated uracil, isocytosine, 1-methylpseudouracil, orotic acid, agmatidine, lysidine, 2-thiouracil, 2-thiothymine, 5-substituted pyrimidines (e.g., 5-halouracil, 5-halomethyluracil, 5-trifluoromethyluracil, 5-propynyluracil, 5-propynylcytosine, 5-aminomethyluracil, 5-hydroxymethyluracil, 5-formyluracil, 5-aminomethylcytosine, 5-formylcytosine), 5-hydroxymethylcytosine, 7-deazaguanine, 7-deazaadenine, 7-deaza- 8-aza-2,6-diaminopurine, 8-aza-7-deazaguanine, 8-aza-7-deazaadenine, 8-aza-7-deaza-2,6-diaminopurine, 8-oxo-adenine, 3-deazapurines (e.g., 3-deazaadenosine), pseudoisocytosine, N4-ethylcytosine, N2-cyclopentylguanine, N2-cyclopentyl-2-aminopurine, N2-propyl-2-aminopurine, 2,6-diaminopurine, 2-aminopurine, G-clamps and derivatives thereof, super A, super T, super G, amino-modified nucleobases or derivatives thereof; and modified or universal bases (e.g., 2,6-difluorotoluene), or deletions such as abasic sites (e.g., 1-deoxyribose, 1,2-dideoxyribose, 1-deoxy-2-O-methylribose, azaribose).

[0101] In some embodiments, the nucleotide analog is an analog of a nucleic acid nucleotide. In some embodiments, the nucleotide analog is an analog of adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine, or deoxyuridine. In some embodiments, the nucleotide analog is not guanosine or deoxyguanosine. In some embodiments, the nucleotide analog is not a nucleic acid nucleotide. In some embodiments, the nucleotide analog is not adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine, or deoxyuridine.

[0102] Nucleotides are typically linked to adjacent nucleotides through condensation of their 5'-phosphate moiety to the 3'-hydroxyl moiety of the adjacent nucleotide monomer. Similarly, their 3'-hydroxyl moiety is typically linked to the 5'-phosphate of the adjacent nucleotide monomer. This forms a phosphodiester bond. The phosphodiester and scaffold form an alternating copolymer. The bases are attached to this copolymer, i.e., the scaffold moiety. Because of this characteristic, the alternating copolymer formed by the linked scaffolds of an oligonucleotide is often referred to as the "backbone" of the oligonucleotide. Because phosphodiester bonds connect adjacent monomers, they are often referred to as "backbone linkages." It is understood that even if the phosphate group is modified and replaced with an analogous moiety, such as PS, such moiety is still referred to as the backbone linkage of the monomer. This is referred to as "backbone linkage modification." In a general sense, the backbone of an oligonucleotide comprises alternating scaffolds and backbone linkages.

[0103] EONs according to the present invention may include linkage modifications, such as, but not limited to, modified forms of phosphodiesters present in RNA, including, but not limited to, PS, chiral pure PS, (R)-PS, (S)-PS, methyl phosphonate (MP), chiral pure methyl phosphonate, (R)-methyl phosphonate, (S)-methyl phosphonate, phosphoryl guanidine (e.g., PNdmi), chiral pure phosphoryl guanidine, (R)-phosphoryl guanidine, (S)-phosphoryl guanidine, phosphorodithioate (PS2), phosphonoacetate, e) (PACE), phosphonoacetamide (PACA), thiophosphonoacetate, thiophosphonoacetamide, methyl phosphorothioate, methylthiophosphonate, PS prodrugs, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, boranophosphate, borano PS, methylboranophosphate, methylborano PS, methylboranophosphonate, methylboranophosphothioate, phosphate, phosphotriester, aminoalkylphosphotriester, and derivatives thereof. Other modifications include phosphoramidites, phosphoramidates, N3'→P5' phosphoramidates, phosphorodiamidates, phosphorothiodiamidates, sulfamates, diethylene sulfoxides, amides, sulfonates, siloxanes, sulfides, sulfones, formacetyls, alkenyls, methylenehydrazinos, sulfonamides, triazoles, oxalyls, carbamates, methyleneiminos (MMIs), and thioacetamido nucleic acids (TANAs); and derivatives thereof. Various salts, mixed salts, and free acid forms are also included, including 3'→3' and 2'→5' linkages.

[0104] In one embodiment, the EON contains a substitution of one of the non-bridging oxygens in the phosphodiester bond. This modification slightly destabilizes base pairing but confers significant resistance to nuclease degradation. Preferred nucleotide analogs or equivalents include PS, phosphonoacetate, phosphorodithioate, phosphotriester, aminoalkylphosphotriester, H-phosphonate, methyl and other alkylphosphonates (including 3'-alkylenephosphonate, 5'-alkylenephosphonate, and chiral phosphonate), phosphinate, phosphoramidate (including 3'-aminophosphoramidate and aminoalkylphosphoramidate), thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, selenophosphate, or boranophosphate. Particularly preferred are internucleoside linkages modified to contain PS. Many of these non-natural bond modifications (e.g., PS) are chiral, meaning that they have both Rp and Sp configurations, and are known to those skilled in the art. In one embodiment, the chirality of the PS bond is controlled, meaning that each bond has either the preferred Rp or Sp configuration. The selection of the Rp or Sp configuration at a particular bond position may depend on the target sequence and the binding efficiency and induction efficiency resulting in RNA editing. However, if this is not particularly desired, the composition may contain an AON having both the Rp and Sp configurations at a particular bond position as an active compound. A mixture of EONs is also possible, where one configuration is preferred at some positions but not at other positions.

[0105] Again, in all cases, the modification must be compatible with editing, so that the EON serves as an editing oligonucleotide, and when bound to its target sequence, the properties of dsRNA are generated, thereby recruiting adenosine deaminase.In all aspects of the present invention, the enzyme with adenosine deaminase activity is preferably ADAR1, ADAR2, or ADAT.In a highly preferred embodiment, the EON is an RNA editing oligonucleotide that targets pre-mRNA or mRNA, wherein the target nucleotide is adenosine in the target RNA, and the adenosine is deaminated to inosine, which is read as guanosine by the translational machinery.The present invention also relates to a pharmaceutical composition comprising the EON characterized herein and a pharmaceutically acceptable carrier.

[0106] Other chemical modifications of EONs according to the present invention include replacing any one or more hydrogen atoms with deuterium or tritium, examples of which can be found, for example, in WO2014 / 022566 or WO2015 / 011694.

[0107] The EONs of the present invention preferably do not contain a 5'-terminal O6-benzylguanosine or a 5'-terminal amino modification, and preferably do not contain a SNAP tag domain (modified O6-alkylguanosine-DNA-alkyltransferase) covalently linked to it. The EONs of the present invention preferably do not contain a boxB RNA hairpin sequence. In one embodiment, the EONs of the present invention contain 0, 1, 2, or 3 wobble base pairs with the target sequence and / or 0, 1, 2, 3, 4, 5, 6, 7, or 8 mismatch base pairs with the target RNA sequence. When the isolated nucleotide is a uridine, no mismatch occurs. One alternative to uridine is to place an isouridine opposite the target adenosine, which likely does not pair with a G as a U. Preferably, the target adenosine in the target sequence forms a mismatch base pair with the nucleoside in the EON directly opposite the target adenosine.

