Antisense oligonucleotides for treatment of liver diseases

Abstract

The present invention relates to antisense oligonucleotides (AONs) that can mediate RNA editing by binding to a target RNA nucleic acid molecule (preferably an RNA transcript molecule) in a cell and recruiting an endogenous deaminase in the cell to deaminize one or more target adenosine nucleotides in the target RNA molecule to inosine. The target RNA molecule is a transcript molecule of the SLC10A1 gene encoding a Na < + > / taurocholic acid co-transporter polypeptide (NTCP). The RNA editing of one or more target adenosines will result in a loss of function of the NTCP protein, which will result in a reduced bile acid uptake circulating from the portal vein to the liver, thereby reducing the risk of developing conditions associated with bile accumulation in the liver.

Description

Technical Field

[0001] This invention relates to the medical field, and more particularly to the field of (chronic) liver diseases, such as cirrhosis, which is caused, for example, by non-alcoholic fatty liver disease (NAFLD) and cholestasis. This invention describes antisense oligonucleotides that mediate nucleotide-specific RNA editing in the SLC10A1 gene transcript to achieve the encoded Na+ + Amino acid alterations in taurocholic acid cotransport polypeptide (NTCP) can affect its activity. Background Technology

[0002] Cholestatic diseases are caused by the accumulation of bile acids in the liver due to bile duct dysfunction, leading to liver cell damage. The consequences of these diseases can be devastating and severely impact quality of life. Clinical manifestations include itching, dry skin, fatigue, pain, and weight loss. Without treatment, the damage progresses through various stages, from liver fibrosis to cirrhosis, ultimately leading to liver failure and increasing the risk of liver cancer. Primary sclerosing cholangitis (PSC) and biliary atresia (BA) are two forms of cholestatic diseases with high unmet medical needs, often requiring liver transplantation. PSC is an inflammatory disease that is typically diagnosed in people aged 30 to 40, with a higher incidence in men (66%). It is estimated that approximately 80,000 people in North America and Europe have PSC, with a prevalence of 1 to 9 per 100,000. The disease causes bile duct fibrosis and sclerosis, resulting in the toxic accumulation of bile acids in the liver. BA is a pediatric disease affecting newborns, caused by the absence or defect of bile ducts. This disease causes harmful bile acids to accumulate in the liver, which can rapidly progress to cirrhosis in childhood. It is estimated that about 20,000 people in North America and Europe have BA, and the prevalence in the Western world is 1 in 10,000 to 15,000 newborns.

[0003] Currently, there are no approved drugs for treating PSC or BA. For PSC, liver transplantation is the only treatment option with evidence of prolonged survival. However, 20% to 40% of patients who receive a liver transplant experience PSC recurrence, and the median survival for those who do not receive a transplant is only about 21 years. The standard treatment for BA is surgery performed in the first few weeks after birth. However, most patients who undergo this surgery still require a liver transplant in early childhood.

[0004] Hepatocytes in the liver primarily acquire bile acids through the enterohepatic reuptake cycle. The process of bile acid uptake from the portal circulation into hepatocytes is mainly facilitated by Na+ encoded by the SLC10A1 gene. + / Taurocholic acid cotransport polypeptide (NTCP) is involved. NTCP protein has received much attention because it has been identified as a major protein for recognition and entry by hepatitis B virus (HBV) and hepatitis D virus (HDV). However, the normal function of NTCP is at least related to the uptake of conjugated bile acids from circulation into hepatocytes. As previously mentioned, the accumulation of bile acids in the liver is associated with a variety of liver diseases, and studies have shown that inhibiting NTCP can improve liver function by reducing the level of toxic bile acids in hepatocytes and preventing the aggravation of liver damage through certain available markers such as fibrosis, bile duct cell proliferation, alkaline phosphatase (ALP), alanine aminotransferase (ALT), and inflammatory biomarkers such as cytokines (Slijepcevic D and Van de Graaf 2017. DigDis.35(3):251-258; Slijepcevic D et al. 2018. Hepatology 68(3):1057-1069).

[0005] This invention relates to a completely different method for reducing NTCP activity, namely, by using antisense oligonucleotides (AONs) and the cell's own nucleic acid editing mechanisms to specifically target and modify one or more nucleotides in the SLC10A1 transcript, thereby providing a loss-of-function NTCP protein. This reduces the ability of hepatocytes to take up bile acids from the portal circulation, thereby treating conditions associated with bile accumulation in the liver. The technology involved in this invention is commonly referred to as "RNA editing."

[0006] RNA editing is a natural process by which eukaryotic cells alter the sequence of their RNA molecules in a site-specific and precise manner, thereby expanding the RNA library encoded by the genome by orders of magnitude. RNA editing enzymes have been described in eukaryotes throughout the animal and plant kingdoms, and these processes play a crucial role in maintaining cellular homeostasis in metazoans ranging from the simplest life forms (such as *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, respectively, by enzymes called RNA-acting adenosine deaminase (ADAR) and APOBEC / AID (RNA-acting cytidine deaminase).

[0007] ADAR is a multidomain protein containing one catalytic domain and two to three double-stranded RNA recognition domains (depending on the specific enzyme). Each recognition domain recognizes a specific double-stranded RNA (dsRNA) sequence and / or conformation. The catalytic domain also plays a role in recognizing and binding a portion of the dsRNA helix, but its key function is to convert an A-to-I sequence at a predetermined position in the target RNA via nucleotide deamination. Inosine is read as guanosine by the cell's translation mechanisms, meaning that if the edited adenosine is located in the coding region of mRNA or pre-mRNA, it can re-encode a protein sequence. A-to-I conversion can also occur in the 5' non-coding sequence of the target mRNA, creating a new translation initiation site upstream of the original initiation site, resulting in an N-terminal extended protein; or in the 3' untranslated region (UTR) or other non-coding regions of the transcript, affecting RNA processing and / or stability. Furthermore, A-to-I conversion can occur in splicing elements of pre-mRNA introns or exons, altering the splicing pattern. As a result, exons may be included or skipped. Enzymes that catalyze the deamination of adenosine belong to the ADAR enzyme family, including human deaminases hADAR1, hADAR2, and hADAR3. However, hADAR3 has not yet been shown to possess deaminase activity.

[0008] The use of oligonucleotides to edit target RNA with adenosine deaminase has been described (e.g., Woolf et al., Proc Natl Acad Sci USA 1995, 92:8298-8302; Montiel-Gonzalez et al., Proc Natl Acad Sci USA 2013, 110(45):18285–18290; Vogel et al., Angewandte Chemie Int 2014, Ed 53:267-271). A drawback of the method described by Montiel-Gonzalez et al. (2013) is the requirement for a fusion protein formed by fusing the boxB recognition domain of a phage λN protein with the adenosine deaminase domain of a truncated native ADAR protein. This necessitates either transduction of the target cell by the fusion protein (a major obstacle) or transfection with a nucleic acid construct encoding the engineered adenosine deaminase fusion protein to achieve expression. The system described by Vogel et al. (2014) has similar limitations because it is unclear how to apply the system without first genetically modifying ADAR and then transfecting or transforming cells carrying the target RNA to provide cells with this genetically engineered protein. A similar system is described in US 9,650,627. The oligonucleotides of Woolf et al. (1995) are 100% complementary to the target RNA sequence, but suffer from a serious lack of specificity: almost all adenosines complementary to AON in the target RNA chain are edited.

[0009] ADAR is known to act on any dsRNA. Through a process sometimes referred to as “promiscuous editing,” the enzyme edits multiple A's in the dsRNA. Therefore, methods and means are needed to avoid this promiscuous editing and target only specific adenosines in the target RNA molecule for therapeutic applications. Vogel et al. (2014) showed that this off-target editing can be inhibited by using a 2'-O-methyl (2'-OMe) modified nucleoside at a position in the oligonucleotide opposite to the adenosine that should not be edited, and an unmodified nucleoside at a position directly opposite the adenosine specifically targeted on the target RNA. However, specific editing effects at the target nucleotide have not been shown without the use of recombinant ADAR enzymes covalently bound to AON. Several publications now demonstrate that recruiting endogenous ADAR (and thus without the need for exogenous and / or recombinant sources) while maintaining specificity, i.e., targeting a single adenosine in the target RNA molecule and deaminoing it to inosine, is feasible. WO2016 / 097212 discloses an AON for RNA-targeted editing, characterized by a sequence complementary to the target RNA sequence (referred to herein as the "targeting moiety") and the presence of a stem-loop / hairpin structure (referred to herein as the "recruiting moiety"), which preferably is not complementary to the target RNA. This oligonucleotide is called a "self-circularized AON". The recruitment moiety functions to recruit naturally occurring (endogenously present) ADAR enzymes in the cell to dsRNA formed by hybridization of the target sequence and the targeting moiety. Due to the recruitment moiety, the presence of a conjugated entity or modified recombinant ADAR enzyme is unnecessary. WO2016 / 097212 describes the recruitment moiety as a stem-loop structure mimicking the structure of a natural substrate (e.g., the GluB receptor) or Z-DNA, known to be recognized by the dsRNA-binding domain or Z-DNA-binding domain of an ADAR enzyme. 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 by a single nucleic acid strand). The described stem-loop structure of the recruitment portion is an intramolecular stem-loop structure, formed within the AON itself, and is believed to attract (endogenous) ADAR. Subsequently, similar RNA editing systems containing stem-loop structures have been described in WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, US11,390,865, WO2020 / 246560, and WO2022 / 078995.

[0010] WO2017 / 220751 and WO2018 / 041973 describe next-generation types of AONs that do not contain this stem-loop structure but are (almost completely) complementary to the target region and appear to still attract endogenous ADAR enzymes. In one embodiment, there are one or more mismatched nucleotides, wobble, or bumps between the oligonucleotide and the target sequence. The unique mismatch may be located at the nucleoside site opposite the target adenosine, but in other embodiments, AONs (or “RNA-editing oligonucleotides”—although the deamination reaction is performed by ADAR enzymes—often abbreviated as “EONs”) are described as having multiple bumps and / or wobbles when binding to the target sequence region. When the sequence of the AON is carefully selected to attract / recruit ADAR, RNA editing appears to be achievable in vitro, ex vivo, and in vivo using AONs lacking stem-loop structures and endogenous ADAR enzymes. An “orphan nucleoside” is defined as the nucleoside in the AON that is directly opposite the target adenosine in the target RNA molecule, and is a nucleotide with an unmodified cytosine nucleobase and not carrying a 2′-OMe modification. Orphan nucleotides can be deoxyribonucleosides (DNA), where the remainder of the AON may still carry a 2'-O-alkyl modification (such as 2'-OMe) on the sugar moiety, or the nucleotides directly surrounding the orphan nucleotide may contain chemical modifications (such as DNA compared to RNA) to further improve RNA editing efficiency and / or increase resistance to nucleases. These effects can be further improved by using positive oligonucleotides (SON) to “protect” the AON from degradation after delivery to the cell (described in WO2018 / 134301 and US11,274,300).

[0011] The use of chemical modifications and specific structures in oligonucleotides that can be used for specific adenosine editing in ADAR-mediated target RNAs has become a widely disclosed topic in the field, as illustrated in WO2019 / 111957, WO2019 / 158475, WO2020 / 165077, WO2020 / 201406, WO2020 / 211780, WO2021 / 008447, WO2021 / 020550, and WO2021 / 0 60527、WO2021 / 117729、WO2021 / 136408、WO2021 / 182474、WO2021 / 216853、WO2021 / 242778、WO20 21 / 242870, WO2021 / 242889, WO2022 / 007803, WO2022 / 018207, WO2022 / 026928 and WO2022 / 124345.The use of specific sugar moieties has been disclosed in, for example, WO2020 / 154342, WO2020 / 154343, WO2020 / 154344, WO2022 / 103839, and WO2022 / 103852, while the use of stereodefined linker moieties (typically for oligonucleotides that can be used for, for example, exon skipping, interstitial bodies, siRNA, or specific RNA editing oligonucleotides involving multiple target sequences) has been disclosed in WO2011 / 005761, WO2014 / 010250, WO2014 / 012081, and 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(PN PLA3), WO2018 / 237194, WO2019 / 032607(C9orf72), WO2019 / 055951, WO2019 / 075357(SMA / ALS), WO2019 / 200185(DM1), WO 2019 / 217784(DM1), WO2019 / 219581, WO2020 / 118246(DM1), WO2020 / 160336(HTT), WO2020 / 191252, WO2020 / 196662, WO20 The following publications describe this: WO2020 / 219981 (USH2A), WO2020 / 219983 (RHO), WO2020 / 227691 (C9orf72), WO2021 / 071788 (C9orf72), WO2021 / 071858, WO2021 / 178237 (MAPT), WO2021 / 234459, WO2021 / 237223, WO2022 / 099159, WO2021 / 030778, WO2022 / 174053, and WO2023 / 278589. In addition to these publications, numerous other publications address the targeting of specific RNA target molecules or specific adenosines within such RNA target molecules, whether for the repair of mutations leading to premature stop codons or other pathogenic mutations.Publicly disclosed examples of targeting adenosine in specific target RNA 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 / 231679 (G JB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 and WO2021 / 231692 (OTOF, autosomal recessive nonsyndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (arginine succinate lyase deficiency); WO2021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1).

[0012] The present invention aims to provide one or more alternative and / or improved compounds or compositions for treating liver diseases, such as cholestatic diseases caused by the accumulation of bile acids in the liver. Summary of the Invention

[0013] This paper discloses an antisense oligonucleotide (AON) that can recruit endogenous ADAR enzymes in human cells, provided that the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in the cell, wherein the region contains target adenosine, and the nucleotide in the AON corresponding to the target adenosine is an orphan nucleotide. The ADAR enzyme can deaminate the target adenosine to inosine after binding to the double-stranded complex, and the target RNA nucleic acid molecule encodes Na+. +The transcript molecule of the human SLC10A1 gene containing taurocholic acid cotransport polypeptide (NTCP). Preferably, the transcript molecule is pre-mRNA or mRNA. Preferably, the cell is a liver cell, more preferably a hepatocyte. In one aspect, the SLC10A1 gene is wild-type, and the target adenosine is selected from: (i) adenosine in the CAG codon encoding glutamine (Q) at position 68 of the NTCP protein, wherein deamination of the adenosine results in the amino acid being arginine (R); (ii) the first adenosine in the CAA codon encoding glutamine (Q) at position 261 of the NTCP protein, wherein deamination of the adenosine results in the amino acid being arginine (R); (iii) adenosine in the GAG ​​codon encoding glutamate (E) at position 257 of the NTCP protein, wherein deamination of the adenosine results in the amino acid being glycine (G); and (iv) the first adenosine in the AAG codon encoding lysine (K) at position 314 of the NTCP protein, wherein deamination of the adenosine results in the amino acid being glutamate (E); wherein deamination of the target adenosine results in impaired function of the NTCP protein in transporting bile acids from the portal circulation into the cell. In another aspect, the orphan nucleotide is deoxycytidine or deoxyuridine. In one aspect, orphan nucleotides are cytidine analogs, such as deoxynucleotides containing a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase. In another aspect, orphan nucleotides are uridine analogs, such as deoxynucleotides containing an isouriacritic nucleobase. In one aspect, the nucleotides in AON are numbered such that the orphan nucleotide is numbered 0, increasing positively (+) towards the 5' end and negatively (-) towards the 3' end, and when the nucleotide opposite the first nucleotide (-1 position) at the 3' end of the target RNA nucleic acid molecule is cytidine, the nucleotide at that position in the orphan nucleotide' is deoxyinosine.

[0014] An AON is disclosed, wherein the AON comprises modifications of one or more linking portions, each modification being independently selected from phosphate thioester (PS), phosphonoacetate, dithiophosphate, methylphosphonate (MP), sulfonylphosphonamide, (1,3-dimethylimidazoline-2-yl)phosphonamide (PNdmi) linking bonds or PNms linking bonds having a structure according to formula (I), wherein: X = O or S; and R = aryl, substituted aryl, heterocycle, substituted heterocycle, aromatic heterocycle, substituted aromatic heterocycle, C1-C6 alkoxy, substituted C1-C6 alkoxy, C1-C 20 Alkyl, substituted C1-C 20 Alkyl, C1-C6 alkenyl, C1-C6 substituted alkenyl, C1-C6 alkynyl, substituted C1-C6 alkynyl or conjugated group.

[0015] An AON is disclosed, wherein the internucleotide linker bond numbering in the AON is such that linker bond number 0 is the linker bond at the 5' end of the orphan nucleotide, the linker bond positions in the oligonucleotide increase positively (+) towards the 5' end and negatively (-) towards the 3' end, and the linker bond position -2 is an MP linker bond.

[0016] Disclosed is an AON comprising one or more nucleotides having a single or double substitution at the 2′, 3′ and / or 5′ position of the ribose, each substitution being independently selected from: -OH; -F; substituted or unsubstituted linear or branched lower (C1-C1) nucleotides. 10 Alkyl, alkenyl, alkynyl, alkylaryl, allyl, 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-allyl; -O-alkyl; -O-alkyl; -methoxy; -aminopropoxy; -methoxyethoxy; -dimethylaminooxyethoxy; and -dimethylaminoethoxyethoxy. In one aspect, AON is partially (non)covalently bonded to GalNAc.

