Antisense oligonucleotides for the treatment of cardiovascular disease

Abstract

The disclosure relates to the field of diseases caused by high levels of LDL-C and / or fibrinogen, such as cardiovascular disease (CVD). The disclosure involves antisense oligonucleotides (AONs) for RNA editing technology in deaminating a specific adenosine in transcript molecules of the human Asparagine-Linked Glycosylation 12 homolog glycosyltransferase (ALG12) gene, thereby rendering a mutation from a serine residue at position 275 in the human ALG12 amino acid sequence to a glycine residue (S275G).

Description

ANTISENSE OLIGONUCLEOTIDES FOR THE TREATMENT OF CARDIOVASCULAR DISEASETECHNICAL FIELD

[0001] This disclosure relates to the field of medicine and in particular diseases caused by high plasma levels of Low-Density Lipoprotein cholesterol (LDL-C) and Apolipoprotein-B (Apo-B), such as cardiovascular disease (CVD). The disclosure involves the use of nucleotide editing technology in targeting the transcript for the human dolichyl-phosphate-mannose (Man):Man7- N-acetylglucosamine (GlcNAc)2 -pyrophosphate-dolichyl - alpha - 1 ,6-mannosyltransferase encoded by the Asparagine-Linked Glycosylation 12 homolog glycosyltransferase gene ALG12 to bring about amino acid changes that reduce ALG12 protein function.BACKGROUND

[0002] The synthesis of N-linked glycoproteins starts at the cytosolic surface of the endoplasmic reticulum (ER) with the sequential placement of monosaccharides to dolichol-diphosphate to form the lipid-linked oligosaccharide (LLO) precursor Glc3Man9GlcNAc2-PP-dolichol. This complex and multistep process is catalyzed by different family members of the Asparagine- Linked Glycosylation (ALG) glycosyltransferases. Once the oligosaccharide complex is completed, it is transferred from the dolichol anchor onto an asparagine residue on a newly synthesized protein by the oligosaccharyl-transferase (OST) complex in the lumen of the ER.

[0003] Congenital disorders of glycosylation (CDG) are a group of inherited disorders characterized by defects in protein or lipid glycosylation. CDGs are divided into two subgroups: (i) those that affect the synthesis of the LLOs and their subsequent transfer onto peptide chains (referred to as CDG-1A to CDG-1 L); and (ii) those that affect the further processing of the protein-linked oligosaccharides (referred to as CDG-2A to 2E). CDG-1A is also referred to as PMM2-CDG. CDG-1 B is also referred to as MPI-CDG. CDG-1C is also referred to as ALG6- CDG, whereas CDG-1G is also referred to as ALG12-CDG, which is caused by mutations in the human gene coding for ALG12 glycosyltransferase, a mannosyltransferase responsible for addition of the eighth mannose residue to the growing LLO. To date, only very few CDG-1G patients have been described in literature, owing to the importance of a functional ALG 12 protein. The subjects that have been documented display a wide plethora of clinical symptoms like dysmorphic features, failure to thrive, motor development impairments and decreased immunoglobulins (see e.g., Chantret I et al. 2002. J Biol Chem. 277:25815-25822; Thiel C et al. 2002. Biochem J. 367:195-201 ; Kranz C et al. 2007. Am J Med Genet. 143A: 1371 -1378; Murali C et al. 2014. Molec Genet Metab Rep. 1 :213-219; Tahata S et al. 2019. Molec Genet Metab. 128:409-414; Eklund EA et al. 2005. Molec Genet Metab. 84:25-31).

[0004] Interestingly, measurements of plasma Low-Density Lipoprotein cholesterol (LDL-C) and Apolipoprotein-B (Apo-B) levels in heterozygous CDG-IG carriers showed a consistent decrease in comparison to control subjects not carrying ALG12 mutations with an average decrease of around 50% (Van den Boogert AW et al. 2019. Circulation 140:280-292). Experiments using a human liver cell model, patient-derived fibroblasts, and hepatocyte-like cells derived from induced pluripotent stem cells, demonstrated an increased LDL receptor abundance on the cell surface. It was concluded that the increased level of receptors likely mediated the observed decreased plasma LDL cholesterol phenotype (Van den Boogert et al. 2019).