[0108] It should be noted that when EON is delivered through vector (for example, AAV vector), there is no chemical modification in the EON that acts on target RNA molecule.According to the present invention, it is preferred to use " naked " EON with chemical modification as outlined herein.The EON that has circular or hairpin structure (recruitment part, for example, as disclosed in WO2016 / 097212, WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560 and WO2022 / 078995) is also included in the present invention.Because they can also be applied to edit adenosine in target RNA molecule.

[0109] EON of the present invention can utilize endogenous cellular pathway and naturally occurring ADAR enzyme to specifically edit target adenosine in target RNA sequence.EON of the present invention can recruit ADAR and form complex with it, and then promote the deamination of (single) specific target adenosine nucleotide in target RNA sequence.Ideally, only one adenosine is deaminate.Preferably, when EON of the present invention forms complex with ADAR, it causes the deamination of single target adenosine.

[0110] Analysis of natural targets of ADAR enzymes has shown that they generally contain mismatches between the two strands that form the RNA helices edited by ADAR1 or 2. These mismatches have been suggested to enhance the specificity of the editing reaction (Stefl et al., 2006. Structure 14(2):345-355; Tian et al. 2011. Nucleic Acids Res 39(13):5669-5681). Characterization of the optimal pattern of paired / mismatched nucleotides between EONs and target RNAs is also likely to be important for the development of effective ADAR-based EON therapies.

[0111] As outlined above, the EONs of the present invention utilize specific nucleotide modifications at predetermined locations to ensure stability and proper ADAR binding and activity. As described in detail herein, these modifications may vary and include modifications in the EON backbone, in the sugar moiety of the nucleotide, and in the nucleobase or phosphodiester bond. They may also be variably distributed throughout the sequence of the EON. Specific modifications may be required to support interactions with various amino acid residues in the RNA-binding domain of the ADAR enzyme and those in the deaminase domain. For example, internucleotide PS bonds or 2'-OMe or 2'-MOE modifications may be tolerated in some EONs, but should be avoided in others to avoid interfering with the important interactions between the enzyme and the phosphate and 2'-OH groups. Specific nucleotide modifications may also be required to enhance editing activity for substrate RNAs whose target sequences are not optimal for ADAR editing. Previous studies have established that certain sequence contexts are more susceptible to editing. For example, the target sequence 5′-UAG-3′ (with a central target A) contains the most favorable nearest neighbor nucleotide for ADAR2, whereas the target sequence 5′-CAA-3′ is unfavorable (Schneider et al. 2014. Nucleic Acids Res 42(10):e87). Structural analysis of the ADAR2 deaminase domain suggests the possibility of enhancing editing by carefully selecting the nucleotide opposite the target trinucleotide. For example, the 5′-CAA-3′ target sequence pairs with the 3′-GCU-5′ sequence on the opposite strand (creating a central AC mismatch), but this is unfavorable because the guanosine base clashes sterically with the amino acid side chain of ADAR2.

[0112] Mutagenesis studies of human ADAR2 revealed that a single mutation at residue 488 (E488Q) from glutamic acid to glutamine increased the deamination rate constant by 60-fold compared to the wild-type enzyme (Kuttan and Bass. Proc Natl Acad Sci USA 2012. 109(48):3295-3304). During the deamination reaction, ADAR flips the base to be edited from its RNA duplex and places it in the enzyme's active site (Matthews et al. 2016). When ADAR2 edits an adenosine in a favorable context (A:C mismatch), the nucleotide opposite the target adenosine is often referred to as a "lone cytidine." The crystal structure of ADAR2 E488Q bound to double-stranded RNA (dsRNA) revealed that the glutamine (Gln) side chain at position 488 can donate a hydrogen bond to the N3 position of the lone cytidine, leading to an improved catalytic rate of ADAR2 E488Q. In the wild-type enzyme, a glutamic acid (Glu) exists at position 488 instead of a glutamine (Gln), and the amide group of the glutamine is absent, becoming a carboxylic acid instead. Therefore, to achieve the same contact with the lone cytidine using the E488Q mutant in the wild-type context, protonation is required for this contact to occur. When using endogenously expressed ADAR2 to correct disease-associated mutations, it is important to maximize the editing efficiency of the wild-type ADAR2 enzyme present in cells. WO 2020 / 252376 discloses the use of EONs, specifically modified RNA bases at the lone cytidine position, to mimic the hydrogen bonding pattern observed with the E488Q ADAR2 mutant. It was hypothesized that by replacing the nucleotide opposite the target adenosine in the EON with a cytidine analogue that functions as a hydrogen bond donor at N3, it would be possible to stabilize the same contacts that would improve catalytic rate in the mutant enzyme.Two cytidine analogs were of particular interest: pseudoisocytidine (also referred to as "piC"; Lu et al. J Org Chem 2009. 74(21):8021-8030; Burchenal et al. (1976) Cancer Res 36:1520-1523) and Benner's base Z (also referred to as "dZ"; Yang et al. Nucl Acid Res 2006. 34(21):6095-6101), which were initially selected because they donate a hydrogen bond at N3 with minimal perturbation to the nucleobase geometry. Benner's base is also chemically referred to as 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase. The presence of cytidine analogs within AONs can be in addition to modifications to the ribose 2' group. The ribose 2' groups in the AON can be independently selected from the following: 2'-H (i.e., DNA), 2'-OH (i.e., RNA), 2'-OMe, 2'-MOE, 2'-F, or a 2'-4' linkage (i.e., bridged nucleic acids such as locked nucleic acids (LNAs)), or other 2' substituents. The 2'-4' linkage can be selected from linkers known in the art, such as a methylene linker or a constrained ethyl linker.

[0113] In one embodiment, the EON comprises one or more sugar moieties that are mono- or di-substituted at the 2′, 3′, and / or 5′ positions, as exemplified below: —OH; —H; —F; substituted or unsubstituted, straight or branched lower (C1-C 10 ) alkyl, alkenyl, alkynyl, alkaryl, aryl, or aralkyl, which may be interrupted by one or more heteroatoms; -O-, S-, or N-alkyl; -O-, S-, or N-alkenyl; -O-, S-, or N-alkynyl; -O-, S-, or N-aryl; -O-alkyl-O-alkyl; -methoxy; -aminopropoxy; -methoxyethoxy; -dimethylaminooxyethoxy; and -dimethylaminoethoxyethoxy.