[0017] This document discloses pharmaceutical compositions comprising the AON disclosed herein or the carriers disclosed herein, as well as pharmaceutically acceptable carriers. This document discloses AONs for treating diseases caused by cholestasis in the liver, such as cholestasis, primary sclerosing cholangitis (PSC), biliary atresia (BA), and cirrhosis. This document also discloses the use of AONs in the manufacture of medicaments for treating diseases caused by cholestasis in the liver, such as cholestasis, PSC, BA, and cirrhosis.

[0018] This article discloses a method for editing human SLC10A1 polynucleotide in cells (preferably hepatocytes), wherein the human SLC10A1 polynucleotide is a pre-mRNA or mRNA molecule. The method involves contacting the SLC10A1 polynucleotide with an AON that can trigger ADAR-mediated adenosine to inosine deamination, thereby editing the SLC10A1 polynucleotide to encode an NTCP protein with impaired, reduced, or lost bile acid uptake function.

[0019] This article discloses a method for treating, improving, or slowing the progression of diseases (such as cholestasis, PSC, BA, and cirrhosis) caused by cholestasis in the liver in human subjects in need. The method comprises administering the AON disclosed herein, the carrier disclosed herein, or the pharmaceutical composition disclosed herein to the subject, thereby contacting the SLC10A1 polynucleotide in the subject's cells with an AON capable of achieving ADAR-mediated adenosine-to-inosine deamination, thereby editing the SLC10A1 polynucleotide to encode an NTCP protein with impaired, reduced, or lost bile acid uptake function, thereby treating the subject. Attached Figure Description

[0020] One or more embodiments of the invention will now be described by way of example only, with reference to the accompanying drawings, wherein:

[0021] Figure 1 The partial 5' to 3' sequence of the human SLC10A1 mRNA transcript (SEQ ID NO:1) is shown, with the CAG codon encoding glutamine (Q) at position 68 of the NTCP protein in bold. The underlined adenosine is the RNA editing target disclosed herein, which, upon editing, generates a codon (CIG / CGG) encoding an arginine (R) residue at this position. Below the target sequence are the 5' to 3' sequences of 30 initial AONs designed to target the target adenosine in SEQ ID NO:1. AONs T1-01 to T1-30 are SEQ ID NO:5 to SEQ ID NO:34, as shown in the figure. The chemical modifications are as follows: Ae and Ge are 2'-MOE-modified adenosine and guanosine, respectively; Cm, Am, Um, and Gm are 2'-OMe-modified cytidine, adenosine, uridine, and guanosine, respectively; Gf, Cf, Af, and Uf are 2'-F-modified guanosine, cytidine, adenosine, and uridine, respectively; m5Ce is 2'-MOE-modified 5-methylcytidine; Zd (orphan nucleotide) is a deoxynucleotide carrying a Benner base (a deoxycytidine analog); Id is deoxyinosine; Te (or m5Ue) is 2'-MOE-modified thymidine (identical to 5-methyluridine with 2'-MOE substitution); "!" indicates a PNdmi linker; "^" indicates an MP linker; "*" indicates a PS linker. All other internucleotide links are phosphodiester (PO) links.

[0022] Figure 2 The partial sequence of the human SLC10A1 mRNA transcript (SEQ ID NO:2) is shown, with the CAA codon encoding glutamine (Q) at position 261 of the NTCP protein in bold. The underlined adenosine is the RNA editing target disclosed herein, which, after editing, generates a codon encoding an arginine (R) residue (CIA / CGA) at this position. Below the target sequence are the 5′ to 3′ sequences of 30 initial AONs designed to target the adenosine in SEQ ID NO:2. AONs T2-01 to T2-30 are SEQ ID NO:35 to SEQ ID NO:64, as shown in the figure. Chemical modifications are as follows. Figure 1 As shown.

[0023] Figure 3The partial sequence of the human SLC10A1 mRNA transcript (SEQ ID NO:3) is shown, with the GAG ​​codon encoding glutamate (E) at position 257 of the NTCP protein in bold. The underlined adenosine is the RNA editing target disclosed herein, which, after editing, generates a codon encoding a glycine (G) residue at this position (GIG / GGG). Below the target sequence are the 5′ to 3′ sequences of 30 initial AONs designed to target the adenosine in SEQ ID NO:3. AONs T3-01 to T3-30 are SEQ ID NO:65 to SEQ ID NO:94, as shown in the figure. Chemical modifications are as follows. Figure 1 As shown.

[0024] Figure 4 The partial sequence of the human SLC10A1 mRNA transcript (SEQ ID NO:4) is shown, with the AAG codon encoding lysine (K) at position 314 of the NTCP protein in bold. The underlined adenosine is the RNA editing target disclosed herein, which, after editing, generates a codon (IAG / GAG) encoding a glutamate (E) residue at this position. Below the target sequence are the 5′ to 3′ sequences of 30 initial AONs designed to target the adenosine in SEQ ID NO:4. AONs T4-01 to T4-30 are SEQ ID NO:95 to SEQ ID NO:124, as shown in the figure. Chemical modifications are as follows. Figure 1 As shown.

[0025] Figure 5 This shows another group of AONs designed to target adenosine in SEQ ID NO:1 (Q68R). The corresponding SEQ ID NO is shown in the figure. Chemical modifications are as follows. Figure 1 As given, the symbol "#" represents the PNms connection key, and the symbol "θ" represents the PO connection key.

[0026] Figure 6 This shows another group of AONs designed to target adenosine in SEQ ID NO:3 (E257G). The corresponding SEQ ID NO is shown in the figure. Chemical modifications are as follows. Figure 1 As given.

[0027] Figure 7 A and Figure 7 B represents the percentage of c.203A>G editing (Q68R) measured in two independent experiments after co-treatment with AG1856 saponin in primary human hepatocytes (PHH) using 16 AONs as shown in the figure. Two unrelated AONs (RM4777 and RM4266), AG1856 alone, and untreated samples served as negative controls.

[0028] Figure 8 A and Figure 8 The values ​​of B are shown in two independent experiments, representing the percentage of c.770A>G editing (E257G) measured after co-treatment with AG1856 saponin in PHH using 17 AONs as shown in the figure. Two unrelated AONs (RM4777 and RM4266), AG1856 alone, and untreated samples served as negative controls.

[0029] Figure 9 A and Figure 9 The values ​​of B are shown in two independent naked uptake experiments, using the 18 AONs shown in the figure in saponin-free PHH, to measure the percentage of c.203A>G editing (Q68R). Two unrelated AONs (RM4777 and RM4266) and the untreated sample were used as negative controls.

[0030] Figure 10 A and Figure 10 The figures for B are shown in two independent naked uptake experiments, using the 16 AONs as shown in the figure in saponin-free PHH, to measure the percentage of individual c.770A>G editing (E257G). Two unrelated AONs (RM4777 and RM4266) and the untreated sample were used as negative controls.

[0031] Figure 11 The results showed that in both experiments, co-treatment with AG1856 saponins in hepatocytes using 16 types of AONs as shown in the figure ( Figure 11 A) and not co-treated with saponins ( Figure 11 The percentage of c.203A>G edits (Q68R) measured in B). Untreated samples and AG1856 alone served as negative controls.

[0032] Figure 12 The results showed that in both experiments, co-treatment with AG1856 saponins in hepatocytes using 17 types of AONs as shown in the figure ( Figure 12 A) and not co-treated with saponins ( Figure 12 The percentage of edits (C.770A>G edits, E257G) measured in sample B). Untreated samples and AG1856 alone served as negative controls.

[0033] Figure 13 The figure shows the percentage of c.203A>G editing (Q68R) measured after co-treatment with AG1856 saponin in HepG2 cells (HepG2NTCP) overexpressing human NTCP using 11 AONs as shown in the figure. Untreated samples served as negative controls.

[0034] Figure 14This figure shows the percentage of c.770A>G editing (E257G) measured after co-treatment with AG1856 saponin in HepG2NTCP cells using 10 AONs as shown in the figure. Untreated samples and AG1856-only samples served as negative controls.

[0035] Figure 15 shows another experiment using 25 asymmetric AONs as shown in the figure. Figure 15A ) and using the 34 types of symmetrical AONs shown in the figure. Figure 15B The percentage of c.203A>G editing (Q68R) after co-treatment with AG1856 saponin in PHH was measured. Irrelevant AON (RM4777) and untreated samples served as negative controls.

[0036] Figure 16 shows another experiment using 21 asymmetric AONs as shown in the figure. Figure 16A ) and using the 38 symmetrical AONs shown in the figure. Figure 16B The percentage of c.770A>G edits (E257G) was measured after co-treatment with AG1856 saponin in PHH. Irrelevant AON (RM4777) and untreated samples served as negative controls.

[0037] Figure 17 The results show that in human U2OS cells, after (independent) transfection with plasmids expressing seven human NTCP mutants, as shown in the figure, at 1.0 μM TCA ( Figure 17 A) and 10μM TCA ( Figure 17 The uptake (pmol) of radiolabeled taurocholic acid (TCA; bile acid) under treatment B). Wild-type NTCP expression plasmid was used as a positive control, and untreated samples were used as negative controls.

[0038] Figure 18 This shows another group of AONs designed to target adenosine in SEQ ID NO:1 (Q68R), using multiple linker bonds to replace the linker bonds in the earlier AON group. The corresponding SEQ ID NO is shown in the second column. Chemical modifications such as... Figure 1 As given, the symbol "#" represents the PNms connection key, and the symbol "θ" represents the PO connection key.

[0039] Figure 19 The c.203A>G editing (Q68R) experiment shows the use of 12 AONs as shown in the figure (see details). Figure 18 Percentage of edits after transfection with liposomes in PHH. AON (RM4777), liposome-only, and untreated (NT) PBS samples served as negative controls.

[0040] Figure 20The experiment shown in c.203A>G edit (Q68R) uses 24 types of AON as shown in the figure (see details). Figure 18 Percentage of edits after transfection with liposomes in PHH. Irrelevant AON (CNTRL), liposome-only (Lipo2000), and untreated (NT) PBS samples served as negative controls.

[0041] Figure 21 The experiment shown in c.203A>G edit (Q68R) uses 9 types of AON as shown in the figure (see details). Figure 18 The percentage of edits after co-treatment with AG1856 in PHH. Irrelevant AON (control-ON), saponin-only (AG1856), and untreated (NT) samples served as negative controls.

[0042] Figure 22 The experiment shown in c.203A>G edit (Q68R) uses 26 types of AON as shown in the figure (see details). Figure 18 Percentage of edits after liposome transfection in PHH. Irrelevant control (RM4777), liposome-only (MOCK Lipo), and untreated (NT) samples served as negative controls. Detailed Implementation

[0043] The AON disclosed herein can recruit endogenously present deaminases in cells, such as ADAR1 and / or ADAR2. Upon binding to a target RNA molecule, the AON disclosed herein can mediate RNA editing of the target adenosine within the target RNA, as the deaminase is recruited to the double-stranded AON / target RNA molecule complex and subsequently deaminates the target adenosine to inosine.

[0044] In this article, oligonucleotides are abbreviated as "AON," but are sometimes also referred to as "editing oligonucleotides" or "EON," although RNA editing events are performed by deaminases, and the role of oligonucleotides is only to trigger RNA editing. There is a growing need to improve the pharmacokinetic properties of AONs without negatively impacting the editing efficiency of target adenosine in the target RNA, and / or without negatively impacting the stability of the AON itself (AONs are readily degraded by nucleases present in natural cells). Many chemical modifications can be used to generate AONs (and many are already in use in this field). However, many of these properties are not always compatible with the expectation of achieving efficient RNA editing. In the search for better pharmacokinetic properties, it was early discovered that 2′-O-methoxyethyl (or 2′-methoxyethoxy, or 2′-MOE) modifications of the ribose in some (but not all) nucleotides were unexpectedly compatible with efficient ADAR linking and editing (WO2019 / 158475). Similarly, it was early discovered that PS linkages between some (but not all) nucleosides were unexpectedly compatible with efficient ADAR linking and editing (WO2019 / 219581). Furthermore, early findings indicated that phosphonoacetate linkage modifications and / or unlocking of nucleic acid (UNA) ribose modifications at some (but not all) positions in AON are compatible with efficient linking and subsequent deamination of enzymes with nucleotide deamination activity (WO2020 / 165077). Although the properties of phosphonoacetate and UNA modifications are known in themselves, their compatibility with linking and deamination reactions of enzymes with nucleotide deamination activity was unknown.

[0045] This document discloses an AON that can provide (mediate, induce, or trigger) RNA editing of target adenosine in target transcript molecules (such as pre-mRNA and / or mRNA). The target transcript molecule may be encoded by a mutated gene, wherein the mutation is the cause of disease, and wherein the editing can reverse the mutation to produce a wild-type protein or a protein with wild-type function (e.g., when the mutated amino acid is replaced with an amino acid that does not cause disease or provides an improved phenotype). As disclosed in more detail herein, the target transcript molecule may also be encoded by a wild-type gene, for example, in a preferred aspect of the invention, wherein the target nucleic acid molecule is a transcript of the wild-type human SLC10A1 gene as shown in this invention, wherein RNA editing causes the encoded NTCP protein to lose function but improves the disease state of the treated subject.

[0046] Non-restricted examples of using RNA editing to target transcript molecules for multiple therapies include SERPINA1 (for the treatment of α1-antitrypsin (A1AT) deficiency; see, for example, WO2016 / 097212, WO2017 / 220751, WO2018 / 041973 and WO2021 / 243023) and IDUA (for the treatment of Hurler syndrome; see, for example, WO2017 / 220751, WO2021 / 097212, WO2017 / 220751, WO2018 / 041973 and WO2021 / 243023), WO2018 / 041973 and WO2021 / 243023), and IDUA (for the treatment of Hurler syndrome; see, for example, WO2017 / 220751, WO2018 / 041973, WO2021 / 097212, WO2017 / 220751, WO2018 / 041973 and WO2021 / 243023). 2018 / 041973 and WO2021 / 209010), LRRK2 (for the treatment of Parkinson's disease; WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, WO2021 / 231673 and WO2021 / 242903), ABCA4 (for the treatment of Stargardt's disease; see, for example, WO2021 / 130313 and WO2021 / 2). 31830), USH2A (for the treatment of Usher syndrome; e.g. WO2020 / 157008, WO2020 / 219981 and WO2021 / 136404), APP (e.g. WO2021 / 113270), CMT1A (e.g. WO2021 / 113390), ASS1 (e.g. WO2021 / 231675), GJB2 (e.g. WO2021 / 231679), M ECP2 (used to treat Rett syndrome; e.g., WO2019 / 071274 and WO2021 / 231680), OTOF (used to treat autosomal recessive nonsyndromic hearing loss; e.g., WO2021 / 231685 and WO2021 / 231692), XLRS (e.g., WO2021 / 231691), and PCSK9 (used to treat hypercholesterolemia; e.g., WO2023 / 152371).

[0047] This invention relates to an AON that mediates RNA editing of one or more adenosines present in the SLC10A1 gene transcript using an endogenous (naturally occurring) ADAR enzyme in a host cell (preferably hepatocytes). The two-dimensional structure of the NTCP protein in the cell membrane of hepatocytes is known in the art (Ho RH et al., 2004; J Biol Chem. 279(8):7213-7222). The AON disclosed herein aims to reduce the reabsorption of bile acids in the liver by inhibiting NTCP function. Several loss-of-function variants of NTCP have been identified in the art. Therefore, these mutations occur naturally in a small number of individuals without causing cholestasis-related symptoms (Vaz et al., 2015; Hepatology. 61(1):260-267; Schneider et al., 2022; Clin Res Hepatol Gastroenterol. 46(3):101824), although the concentration of circulating bile acids remains relatively high. Various amino acid substitutions in the sodium-binding pocket have been described in the art that result in loss-of-function mutations in the NTCP protein, leading to the inability to transport bile acids but without affecting protein expression or localization (Huan Yan et al., 2014; J Virology. 88(6): 3273-3284). This finding suggests that treatment of subjects with cholestasis-related disorders using the AONs disclosed herein will reduce the accumulation of these toxic bile acids in the liver. Furthermore, the generation of loss-of-function variants of NTCP is expected to promote the elimination of bile acids from the body by increasing the excretion of bile acids in feces and urine. This process, known as bile acid sulfation, enhances the solubility of bile acids and reduces their absorption in the intestine. The present invention relates to a variety of AONs intended to deaminate multiple adenosines in the SLC10A1 transcript, each believed to independently cause loss of function of the NTCP protein, and each AON is believed to be independently usable for the treatment of the bile acid accumulation disorders disclosed herein. However, it is not excluded that two or more adenosines may be targeted for deamination in a single treatment. To avoid being bound by theory, by combining the AON targeting multiple adenosines disclosed in this paper to target multiple amino acids within a single NTCP protein, synergistic or additive effects can be obtained to improve therapeutic efficacy.

