Antisense oligonucleotides for treatment of cardiovascular diseases

Abstract

The present invention relates to the field of diseases caused by high LDL-C and / or fibrinogen levels, such as cardiovascular diseases. The invention relates to oligonucleotides for RNA (Ribonucleic Acid) editing technology. The oligonucleotides are used for deamination of target adenosine nucleotides, such as 1055th adenosine, in human B4GALT1 gene transcripts.

Description

Technical Field

[0001] The present invention relates to the medical field, particularly cardiovascular diseases (CVDs). The present invention relates to the use of RNA editing technology to target transcripts encoding β-1,4-galactosyltransferase 1 (B4GALT1), thereby causing amino acid changes that reduce B4GALT1 function. Background Art

[0002] CVDs are one of the leading causes of death and disability worldwide. Therefore, the treatment of CVDs represents a major unmet medical need. The occurrence of CVDs has multiple risk factors. These risk factors include elevated blood levels of low-density lipoprotein cholesterol (LDL-C) and / or elevated fibrinogen levels (Montasser et al. 2021. Science 374:1221–1227). Elevated LDL-C concentrations increase the likelihood of arterial plaque formation. Atherosclerosis and fibrinogen increase the risk of blood clotting and thrombosis. Elevated LDL-C is an established risk factor for coronary artery disease (CAD). Therefore, identifying mechanisms to reduce the blood levels of LDL-C and / or fibrinogen would provide potential targets for the prevention, amelioration, or treatment of CVDs.

[0003] A recently discovered target for combating CVDs is the enzyme B4GALT1, which is involved in the processing of biologically important biomolecules, including those involved in lipid metabolism and blood clotting. B4GALT1 is widely expressed and plays a key role in the processing of the N-linked oligosaccharide moiety in glycoproteins, transferring galactose from uridine diphosphate galactose (UDP-Gal) to specific glycoprotein substrates. Therefore, B4GALT1 is crucial for the biological activities associated with correctly constructed oligosaccharides.

[0004] Recent studies have found an association between the B4GALT1 mutation present in the Old Order Amish population and the reduced CVD in that population (Montasser et al. 2021, ibid.). Importantly, genetic analysis of nearly 600,000 subjects showed that this gene is associated with reduced CAD. It was found that this population has abundant missense variants in the B4GALT1 protein, called p.Asn352Ser (asparagine changed to serine; also referred to as p.N352S or simply N352S in this article), and the reduced blood concentrations of LDL-C and fibrinogen are related to this variant. The effects of this change were systematically analyzed by multiple techniques. Knock-in mouse studies provided evidence that the N352S variant is associated with reduced levels of LDL-C and fibrinogen in the blood. N-linked polysaccharide analysis of human subject serum with this mutation was related to reduced galactosylation and sialylation of apolipoprotein B100, fibrinogen, immunoglobulin G, and transferrin. Enzymatic assays showed that this mutation results in a 50% reduction in galactosyltransferase activity compared to the unmutated protein. Structural studies showed that position 352 of B4GALT1 is within the functional domain of B4GALT1, and the mutation of asparagine to serine is thought to hinder the conformational changes required for enzyme activity, thus affecting the glycosylation efficiency of B4GALT1.

[0005] Therefore, it was observed that the B4GALT1 N352S mutation affects at least two independent risk factors related to CVD, namely reducing blood LDL-C and fibrinogen concentrations, and thus may be one of the first targets with the ability to prevent CVD pleiotropically. Importantly, no association was found between the B4GALT1 N352S missense variant and any severe phenotypes.

[0006] Targeting the regulation of galactosylation has been identified as a potential therapeutic target for preventing, treating, or improving CVD while avoiding significant side effects, although no mechanism for regulating galactosylation has been proposed. The present disclosure aims to provide one or more alternative and / or improved techniques, compounds, and / or compositions for treating CVD. SUMMARY OF THE INVENTION

[0007] The present disclosure provides an RNA editing oligonucleotide (EON) that is capable of forming a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell, wherein the double-stranded complex is capable of recruiting an endogenous ADAR enzyme that naturally exists in the cell, wherein the region contains a target adenosine, wherein the nucleotide opposite the target adenosine in the EON is an orphan nucleotide, wherein the ADAR enzyme deaminates the target adenosine to inosine, and wherein the target RNA nucleic acid molecule is a transcript of the human β-1,4-galactosyltransferase 1 (B4GALT1) gene. In a preferred aspect, the B4GALT1 transcript is a pre-mRNA or an mRNA molecule. In another preferred aspect, the cell is a human liver cell, preferably a hepatocyte.

[0008] In one aspect, the target adenosine in the B4GALT1 transcript is located at a position where guanosine would encode a B4GALT1 protein variant with a reduced enzyme turnover rate. In a preferred aspect, the target adenosine is located at position c.1055A in the B4GALT1 transcript, and wherein deamination results in the change of asparagine (N; Asn; encoded by the AAU codon) at position 352 in the human wild-type B4GALT1 amino acid sequence to serine (S; Ser; encoded by the AGU codon).

[0009] The present disclosure provides an EON as disclosed herein, wherein at least one nucleotide contains one or more non-naturally occurring chemical modifications in the ribose, linkage, or base moiety, provided that the orphan nucleotide is not a cytidine containing a 2'-OMe ribose substitution. In a preferred aspect, one or more of the modifications in the linkage are independently selected from phosphorothioate (PS), phosphonoacetate, dithiophosphonate, methylphosphonate (MP), sulfonyl phosphoramidate, or PNdmi internucleoside linkages. In a preferred aspect, one or more of the modifications in the ribose moiety are single or double substitutions at the 2', 3', and / or 5' positions of the ribose, each substitution independently selected from: -OH; -F; substituted or unsubstituted, straight-chain or branched lower (C1-C 10 ) alkyl, alkenyl, alkynyl, alkaryl, 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.

[0010] The present disclosure provides an EON, wherein the EON includes an EON selected from those provided in SEQ ID NO: 3 to 42, 59 to 1069, and 1078 to 1190, preferably an EON selected from those provided in SEQ ID NO: 23, 19, 31, 27, 35, 39, 69, 70, 71, 72, 73, 93, 94, 95, 1079, 1084, 1093, 1095, 1100, 1102, 1115, 1121, 1123, 1124, and 1139 to 1190, or consisting of the same. Importantly, the present disclosure provides an EON that contains the base nucleotide sequence of these preferred EONs and is further modified according to the outline herein, which means that, using the teachings provided herein, these preferred EONs can even be further optimized to achieve a more efficient RNA editing effect. In a preferred aspect, the orphan nucleotides in the EONs disclosed herein are deoxynucleotides carrying 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase (Benner's base Z) or isouracil nucleobase (isoU).

[0011] The present disclosure also provides a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, which contains a nucleic acid molecule encoding the EON disclosed herein. The present disclosure also provides a nanoparticle delivery carrier formulation containing the EON disclosed herein. In a preferred aspect, the nanoparticle delivery carrier is a lipid nanoparticle (LNP). The present disclosure also provides a pharmaceutical composition, which contains the EON, vector, or nanoparticle delivery carrier formulation disclosed herein, and a pharmaceutically acceptable carrier.

[0012] In one aspect, the present disclosure provides the EON, vector, nanoparticle delivery carrier formulation, or pharmaceutical composition disclosed herein for treating CVD.

[0013] In one aspect, the present disclosure provides the use of the EON, vector, nanoparticle delivery carrier formulation, or pharmaceutical composition disclosed herein in the preparation of a drug for treating CVD.

[0014] In one aspect, the present disclosure provides a method for editing a B4GALT1 transcription molecule in vitro, ex vivo, or in vivo, the method comprising contacting the B4GALT1 transcription molecule or a part thereof with the EON disclosed herein, thereby forming a double-stranded complex of the EON and the B4GALT1 transcription molecule, and further allowing recruitment of ADAR1 or ADAR2 deaminase that binds to the double-stranded complex, and thereby specifically editing adenosine in the B4GALT1 transcription molecule to inosine by the deaminase.

[0015] In one aspect, the present disclosure provides a method of treating, alleviating, or improving CVD in a patient in need thereof, the method comprising contacting a B4GALT1 transcript molecule in a subject's cells with an EON disclosed herein, thereby treating the patient.

[0016] In one aspect, the present disclosure provides a method for deaminating a target adenosine in a B4GALT1 transcript molecule in a cell, the method comprising the steps of: (i) providing to the cell an EON, a vector, or a nanoparticle delivery vehicle formulation disclosed herein; (ii) allowing the cell to uptake the EON, vector, or nanoparticle delivery vehicle; (iii) annealing the EON to the B4GALT1 transcript molecule; (iv) allowing an endogenous ADAR enzyme naturally present in the cell to deaminate the target adenosine in the B4GALT1 transcript molecule to inosine; and optionally (v) identifying the presence of inosine in the target RNA molecule. In a preferred aspect, the cell is a human cell, preferably a liver cell, more preferably a hepatocyte, wherein the target adenosine is at position c.1055A in the B4GALT1 transcript, and wherein the deamination results in the change of asparagine (N; Asn) at position 352 in the human wild-type B4GALT1 amino acid sequence to serine (S; Ser). In a preferred aspect, step (v) of the method disclosed herein comprises: a) sequencing the B4GALT1 pre-mRNA or mRNA molecule or its derived cDNA; b) assessing the presence of the 352Ser B4GALT1 protein variant; or c) using a functional readout, preferably assessing the rate of decrease of UDP-Gal, or assessing the glycosylation level of transferrin in serum. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] One or more aspects will now be described by way of example only in conjunction with the accompanying drawings:

[0018] Figure 1A The nucleotide sequence of the wild-type human B4GALT1 RNA transcript from the start codon to the stop codon is shown, showing the start codon and the stop codon, wherein the target adenosine at nucleotide position 1055 (in the AAU codon encoding asparagine at amino acid position 352) is shown in bold. Figure 1B The amino acid sequence of the human wild-type B4GALT1 protein is shown, wherein the asparagine (N) at amino acid position 352 is shown in bold.

[0019] Figure 2Shown are EON sequences (5’ to 3’) designed to edit the B4GALT1 transcript, with their respective SEQ ID NOs shown below. RM4838 / EON13 is also known as B4GALT1-13. RM4830 / EON05 is also known as B4GALT1-05. Chemical modifications of the EONs are as follows: m5Ue is 2’-MOE modified 5-methyluridine (similar to 2’-MOE modified thymidine); m5Ce is 2’-MOE modified 5-methylcytidine; Ae and Ge are 2’-MOE modified adenosine and guanosine, respectively; Gm, Am, Um, and Cm are 2’-OMe modified guanosine, adenosine, uridine, and cytidine, respectively; Af, Uf, Gf, and Cf are 2’-F modified adenosine, uridine, guanosine, and cytosine, respectively; Zd (at an orphan nucleotide position) is a cytidine analogue also known as a Benner base-carrying nucleoside (which will be further outlined herein), having a deoxy moiety at the 2’ ribose position (=DNA); C2f (at an orphan nucleotide position) is 2’,2’-difluoro modified cytidine; m5Ud (or simply Ud) is a deoxynucleotide with a 5-methyluridine base; Cd (at an orphan nucleotide position) is deoxycytidine; an asterisk * indicates a PS linkage; “!” indicates a PNdmi linkage; “^” indicates an MP linkage. All other linkages are phosphodiester linkages.

[0020] Figure 3 Shown is the percentage of editing of the endogenous B4GALT1 transcript in human HepG2 cells after treatment with the indicated EONs and saponin (AG1856) at two different concentrations (1 and 5 μM). Negative controls are cells treated with AG1856 alone and untreated cells.

[0021] Figure 4 Shown is the percentage of exon 5 skipping of the endogenous B4GALT1 pre-mRNA in human HepG2 cells after treatment with the indicated EONs and saponin (AG1856) at two different concentrations (1 and 5 μM). Negative controls are cells treated with AG1856 alone and untreated cells.

[0022] Figure 5 Shown are (A) the percentage of editing of the endogenous B4GALT1 transcript in hepatospheres generated from primary human hepatocytes after treatment with the four indicated EONs; and (B) the percentage of exon 5 skipping in these same samples. Negative controls are cells treated with saponin (NT+AG) alone and untreated cells.

[0023] Figure 6Shows a set of EONs (as shown, SEQ ID NO: 62 to 108), which are designed to have a GalNAc moiety attached to the 5'-end of the oligonucleotide. The EON of SEQ ID NO: 1069 (B4GALT1-134(-)) is the same as the EON of SEQ ID NO: 72 (B4GALT1-134), but without the GalNAc moiety and the linker between GalNAc and the oligonucleotide. Chemical modifications are as Figure 2 shown. L001 is a three-branched GalNAc moiety (OP-042; Hongene Biotech). L103 is a TEG linker that connects the GalNAc moiety to the first nucleotide at the 5'-end.

[0024] Figure 7(A) shows the editing percentage of the human B4GALT1 target transcript in primary human hepatocytes (PHH) treated with 5 μM EON in the presence of 1 μM saponin (AG1856), using the EONs in Figure 6 and EON01 and EON05 (see Figure 2 ). An untreated (NT) sample was used as a negative control. (B) shows the exon 5 skipping percentage observed in the same samples.

[0025] Figure 8(A) shows the editing percentage of the human B4GALT1 target transcript in PHH treated with 5 μM EON by naked uptake, i.e., in the absence of saponin, using the EONs in Figure 6 and EON01 and EON05 (see Figure 2 ). An untreated (NT) sample was used as a negative control. (B) shows the exon 5 skipping percentage observed in the same samples.

[0026] Figure 9(A) shows the editing percentage of the B4GALT1 target transcript in primary mouse hepatocytes (PMH) treated with 5 μM EON by naked uptake, i.e., in the absence of saponin, using the EONs in Figure 6 and EON01 and EON05 (see Figure 2 ). An untreated (NT) sample was used as a negative control. (B) shows the exon 5 skipping percentage observed in the same samples.

[0027] Figure 10 (A) shows the editing percentage of the B4GALT1 target transcript in PHH treated with EON formulated in LNP. The LNP preparation containing the indicated EON was administered to the cells at the concentrations shown on the right. The untreated sample was a single sample as no LNP was administered. (B) shows the exon 5 skipping percentage observed in the same samples.

[0028] Figure 11 (A) shows the in vivo B4GALT1 target transcript editing percentages in mouse hepatocytes at days 2, 4, 7, and 30 after administration of LNP formulations (as shown in Example 8) containing RNA editing oligonucleotides EON05, EON13, or EON134. PBS administration and an actin B-targeting oligonucleotide (RM3891) formulated in LNP were generally used as negative controls for B4GALT1 editing and as a positive control for editing (for the actin B target), with an editing percentage of approximately 40% at day 4 after administration of the latter (data not shown). (B) shows the exon 5 skipping percentages observed in the same samples.

[0029] Figure 12 Shows an additional set of 960 EONs with chemical modifications as Figure 2 shown. The corresponding SEQ ID NOs are listed next to the RM numbers.

[0030] Figure 13 Shows an additional set of 22 EONs with chemical modifications as Figure 2 shown. The corresponding SEQ ID NOs are listed next to the RM numbers. "#" indicates the PNms linker bond.

[0031] Figure 14 Shows the editing percentages of the B4GALT1 transcript in PHHs after treatment with the indicated EONs. Figure 2 For B4GALT1-69 in = RM107689, B4GALT1-84 = RM107704.

[0032] Figure 15(A) shows the schematic positions of nine sequence groups (EONs) towards the exon 5-exon 6 boundary in the B4GALT1 transcript, with group 9 being the outermost in exon 5. Several EONs were designed for each group (details not shown), which had various modifications outlined herein, and the effect on exon 5 skipping was tested for each group. (B) shows the results of these exon skipping experiments, with all generated EONs having individual bar graphs, but nine independent sequences are shown below the figure. RM105550 (= B4GALT-13, with GalNAc on the 5' side but without a TEG linker) was used as a positive control. Untreated samples (NT) were used as negative controls.

[0033] Figure 16Show a set of EONs with various modifications related to 2'-F at different positions, where SEQ ID NOs: 1101 - 1120 are related to the original design of SEQ ID NO: 1100 (B4GALT1 - 13), and SEQ ID NOs: 1122 - 1138 are related to the original design of SEQ ID NO: 1121 (B4GALT1 - 21). All modifications are as Figure 2 and Figure 6 shown. There is no TEG linker between the L001GalNAc moiety and the most terminal 5'-nucleotide.

[0034] Figure 17 (A) Shows the percentage of editing (left y-axis) and the percentage of exon 5 skipping (right y-axis) in PHH treated with the indicated EONs. Untreated (NT) samples serve as negative controls, and RM106564 (EON13) serves as a reference control. (B) Shows the results of the same experiment conducted using different EON groups, where two untreated samples serve as negative controls, and RM106566 (EON21) serves as a reference control. Detailed Description

[0035] It has been realized that it is possible to target the B4GALT1 transcript to regulate the activity of the B4GALT1 enzyme, thereby preventing, ameliorating, or treating CVD. This technology is generally referred to as "RNA editing". Disclosed herein are oligonucleotides that can be used to specifically deaminate a specific target adenosine in (human) B4GALT1 transcripts (pre-mRNA and / or mRNA) in vivo, preferably using endogenous deaminases, to produce a B4GALT1 enzyme variant with reduced galactosyltransferase activity. A particularly preferred target adenosine is the adenosine in the asparagine (Asn352) codon at amino acid position 352, where deamination produces a codon encoding serine (Ser352), but the RNA editing technology disclosed herein is also applicable to other target adenosines within B4GALT1 that can be targeted to reduce the turnover rate of galactosyltransferase.

[0036] RNA editing is a natural process by which eukaryotic cells alter the sequence of their RNA molecules, typically in a site-specific and precise manner, thereby increasing the genomic-encoded RNA repertoire by several orders of magnitude. RNA editing enzymes have been described for eukaryotic species in the plant and animal kingdoms, and these processes play important roles in managing cellular homeostasis in metazoans from the simplest forms of life, such as Caenorhabditis elegans, to humans. Examples of RNA editing include the conversion of adenosine (A) to inosine (I), and the conversion of cytidine (C) to uridine (U), which are mediated by adenosine deaminases acting on RNA ( Aadenosine D eaminases a cting on R NA, ADAR), and APOBEC / AID (cytidine deaminases acting on RNA).