[0114] In one embodiment, the nucleotide analog or equivalent in EON contains one or more base modifications or substitutions. Modified bases include synthetic and natural bases, such as inosine, xanthine, hypoxanthine, and other aza, deaza, hydroxy, halo, thio, thiol, alkyl, alkenyl, alkynyl, and thioalkyl derivatives of pyrimidine and purine bases, which are known or will be developed in the art. Purine nucleobases and / or pyrimidine nucleobases can be modified, for example, by amination or deamination of heterocycles, to change their properties. The exact chemical structure and format can vary from oligonucleotide construction to oligonucleotide construction and application to application, and can be determined according to the wishes and preferences of those skilled in the art.

[0115] EONs according to the present invention are typically longer than 10 nucleotides, preferably greater than 11, 12, 13, 14, 15, 16, and even more preferably greater than 17 nucleotides. In one embodiment, AONs according to the present invention are longer than 20 nucleotides. Oligonucleotides according to the present invention are preferably shorter than 100 nucleotides, more preferably shorter than 60 nucleotides, and even more preferably shorter than 50 nucleotides. In preferred embodiments, oligonucleotides according to the present invention comprise 18 to 70 nucleotides, more preferably 18 to 60 nucleotides, and even more preferably 18 to 50 nucleotides. Thus, in particularly preferred embodiments, oligonucleotides according to the present invention comprise 15, 16, 17, 18, 19, 20, 21, 22, 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, or 50 nucleotides. In one embodiment, the AON is 27, 28, 29, or 30 nucleotides in length.

[0116] In one embodiment, an inverted deoxyT or dideoxyT nucleotide is incorporated at each or both ends of an EON according to the invention. [Example]

[0117] Example 1. Editing of target adenosines in human APP RNA molecules using triterpene saponin AG1856 in vitro. The human retinal pigment epithelial cell line ARPE-19, which harbors the wild-type APP gene, was obtained from ATCC (Lot No. #70013110). Briefly, 5.0 × 10 4 ARPE-19 cells were seeded per 24-well plate 4 hours before oligo treatment. 1 μM EON hAPPex17-1dZ was then added to the cells. This EON (RM3023) has the following sequence: JPEG2026503580000006.jpg7166 (sequence number 1) where * represents a PS bond, dZ represents a deoxynucleotide with a benign base, C and A represent cytidine and adenosine (DNA), respectively, and lowercase nucleotides represent 2'-OMe-modified nucleotides. This EON targets adenosine at position 2286 in exon 17 of human APP mRNA. Untreated (NT) ARPE-19 cells were used as a negative control, while another control was cells treated with EON alone (control). 72 hours after EON treatment, 2, 4, or 8 μM AG1856 (purified as disclosed in WO2021 / 122998) in fresh medium was added to the cells. RNA was extracted from cells 120 hours after oligo treatment using the Direct-zol RNA MircoPrep (Zymo Research, R2062) kit according to the manufacturer's instructions, and cDNA was prepared using the Maxima Reverse Transcriptase Kit (Thermo Fisher) with a combination of random hexamers and oligo-dT primers according to the manufacturer's instructions. The cDNA was diluted 10-fold, and 1 μL of this dilution was used as a template for digital droplet PCR (ddPCR) using 2.5 ng of total RNA input. ddPCR assays for absolute quantification of nucleic acid target sequences were performed using BioRad's QX-200 Droplet Digital PCR System. 1 μL of diluted cDNA obtained from the RT cDNA synthesis reaction was used in a total reaction mixture of 21 μL, including ddPCR Supermix for Probes no dUTP (BioRad) and the Taqman SNP genotype assay with the following forward and reverse primers and gene-specific probes: Hexachlorofluorescein, a non-fluorescent quencher-labeled wild-type probe 5'- / 5HEX / TGTT+GTCAT+A+G+CGACAGT / 3IABkFQ / -3' (SEQ ID NO: 2) Fluorescein and non-fluorescent quencher-labeled mutant probes 5'- / 56-FAM / TGTTGTCAT+G+GCGACAGT / 3IABkFQ / -3' (SEQ ID NO: 3)

[0118] A total volume of 21 μL of PCR mix containing cDNA was loaded into the center column of a ddPCR cartridge (BioRad) using a multichannel pipette. Replicate samples were divided into two cartridges. The bottom column was filled with 70 μL of probe droplet generation oil (BioRad). After replacing the rubber gasket, droplets were generated using a QX200 droplet generator. 42 μL of oil emulsion from the top column of the cartridge was transferred to a 96-well PCR plate. The PCR plate was sealed with tin foil using a PX1 plate sealer at 170°C for 4 seconds, followed by the following PCR program: one cycle of enzyme activation at 95°C for 10 minutes; 40 cycles of denaturation at 95°C for 30 seconds and annealing / extension at 53.8°C for 1 minute; one cycle of enzyme inactivation at 98°C for 10 minutes; and then stored at 8°C. The APP forward sequence primer was 5′-CATTGGACTCATGGTGG-3′ (SEQ ID NO: 5). The APP reverse primer was 5'-CAGCATCACCAAGGTG-3' (SEQ ID NO: 6). After PCR, plates were read and analyzed on a QX200 droplet reader.

[0119] The effect on RNA editing is shown in Figure 1 and clearly demonstrates a dose-dependent increase in RNA editing of endogenous APP transcripts in ARPE-19 cells when increasing amounts of AG1856 were delivered separately from EON.

[0120] Example 2. Editing of target adenosines in mouse App RNA molecules using triterpene saponin AG1856 in vivo. Female C57BL / 6JRj mice were randomly assigned and all were approximately 10 weeks old at the time of the first dose. Mice were group-housed (5 per cage) in standard open polysulfone IIL cages under a regular light-dark cycle (12:12 h light:dark, lights on at 7:00 AM) with free access to standard chow and water. A total of 12 female wild-type C57BL / 6JRj mice were treated with a single subcutaneous (SC) injection of AG1856 (3.6 mg / kg) in sterile PBS on day 0, followed one hour later by an intravenous (IV) injection of EON (35 mg / kg) in sterile PBS targeting the mApp adenosine at position 2344 within exon 17. The EON was designated RM3835 and has the following sequence: JPEG2026503580000007.jpg8166 (sequence number 7) Here, lowercase letters represent 2′-OMe modified nucleotides, bold A represents deoxyadenosine, underlined capital letters represent 2′-F modified nucleotides, dZ represents deoxynucleotides with a benone base (DNA), underlined lowercase letters represent 2′-MOE modified nucleotides, an asterisk represents a PS bond, (MeP) represents a methylphosphonate bond, and "!" represents a PNdmi bond.