[0048] Cholestasis can cause inflammation and lead to the development of liver fibrosis, cirrhosis, hepatocellular carcinoma, and / or liver failure. The AON disclosed in this article aims to reduce the toxic accumulation of bile acids to alleviate inflammation and reduce or prevent fibrosis and cirrhosis, thereby protecting hepatocytes and liver function.

[0049] Cholestasis can be caused by chronic or acute factors. Chronic liver disease leads to liver fibrosis and cirrhosis (scarring), reducing liver function. Chronic diseases can include infectious diseases such as hepatitis viruses (hepatitis B virus, hepatitis C virus, hepatitis D virus); alcoholic hepatitis; autoimmune and autoinflammatory diseases, including primary sclerosing cholangitis, primary biliary cholangitis, and autoimmune hepatitis; metabolic diseases, including non-alcoholic fatty liver disease, metabolic-associated fatty liver disease, and non-alcoholic steatohepatitis; and genetic diseases, including Wilson's disease, hereditary hemochromatosis, and alpha-1 antitrypsinopathy. Acute hepatitis and cholestasis can be caused by infections (including hepatitis viruses, mononucleosis, HIV, cytomegalovirus, sepsis, gallbladder infection), alcohol poisoning or toxic hepatitis, liver cancer or lymphoma, medication use (including contraceptives, anabolic steroids, penicillin antibiotics (including amoxicillin), azathioprine, imipramine, estradiol, cimetidine, chlorpromazine, prochlorperazine, tolbutamide, terbinafine), and cholestasis of pregnancy.

[0050] Cholestasis can also be caused by extrahepatic causes, such as bile duct obstruction or stenosis, including gallstones in the common bile duct, cystic duct, or Hartmann's pouch; pancreatic cysts and pseudocysts; extrahepatic bile duct tumors; chronic pancreatitis; pancreatic cancer; bile duct cancer; cholangitis; biliary atresia; and previous injury or surgery. In addition, cholestasis can also occur in newborns due to causes including infectious agents such as viruses, bacteria, spirochetes, and parasites; drug toxins, endotoxins, total parenteral nutrition-related cholestasis, or herbal products; metabolic causes, including hypothyroidism or hypopituitarism; immune-related alloimmune liver disease during pregnancy; anatomical obstruction, including biliary atresia, choledochal cysts, cholelithiasis, biliary sludge, concentrated bile, spontaneous perforation of the common bile duct, or tumors; idiopathic neonatal hepatitis (transient neonatal cholestasis), cardiovascular and circulatory system diseases, hemophagocytic lymphohistiocytosis, cholestasis caused by malignant tumors or congenital lupus; or genetic and metabolic causes. This includes α1-antitrypsin (A1AT) deficiency, Alagille syndrome, arthritic-renal-cholestasis syndrome, Caroli's disease, congenital liver fibrosis, Down syndrome, Turner syndrome, citrate deficiency, cystic fibrosis, bile acid synthesis disorders, bile acid conjugation disorders, fatty acid oxidation defects, galactosemia, type IV glycogen storage disease, hereditary fructose intolerance, mitochondrial respiratory chain disorders, neonatal ichthyosis-sclerosing cholangitis syndrome, neonatal sclerosing cholangitis, Niemann-Pick disease type C, peroxidase disorders, progressive familial intrahepatic cholestasis, lipid storage diseases, tyrosinemia, or urea cycle defects. The AON disclosed herein is intended to prevent the accumulation of bile acids in the liver and the resulting cholestasis for all the aforementioned diseases, as described in all pharmaceutical formulations, uses, and treatments herein. The AON disclosed herein can also be used to treat hepatitis B virus and hepatitis D virus infections by altering the NTCP structure and blocking uptake by hepatocytes.

[0051] definition

[0052] Whenever oligonucleotide, oligo, ON, ASO, oligonucleotide composition, antisense oligonucleotide, AON, (RNA) editing oligonucleotide, EON, and RNA (antisense) oligonucleotide are mentioned, unless the context otherwise requires, they refer to oligonucleotides and deoxyoligonucleotides. Oligonucleotides may completely lack (naturally occurring) RNA and DNA nucleotides and may consist entirely of modified nucleotides. Whenever "oligonucleotide" is mentioned, it may contain the bases A, G, C, U, or I. Whenever "deoxyoligonucleotide" is mentioned, it may contain the bases A, G, C, T, or I. However, the AON disclosed herein may contain a mixture of ribonucleotides and deoxyribonucleotides. When deoxyribonucleotides are used, that is, when the sugar is not modified at the 2′ position, the nucleotide is usually abbreviated as dA (or Ad), dC (or Cd), dG (or Gd), or T, where “d” indicates the deoxy nature of the nucleoside. Normal RNA or ribonucleotides modified at the 2′ position are usually abbreviated as those without “d”, and are usually abbreviated with their respective modifications, as explained in this article.

[0053] The term "nucleoside" refers to a nucleobase linked to a (deoxy)ribosyl sugar and without a phosphate group. A "nucleotide" consists of a nucleoside and one or more phosphate groups. Therefore, the term "nucleotide" refers to a respective nucleobase-(deoxy)ribosyl-phosphate linker, and any chemical modification of the ribose moiety or phosphate group. Thus, the term will include nucleotides containing a locked ribose moiety (containing a 2′–4′ bridge, containing a methylene or any other group), an unlocked nucleic acid (UNA), a threonine nucleic acid (TNA), or a linker containing a phosphodiester, phosphonoacetate, phosphotriester, PS, (di)thiophosphate, MP (or MeP), methylthiophosphonate, phosphoramide linker, PNdmi, and linkers according to formula (I) as described herein. Sometimes, the terms nucleobase, nucleoside, and nucleotide are used interchangeably unless the context explicitly requires otherwise, such as when a nucleoside is linked to an adjacent nucleoside and the linker between these nucleosides is modified. As described herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms “ribonucleoside” and “deoxyribonucleoside”, or “ribose” and “deoxyribose”, are used as in the art.

[0054] Sometimes, the terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, inosine and hypoxanthine are used interchangeably, referring to the corresponding nucleobase on one hand, and the nucleoside or nucleotide on the other. The nucleobase thymine (T) is also called 5-methyluracil (m). 5 5-methyluracil (U) is a derivative of uracil (U); in this article, thymine and 5-methyluracil are used interchangeably. Similarly, thymidine, also known as 5-methyluridine, is a derivative of uridine; in this article, thymidine and 5-methyluridine are used interchangeably.

[0055] When referring to nucleotides in oligonucleotides, this includes cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5-hydroxycytosine, and β-D-glucosyl-5-hydroxymethylcytosine. When referring to adenine, this includes N6-methyladenine, 8-oxoadenine, 2,6-diaminopurine, and 7-methyladenine. When referring to uracil, this includes dihydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1-methylpseudouracil, 4-thiouracil, and 5-hydroxymethyluracil. When referring to guanine, this includes 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7-trimethylguanosine, and N2,7-dimethylguanosine. Whenever nucleosides or nucleotides are mentioned, this includes furanose derivatives such as 2′-deoxy, 2′-hydroxy, and 2′-O-substituted variants such as 2′-O-methyl (2′-OMe), as well as other modifications, including 2′-4′ bridging variants. Whenever oligonucleotides are mentioned, one or more linking bonds can be naturally occurring phosphodiester links, while the remaining links between the two mononucleotides can be modified links. Examples of such modified links include phosphonoacetates, phosphotriesters, PS, (di)thiophosphates, MP, phosphoramide links, phosphoroguanidine, thiophosphoroguanidine, sulfonylphosphoramide, PNdmi, and linking structures according to formula (I), which are described in detail below.

[0056] The term "comprising" encompasses both "including" and "consisting of," for example, a composition "comprising X" may consist of only X, or may include other substances, such as X+Y. The term "about" is optional when related to the numerical value x, meaning, for example, x ± 10%.

[0057] The term "substantially" does not exclude "completely," for example, a composition that is "substantially free of Y" can be completely free of Y. In relevant cases, the term "substantially" may be omitted from the definition of this invention.

[0058] The terms “contribute to” or “mediate” are used interchangeably with “can promote”. When used in the context of AONs that contribute to (or mediate) ADAR editing, this means that after entering the cell, the AON interacts with the target RNA sequence to form a double-stranded structure recognized by an ADAR enzyme, which then deaminates the target adenosine to inosine. Therefore, AONs themselves do not possess enzymatic function (ADAR enzymes do), but upon binding to the target RNA molecule, they can trigger, induce, cause, organize, mediate, provide, give, generate, promote, or lead to RNA editing.

[0059] The term “mismatch” is used herein to refer to a relative nucleotide in a double-stranded RNA complex that does not form a perfect base pair according to the Watson-Crick base pairing rules. Historically, mismatched nucleotides were GA, CA, UC, AA, GG, CC, and UU pairs. In some embodiments, the AONs disclosed herein have fewer than four mismatches with the target sequence, such as 0, 1, or 2 mismatches. “Wag” base pairs are GU, IU, IA, and IC base pairs. When U is opposite the target A, there is no mismatch, and the AON can be 100% complementary. When C is opposite the target A, there is at least one mismatch between the AON and the target sequence. Although G:G pairing is considered a mismatch, this does not necessarily mean that the interaction is unstable. This means that, based on the current disclosure, the term “mismatch” may be somewhat outdated, where, based on the source of the nucleotide, Hoogsteen base pairings may be considered mismatches but still relatively stable. For example, isolated G:G pairings in double-stranded RNA may be quite stable but are still defined as mismatches. Analysis of the natural targets of ADAR enzymes has shown that these typically include mismatches between the two strands of the RNA helix that forms ADAR1 or ADAR2 edits. These mismatches have been proposed to enhance the specificity of the editing response (Stefl et al., 2006. Structure 14(2):345-355; Tian et al., 2011. Nucleic Acids Res 39(13):5669-5681). Characterizing the optimal pattern of paired / mismatched nucleotides between AON and target RNA is also important for developing highly effective ADAR-based AON therapies.

[0060] The term “complementary” as used herein refers to the hybridization of an AON with a second nucleic acid strand under physiological conditions. Examples include (i) when an AON, as the first nucleic acid strand (=guide oligonucleotide), forms a heteroduplex RNA editing oligonucleotide complex with a second complementary nucleic acid strand (in vitro), or (ii) when it forms a double-stranded complex with a target RNA molecule. This term does not necessarily mean that every nucleotide in the nucleic acid strand is perfectly paired with its corresponding nucleotide in its relative sequence. In other words, although an AON may be complementary to the target sequence, there may be mismatches, wobbles, and / or bulges between the AON and the target sequence, and under physiological conditions, the AON will still hybridize with the target sequence, allowing cellular RNA editing enzymes to deamination the target adenosine to inosine. Therefore, the term “substantially complementary” also means that despite the presence of mismatches, wobbles, and / or bulges, the AON has enough matching nucleotides with the target sequence to hybridize with the target RNA molecule under physiological conditions. As shown herein, an AON may be complementary if it can hybridize with its target under physiological conditions, but it may also contain one or more mismatches, wobbles, and / or bulges with the target sequence.

[0061] The term "orphan nucleotide" refers to the nucleotide in AON that is directly opposite to target adenosine, which is adenosine deaminated by a deaminase. Orphan nucleotides can be native cytidine or deoxycytidine, or uridine or deoxyuridine. They can also be chemically modified nucleotides, as described further in detail below, or known or chemically modified analogues of native (deoxy)cytidine, such as nucleotides carrying the Benner base, or known or chemically modified analogues of native (deoxy)uridine, such as isouridine, as outlined in further detail below.

[0062] "Nucleotide analogues" refer to analogues of nucleic acid nucleotides. Nucleotide analogues are analogues of adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine, or deoxyuridine.

[0063] The term "downstream" in relation to a nucleic acid sequence refers to the sequence extending further along the 3′ direction; the term "upstream" means the opposite direction. Therefore, in any sequence encoding a polypeptide, the start codon in the sense strand is upstream of the stop codon, but the start codon in the antisense strand is downstream of the stop codon. This is also true for the AONs disclosed herein. In antisense oligonucleotides, the nucleotide upstream of the orphan nucleotide is located at the 5′ end, and the nucleotide downstream of the orphan nucleotide is located at the 3′ end.

[0064] The nucleotide numbering system disclosed in this paper for AONs is such that orphan nucleotides are numbered 0, and the nucleotide at the 5' end of an orphan nucleotide is numbered +1. The counting increases positively (+) towards the 5' end and negatively (-) towards the 3' end, with the first nucleotide at the 3' end of an orphan nucleotide being numbered -1. The numbering method for internucleotide linkages in AONs is as follows: linkage number 0 is the linkage at the 5' end of an orphan nucleotide, and the linkage position in oligonucleotides increases positively (+) towards the 5' end and negatively (-) towards the 3' end.

[0065] The term "hybridization" usually refers to specific hybridization, excluding non-specific hybridization. Specific hybridization can occur under selected experimental conditions, using techniques well known in the art, ensuring that the most stable interaction between the probe and target occurs when the probe and target have at least 70%, preferably at least 80%, more preferably at least 90% sequence identity.

[0066] The term "splicing mutation" refers to a mutation in a gene encoding pre-mRNA in which the splicing mechanism is dysfunctional in a sense, meaning that the intron is interfered with from the exon, and due to the aberrant splicing, the subsequent translation frame shift leads to premature termination of the encoded protein. Typically, this shortened protein degrades rapidly and loses all functional activity.

[0067] When referring to the “naked” form of AON disclosed herein, it means that the AON is manufactured in a laboratory or manufacturing facility, where it is typically chemically modified to prevent rapid degradation after administration into mammalian bodies, tissues, or cells. Therefore, the naked form of AON differs from the form encoded (and delivered) by a viral genome or contained within a plasmid vector. When such viral or plasmid vectors are administered, the encoded AON is expressed from the viral vector genome or plasmid into the cells to which the viral or plasmid vector is delivered. Thus, the AON is then unmodified and contains only naturally occurring RNA nucleotides.

[0068] The length of the AON disclosed herein, when delivered in naked form, is preferably 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. However, when the AON disclosed herein is delivered via expression in a viral vector, the AON can be longer, for example, 70, 80, 90, 100, 150, or 200 or more nucleotides.

[0069] The term "HEON" refers to a heteroduplex RNA editing oligonucleotide complex molecule in which the AON disclosed herein hybridizes with a partially or completely complementary, or partially or completely overlapping, positive oligonucleotide. Because the AON disclosed herein typically has specific chemical modifications that differ from those in the positive strand, the two strands form such heteroduplex RNA editing oligonucleotide complexes. The positive strand can be chemically modified almost entirely, similar to or different from those performed in the AON disclosed herein, for example by providing a ribose moiety carrying a nucleotide with a 2′-OMe substitution, a 2′-F substitution, or a 2′-MOE substitution. It should be understood that the positive strand present in the HEON is a distinct entity from the target RNA molecule in the cell. The positive strand of HEON is preferably 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 in length. HEON is typically produced in vitro and used as a delivery vehicle to protect AON from degradation before application to cells. In other words, HEON is preferably formed before AON is applied to cells.

[0070] This article discloses an AON that recruits endogenous ADAR enzymes in human cells after forming a double-stranded complex with a region of a target RNA nucleic acid molecule in the cell. This region contains a target adenosine, and the nucleotides in the AON corresponding to the target adenosine are orphan nucleotides. The ADAR enzyme deaminates the target adenosine to inosine after binding to the double-stranded complex, and the target RNA nucleic acid molecule is a transcript molecule encoding the human SLC10A1 gene encoding NTCP. Preferably, the transcript molecule is pre-mRNA or mRNA. Preferably, the cell is a liver cell, more preferably a hepatocyte. In one aspect, the nucleotides in the AON are numbered such that the orphan nucleotide is numbered 0, and the nucleotides are further positively (+) incremented towards the 5′ end and negatively (-) incremented towards the 3′ end, and the orphan nucleotides are deoxyribonucleotides containing cytosine, cytosine analogs, uracil, or isouriacil. In one aspect, an orphan nucleotide is a deoxynucleotide containing a cytosine analogue, wherein the cytosine analogue is a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase, also known as a Benner base. In some aspects of the AON disclosed herein, when the nucleotide opposite the first nucleotide at the 3′ end of the target RNA nucleic acid molecule is cytidine, the first nucleotide at the 3′ end of the orphan nucleotide is deoxyinosine. This is the case when the target adenosine is a) the first adenosine in the codon encoding glutamine at position 68, b) the first adenosine in the codon encoding glutamine at position 261, or c) the first adenosine in the codon encoding lysine at position 314. In one aspect, the length of the AON is 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. In another aspect, the AON contains one or more modifications to the linker portion, each modification being independently selected from: PS, phosphonoacetate, dithiophosphate, MP, sulfonylphosphamide, PNdmi, and PNms. In one aspect, the nucleotide linkers in the AON are numbered such that linker number 0 is the linker at the 5′ end of an orphan nucleotide, and the linker positions in the oligonucleotide increase positively (+) towards the 5′ end and negatively (-) towards the 3′ end, wherein linker position -2 is an MP linker or a PNms linker. In another aspect, the linker between the last two nucleotides at the 5′ and / or 3′ ends of the AON is a PNdmi linker or a PNms linker. In another aspect, the AON contains one or more nucleotides having a single or disubstituted nucleotide at the 2′, 3′, and / or 5′ position of the ribose, each substitution being independently selected from: -OH; -F; substituted or unsubstituted linear or branched lower (C1-C1) nucleotides.10 Alkyl, alkenyl, alkynyl, alkylaryl, allyl, 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-allyl; -O-alkyl; -O-alkyl; -methoxy; -aminopropoxy; -methoxyethoxy; -dimethylaminooxyethoxy; and -dimethylaminoethoxyethoxy. In one aspect, AON is covalently or non-covalently bound to the GalNAc moiety directly or via a linker. When AON needs to be delivered to hepatocytes (especially hepatocytes), those skilled in the art can select a suitable linker and the need for covalent or non-covalent binding of the GalNAc moiety. In one aspect, AON is covalently or non-covalently bound to a triterpenoid glycoside (preferably AG1856) directly or via a linker. As described in PCT / EP2024 / 051278 (unpublished), RNA editing can be significantly enhanced when AON is linked to a saponin in a 1:1 ratio, especially when the saponin is AG1856. Therefore, to increase the endosomal release (intracellular) of AON and make it available for RNA target hybridization, it is preferable to link AON to AG1856 (covalently or non-covalently) before administration to cells or the subject to be treated.