[0005] Cardiovascular disease (CVD) is one of the leading causes of death and disability worldwide. The provision of therapies for CVD therefore represents a significant unmet medical need. A variety of risk factors exists for the development of CVD. These risk factors include elevated blood levels of LDL-C and / or elevated levels of fibrinogen. Increased LDL-C concentration rises the likelihood of arterial plaque formation. Atherosclerosis and fibrinogen increase the risk of blood clotting and thrombosis. LDL-C is an established risk factor for coronary artery disease (CAD). The identification of mechanisms by which to reduce the blood levels of LDL-C and / or fibrinogen would therefore provide potential targets by which to prevent, ameliorate or treat CVD.

[0006] The disclosure provides one or more alternative and / or improved techniques, compounds and / or compositions, based on RNA editing, for use in the treatment of CVD.SUMMARY

[0007] Disclosed herein is an antisense oligonucleotide (AON) that can form a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell, wherein the double-stranded complex can recruit an endogenous ADAR enzyme naturally present in the cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine, wherein the target RNA nucleic acid molecule is a transcript molecule of the human Asparagine-Linked Glycosylation 12 homolog protein ALG12 gene.

[0008] Preferably, the deamination results in a change from serine to glycine at position 275 in the human wildtype ALG12 amino acid sequence (S275G).

[0009] In one aspect, the ALG12 transcript molecule is a pre-mRNA or an mRNA molecule.

[0010] In one aspect, the orphan nucleotide is a deoxycytidine or a deoxyuridine.

[0011] In one aspect, the orphan nucleotide is a cytidine analog such as a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1 H)-pyridone nucleobase (Zd, also referred to as Benner’s base).

[0012] In one aspect, the orphan nucleotide is a uridine analog such as a deoxynucleotide comprising an iso-uracil nucleobase.

[0013] In one aspect the nucleotide numbering in the AON is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end.

[0014] In one aspect, the first nucleotide 3’ from the orphan nucleotide (position -1 in the AON, which is positioned opposite the cytidine that is 5’ of the target adenosine in the target sequence) is a deoxyinosine (Id, as shown in Fig. 1).

[0015] In one aspect, at least one nucleotide in the AON comprises one or more non-naturally occurring chemical modifications, or one or more additional non-naturally occurring chemical modifications, in the ribose, linkage, or base moiety, with the proviso that the orphan nucleotide is not a cytidine comprising a 2’-OMe ribose substitution.

[0016] In one aspect, the AON comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methylphosphonate (MP), sulfonylphosphoramidate, or a (1 ,3- dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi) internucleotide linkage.

[0017] In one aspect, the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end, and wherein linkage position -2 is an MP or a PNms linkage.

[0018] In one aspect, the AON comprises one or more nucleotides comprising a mono- or disubstitution at the 2', 3' and / or 5' position of the ribose, each independently selected from the group consisting of: -OH; -F; substituted or unsubstituted, linear or branched lower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, that 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; -dimethylamino oxyethoxy; and - dimethylaminoethoxyethoxy.

[0019] In one aspect, the AON comprises or consists of an AON, individually selected from the group consisting of the AONs as provided in detail in SEQ ID NO:2 to 51 , preferably selected from the group consisting of SEQ ID NO:7, 6, 8, 5, 10, 3, 47, 4, 38, 40, 35, 15, 37, 18, 19, 11 , and 17.

[0020] In one aspect, herein disclosed is a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON as disclosed herein.

[0021] In one aspect, herein disclosed is a pharmaceutical composition comprising an AON or a vector according as disclosed herein, and a pharmaceutically acceptable carrier.

[0022] In one aspect, herein disclosed is an AON, a vector, or a pharmaceutical composition as disclosed herein, for use in the treatment of CVD.

[0023] In one aspect, herein disclosed is the use of an AON, or a vector as disclosed herein in the manufacture of a medicament for the treatment of CVD.

[0024] In one aspect, the disclosure provides a method of editing an ALG12 polynucleotide, the method comprising contacting the ALG12 polynucleotide with an AON as disclosed herein, which AON can trigger an adenosine deaminase acting on RNA (ADAR)-mediated adenosine to inosine alteration of an adenosine associated with CVD, thereby editing the ALG12 polynucleotide.

[0025] In one aspect, the disclosure provides a method of treating CVD in a patient in need thereof, the method comprising contacting a ALG12 polynucleotide in a cell of the subject with an AON as disclosed herein, thereby treating the patient.

[0026] In one aspect, the disclosure provides a method of treating, preventing, slowing down or ameliorating CVD in a human subject, the method comprising administering to a human subject (or patient) in need thereof a therapeutically effective amount of an AON as disclosed herein, a vector as disclosed herein, or a pharmaceutical composition as disclosed herein.