[0037] ADAR is a multi-domain protein that contains a catalytic domain and two to three double-stranded RNA recognition domains (depending on the enzyme in question). Each recognition domain recognizes a specific double-stranded RNA (dsRNA) sequence and / or conformation. The catalytic domain also plays a role in recognizing and binding to a partially double-stranded RNA helix, although the key function of the catalytic domain is to convert A to I at a predetermined position near the target RNA by deamination of the nucleobase. The cell's translation machinery reads inosine as guanosine, which means that if the edited adenosine is in the coding region of an mRNA or pre-mRNA, it can recode the protein sequence. The A-to-I conversion can also occur in the 5' untranslated sequence of the target mRNA, generating a new translation start site upstream of the original start site, resulting in a protein with an N-terminal extension; or in the 3' untranslated region or other non-coding regions of the transcript, which may affect RNA processing and / or stability. In addition, the A-to-I conversion can occur on splicing elements in introns or exons of pre-mRNA, thereby altering the splicing pattern. As a result, exons may be added or skipped. Enzymes that catalyze adenosine deamination belong to the ADAR enzyme family, which includes the human deaminases hADAR1 and hADAR2, as well as hADAR3. However, for hADAR3, its deaminase activity has not been confirmed.

[0038] Existing literature has described the use of adenosine deaminase to edit target RNA with oligonucleotides (e.g., Woolf et al. 1995. Proc Natl Acad Sci USA 92:8298-8302; Montiel-Gonzalez et al. 2013. Proc Natl Acad Sci USA 110(45):18285–18290; Vogel et al. 2014. Angewandte Chemie Int Ed 53:267-271). A drawback of the method described by Montiel-Gonzalez et al. (2013, ibid.) is the need for a fusion protein that is genetically engineered by fusing the boxB recognition domain of the bacteriophage λN protein with the adenosine deaminase domain of a truncated native ADAR protein. It requires transduction of the fusion protein into target cells (which is a major obstacle), or transfection of a nucleic acid construct encoding the engineered adenosine deaminase fusion protein into target cells for expression. The system described by Vogel et al. (2014, ibid.) has a similar drawback, i.e., it is not clear how to apply the system without first genetically modifying ADAR and then transfecting or transforming cells containing the target RNA, thus providing this genetically engineered protein to the cells. US 9,650,627 describes a similar system. The oligonucleotides of Woolf et al. that are 100% complementary to the target RNA sequence (1995, ibid.) are severely lacking in specificity: almost all adenosines complementary to the antisense oligonucleotides in all target RNA strands are edited.

[0039] It is known that ADAR can act on any double-stranded RNA. Through a process sometimes referred to as "promiscuous editing", this enzyme edits multiple As in double-stranded RNA. Therefore, there is a need to find methods and means that can circumvent this promiscuous editing and specifically target specific adenylates in the target RNA molecule for therapeutic applications. Vogel et al. (2014, ibid.) showed that by using 2'-O-methyl (2'-OMe)-modified nucleosides at positions opposite the adenines that should not be edited in the oligonucleotide, and unmodified nucleosides at positions directly opposite the specifically targeted adenine on the target RNA, this off-target editing can be inhibited. However, specific editing effects on target nucleotides have not been shown without using a recombinant ADAR enzyme that forms a covalent bond with EON. There are currently several publications indicating that it is feasible to recruit endogenous ADAR (thus eliminating the need for exogenous and / or recombinant sources) while maintaining the specificity where a single adenine in the target RNA molecule can be targeted and deaminated to inosine. WO2016 / 097212 discloses antisense oligonucleotides (AONs) for targeted RNA editing, where the AON is characterized by its sequence complementary to the target RNA sequence (referred to herein as the "targeting portion") and the presence of a stem-loop / hairpin structure (referred to herein as the "recruitment portion"), which is preferably not complementary to the target RNA. Such oligonucleotides are referred to as "self-loop AONs". The role of the recruitment portion is to recruit the natural ADAR enzyme present in the cell to the double-stranded RNA formed by hybridization of the target sequence with the targeting portion. Due to the presence of the recruitment portion, there is no need for a binding entity or a modified recombinant ADAR enzyme. WO2016 / 097212 describes the recruitment portion as a stem-loop structure that mimics a natural substrate (e.g., the GluB receptor) or a Z-DNA structure known to be recognized by the double-stranded RNA-binding domain or the Z-DNA-binding domain of the 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 within a single nucleic acid strand. The stem-loop structure of the recruitment portion is an intramolecular stem-loop structure formed within the AON itself and is thought to attract (endogenous) ADAR. Similar stem-loop structure-containing systems for RNA editing are described in WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560, and WO2022 / 078995.

[0040] WO2017 / 220751 and WO2018 / 041973 describe a new generation of AONs that do not contain such stem-loop structures but are (almost completely) complementary to the target region. In one embodiment, there is one or more mismatched nucleotides, wobbles, or bulges between the oligonucleotide and the target sequence. The only mismatch can be at the nucleoside site opposite the target adenosine, but in other embodiments, the AON (or RNA editing oligonucleotide, abbreviated as "EON") is described as having multiple bulges and / or wobbles when linked to the target sequence region. When the sequence of the EON is carefully selected to attract / recruit ADAR, it seems that RNA editing can be achieved in vitro, ex vivo, and in vivo using EONs lacking stem-loop structures and endogenous ADAR enzymes. An "orphan nucleoside" is defined as a nucleoside in the EON that is positioned opposite the target adenosine in the target RNA molecule and does not carry a 2'-OMe modification. The orphan nucleoside can be a deoxyribonucleoside (DNA), where the rest of the EON can still carry a 2'-O-alkyl modification (such as 2'-OMe) at the sugar entity, or the nucleotides directly surrounding the orphan nucleoside contain chemical modifications (such as modifications of DNA compared to RNA), thereby further improving RNA editing efficiency and / or enhancing resistance to nucleases. Such effects can even be further enhanced by "protecting" the EON from degradation by using sense oligonucleotides (SONs) (see WO2018 / 134301). Chemical modifications and special structures in oligonucleotides for ADAR-mediated editing of specific adenosines in target RNA have been the subject of numerous publications in the field, such as WO2019 / 111957, WO2019 / 158475, WO2020 / 165077, WO2020 / 201406, WO2020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, WO2022 / 007803, WO2022 / 018207, WO2022 / 026928, and WO2022 / 124345.For example, the use of specific sugar moieties is disclosed in WO2020 / 154342, WO2020 / 154343, WO2020 / 154344, WO2022 / 103839 and WO2022 / 103852, while the use of stereodefined linker moieties (generally for oligonucleotides, such as for exon skipping, gapmer, siRNA, or particularly for RNA editing oligonucleotides, related to a variety of target sequences) is described in WO2011 / 005761, WO2014 / 010250, WO2014 / 012081, WO2015 / 107425, WO2017 / 015575 (HTT), WO2017 / 062862, WO2017 / 160741, WO2017 / 192664, WO2017 / 192679 (DMD), WO2017 / 198775, WO2017 / 210647, WO2018 / 067973, WO2018 / 098264, WO2018 / 223056 (PNPLA3), WO2018 / 223073 (APOC3), WO2018 / 223081 (PNPLA3), WO2018 / 237194, WO2019 / 032607 (C9orf72), WO2019 / 055951, WO2019 / 075357 (SMA / ALS), WO2019 / 200185 (DM1), WO2019 / 217784 (DM1), WO2019 / 219581, WO2020 / 118246 (DM1), WO2020 / 191252, WO2020 / 196662, WO2020 / 219981 (USH2A), WO2020 / 219983 (RHO), WO2020 / 227691 (C9orf72), WO2021 / 071788 (C9orf72), WO2021 / 071858, WO2021 / 178237 (MAPT), WO2021 / 234459, WO2021 / 237223 and WO2022 / 099159. In addition to these disclosures, a large number of publications relate to targeting specific RNA target molecules or specific adenosines in such RNA target molecules, either to repair mutations that cause premature termination codons or other pathogenic mutations.Examples of such disclosures targeting adenosines 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 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 and WO2021 / 231692 (OTOF, autosomal recessive nonsyndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); WO2021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1).

[0041] Embodiment

[0042] According to one aspect, the present disclosure provides an EON that is capable of forming a double-stranded complex with a region of an endogenous human B4GALT1 transcript molecule in a cell, wherein the region or a portion of the B4GALT1 transcript molecule contains a target adenosine, wherein the nucleotide opposite the target adenosine in the EON is an orphan nucleotide, and wherein the double-stranded complex can recruit an endogenous ADAR enzyme present in the cell to deaminate the target adenosine to inosine, thereby editing the B4GALT1 transcript molecule.

[0043] According to one aspect, the present disclosure provides an EON that is capable of forming a double-stranded complex with a region of an endogenous human B4GALT1 transcript molecule in a cell, wherein the region or a portion of the B4GALT1 transcript molecule contains a target adenosine, wherein the nucleotide opposite the target adenosine in the EON is an orphan nucleotide, and wherein the double-stranded complex can recruit an endogenous ADAR enzyme present in the cell to deaminate the target adenosine to inosine, thereby editing the B4GALT1 transcript molecule, and wherein the EON causes a splicing regulatory event, preferably wherein exon 5 is skipped from the pre-mRNA, thereby removing the exon containing the target adenosine (whether edited or not) from the transcript, leaving an mRNA that cannot be translated into a functional B4GALT1 protein. The splicing regulatory event can be exon 5 skipping, can be aberrant exon skipping, can be exon 4 + exon 5 skipping, can be intron insertion, or can be (partial) skipping of exon 6.

[0044] In one aspect, the B4GALT1 transcript molecule is a pre-mRNA or mRNA molecule.

[0045] In one aspect, the cell is a human liver cell, preferably a hepatocyte.

[0046] In one aspect, the target adenosine is located at a position in the B4GALT1 transcript where guanosine results in a B4GALT1 protein variant with reduced enzyme turnover.

[0047] In one aspect, the target adenosine is located at position c.1055A in the B4GALT1 transcript.

[0048] In one aspect, the region or a part thereof of the B4GALT1 transcript molecule comprises the following sequence: 5'-...CCCAAUCCU...-3', where A is the target adenosine.

[0049] In one aspect, the EON comprises or consists of the following EONs: the EONs are independently selected from the EONs detailed in SEQ ID NOs: 3 to 42, 59 to 1069, and 1078 to 1190, preferably selected from SEQ ID NOs: 23, 19, 31, 27, 35, 39, 69, 70, 71, 72, 73, 93, 94, 95, 1079, 1084, 1093, 1095, 1100, 1102, 1115, 1121, 1123, 1124, and 1139 to 1190.

[0050] In one aspect, the EON comprises or consists of the basic nucleotide sequence provided in any one of SEQ ID NOs: 1139 to 1190, wherein:

[0051] - The EON may comprise a GalNAc moiety (such as L001 disclosed herein), wherein the GalNAc moiety may be linked to the EON via a linker (such as the L103 TEG linker disclosed herein);

[0052] - The EON may comprise nucleotides modified with sugar moieties disclosed herein, and the modification is preferably selected from LNA, 2'-MOE, 2'-OMe, 2'-F, and 2'-H (DNA);

[0053] - The EON may comprise nucleotides modified with nucleobase moieties disclosed herein;

[0054] - The EON may comprise nucleosides linked to each other via a linking bond, and the linking bond is preferably selected from PS, PNdmi, MP, and PNms;

[0055] - The EON contains orphan nucleotides, which are preferably deoxynucleotides containing a cytosine base, a uracil base, an isocytosine base, or a cytosine analogue, more preferably a Benner base;

[0056] - The EON contains a nucleotide at the -1 position, which is preferably a deoxynucleotide;

[0057] - The EON contains a nucleotide at the +1 position, which is preferably a deoxynucleotide or a 2'-MOE modified nucleotide;

[0058] - The EON contains a linking bond at the -2 position, which is preferably an MP linking bond or a PNms linking bond;

[0059] - The EON contains a linking bond at the -1 position, which is preferably a PS linking bond;

[0060] - The EON contains a 5'-AXU-3' sequence at the +1, 0, and -1 positions, where

[0061] a) A is a nucleoside with an adenine base, preferably where A is deoxyadenosine (Ad)

[0062] or 2'-MOE modified adenosine (Ae);

[0063] b) X is an orphan nucleotide, which is a deoxynucleotide containing the modifications described herein, preferably containing a Benner base (Zd) or isocytosine (8d); and

[0064] c) U is a nucleoside with a uracil base, preferably where U is deoxyuridine (Ud or m5Ud);

[0065] - The EON may contain a PNdmi linking bond connecting the terminal 5' nucleoside to its adjacent nucleoside; and / or

[0066] - The EON may contain a PNdmi linking bond connecting the terminal 3' nucleoside to its adjacent nucleoside.

[0067] SEQ ID NO: 1139 to 1151 are shown below (from 5' to 3'), where X is an orphan nucleotide, preferably a deoxynucleotide carrying a Benner base (Zd), and the sequences at the +1, 0, and -1 positions (sometimes referred to as the "central triplet") are underlined:

[0068]

[0069] SEQ ID NO: 1152 to 1164 are shown below (from 5' to 3'), where X is an orphan nucleotide, preferably deoxycytidine, and the sequences at the +1, 0, and -1 positions (sometimes referred to as the "central triplet") are underlined:

[0070]

[0071] SEQ ID NOs: 1165 to 1177 are shown below (from 5' to 3'), where X is an orphan nucleotide, preferably deoxyuridine, and the sequences at positions +1, 0, and -1 (sometimes referred to as the "central triplet") are underlined:

[0072]

[0073] SEQ ID NOs: 1178 to 1190 are shown below (from 5' to 3'), where X is an orphan nucleotide, preferably a deoxynucleotide carrying an isouridine base, and the sequences at positions +1, 0, -1 (sometimes referred to as the "central triplet") are underlined:

[0074]

[0075] In one aspect, at least one nucleotide of the EONs disclosed herein comprises one or more non-naturally occurring chemical modifications in the ribose, linkage, or base moiety, or one or more additional non-naturally occurring chemical modifications, provided that the orphan nucleotide is not a cytidine containing a 2'-OMe ribose substitution.

[0076] In one aspect, one or more additional modifications in the linkage moiety are each independently selected from PS, phosphonoacetate, phosphorothioate, MP, sulfonyl phosphoramidate, or PNdmi internucleotide linkages.

[0077] In one aspect, the EONs disclosed herein comprise one or more nucleotides containing a mono- or di-substitution at the 2', 3', and / or 5'-positions of the ribose, each substitution independently selected from: -OH; -F; substituted or unsubstituted, straight-chain or branched lower (C1-C 10 ) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, 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.

[0078] According to one aspect, a vector is disclosed herein, preferably a viral vector, more preferably an AAV vector, which comprises a nucleic acid molecule encoding the EONs disclosed herein.

[0079] According to one aspect, a nanoparticle delivery vehicle formulation comprising the EONs disclosed herein is disclosed. Preferably, the nanoparticle delivery vehicle formulation disclosed herein is an LNP formulation.

[0080] According to one aspect, a method of editing a B4GALT1 polynucleotide is disclosed herein, the method comprising contacting the B4GALT1 polynucleotide with an EON disclosed herein to effect an adenosine-to-inosine conversion of an adenosine associated with CVD mediated by an adenosine deaminase acting on RNA (ADAR), thereby editing the B4GALT1 polynucleotide. The polynucleotide is preferably a pre-mRNA or mRNA molecule.

[0081] According to one aspect, a method of treating CVD in a patient in need thereof is disclosed herein, the method comprising contacting the B4GALT1 polynucleotide in a subject's cells with an EON, a vector, a nanoparticle delivery vehicle formulation or a pharmaceutical composition disclosed herein to effect an ADAR-mediated adenosine-to-inosine conversion of an adenosine associated with CVD, thereby treating the patient.

[0082] According to one aspect, a method of treating CVD is disclosed herein, the method comprising administering to a patient in need thereof a therapeutically effective amount of an EON, a vector, a nanoparticle delivery vehicle formulation or a pharmaceutical composition disclosed herein, thereby treating CVD.

[0083] According to one aspect, a method for deaminating a target adenosine in a human B4GALT1 pre-mRNA or mRNA molecule in a cell is provided herein, the method comprising the steps of: (i) providing to the cell an EON disclosed herein; (ii) causing the cell to uptake the EON; (iii) annealing the EON to the B4GALT1 pre-mRNA or mRNA molecule; (iv) causing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to inosine; and optionally (v) identifying the presence of inosine in the target RNA molecule. In one aspect, the target adenosine is at position c.1055A of the B4GALT1 pre-mRNA or mRNA molecule. In one aspect, step (v) comprises: a) determining the sequence of the B4GALT1 pre-mRNA or mRNA molecule; b) assessing the presence of the 352Ser B4GALT1 protein variant; or c) using a functional readout, preferably assessing the rate of decrease of UDP-Gal, or assessing the level of glycosylation of transferrin in serum.

[0084] The present disclosure provides EONs that are capable of RNA editing of target adenosines in human B4GALT1 transcripts (pre-mRNA and / or mRNA) to produce B4GALT1 enzyme variants with reduced turnover rates. In one aspect, the EONs result in at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 60%, 70%, 80%, 90%, or 100% of the B4GALT1 transcript molecules encoding variant B4GALT1. The target adenosine can be located at any position where editing at that position results in a reduced turnover rate. Any reduction in turnover rate relative to the wild-type enzyme can be beneficial. In one embodiment, the turnover rate of the B4GALT1 variant is reduced by at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 60%, 70%, 80%, 90%, or 100% relative to the wild-type enzyme. In one embodiment, the reduction in turnover rate can be measured by determining the rate at which the B4GALT1 enzyme transfers galactose from UDP-Gal to GlcNAc as a receptor, as described by Montasser et al. (2021, supra). Editing the target adenosine may reduce the turnover rate of the B4GALT1 enzyme variant through different mechanisms, including, for example, amino acid residue mutations, generation of premature stop codons, or pre-mRNA splicing site variations.