[0071] This paper discloses an AON that recruits endogenous ADAR enzymes in human cells after forming a double-stranded complex with a region of a target RNA nucleic acid molecule in the cell. This region contains a target adenosine, and the nucleotide corresponding to the target adenosine in the AON is an orphan nucleotide. The ADAR enzyme, upon binding to the double-stranded complex, deaminates the target adenosine to inosine. The SLC10A1 gene is wild-type, and the target adenosine is selected from: (i) adenosine in the CAG codon encoding glutamine (Q) at position 68 of the NTCP protein, wherein deamination of the adenosine converts the amino acid to arginine (R); (ii) adenosine encoding N... The first adenosine in the CAA codon encoding glutamine (Q) at position 261 of the NTCP protein, wherein deamination of the adenosine changes the amino acid to arginine (R); (iii) the adenosine in the GAG ​​codon encoding glutamate (E) at position 257 of the NTCP protein, wherein deamination of the adenosine changes the amino acid to glycine (G); and (iv) the first adenosine in the AAG codon encoding lysine (K) at position 314 of the NTCP protein, wherein deamination of the adenosine changes the amino acid to glutamate (E); wherein deamination of the target adenosine results in impaired function of the NTCP protein in transporting bile acids from the portal circulation into the cell. In a preferred aspect, the target adenosine is located in the CAG codon encoding glutamine at position 68 of the NTCP protein, wherein AON comprises or is composed of a compound selected from SEQ ID NO. NO:150, 151, 152, 154, 156, 158, 159, 163, 164, 165, 166, 1127, 1128, 1129, 1130, 1131, 1133, 1137, 1138, 1139, 1140, 1141, 1142, 1143, 1144, 1145, 1146, 1150, 1151, 1152, 1153, 1154, 1155, 1156, 1157, 1158, 1159, 1160, 116 The sequence and modifications of 1167, 1168, 1169, 1170, 1171, 1172, 1173, 1174, 1175, 1176, 1177, 1178, 1180, 1181, 1182, 1183, 1260, 1283, 1284, 1285, 1286, 1287, 1288, 1295, 1296, 1297, 1298, 1299, 1300, 1301, 1302, 1303, 1304, 1305, and 1306 are composed of.In another preferred aspect, the target adenosine is located in the GAG ​​codon encoding glutamate at position 257 of the NTCP protein, and wherein AON comprises or is composed of a protein selected from SEQ ID NO: 1193, 1194, 1197, 1198, 1199, 1200, 1201, 1202, 1203, 1204, 1205, 1206, 1207, 1208, 1209, 1210, 1211, 1212, 1213, 1215, 1216, 1217, 1218, 1219, 1220, 1221, 1222, 1 The sequences and modifications of 223, 1224, 1225, 1226, 1227, 1228, 1229, 1230, 1231, 1232, 1233, 1234, 1235, 1236, 1237, 1238, 1239, 1240, 1241, 1242, 1243, 1244, 1245, 1246, 1247, 1253, and 1254 constitute the present invention. The present invention also relates to vectors comprising nucleic acid molecules encoding the AON disclosed herein, preferably viral vectors, more preferably adeno-associated virus (AAV) vectors. When a coding vector is used and the AON is encoded by a viral vector (genome), the orphan nucleotide is cytidine, and when the relative nucleotide in the target sequence is cytidine, the nucleotide at position -1 is guanosine. Furthermore, since the AON is transcribed from a viral vector (genome), there are no chemical modifications in the backbone, linkages, and glycoribose. This invention also relates to nanoparticle delivery carrier formulations comprising the AON disclosed herein. In a preferred aspect, the nanoparticle delivery carrier is a lipid nanoparticle (LNP). LNPs that can be used in the context of the AON of this invention are those used in the art for delivering small and large RNA molecules, such as those used for delivering mRNA-based vaccines against Covid-19 coronavirus. LNPs already used for delivering other types of RNA (such as siRNA) can also be used for delivering the AON disclosed herein. If an LNP or any other similar type of carrier is applied, the AON is still considered naked because it is not transcribed from a polynucleotide encoding a nucleotide (as in the case of plasmids or vectors, where the AON is not considered "naked"). Therefore, although the chemically modified AON is encapsulated in a carrier (preferably an LNP), it is still considered naked because it is manufactured in a laboratory setting and subsequently encapsulated in a carrier using methods known to those skilled in the art.The present invention also relates to delivery vectors, preferably LNPs, comprising “naked” and chemically modified AONs disclosed herein, more preferably such as SEQ ID NO: 150, 151, 152, 154, 156, 158, 159, 163, 164, 165, 166, 1127, 1128, 1129, 1130, 1131, 1133, 1137, 1138, 1139, 1140, 1141, 1142, 1143, 1144, 1145, 1146, 1150, 1151, 1152, 1153, 1154, 1155, 11 56, 1157, 1158, 1159, 1160, 1161, 1167, 1168, 1169, 1170, 1171, 1172, 1173, 1174, 1175, 1176, 1177, 1178, 1180, 1181, 1182, 1183, 1260, 1283, 1284, 1285, 1286, 1287, 1288, 1295, 1296, 1297, 12 98, 1299, 1300, 1301, 1302, 1303, 1304, 1305, 1306, 1193, 1194, 1197, 1198, 1199, 1200, 1201, 1202, 1203, 1204, 1205, 1206, 1207, 1208, 1209, 1210, 1211, 1212, 1213, 1215, 1216, 1217, 1218, 12 The invention relates to any one of the following: 19, 1220, 1221, 1222, 1223, 1224, 1225, 1226, 1227, 1228, 1229, 1230, 1231, 1232, 1233, 1234, 1235, 1236, 1237, 1238, 1239, 1240, 1241, 1242, 1243, 1244, 1245, 1246, 1247, 1253, and 1254. The invention also relates to pharmaceutical compositions comprising the AON disclosed herein, a carrier or nanoparticle delivery carrier formulation, and a pharmaceutically acceptable carrier. In one aspect, the AON disclosed herein is in naked form. In another aspect, the AON disclosed herein is in circular form. In another aspect, the AON disclosed herein is not in naked form but is expressed from the genome of a viral vector. In one aspect, the AON disclosed herein is not in naked form, but is expressed from an expression vector such as a plasmid. In another aspect, when the AON disclosed herein is not in naked form, the length of the AON is 15 to 60 nucleotides as described above, or in another embodiment, the length is 61 to 300 nucleotides. It should be noted that when the AON is delivered via a vector (e.g., an AAV vector), there is no chemical modification in the AON acting on the target RNA molecule.While “naked” AONs with the chemical modifications outlined herein are preferred, AONs delivered by other means, such as expression via AAV vectors, or circular editing molecules, or editing molecules with hairpin structures (recruitment portions, e.g., disclosed in WO2016 / 097212, WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560, and WO2022 / 078995), are also included in this invention because these can also be used to edit adenosine in target SLC10A1 RNA molecules to produce a defunctionalized NTCP protein. Those skilled in the art will understand that AONs are also considered naked when the delivery portion or attachment of the AON (e.g., for targeting hepatocytes in the liver) is used, as is the case when the GalNAc-AON is encapsulated in a delivery vector such as an LNP. This invention also relates to the use of the AON disclosed herein in the treatment of diseases caused by cholestasis in the liver, such as cholestasis, primary sclerosing cholangitis (PSC), biliary atresia (BA), and cirrhosis. This invention also relates to the use of the AON disclosed herein in the manufacture of medicaments for the treatment of diseases caused by cholestasis in the liver, such as cholestasis, PSC, BA, and cirrhosis. This invention further relates to a method for editing human SLC10A1 pre-mRNA or mRNA molecules in liver cells (preferably hepatocytes), the method comprising contacting the SLC10A1 pre-mRNA or mRNA molecule with an AON capable of triggering ADAR-mediated adenosine to inosine deamination, thereby editing the SLC10A1 pre-mRNA or mRNA molecule to encode an NTCP protein with impaired, reduced, or lost bile acid uptake function, wherein the AON is as disclosed herein. This invention also relates to a method for treating, improving, or slowing the progression of diseases (such as cholestasis, PSC, BA, and cirrhosis) caused by cholestasis in the liver in human subjects in need. The method comprises administering to the subject an AON, carrier, or nanoparticle delivery carrier formulation disclosed herein, such that the AON hybridizes with a complementary portion of a region of the SLC10A1 pre-mRNA or mRNA molecule containing the target adenosine in the subject's cells. This achieves ADAR-mediated adenosine-to-inosine deamination, thereby altering the SLC10A1 pre-mRNA or mRNA molecule to encode an NTCP protein with reduced, decreased, or lost bile acid uptake function, thereby treating the subject. RNA editing can, in principle, be performed at multiple locations within the SLC10A1 transcript. Examples of target adenosines disclosed herein include, preferably, the adenosine in the CAG codon encoding glutamine (Q) at position 68 of the NTCP protein, the first adenosine in the CAA codon encoding glutamine (Q) at position 261, the adenosine in the GAG ​​codon encoding glutamate (E) at position 257, and the first adenosine in the AAG codon encoding lysine (K) at position 314.Highly preferred are the AONs disclosed herein, and methods for using these AONs to achieve Q68R alterations and / or E257G alterations in human NTCP proteins. A method for deamination of target adenosine in human SLC10A1 pre-mRNA or mRNA molecules in hepatocytes (preferably hepatocytes) in vitro, ex vivo, or in vivo is also disclosed, the method comprising the steps of: (i) providing the cells with the AONs disclosed herein; (ii) allowing the cells to take up the AONs; (iii) annealing the AONs with the SLC10A1 pre-mRNA or mRNA molecules; (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the SLC10A1 pre-mRNA or mRNA molecules to inosine; and optionally (v) identifying the presence of inosine in the SLC10A1 pre-mRNA or mRNA molecules using functional readouts. In one aspect, the method disclosed herein includes the step of administering a triterpenoid glycoside before, after, or simultaneously with the administration of the AON, wherein, in a preferred aspect, the triterpenoid glycoside is AG1856. However, in a preferred aspect, the triterpenoid glycoside (or commonly referred to as "saponin") is physically bound to the AON. In one aspect, the AON disclosed herein comprises a linker portion having a structure according to formula (I).

[0072]

[0073] Where: X = O or S; and R = aryl, substituted aryl, heterocycle, substituted heterocycle, aromatic heterocycle, substituted aromatic heterocycle, C1-C6 alkoxy, substituted C1-C6 alkoxy, C1-C 20 Alkyl, substituted C1-C 20 Alkyl, C1-C6 alkenyl, C1-C6 substituted alkenyl, C1-C6 alkynyl, substituted C1-C6 alkynyl, or conjugated groups. In a preferred embodiment, X = O and R = methyl, and the linker modification is referred to as methanesulfonylphosphamide, MsPA, or "PNms". In one embodiment, PNms linkers are used instead of MP and / or PNdmi linkers.

[0074] Chemical modification

[0075] Various chemicals and modifications readily available for use according to the present invention are known in the art. All chemical modifications listed herein that can be used with the AON disclosed herein can also be used with the positive chain complementary to the AON, when the AON and the positive chain form a HEON complex, as described in GB 2215614.5 (unpublished) and as disclosed above, except that the opposing positive chain does not contain an orphan nucleotide. Therefore, modifications related to orphan nucleotides relate only to the AON disclosed herein, but all other modifications relate to the AON disclosed herein and any (protective) positive oligonucleotide that can be used with the AON in pharmaceutical products. This includes the use of hydrophobic moieties (such as tocopherol and cholesterol) and cell-specific ligands (such as the GalNAc moieties), which have also been described herein and are detailed in PCT / EP2023 / 079290 (unpublished), which can bind to the AON or its opposing chain, or both. Preferred GalNAc moieties that can be used in the context of the AON disclosed herein are disclosed in WO2022 / 271806.

[0076] Those skilled in the art will recognize that oligonucleotides (such as AONs as outlined herein) are typically composed of repeating monomers. These monomers are most commonly 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). These consist of a pentose sugar, a ribose sugar, a 5′-linked phosphate group (linked by a phosphate ester), and a 1′-linked base. The sugar links the base and the phosphate, and is therefore often referred to as the “scaffold” of the nucleotide. Therefore, modifications to the pentose sugar are often referred to as “scaffold modifications.” The original pentose sugar can be completely replaced by another portion that similarly links the base and the phosphate. Therefore, it should be understood that while pentose sugars are often scaffolds, a scaffold is not necessarily a pentose sugar. Examples of scaffold modifications applicable to the AON monomers disclosed herein are disclosed in WO2020 / 154342, WO2020 / 154343, and WO2020 / 154344.

[0077] The nucleosides in the AONs disclosed herein can be natural nucleosides (deoxyribonucleosides or ribonucleosides) or non-natural nucleosides. It should be noted that for RNA editing (where double-stranded RNA is typically a substrate of enzymes with deamination activity, such as ADAR), ribonucleosides are considered "natural," while deoxyribonucleosides may be considered non-natural or modified for evidentiary purposes simply because DNA is not present in the RNA-RNA double-stranded (natural) substrate conformation. Those skilled in the art will understand that when a nucleotide has a natural ribose moiety, it can still be non-naturally modified in its bases and / or linkages.

[0078] It is recognized in the art that common limiting factors for oligonucleotide-based therapies are the ability of oligonucleotides to be taken up by cells (when delivered spontaneously or “naked” without the application of delivery vectors such as viral vectors or plasmids), biodistribution, and resistance to nuclease-mediated degradation. Those skilled in the art will know, and have described in detail in the art, that various chemical modifications can help overcome these limitations. Examples of such commonly used chemical modifications are 2′-OMe, 2′-F, and 2′-MOE modifications of sugars and the use of PS linkages between nucleosides, as described herein.

[0079] Scaffold modification (ribose)

[0080] The 2′ ribose group in all nucleotides of the AON disclosed herein, except for the ribose portion of orphan nucleotides (which has certain limitations in terms of RNA editing compatibility), can be independently selected from 2′-H (i.e., DNA), 2′-OH (i.e., RNA), 2′-OMe, 2′-MOE, 2′-F, or 2′-4′-linked (e.g., locked nucleic acid (LNA)), or other ribosyl 1′-substituted, 2′-substituted, 3′-substituted, 4′-substituted, or 5′-substituted. When the nucleobase is naturally occurring cytosine, orphan nucleotides in AONs that do not contain other chemical modifications to the ribose, base, or linker preferably do not carry 2′-OMe or 2′-MOE substitutions, but may carry 2′-F, 2′,2′-difluoro (diF), or 2′-ara-F (FANA) substitutions or may be DNA. WO2024 / 013360 discloses modification of the 2′ position of the ribose moiety of orphan nucleotides by substitution with 2′,2′-disubstituted nucleotides (such as diF), and this also applies to the disclosure herein. The 2′-4′ linker may be selected from many linkers known in the art, such as methylene linkers, amide linkers, or constrained ethyl linkers (cEt).