[0027] In one aspect, the disclosure provides a method for the deamination of a target adenosine in an ALG12 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the ALG12 pre-mRNA or mRNA molecule; (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine; and optionally (v) identifying the presence of the inosine in the target RNA molecule. Preferably, the target adenosine is the adenosine in the AGC codon coding for serine at position 275 in the wildtype ALG12 protein. In one embodiment, step (v) comprises: a) determining the sequence of the ALG12 pre-mRNA or mRNA molecule; b) assessing the presence of a glycine residue at position 275 (275Gly) in the ALG12 protein variant; or c) using a functional read-out.BRIEF DESCRIPTION OF THE DRAWINGS

[0028] One or more embodiments will now be described, by way of example only, with reference to the accompanying drawings:

[0029] Fig. 1 shows part of the nucleotide sequence of a wildtype human ALG 12 mRNA transcript (SEQ ID NO:1), wherein the AGC codon for serine at position 275 in the wild type ALG12 protein is in capitals. The target adenosine is underlined. Below the target sequence the sequence and chemical modifications are shown of 50 AONs designed for editing the target adenosine, with their respective RM code names and SEQ ID NO’s as shown. The chemical modifications are as follows: Um, Am, Gm, and Cm are 2’-OMe modified uridine, adenosine, guanosine, and cytidine, respectively; Ae and Ge are 2’-MOE modified adenosine and guanosine, respectively; m5Ce is 2’-MOE modified 5-methyl-cytidine; m5Ue is 2’-MOE modified 5-methyluridine (alsosometimes named “Te”; 2’-MOE modified thymidine); m5CI is a 2’-4’ locked 5-methyl-cytidine; Al and Gl are 2’-4’ locked adenosine and guanosine, respectively; Gf, Of, Af, and Ilf are 2’-F modified guanosine, cytidine, adenosine, and uridine, respectively; Zd is a deoxynucleotide (deoxycytidine analog) carrying a Benner’s base; Id is deoxyinosine; Cd is deoxycytidine; “e” refers to a PO linkage; “I” refers to a PNdmi linkage; “A” refers to a MP linkage; “*” refers to a PS linkage.

[0030] Fig. 2 shows the percentage editing of human wildtype ALG12 transcripts of the target adenosine shown in Fig. 1 in Huh7 cells after transfection of 30 AONs (names given below the graph), determined by ddPCR. RM4266 represents a transfection with an unrelated AON. Two other negative controls were transfection mixture only (TF mix) and cells that were not treated (unt cells).DETAILED DESCRIPTION

[0031] The present disclosure relates to the realization that lowering plasma LDL-C and Apo-B levels in principle should provide for a treatment of CVD. The observation that heterozygous (non-diseased) carriers of ALG12 mutations displayed a significant lower plasma LDL-C and Apo-B level in comparison to non-mutant control subjects, led to the realization that introducing a mutation, without completely abolishing the ALG12 functionality would add in the increase in LDL receptor expression and therethrough lowering the LDL-C levels in plasma, and subsequently provide a preventive treatment for CVD. It was realized that ALG12 functionality, in view of the disorders observed in patients carrying homozygous loss-of-function mutations in the ALG12 gene, should not be totally abolished, but rather that a ‘transient’ knock-off would be preferred. Hence, it was realized that gene therapy or complete gene replacement with mutant ALG12 variant was not the way forward.

[0032] As mentioned above, not many CDG-1G patients have been described in literature, let alone the specific mutations that resulted in the occurrence of the disease. One of the few mutations described is a missense mutation in exon 7 and is generally referred to as p.S275N, which is a serine (S) to asparagine (N) change at position 275 of the human ALG12 protein. This position is in the 8thtransmembrane domain (TMD) of ALG12. The serine residue at position 275 is highly conserved among species and 3D-modeling shows it forms a hydrogen bond using the free OH-group together with an asparagine residue at position 355, thereby stabilizing the interaction with the 11thTMD. Apparently, replacing the serine at position 275 with an asparagine residue disrupts this interaction. Without wishing to be bound by theory, this presumably leads to destabilization of the ALG12 protein. Hence, the p.S275N change is considered a loss-of-function mutation.