[0085] It should also be noted that a reduction in turnover rate of the system as a whole (e.g., in a cellular environment) may depend on multiple factors, including the reduction in turnover rate of any particular enzyme variant and the ratio of the enzyme variant produced to the wild-type enzyme in the system, and these factors can be balanced to produce the desired effect. For example, Montasser et al. (2021, supra) described the application of a different editing technique, namely a gene editing technique that uses CRISPR-Cas9 to knockdown the B4GALT1 gene (i.e., encoding a completely inactivated B4GALT1). In this model, the gene editing was not complete, resulting in 20.7% of the transcripts being the active wild-type B4GALT1. This also led to a significant reduction in LDL-C (50%). One advantage of the RNA editing method of the present disclosure is that RNA editing is transient, allowing for simpler adjustment of the level of the edited transcript. This is particularly useful when continuous re-evaluation and adjustment of the reduction level is required (e.g., depending on disease progression / regression and / or changes in substrate levels in the system).

[0086] In one aspect, the turnover rate of the system as a whole is reduced such that the turnover rate of the system comprising the EON is reduced by at least 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% or 50% relative to the system not comprising the EON. The ratio of the enzyme variant produced in the system to the wild-type enzyme can be controlled by selecting reaction conditions (such as the selection of reactant concentration or temperature), but in a therapeutic setting, this generally depends on the EON dose provided. Those skilled in the art can accordingly adjust and optimize these factors to achieve the best reduction in the overall system turnover rate. In one embodiment, the B4GALT1 enzyme variant can be non-functional, but the EON dose can be controlled such that the system as a whole can still retain a certain level of functional wild-type B4GALT1. This is an important advantage of RNA editing over DNA editing because editing DNA to encode a non-functional B4GALT1 would result in all B4GALT1 enzyme molecules being non-functional. In one aspect, the wild-type enzyme has Figure 1B the sequence shown.

[0087] In one aspect, the EON causes deamination of the adenosine at position 1055 of human mRNA, resulting in inosine. In other words, the AAU codon encoding the 352nd amino acid asparagine is converted to an AIU codon, which is read as an AGU codon encoding serine. In another aspect, the EONs disclosed herein cause deamination of another adenosine present in the B4GALT1 transcript, which can be any adenosine, and when deaminated to inosine, it results in a reduction in the turnover rate of the B4GALT1 enzyme. In addition, there may be other mutations in the B4GALT1 gene (and transcript), which may be caused by RNA editing, thereby reducing normal B4GALT1 function. It should be understood that "causing", "triggering" or "generating" deamination does not mean that the EON itself is editing the adenosine (or has enzymatic activity). The double-stranded complex of the EON with the target RNA molecule binds to the ADAR enzyme, which acts as the deaminating entity. ADAR is the enzyme that performs deamination, and the EON is responsible for triggering deamination at the target site.

[0088] In one aspect, the EONs disclosed herein are single-stranded oligonucleotides that contain an orphan nucleotide with a position opposite to the target adenosine, wherein the orphan nucleotide is chemically modified as disclosed herein, and the rest of the oligonucleotide is also chemically modified as disclosed herein to prevent its degradation by nucleases. In one aspect, the present disclosure relates to any type of oligonucleotide or heteroduplex oligonucleotide complex that may or may not be associated with a hairpin structure (internal or terminal), may bind to ADAR or its catalytic domain, or wherein the oligonucleotide is expressed by a vector (such as AAV), or wherein the oligonucleotide is in a circular form. It should be understood that any oligonucleotide-based RNA editing that involves deamination of a nucleotide in the B4GALT1 transcript (preferably resulting in the enzyme variant p.Asn352Ser) and leads to a decrease in B4GALT1 enzyme function is encompassed. The protein mutation called p.Asn352Ser may also be referred to as N352S, and the change of adenosine at position 1055 in the B4GALT1 transcript to guanosine may also be referred to as c.1055A>G. In one aspect, the EONs disclosed herein are "naked" oligonucleotides in which the ribose, base, and / or internucleoside linkages of one or more nucleotides in the sequence contain various chemical modifications, can hybridize to the B4GALT1 transcript or a portion thereof containing the target adenosine, and can recruit endogenous ADAR for deamination of the target adenosine.

[0089] Definitions

[0090] The term "nucleoside" refers to a nucleobase linked to a (deoxy)ribosyl sugar and does not contain a phosphate group. A "nucleotide" consists of a nucleoside and one or more phosphate groups. Thus, a "nucleotide" refers to the corresponding nucleobase-(deoxy)ribosyl-phosphate linker, as well as any chemical modifications of the ribose moiety or the phosphate group. Thus, the term will include nucleotides containing a locked ribosyl moiety (containing a 2'-4' bridge, containing a methylene or any other group), unlocked nucleic acids (UNA), threose nucleic acids (TNA), and nucleotides containing linkers including phosphodiester, phosphonoacetate, phosphotriester, PS, (di)thiotriphosphate, MP, methyl thiophosphonate, phosphoramidate linkages, etc. Sometimes, the terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, inosine and hypoxanthine, etc. may be used interchangeably, referring on the one hand to the corresponding nucleobases and on the other hand to the nucleosides or nucleotides. Thymine (T) is also known as 5-methyluracil (m 5U), is a derivative of uracil (U); thymine, 5-methyluracil, and uracil may be used interchangeably throughout this disclosure. Similarly, thymidine, also known as 5-methyluridine, is a derivative of uridine; thymidine, 5-methyluridine, and uridine may be used interchangeably in this disclosure. Sometimes, unless the context clearly requires otherwise, such as when a nucleoside is linked to an adjacent nucleoside and the linkage between these nucleosides is modified, the terms nucleobase, nucleoside, and nucleotide may be used interchangeably. As used herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms "ribonucleoside" and "deoxyribonucleoside", or "ribose" and "deoxyribose" are used in the same sense as in the art.

[0091] Unless the context otherwise indicates, when referring to oligonucleotides, oligonucleotides, ONs, ASOs, oligonucleotide compositions, antisense oligonucleotides, AONs, (RNA) editing oligonucleotides, EONs, and RNA (antisense) oligonucleotides, they all refer to oligoribonucleotides and deoxyoligoribonucleotides. Oligonucleotides may be completely lacking in RNA or DNA nucleotides as they occur in nature, or they may consist entirely of modified nucleotides. When referring to "oligoribonucleotides", they may contain the bases A, G, C, U, or I. When referring to "deoxyoligoribonucleotides", they may contain the bases A, G, C, T, or I. However, the oligonucleotides disclosed herein may contain a mixture of ribonucleosides and deoxyribonucleosides. When deoxyribonucleotides are used, the 2'-position of the sugar is not modified, and the nucleotide is usually abbreviated as Ad, Cd, Gd, Id, Ud, or T, where "d" represents the deoxy nature of the nucleoside; while normal RNA or ribonucleosides with a modified 2'-position are usually abbreviated without "d" and are often abbreviated with their respective modifications as described herein.

[0092] When nucleotides are mentioned in oligonucleotides, for example, cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5-hydroxycytosine, and β-D-glucosyl-5-hydroxymethylcytosine are all included. When adenine is mentioned, N6-methyladenine, 8-oxoadenine, 2,6-diaminopurine, and 7-methyladenine are included. When uracil is mentioned, dihydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1-methylpseudouracil, 4-thiouracil, and 5-hydroxymethyluracil are included. When guanine is mentioned, 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7-trimethylguanosine, and N2,7-dimethylguanosine are included. When nucleosides or nucleotides are mentioned, furanose ribose derivatives are included, such as 2'-deoxy, 2'-hydroxy, and 2'-O-substituted variants (e.g., 2'-OMe), as well as other modifications, including 2'-4' bridged variants. When oligonucleotides are mentioned, the linkage between two single nucleotides can be a phosphodiester linkage and its modifications, including phosphonoacetate, phosphotriester, PS, (di)thiophosphonate, MP, phosphoramidate linker, phosphoguanidine, thiophosphoguanidine, sulfono phosphoramidate, etc.

[0093] The term "comprising" encompasses "including" as well as "consisting of", for example, a composition "comprising X" can consist only of X or can contain other components, such as X + Y. The term "about" associated with a numerical value x is optional and means, for example, x ± 10%.

[0094] The term "substantially" does not exclude "completely", for example, a composition "substantially free of Y" may be completely free of Y. Where applicable, the term "substantially" can be omitted from the definition.

[0095] As used herein, the term "complementary" refers to the hybridization of an EON with a second nucleic acid strand under physiological conditions (e.g., when an oligonucleotide, as the first nucleic acid strand (= guiding oligonucleotide), forms a heteroduplex RNA editing oligonucleotide complex, or HEON, with another complementary nucleic acid strand), or when it forms a duplex complex with a target RNA sequence. The term does not necessarily mean that every nucleotide in the nucleic acid strand pairs perfectly with the opposing nucleotide in the opposing sequence. In other words, although an EON may be complementary to a target sequence, there may be mismatches, wobbles, and / or bulges between the oligonucleotide and the target sequence, and under physiological conditions, the EON still hybridizes with the target sequence such that cellular RNA editing enzymes can edit the target adenosine. Thus, the term "substantially complementary" also means that, despite the presence of mismatches, wobbles, and / or bulges, the EON has sufficient matching nucleotides between the EON and the target sequence such that the EON can hybridize with the target RNA under physiological conditions. As shown herein, if an EON can hybridize with its target under physiological conditions but may also contain one or more mismatches, wobbles, and / or bulges, the EON can be complementary to the target sequence.

[0096] In terms of a nucleic acid sequence, the term "downstream" refers to a more distant position in the 3' direction along the sequence; "upstream" refers to the opposite direction. Thus, in any sequence encoding a polypeptide, the start codon is upstream of the stop codon in the sense strand but downstream of the stop codon in the antisense strand.

[0097] "Hybridization" generally refers to specific hybridization and does not include non-specific hybridization. Using techniques well known in the art, specific hybridization can be carried out under selected experimental conditions to ensure that the most stable interaction between the probe and the target occurs when the probe and the target have at least 70%, preferably at least 80%, more preferably at least 90% sequence identity.

[0098] The term "mismatch" as used herein refers to opposing nucleotides in a double-stranded RNA complex that do not form perfect base pairs according to the Watson-Crick base pairing rules. In the traditional sense, mismatched nucleotides are G-A, C-A, U-C, A-A, G-G, C-C, U-U pairs. In some aspects, the EONs disclosed herein have fewer than four mismatches with the target sequence, such as 0, 1, or 2 mismatches. "Wobble" base pairs include G-U, I-U, I-A, and I-C base pairs. Although G:G pairing is considered a mismatch, this does not necessarily mean that the interaction is unstable, which means that based on the prior art and the current disclosure, the term "mismatch" may be somewhat outdated because Hoogsteen base pairing may be considered a mismatch based on the nucleotide source but is still relatively stable. For example, an isolated G:G pairing in double-stranded RNA may be very stable but is still defined as a mismatch.

[0099] The term "splicing mutation" refers to a mutation in a gene encoding pre-mRNA, in which the splicing mechanism malfunctions, resulting in disordered splicing of introns from exons, and due to abnormal splicing, subsequent translation shows an out-of-frame phenomenon, leading to premature termination of the encoded protein. This truncated protein is usually rapidly degraded and loses any functional activity.

[0100] The EONs disclosed herein (and the complementary nucleic acid strands when two oligonucleotides form HEONs) can be almost completely chemically modified, for example, by providing ribose moieties of nucleotides with 2'-OMe substitution, 2'-F substitution, or 2'-O-methoxyethyl (2'-MOE) substitution. The orphan nucleotides in the EONs are preferably cytidine or its analogs (such as nucleotides carrying Benner bases), or uridine or its analogs (such as isouridine), and / or in one aspect, contain a diF (diF) modification at the 2'-position of the sugar, in another aspect, contain deoxyribose (2'-H, DNA), and in yet another aspect, at least one of the two adjacent nucleotides flanking the orphan nucleotide does not contain a 2'-OMe modification; in another embodiment, neither of the two adjacent nucleotides flanking the orphan nucleotide contains a 2'-OMe modification. Complete modification, that is, all nucleotides of the oligonucleotide carry 2'-OMe modifications and carry natural bases, will result in the oligonucleotide being non-functional in RNA editing (known in the art), which may be because it hinders the ADAR activity at the target position. Generally, protecting adenosine in the target RNA from editing can be achieved by providing a 2'-OMe group for the opposing nucleotide (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing guanine or adenine as the opposing base, since these two nucleobases can also reduce the editing of the opposing adenosine.

[0101] Various chemical methods and modifications are known in the field of oligonucleotides and can be easily used according to the present disclosure. The conventional internucleoside linkage between nucleotides can be altered by mono-thioation or di-thioation of the phosphodiester bond, generating PS esters or dithiophosphates, respectively. Other modifications of the internucleoside linkage are possible, including amidation and peptide linkers.

[0102] In one aspect, the EONs disclosed herein contain 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.

[0103] RNA editing entities (such as human ADAR enzymes) are known in the art to edit double-stranded RNA structures with different specificities, depending on a variety of factors. One important factor is the degree of complementarity between the two strands that make up the double-stranded RNA sequence. Perfect complementarity between the two strands generally results in the catalytic domain of human ADAR deaminating adenosine in an unbiased manner, reacting with any adenosine it encounters. Specificity of hADAR1 and hADAR2 can be increased by introducing chemical modifications and / or ensuring the presence of multiple mismatches in the double-stranded RNA, which may help to localize the double-stranded RNA binding domain in a manner that is not yet well understood. In addition, the deamination reaction itself can be enhanced by providing an oligonucleotide containing a mismatch opposite the adenosine to be edited. In accordance with the teachings of the present disclosure, one of ordinary skill in the art will be able to design the complementary portion of the oligonucleotide according to their needs.

[0104] The most well-known RNA editing proteins in cells for use with the EONs of the present disclosure are human ADAR1 and / or ADAR2. One of ordinary skill in the art will understand that the extent to which an editing entity can be redirected to other target sites within a cell can be modulated by altering the affinity of the EON for the editing enzyme recognition domain. The exact modifications can be determined by trial and error and / or computational methods based on the structural interactions between the EON and the editing enzyme recognition domain. Additionally, or alternatively, the extent to which an editing enzyme resident in the cell can be recruited and redirected can be modulated by the dose and administration regimen of the EON. This is sometimes determined by the experimenter (in vitro) or clinician, typically in Phase I and / or Phase II clinical trials.

[0105] The present disclosure provides for the modification of target RNA sequences in eukaryotic cells, preferably metazoans, more preferably mammals, and most preferably human cells. The present disclosure is particularly applicable to modifying RNA sequences in cells and tissues in which B4GALT1 is expressed and the enzyme functions. B4GALT1 is an enzyme that plays a key role in the synthesis of complex carbohydrates called glycoconjugates. While it has been shown that a reduction in B4GALT1 enzyme activity can confer a protective effect against CVD, the exact mechanism by which this occurs is not fully understood. B4GALT1 is involved in constructing oligosaccharides with specific sequences and structures, particularly those that are part of peptidoglycans. These oligosaccharides are essential for the normal function and stability of proteins. A reduction in B4GALT1 activity results in a decrease in the levels of intact oligosaccharides and downstream biological activities. One important class of oligosaccharides that B4GALT1 is involved in constructing are those terminated with sialic acid groups, where sialylation is dependent on a prior galactosylation step. Montasser et al. (2021, supra) found that a reduction in B4GALT1 activity was associated with a decrease in the galactosylation and sialylation of apolipoprotein B100, fibrinogen, immunoglobulin G, and transferrin.

[0106] Target cells can be located in vitro, ex vivo or in vivo. An advantage of the disclosed materials is that they can be used for in situ cells in a living organism as well as for cultured cells. In some aspects, the cells are processed ex vivo and then introduced into a living organism (e.g., reintroduced into the organism from which they were originally sourced). The present disclosure can also be used to edit target RNA sequences in cells from a transplant or so-called organoids (such as liver tissue organoids or "spheroids"). Organoids can be considered three-dimensional in vitro-derived tissues, but specific conditions are required to generate separate, isolated tissues. They are very useful in a therapeutic setting because they can be extracted ex vivo from patient cells and then the organoids can be reintroduced into the patient as autologous material, which is less likely to be rejected compared to a normal transplant.

[0107] Without being bound by theory, RNA editing by ADAR is thought to occur on primary transcripts in the nucleus (during transcription or splicing), or in the cytoplasm (e.g., mature mRNA, miRNA or ncRNA can be edited in these cytoplasmic compartments).

[0108] It should be clear that the targeted editing of the present disclosure can be applied to any adenosine in the B4GALT1 transcript if deamination of that adenosine results in a decrease in B4GALT1 function. However, as described herein, it is preferred to target the adenosine at position 1055 in the wild-type B4GALT1 transcript to change it from the AAU codon (encoding asparagine) to AIU (or AGU, encoding serine). Generally speaking, RNA editing can be used to create RNA sequences with different properties. These properties can be coding properties (creating proteins with different sequences or lengths, thereby resulting in changes in protein properties or functions), or binding properties (resulting in inhibition or overexpression of the RNA itself or the target or binding partner; the entire expression pathway can be altered by recoding miRNA or its homologous sequence on the target RNA). The function or localization of a protein can be arbitrarily changed through functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co-translational or post-translational modifications, catalytic sites of enzymes, binding sites of binding partners, degradation or activation signals, and so on. As described herein, EON may also cause splicing effects, such as exon skipping (e.g., skipping of exon 5), but this is not necessarily a bad thing because the resulting mRNA may transiently encode an inactive B4GALT1 protein (since the original DNA encoding the protein remains unchanged). The present disclosure encompasses these and other forms of RNA and protein "engineering", whether for the prevention, delay, or treatment of diseases, or for any other purpose in the medical or biotechnology fields, as diagnostic, preventive, therapeutic, research tools, or other uses. Therefore, any RNA editing targeting the target adenosine in the B4GALT1 transcript that can reduce the B4GALT1 turnover rate is encompassed within the scope disclosed herein.