[0121] Mice were sacrificed on day 14, and tissues were collected at necropsy. Rapidly frozen tissue samples were thawed and disrupted using a MagNA lyzer (Roche) in TRIzol reagent (Roche). The samples were subjected to two 30-second passes (6500 rpm) with a 90-second cooling period between passes. The samples were then incubated at room temperature for 2 minutes to allow complete dissociation of nuclear proteins. Chloroform was then added to the suspension in preparation for phase separation. After centrifugation at 12,000 × g (4°C) for 15 minutes, the aqueous layer containing RNA was used for further processing. RNA was isolated using the ReliaPrep RNA Cell Miniprep System according to the manufacturer's protocol. cDNA synthesis was performed using the Maxima Reverse Transcriptase Kit (Thermo Scientific) according to the manufacturer's protocol. Briefly, to avoid interference of cDNA synthesis by secondary structures, 500 ng of RNA was first incubated with a dNTP mix (10 mM each), random hexamers, and oligo Dt at 70°C for 5 minutes, followed by slow cooling to 10°C at a 10°C drop per 15 seconds. Reverse transcriptase buffer and enzyme were then added, and the sample was incubated at 25°C for 10 minutes, 50°C for 30 minutes, and 80°C for 5 minutes (to inactivate the enzyme). For dPCR analysis, cDNA samples (undiluted for pancreas and diluted 40-fold for other tissues) were incubated with a mixture of primers and probes specific for wild-type mApp exon 17 (HEX), mutant mApp exon 17 (FAM), all mApp exons 4-5 (Cy5), and mRps19. The primer and probe sequences were as follows:

[0122] JPEG2026503580000008.jpg82166

[0123] 12 μL of each sample was loaded into a QIAcuity Nanoplate 8.5K 96-well and run on the QIAcuity (Qiagen) including sample splitting, PCR, and imaging.

[0124] The results are shown in Figure 2. While no editing was observed in the control group administered triterpene saponin AG1856 alone, some editing (0.1–1%) was observed in tissues from animals treated with RM3835 alone. This increased to approximately 2%, 6%, and 9% in the pancreas, liver, and kidney, respectively, when pretreated with triterpene saponin, demonstrating the strong effect of AG1856 on EON- and ADAR-mediated RNA editing efficiency in vivo.

[0125] Example 3. Editing of human B4GALT1 transcripts in HepG2 cells using AG1856. Recently identified targets in the fight against CVD include the enzyme B4GALT1, which is involved in the processing of biologically important biomolecules, including those involved in lipid metabolism and coagulation. B4GALT1 is ubiquitously expressed and plays a key role in the processing of N-linked oligosaccharide moieties on glycoproteins, transferring galactose from uridine diphosphate galactose (UDP-Gal) to specific glycoprotein substrates. B4GALT1 is therefore important for the biological activities associated with correctly assembled oligosaccharides.

[0126] Editing of endogenous human B4GALT1 transcripts was investigated in human cells in the presence of AG1856. For this purpose, human HepG2 hepatocellular carcinoma cells were cultured in EMEM + 10% FBS + 1% P / S. Cells were maintained at 37°C in a 5% CO2 atmosphere. In the first experiment, EONs designated RM4826-RM4849 (see Figure 3) were tested for editing efficiency. A total of 0.75 x 10 5HepG2 cells were seeded and treated with 1 or 5 μM EON, and 1 μM AG1856 was added per well. This mixture was left on the cells for 72 hours. Afterwards, the cells were harvested, and total RNA was isolated from the transfected cells using the ReliaPrep™ RNA Miniprep Kit. After removing the culture medium, the cells were washed once with PBS. After completely aspirating the PBS, 100 μL of lysis buffer was added to the well to lyse the cells and recover the intracellular material. After adding 35 μL of isopropanol, the mixture was loaded onto a column and subjected to multiple washing steps and DNase I treatment. After elution in a total volume of 15 μL of DNase / RNase-free water, the RNA yield was determined using spectrophotometry (NanoDrop) and stored at -80°C.

[0127] Maxima reverse transcriptase (RT, Thermo Fisher Scientific) was used to generate complementary DNA (cDNA). Typically, 500 ng of total RNA was used in a reaction mixture containing 4 μL of 5x RT buffer, 2 μL of dNTP mix (10 mM each), 0.5 μL of oligo(dT), 0.5 μL of random hexamers, and 0.5 μL of Maxima reverse transcriptase (all Thermo Fisher Scientific). DNase- and RNase-free water was added to a total volume of 20 μL. The sample was loaded into a T100 thermal cycler (Bio-Rad) and initially incubated at 25°C for 10 minutes, followed by a 30-minute cDNA reaction at 50°C and a 5-minute stop step at 85°C. The sample was cooled to 4°C and then stored at -20°C.

[0128] To determine the editing efficiency, cDNA samples were used in a multiplex digital PCR (dPCR) assay. HepG2 cDNA samples were diluted 5-fold before dPCR. dPCR was designed to distinguish between cDNA species containing the original adenosine and those containing edited inosine (which is converted to guanidine during cDNA synthesis). dPCR also quantifies the amount of B4GALT1-specific cDNA molecules in the mixture using a primer / probe set targeting exons 1 and 2. The primer and probe sequences are listed in Table 1.

[0129] JPEG2026503580000009.jpg90166

[0130] Digital PCR was performed using a QIAcuity 4 (5-plex) system, a QIAcuity PCR kit, and a 96-well 8.5K nanoplate (Qiagen). A total of 1.2 μL of diluted cDNA mixture was used in a dPCR mixture containing: 3 μL of 4x QIAcuity Mastermix, 0.6 μL per primer (10 μM stock concentration), and 0.3 μL per probe (10 μM stock concentration), plus DNase- and RNase-free water for a total volume of 12 μL. The dPCR mixture was prepared in a preplate and then transferred to a 96-well 8.5K nanoplate and sealed with a nanoplate seal. The plate was then transferred to the QIAcuity 4 instrument. First, a priming and rolling step was performed to form and separate the chamber compartments, followed by an amplification step using the following cycling protocol: 95°C for 2 minutes for enzyme activation, 95°C for 15 seconds for denaturation, and 60°C for 30 seconds for annealing / extension, for 40 cycles. After the amplification step, an image acquisition step was performed for all wells. Data were analyzed using QIAcuity Suite Software (Qiagen).

[0131] The percentage of A to I editing was determined by dividing the number of G-containing segments per ng of RNA by the total (G+A-containing segments) and multiplying by 100.

[0132] The results of RNA editing of the endogenous B4GALT1 transcript are shown in Figure 4 and indicate that, although efficiency varied significantly between EONs, increasing EON concentration from 1 μM to 5 μM resulted in higher editing levels, and importantly, adequate editing levels were obtained when EONs were administered in the presence of AG1856. No editing was observed in negative controls (AG1856 only and non-treated samples (NT)).

[0133] Example 4. Editing of human B4GALT1 transcripts in liver spheroids. Next, we investigated whether RNA editing of endogenous B4GALT1 transcripts could be achieved in hepatic spheroids cultured from primary human hepatocytes. Female primary human hepatocytes (PHH) (BioIVT) were used to generate spheroids. Following the supplier's protocol, PHH cell suspensions (15,000 cells / mL) were seeded into Nuclon Sphera low-attachment U-bottom 96-well plates at 1,500 cells / well. The plates were then incubated at 37°C in a 5% CO2 atmosphere for 5 days.