[0081] The AON disclosed herein may contain one or more nucleotides carrying a 2′-MOE ribose modification. Furthermore, the AON disclosed herein may contain one or more nucleotides without a 2′-MOE ribose modification, or wherein the 2′-MOE ribose modification is located at a position that does not impede the deamination of target adenosine by an enzyme with adenosine deaminase activity. The AON disclosed herein may contain a 2′-OMe ribose modification at a position not containing a 2′-MOE ribose modification. The AON disclosed herein may contain a deoxynucleotide at a position not containing a 2′-MOE or 2′-OMe ribose modification or other 2′ ribose substitution. The AON disclosed herein may comprise one or more nucleotides containing a 2′ substitution, such substitution being 2′-MOE, 2′-OMe, 2′-OH, 2′-deoxy, TNA, 2′-fluoro(2′-F), 2′,2′-difluoro(diF) modification, 2′-fluoro-2′-C-methyl modification, or a 2′-4′-link (i.e., a bridging nucleic acid, such as a locked nucleic acid (LNA or, for example, an example mentioned in WO2018 / 007475)). Other nucleic acid monomers that may be used in the AON disclosed herein are arabinonucleotides and 2′-deoxy-2′-fluoroarabinonucleotides (FANA), for example, to improve affinity. The 2′-4′ link may be selected from adapters known in the art, such as methylene adapters or constrained ethyl adapters. Various 2' modifications that can exist in the AONs disclosed herein are known in the art, including but not limited to the modifications detailed in WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, WO2018 / 134301, WO2019 / 219581, WO2019 / 158475, and WO2022 / 099159. In all cases, the modifications should be compatible with RNA editing, enabling the AON to fulfill its role as an oligonucleotide, capable of forming a double-stranded complex with the target RNA, and recruiting deaminases by generating such a double-stranded nucleic acid complex, which can then deaminate the target adenosine. When the monomer in the AON disclosed herein contains an unlocking nucleic acid (UNA) ribose modification, the monomer may have a 2' position containing the same modification discussed above, such as 2'-MOE, 2'-OMe, 2'-OH, 2'-deoxy, 2'-F, 2',2'-diF, 2′-fluoro-2'-C-methyl, arabinonucleotide, FANA, or a 2'-4'-link (i.e., a bridging nucleic acid, such as a locked nucleic acid (LNA)). In one aspect, the AON disclosed herein contains at least one nucleotide containing a threononucleotide (TNA) ribose modification. In another aspect, the AON disclosed herein contains at least one nucleotide having a sugar moiety containing a 2'-fluoro (2'-F) modification. The preferred position of the nucleotide carrying the 2'-F modification is the -3 position in the AON, which may be present together with the same 2' modification in the orphan nucleotides discussed above.

[0082] Base modification

[0083] A nucleobase (sometimes called a nucleobase) is typically adenine, cytosine, guanine, thymine, or uracil, or a derivative thereof. A nucleobase is defined as the portion that can bind to another nucleobase via an H bond, a polarized bond (such as through the CF moiety), or an aromatic electron interaction. Cytosine, thymine, and uracil are pyrimidine bases, typically linked to a scaffold via their 1-nitrogen atom. Adenine and guanine are purine bases, typically linked to a scaffold via their 9-nitrogen atom. The terms “adenine,” “guanine,” “cytosine,” “thymine,” “uracil,” and “hypoxanthine” as used herein refer to the nucleobase itself. The terms “adenosine,” “guanosine,” “cytosine,” “thymine,” “uridine,” and “inosine” refer to the nucleobase linked to a (deoxy)ribosyl sugar. The nucleobases in the AON disclosed herein can be adenine, cytosine, guanine, thymine, or uracil, or any other part capable of interacting with another nucleobase via H bonds, polarization bonds (such as CF), or aromatic electron interactions. Nucleobases at any position in the AON disclosed herein can be modified forms of adenine, cytosine, guanine, or uracil, such as hypoxanthine (the nucleobase in inosine), pseudouracil, pseudocytosine, isouracil, N3-glycosylated uracil, 1-methylpseudouracil, orotic acid, guanidine, lysine nucleoside, 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-aminomethyluracil). Cytosine, 5-formylcytosine, 5-hydroxymethylcytosine, 7-deazoguanine, 7-deazoadenine, 7-deazo-2,6-diaminopurine, 8-aza-7-deazoguanine, 8-aza-7-deazoadenine, 8-aza-7-deazo-2,6-diaminopurine, 8-oxoadenine, 3-deazopurine (e.g., 3-deazoadenosine), pseudoisocytosine, N4-ethylcytosine, N2-cyclopentylguanine, N2-cyclopentyl-2-aminopurine, N2-propyl-2-aminopurine, 2,6-diaminopurine, 2-aminopurine, G-clasts and their derivatives, Super A, Super T, Super G, amino-modified nucleobases or their derivatives; and degenerate or universal bases, such as 2,6-difluorotoluene, or those absent (e.g., without a base site) (e.g., 1-deoxyribose, 1,2-dideoxyribose, 1-deoxy-2-O-methylribose, azaribose). Modified bases include synthetic and natural bases, such as inosine, xanthine, hypoxanthine, and other -aza, denitro, -hydroxy, -halogenated, -thio, thiol, -alkyl, -alkenyl, -ynyl, thioalkyl derivatives of pyrimidine and purine bases, which are known or will be known in the art. Purine nucleobases and / or pyrimidine nucleobases may be modified to alter their properties, for example, through amination or deamination of heterocycles.The exact chemistry and form may vary depending on the oligonucleotide construct and application, and may be determined according to the wishes and preferences of those skilled in the art.

[0084] Scaffold modification refers to a modified version of the naturally occurring ribose moiety in RNA (i.e., the pentose moiety), such as bicyclic sugars, tetrahydropyran, hexoses, morpholino, 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 groups, such as 2'-OMe, 2'-O-(2-cyanoethyl), 2'-MOE, 2'-O-(2-thiomethyl)ethyl, 2'-O-butyryl, 2'-O-propynyl, 2'-O-allyl, 2'-O-(2-aminopropyl), 2'-O-(2-(dimethylamino)propyl), 2'-... '-O-(2-amino)ethyl, 2'-O-(2-(dimethylamino)ethyl); 2'-deoxy (DNA); 2'-O-(haloalkyl)methyl, such as 2'-O-(2-chloroethoxy)methyl (MCEM), 2'-O-(2,2-dichloroethoxy)methyl (DCEM); 2'-O-alkoxycarbonyl, such as 2'-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2'-O-[2-N-methylaminomethyl] Acyl)ethyl](MCE), 2'-O-[2-(N,N-dimethylcarbamoyl)ethyl](DCME); 2'-halogens, such as 2'-F, FANA; 2'-O-[2-(methylamino)-2-oxoethyl](NMA); bicyclic or bridging nucleic acid (BNA) scaffold modifications, such as conformationally restricted nucleotide (CRN) monomers, locked nucleic acid (LNA) monomers, xylose-LNA monomers, α-LNA monomers, α-l- LNA monomers, β-d-LNA monomers, 2'-amino-LNA monomers, 2'-(alkylamino)-LNA monomers, 2'-(acylamino)-LNA monomers, 2'-N-substituted 2'-amino-LNA monomers, 2'-thio-LNA monomers, (2'-O,4'-C)-constrained ethyl (cEt)BNA monomers, (2'-O,4'-C)-constrained methoxyethyl (cMOE)BNA monomers, 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-pyranocyanate (DpNA) monomers, 2'-C-bridged bicyclic nucleotide (CBBN) monomers, oxo-CBBN monomers, heterocyclic-bridged BNA monomers (e.g., triazole or tetraazole linkages), 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. Monomers, F-tcDNA monomers, α-anomeric bicyclic DNA (abcDNA) monomers, oxobutane nucleotide monomers, locked PMO monomers derived from 2'-amino-LNA, guanidine-bridged nucleic acid (GuNA) monomers, spirocyclopropene-bridged nucleic acid (scpBNA) monomers and their derivatives; cyclohexenyl nucleic acid (CeNA) monomers, altriol nucleic acid (ANA) monomers, hexitol nucleic acid (HNA) monomers, fluorinated HNA (F-HNA) monomers, pyranosyl-RNA (p-RNA) monomers, 3'-deoxypyranosyl DNA (p-DNA), unlocking nucleic acid (UNA); inverted versions of any of the above monomers. All of these modifications are known to those skilled in the art.

[0085] Orphan nucleotides

[0086] Mutagenic studies of human ADAR2 have shown that a single mutation at residue 488, from glutamate to glutamine (E488Q), increases 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 edited base out of its RNA double strand and into the enzyme's active site (Matthews et al. Nat Struct Mol Biol. 2016. 23(5):426-433). When ADAR2 edits adenosine in a preferred context (A:C mismatch), the nucleotide opposite the target adenosine is usually referred to as an "orphan nucleotide" (or "orphan cytidine," depending on the context), as described above. Crystal structures of ADAR2 E488Q bound to double-stranded RNA (dsRNA) show that the glutamine (Gln; Q) side chain at position 488 can provide an H bond to the N3 position of orphan cytidine, leading to an increased catalytic rate of ADAR2 E488Q. In the wild-type enzyme, glutamate (or glutamic acid; Glu; E) is present at position 488 instead of glutamine (Gln), and the amide group of glutamine is absent; instead, a carboxylic acid is present. To achieve the same contact between orphan cytidine and the E488Q mutant in the wild-type case, protonation is required for this contact to occur. To utilize endogenously expressed ADAR2 to correct disease-related mutations, the editing efficiency of the wild-type ADAR2 enzyme in cells must be maximized. WO2020 / 252376 discloses the use of AONs with modified RNA bases, particularly at the orphan cytidine position, to mimic the hydrogen bonding pattern observed in the E488Q ADAR2 mutant. By replacing the nucleotide in AON opposite the target adenosine with a cytidine analog that acts as an H bond donor at N3, it is expected that the same contact, which is believed to provide increased catalytic rates for the mutant enzyme, can be stabilized. Two cytidine analogs of particular interest are pseudocytidine (also known as “piC”; Lu et al. JORG Chem 2009. 74(21): 8021-8030; Burchenal et al. (1976) Cancer Res 36: 1520-1523) and Benner base Z (also known as “dZ”; Yang et al. Nucleic Acid Res 2006. 34(21): 6095-6101), which were initially chosen because they provide a hydrogen bond donor at N3 and have minimal interference with nucleobase shape. The Benner base is also known as the 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase. The presence of cytidine analogs in AON can occur outside of modifications to the 2′ ribose group.The ribose 2′ group in the orphan nucleotide can be independently selected from 2′-H (i.e., DNA), 2′-OH (i.e., RNA), 2′-OMe, 2′-MOE, 2′-F, or 2′-4′-linked (i.e., bridging nucleic acids, such as locked nucleic acids (LNA)), or other 2′ substitutions. The 2′-4′ linker can be selected from adapters known in the art, such as methylene adapters or constrained ethyl adapters.

[0087] The orphan nucleotides in the AONs disclosed herein are preferably cytidine or analogues thereof (such as nucleotides carrying a Benner base) or uridine or analogues thereof (such as isouridine). Whether the orphan nucleotide is cytidine or analogues thereof, or uridine or analogues thereof, it preferably contains deoxyribose (2′-H;=DNA), but may also contain diF modification at the 2′ position of the sugar. In one embodiment, at least one adjacent (directly adjacent) nucleotide of the orphan nucleotide, and in another embodiment, two adjacent nucleotides, do not contain 2′-OMe modification. All nucleotides of the oligonucleotide (including the orphan nucleotide) have 2′-OMe modification and complete modification of the native base, resulting in the oligonucleotide being nonfunctional for RNA editing (known in the art), possibly because it inhibits ADAR activity at the target site. Typically, adenosine in the target RNA can be protected from editing by providing a relative nucleotide with a 2′-OMe group (at least when there are no other chemical substitutions or modifications in the nucleotide), or by providing guanine or adenine as a relative base, as these two nucleobases can also reduce editing relative to adenosine.

[0088] Connect key modifier

[0089] Nucleosides are typically linked to adjacent nucleotide monomers via the condensation of their 5'-phosphate moiety with the 3'-hydroxy moiety of the adjacent nucleotide monomer. Similarly, their 3'-hydroxy moiety is typically linked to the 5'-phosphate of the adjacent nucleotide monomer. This forms a phosphodiester bond. The phosphodiester and the scaffold form an alternating copolymer. Bases are grafted onto this copolymer, i.e., onto the scaffold moiety. Due to this property, the alternating copolymer formed by the linking scaffold of oligonucleotides is often referred to as the "backbone" of the oligonucleotide. Because the phosphodiester bonds link adjacent monomers together, they are often called "backbone linkages." It should be understood that when the phosphate group is modified so that it is replaced by a similar moiety (such as a thiophosphate), this moiety is still called a backbone linkage of the monomer. This is called "backbone modification." In general, the backbone of an oligonucleotide consists of an alternating scaffold and backbone linkages.

[0090] As detailed herein, the naked AONs disclosed herein contain at least one, preferably multiple, linker modifications. More preferably, the AONs disclosed herein contain linker modifications at most positions, and may contain linker modifications at all positions if the AON is capable of deaminase-mediated RNA editing upon binding to a target RNA nucleic acid molecule. Linker modifications can be, but are not limited to, modified versions of phosphodiester present in RNA, such as thiophosphate (PS), chiral pure PS, (R)-PS, (S)-PS, methylphosphonate (MP or MeP), chiral pure MP, (R)-MP, (S)-MP, phosphorylguanidine (e.g., PNdmi), chiral pure phosphorylguanidine, (R)-phosphonylguanidine, (S)-phosphonylguanidine, dithiophosphate (PS2), phosphonoacetate (PACE), phosphonoacetamide (PACA), thiophosphonoacetate, thiophosphonoacetamide, methylthiophosphate, methylthiophosphonate, PS prodrug, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, boran phosphate, boran PS, methylborane phosphate, methylborane PS, methylborane phosphonate, methylborane thiophosphate, phosphate ester, phosphate triester, aminoalkyl phosphate triester and its derivatives. Another type of modification includes phosphoramide, phosphoramide, N3'→P5' phosphoramide, diphosphodiamid, thiophosphodiamid, sulfonamide ester, diethylene sulfoxide, amide, sulfonate ester, siloxane, sulfide, sulfone, formylacetyl, alkenyl, methylenehydrazine, sulfonamide, triazole, oxalyl, carbamate, methyleneimine (MMI), and thioacetamide nucleic acid (TANA); and their derivatives. It also includes various salts, mixed salts, deprotonated, protonated, tautomerized, and free acid forms, as well as 3'→3' and 2'→5' linkages. The AON disclosed herein may also contain one or more linkage modifications according to the structure of formula (I), (II), (III), (IV), or (V).

[0091] In a preferred aspect, the AON disclosed herein comprises an internucleotide linker having the structure of formula (I), wherein X = O and R = CH3, and this linker is generally referred to herein as a PNms linker. In other preferred aspects, R is equal to one of the following structures: (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0092]

[0093] The AON disclosed herein contains one or more PN connection bonds as shown in equation (I), which can be independently R. P or S P Chirality, or stereo randomness.

[0094] In the AON disclosed herein, one or more PN linkages as shown in formula (I) may be in tautomeristic and / or pH-dependent (de)protonated forms, including but not limited to structures (A), (B), (C), (D), and (E):

[0095]

[0096]

[0097] X and R are as shown in equation (I) above.

[0098] This paper also discloses an AON that, after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in the cell, mediates the deamination of adenosine by recruiting a deaminase in the cell, wherein the region contains the target adenosine, wherein the deaminase can deaminate the target adenosine to inosine, and wherein the AON contains a portion having a structure according to formula (II) at one and / or both ends:

[0099]

[0100] Where: X = O or S;

[0101] Y = O - or S - ;and

[0102] R = aryl, substituted aryl, heterocycle, substituted heterocycle, aromatic heterocycle, substituted aromatic heterocycle, C1-C6 alkoxy, substituted C1-C6 alkoxy, C1-C 20 Alkyl, substituted C1-C 20 Alkyl, C1-C6 alkenyl, C1-C6 substituted alkenyl, C1-C6 ynyl, substituted C1-C6 ynyl, or conjugated groups. In a preferred embodiment, X = O and R = methyl. Preferred internucleotide linker modifications for the AON disclosed herein have a structure of formula (III):

[0103]

[0104] It is also known as the PNms linker. The PNms linker and its application in oligonucleotides, as well as its alternative to the PS linker, have been described (Chelobanov BP et al. Russian J Bioorganic Chemistry. 2017. 43(6):664-668; DOI:10.1134 / S1068162017060024; Klabenkova K et al. Molecules. 2021. 26(17):5420; Miroshnichenko SK et al. Proc Natl Acad Sci USA. 2019. 116(4):1229-1234), for example in oligonucleotides that provide splicing conversion (Hammond SM et al. Nucleic Acid Ther. 31(3):190-200).

[0105] The AON disclosed herein may contain a non-bridging oxygen substitution in the phosphodiester linkage. This modification slightly disrupts base pairing but significantly increases resistance to nuclease degradation. Preferred nucleotide analogs or equivalents include PS, phosphonoacetate, dithiophosphate, phosphate triester, aminoalkyl phosphate triester, H-phosphonate, methyl and other alkylphosphonates (including 3′-alkylphosphonene, 5′-alkylphosphonene and chiral phosphonates), phosphonates, phosphoramides (including 3′-aminophosphoramide and aminoalkylphosphoramide), thiophosphoramide, thioalkylphosphonate, thioalkyl phosphate triester, selenophosphate, or boron phosphate. Particularly preferred are modifications to include PS in the internucleotide linkage. Particularly preferred are modifications to include PNms in the internucleotide linkage. Particularly preferred are modifications to include PNdmi in the internucleotide linkage. Conventional internucleotide linkages between nucleotides can be altered by monothiolation or dithiolation of the phosphodiester bond to produce PS esters or dithiophosphate esters, respectively. Other modifications to the internucleotide linker are possible, including amidation and peptide linkers. Those skilled in the art can determine for which target RNA nucleic acid molecules the AON contains specific linker modifications at each linker position of the AON disclosed herein to produce the most efficient and stable oligonucleotide compound.