[0109] The present disclosure relates to a new field of using gene editing techniques to treat CVD. The gene editing techniques are not particularly limited. Suitable techniques include known gene therapy techniques, including DNA editing techniques such as CRISPR / Cas, ZFN, TALEN, and meganucleases, and preferably RNA editing techniques such as ADAR-mediated editing techniques, as further outlined in detail herein.

[0110] The amount, dosage, and dosing regimen of the EON to be administered may vary depending on the cell type, the disease to be treated, the target population, the mode of administration (e.g., systemic or local), the disease severity, and the level of acceptable side effects, but these can and should be evaluated through trial and error in in vitro studies, preclinical, and clinical trials. The trials are particularly straightforward when the modified sequence results in an easily detectable phenotypic change or a change in a specific biomarker (level or activity). Higher doses of the EON may compete for binding to ADARs within the cell, thus depleting the number of entities that can freely participate in RNA editing, but routine dosage trials will reveal any such effects for a given EON and a given target.

[0111] A suitable assay technique is to deliver the EON to a cell line or test organism and then collect biopsy samples at different time points thereafter. The sequence of the target RNA can be evaluated in the biopsy samples, and the proportion of cells with the modification can be easily tracked. After performing the assay once, the information can be retained, and subsequent deliveries can be made without collecting biopsy samples. Thus, the methods disclosed herein can include the step of identifying whether the desired change is present in the cell target RNA sequence, thereby verifying that the target RNA sequence has been modified. This step typically involves sequencing the relevant portion of the target RNA or its cDNA copy (if the target RNA is pre-mRNA, the cDNA copy of its spliced product), as described above, which allows for easy verification of the sequence change. Alternatively, such a change can be evaluated through the function of the protein, e.g., by measuring the rate of decrease of UDP-Gal or by assessing the glycosylation level of transferrin in serum or any other potential marker before and after treatment, and these measurements are preferably performed in vitro on samples obtained from the treated subject.

[0112] After RNA editing occurs in the cell, the modified RNA is diluted over time, e.g., due to cell division, the limited half-life of the edited RNA, etc. Thus, in terms of actual treatment, the methods disclosed herein may involve repeated delivery of the EON until sufficient amounts of the target RNA are modified to provide a tangible benefit to the patient and / or to maintain the benefit over time.

[0113] The EONs disclosed herein are particularly suitable for therapeutic use, and thus the present disclosure also relates to a pharmaceutical composition comprising an EON disclosed herein or a vector or plasmid encoding an EON disclosed herein and a pharmaceutically acceptable carrier. In some aspects, the pharmaceutically acceptable carrier can simply be a salt solution. This can be isotonic or hypotonic and is particularly suitable for pulmonary delivery. The present disclosure also provides a delivery device (e.g., a syringe, an inhaler, a nebulizer) comprising the pharmaceutical composition disclosed herein.

[0114] The present disclosure also provides the EONs disclosed herein for treating CVD. Such treatment can be achieved by altering the target B4GALT1 RNA sequence in a mammal (such as a human liver cell, preferably a hepatocyte), as B4GALT1 is highly expressed in the liver, although B4GALT1 has been found to be ubiquitously expressed in the human body. Similarly, the present disclosure also provides the use of the EONs disclosed herein in the preparation of a medicament for altering the target B4GALT1 RNA sequence (as described herein) in a mammal (preferably a human liver cell, more preferably a hepatocyte), thereby treating, preventing or ameliorating CVD.

[0115] The present invention also provides a method for deaminating at least one specific target adenosine present in a target B4GALT1 RNA sequence in a cell, the method comprising the steps of: providing to the cell the EONs disclosed herein (naked or delivered via a vector); causing the cell to uptake the EON (or vector); annealing the EON to the target RNA molecule; causing an endogenous mammalian ADAR enzyme to deaminate the target adenosine (preferably the adenosine at position 1055 in the B4GALT1 transcript) in the target RNA molecule to inosine; and optionally identifying the presence of inosine in the RNA sequence.

[0116] The term CVD includes diseases such as CAD (sometimes also referred to as coronary heart disease), stroke and transient ischemic attack (TIA; or mini-stroke), peripheral artery disease and / or aortic disease. The EONs disclosed herein can be used to treat, prevent or ameliorate any or all of these diseases. In a preferred embodiment, the CVD for treatment according to the present disclosure is CAD.

[0117] The methods disclosed herein can be applied to subjects in which the target is adenosine. For example, when the target adenosine is at position 1055 in the B4GALT1 transcript, patients who have been identified as having or likely to have the c.1055A>G variant are generally not treated.

[0118] The present disclosure also provides a method for deaminating at least one specific target adenosine present in a target B4GALT1 RNA sequence in a cell, the method comprising the steps of: providing to the cell a vector or plasmid encoding the EONs described herein; causing the cell to uptake the vector or plasmid; annealing the EON to the target RNA molecule; causing an endogenous mammalian ADAR enzyme to deaminate the target adenosine (preferably the adenosine at position 1055 in the B4GALT1 transcript) in the target RNA molecule to inosine; and optionally identifying the presence of inosine in the RNA sequence.

[0119] In a preferred aspect, depending on the ultimate effect of the A-to-I conversion, the identification step comprises the following steps: sequencing the target RNA; sequencing the cDNA derived from the target RNA; assessing the presence of A-to-G conversion in the target RNA-derived cDNA; assessing the presence of a functional protein; assessing whether deamination has altered the splicing of the pre-mRNA; or using a functional readout, wherein the target RNA after deamination should encode an enzyme with a reduced enzyme turnover rate. Examples include assessing the rate of decrease of UDP-Gal and / or assessing biomarkers in serum and / or plasma. For example, a decrease in the enzyme turnover rate of B4GALT1 can be detected by a decrease in fibrinogen in plasma, a decrease in LDL-C in serum, or a decrease in the level of tetra-sialylated transferrin in serum and a corresponding increase in the level of lower sialylation (e.g., an increase in the level of tri-sialylated transferrin).

[0120] The best method for identifying the presence of inosine after target adenosine deamination is of course to perform dPCR or even sequencing using methods well known to those skilled in the art and as described herein. However, those skilled in the art of liver disease can apply assays to monitor certain biomarkers related to LDL-C and / or fibrinogen levels, as described above.

[0121] Montasser et al. (2021, ibid.) described in detail suitable functional assays for testing the decrease in B4GALT1 turnover rate. Briefly, one in vivo assay is the carbohydrate-deficient transferrin (CDT) assay. The CDT assay is clinically used to diagnose patients with congenital glycosylation diseases. The CDT assay assesses the sialylation level of the protein transferrin. Under normal conditions, transferrin is mainly tetra-sialylated. As B4GALT1 activity decreases, the level of tetra-sialylated transferrin decreases, and correspondingly, the level of transferrin with lower sialylation (e.g., tri-sialylated transferrin) increases. This assay can be used to determine the efficacy of EON in an in vivo environment, analyze the combined effect of the decrease in B4GALT1 turnover rate and the percentage of B4GALT1 variants relative to the wild type. Another assay is to assess the rate of decrease of UDP-Gal in a biochemical analysis. Any receptor can be used for this determination, but a convenient receptor is the monosaccharide N-acetylglucosamine (NGlcNAc). Standard assay conditions are described in Montasser et al. (2021, ibid.). This assay can be used to detect the kinetic parameters of pure B4GALT1 mutants or cell isolates after EON-mediated ADAR editing. A further assay is to knock in the B4GALT1 enzyme or variant, e.g., into a mouse model, and assess the serum and / or plasma concentration of biomarkers (e.g., LDL-C and / or fibrinogen) for B4GALT1 activity.

[0122] The EON disclosed herein is suitable for administration in the form of an aqueous solution (e.g., saline) or suspension, optionally containing additives, excipients, and other ingredients compatible with pharmaceutical use, in a concentration range of 1 ng / ml to 1 g / ml, preferably 10 ng / ml to 500 mg / ml, more preferably 100 ng / ml to 100 mg / ml. The suitable range of the dose is 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 mode of administration can be inhalation (e.g., by nebulization), intranasal, oral, injection or infusion, intravenous, subcutaneous, intradermal, intramuscular, intratracheal, intraperitoneal, rectal, intrathecal, intracisternal, parenteral, etc. The administration can be in solid form, in the form of powder, pills, gels, solutions, sustained-release preparations, or any other form suitable for human medication.

[0123] In one aspect, the method disclosed herein includes the steps of: administering to a subject the EON or pharmaceutical composition disclosed herein, such that the EON forms a double-stranded nucleic acid complex with its specifically complementary target nucleic acid molecule in the cells of the subject; involving endogenous adenosine deaminase (e.g., ADAR1 and / or ADAR2); and causing the enzyme to deaminate the target adenosine in the target nucleic acid molecule to inosine, thereby alleviating, preventing, or improving CVD.

[0124] RNA editing molecules present in cells are generally proteins in nature, such as the ADAR enzymes found in metazoans (including mammals). Preferably, the cellular editing entity is an enzyme, more preferably an adenosine deaminase or a cytidine deaminase, and even more preferably an adenosine deaminase. These are enzymes with ADAR activity. Of greatest interest are human ADAR, hADAR1 and hADAR2, including any of their isoforms. Oligonucleotide constructs disclosed herein can be conveniently designed against RNA editing enzymes known in the art, said editing enzymes including adenosine deaminases acting on RNA (ADAR), such as hADAR1 and hADAR2 in humans or human cells, and cytidine deaminases. It is known that there are two isoforms of hADAR1; a longer 150 kDa interferon-induced version and a shorter 110 kDa version, the latter being produced by alternative splicing of a common pre-mRNA. Thus, the levels of the 150 kDa isoform available in cells may be affected by interferons, particularly interferon-γ (IFN-γ). hADAR1 can also be induced by TNF-α. This provides an opportunity for the development of combination therapies, namely that IFN-γ or TNF-α and EON disclosed herein can be administered to a patient simultaneously or sequentially in any order as a combination product or as separate products. Certain disease conditions may already occur concomitantly with elevated levels of IFN-γ or TNF-α in certain tissues of a patient, creating a further opportunity to make the editing more specific to the diseased tissue. One of ordinary skill in the art will appreciate that the extent of redirection of the intracellular editing entity 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.

[0125] Chemical modification

[0126] When EON forms a so-called heteroduplex RNA editing oligonucleotide (HEON) complex with the complementary strand, all chemical modifications available for the EONs disclosed herein can also be used for the sense strand complementary to the EON, as described in PCT / EP2023 / 079290 (not published), provided that the opposite sense strand does not contain orphan nucleotides. Thus, modifications related to orphan nucleotides are only relevant to the EONs disclosed herein, while all other modifications are relevant to the EONs disclosed herein and any (protective) sense oligonucleotides that can be used with the EONs for pharmaceuticals. This includes the use of hydrophobic moieties (such as tocopherol and cholesterol) and cell-specific ligands (such as GalNAc moieties), which have also been described herein and are described in detail in PCT / EP2023 / 079290 (not published), and which can bind to the EON or its opposite strand, or both.

[0127] The internucleoside linkage in the oligonucleotides disclosed herein can comprise one or more naturally occurring internucleoside linkages and / or modified internucleoside linkages. Without limitation, at least one, at least two, or at least three internucleoside linkages from the 5' and / or 3' end of the EON are preferably modified internucleoside linkages. A preferred modified internucleoside linkage is a PS linkage. In one aspect, all internucleoside linkages of the EON are modified internucleoside linkages. In one aspect, the EON comprises a PNdmi linkage that connects the outermost nucleosides at the 5' end and / or 3' end, and the penultimate nucleosides at each of these two ends, respectively. The PNdmi linkage preferably used in the EONs disclosed herein has the structure of formula (I):

[0128]

[0129] A common limiting factor in oligonucleotide therapy is the ability of the oligonucleotide to be taken up by cells (when delivered in its native form, or in "naked" form without a delivery vehicle), its biodistribution, and its resistance to nuclease-mediated degradation. Those skilled in the art are aware, and a variety of chemical modifications have been described in detail in the art, which can help overcome these limitations. Examples of commonly used chemical modifications currently include 2'-O-methyl (commonly abbreviated as 2'-OMe or 2'-O-Me), 2'-F, and 2'-O-methoxyethyl (also commonly referred to as 2'-methoxyethoxy or 2'-MOE) modifications of sugars, and the use of PS linkages between nucleosides. WO2020 / 201406 discloses the use of MP linkages for modification at certain positions around orphan nucleotides in a first nucleic acid strand. The ribose 2'-groups of all nucleotides in the EON can independently be selected from 2'-H (i.e., DNA), 2'-OH (i.e., RNA), 2'-OMe, 2'-MOE, 2'-F, or 2'-4'-linkage (e.g., locked nucleic acid (LNA)), or other ribosyl 1'-substituted, 2'-substituted, 3'-substituted, 4'-substituted, or 5'-substituted, except that the ribose moiety of the orphan nucleotide has certain limitations in terms of compatibility with RNA editing. Orphan nucleotides in the EON that are not otherwise chemically modified at the ribose, base, or linkage preferably do not carry 2'-OMe or 2'-MOE substitutions, but can carry 2'-F, 2',2'-difluoro (diF), or 2'-ara-F (FANA) substitutions, or can be DNA. PCT / EP2023 / 069609 (not published) describes the modification of the 2'-position of the ribose moiety of orphan nucleotides by 2',2'-disubstituted substitutions (e.g., diF), which is also applicable. The 2'-4'-linkage can be selected from a variety of linkages known in the art, such as a methylene linker, an amide linker, or a constrained ethyl linker (cEt).

[0130] The present disclosure provides an EON for deaminating a target nucleotide (preferably adenosine) in a target RNA, wherein the EON is complementary to a segment of nucleotides in the target RNA that includes the target adenosine, and wherein the nucleotide in the first nucleic acid strand that is opposite to the target nucleotide is an orphan nucleotide; when the target nucleotide is adenosine, the orphan nucleotide preferably comprises a base, a modified base, or a base analog that has an NH moiety at a position similar to the position of the ring nitrogen atom (e.g., the Benner base Z). The nucleotide numbering in the EON is as follows: the orphan nucleotide is numbered 0, and the nucleotide at the 5'-end of this orphan nucleotide is numbered +1. The numbering increases further in the positive (+) direction towards the 5'-end and in the negative (-) direction towards the 3'-end, wherein the first nucleotide at the 3'-end of the orphan nucleotide is numbered -1. The internucleoside linkages in the EON are numbered as follows: the linkage numbered 0 is the linkage at the 5'-end of the orphan nucleotide, and the linkage positions in the oligonucleotide increase in the positive (+) direction towards the 5'-end and in the negative (-) direction towards the 3'-end.

[0131] Preferably, the EON comprises one or more (chirally pure or chiro-mixed) PS linkages. In one aspect, the PS linkages connect 3, 4, 5, 6, 7, or 8 nucleotides at both ends of the first nucleic acid strand. In one aspect, the EON comprises one or more phosphoramidate (PN) linkages. In one aspect, the PN linkages connect the last two nucleotides at both ends of the EON, as described herein.

[0132] The nucleosides in the EON can be natural nucleosides (deoxyribonucleosides or ribonucleosides) or unnatural nucleosides. It should be noted that for RNA editing, double-stranded RNA is typically a substrate for enzymes with deaminase activity (e.g., ADAR), and thus ribonucleosides are considered "natural", while deoxyribonucleosides may be considered unnatural or modified merely because DNA does not exist in the RNA-RNA duplex substrate configuration (for the purpose of argument). Those skilled in the art understand that when a nucleotide has a natural ribose moiety, it can still be non-naturally modified in the base and / or the linkage.

[0133] In addition to the specific preferred chemical modifications at certain positions in the compounds disclosed herein, the compounds disclosed herein may also contain one or more (additional) modifications to the nucleobases, scaffolds, and / or backbone linkages, which may or may not be present in the same monomer, e.g., at the 3' and / or 5' positions. Scaffold modifications refer to modified forms of the ribosyl moiety (i.e., the pentose moiety) that are naturally present in RNA, such as bicyclic sugars, tetrahydropyrans, hexoses, morpholinos, 2'-modified sugars, 4'-modified sugars, 5'-modified sugars, and 4'-substituted sugars. Examples of suitable modifications include, but are not limited to, 2'-O-modified RNA monomers, such as 2'-O-alkyl or 2'-O-(substituted) alkyl, e.g., 2'-OMe, 2'-O-(2-cyanoethyl), 2'-MOE, 2'-O-(2-thiomethyl)ethyl, 2'-O-butyryl, 2'-O-propargyl, 2'-O-allyl, 2'-O-(2-aminopropyl), 2'-O-(2-(dimethylamino)propyl), 2'-O-(2-amino)ethyl, 2'-O-(2-(dimethylamino)ethyl); 2'-deoxy (DNA); 2'-O-(haloalkyl)methyl, e.g., 2'-O-(2-chloroethoxy)methyl (MCEM), 2'-O-(2,2-dichloroethoxy)methyl (DCEM); 2'-O-alkoxycarbonyl, e.g., 2'-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2'-O-[2-N-methylcarbamoyl)ethyl] (MCE), 2'-O-[2-(N,N-dimethylcarbamoyl)ethyl] (DCME); 2'-halo, e.g., 2'-F, FANA; 2'-O-[2-(methylamino)-2-oxoethyl] (NMA); bicyclic or bridged nucleic acid (BNA) scaffold modifications, such as conformationally restricted nucleotide (CRN) monomers, locked nucleic acid (LNA) monomers, xylose-LNA (xylo-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) monomers, 2',4'-BNA NC (NMe) monomers, 2',4'-BNA NC(NBn) monomers, ethylene-bridged nucleic acid (ENA) monomers, carba-LNA (cLNA) monomers, 3,4-dihydro-2H-pyran nucleic acid (DpNA) monomers, 2'-C-bridged bicyclic nucleotide (CBBN) monomers, oxo-CBBN monomers, heterocyclic-bridged BNA monomers (e.g., triazolyl- or tetrazolyl-linked), amide-bridged BNA monomers (e.g., AmNA), urea-bridged BNA monomers, sulfonamide-bridged BNA monomers, bicyclic carbocyclic nucleotide monomers, TriNA monomers, α-l-TriNA monomers, bicyclic DNA (bcDNA) monomers, F-bcDNA monomers, tricyclic DNA (tcDNA) monomers, F-tcDNA monomers, α-anomeric bicyclic DNA (abcDNA) monomers, oxetane nucleotide monomers, locked PMO monomers derived from 2'-amino LNA, guanidine-bridged nucleic acid (GuNA) monomers, spiropropylene-bridged nucleic acid (scpBNA) monomers, and derivatives thereof; 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), unlocked nucleic acid (UNA); and inverted versions of any of the above monomers. All of these modifications are known to those skilled in the art.