[0134] After incubation, the spheroids were transferred to a flat-bottom 96-well plate and pooled. Pooling resulted in eight spheroids per well, with a total medium volume of 100 μL per well. To maintain the spheroid cultures, 100 μL of maintenance medium was added to each well and incubated for an additional 48 hours at 37°C in a 5% CO2 atmosphere. The maintenance medium consisted of INVITROGRO Spheroid Maintenance Medium (BioIVT) combined with Spheroid Medium Supplement A, which was also used for spheroid treatment. 100 μL of medium was then removed from each well (containing day 7 spheroids) and 100 μL of treatment medium was added. Each well was treated with 1 or 5 μM EON and 1 μM AG1856 saponin for 72 hours. For this initial spheroid experiment, four of the best performing EONs (see above) were selected: RM4834, RM4838, RM4842, and RM4846. Negative controls were saponin only and NT samples.

[0135] After exposure to EON / saponin, spheroids were harvested and washed once with PBS, followed by the addition of 300 μL of lysis buffer. RNA isolation, RNA yield determination, cDNA synthesis, and editing efficiency using dPCR were performed as described above.

[0136] The results are shown in Figure 5. Consistent with the results observed in HepG2 cells, all four EONs were able to mediate relatively high levels of RNA editing of the endogenous B4GALT1 transcript in PHH-cultured spheroids when co-administered with AG1856, reaching levels above 30%. All four EONs functioned in a similar range.

[0137] Example 5. Editing of human B4GALT1 transcripts in primary human hepatocytes. Next, a new set of EONs was designed based on the best-performing editors from the initial editing screen (EON01 (RM4826), EON05 (RM4830), RM4834, RM4838, RM4842, and RM4846). This new set of EONs and their respective chemical modifications are shown in Figure 6. Some of these EONs differ in their 5'-end portions. This is because some EONs are complementary to exon 6 (i.e., after intron 5 is spliced ​​from the pre-mRNA), while others are complementary to the 5'-end portion bearing intron 5 (i.e., before splicing). See, for example, the difference in the 5'-end portions between B4GALT1-32 (RM106386) and B4GALT1-218 (RM106292). Each of the EONs in Figure 6 contains a triantennary GalNAc modification (L001 = OP-042; Hongene Biotech) at the 5' end to stimulate hepatocyte entry, and is linked to the most terminal 5' nucleotide through a TEG linker (L103). EONs and their adducts were prepared according to standard protocols known to those skilled in the art. The TEG linker was as follows: JPEG2026503580000010.jpg18166

[0138] The new set of EONs was used in conjunction with EON01 (RM4826) and EON05 (RM4830) to examine the editing of endogenous human B4GALT1 transcripts in primary human hepatocytes (PHH). A non-treated (NT) sample was also used as a negative control. For this purpose, a total of 0.5 x 10 5PHH cells (BioIVT) were seeded using Invitrogen CP medium supplemented with TORPEDO Antibiotic Mix. The cells were maintained at 37°C in a 5% CO2 atmosphere. Four hours after seeding, the medium was replaced with Invitrogen HI medium supplemented with TORPEDO Antibiotic Mix and cultured. The next day, the cells were treated with 5 μM EON + 1 μM AG1856 per well. This mixture was left on the cells for 72 hours. Afterwards, the cells were washed once with PBS, and 100 μL of lysis buffer was added. Total RNA was isolated using the Direct-zol™ RNA Microprep Kit (Zymo Research). After adding 100 μL of ethanol (95–100%), the mixture was loaded onto a column and subjected to multiple washing steps and DNase I treatment. After elution in a total volume of 15 μL of DNase / RNase-free water, RNA yields were determined using spectrophotometric analysis (NanoDrop) and stored at −80° C. RT reactions and dPCR were subsequently performed as described above using the indicated primers and probes.

[0139] The results are shown in Figure 7 and generally demonstrate 20-40% editing levels in most EONs. To determine the effect of saponin, we investigated whether RNA editing could be achieved in PHHs without the addition of saponin, i.e., by simply co-incubating oligonucleotides in cell culture medium (i.e., gymnotic uptake, or "gymnosis"). The overall experimental setup was identical to that described above, with the only exception that saponin was not added along with the EONs. All subsequent RNA purification and dPCR steps were similar to those described above.

[0140] The results of this experiment are shown in Figure 8, which shows that the percentage of RNA editing shown after gymnotic incorporation of EON (i.e., without the addition of AG1856) was significantly lower than that shown in Figure 7. This clearly demonstrates the positive effect of co-administration of AG1856 to achieve RNA editing even for a different sequence, in this case the human B4GALT1 transcript.

[0141] Example 6. Editing of target adenosines in endogenous ANGPTL3 target RNA molecules in human Huh-7 cells and liver spheroids. Elevated plasma levels of low-density lipoprotein cholesterol (LDL-C, also known as "bad" cholesterol) are a well-known risk factor for atherosclerotic cardiovascular disease (ASCVD). The strong causal relationship between plasma LDL-C and ASCVD provides the rationale for aggressive LDL-lowering therapy in individuals at high risk for ASCVD. Much attention has focused on the enzyme lipoprotein lipase (LPL), which catalyzes the hydrolysis of plasma triglycerides and is the rate-limiting step for triglyceride uptake into muscle, heart, and adipose tissue. Due to its importance in plasma lipid metabolism, LPL activity in various tissues is tightly regulated to enable lipid uptake in response to local lipid demand. The angiopoietin-like 3 (ANGPTL3) protein is one of the proteins involved in regulating lipoprotein lipase (LPL) activity. Using nucleotide editing techniques to target the gene transcript encoding the angiopoietin-like 3 (ANGPTL3) protein can result in amino acid changes that result in an ANGPTL3 protein with reduced ability to inhibit lipolysis.