[0106] Many non-naturally occurring linkers (such as PS) are modified to be chiral. This means that Rp and Sp configurations exist, as is known to those skilled in the art. In one embodiment, the chirality of the PS linker is controlled, meaning that each linker is in either an Rp or Sp configuration, whichever is preferred. The selection of an Rp or Sp configuration at a given linker position may depend on the target sequence and the efficiency of binding and inducing RNA editing that results in the target adenosine. However, unless there is a particular expectation, the composition may comprise an AON as the active compound having both Rp and Sp configurations at a specific given linker position. Mixtures of such AONs are also feasible, wherein some positions preferably have either configuration, while others are irrelevant. In one aspect, the AON disclosed herein comprises one or more (chiral pure or chiral mixed) PS linkers. In one aspect, the AON disclosed herein comprises one or more (chiral pure or chiral mixed) phosphoramide (PN) linkers. In one aspect, the AON disclosed herein comprises one or more (chiral pure or chiral mixed) PNms linkers. In one aspect, the PN linkers connect the terminal two nucleotides at each end of the AON. The AON disclosed herein may also contain linker modifications at all non-chirally controlled positions. The AON disclosed herein may also contain one or more naturally occurring internucleotide linkers. The selection and number of modified linkers may depend on the specific target, sequence, length, and stability of the AON observed in a particular cell type of interest, and can be assessed using methods known to those skilled in the art. In one aspect, at least two, at least three, or at least four internucleotide linkers between two, three, four, or five nucleotides at the 5' and / or 3' ends of the AON disclosed herein are modified internucleotide linkers. In one aspect, the AON disclosed herein contains at least one MP internucleotide linker according to the structure of formula (IV):

[0107]

[0108] As described in the art, the preferred position of the MP linker in an AON is linker position -2, thereby linking the nucleotide at position -1 to the nucleotide at position -2. In a preferred embodiment, this position in the AON disclosed herein contains a linker modification according to the structure of formula (I), more preferably a linker modification according to the structure of formula (III), rather than an MP linker. WO2020 / 201406 discloses the use of MP linker modifications at certain positions around orphan nucleotides in the first nucleic acid strand. Although the presence of MP linkers is compatible with RNA editing by human ADAR enzymes, introducing MP linkers during oligonucleotide fabrication requires additional fabrication (purification) steps during coupling and decoupling, and is therefore challenging. In one aspect, the AON does not contain an MP linker.

[0109] In one aspect, the AON disclosed herein comprises at least one PNdmi linker, preferably linking the two terminal nucleosides at the 5′ and / or 3′ ends of the AON. The PNdmi linker used in the AON disclosed herein has the structure of formula (V):

[0110]

[0111] In one aspect, an inverted deoxy-T or dideoxy-T nucleotide may be incorporated into either or both ends of the AON disclosed herein. Other nucleotide-linking bonds that may be used in the AON disclosed herein are those disclosed in WO2023 / 278589.

[0112] In one aspect, the AON disclosed herein comprises at least one phosphonoacetate and / or at least one phosphonoacetamide nucleoside linker.

[0113] Conjugate Chemistry

[0114] In one aspect, the AON disclosed herein, or its sense strand which can be annealed before entering the target cell (in the HEON disclosed herein), binds to a hydrophobic moiety, such as palmitoyl or an analogue thereof, cholesterol or an analogue thereof, or tocopherol or an analogue thereof. Preferably, it binds to the 5' end. If the hydrophobic moiety binds to both the 5' and 3' ends, these hydrophobic moieties can be the same or different. The hydrophobic moiety bound to the oligonucleotide can bind directly or indirectly mediated by another substance. When the hydrophobic moiety binds directly, it can be bound by covalent bonds, ionic bonds, hydrogen bonds, etc. When the hydrophobic moiety binds indirectly, it can be bound by a linker group (connector). The linker can be cleavable or non-cleavable. A cleavable linker is one that can be cleaved under physiological conditions (e.g., in cells or animals, such as humans). Cleavable linkers can be selectively cleaved by endogenous enzymes (such as nucleases) or by the physiological environment of a specific site in the body or cell (such as pH or a reducing environment, such as glutathione concentration). Examples of cleavable adapters include, but are not limited to, amides, esters, one or two esters of phosphodiesters, phosphate esters, carbamates, disulfide bonds, and native DNA adapters. Cleavable adapters also include self-eliminating adapters. Non-cleavable adapters are adapters that do not cleave under physiological conditions, or that cleave very slowly compared to cleavable adapters, such as PS linkages, modified or unmodified deoxyribonucleosides linked by PS linkages, spacers linked by PS linkages, and adapters composed of modified or unmodified ribonucleosides. When the adapter is a nucleic acid (such as DNA) or an oligonucleotide, there is no limitation on the chain length. However, its length is typically 2 to 20 bases, 3 to 10 bases, or 4 to 6 bases. There are no limitations on the length or composition of the spacer linking the ligand and the oligonucleotide, and it may include, for example, ethylene glycol, triethylene glycol (TEG), HEG, alkyl chains, propyl, 6-aminohexyl, or dodecyl. In one embodiment, the GalNAc moiety binds to the AON disclosed herein via a TEG adapter. One or more other types of molecules can bind to AON through one or more linkers, including peptides, sugars, vitamins, polymers, aptamers, fragments of antibodies, small molecules, etc.

[0115] Overview

[0116] In addition to specific preferred chemical modifications at certain positions in the compounds disclosed herein, the AONs disclosed herein may contain one or more (additional) modifications to nucleobases, scaffolds, and / or backbone linkages, which may or may not be present in the same monomer, for example at the 3' and / or 5' positions. In one aspect, the AONs disclosed herein contain at least one nucleoside linkage according to formula (I), and / or the AON further contains at least one nucleotide with a sugar moiety comprising a 2'-OMe modification, and / or the AON contains at least one nucleotide with a sugar moiety comprising a 2'-MOE modification, and / or the AON contains at least one nucleotide with a sugar moiety comprising a 2'-F modification, and / or the AON contains an orphan nucleotide with a 2'-H sugar moiety, thus referred to as a DNA nucleotide, even if other modifications may be present in its bases and / or linkages with adjacent nucleosides. In one aspect, the sugar moiety of the orphan nucleotide has a 2'-F sugar moiety. In one aspect, the sugar moiety of the orphan nucleotide has a diF substitution. In one aspect, the sugar moiety of the orphan nucleotide has both a 2'-F and a 2'-C-methyl sugar moiety. In one respect, the sugar moiety of orphan nucleotides contains 2'-F in the arabinose configuration (FANA).

[0117] In one respect, AON is an antisense oligonucleotide that can form a double-stranded nucleic acid complex with a target RNA molecule, wherein the double-stranded nucleic acid complex recruits adenosine deaminase to deaminate the target adenosine in the target RNA molecule, wherein the nucleotide opposite to the target adenosine in AON is an orphan nucleotide, and wherein the orphan nucleotide has the structure of formula (VI):

[0118]

[0119] Wherein: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from cytosine, uracil, isouriacil, N3-glycosylated uracil, pseudoisocytosine, 8-oxoadenine, and 6-amino-5-nitro-3-yl-2(1H)-pyridone; R1 and R2 are independently selected from H, OH, F, or CH3; R3 is the portion of AON located at the 5' end of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4 is the portion of AON located at the 3' end of the orphan nucleotide, consisting of 4 to 25 nucleotides. The nucleotides at the 3' and / or 5' ends of the orphan nucleotides may (more preferably, the nucleotides at the 3' end (-1 position)) be DNA.

[0120] Other chemical modifications of the AON disclosed herein include replacing one or more of any hydrogen atoms with deuterium or tritium, examples of which can be found, for example, in WO2014 / 022566 or WO2015 / 011694. Similarly, in all cases, the modifications should be editable, enabling the AON to function as an oligonucleotide, i.e., recruiting adenosine deaminases upon binding to a target sequence due to the resulting double-stranded nucleic acid entity. In all aspects of this disclosure, the enzyme having adenosine deaminase activity is preferably ADAR1, ADAR2, or ADAT.

[0121] The AON disclosed herein preferably does not contain 5′-terminal O6-benzylguanosine or 5′-terminal amino modification, and preferably is not covalently linked to a SNAP tag domain (an engineered O6-alkylguanosine-DNA-alkyltransferase). The AON disclosed herein preferably does not contain a boxB RNA hairpin sequence. In one aspect, the AON disclosed herein contains 0, 1, 2, or 3 wobble base pairs with the target sequence, and / or 0, 1, 2, 3, 4, 5, 6, 7, or 8 mismatched base pairs with the target RNA sequence. When the orphan nucleotide is uridine, there is no mismatch; the definition may differ when the orphan nucleotide is a uridine analog or derivative. An alternative to uridine is to place an isouridine opposite the target adenosine, which may not pair like G and U. Preferably, the target adenosine in the target sequence forms a mismatched base pair with the nucleoside in the AON that directly opposite the target adenosine.

[0122] As described above, the AON disclosed herein undergoes specific nucleotide modifications at specific locations to ensure stability and proper ADAR binding and activity. These modifications may vary and may include modifications to the AON backbone, nucleotide sugar moiety, and nucleobase or phosphodiester linkages, as detailed herein. They may also be variable in the sequence of the AON. Specific modifications may be required to support the interaction of different amino acid residues within the RNA-binding domain and deaminase domain of the ADAR enzyme. For example, PS linkages between nucleosides or 2′-OMe or 2′-MOE modifications may be tolerable in some parts of the AON, but should be avoided in others to prevent disruption of the enzyme's critical interactions with the phosphate and 2′-OH groups. Specific nucleotide modifications may also be necessary to enhance the editing activity of the substrate RNA, where the target sequence is not optimal for ADAR editing. Previous studies have shown that certain sequence contexts are more conducive to editing. For example, the target sequence 5'-UAG-3' (with target A in the middle) contains the most preferred adjacent nucleotide of ADAR2, while the 5'-CAA-3' target sequence is less desirable (Schneider et al., 2014. Nucleic Acids Res 42(10):e87). Structural analysis of the ADAR2 deaminase domain suggests that editing may be enhanced by carefully selecting the nucleotide opposite the target trinucleotide. For example, the pairing of the 5'-CAA-3' target sequence with the 3'-GCU-5' sequence on the opposite strand (forming an AC mismatch in the middle) is less desirable because the guanine base sterically conflicts with the amino acid side chain of ADAR2. In this case, the guanine opposite C is preferably replaced by inosine (therefore at the -1 position of AON), and more preferably deoxyinosine.

[0123] The AON disclosed herein, unlike descriptions of siRNA or gapmers and their relationship to RNase degradation, and the use of such gapmers in double-stranded complexes (see, for example, EP 3954395 A1), does not contain a DNA nucleotide segment that would make the target sequence (or sense nucleic acid strand) a target for RNase-mediated degradation. Degradation of the target transcript molecule by binding to the AON is not desired. In one embodiment, the AON does not contain four or more consecutive DNA nucleotides at any position within its sequence. In embodiments, the AON consists of as many (chemically) modified nucleotides as possible to enhance resistance to RNase-mediated degradation while producing RNA editing effects as efficiently as possible. This means that the orphan nucleotide and several other nucleotides in the AON can be DNA, but there is no segment of four or more consecutive DNA nucleotides within the AON. Therefore, the AON disclosed herein is not a gapmer. Gapmers reduce the expression of the target transcript but do not produce RNA editing of specific adenosines within the target transcript. Interstitial oligonucleotides are, in principle, single-stranded nucleic acids consisting of a central region (a DNA interstitial region with at least four consecutive deoxyribonucleotides) and flanking regions located directly at their 5' end (5' flanking region) and 3' end (3' flanking region). In contrast, the AONs disclosed herein can be any oligonucleotide that produces an RNA editing effect, wherein the target adenosine in the target RNA molecule is deaminated to inosine, thus exhibiting the highest possible resistance to RNase-mediated degradation to produce this effect and allow the mRNA transcript to be translated into protein.

[0124] The AONs disclosed herein can also be administered in the context of adjuvants that increase AON entry into target cells and / or endosome escape after entry. A subset suitable for such applications is, for example, a group of chemical compounds (typically purified from nature) referred to as “saponins” or “triterpenoid glycosides.” A preferred saponin suitable for use in the methods disclosed herein is AG1856, disclosed in WO2021 / 122998, and further described in PCT / EP2024 / 051278 (not disclosed) for its use with oligonucleotides that produce RNA editing.

[0125] This document also discloses pharmaceutical compositions comprising the AON disclosed herein, and further comprising pharmaceutically acceptable carriers, solvents, diluents, and / or other additives (such as saponins or triterpenoid glycosides discussed above, such as AG1856, which may also be administered separately from the AON), and soluble in pharmaceutically acceptable organic solvents, etc. The dosage form for administration of the AON or pharmaceutical composition may depend on the disease to be treated and the tissue to be targeted, and may be selected according to conventional procedures in the art. The pharmaceutical composition may be administered by single-dose or multiple-dose administration. It may be administered daily or at appropriate intervals, as determined using common knowledge in the art, and may be adjusted according to the disease and the potency of the active ingredient.

[0126] Although in a preferred embodiment, the AON disclosed herein is a single-stranded oligonucleotide comprising an orphan nucleotide opposite to the target adenosine, wherein the orphan nucleotide is chemically modified as disclosed herein, and wherein the remainder of the oligonucleotide is also chemically modified to prevent its degradation by nucleases, as disclosed herein, in another embodiment, any kind of oligonucleotide or heteroduplex oligonucleotide complex is disclosed, which may or may not be bound to a hairpin structure (internal or terminal), which may be bound to ADAR or its catalytic domain, or wherein the oligonucleotide is in a cyclic form. In a preferred aspect, the AON disclosed herein is a “naked” oligonucleotide containing various chemical modifications in the ribose and / or bases of one or more nucleotides within the sequence, preferably containing at least one linker according to formula (I) as disclosed herein, which may hybridize to the target transcript or its portion containing the target adenosine, and may recruit endogenous (naturally occurring) ADAR in the target cell to deaminate the target adenosine. In another respect, the AONs delivered in “naked” form disclosed herein do not contain stem-loop structures for recruiting deaminases, which allows for shorter AONs and improved cell delivery and transport.

[0127] RNA editing entities known in the art (such as human ADAR enzymes) edit dsRNA structures with varying degrees of specificity, depending on several factors. One important factor is the degree of complementarity between the two strands that make up the dsRNA sequence. Perfect complementarity of the two strands typically results in the catalytic domain of human ADAR deaminating adenosine in a non-discriminatory manner, reacting with any adenosine encountered. The specificity of hADAR1 and hADAR2 can be enhanced by introducing chemical modifications into the dsRNA and / or ensuring several mismatches, which may facilitate the localization of the dsRNA-binding domain in a manner not yet well-defined. Furthermore, the deamination reaction itself can be enhanced by providing oligonucleotides containing mismatches opposite the adenosine to be edited. Following the description in this application, those skilled in the art will be able to design complementary portions of oligonucleotides according to their needs.

[0128] Those skilled in the art will understand that the extent to which intracellular editing enzymes are redirected to other target sites can be modulated by altering the affinity of the first nucleic acid strand for the editing enzyme recognition domain. The exact modification can be determined through trial and error and / or computational methods based on the structural interactions between the AON and the editing enzyme recognition domain. Alternatively, the degree of recruitment and redirection of intracellular editing enzymes can be modulated by the dosage and administration regimen of the AON. This will be determined by the experimenter (in vitro) or clinician, typically in Phase I and / or Phase II clinical trials.

[0129] This document discloses site-specific editing of target adenosine sequences in RNA sequences of eukaryotic cells, preferably metazoan cells, more preferably mammalian cells, more preferably human cells, more preferably human liver cells, and most preferably human hepatocytes. Target cells can be located in vitro, ex vivo, or in vivo. One advantage of the AON disclosed herein is its applicability to in situ cells in living organisms, but also to cells in culture. In some embodiments, cells are treated ex vivo and then introduced into a living organism (e.g., reintroduced from the organism of their original source). The AON disclosed herein can also be used to edit target RNA sequences in graft cells or so-called organoids (e.g., liver tissue organoids). Organoids can be considered three-dimensional in vitro derived tissues, but are generated as single, isolated tissues using specific conditions. They are useful in therapeutic settings because they can be derived from patient cells in vitro, and the organoids can then be reintroduced into the patient as autologous material, less likely to be rejected compared to normal grafts.