[0134] The base sequence of the EON disclosed herein is complementary to a partial base sequence of the target B4GALT1 transcript and can thus anneal (or hybridize) with the target transcript, and the above partial base sequence contains at least the target adenosine to be deaminated to inosine (preferably adenosine at position 1055). The complementarity of the base sequences can be determined by programs such as BLAST. Those skilled in the art can easily determine the conditions (temperature, salt concentration, etc.) under which two strands can hybridize based on the interstrand complementarity.

[0135] Unlike the gapmers described previously, their relationship with RNase cleavage, and the use of such gapmers in double-stranded complexes (see, for example, EP 3954395 A1), the EONs disclosed herein do not contain a stretch of DNA nucleotide sequence that makes the target sequence (or sense nucleic acid strand) a target for RNase-mediated cleavage. In one aspect, the EONs disclosed herein do not contain four or more consecutive DNA nucleotides anywhere in their sequences. In another aspect, the EONs disclosed herein are composed of as many (chemically) modified nucleotides as possible to enhance resistance to RNase-mediated cleavage while producing an RNA editing effect as efficiently as possible. This means that the orphan nucleotides and several other nucleotides within the EONs may be DNA, but there are no stretches of four or more consecutive DNA nucleotides within the EONs. Thus, the EONs disclosed herein are not gapmers. In principle, a gapmer is a single-stranded nucleic acid composed of a central region (a DNA spacer having at least four consecutive deoxyribonucleotides) and wing regions directly at its 5'-end (5'-wing region) and 3'-end (3'-wing region). In contrast, the EONs disclosed herein can be any oligonucleotide that produces, causes, triggers, or allows the following RNA editing effect, where the target adenosine in the target RNA molecule is deaminated to inosine and thus is as resistant as possible to RNase-mediated cleavage to produce this effect.

[0136] In one aspect, the EON disclosed herein, or the sense strand that may anneal thereto prior to entry into the target cell, binds to a hydrophobic moiety, such as palmitoyl or an analogue thereof, cholesterol or an analogue thereof, tocopherol or an analogue thereof. It preferably binds to the 5' end. If the hydrophobic moieties bind to the 5' end and the 3' end, these hydrophobic moieties may be the same or different. The hydrophobic moiety bound to the oligonucleotide may bind directly or indirectly through other substances. When the hydrophobic moiety binds directly, this moiety can bind through covalent bonds, ionic bonds, hydrogen bonds, etc. When the hydrophobic moiety binds indirectly, it can bind through a linker. The linker can be cleavable or non-cleavable. A cleavable linker refers to a linker that can be cleaved under physiological conditions (e.g., in a cell or an animal body (e.g., a human body)). Cleavable linkers can be selectively cleaved by endogenous enzymes (e.g., nucleases) or by the physiological environment specific to a body or cell site (e.g., pH or reducing environment, such as glutathione concentration). Examples of cleavable linkers include, but are not limited to, one or two esters in amides, esters, phosphodiesters, phosphates, carbamates, disulfide bonds, and natural DNA linkers. Cleavable linkers also include self-destructive linkers. A non-cleavable linker refers to a linker that does not cleave under physiological conditions or cleaves at a very slow rate compared to cleavable linkers, such as PS linkages, modified or unmodified deoxyribonucleosides linked by PS linkages, spacers linked by PS bonds, and linkers composed of modified or unmodified ribonucleosides. When the linker is a nucleic acid (e.g., DNA) or an oligonucleotide, the chain length is not limited. However, its length is generally 2 to 20 bases, 3 to 10 bases, or 4 to 6 bases. The length or composition of the spacer linking the ligand and the oligonucleotide is not limited and can include, for example, ethylene glycol, TEG, HEG, alkyl chains, propyl, 6-aminohexyl, or dodecyl.

[0137] The present invention also provides a pharmaceutical composition, which comprises the EON described herein, further comprises a pharmaceutically acceptable carrier and / or other additives, and is soluble in a pharmaceutically acceptable organic solvent, etc. The dosage form of the EON or the pharmaceutical composition depends on the disease to be treated and the tissue to be targeted and can be selected according to conventional methods in the art. The pharmaceutical composition can be administered in a single dose or multiple doses. The administration mode can be daily administration or administration at appropriate time intervals, which can be determined using general knowledge in the art and can be adjusted according to the disease and the efficacy of the active ingredient.

[0138] In one aspect, the EONs described herein comprise at least one nucleotide having a sugar moiety comprising a 2'-OMe modification. In one aspect, the EONs described herein comprise at least one nucleotide having a sugar moiety comprising a 2'-MOE modification. In one aspect, the EONs described herein comprise at least one nucleotide having a sugar moiety comprising a 2'-F modification. In one aspect, an orphan nucleotide carries a 2'-H at the sugar moiety and is thus referred to as a DNA nucleotide, even though there may be other modifications to its base and / or the linkage to an adjacent nucleoside. In one aspect, an orphan nucleotide carries 2'-F at the sugar moiety. In one aspect, an orphan nucleotide carries a diF substitution at the sugar moiety. In one aspect, an orphan nucleotide carries 2'-F and 2'-C-methyl at the sugar moiety. In one aspect, an orphan nucleotide comprises 2'-F in an arabinose configuration (FANA) at the sugar moiety. In one aspect, the EON is an antisense oligonucleotide capable of forming a double-stranded nucleic acid complex with a target RNA molecule, wherein the double-stranded nucleic acid complex can recruit adenosine deaminase to deaminate a target adenosine in the target B4GALT1 RNA molecule, wherein the nucleotide opposite the target adenosine in the EON is an orphan nucleotide, and wherein the orphan nucleotide has the following formula (II):

[0139]

[0140] Wherein: X is O, NH, OCH2, CH2, Se or S; B is a nitrogenous base selected from: cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxoadenine, and 6-amino-5-nitro-3-yl-2(1H)-pyridone; R1 and R2 are each independently selected from H, OH, F or CH3; R3 is the EON moiety at the 5' end of the orphan nucleotide and consists of 7 to 30 nucleotides; R4 is the EON moiety at the 3' end of the orphan nucleotide and consists of 4 to 25 nucleotides. The nucleotides at the 3' and / or 5' ends of the orphan nucleotide can be DNA, more preferably the nucleotide at the 3' end (-1 position).

[0141] In one aspect, the EONs disclosed herein comprise at least one MP internucleoside bond according to the following formula (III):

[0142]

[0143] The preferred position of the MP linkage in the EONs disclosed herein is linkage -2, thereby linking the nucleoside at the -1 position to the nucleoside at the -2 position, although other positions of the MP linkage are not explicitly excluded.

[0144] The EONs disclosed herein may further comprise one or more linkage modifications according to the following formula (IV) structure:

[0145]

[0146] Wherein:

[0147] X = O or S; and

[0148] 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. In a preferred embodiment, X = O, R = methyl, and the linking bond modification is called methylsulfonyl phosphoramidate (MsPA) or PNms. In a preferred aspect, the PN methylsulfonyl (or PNms) linking bond is at the -2 position linking bond position in the EON disclosed herein (at this position, it replaces the PS, PO or MP linking bond), for example EON B4GALT1-65 (SEQ ID NO: 1079).

[0149] In a preferred aspect, the EON disclosed herein comprises an internucleoside linking bond of formula (IV), wherein X = O, R = CH3, and this linking bond is generally referred to as a PNms linking bond (methylsulfonyl phosphoramidate) herein. In other preferred aspects, R is equal to one of the following structures (a), (b), (c), (d), (e), (f), (g), (h) or (i):

[0150]

[0151] Also disclosed herein is an EON that, after forming a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, is capable of mediating adenosine deamination by recruiting a deaminase in the cell, wherein the region contains a target adenosine, wherein the deaminase can deaminate the target adenosine to inosine, and wherein the EON comprises a moiety having a structure according to formula (V):

[0152]

[0153] Wherein: X = O or S;

[0154] Y = O - or S - ; and

[0155] 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 20An alkyl group, a C1-C6 alkenyl group, a C1-C6 substituted alkenyl group, a C1-C6 alkynyl group, a substituted C1-C6 alkynyl group, or a conjugated group. In a preferred embodiment, X = O and R = methyl.

[0156] The EONs disclosed herein may include substitution of one of the non-bridging oxygen atoms in the phosphodiester linkage. Such modification slightly reduces the stability of base pairing but significantly enhances resistance to nuclease degradation. Preferred nucleotide analogs or equivalents include PS, phosphonoacetate, dithiophosphate, phosphotriester, aminoalkyl phosphotriester, H-phosphonate, methylphosphonate, and other alkylphosphonates (including 3'-alkylene phosphonates, 5'-alkylene phosphonates, and chiral phosphonates), phosphinates, phosphoramidates (including 3'-aminophosphoramidates and aminoalkyl phosphoramidates, thionophosphoramidates), thionoalkylphosphonates, thionoalkylphosphotriesters, selenophosphates, or boranophosphates. Particularly preferred are internucleoside linkages modified to include PS. Particularly preferred are internucleoside linkages modified to include MP. Particularly preferred are internucleoside linkages modified to include PNms. Particularly preferred are internucleoside linkages modified to include PNdmi. Conventional internucleoside linkages between nucleotides can be altered by mono-thioation or di-thioation of the phosphodiester bond to produce PS esters or dithiophosphates, respectively. Other modifications of the internucleoside linkage are possible, including amidation and peptide linkers. Those skilled in the art can determine for which target RNA nucleic acid molecule an EON contains a certain linkage modification at each linkage position described herein to produce the most effective and stable oligonucleotide compound.

[0157] In one aspect, the EONs described herein include at least one nucleotide with a sugar moiety that includes a 2'-fluoro (2'-F) modification. The preferred position of the nucleotide with the 2'-F modification is the -3 position in the EON, which may be present simultaneously with the same 2' modification in the orphan nucleotide as described above.

[0158] In one aspect, the EONs disclosed herein include at least one phosphonoacetate or phosphonoacetamide internucleoside linkage.

[0159] In one aspect, the EONs disclosed herein comprise at least one nucleotide comprising an LNA ribose modification or a UNA ribose modification. In one aspect, the EONs disclosed herein comprise at least one nucleotide comprising a TNA ribose modification.

[0160] As is known to those skilled in the art, oligonucleotides (such as the EONs outlined herein) are generally composed of repeating monomers. Such monomers are typically nucleotides or chemically modified nucleotides. The most common natural nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (U). They consist of a pentose, ribose, a 5'-linked phosphate group linked by a phosphoester, and a 1'-linked base. The sugar links the base and the phosphate and is thus often referred to as the "backbone" of the nucleotide.

[0161] Thus, modifications of the pentose are generally referred to as "backbone modifications". The original pentose can be completely replaced by another moiety that similarly links the base and the phosphate group. Thus, it can be understood that while the pentose is typically the backbone, the backbone is not necessarily a pentose. Examples of backbone modifications useful for the EON monomers disclosed herein are disclosed in WO2020 / 154342, WO2020 / 154343, and WO2020 / 154344.

[0162] In one aspect, the EONs disclosed herein can comprise one or more nucleotides with 2'-MOE ribose modifications. Additionally, in one aspect, the EONs disclosed herein comprise one or more nucleotides without 2'-MOE ribose modifications, and wherein the 2'-MOE ribose modification is located at a position that does not prevent an enzyme with adenosine deaminase activity from deaminating the target adenosine. In another aspect, the EONs disclosed herein comprise 2'-OMe ribose modifications at positions that do not contain 2'-MOE ribose modifications, and / or wherein the EONs comprise deoxynucleotides at positions that do not contain 2'-MOE ribose modifications. In one aspect, the EONs disclosed herein comprise one or more nucleotides that comprise a 2' position that contains 2'-MOE, 2'-OMe, 2'-OH, 2'-deoxy, TNA, 2'-fluoro (2'-F), 2',2'-disubstituted modifications (such as 2',2'-difluoro (diF) modifications, 2'-fluoro-2'-C-methyl modifications, or other modifications as noted, for example, by Grosse et al. (2022. ACS Med Chem Lett DOI:10.1021 / acsmedchemlett.2c00372), including 2'-spiro modifications) or a 2'-4'-linkage (i.e., bridged nucleic acids, such as LNA or examples as mentioned in, for example, WO2018 / 007475). In another aspect, for example, to increase affinity, other suitable nucleic acid monomers are arabinonucleic acid and 2'-deoxy-2'-fluoroarabinonucleic acid (FANA). The 2'-4' linkages can be selected from linkers known in the art, such as methylene linkers or constrained ethyl linkers. A variety of 2' modifications are known in the art. More examples are disclosed in more detail in, for example, WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, WO2018 / 134301, WO2019 / 219581, WO2019 / 158475, and WO2022 / 099159. In all cases, the modifications should be compatible with editing such that the EONs can function as editing trigger oligonucleotides that can form a duplex complex with the target RNA and recruit a deaminase that can then deaminate the target adenosine. When the monomer comprises a UNA ribose modification, the 2' position of the monomer can contain the same modifications as described above, such as 2'-MOE, 2'-OMe, 2'-OH, 2'-deoxy, 2'-F, 2',2'-diF, 2'-fluoro-2'-C-methyl, arabinonucleic acid, FANA, or a 2'-4'-linkage (i.e., bridged nucleic acid, such as LNA).

[0163] Bases, sometimes referred to as nucleobases, are generally adenine, cytosine, guanine, thymine, or uracil, or derivatives thereof. A base, sometimes referred to as a nucleobase, is a moiety capable of binding to another nucleobase through hydrogen bonding, polar bonds (such as through a CF moiety), or aromatic electronic interactions. Cytosine, thymine, and uracil are pyrimidine bases and are generally linked to the backbone through their 1-nitrogen. Adenine and guanine are purine bases and are generally linked to the backbone through their 9-nitrogen. The terms "adenine", "guanine", "cytosine", "thymine", "uracil", and "hypoxanthine" as used herein refer to the nucleobases themselves. The terms "adenosine", "guanosine", "cytidine", "thymidine", "uridine", and "inosine" refer to nucleobases linked to a (deoxy)ribosyl group.

[0164] The nucleobases in the EONs disclosed herein can be adenine, cytosine, guanine, thymine, inosine, or uracil, or any other moiety capable of interacting with another nucleobase through hydrogen bonding, polar bonds (such as CF), or aromatic electronic interactions. The nucleobase at any position in a nucleic acid strand can be a modified form of adenine, cytosine, guanine, or uracil, such as hypoxanthine (the nucleobase in inosine), pseudouracil, pseudocytosine, isouridine, N3-glycosylated uracil, 1-methylpseudouracil, orotic acid, agmatidine, lysine, 2-thiouracil, 2-thiothymine, 5-substituted pyrimidines (e.g., 5-halouracil, 5-halomethyluracil, 5-trifluoromethyluracil, 5-propynyluracil, 5-propynylcytosine, 5-aminomethyluracil, 5-hydroxymethyluracil, 5-formyluracil, 5-aminomethylcytosine, 5-formylcytosine), 5-hydroxymethylcytosine, 7-deazaguanine, 7-deazaadenine, 7-deaza-2,6-diaminopurine, 8-aza-7-deazaguanine, 8-aza-7-deazaadenine, 8-aza-7-deaza-2,6-diaminopurine, 8-oxoadenine, 3-deazapurines (e.g., 3-deazoadenosine), pseudoisocytosine, N4-ethylcytosine, N2-cyclopentylguanine, N2-cyclopentyl-2-aminopurine, N2-propyl-2-aminopurine, 2,6-diaminopurine, 2-aminopurine, G-clamp (G-clamp) and its derivatives, Super A, Super T, Super G, amino-modified nucleobases or their derivatives; and degenerate bases or universal bases such as 2,6-difluorotoluene, or a missing abasic site (e.g., 1-deoxyribose, 1,2-dideoxyribose, 1-deoxy-2-O-methylribose, and azaribose).

[0165] In one aspect, a nucleotide analogue is an analogue of a nucleic acid nucleotide. In one aspect, the nucleotide analogue is an analogue of adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine or deoxyuridine. In one aspect, the nucleotide analogue is not guanosine or deoxyguanosine. In one aspect, the nucleotide analogue is not a nucleic acid nucleotide. In one aspect, the nucleotide analogue is not adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine or deoxyuridine.

[0166] Nucleotides are typically linked to adjacent nucleotides by condensation of their 5'-phosphate moiety with the 3'-hydroxyl moiety of an adjacent nucleotide monomer. Similarly, its 3'-hydroxyl moiety is typically linked to the 5'-phosphate of an adjacent nucleotide monomer. This forms a phosphodiester bond. The phosphodiester and the backbone form an alternating copolymer. The bases are grafted onto this copolymer, i.e., onto the backbone moiety. Due to this property, the alternating copolymer formed by the oligonucleotide linking backbone is commonly referred to as the "backbone" of the oligonucleotide. Since the phosphodiester bond joins adjacent monomers together, they are commonly referred to as "backbone linkage bonds". It will be understood that when the phosphate group is modified to a similar group (e.g., PS), the group is still referred to as the backbone linkage bond of the monomer. This is called "backbone linkage bond modification". Generally speaking, the backbone of an oligonucleotide consists of alternating backbone and backbone linkage bonds.