[0142] A set of 30 ANGPTL3-targeting EONs (see Figure 9; SEQ ID NOS: 91-120 correspond to RM5035-RM5064, respectively) was tested to assess the level of editing of human ANGPTL3 target (pre-)mRNA in cells after gymnotic incubation with EONs (gymnotic = without any transfection procedure) and in the presence of saponin AG1856. To this end, Huh-7 human hepatocyte-derived cellular carcinoma cells (CLS Cell Lines Service GmbH) were cultured in RPMI 1640 supplemented with 10% FBS / 2 mM L-glutamine and maintained at 37°C in a 5% CO atmosphere. Primary human hepatocytes (BioIVT) were cultured in the corresponding INVITROGRO seeding and maintenance medium (BioIVT) and maintained at 37°C in a 10% CO atmosphere. Human primary hepatocyte-derived hepatic spheroids were generated using the manufacturer's protocol (BioIVT). A total of 0.5 × 10 5 Huh-7 cells were seeded into each well of a 24-well plate one day before exposure to EON. After ON incubation, the seeding medium was aspirated, and the EON mixture in fresh medium was added to the cells. In experiments in which saponin AG1856 was added, 1 μM AG1856 was added to the mixture containing 1 μM EON. Cells were incubated with the inoculum for 72 hours, after which total RNA was isolated. For gymnotic treatment of spheroids derived from human primary hepatocytes, 1.5 × 10 cells were cultured. 3Cells were seeded into 96-well plates in a total volume of 100 μL of seeding medium. The plates were centrifuged at 250 g for 2 minutes to settle the cells to the bottom of the wells. Spheroids were formed over a 5-day incubation period. Before exposure, the medium was aspirated, and a mixture containing 5 μM EON and 1 μM AG1856 in fresh maintenance medium was added. After 72 hours of incubation, the medium was aspirated, and total RNA was isolated. 72 hours after exposure to EON, cells were harvested, and total RNA was isolated from the cells using the Direct-zol RNA Microprep Kit (Zymo Research). After removing the culture medium, the cells were washed once with PBS. After completely aspirating the PBS, 100 μL of TRIreagent (Zymo Research) was added to lyse the cells and recover intracellular material. For spheroids, 300 μL of TRIreagent was used. After adding 100 μL of ethanol (300 μL for spheroid samples), the mixture was loaded onto a column and subjected to multiple wash steps and DNase I treatment. After elution in a total volume of 15 μL of DNase / RNase-free water, RNA yield was determined using spectrophotometry (NanoDrop) and stored at -80°C. Maxima reverse transcriptase (RT, Thermo Fisher Scientific) was used to synthesize cDNA. Typically, 100 ng of total RNA was used in a reaction mixture containing: 4 μL of 5x RT buffer, 1 μL of dNTP mix (10 mM each), 0.5 μL of oligo(dT), 0.5 μL of random hexamers (all Thermo Fisher Scientific), and DNase- and RNase-free water to a total volume of 20 μL. The samples were loaded into a T100 thermal cycler (Bio-Rad) and initially incubated at 25°C for 10 min, followed by a cDNA reaction at 50°C (30 min) and a 5-min stop step at 85°C. The samples were cooled to 4°C and then stored at -20°C.

[0143] To determine editing efficiency, cDNA samples were subjected to multiplex dPCR (Qiagen) assays. The first assay was designed to distinguish between cDNA species containing the original adenosine and those containing edited inosines (which are converted to guanidine during cDNA synthesis and subsequent PCR). The second multiplex ddPCR quantified the amount of ANGPTL3 transcripts by measuring exon 6-7-specific fragments. A separate HPRT1-specific dPCR was performed using an HPRT1-specific primer / probe set to correct for variations in sample isolation or potential effects during exposure. Primer and probe sequences are shown in Table 2, and cycling conditions are shown in Table 3.

[0144] JPEG2026503580000011.jpg90166

[0145] JPEG2026503580000012.jpg42166

[0146] A total of 1 μL of the cDNA mixture was used in a dPCR mixture containing the following: 3 μL of 4× dPCR QIAcuity Mastermix for probes (Qiagen), 0.6 μL of primers and 0.3 μL of probe (all at 10 μM stock concentration), and 4.5 μL of DNase- and RNase-free water for a total volume of 12 μL. The resulting mixture was mixed thoroughly, transferred to a well of a QIAcuity 96-well 8.5K nanoplate (Qiagen), and loaded into the QIAcuity dPCR instrument. Data were analyzed using QIAcuity Suite Software (Qiagen). The percentage of A-to-I editing was determined by dividing the number of G-containing molecules by the total (G+A-containing species) and multiplying by 100.

[0147] Figure 10 shows the percentage of A to I editing determined in human Huh-7 cells after incubation with the 30 EONs shown using the gymnotic method without saponin application, as described above. Most EONs showed detectable but low levels of editing. Figure 11 shows the results of the same experiment, but in this case Huh-7 cells were co-incubated with the triterpene glycoside AG1856. Notably, no editing was observed in the untreated (NT) sample and the AG1856-only control, but editing levels increased dramatically compared to experiments without saponin. Some editing levels reached 60%, clearly demonstrating the beneficial properties of co-administration of saponin. Figure 12 shows the results of an experiment in which the same 30 EONs were tested for A to I editing of the ANGPTL3 transcript in hepatic spheroids generated from primary human hepatocytes. All EON incubations were performed in conjunction with 5 μM AG1856 incubation, and all EONs showed very high editing levels (except RM5041, which is likely due to experimental error as editing of this EON was confirmed in Huh-7 cells), with the editing percentage reaching over 60% of ANGPTL3 transcripts. To confirm the effect of AG1856 saponin, the same experiment was performed in liver spheroids made from primary human hepatocytes using 5 μM incubation of six selected EONs: RM5059 (SEQ ID NO: 115), RM5060 (SEQ ID NO: 116), RM5061 (SEQ ID NO: 117), RM5062 (SEQ ID NO: 118), RM5063 (SEQ ID NO: 119), and RM5064 (SEQ ID NO: 120), but without the addition of saponin. The results are shown in Figure 13 and show that RNA editing of the target adenosine in the endogenous ANGPTL3 transcript was achieved, but to a much lower extent in the absence of saponin, further confirming the beneficial effect of co-administration of saponin.

[0148] Example 7. Editing of target adenosines in endogenous actin B-target RNA molecules in vivo. Similar to Example 2, a study was conducted to investigate the beneficial properties of AG1856 in in vivo RNA editing experiments, this time targeting the endogenous mouse actin B (mActB) target transcript. To this end, female C57BL / 6J mice (n=4-6, 9-16 weeks old) received 5 daily doses of EON RM3891 (10 mg / kg each day, days 1-5), which targets wild-type mouse Actb, or PBS. These were administered with or without a single subcutaneous (SC) dose of AG1856 (3.6 mg / kg on day 5, 1 hour after RM3891 treatment), all administered SC. Animals were sacrificed 3 days after the final dose (=day 8), and tissues were isolated. RM3891 has the following sequence (5'→3'; SEQ ID NO: 132; chemical modifications are as shown in Figure 3): Um!Cm*Cm*Um*Gm*Um*Am*Af*Cm*Cf*Am*Cf*m5UeZd*Ad^UmUf*m5Ue*Cf*Am*Uf*Gm*Gf*Am*Um*Am!Cm

[0149] Snap-frozen tissue samples were thawed and disrupted in TRIzol reagent (Roche) using a MagNA lyzer (Roche). The samples were subjected to two 30-second passes (6500 rpm) with a 90-second cooling period between passes. The samples were then incubated at room temperature for 2 minutes to allow complete dissociation of nuclear proteins. Chloroform was then added to the suspension in preparation for phase separation. After centrifugation at 12000g (4°C) for 15 minutes, the aqueous layer containing RNA was used for further processing. RNA was isolated using the ReliaPrep RNA Cell Miniprep System according to the manufacturer's protocol. cDNA synthesis was performed using the Maxima Reverse Transcriptase Kit (Thermo Scientific) according to the manufacturer's protocol. Briefly, to avoid interference of cDNA synthesis by secondary structures, 500 ng of RNA was first incubated with a dNTP mix (10 mM each), random hexamers, and oligo Dt at 65°C for 5 min, followed by slow cooling to 10°C at a 10°C drop per 15 s. Reverse transcriptase buffer and enzyme were then added, and the sample was incubated at 25°C for 10 min, 50°C for 30 min, and 80°C for 5 min. For dPCR analysis, the cDNA samples were incubated with a mixture of primers and probes specific for wild-type mActB 3′-UTR (HEX), mutant mActB 3′-UTR (FAM), the entire mActB exons 2-3 (Cy5), and the housekeeping gene mRps19 (Table 4). 12 μL of each sample was loaded into a QIAcuity Nanoplate 8.5K 24-well and run on the QIAcuity (Qiagen) including sample splitting, PCR (Table 2), and imaging.