[0130] Not wishing to be bound by theories, such as RNA editing via human ADAR2, which is thought to occur on the primary transcript in the cell nucleus during transcription or splicing, or in the cytoplasm, where, for example, mature mRNA, miRNA, or ncRNA can be edited. Generally, RNA editing can be used to produce RNA sequences with different properties. Such properties can be coding properties (producing proteins with different sequences or lengths, resulting in altered protein properties or functions) or binding properties (resulting in inhibition or overexpression of the RNA itself or its target or binding chaperone; this can be achieved by recoding miRNAs or their homologous sequences on target RNAs). Protein function or localization can be arbitrarily altered through functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co-translation or post-translational modifications, catalytic sites of enzymes, binding sites of binding chaperones, degradation or activation signals, etc. These and other forms of RNA and protein “engineering,” whether for the prevention, delay, or treatment of disease, or for any other medical or biotechnological purpose, as diagnostic agents, preventative agents, therapeutic agents, research tools, or otherwise, are included within the scope of this disclosure.

[0131] The amount, dosage, and administration regimen of AON to be administered may vary depending on cell type, the disease to be treated, the target population, the route of administration (e.g., systemic vs. local), disease severity, and acceptable levels of side effects; however, these can and should be assessed through trial and error during in vitro studies, preclinical trials, and clinical trials. Trials are particularly definitive when the modified sequence results in easily detectable phenotypic changes or changes in the level or activity of specific biomarkers (e.g., plasma bile acid levels). Higher doses of AON may competitively bind to intracellular ADAR enzymes, thereby depleting the amount of enzyme freely available for RNA editing; however, standard dosing studies will reveal any such effects of a given AON and a given target.

[0132] A suitable assay technique involves delivering AON to a cell line or test organism and then collecting biopsy samples at different time points. The sequence of the target RNA can be evaluated in the biopsy samples, and the proportion of cells with modification can be easily tracked. As mentioned above, the plasma concentration of bile acids in samples from treated subjects, before and after treatment, or with or without the use of the AON disclosed herein, is an appropriate biomarker for assessing the function of certain proteins in the subject. Once the assay is performed, this knowledge can be retained, and future deliveries may not require the collection of biopsy samples. Therefore, the methods disclosed herein can include identifying the presence of desired changes in the cellular target RNA sequence to verify that the target RNA sequence has been modified. This step would typically involve sequencing the relevant portion of the target RNA or a copy of its cDNA (or a copy of the cDNA of its splice product, if the target RNA is pre-mRNA), as mentioned above, thus allowing for easy verification of sequence changes. Alternatively, as mentioned above, the changes can be assessed by evaluating protein function or any other potential biomarkers before, during, and / or after treatment, preferably in vitro on samples from treated subjects.

[0133] Following RNA editing in a cell, the modified RNA can be diluted over time, for example, due to cell division or the limited half-life of the edited RNA. Therefore, in practical therapeutic applications, the methods disclosed herein may include repeated delivery of AON until sufficient target RNA is modified to provide tangible benefits to the patient and / or maintain those benefits over time.

[0134] The AON disclosed herein is particularly suitable for therapeutic use, and therefore pharmaceutical compositions comprising the AON disclosed herein and a pharmaceutically acceptable carrier, solvent, or diluent are also disclosed. In some embodiments, the pharmaceutically acceptable carrier may simply be an aqueous saline solution. Especially for pulmonary delivery, it may be isotonic or hypotonic. The AON disclosed herein is suitably administered in aqueous solution (e.g., saline) or suspension form, optionally containing additives, excipients, and other components compatible with pharmaceutical use, at concentrations ranging from 1 ng / ml to 1 g / ml, preferably from 10 ng / ml to 500 mg / ml, more preferably from 100 ng / ml to 100 mg / ml. Dosage ranges suitably from about 1 μg / kg to about 100 mg / kg, preferably from about 10 μg / kg to about 10 mg / kg, more preferably from about 100 μg / kg to about 1 mg / kg. The administration can be by inhalation (e.g., via nebulization), intranasal, oral, by injection or infusion, intravenous, subcutaneous, intradermal, intramuscular, intratracheal, intraperitoneal, rectal, intrathecal, intracerebrospinal, parenteral, etc. The form of administration can be solid, powder, pill, gel, solution, sustained-release formulation, or any other form compatible with human drug use.

[0135] In one implementation, depending on the final deamination effect of the A-to-I conversion, the identification step of whether editing has occurred includes the following steps: sequencing the target RNA; assessing the presence or absence of nonfunctional or reduced-function proteins; assessing whether the splicing of the pre-mRNA has been altered by deamination; or using functional readouts, since the deamination-encoded target RNA should encode a protein with reduced or missing function, or conversely, a protein with increased or restored function. Identification of inosine deamination can be performed using appropriate biomarkers via functional readouts. Functional assessment will generally be performed according to methods known to those skilled in the art. A suitable way to identify the presence of inosine after deamination of target adenosine is, of course, dPCR or even sequencing, using methods well known to those skilled in the art. However, those skilled in the art of liver disease will prefer to apply tests to monitor certain biomarkers related to liver function.

[0136] In one embodiment, the method disclosed herein includes the following steps: administering the AON or pharmaceutical composition disclosed herein to a subject; forming a double-stranded nucleic acid complex between the AON and a specific complementary target nucleic acid molecule in the subject's cells; ligating an endogenously present adenosine deaminase (e.g., ADAR1 or ADAR2); and causing the enzyme to deaminate the target adenosine in the target nucleic acid molecule to inosine, thereby alleviating, treating, improving, or slowing disease progression.

[0137] RNA editing molecules present in cells are typically protein-based, such as ADAR enzymes found in metazoans, including mammals. Human ADAR, hADAR1, and hADAR2, including any of their isoforms, are of particular interest. RNA editing enzymes known in the art can be readily engineered into the oligonucleotide constructs disclosed herein, including RNA-acting adenosine deaminases (ADARs), such as hADAR1 and hADAR2 in humans or human cells, as well as cytidine deaminases. hADAR1 is known to exist in two isoforms: a long 150 kDa interferon-inducible form and a short 110 kDa form, produced via alternative splicing of a common pre-mRNA. Therefore, the level of the 150 kDa isoform in cells can be influenced by interferon, particularly interferon-γ (IFN-γ). hADAR1 can also be induced by TNF-α. This provides an opportunity to develop combination therapies, whereby IFN-γ or TNF-α is administered to patients as a combination product or as a single product with the AON disclosed herein, concurrently or subsequently in any order. Certain disease conditions may be associated with elevated levels of IFN-γ or TNF-α in certain tissues of patients, creating further opportunities for more specific editing targeting of diseased tissues. Those skilled in the art will understand that the extent to which intracellular editing entities are redirected to other target sites can be modulated by altering the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0138] The AON disclosed herein utilizes endogenous cellular pathways and naturally occurring ADAR enzymes to specifically edit target adenosines in target RNA sequences. The AON disclosed herein recruits ADAR and forms a complex with it, then promotes the deamination of a (single) specific target adenosine nucleotide in the bound target RNA sequence to inosine. Ideally, only one adenosine is deaminated. The AON disclosed herein, upon complexing with ADAR, preferably induces the deamination of a single target adenosine.

[0139] The AONs disclosed herein, especially when in naked form, are typically longer than 10 nucleotides, preferably more than 11, 12, 13, 14, 15, or 16 nucleotides, and even more preferably more than 17 nucleotides. In one aspect, the AONs disclosed herein are longer than 20 nucleotides. The AONs disclosed herein are preferably shorter than 100 nucleotides, even more preferably shorter than 60 nucleotides, and even more preferably shorter than 50 nucleotides. In a preferred aspect, the AONs disclosed herein contain 18 to 70 nucleotides, more preferably 18 to 60 nucleotides, and even more preferably 18 to 50 nucleotides. Therefore, in a particularly preferred aspect, the AON disclosed herein 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. In one embodiment, the AON is 27, 28, 29, or 30 nucleotides in length.

[0140] Example

[0141] Example 1. RNA editing of SLC10A1 transcripts using multiple AONs.

[0142] An initial set of 4 × 30 AONs was designed to target each of the four target adenosines in the human SLC10A1 transcript (respectively). The design and chemical modification of these 120 AONs are as follows: Figure 1 , Figure 2 , Figure 3 and Figure 4 Provided. Subsequently, a large number of additional AONs were designed to target adenosine at the CAG codon encoding glutamine (Q) in SEQ ID NO:1 and convert it to the CGG codon encoding arginine (R), thereby introducing the c.203A>G(Q68R) mutation. These additional AONs with their own chemical modifications are as follows: Figure 5 Provided. In addition, a large number of additional AONs were designed to target adenosine at the GAG ​​codon encoding glutamate (E) in SEQ ID NO:3 and convert it to the GGG codon encoding glycine (G), thereby introducing the c.770A>G (E257G) mutation. These additional AONs, each with its own chemical modifications, are as follows: Figure 6 Provided.

[0143] For initial screening of AON, the human hepatoblastoma cell line HuH-6 (Cell Lines Service) expressing SLC10A1 mRNA was transfected with 100 nM AON using a Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer's instructions. In addition, HuH-6 cells were subjected to gymnotic exposure with different concentrations of EON (with or without the chemical reagent). After 72 hours of incubation, the culture medium was removed, and total RNA was isolated using the RNeasyMicro kit (Qiagen). cDNA was synthesized using a mixture of random hexamer and oligo(dT) primers using the Maxima Reverse Transcriptase kit. The percentage of ADAR-mediated A to I conversion was determined by quantitative digital PCR (dPCR) assays designed for different target sites, using primers and probes provided in Table 1. The percentage was calculated by dividing the number of guanine-containing cDNA species by the total number of target copies and then multiplying by 100. Control dPCR was performed upstream or downstream of the transcript for normalization.

[0144] Table 1: Primers and probes used for quantitative PCR assays and their respective sequences. Primers and probes for detecting exon 1 and exon 4 skipping are also provided. Four different targets are given by their amino acid variations. The SEQ ID NO for each primer or probe is given in parentheses after its respective sequence in the middle column. + indicates locked nucleic acid (LNA) on the 3' side.

[0145]

[0146]

[0147] Example 2. Using saponins as transfection helpers, RNA editing of two adenosines in human SLC10A1 transcripts was performed in primary human hepatocytes.

[0148] Primary human hepatocytes (PHH) were cultured and plated at approximately 700,000 cells per well, and treated with 5 μM AON in the presence of 0.5 μM MAG1856 for 72 h. Cells were then harvested, and RNA was isolated as described above. Subsequently, dPCR was performed using the appropriate primers, and the percentage of editing of the A>G change at position 203 of human SLC10A1 mRNA (associated with the Q68R mutation in the protein) and the A>G change at position 770 of human SLC10A1 mRNA (associated with the E257G mutation in the protein) were determined. Both experiments were performed twice. Figure 7 A and Figure 7B shows the use of 16 AONs as shown in the figure (e.g. Figure 5 The percentage of editing (Q68R) of c.203A>G was measured in two experiments. Although the percentage was slightly lower in the second experiment, the editing percentage was as high as 65%, with RM106622, RM106624, RM106626, RM106631, RM106632, RM106633 and RM106634 showing the best performance. Figure 8 A and Figure 8 B shows the use of 17 AONs (e.g.) Figure 6 The percentage of edits (E257G) for c.770A>G edits is measured in two experiments. Although the percentage is slightly lower in the second experiment, the edit percentage is as high as 55%, with the four asymmetric AONs (where the length of the AON 5' portion calculated from the orphan position is much longer than that of the AON 3' portion) RM106580, RM106581, RM106582 and RM106583 performing best, as shown in the figure.

[0149] Example 3. RNA editing of two adenosines in human SLC10A1 transcripts using naked uptake in primary human hepatocytes.

[0150] The experiments of Example 2 were repeated in PHH cells, but without the aid of AG1856 saponin during incubation. This method of oligonucleotide entry into cells is also known as "naked uptake" or "naked transfection" (gymnosis). The incubation time was again 72 hours, but with 10 μM AON and approximately 200,000 cells per well. For c.203A>G (Q68R), a total of 18 different AONs were tested; for c.770A>G (E257G), a total of 16 AONs were tested. Figure 9 A and Figure 9 B shows the use of 18 AONs as shown in the figure (e.g. Figure 5 The percentage of editing (Q68R) of c.203A>G was measured in two naked intake experiments. Although the percentage was slightly lower again in the second experiment, the percentage of editing was as high as 2%, with RM106620, RM106622, RM106873 and RM106874 showing the best performance. Figure 10 A and Figure 10 B shows the use of 16 AONs (e.g.) Figure 6The percentage of editing (E257G) of c.770A>G was measured in two naked uptake experiments. Although the percentage in the second experiment was also slightly lower, the percentage of editing was as high as 1.7%, where RM106867 and the four asymmetric AONs (where the length of the AON 5' portion calculated from the orphan position is much longer than that of the AON 3' portion) RM106580, RM106581, RM106582 and RM106583 again showed significantly good results, as shown in the figure.

[0151] Example 4. RNA editing of two adenosines in human SLC10A1 transcripts in hepatocytes using saponin-supported uptake and naked uptake.

[0152] The experiments described in Examples 2 and 3 were repeated for both target sites, but now hepatic spheroid organoids generated from PHH cells using culture conditions known to those skilled in the art were used. Eight spheroids were used per incubation, incubated for 120 hours with 5 μM AON in each case, with or without 0.5 μM AG1856. RNA purification, cDNA generation, and dPCR were performed as described above. Figure 11 A and Figure 11 B shows the use of 16 AONs (such as...) in experiments with and without AG1856, respectively. Figure 5 The percentage of editing (Q68R) measured for c.203A>G editing is given. Similarly, in these liver organoids, the editing percentage reached as high as 75% in the presence of saponins, with RM106622, RM106873, RM10631, RM106632, RM106633, and RM106634 showing the best performance. However, when saponins were not used, i.e., without any cell entry support, RM106622 outperformed all other AON tests, reaching a level of nearly 8%. Figure 12 A and Figure 12 B shows the use of 17 AONs (such as...) in experiments with and without AG1856, respectively. Figure 5 The percentage of editing (as shown) for c.770A>G editing (E257G) was measured. Similarly, in these liver organoids, the editing percentage reached as high as 55% in the presence of saponins, with the “asymmetric” AONs RM106580, RM106581, RM106582, and RM106583 showing the best performance, also reaching nearly 4.5% in the absence of saponins.

[0153] Example 5. RNA editing of two adenosines in human SLC10A1 transcript in human HepG2 cells overexpressing NTCP using saponin-supported uptake.

[0154] Human HepG2 cells stably expressing wild-type human NTCP (HepG2NTCP) were cultured and plated at approximately 100,000 cells per well, and treated with 5 μM AON in the presence of 0.5 μM AG1856 for 72 h. Cells were then harvested, and RNA was isolated as described above. Subsequently, dPCR was performed using the appropriate primers, and the percentage of editing of the A>G change at position 203 of human SLC10A1 mRNA (associated with the Q68R mutation in the protein) and the A>G change at position 770 of human SLC10A1 mRNA (associated with the E257G mutation in the protein) were determined. Both experiments were performed twice. Figure 13 The diagram shows the use of 11 AONs (e.g., ...). Figure 5 The percentage of edits (Q68R) measured for c.203A>G edits are given. The percentage of edits is as high as 25%, with RM106631, RM106632, RM106633 and RM106634 performing best. Figure 14 This shows the use of 10 AONs as shown in the figure (e.g.) Figure 6 The percentage of edits (as shown) for c.770A>G edits (E257G) is measured. The percentage of edits is as high as 25%, with the four asymmetric AONs (where the length of the AON 5' portion calculated from the orphan position is much longer than that of the AON 3' portion) RM106580, RM106581, RM106582 and RM106583 performing best, as shown in the figure.

[0155] Example 6. RNA editing of two adenosines in human SLC10A1 transcript in primary human hepatocytes using saponin-supported uptake.

[0156] PHH cells were cultured in a smaller format and plated in 12-well plates to approximately 200,000 cells per well, and treated with 5 μM AON in the presence of 0.5 μM AG1856 for 72 h. Cells were then harvested, and RNA was isolated as described above. Subsequently, dPCR was performed using the appropriate primers, and the percentage of editing of the A>G change at position 203 of human SLC10A1 mRNA (associated with the Q68R mutation in the protein) and the A>G change at position 770 of human SLC10A1 mRNA (associated with the E257G mutation in the protein) was determined. Figure 15A This shows the use of 25 asymmetric AONs (e.g., as shown in the figure). Figure 5 The percentage of edits (Q68R) measured for c.203A>G edits are given. The percentage of edits is as high as 45%, with RM107341, RM107346, RM107350 and RM107354 performing best. Figure 15B This shows the use of 32 symmetrical AONs (as shown in the figure). Figure 5 The percentage of edits (Q68R) measured for c.203A>G edits are given. The percentage of edits is as high as 45%, with RM107357, RM107362, RM107377, RM107381, RM107382 and RM107385 performing best. Figure 16A This shows the use of 21 asymmetric AONs (as shown in the figure). Figure 6 The percentage of edits (as shown) for c.770A>G edits (E257G) is measured. The percentage of edits is as high as 35%, with RM107275, RM107276, RM107278, RM107284, RM107291 and RM107293 performing best. Figure 16B This shows the use of 38 symmetrical AONs (as shown in the figure). Figure 6 The percentage of edits measured for c.770A>G edits (E257G) is given. The percentage of edits is as high as 15%, with RM107295 and RM107298 performing best, as shown in the figure. These experiments appear to confirm the earlier findings mentioned above that c.203A>G edits (Q68R) are more effective using symmetric AONs, while c.770A>G edits (E257G) are more effective using asymmetric AONs.