[0167] The EONs disclosed herein can include linkage modifications. The linkage modifications can be, but are not limited to, modified forms of the phosphodiesters present in RNA, such as PS, chirally pure PS, (R)-PS, (S)-PS, MP, chirally pure MP, (R)-methylphosphonate, (S)-methylphosphonate, phosphoramidoguanidine (e.g., PNdmi), chirally pure phosphoramidoguanidine, (R)-phosphoramidoguanidine, (S)-phosphoramidoguanidine, dithiophosphate (PS2), phosphonacetate (PACE), phosphonoacetamide (PACA), thiophosphonacetate, thiophosphonoacetamide, methyl phosphorohioate, methylthiophosphonate, PS prodrug, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, borane phosphate, borane PS, methylborane phosphate, methylborane PS, methylborane phosphonate, methylborane thiophosphate, phosphate, phosphotriester, aminoalkyl phosphotriester, and their derivatives. Other modifications include phosphoramidite, phosphoramidate, N3’→P5’ phosphoramidate, phosphordiamide, thiophosphordiamide, sulfamate, diethylsulfoxide, amide, sulfonate, siloxane, sulfide, sulfone, formacetyl, alkenyl, methylenehydrazino, sulfonamide, triazole, oxalyl, carbamate, methyleneimino (MMI) and thioacetamide nucleic acid (TANA); and their derivatives. In addition, various salts, mixed salts and free acid forms, as well as 3’→3’ and 2’→5’ linkages are also included.

[0168] In one aspect, the EONs disclosed herein involve substitution of a non-bridging oxygen in a phosphodiester linkage. Such modification slightly disrupts the stability of base pairing but significantly enhances resistance to nuclease degradation. Preferred nucleotide analogs or equivalents include PS, phosphonoacetate, phosphorodithioate, phosphotriester, aminoalkyl phosphotriester, H-phosphonate, methylphosphonate, and other alkylphosphonates (including 3'-alkylphosphonate, 5'-alkylphosphonate, and chiral phosphonates), phosphinate, phosphoramidate (including 3'-aminophosphoramidate and aminoalkylphosphoramidate, thionophosphoramidate), thionoalkylphosphonate, thionoalkylphosphotriester, selenophosphate, or boranophosphate. Particularly preferred are internucleoside linkages modified to incorporate PS. Many such non-naturally occurring linkage modifications (e.g., PS) are chiral, meaning there are Rp and Sp configurations known to those skilled in the art. In one aspect, the chirality of the PS linkage is controllable, meaning each linkage is either in the Rp configuration or the Sp configuration, whichever is preferred. Selection of the Rp or Sp configuration at a particular linkage position may depend on the target sequence, the EON sequence, and the binding and induction efficiency for providing RNA editing. However, if not particularly required, the composition may contain an EON as the active compound, which has both Rp and Sp configurations at a particular linkage position. Mixtures of such EONs are also feasible, where certain positions preferably have one of the configurations, and it does not matter for other positions.

[0169] Similarly, in all cases, the modification should be compatible with editing so that the EON can function as an oligonucleotide that produces editing, which, when linked to the target sequence, can recruit adenosine deaminase due to the resulting double-stranded RNA property. In all aspects, the enzyme having adenosine deaminase activity is preferably ADAR1 or ADAR2. In a highly preferred aspect, the EON is an RNA editing oligonucleotide targeting pre-mRNA or mRNA, where the target nucleotide is adenosine in the target RNA, and the adenosine is deaminated to inosine, which is read as guanosine by the translation machinery. The present disclosure also provides a pharmaceutical composition comprising the EONs described herein and a pharmaceutically acceptable carrier.

[0170] Other chemical modifications of the EONs disclosed herein include substitution of one or more hydrogen atoms with deuterium or tritium, examples of which can be found in, for example, WO2014 / 022566 or WO2015 / 011694.

[0171] The present disclosure also provides the EONs disclosed herein or pharmaceutical compositions comprising the EONs disclosed herein for treating, preventing or ameliorating CVD, such as CAD. In one aspect, the present disclosure provides EONs or pharmaceutical compositions comprising the EONs disclosed herein for treating, preventing or ameliorating diseases in which B4GALT1 functions in a wild-type manner. In one aspect, the present disclosure provides EONs or pharmaceutical compositions comprising the EONs disclosed herein for treating, preventing or ameliorating diseases associated with high LDL-C and / or fibrinogen levels. In one aspect, the present disclosure provides EONs or pharmaceutical compositions comprising the EONs disclosed herein for treating or preventing CVD, such as CAD.

[0172] The EONs disclosed herein preferably do not contain a 5'-terminal O6-benzylguanosine or 5'-terminal amino modification and are preferably not covalently linked to an SNAP-tag domain (an engineered O6-alkylguanine-DNA-alkyltransferase). The EONs disclosed herein preferably do not contain a boxB RNA hairpin sequence. In one aspect, the EONs disclosed herein have 0, 1, 2 or 3 wobble base pairs with a target sequence and / or 0, 1, 2, 3, 4, 5, 6, 7 or 8 mismatched base pairs with a target RNA sequence. When the orphan nucleotide is uridine, no mismatches may be present. An alternative to uridine is to place isouridine opposite the target adenosine, and isouridine does not pair like U with A and is thus considered a mismatch. Ideally, the target adenosine in the target sequence forms a mismatched base pair with the nucleoside in the EON opposite the target adenosine.

[0173] It should be noted that when the EONs are delivered by a vector (such as an AAV vector), there are no chemical modifications in the EONs acting on the target RNA molecule. Although "naked" EONs with the chemical modifications described herein are preferably used, EONs delivered by other means, such as expression by an AAV vector, or editing of circular or hairpin-structured molecules (recruiting moieties, such as those disclosed in WO2016 / 097212, WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560 and WO2022 / 078995) are also included in the present disclosure, as these can also be used to edit the adenosine in the target B4GALT1 RNA molecule to generate a B4GALT1 protein with reduced function.

[0174] It is noted that when the EON contains the chemical modifications detailed herein, it can still be delivered via a delivery vehicle. Suitable delivery vehicles are nanoparticle delivery vehicles such as polymeric nanoparticles, dendrimers, inorganic nanoparticles and nanocrystals, organic nanocrystals, and liposomes. Preferred nanoparticles are lipid nanoparticles (LNPs), which are nanosized lipid vesicles that carry the EONs of the present disclosure and facilitate delivery to target cells. If LNPs or any other similar type of carrier is used, the EON is still considered naked because it is not transcribed from an encoding polynucleotide (e.g., in the case of a plasmid or vector, the EON is not considered "naked"). Thus, even if the chemically modified EON is encapsulated by a carrier (preferably an LNP), it is still considered naked because it is manufactured in a laboratory environment and then encapsulated in the carrier using methods known to those of skill in the art. The present disclosure also relates to a delivery vehicle, preferably an LNP, that contains the 'naked' and chemically modified EONs as disclosed herein, even more preferably as disclosed in any one of SEQ ID NO: 3 to 42, 59 to 1069, and 1078 to 1190, preferably selected from SEQ ID NO: 23, 19, 31, 27, 35, 39, 69, 70, 71, 72, 73, 93, 94, 95, 1079, 1084, 1093, 1095, 1100, 1102, 1115, 1121, 1123, 1124, and 1139 to 1190. Those of skill in the art understand that when a delivery moiety or attachment to the EON (such as a GalNAc moiety to target hepatocytes in the liver) is used, the EON is still considered naked even if the GalNAc-(linker)-EON is encapsulated in a delivery vehicle such as an LNP.

[0175] The EONs disclosed herein can utilize endogenous cellular pathways and naturally occurring ADAR enzymes (= endogenous) to specifically edit target adenosines in target RNA sequences. The EONs disclosed herein can recruit endogenous ADAR and complex with it when forming a double-stranded complex with the target RNA molecule, and then facilitate the deamination of a (single) specific target adenosine nucleotide in the target RNA sequence. Ideally, only one adenosine is deaminated. When the EONs disclosed herein are complexed with ADAR, preferably a single target adenosine is deaminated.

[0176] Analysis of the natural targets of ADAR enzymes has shown that these targets typically contain mismatches between the two strands that form the RNA helices edited by ADAR1 or ADAR2. It has been shown that these mismatches enhance the specificity of the editing reaction (Stefl et al. 2006. Structure 14(2):345-355; Tian et al. 2011. Nucleic Acid Res 39(13):5669-5681). Characterization of the optimal pattern of paired / mismatched nucleotides between EON and the target RNA is also crucial for the development of effective ADAR-based EON therapies.

[0177] As described above, the EONs disclosed herein utilize specific nucleotide modifications at pre-determined positions to ensure stability as well as proper binding and activity of ADAR. These modifications can vary and may include modifications in the EON backbone, nucleotide sugar moieties, and nucleobases or phosphodiester linkages, as described herein. These modifications may also be distributed in a variable manner throughout the EON sequence. Specific modifications may be required to support interactions between different amino acid residues in the RNA-binding domain and deaminase domain of the ADAR enzyme. For example, PS linkages or 2'-OMe or 2'-MOE modifications between nucleotides may be allowed in certain portions of the EON, while these modifications should be avoided in other portions to prevent disruption of critical interactions of the enzyme with phosphate groups and 2'-OH groups. Specific nucleotide modifications may also be required to enhance the editing activity of the substrate RNA when the target sequence is not an optimal sequence for ADAR editing. Previous studies have demonstrated that certain sequence contexts are more amenable to editing. For example, the target sequence 5'-UAG-3' (with target A in the middle) contains the most preferred nearest neighbor nucleotides for ADAR2, while the 5'-CAA-3' target sequence is less preferred (Schneider et al., 2014. Nucleic Acids Res 42(10):e87). Structural analysis of the ADAR2 deaminase domain suggests that it is possible to enhance editing by carefully selecting the nucleotide opposite the target trinucleotide. For example, the 5'-CAA-3' target sequence paired with the 3'-GCU-5' sequence on the opposite strand (forming an A-C mismatch in the middle) is unfavorable due to steric clashes between the guanosine base and amino acid side chains of ADAR2. Although other adenosines in the B4GALT1 transcript may also be targeted to attenuate protein function, the adenosine at position 1055 is preferably deaminated. The present disclosure also provides RNA editing oligonucleotides (collectively referred to herein as EONs) that can deaminate adenosines in the B4GALT1 transcript, resulting in a B4GALT1 enzyme with reduced turnover. This means that it is not strictly limited to deamination of the adenosine at position 1055, and other (one or more) adenosines may also be targeted, which may also lead to a decrease in B4GALT1 enzyme function. Other adenosines that are equally important for B4GALT1 function can be identified by methods such as population gene screening or in silico simulations, and these adenosines can also be targeted by RNA editing and follow the teachings of the present disclosure. All RNA events and oligonucleotides that can be used for such targeting are encompassed by the present disclosure, regardless of the specific nucleic acid molecule or EON.

[0178] Mutagenesis studies of human ADAR2 have shown that a single mutation from glutamate to glutamine at residue 488 (E488Q) increases the rate constant of the deamination reaction by 60-fold compared to the wild-type enzyme (Kuttan & Bass. 2012. Proc Natl Acad Sci USA. 109(48):3295-3304). During the deamination reaction, ADAR flips the edited base out of its RNA duplex and into the enzyme active site (Matthews et al. 2016. Nat Struct Mol Biol. 23(5):426-433). When ADAR2 edits adenosine in a preferred context (A:C mismatch), the nucleotide opposite the target adenosine is referred to as the orphan cytidine. The crystal structure of ADAR2 E488Q bound to double-stranded RNA shows that the glutamine (Gln) side chain at position 488 can provide a hydrogen bond to the N3 position of the orphan cytidine, thereby increasing the catalytic rate of ADAR2 E488Q. In the wild-type enzyme, position 488 is glutamate (Glu) rather than glutamine (Gln), so the amide group of glutamine is absent and instead there is a carboxylic acid. To achieve the same contact for the orphan cytidine in the wild-type situation, protonation is required. To utilize endogenous expressed ADAR2 to correct disease-related mutations, the editing efficiency of wild-type ADAR2 enzyme in cells must be maximized. WO2020 / 252376 discloses the use of EONs with modified RNA bases, especially at the position of the orphan cytidine, to mimic the hydrogen bond pattern observed in the E488Q ADAR2 mutant. By replacing the nucleotide opposite the target adenosine in the EON with a cytidine analogue (as an N3 hydrogen bond donor), it is envisaged that the same contact can be stabilized, thereby increasing the catalytic rate of the mutant enzyme. Two cytidine analogues are of particular interest: pseudoisocytidine (also known as "piC"; Lu et al., 2009. J Org Chem. 74(21):8021-8030; Burchenal et al., 1976. Cancer Res 36:1520-1523) and Benner's base Z (also known as "Zd"; Yang et al., 2006, Nucl Acid Res. 34(21):6095-6101), which were initially selected because they provide hydrogen bond donation at N3 while having minimal impact on the shape of the nucleobase. Benner's base is also known as 6-amino-5-nitro-3-yl-2(1H)-pyridone. In addition to modifications of the ribose 2'-group, cytidine analogues may also be present in the EON. The ribose 2'-group in the EON can independently be selected from 2'-H (i.e., DNA), 2'-OH (i.e., RNA), 2'-OMe, 2'-MOE, 2'-F or 2'-4'-linkage (i.e., bridged nucleic acid, such as LNA) or other 2'-substitutions.The 2'-4' linkage can be selected from linkers known in the art, such as a methylene linker or a constrained ethyl linker.

[0179] In one aspect, the EON comprises one or more sugar moieties mono- or di-substituted at the 2', 3' and / or 5' positions, such as: -OH; -H; -F; substituted or unsubstituted, straight-chain or branched lower (C1-C 10 ) alkyl, alkenyl, alkynyl, alkaryl, 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.

[0180] In one aspect, the nucleotide analogs or equivalents within the EON comprise one or more base modifications or substitutions. Modified bases include synthetic and natural bases such as inosine, xanthine, hypoxanthine, and other -aza, deaza, -hydroxy, -halo, -thio, thiol, -alkyl, -alkenyl, -alkynyl, thioalkyl derivatives of pyrimidine and purine bases known or to be known in the art. Purine nucleobases and / or pyrimidine nucleobases can be modified to alter their properties, such as by amination or deamination of the heterocycle. Specific chemical properties and forms may vary depending on the oligonucleotide construct and application and can be designed according to the wishes and preferences of those skilled in the art.

[0181] The EONs disclosed herein are generally longer than 10 nucleotides, preferably longer than 11, 12, 13, 14, 15, 16 nucleotides, more preferably longer than 17 nucleotides. In one aspect, the EONs disclosed herein are longer than 20 nucleotides. The EONs disclosed herein are preferably shorter than 100 nucleotides, more preferably shorter than 60 nucleotides, more preferably shorter than 50 nucleotides. In a preferred aspect, the EONs disclosed herein comprise 18 to 70 nucleotides, more preferably comprise 18 to 60 nucleotides, even more preferably comprise 18 to 50 nucleotides. Thus, in a particularly preferred aspect, the EONs disclosed herein comprise 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 or 50 nucleotides. In one embodiment, the length of the EON is 27, 28, 29 or 30 nucleotides.

[0182] In one aspect, at either or both ends of the EONs disclosed herein, reverse deoxy T or dideoxy T nucleotides are incorporated.

[0183] Example

[0184] Example 1. Editing of adenosine in the human B4GALT1 target RNA molecule using an in vitro biochemical editing assay.

[0185] First, a set of initial B4GALT1-targeting EONs (as Figure 2 shown) were tested for editing of the human B4GALT1 target (precursor) mRNA in an in vitro biochemical editing assay. To obtain the B4GALT1 target RNA, the B4GALT1 G region (IDT) containing the T7 promoter sequence and the (partial) HFE sequence was used as a template, and PCR was performed with the forward primer 5’-CTC GAC GCA AGC CAT AAC AC-3’ (SEQ ID NO:43) and the reverse primer 5’-TGG ACC GAC TGG AAA CGT AG-3’ (SEQ ID NO:44). The 5’ to 3’ G region sequence (SEQ ID NO:45) is as follows, where the target adenosine is underlined and bolded, and the primer sequences are underlined:

[0186]

[0187] The PCR product was then used as a template for in vitro transcription. The reaction used the MEGAscript T7 Transcription Kit. The RNA was purified on a urea gel and then extracted in a buffer of 50 mM Tris-Cl pH 7.4, 10 mM EDTA, 0.1% SDS, and 0.3 M NaCl, and subsequently purified with phenol-chloroform. The purified RNA was used as the target for the biochemical editing experiment.

[0188] First, EON RM4439-RM4454 and RM4826-RM4849 were annealed to the B4GALT1 target RNA, respectively. The annealing process was carried out in a buffer (5 mM Tris-Cl pH 7.4, 0.5 mM EDTA, and 10 mM NaCl), and the ratio of target RNA to oligonucleotide was 1:3 (600 nM oligonucleotide and 200 nM target). The samples were heated at 95 °C for 3 minutes and then slowly cooled to room temperature. Next, the editing reaction was carried out. The annealed oligonucleotide / target RNA was combined with a protease inhibitor (cOmplete TM, Mini, EDTA-free Protease I, Sigma-Aldrich), RNase inhibitor (RNasin, Promega), poly A (Qiagen), tRNA (Invitrogen), and editing reaction buffer (15 mM Tris-Cl pH 7.4, 1.5 mM EDTA, 3% glycerol, 60 mM KCl, 0.003% NP-40, 3 mM MgCl2, and 0.5 mM DTT) were mixed to a final concentration of 6 nM oligonucleotide and 2 nM target RNA. At the start of the reaction, purified ADAR2 (GenScript) was added to the mixture at a final concentration of 6 nM and incubated at 37 °C for a predetermined time point. Each reaction was terminated by adding 95 μl of 3 mM EDTA solution at 95 °C. Then, 6 μl of the terminated reaction mixture was taken as a template, and cDNA synthesis was performed using the Maxima reverse transcriptase kit (Thermo Fisher) and random hexamer primers (Thermo Fisher Scientific). In the presence of primers and dNTPs, the RNA was initially denatured at 95 °C for 5 minutes and then slowly cooled to 10 °C, after which first-strand synthesis was carried out at an extension temperature of 62 °C in a total volume of 20 μl according to the manufacturer's instructions. Using the Amplitaq gold 360 DNA polymerase kit (Applied Biosystems), pyrosequencing analysis of the PCR amplification products was performed with 1 μl of cDNA as a template according to the manufacturer's instructions. Then, the following thermal cycling protocol was used for PCR: initial denaturation: 95 °C, 5 minutes; then 40 cycles: 95 °C, 30 seconds; 58 °C, 30 seconds; 72 °C, 30 seconds; and finally an extension at 72 °C for 7 minutes.