[0150] JPEG2026503580000013.jpg104166

[0151] The results of these in vivo experiments are shown in Figure 14A: liver, Figure 14B: kidney, and Figure 14C: spleen. Whereas no editing was observed in control mice (PBS alone or PBS + AG1856), significant percentage editing of endogenous mActB transcripts was detectable in all three tissues when RM3891 was used alone. Furthermore, when AG1856 was co-administered, it was significantly increased in the liver and kidney, confirming the in vitro results described above and the in vivo results shown above and in Figure 2.

[0152] Example 8. Editing of target adenosines in mApp target RNA molecules in primary mouse hepatocytes using AG1856-EON conjugates. Next, we investigated whether conjugating saponin molecules to RNA-editing oligonucleotides could further improve editing efficiency and / or levels. The first structure prepared is shown in Figure 15, showing saponin (AG1856) conjugated to the 5' end of EON. This was achieved using an N-ε-maleimidocaproic acid hydrazide (EMCH) group pre-attached to the saponin and a CS linker pre-attached to the EON.

[0153] The AG1856-EMCH moiety was prepared as follows: A stock solution of AG1856 (Clochard et al. 2020) in EtOH (672.8 μM) and a stock solution of EMCH hydrazide trifluoroacetate in CHCl3 (17.7 mM) were prepared. EMCH (3 equivalents) was added to the AG1856 solution (1 equivalent) and incubated at room temperature for 24 hours. The reaction was monitored by thin-layer chromatography (CHCl3 / MeOH / H2O / CH3COOH, 50:40:10:5, stain: vanillin). The final product was purified by HPLC (C 18 The crude product was purified by elution with 70% HO (+0.01% TFA) / MeCN (+0.01% TFA) for 10 min, followed by 42% in 5 min, and then 35% in 20 min. The solvent was removed by speedvac and then lyophilized, giving a typical yield of 75%.

[0154] The EON-C6S moiety is generated using standard methods known to those skilled in the art, wherein the thiol modifying group C6 SS of formula (VI): JPEG2026503580000014.jpg19166 was attached to the 5' end of the EON. The EON selected for editing the target adenosine in the mouse App (mApp) transcript (see Example 1) was RM5522, which has the following sequence (5'->3'; SEQ ID NO: 133; chemical modification is as shown in Figure 3, where L101 represents a C6S-S linker, attached to the 5'-terminal 2'-OMe-modified adenosine (Am) by a phosphodiester bond): JPEG2026503580000015.jpg13166

[0155] The conjugation of AG1856-EMCH with EON-C6-6 involves two reaction steps: i) disulfide reduction and ii) Michael addition. The reduction was performed as follows: Degassed PBS buffer (10 mL each) was prepared by sparging with nitrogen for 25–30 min.

[0156] EON-C6S was prepared at 0.164 μmol / 160 μL in degassed PBS, and tris(2-carboxyethyl)phosphine hydrochloride (TCEP; CAS: 51805-45-9) stock solution was prepared at 1.64 μmol / 40 μL in degassed PBS. All stock solutions were flushed with nitrogen until use. 160 μL of RM5522 stock solution (1.0 equiv., 0.164 μmol, 1.5 mg) and 40 μL of TCEP stock solution (10.0 equiv., 1.64 μmol, 0.47 mg) were mixed in a 2 mL reaction tube equipped with a stir bar, purged with nitrogen, and stirred at ambient temperature for approximately 90 minutes. To maintain the reduced state of EON, each tube or filter was thoroughly purged with nitrogen after opening and before reclosing. An Amicon Ultra-4 centrifugal filter (3K) was pre-washed with 2 mL of PBS by centrifugation for approximately 30 minutes. After the reaction, the reaction mixture was diluted with 1600 μL of degassed PBS, and the stir bar was removed. To remove excess TCEP and free thiol residues from the RM5522-SS protecting group, the diluted reaction mixture (1800 μL) was transferred to a pre-washed Amicon Ultra centrifugal filter and centrifuged for approximately 40 minutes. The remaining solution was diluted once more with 1800 μL of degassed PBS and concentrated again by centrifugation for approximately 40 minutes. The remaining solution was diluted to 160 μL with degassed PBS, collected by pipette, and transferred to a Michael addition reaction vessel.

[0157] The Michael addition was carried out as follows. First, a 0.328 μmol / 40 μL stock solution of AG1856-EMCH maleimide was prepared in anhydrous N,N-dimethylformamide (DMF; CAS: 68-12-2) and purged with nitrogen until use. In a 2 mL reaction tube equipped with a seal and a stir bar, EON (post-treatment reaction mixture from the reduction step) bearing a free thiol group was mixed with 40 μL (2.0 equiv., 0.328 μmol, 0.677 mg) of the AG1856-maleimide stock solution, purged with nitrogen, covered with parafilm, and stirred overnight (approximately 15 h) at ambient temperature. PBS, pH 6.9, was prepared by adding HCl to the PBS stock solution. An Amicon Ultra-4 centrifugal filter (3K) was pre-washed with 1.8 mL of PBS (pH 6.9) for 40 min. The reaction mixture was diluted with 1600 μL of PBS (pH 6.9) to reduce the proportion of DMF that the filter membrane cannot tolerate. The diluted reaction mixture was then transferred to an Amicon filter and concentrated for >40 min (to remove DMF and excess maleimide). The remaining solution (approximately 100 μL) was diluted three times with 1800 μL of PBS and concentrated for 45 min. The remaining solution was diluted to 250 μL with PBS (pH 6.9) and collected with a pipette. Mass spectrometry and native PAGE were performed on the crude product using standard procedures known to those skilled in the art, confirming the conjugation of EON to AG1856 saponin (data not shown).

[0158] To determine whether EONs conjugated to saponins (exemplified here by AG1856) and prepared as described above can still cause RNA editing in cells (and thus recruit endogenous ADARs), the conjugates were compared to RM5522 administered alone (and RM3835, see Example 2), and further compared to administering RM5522 and AG1856 separately, as well as to RM5522 alone.