[0033] It was realized that introducing a similar mutation at this position, without inducing the CDG-1G disease type (like what is ongoing in heterozygous carriers of the p.S275N mutation,see Van den Boogert et al. 2019), would then provide a treatment for CVD as it lowers LDL-C and Apo-B levels.

[0034] Introduction of a loss-of-function mutation in the ALG12 transcript using ADAR-mediated RNA editing (see below) of a target adenosine in a specified codon, based on the evidence provided by scarce naturally occurring variants, is challenging. The abovementioned p.S275N variant perse cannot be engineered through ADAR-mediated deamination, because - by using RNA editing - the AGC codon for serine cannot be changed to AAC (for asparagine), but can only be changed to GGC, which means the introduction of a glycine residue (G) at position 275. Hence, instead of the known p.S275N variant, a new unknown variant, referred to herein as “p.S275G” is generated. Importantly, based on the clinical symptoms of homozygous p.S275N CDG-1G patients, the published heterozygous carrier analysis, and the 3D-modelling highlighting the importance of the presence of the free hydrogen donor of serine at this position, it is postulated here that the introduction of glycine (G; instead of asparagine, N) at position 275 would lead to the same disruption of the functionality of the ALG12 protein and thereby provide increased levels of the LDL receptor, therethrough allowing the decrease in plasma LDL-C levels and hence allowing for a treatment for people suffering from CVD and / or people at risk of developing CVD.

[0035] Hence, it was realized that it is possible to target the ALG12 transcript to modulate the activity of the ALG12 protein, and thereby prevent, ameliorate, or treat CVD, by applying a technology that is generally referred to as “RNA editing”, using oligonucleotides to specifically deaminate a specific target adenosine in the transcript of the (human) ALG12 transcript (pre- mRNA and / or mRNA) in vivo, preferably using endogenous deaminating enzymes, such as human ADAR1 and / or ADAR2 to produce a p.S275G ALG12 protein variant with reduced activity.

[0036] RNA editing is a natural process through which eukaryotic cells alter the sequence of their RNA molecules, often in a site-specific and precise way, thereby increasing the repertoire of genome encoded RNAs by several orders of magnitude. RNA editing enzymes have been described for eukaryotic species throughout the animal and plant kingdoms, and these processes play an important role in managing cellular homeostasis in metazoans from the simplest life forms (such as Caenorhabditis elegans) to humans. Examples of RNA editing are adenosine (A) to inosine (I) conversions and cytidine (C) to uridine (U) conversions, which occur through enzymes called Adenosine Deaminases acting on RNA (ADAR) and APOBEC / AID (cytidine deaminases that act on RNA), respectively.

[0037] ADAR is a multi-domain protein, comprising a catalytic domain, and two to three doublestranded 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 does also play a role in recognizing and binding a part of the dsRNA helix, although the key function of the catalytic domain is to convert an A into I in a nearby, predefined,position in the target RNA, by deamination of the nucleobase. Inosine is read as guanosine by the translational machinery of the cell, meaning that, if an edited adenosine is in a coding region of an mRNA or pre-mRNA, it can recode the protein sequence. A to I conversions may also occur in 5’ non-coding sequences of a target mRNA, creating new translational start sites upstream of the original start site, which gives rise to N-terminally extended proteins, or in the 3’ UTR or other non-coding parts of the transcript, which may affect the processing and / or stability of the RNA. In addition, A to I conversions may take place in splice elements in introns or exons in pre-mRNAs, thereby altering the pattern of splicing. As a result, exons may be included or skipped. The enzymes catalysing adenosine deamination are within an enzyme family of ADARs, which include human deaminases hADARI and hADAR2, as well as hADAR3. However, for hADAR3 no deaminase activity has been demonstrated.

[0038] The use of oligonucleotides to edit a target RNA applying adenosine deaminase has been described (e.g., Woolf et al. 1995. PNAS 92:8298-8302; Montiel-Gonzalez et al. 2013. PNAS 110(45): 18285-18290; Vogel et al. 2014. Angewandte Chemie Int Ed 53:267-271). A disadvantage of the method described by Montiel-Gonzalez et al. (2013, supra) is the need for a fusion protein consisting of the boxB recognition domain of bacteriophage lambda N-protein, genetically fused to the adenosine deaminase domain of a truncated natural ADAR protein. It requires target cells to be either transduced with the fusion protein, which is a major hurdle, or that target cells are transfected with a nucleic acid construct encoding the engineered adenosine deaminase fusion protein for expression. The system described by Vogel et al. (2014, supra) suffers from similar drawbacks, in that it is not clear how to apply the system without having to genetically modify the ADAR first and subsequently transfect or transform the cells harboring the target RNA, to provide the cells with this genetically engineered protein. US 9,650,627 describes a similar system. The oligonucleotides of Woolf et al. (1995, supra) that were 100% complementary to the target RNA sequences suffered from severe lack of specificity: nearly all adenosines in the target RNA strand that was complementary to the antisense oligonucleotide were edited.