[0189] Since inosine pairs with cytidine during cDNA synthesis in the reverse transcription reaction, the nucleotide incorporated at the edited position during PCR will be guanosine. The percentage of guanosine (edited) to adenosine (unedited) is determined by pyrosequencing. The PCR products were subjected to pyrosequencing and data analysis using a PyroMark Q48 Autoprep instrument (QIAGEN) according to the manufacturer's instructions, with 10 μl of PCR product and 4 μM sequencing primer input: the analysis performed by the instrument provides the results of the selected nucleotide, i.e., the percentage of adenosine and guanosine detected at that position. Thus, the degree of A-to-I editing at the selected position is measured by the percentage of guanosine at that position.

[0190] Example 2. Editing of the human B4GALT1 transcript in HepG2 cells.

[0191] Study the editing of endogenous human B4GALT1 transcripts in human cells. To this end, human HepG2 hepatocellular carcinoma cells were cultured in EMEM + 10% FBS + 1% P / S. The cells were stored in an environment of 37 °C and 5% CO2. In the initial experiment, the editing efficiency of EONs named RM4826 to RM4849 (see Figure 2 ) was tested. A total of 0.75×10 5 HepG2 cells were seeded and treated with 1 or 5 μM EON, and 1 μM AG1856 (a saponin, also known as triterpenoid glycoside; see WO2021 / 122998) was added to each well. The mixture was stored in the cells for 72 hours. Then the cells were collected, and total RNA was isolated from the transfected cells using the ReliaPrep TM RNA Miniprep Kit. After removing the medium, the cells were washed once with PBS. After completely sucking dry the PBS, 100 μL of lysis buffer was added to the wells to lyse the cells and collect the intracellular substances. After adding 35 μL of isopropanol, the mixture was loaded onto the column, subjected to multiple washing steps, and treated with DNase I. After eluting with 15 μL of DNase / RNase-free water in total volume, the RNA yield was determined using spectrophotometric analysis (NanoDrop) and stored at -80 °C.

[0192] Complementary DNA (cDNA) was generated using Maxima reverse transcriptase (RT, Thermo Fisher). Generally, 500 ng of total RNA was used in a reaction mixture containing 4 μL of 5xRT buffer, 2 μL of dNTP mixture (10 mM each), 0.5 μL of Oligo(dT), 0.5 μL of random hexamer, and 0.5 μL of Maxima reverse transcriptase (all products of Thermo Fisher), and DNase- and RNase-free water was added to a total volume of 20 μL. The samples were loaded into a T100 thermal cycler (Bio-Rad), initially incubated at 25 °C for 10 minutes; subsequently, a cDNA reaction was carried out at 50 °C for 30 minutes, and the reaction was terminated at 85 °C for 5 minutes. The samples were cooled to 4 °C and then stored at -20 °C.

[0193] To determine the editing efficiency, the cDNA samples were subjected to two multiplex digital PCR (dPCR) assays. Before performing the dPCR measurements, the HepG2 cDNA samples were diluted 5-fold. The first dPCR was designed to distinguish cDNA species containing either the original adenosine or the edited inosine (converted to guanosine during cDNA synthesis). The first dPCR also used primer / probe sets targeting exons 1 and 2 to quantify the amount of B4GALT1-specific cDNA molecules in the mixture. The second dPCR was designed to measure exon 5 skipping of B4GALT1 and the housekeeping gene HPRT1. The primer and probe sequences are listed in Table 1.

[0194] Table 1. Primer and probe names and sequences (+ indicates LNA nucleotides on the 3’ side)

[0195]

[0196] Digital PCR was performed using the QIAcuity 4,5-plex, QIAcuity PCR kit, and 96-well 8.5K nanoplate (Qiagen). A total of 1.2 μL of the diluted cDNA mixture was used in the dPCR mix, which contained 3 μL of 4xQIAcuity Mastermix, 0.6 μL of each primer (10 μM stock concentration), and 0.3 μL of each probe (10 μM stock concentration), and nuclease-free and RNase-free water was added to a total volume of 12 μL. The dPCR mixture was prepared in a pretreatment plate and then transferred to a 96-well 8.5K nanoplate and sealed with a nanoplate sealing film. The plate was then transferred to the QIAcuity Four instrument. First, the priming and rolling amplification steps were performed to generate and separate the chamber partitions, and then the amplification step was carried out with the following amplification program: enzyme activation at 95 °C for 2 minutes, denaturation at 95 °C for 15 seconds, annealing / extension at 60 °C for 30 seconds, for 40 cycles. After the amplification step was an image acquisition step for all wells. Data analysis was performed using the QIAcuity Suite software (Qiagen).

[0197] The total copy number per ng of RNA was determined by calculating the sum of the partitions containing A, G, and exon 5 skipping in each ng of RNA. The percentage of A-to-I editing was determined by dividing the number of partitions containing G in each ng of RNA by the total number (partitions containing G plus A) and multiplying by 100. The percentage of exon 5 skipping was determined by dividing the number of partitions containing exon 5 skipping by the total copy number per ng of RNA.

[0198] Figure 3The RNA editing results of endogenous B4GALT1 transcripts were provided, and the results showed significant differences in efficiency among EONs, but higher editing levels were obtained when increasing from 1 μm EON to 5 μm EON. No editing was observed in the negative controls (only AG1856 and untreated samples (NT)). RM4826, RM4830, RM4834, RM4838, RM4842, and RM4846 seemed to produce the highest RNA editing levels.

[0199] Since the binding of EON to the target transcript may also interfere with splicing and may or may not induce exon skipping events, it was investigated whether exon 5 of the B4GALT1 pre-mRNA was skipped during splicing. The c.1055A target adenosine is located in exon 5 of the B4GALT1 gene. The results are as Figure 4 shown, indicating that the exon skipping levels of some EONs are very high (exceeding 70% in some cases), but these levels do not seem to be related to the editing levels of the above 6 EONs with the best editing effects. The highest exon skipping percentages were observed in RM4832, RM4834, RM4836, RM4838, RM4839, RM4840, and RM4841. Lower exon skipping levels were observed in RM4826, RM4827, RM4828, RM4829 (the 5 μm sample was lost during this process), RM4842, RM4843, RM4844, RM4845, RM4846, RM4847, RM4848, and RM4849. Interestingly, for EONs with lower skipping percentages, these lower exon skipping levels were observed in samples with higher EON concentrations (5 μm), rather than in samples with lower EON concentrations (1 μm). It remains to be determined whether exon skipping induced by RNA editing oligonucleotide therapy is an unwanted effect or, on the contrary, an additive effect as described herein. The main reason for targeting c.1055A in the B4GALT1 transcript is to reduce the functionality of the B4GALT1 protein. Skipping exon 5 of the transcript (which is an in-frame event) may also result in a shorter and less active protein, which may be another beneficial effect of EON treatment.

[0200] Example 3. Editing of human B4GALT1 transcripts in hepatic spheroids.

[0201] Next, it was investigated whether endogenous B4GALT1 transcripts could also be RNA-edited in liver spheroids cultured from primary human hepatocytes. To generate cell spheroids, primary human hepatocyte (PHH) cells from females (BioIVT) were used. According to the supplier's protocol, a PHH cell suspension of 1,500 cells / well (15,000 cells / mL) was seeded into Nuclon Sphera low-attachment U-bottom 96-well plates, using INVITROGRO Cell Spheroid Plating Medium, combined with Spheroid Medium Supplement A, TORPEDO Antibiotic Mix, and INVITROGRO Cell Spheroid Spin Medium (all from BioIVT). The culture plates were then incubated for 5 days at 37 °C in an environment of 5% CO2.

[0202] After incubation, the spheroids were transferred and combined into flat-bottom 96-well plates. After combination, there were 8 spheroids per well, and the total medium volume was 100 μL / well. To maintain spheroid culture, 100 μL of maintenance medium was added to each well, and incubation was continued for 48 hours at 37 °C in an environment of 5% CO2. The maintenance medium consisted of INVITROGRO Spheroid Maintenance Medium (BioIVT) and Spheroid Medium Supplement A, and was also used during spheroid treatment. Then, 100 μL of medium was taken from each well (containing 7-day-old spheroids), and 100 μL of treatment condition was added. Each well was treated with 1 or 5 μM EON and 1 μM AG1856 saponin (see above) for 72 hours. In this initial spheroid experiment, we selected the four EONs (see above) that performed best in the cells: RM4834, RM4838, RM4842, and RM4846. The negative control was saponin and NT samples only.

[0203] After EON / saponin treatment, the spheroids were collected, washed once with PBS, and then 300 μL of lysis buffer was added. RNA isolation, RNA yield determination, cDNA generation, editing efficiency using dPCR, and exon 5 skipping effect evaluation were all carried out according to Example 2.

[0204] The results are as Figure 5As shown. Consistent with the results observed in HepG2 cells, all four EONs were able to mediate relatively high levels of RNA editing of the endogenous B4GALT1 transcript in the spheres grown from PHH, with levels up to over 30% ( Figure 5 A). All four EONs showed a similar range. Similar to that shown in Example 2, exon 5 skipping was also evaluated in these spheres, and the results are as Figure 5 shown in B. Interestingly, treatment with RM4834 and RM4838 led to relatively high levels of exon 5 skipping (over 40%), while RM4842 and RM4846 only led to a skipping percentage of approximately 8%. No editing or exon skipping was detected in samples treated with only saponin ( Figure 5 NT+AG in) or in untreated samples. This indicates that significant levels of RNA editing can be achieved using multiple antisense oligonucleotides, which unexpectedly showed significant differences in their ability to affect the splicing of human B4GALT1 pre-mRNA.

[0205] Example 4. Effect on fibrinogen levels after in vitro editing of human B4GALT1 transcript.

[0206] To study the effect of RNA editing of the c.1055A target adenosine in the human B4GALT1 transcript on fibrinogen secretion, we tested whether this RNA editing would reduce the fibrinogen content in the supernatants of human hepatocellular carcinoma cell lines HepG2 and Huh-7 cells after treatment with EONs. For this purpose, HepG2 cells were cultured in EMEM + 10% FBS + 1% P / S, and Huh-7 cells were cultured in RPMI + 10% FBS + 1% P / S. The cells were placed in an environment of 37 °C and 5% CO2. A total of 0.75×10 5 HepG2 cells and 0.5×10 5Individual Huh-7 cells were seeded in 24-well plates and treated with 5 μM EON and 1 μM AG1856 (see above) in each well for 72 hours. Then, the cell culture supernatant was collected. These samples were centrifuged at 2000 g for 10 minutes. Subsequently, the supernatant was transferred to a new tube. These samples were diluted 1:500, and the fibrinogen level was measured using a high-sensitivity fibrinogen ELISA (ab241383, Abcam) as follows. 50 μL of the sample or standard was added to each well. Subsequently, 50 μL of the antibody mixture was added to the wells. After incubation at room temperature for 1 hour, the wells were washed 3 times with the wash buffer. Then, 100 μL of the TMB developing solution was added to each well, and the plate was incubated in the dark for 10 minutes. Finally, 100 μL of the stop solution was added to each well. The plate was measured at 450 nm using a microplate reader (SpectraMax M5). A standard curve was generated using the 4PL method in GraphPad (version 9.0.1). The raw values were corrected according to the blank value, and the mean was calculated by repeated measurements. The fibrinogen level was determined by interpolation of the standard curve. The fibrinogen level was normalized according to the RNA yield under the NT+ medium condition.

[0207] Example 5. Editing of the human B4GALT1 transcript in primary human hepatocytes.

[0208] Next, a new set of EONs was designed based on the best-performing editors (EON01 (RM4826), EON05 (RM4830), EON09 (RM4834), EON13 (RM4838), EON17 (RM4842), and EON21 (RM4846)) in the first editing screen. Figure 6 This new set of EONs and their respective chemical modifications are presented. Some of these EONs differ in the 5'-terminal portion because some EONs are complementary to exon 6 (therefore, after splicing intron 5 from the pre-mRNA), while some are complementary to intron 5 in the 5'-terminal portion (therefore, before splicing), for example, see the differences in the 5'-terminal portions of B4GALT1-32 (RM106386) and B4GALT1-218 (RM106292). Figure 6 Each of the EONs in (except for B4GALT1-134(-)) contains a tri-antennary GalNAc modification (L001 = OP-042; HongeneBiotech) at the 5'-terminal, which is linked to the most terminal 5'-nucleotide via a TEG linker (L103) to stimulate entry into hepatocytes. The EONs and their attachments were all manufactured according to standard protocols known to those skilled in the art.

[0209] Using the new EON groups, as well as EON01 (RM4826) and EON05 (RM4830), the editing of the endogenous human B4GALT1 transcript was studied in primary human hepatocytes (PHH). Untreated (NT) samples were taken as negative controls. For this purpose, a total of 0.5×10 5 PHH cells (BioIVT) were seeded in INVITGRO CP medium supplemented with the TORPEDO antibiotic mixture. The cells were placed in an environment of 37 °C and 5% CO2. Four hours after seeding, the medium was updated to INVITROGRO HI medium supplemented with the TORPEDO antibiotic mixture for culture. The next day, the cells were treated with 5 μM EON + 1 μM AG1856 per well. This mixture was kept on the cells for 72 hours. Then, the cells were washed once with PBS and 100 μL of lysis buffer was added. Total RNA was isolated using the Direct-zol TM RNA Microprep kit (Direct-zol TM RNA Microprep kit) (Zymo Research). After adding 100 μL of ethanol (95 - 100%), the mixture was loaded onto the column and subjected to multiple washing steps and DNase I treatment. After elution with a total volume of 15 μL of DNase / RNase-free water, the RNA yield was determined using spectrophotometric analysis (NanoDrop) and stored at -80 °C. Subsequently, RT reactions and dPCR were performed as described in Example 2 using the indicated primers and probes.

[0210] The results are shown in Figure 7, where (A) shows the editing percentages obtained in these PHH, clearly indicating that EON05 (without the GalNAc moiety) remains one of the best-performing EONs. This can be explained by the possibility that the interaction of GalNAc with its respective receptor is less important for facilitating entry in vitro, while it may be significantly increased in the in vivo environment. In any case, some new EONs clearly show more favorable effects than some other EONs. For example, B4GALT1-33 is one of the best performers. Figure 7(B) shows the exon 5 skipping percentages in the same samples, clearly indicating significant differences between some EON treatments. For example, after incubation with B4GALT1-212 and -213 EONs, exon skipping is close to zero, while B4GALT1-175 and -176 EONs result in relatively high skipping percentages. Notably, although EON01 and EON05 perform similarly in terms of editing, these two (GalNAc-free) EONs differ greatly in causing exon 5 skipping.

[0211] Subsequently, in the next experiment, we investigated whether RNA editing could also be achieved in PHH without adding saponin, i.e., by simply co-culturing with oligonucleotides in the cell culture medium (= naked uptake, or "naked transfection"). The entire experimental setup was the same as the above setup, except that saponin was not added together with the EON. All downstream RNA purification and dPCR steps were as described above.

[0212] The results of this experiment are shown in Figure 8, where (A) shows the percentage of RNA editing after naked uptake of the indicated EON, which is significantly (and as expected) lower than that in Figure 7 (where saponin was used as the "transfection agent"), but still significantly higher than the background value. It should be noted that a low level of editing does not necessarily mean a low downstream effect on the protein and its function, as described above. The percentage of editing induced by the EON is in the range of 1%, except for B4GALT1-132, which shows a level as high as 4%. The percentage of exon skipping (exon 5) was also determined in these samples. Consistent with the lower editing level, a lower level of exon skipping was also found, see Figure 8(B), and its pattern is similar to that found in the experiment using saponin to improve oligonucleotide entry into cells.

[0213] Example 6. Editing of the B4GALT1 transcript in primary mouse hepatocytes.

[0214] A similar experiment to Example 5 was conducted using primary mouse hepatocytes (PMH), but without using saponin. Primary mouse hepatocytes were isolated from mouse livers using a Liver perfusion kit (Miltenyi) and a GentleMACS Octo dissociator with heater (Miltenyi) according to the manufacturer's instructions. Specifically, the liver was excised from the mouse, washed with PBS, and then transferred to the GentleMACS Octo dissociator with heater. The automated program 37C_m_LIPK_1 was run, which consisted of different steps including initiation, initial perfusion, washing, equilibration, and enzymatic perfusion. After perfusion, the perfused liver lobe and the used enzyme solution were transferred to a gentleMACS C tube (Miltenyi). Subsequently, the gentleMACS C tube was transferred back to the Octo dissociator, and the LIPK_HR_1 program was run to further release hepatocytes from the perfused liver. Finally, the cell solution was passed through a MACS 100 μM SmartStrainer (Miltenyi) to enrich hepatocytes. Cell debris in the cell solution was removed using a Debris Removal kit (Miltenyi). The cell pellet was resuspended in 5 mL DMEM + 5% FBS + 1% P / S. The cells were counted and a total of 0.75×10 5Cells. Cells were treated with 5 μM EON using Williams E medium + 2 mM glutamine + 1% P / S, and the mixture was maintained on the cells for 72 hours. RNA isolation, RNA yield determination, cDNA generation, editing efficiency using dPCR, and exon 5 skipping effect assessment were all performed as in Example 2. However, only one dPCR mix was used, which combined the detection of editing, exon 5 skipping, and B4GALT1 expression in exons 1-2, using mouse-specific primers and probes, as shown in Table 2.