[0159] Mouse hepatocytes were isolated using a Miltenyi Biotec GentleMACS Dissociator and Liver Perfusion Kit (130-128-030) according to the manufacturer's protocol. Briefly, the liver was perfused using the LIPK_HR-1 program on the GentleMACS Dissociator, where it was first washed with predigestion buffer and finally perfused with enzymatic digestion solution. The liver was then transferred to a C-tube on the GentleMACS Dissociator along with the digestion solution from the previous step and dissociated using the LIPK_HR-1 program. To obtain a single-cell suspension, the dissociated liver solution was passed through a 70 μM strainer, carefully centrifuged, and resuspended in DMEM solution (low glucose, glutamine-free, phenol red-free, 5% FBS, and 1% pen / strep). To further enrich for hepatocytes, a debris removal step was performed using density gradient centrifugation with debris removal solution. The supernatant was carefully removed, and the cell pellet was resuspended in DMEM solution. For the A to I editing cell assay, primary mouse hepatocytes were plated onto collagen I-coated 24-well plates at 7.5 × 10 per well. 4 The cells were seeded in 500 μL of DMEM solution at a density of 1000 μg / ml and incubated overnight at 37°C under 5% CO2. Treatments were as follows: 1 μM RM3835 (EON; see Example 2); - 1 μM RM5522(EON); - 1 μM RM5522@AG1856 (EON conjugated with AG1856); - 0.5 μM RM5522@AG1856; - 1 μM RM5522 + 0.5 μM AG1856 (EON + AG1856 combination treatment); and - 0.5μM RM5522+0.2665μM AG1856.

[0160] These treatment solutions were prepared in Williams E medium containing 2 nM glutamine and 1% pen / strep. After 24 hours of cell stabilization, the medium was removed, and 500 μL of the above treatment solution was added to each well. Untreated cells served as a control. The cells were then incubated at 37°C / 5% CO2 for 72 hours before RNA isolation. The RNA isolation procedure, cDNA synthesis, dPCR, and primers and conditions were as described in Example 2. The results of RNA editing assessment following the described treatments are shown in Figure 16. The percentage of editing detected after treatment with RM3835 and RM5522 alone (=gymnotic treatment) was approximately 10-15%. However, the percentage editing obtained using the RM5522@AG1856 conjugate compound at 1.0 μM or 0.5 μM reached a surprisingly significant level of nearly 100%, indicating that not only were all cells likely affected by the conjugate, but that all mApp transcripts in the target cells were targeted and all target adenosines were deaminated. Notably, when RM5522 EON was administered simultaneously with (but not conjugated to) AG1856 at a concentration of 1 μM, editing levels were significantly higher than when AG1856 was not co-administered, at approximately 25%, which was significantly lower than that observed with the conjugate. These results indicate that co-administration of saponin (exemplified herein by AG1856) with EON enhances the RNA editing effect mediated by EON, but when EON is conjugated to saponin, the editing is nearly complete, reaching 100% in vitro, potentially resulting in a much higher percentage of editing in vivo compared to that observed when saponin and EON are administered in an unconjugated state.In this paper, researchers have shown that RNA editing using chemically modified EONs can be increased not only in vitro but also in vivo when saponin is co-administered, resulting in significant editing levels in organs such as the liver and kidney; and most importantly, when EONs are conjugated to EONs, editing levels can be dramatically increased. Importantly, this indicates that AG1856 does not interfere with the enzymatic deamination activity of endogenous ADAR enzymes recruited into cells while bound to oligonucleotides.

Claims

1. 1. A composition comprising a triterpene glycoside and an antisense oligonucleotide (EON) that induces RNA editing, the EON is capable of forming a double-stranded complex with a region of a target RNA molecule in a cell; the region of the target RNA molecule includes a target adenosine; the nucleotide in the EON opposite the target adenosine is a lone nucleotide; and The double-stranded complex is capable of binding to an endogenous ADAR enzyme to deaminate the target adenosine to inosine, thereby editing the target RNA molecule. composition.

2. 2. The composition of claim 1, The triterpene glycoside is conjugated to the EON. composition.

3. 3. The composition according to claim 1 or 2, The triterpene glycoside is AG1856. composition.

4. The composition according to any one of claims 1 to 3, the isolated nucleotide is cytidine, a cytidine analog, a cytidine derivative, a uridine, a uridine analog, or a uridine derivative; composition.

5. The composition according to any one of claims 1 to 4, At least one nucleotide in the EON comprises one or more non-natural chemical modifications in a ribose moiety, a linker moiety, or a base moiety; provided that the isolated nucleotide is not a cytidine containing a 2'-OMe ribose substitution. composition.

6. The composition according to any one of claims 1 to 5, the isolated nucleotide is a deoxynucleotide; composition.

7. The composition according to any one of claims 1 to 6, The target RNA molecule is pre-mRNA or mRNA. composition.

8. The composition according to any one of claims 1 to 7, The endogenous ADAR enzyme is human ADAR1, ADAR2, or ADAT. composition.

9. A composition according to any one of claims 1 to 8, for use in the treatment of cardiovascular disease, a disease involving the liver, a disease involving the kidney, a disease involving the pancreas, or a disorder of the central nervous system; composition.

10. The composition according to any one of claims 1 to 9, the target RNA molecule is endogenously present in the cell; and the target RNA molecule is a human gene selected from the group consisting of: SERPINA1, IDUA, HFE, ABCA4, USH2A, PCSK9, B4GALT1, ALDH2, HTT, DMD, PNPLA3, APOC3, C9orf72, DMPK, RHO, MAPT, OTOF, SMN1, ASL, APP, ANGPTL3, NTCP, PMP22, LRRK2, ASS1, GJB2, MECP2, and RS1; Transcribed from composition.

11. 1. A component kit comprising: (i) a first pharmaceutical composition comprising a triterpene glycoside, preferably AG1856; and (ii) a second pharmaceutical composition comprising an EON; the EON is capable of forming a double-stranded complex with a region of a target RNA molecule in a cell; the region of the target RNA molecule includes a target adenosine; and The double-stranded complex can recruit an endogenous ADAR enzyme to deaminate the target adenosine to inosine, thereby editing the target RNA molecule; Including, Component kit.

12. 1. A method for editing a target adenosine present in an endogenous target RNA molecule in a cell in a subject, comprising: The method includes the following steps: (i) administering to said subject a triterpene glycoside, preferably AG1856; and (ii) administering to said subject an EON; wherein the EON after administration can form a double-stranded complex with a region of the endogenous target RNA molecule containing the target adenosine in the cell; and The double-stranded complex can recruit the endogenous ADAR enzyme to deaminate the target adenosine to inosine; Including, method.

13. 13. The method of claim 12, The triterpene glycoside is conjugated to the EON. method.

14. 14. The method according to claim 12 or 13, The cell is a liver cell, a kidney cell, or a neuron. method.

15. 1. A method for deaminating a target adenosine in a target RNA molecule, preferably a pre-mRNA or mRNA molecule, in a cell, comprising: The method includes the following steps: (i) providing to said cell a conjugate comprising: - triterpene glycosides, preferably AG1856; and an EON linked at its 3' or 5' end to said triterpene glycoside, wherein said EON is capable of forming a double-stranded complex with said target RNA molecule or a region thereof, said region including said target adenosine; (ii) allowing the conjugate to be taken up by the cells; (iii) annealing the EON to the target RNA molecule; (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine to inosine in the target RNA molecule; and (v) identifying the presence of said inosine within said target RNA molecule. may include method.

Images