[0039] It is known that ADAR may act on any dsRNA. Through a process sometimes referred to as ‘promiscuous editing’, the enzyme will edit multiple A’s in the dsRNA. Hence, there was a need for methods and means that circumvent such promiscuous editing and only target specific adenosines in a target RNA molecule to become therapeutic applicable. Vogel et al. (2014, supra) showed that such off-target editing can be suppressed by using 2’-OMe modified nucleosides in the oligonucleotide at positions opposite to adenosines that should not be edited and used a non-modified nucleoside directly opposite to the specifically targeted adenosine on the target RNA. However, the specific editing effect at the target nucleotide has not been shown to take place without the use of recombinant ADAR enzymes having covalent bonds with the AON. Several publications have now shown that the recruitment of endogenous ADAR (hence without the need for an exogenous and / or recombinant source) is feasible while maintaining aspecificity in which a single adenosine within a target RNA molecule can be targeted and deaminated to an inosine. WO2016 / 097212 discloses AONs for the targeted editing of RNA, wherein the AONs are characterized by a sequence that is complementary to a target RNA sequence (therein referred to as the ‘targeting portion’) and by the presence of a stem-loop I hairpin structure (therein referred to as the ‘recruitment portion’), which is preferably non- complementary to the target RNA. Such oligonucleotides are referred to as ‘self-looping AONs’. The recruitment portion acts in recruiting a natural ADAR enzyme present in the cell to the dsRNA formed by hybridization of the target sequence with the targeting portion. Due to the recruitment portion, there is no need for conjugated entities or presence of modified recombinant ADAR enzymes. WO2016 / 097212 describes the recruitment portion as being a stem-loop structure mimicking either a natural substrate {e.g., the GluB receptor) or a Z-DNA structure known to be recognized by the dsRNA binding domains, or Z-DNA binding domains, of ADAR enzymes. A 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 as described is an intramolecular stem-loop structure, formed within the AON itself, and are thought to attract (endogenous) ADAR. Similar stem-loop structure-comprising systems for RNA editing have been described in WO2017 / 050306, W02020 / 001793, WO2017 / 010556, W02020 / 246560, and WO2022 / 078995.

[0040] WO2017 / 220751 and WO2018 / 041973 describe a next generation type of AONs that do not comprise such a stem-loop structure but that are (almost fully) complementary to the targeted area. In one embodiment, one or more mismatching nucleotides, wobbles, or bulges exist between the oligonucleotide and the target sequence. A sole mismatch may be at the site of the nucleoside opposite the target adenosine, but in other embodiments AONs (or RNA editing oligonucleotides, often abbreviated to ‘EONs’) were described with multiple bulges and / or wobbles when attached to the target sequence area. It appeared possible to achieve in vitro, ex vivo and in vivo RNA editing with AONs lacking a stem-loop structure and with endogenous ADAR enzymes when the sequence of the AON was carefully selected such that it could attract / recruit ADAR. The ‘orphan nucleoside’, which is defined as the nucleoside in the AON that is positioned directly opposite the target adenosine in the target RNA molecule, did not carry a 2’-OMe modification. The orphan nucleoside can be a deoxyribonucleoside (DNA), wherein the remainder of the AON could still carry 2’-O-alkyl modifications at the sugar entity (such as 2’-OMe), or the nucleotides directly surrounding the orphan nucleoside contained chemical modifications (such as DNA in comparison to RNA) that further improved the RNA editing efficiency and / or increased the resistance against nucleases.