[0157] Example 7. Expression of the loss-of-function NTCP mutant on the cell membrane of U2OS cells after transfection.

[0158] To test whether the anticipated NTCP mutants outlined in this article truly result in the loss of bile transmembrane transport, human U2OS cells were cultured in 48-well plates and allowed to adhere overnight. The following day, cells were transfected with plasmids encoding wild-type human NTCP protein (as a positive control) and plasmids encoding the following seven mutant human NTCP: E257G, Q68R, I223V, K314E, Q261R, I279V, and T268A. Experiments were performed in triplicate. Following transfection, cells were treated with 1 μM or 10 μM radiolabeled taurocholic acid (TCA), a human primary bile acid, for 48 hours. After TCA treatment, the bile acid and culture medium were washed away, and the cells were washed several times. The radioactivity in each well was then measured, representing a measure of TCA uptake in these cells. Figure 17 A and Figure 17 B shows the pmol TCA uptake in different cells treated with different expression plasmids, where Figure 17 A shows the results of 1μM treatment. Figure 17Figure B shows the results of 10 μM treatment. No bile acid uptake was observed when cells were untransfected, indicating the absence of NTCP protein in U2OS cells. The “normal” levels of TCA uptake using wild-type NTCP expression plasmids were approximately 1.6% and 7.6%, respectively. These levels were also achieved with mutants I223V, K314E, I279V, and T268A. However, a significant loss of bile acid uptake was observed after expression of mutants E257G, Q68R, and Q261R, demonstrating the fundamental principle behind introducing one or more of these mutations into human NTCP transcripts to reduce the transmembrane transport capacity of bile acids in hepatocytes.

[0159] Example 8. c.203A>G editing (Q68R) of SLC10A1 transcript in primary human hepatocytes in high-throughput screening.

[0160] Subsequently, a set of 960 EONs (EON numbers RM108942 to RM109900; represented by SEQ ID NO: 167 to 1126, see...) were designed. Figure 5 ), containing various chemical and other modifications, all targeting the c.203A position (Q68R) of the human SLC10A1 transcript, as discussed in this article. These EONs were tested as follows: On day 0, PHH (5.0 × 10⁻⁶) was... 4 Three copies of EON transfection were prepared (one cell / well), and the transfection was performed simultaneously with inoculation. RNAiMAX Reagent, according to the manufacturer's instructions. Incubate the plate containing cells, culture medium, and EON at 37°C and 5% CO2 for 72 hours, changing the culture medium 24 hours after transfection / platening.

[0161] On day 3 (72 hours after transfection / plating), the supernatant was discarded, and the following analyses were performed. Cells were collected and RNA was isolated using the RNeasy 96 kit (Qiagen-74182) according to the manufacturer's instructions. Extracted RNA was treated with DNase I (ThermoFisher-EN0521) according to the manufacturer's protocol. The sample was incubated at 37°C for 30 min, followed by the addition of 1 μL of 50 mM EDTA and a further incubation at 60°C for 2 min. Total RNA was then reverse transcribed using the Maxima Reverse Transcriptase (Thermo-EP0742) kit with oligo-dT primers, random hexamer primers, and a mixture of dNTPs (10 mM each). Quantitative PCR was then performed using a Digital PCR System (Bio-Rad, QX200) in a 22 μL sample reaction mixture containing cDNA, appropriate primer pairs, and ddPCR Supermix (dUTP-free) (Bio-Rad-1863024) for probes. Using primers targeting the Q68R target (Table 1), the PCR program was as follows: 95°C for 10 minutes; 40 × 94°C for 30 seconds and 63°C for 60 seconds, 98°C for 10 minutes, then held at 4°C. The plate was then placed in a QX200 droplet reader to measure the number of positive droplets. The percentage of edits was calculated by pooling all A and G counts from three replicates for each transfection, and then scored as follows:

[0162] Score = SUM(G) / (SUM(A+G)*100)

[0163] The p-value (p = 0.05) represents the probability that 3 treatment replicates are different from 3 untreated replicates. EON edit scores are ranked from highest to lowest percentage of edits. Of the 960 EONs tested, 415 EONs scored above 0% edits. These scores are provided in Table 2 below. The remaining 545 EONs all gave 0% edits and are not listed in Table 2. Surprisingly, some EONs performed significantly well, with RM109899, ​​RM108970, RM109900, RM109571, RM108832, RM109040, RM109898, RM109306, and RM108977 performing best, all with percentages above 12%, and some exceeding 21%.

[0164] Table 2: Percentage of EONs targeting c.203A(Q68R) edited in HPP after transfection.

[0165]

[0166]

[0167]

[0168] Example 9. c.203A>G editing of SLC10A1 transcripts in PHH (Q68R) in small transfection screening.

[0169] Several additional experiments were performed using the methods outlined in Example 8, employing various additional AONs containing a variety of chemicals, including alternative linkers, transfection with liposomes (Lipofectamine) in PHH, or co-treatment with AG1856. Details of the additional AONs tested are as follows: Figure 18 The provided diagram shows various positional variations of the MP and / or PNdmi linkages, with PNms linkages (indicated by the hash symbol #), and 2'-F and 2'-MOE substitutions. Transfection, incubation, RNA isolation, cDNA generation, and dPCR were performed as described above. Figure 19 Experiments using RM107361, RM107362, RM107363, RM107364, RM107365, RM107376, RM107377, RM107378, RM107379, RM107380, RM107382 and RM107385 (with RM4777 as a negative control) clearly demonstrate the preferred chemical substances, such as RM107378 (equivalent to RM107362 except for the presence of PNms linkages instead of PNdmi and MP linkages). Figure 20 Experiments using RM108820 to RM108843 show that the absence of 2′-MOE-substituted nucleotides on the 5′ arm of AON and the low abundance of 2′-F-substituted nucleotides (RM108826 to RM108831) negatively impact editing of the Q68R NTCP target. In contrast, the use of 2′-MOE and 2′-F patterns in RM108838 to RM108843 appears to improve editing efficiency, with RM108839 showing the best performance (along with RM108821, which contains significantly more 2′-F-substituted nucleotides on the 5′ arm of AON and none of 2′-MOE-substituted nucleotides). Figure 21Additional experiments showing the use of RM108821, RM108826, RM108827, RM108836, RM108838, RM108839, and RM108840 compared to the earlier tested (and relatively poorly performing) RM107352 and RM107368AON, with co-treatment of AG1856 (as described above), clearly demonstrate that the reproducible high edit percentage obtained using RM108821 (SEQ ID NO:1284) and RM108839 (SEQ ID NO:1302) reaches a level of over 55%. Figure 22 The results show a comparison of RM108839 with several other AONs (RM117635 to RM117647 and RM117837 to RM117846) using liposome transfection, again demonstrating that the editing level obtained with RM108839 is significantly higher, and RM117635 (SEQ ID NO: 1260) also performs well. In conjunction with the teachings disclosed herein, the AON of SEQ ID NO: 1302 has been further modified to include it.

[0170] Example 10. C.203A>G editing (Q68R) of SLC10A1 transcripts in non-human primates using AON encapsulated with lipid nanoparticles.

[0171] In subsequent experiments, it was investigated whether the best-performing AON from in vitro screening could also produce c.203A editing in the endogenous wild-type SLC10A1 transcript in vivo using an endogenous ADAR enzyme, providing a model for treatment-related Q68R changes in humans. For this purpose, RM107377 (SEQ ID NO: 1173), RM107378 (SEQ ID NO: 1174), and RM107385 (SEQ ID NO: 1181) – see [link to relevant documentation] – using standard methods known to those skilled in the art. Figure 5 , Figure 15B and Figure 19 – Encapsulated in multiple lipid nanoparticles (LNPs). EON was administered at four time points at the following doses: 1 mg / kg, 2 mg / kg, 2 mg / kg, and finally 4 mg / kg. Liver biopsies were performed at different time points after administration to examine RNA editing using ddPCR, generally following the protocol described above, with a time course >1 month. Several negative controls (non-related AON and untreated subjects) were included. Plasma bile acid concentrations in treated NHPs were assessed as functional readouts.

Claims

1. An antisense oligonucleotide (AON) capable of recruiting an endogenous ADAR enzyme in human cells after forming a double-stranded complex with a region of a target RNA nucleic acid molecule in the cell, wherein the region contains a target adenosine, wherein the nucleotide in the AON opposite to the target adenosine is an orphan nucleotide, wherein the ADAR enzyme is capable of deaminosaccharifying the target adenosine to inosine after binding to the double-stranded complex, and wherein the target RNA nucleic acid molecule encodes Na+. + / Taurocholic acid cotransport polypeptide (NTCP) is a transcript molecule of the human SLC10A1 gene.

2. The AON according to claim 1, wherein the transcript molecule is a pre-mRNA or mRNA molecule.

3. The AON according to claim 1 or 2, wherein the cell is a liver cell, preferably a liver parenchymal cell.

4. The AON according to any one of claims 1 to 3, wherein the nucleotides are numbered such that the orphan nucleotide is numbered 0, the nucleotides are further positively (+) incremented towards the 5' end and negatively (-) incremented towards the 3' end, and wherein the orphan nucleotide is a deoxynucleotide containing cytosine, a cytosine analog, uracil, or isouriacil.

5. The AON according to claim 4, wherein the orphan nucleotide is a deoxynucleotide comprising a cytosine analog, wherein the cytosine analog is a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase.

6. The AON according to any one of claims 1 to 5, wherein when the nucleotide opposite to the first nucleotide at the 3' end of the orphan nucleotide in the target RNA nucleic acid molecule is cytidine, the first nucleotide at the 3' end of the orphan nucleotide is deoxyinosine.

7. The AON according to any one of claims 1 to 6, wherein the length of the AON is 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.

8. The AON according to any one of claims 1 to 7, wherein the AON comprises one or more modifications in the linker portion, each modification being independently selected from: thiophosphate (PS), phosphonoacetate, dithiophosphate, methylphosphonate (MP), sulfonylphosphonamide, (1,3-dimethylimidazoline-2-yl)phosphonamide (PNdmi), and methanesulfonylphosphonamide (PNms).

9. The AON according to claim 8, wherein the internucleotide linker numbering in the AON is such that linker number 0 is the linker at the 5' end of the orphan nucleotide, the linker positions in the oligonucleotide increase positively (+) towards the 5' end and negatively (-) towards the 3' end, and wherein linker position -2 is an MP linker or a PNms linker.

10. The AON according to claim 8 or 9, wherein the linking bond between the last two nucleotides at the 5' and / or 3' ends of the AON is a PNdmi linking bond or a PNms linking bond.

11. The AON according to any one of claims 1 to 10, wherein the AON comprises one or more nucleotides having a single or double substitution at the 2', 3' and / or 5' position of the ribose, each substitution being independently selected from: -OH; -F; substituted or unsubstituted, straight or branched lower (C1-C2) nucleotides. 10 Alkyl, alkenyl, alkynyl, alkylaryl, allyl 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-allyl; -O-alkyl; -O-alkyl; -methoxy; -aminopropoxy; -methoxyethoxy; -dimethylaminooxyethoxy; and -dimethylaminoethoxyethoxy.

12. The AON according to any one of claims 1 to 11, wherein the AON is directly or covalently or non-covalently bonded to the GalNAc portion via a connector.

13. The AON according to any one of claims 1 to 12, wherein the AON is directly or covalently or non-covalently bound to a triterpenoid glycoside, preferably AG1856, either through a linker.

14. The AON according to any one of claims 1 to 13, wherein the SLC10A1 gene is wild-type, and wherein the target adenosine is selected from: -Adenosine in the CAG codon encoding glutamine (Q) at position 68 of the NTCP protein. Furthermore, the deamination of the adenosine therein converts the amino acid into arginine (R); - The first adenosine in the CAA codon encoding glutamine (Q) at position 261 of the NTCP protein, wherein the deamination of the adenosine changes the amino acid to arginine (R). - Adenosine in the GAG ​​codon encoding glutamate (E) at position 257 of the NTCP protein. Furthermore, the deamination of the adenosine converts the amino acid into glycine (G); and - The first adenosine in the AAG codon encoding lysine (K) at position 314 of the NTCP protein, wherein deamination of the adenosine converts the amino acid to glutamic acid (E). The deamination of the target adenosine leads to impaired function of the NTCP protein in transporting bile acids from the portal circulation into cells.

15. The AON of claim 14, wherein the target adenosine is located in the CAG codon encoding glutamine at position 68 of the NTCP protein, and wherein the AON comprises a protein selected from SEQ ID NO: 150, 151, 152, 154, 156, 158, 159, 163, 164, 165, 166, 1127, 1128, 1129, 1130, 1131, 1133, 1137, 1138, 1139, 1140, 1141, 1142, 1143, 1144, 1145, 1146, 1150, 1151, 1152, 1153, 1154, 1155, 1156, 1157, 1158, 1159, 1160, 1161. The sequences and modifications of or composed of the numbers 1167, 1168, 1169, 1170, 1171, 1172, 1173, 1174, 1175, 1176, 1177, 1178, 1180, 1181, 1182, 1183, 1260, 1283, 1284, 1285, 1286, 1287, 1288, 1295, 1296, 1297, 1298, 1299, 1300, 1301, 1302, 1303, 1304, 1305, and 1306.

16. The AON of claim 14, wherein the target adenosine is located in the GAG ​​codon encoding glutamate at position 257 of the NTCP protein, and wherein the AON comprises a subset selected from SEQ ID NO: 1193, 1194, 1197, 1198, 1199, 1200, 1201, 1202, 1203, 1204, 1205, 1206, 1207, 1208, 1209, 1210, 1211, 1212, 1213, 1215, 1216, 1217, 1218, 1219, 1220, 1221, 1222, 1223. The sequence and modifications of or composed of the numbers 1224, 1225, 1226, 1227, 1228, 1229, 1230, 1231, 1232, 1233, 1234, 1235, 1236, 1237, 1238, 1239, 1240, 1241, 1242, 1243, 1244, 1245, 1246, 1247, 1253, and 1254.

17. A vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding the AON according to claim 1.

18. A nanoparticle delivery carrier formulation comprising AON according to any one of claims 1 to 16.

19. The nanoparticle delivery carrier formulation of claim 18, wherein the nanoparticle delivery carrier is a lipid nanoparticle (LNP).

20. A pharmaceutical composition comprising AON according to any one of claims 1 to 16, a carrier according to claim 17, or a nanoparticle delivery carrier formulation according to claim 18 or 19, and a pharmaceutically acceptable carrier.

21. The AON according to any one of claims 1 to 16, for treating diseases caused by cholestasis in the liver, such as cholestasis, primary sclerosing cholangitis (PSC), biliary atresia (BA), and cirrhosis.

22. Use of AON according to any one of claims 1 to 16 in the manufacture of a medicament for treating diseases caused by cholestasis in the liver (e.g., cholestasis, PSC, BA, and cirrhosis).

23. A method for editing human SLC10A1 pre-mRNA or mRNA molecules in liver cells, preferably hepatocytes, the method comprising contacting the SLC10A1 pre-mRNA or mRNA molecule with an AON capable of triggering ADAR-mediated adenosine deamination to inosine, thereby editing the SLC10A1 pre-mRNA or mRNA molecule to encode an NTCP protein with impaired, reduced, or lost bile acid uptake function, wherein the AON is the AON according to any one of claims 1 to 16.

24. A method for treating, improving, or slowing the progression of a disease (e.g., cholestasis, PSC, BA, and cirrhosis) caused by cholestasis in the liver in a subject of need, the method comprising administering to the subject an AON according to any one of claims 1 to 16, a carrier according to claim 17, or a nanoparticle delivery carrier formulation according to claim 18 or 19, thereby contacting the subject's cells with SLC10A1 pre-mRNA or mRNA molecules to achieve ADAR-mediated adenosine-to-inosine deamination, thereby editing the SLC10A1 pre-mRNA or mRNA molecules to encode an NTCP protein with impaired, reduced, or lost bile acid uptake function, thereby treating the subject.

25. An in vitro, ex vivo, or in vivo method for deamination of target adenosine in human SLC10A1 pre-mRNA or mRNA molecules in liver cells, preferably hepatocytes, the method comprising the following steps: (i) Providing the cells with AON according to any one of claims 1 to 16; (ii) to cause the cells to take up the AON; (iii) Anneal the AON with the SLC10A1 pre-mRNA or mRNA molecule; (iv) Allowing an endogenous ADAR enzyme to deaminate the target adenosine in the pre-mRNA or mRNA molecule of the SLC10A1 to inosine; and optionally (v) Use functional readouts to identify the presence of inosine in the pre-mRNA or mRNA molecule of the SLC10A1.

Images