[0215] Table 1. Primer and probe names and sequences (+ indicates LNA nucleotides on the 3’ side)

[0216]

[0217] The results of these experiments are shown in Figure 9, where (A) shows the percentage of editing using the same EONs. These EONs (except for the orphan nucleotide opposite the target adenosine) are 100% complementary to the target sequence in the human B4GALT1 (pre) mRNA transcript. The target sequence in Mus musculus contains a G five nucleotides in the 5’ direction from the target A, which means Figure 6 that the nucleotide at position -5 in the EON (usually uridine) forms a G:U base pair with the mouse target transcript sequence. As shown in WO2017 / 220751, this is not necessarily disadvantageous. On top of the mismatch at the orphan position, additional mismatches, bulges, and / or wobbles may help the ADAR enzyme in the cell better recognize the formed double-stranded RNA complex, which is regarded as a more suitable targeting structure. If this is indeed the case, the data shown in Figure 9(A) also support this view, because in these cases (without adding saponin), the editing levels in these PMHs are higher than those in PHHs (see Figure 8(A)). EON05 (RM4830; SEQ ID NO: 23) performed the best, with an editing level exceeding 10%. Consistent with the results in PHHs, the extent of the contribution of the GalNAc moiety to cell entry under these in vitro conditions remains to be observed. Nevertheless, the RNA editing level of the best-performing EON is around 5%. As shown in Figure 9(B), the pattern of exon 5 skipping percentage is similar to that observed in the above PHHs.

[0218] Example 7. Editing of the human B4GALT1 transcript in primary human hepatocytes using EONs captured by lipid nanoparticles as delivery carriers.

[0219] In the next experiment, we investigated whether a different type of delivery method, namely encapsulating EON in so-called lipid nanoparticles (LNPs), would help deliver oligonucleotides to target cells. For this purpose, we selected EON05 (RM4830; SEQ ID NO:23; see Figure 2 ), EON13 (RM4838; SEQ ID NO:31; see Figure 2 ), and B4GALT1-134(-) (RM107261; SEQ ID NO:1069, which is identical to B4GALT1-134 but does not contain the GalNAc moiety and the TEG linker and is also referred to herein as EON134; see Figure 6 ).

[0220] The LNP formulations containing EON were prepared according to the disclosure in WO2015 / 048020. The lipid stock solutions were first prepared in ethanol. Then, these lipids were mixed in a molar ratio of DLin-MC3-DMA (50), cholesterol (38.5), DSPCV (10), and PEG-200-DMG (1.5). The formulation process can be summarized as follows. In a microfluidic device, the corresponding EON was mixed with the lipids at a ratio of 1:3.6. The formulated LNPs were further concentrated in a tangential flow filtration (TFF) system, dialyzed for 12 hours for buffer exchange, and then sterilized by sterile filtration. The final solution concentration was 1 mg / mL. The LNP formulations containing EON were stored in glass vials at 2-8 °C until further use.

[0221] PHH were obtained and seeded according to the method in Example 5. A total of 0.5×10 5 cells were seeded per well 24 hours before incubation with the LNP formulations. EON05-LNP, EON13-LNP, and EON134-LNP at concentrations ranging from 0.01 to 10 μM in INVITROGRO HI medium + TORPEDO antibiotic mixture + 10% FBS were used per well for 72 hours. Harvesting, RNA isolation, cDNA preparation, and dPCR operations were all carried out according to the instructions in Example 2.

[0222] Figure 10(A) shows the percentage of editing obtained in PHH using three LNP formulations. As shown, the LNP formulations were used at different specified concentrations, with untreated samples as the negative control. This indicates that at the time point of cell harvest, under these conditions, a concentration of 1 μM was optimal, and up to 10% editing levels could be achieved using the EON05-LNP formulation (without saponin assistance). As in the experiment described above, the percentage of exon 5 skipping was also determined in these samples, and the predicted results were given based on the determined editing percentage, see Figure 10 (B).

[0223] Example 8. Editing of B4GALT1 transcript in mouse hepatocytes.

[0224] The three LNP formulations (EON05-LNP, EON13-LNP, and B4GALT1-134(-)(~EON134-LNP)) described in Example 7 were also used in in vivo experiments, in which mice were injected intravenously (IV) with the LNP formulation or control. The in vivo experiment was set up as follows: The oligonucleotide dose level for all mice was 3 mg / kg body weight, but the 6th group was redosed, and the initial dose for the mice in the 5th group on days 2 and 4 was 1.5 mg / kg body weight:

[0225]

[0226] The negative control vector was an LNP formulation without any oligonucleotides, while Actb-LNP was an LNP formulation containing an oligonucleotide (RM3891) targeting the human actin B (Actin B) target sequence, which also served as a negative control in this experiment.

[0227] On the designated necropsy day, the mice were sacrificed according to standard procedures, and then blood samples were collected and organs, including the liver, were excised. RNA was isolated from the liver tissue. 1 mL of Trizol (Thermo Fisher) was added to the liver tissue in a 2 mL tube containing 1.4 mm ceramic beads (ThermoFisher), and the samples were homogenized for 25 seconds at 6 m / s using a Beadmill 24. The tissue homogenate was transferred to a 1.5 mL Eppendorf tube, and 200 μL of chloroform (VWR) was added to each 1 mL of homogenate. After centrifugation at 12,000 g for 15 minutes at 4 °C, 300 μL of the aqueous layer was transferred to a new 1.5 mL Eppendorf tube. Then 300 μL of isopropanol was added, and ReliaPrep was used TMRNA isolation was performed using an RNA Miniprep kit. The mixture was loaded onto the column and subjected to several washing steps and DNase I treatment. Elution was performed using a total volume of 50 μL of DNase / RNase-free water. RNA yield determination, cDNA generation, editing efficiency using dPCR, and exon 5 skipping effect evaluation were all performed as described in Example 2.

[0228] Figure 11 (A) shows the editing percentages in the livers of mice in each treatment group compared to the Actb-LNP control group (RM3891-LNP) and PBS at days 2, 4, 7, and 30 as indicated. These results indicate that the editing level may be as high as 2% two days after dosing, significantly higher than the levels observed in the livers of mice treated with the negative control. The editing percentage decreased over time, as shown by the samples at day 4, day 7, and day 30. Figure 11 (B) shows the percentage of exon 5 skipping in the same samples, again confirming the results observed in the above cells and hepatic spheroids.

[0229] Example 9. Editing of the human B4GALT1 transcript in primary human hepatocytes.

[0230] To explore the potential of the methods described herein and further vary the chemical modifications and lengths of the oligonucleotides to ultimately improve the editing of the B4GALT1 transcript, we designed an additional set of 960 EONs and tested them separately or independently in PHH, PMH, hepatic spheroids, and / or in vivo as described in the previous examples. Figure 12 The 960 EONs and their respective RM numbers and chemical modifications are provided. From this set and other designed EONs outlined herein, the best-performing candidates (with or without the GalNAc moiety (and TEG linker)) were used and formulated into LNP formulations when needed. These best-performing candidate drugs will enter further pre-clinical development stages and ultimately be used to treat CVD patients in need.

[0231] Example 10. Editing of the human B4GALT1 transcript in primary human hepatocytes.

[0232] Furthermore, we designed another set of 22 EONs and tested them together with some of the previously screened better-performing editing-inducing EONs: B4GALT1-05 = EON05 = RM4830 = SEQ ID NO:23; B4GALT1-13 = EON13 = RM4838 = SEQ ID NO:31; B4GALT1-69 = RM107689 = SEQ ID NO:1064; and B4GALT1-84 = RM107704 = SEQ ID NO:1068.

[0233] Figure 13 Shows this new set of 22 EONs and their respective RM numbers and chemical modifications.

[0234] Using these EONs, the editing of the endogenous human B4GALT1 transcript in PHH was investigated. Untreated (NT) samples and samples treated with saponin only were used as negative controls. The cell culture, seeding, and treatment methods were as described in Example 5 above, but 0.5 μM AG1856 was used per well simultaneously. RNA purification, cDNA generation, and dPCR were performed as described in Example 2 and the indicated primers and probes were used.

[0235] The results, as Figure 14 shown, all tested EONs provided relatively high levels of editing, among which B4GALT1-13, B4GALT1-65 (SEQ ID NO: 1079), B4GALT1-71 (SEQ ID NO: 1084), B4GALT1-80 (SEQ ID NO: 1093), and B4GALT1-82 (SEQ ID NO: 1095) performed best.

[0236] Example 11. Exon 5 skipping in the human B4GALT1 transcript after EON treatment.

[0237] As described above, the aim of EON is to edit the human B4GALT1 transcript by introducing A>I deamination, thereby generating an N352S mutation (or "variant") in the protein sequence. The target adenosine at position 1055 in the transcript is relatively close to the 3' end of exon 5 in the pre-mRNA, and it was observed that some of the EONs disclosed herein result in (in addition to triggering the desired RNA editing) skipping of exon 5 from the pre-mRNA, thereby generating an out-of-frame transcript, which in turn also leads to downregulation of active B4GALT1. Thus, although splicing regulation is generally undesirable, in this case it can be regarded as a side effect (or unexpected additional effect), which may be beneficial for the treatment of CVD. Therefore, it was investigated to what extent the position of the complementary sequence of EON to the target RNA affects the skipping of exon 5. For this purpose, a large set of EONs was designed, where 12 individual EONs were of the same length but had different chemical modifications (including GalNac and TEG linkers), and they were in 9 independent groups. Thus, the ability of 9×12 EONs to cause exon skipping was tested, thereby potentially determining the relationship between their position and the exon skipping efficiency. Figure 15(A) shows the exon 5 / exon 6 boundary and the relative position of the target A in exon 5. Below this schematic diagram, the positions of the 9 groups of EONs are provided, where it should be noted that they are schematically represented from 3' to 5', and the 5' end of the EON extends into intron 5, whereby it mainly targets the pre-mRNA. The EONs were tested in PHH cells in a 96-well plate at 5×10 4 cells per well, incubated at a concentration of 5 μM EON, and co-treated with 1 μM AG1856. RM105550 (= B4GALT1-13 with a GalNAc moiety but without a TEG linker; identical to RM106564; SEQ ID NO:1100) was used as a control. The cells were harvested after 72 hours, and exon 5 skipping was then determined as described above. Figure 15(B) shows the results obtained using all individual EONs. Subsets are given below the graph. This result clearly shows that more exon skipping was observed when the EON was complementary to a sequence relatively far from the 3' end of exon 5. However, when the complementarity "moved" towards the exon 5 / intron 5 boundary, the skipping efficiency decreased significantly. This indicates that the position of the EON and its hybridization position to the target sequence not only affect RNA editing but also have an impact on the skipping of exon 5, and these factors together may lead to downregulation of the function of the B4GALT1 protein, as desired and discussed herein.

[0238] Example 12. Editing of the human B4GALT1 transcript in primary human hepatocytes using EONs with different positions of 2'-fluorine modification.

[0239] Test whether the position of the 2'-F modification in EON affects RNA editing and / or exon 5 skipping. For this purpose, a set of 20 EONs were designed based on the sequence and modification of B4GALT1-13 (RM106564; SEQ ID NO: 1100, the same as RM105550 in Figure 15). Figure 16 Different EONs based on B4GALT1-13 and their 2'-F patterns (SEQ ID NOs: 1101 to 1120) were provided. This showed that RM106949 (SEQ ID NO: 1101) was almost completely lacking in 2'-F modification except at one position. Using these EONs, the same experiments as outlined in Example 11 were performed, except that purified RNA was used to determine the percentage of exon 5 skipping for cDNA generation, and dPCR was used to determine the RNA editing of the target adenosine. Figure 17 (A) shows the percentage of editing (black bars) and the percentage of exon 5 skipping (white bars) for each of the 20 EONs relative to B4GALT1-13 (RM106564; SEQ ID NO: 1100 (see also RM4838; SEQ ID NO: 31)). Notably, the editing percentage of RM106564 was approximately 22%, and the exon skipping percentage was 43%, while RM106949 had significant exon skipping (>50%) but little RNA editing. This indicates that the presence of 2'-OMe modification (rather than 2'-F modification) hindered the correct deamination of this target by the ADAR enzyme but did not interfere with splicing regulation. Notably, for some EONs, such as RM106950 (SEQ ID NO: 1102) and RM106963 (SEQ ID NO: 1115), the editing percentage was higher than 30%, while the exon 5 skipping percentage was lower than 30%, indicating that the position of the 2'-F modification was crucial for obtaining different effects. Similar experiments were performed using 17 EONs (SEQ ID NOs: 1122 to 1138) based on B4GALT1-21 EON (RM106566; SEQ ID NO: 1121 (see also RM4846; SEQ ID NO: 39)). The results of the same experiment are as Figure 17 (B) shows, indicating that certain EONs, such as RM106970 (SEQ ID NO: 1123) and RM106971 (SEQ ID NO: 1124), performed comparably to RM106566 in terms of editing (up to 35%) and exon 5 skipping (approx. 2%).

Claims

1. An RNA editing oligonucleotide (EON) that is capable of forming a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell, wherein the double-stranded complex is capable of recruiting an endogenous ADAR enzyme that naturally occurs in the cell, wherein the region contains a target adenosine, wherein the nucleotide in the EON opposite the target adenosine is an orphan nucleotide, wherein the ADAR enzyme deaminates the target adenosine to inosine, and wherein the target RNA nucleic acid molecule is a transcribed molecule of the human β-1,4-galactosyltransferase 1 (B4GALT1) gene.

2. The EON according to claim 1, wherein the B4GALT1 transcribed molecule is a pre-mRNA or mRNA molecule.

3. The EON according to claim 1 or 2, wherein the cell is a human liver cell, preferably a hepatocyte.

4. The EON according to any one of claims 1 to 3, wherein, The target adenosine is located at a position in the B4GALT1 transcript where guanosine would encode a B4GALT1 protein variant with reduced enzyme turnover rate.

5. The EON according to any one of claims 1 to 4, wherein The target adenosine is located at position c.1055A in the B4GALT1 transcript, and wherein the deamination results in the conversion of asparagine (N; Asn) at position 352 in the human wild-type B4GALT1 amino acid sequence to serine (S; Ser).

6. The EON according to any one of claims 1 to 5, wherein, At least one nucleotide contains one or more chemically modified moieties that are not naturally occurring in the ribose, linkage, or base moiety, provided that the orphan nucleotide is not a cytidine containing a 2'-OMe ribose substitution.

7. The EON according to claim 6, wherein one or more of the modifications in the linkage are independently selected from phosphorothioate (PS), phosphonoacetate, dithiophosphate, methylphosphonate (MP), sulfonyl phosphoramidate, or PNdmi internucleoside linkages.

8. The EON according to claim 6 or 7, wherein one or more modifications in the ribose moiety are single or double substitutions at the 2', 3' and / or 5' positions of ribose, each substitution independently selected from: -OH; -F; substituted or unsubstituted, straight-chain or branched lower (C1-C 10 ) alkyl, alkenyl, alkynyl, alkaryl, 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.

9. The EON according to any one of claims 1 to 8, wherein the EON comprises an EON selected from the group provided by SEQ ID NO: 3 to SEQ ID NO: 42, SEQ ID NO: 59 to SEQ ID NO: 1069, and SEQ ID NO: 1078 to SEQ ID NO: 1190, preferably an EON selected from the group provided by SEQ ID NO: 23, SEQ ID NO: 19, SEQ ID NO: 31, SEQ ID NO: 27, SEQ ID NO: 35, SEQ ID NO: 39, SEQ ID NO: 69, SEQ ID NO: 70, SEQ ID NO: 71, SEQ ID NO: 72, SEQ ID NO: 73, SEQ ID NO: 93, SEQ ID NO: 94, SEQ ID NO: 95, SEQ ID NO: 1079, SEQ ID NO: 1084, SEQ ID NO: 1093, SEQ ID NO: 1095, SEQ ID NO: 1100, SEQ ID NO: 1102, SEQ ID NO: 1115, SEQ ID NO: 1121, SEQ ID NO: 1123, SEQ ID NO: 1124, and 1139 to SEQ ID NO: 1190, or consists of the same.

10. The EON according to any one of claims 1 to 9, wherein the orphan nucleotide is a deoxynucleotide carrying a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase or an isouridine nucleobase.

11. A vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding the EON according to any one of claims 1 to 5.

12. A nanoparticle delivery vehicle formulation comprising the EON according to any one of claims 1 to 10.

13. A nanoparticle delivery vehicle formulation according to claim 12, wherein the nanoparticle delivery vehicle is a lipid nanoparticle (LNP).

14. A pharmaceutical composition comprising the EON according to any one of claims 1 to 10, the vector according to claim 11, or the nanoparticle delivery vehicle formulation according to claim 12 or 13, and a pharmaceutically acceptable carrier.

15. The EON according to any one of claims 1 to 10, the vector according to claim 11, the nanoparticle delivery vehicle formulation according to claim 12 or 13, or the pharmaceutical composition according to claim 14, for the treatment of cardiovascular disease (CVD).

16. Use of the EON according to any one of claims 1 to 10, the vector according to claim 11, the nanoparticle delivery vehicle formulation according to claim 12 or 13, or the pharmaceutical composition according to claim 14 in the preparation of a drug for the treatment of CVD.

17. A method for editing a B4GALT1 transcript molecule in vitro, ex vivo or in vivo, the method comprising contacting the B4GALT1 transcript molecule or a part thereof with an AON according to any one of claims 1 to 10, thereby editing the B4GALT1 transcript molecule.

18. A method for treating, alleviating or improving CVD in a patient in need thereof, the method comprising contacting the B4GALT1 transcript molecule in a subject's cells with an EON according to any one of claims 1 to 10, thereby treating the patient.

19. A method for deaminating a target adenosine in a B4GALT1 transcript molecule in a cell, the method comprising the steps of: (i) providing to the cell an EON according to any one of claims 1 to 10; (ii) causing the cell to uptake the EON; (iii) annealing the EON with the B4GALT1 transcript molecule; (iv) causing an endogenous ADAR enzyme naturally present in the cell to deaminate the target adenosine in the target RNA molecule to inosine; and optionally (v) identifying the presence of inosine in the target RNA molecule.

20. The method according to claim 19, wherein the cell is a human cell, preferably a liver cell, more preferably a hepatocyte, wherein the target adenosine is at position c.1055A in the B4GALT1 transcript, and wherein the deamination results in the change of asparagine (N; Asn) at position 352 in the human wild-type B4GALT1 amino acid sequence to serine (S; Ser).

21. The method according to claim 19 or 20, wherein step (v) comprises: a) sequencing the B4GALT1 pre-mRNA or mRNA molecule or its derived cDNA; b) assessing the presence of the 352Ser B4GALT1 protein variant; or c) using a functional readout, preferably assessing the rate of decrease of UDP-Gal, or assessing the glycosylation level of transferrin in serum.

Images