[0041] Such effects could even be further improved by using sense oligonucleotides (SONs) that ‘protected’ the AONs against breakdown (described in WO2018 / 134301). The use of chemical modifications and particular structures in oligonucleotides that could be used in ADAR-mediatedediting of specific adenosines in a target RNA have been the subject of numerous publications in the field, such as WO2019 / 111957, WO2019 / 158475, W02020 / 165077, W02020 / 201406, W02020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, W02022 / 007803, W02022 / 018207, WO2022 / 026928, andWO2022 / 124345. The use of specific sugar moieties has been disclosed in for instance W02020 / 154342, W02020 / 154343, W02020 / 154344, WO2022 / 103839, andWO2022 / 103852, whereas the use of stereo-defined linker moieties (in general for oligonucleotides that for instance can be used for exon skipping, in gapmers, in siRNA, or specifically for RNA-editing oligonucleotides, related to a wide variety of target sequences) has been described in WO2011 / 005761 , W02014 / 010250, W02014 / 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, W02019 / 032607 (C9orf72), WO2019 / 055951 , WO2019 / 075357 (SMA / ALS), W02019 / 200185 (DM1), WO2019 / 217784 (DM1), WO2019 / 219581 , W02020 / 118246 (DM1), W02020 / 160336 (HTT), WO2020 / 191252, W02020 / 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.

[0042] Next to these disclosures, an extensive number of publications relate to the targeting of specific RNA target molecules, or specific adenosines within such RNA target molecules, be it to repair a mutation that resulted in a premature stop codon, or other mutation causing disease. Examples of such disclosures in which adenosines are targeted within specified target RNA molecules are W02020 / 157008 and WO2021 / 136404 (USH2A); WO2021 / 113270 (APP); WO2021 / 113390 (CMT1A); W02021 / 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 non-syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); W02021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1).

[0043] Disclosed herein are AONs that can produce RNA editing of a target adenosine in the human ALG12 transcript (pre-mRNA and / or mRNA) which produces a ALG12 protein variant that has a reduced or abolished activity. In one embodiment, the AON results 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% reduction in activity in comparison to the wildtype protein.

[0044] The target adenosine resides in the AGO codon for serine (S; Ser) at position 275 of the human ALG12 protein. Editing of the target adenosine may reduce activity of the ALG12 proteinvariant through different mechanisms, but likely through the (reduced or complete) inability for the residue at position 275 to form a hydrogen bond using an OH-group with an asparagine residue at position 355.

[0045] It is an important advantage of RNA editing versus DNA editing because editing of the DNA to encode a non-functional ALG12 protein would cause all ALG12 molecules to be nonfunctional.

[0046] In a preferred aspect, the AON causes the deamination of the adenosine present at position 823 of the mRNA, thereby generating an inosine. In other words, the AGO codon encoding serine at amino acid position 275 is converted to an IGC codon, which is read as GGC that encodes glycine. Other mutations may additionally be made in the ALG12 transcript, that may also be brought about through RNA editing, to reduce the ALG12 function even further.

[0047] Although in a preferred embodiment, the AON of the present disclosure is a singlestranded oligonucleotide comprising an orphan nucleotide opposite the target adenosine, wherein the orphan nucleotide is chemically modified as disclosed herein, and wherein the remainder of the oligonucleotide is chemically modified to prevent it from nuclease breakdown also as disclosed herein, in another embodiment, the disclosure relates to any kind of oligonucleotide or heteroduplex oligonucleotide complex, that may or may not be bound to hairpin structures (internally or at the terminal end(s)), that may be bound to ADAR or catalytic domains thereof, or wherein the oligonucleotide is expressed through a vector, such as an AAV, or wherein the oligonucleotide is in a circular format.

[0048] It is to be understood that any kind of oligonucleotide-based RNA editing is encompassed by the present disclosure if it relates to the deamination of a nucleotide in the ALG12 transcript, preferably to generate the enzyme variant p.Ser275Gly, and causes the reduction in or abolishment of ALG12 protein function. The protein mutation referred to as p.Ser275G may also be referred to as p.S275G or simply as S275G, whereas the adenosine mutation to guanosine at position 823 in the ALG12 transcript may also be referred to as c.823A>G. The RNA editing of this position is directed at rendering a loss-of-function (LOF) of the resulting ALG12 variant.

[0049] In a preferred aspect, the AON of the present disclosure is a ‘naked’ oligonucleotide, comprising a variety of chemical modifications in the ribose sugar, the base, and / or the internucleoside linkage of one or more of the nucleotides within the sequence, that can hybridize to the ALG12 transcript or a part thereof that includes the target adenosine, and can recruit endogenous ADAR for the deamination of the target adenosine.

[0050] Embodiments

[0051] In one embodiment, the disclosure relates to an AON that can form a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell, wherein the double-stranded complex can recruit an endogenous ADAR enzyme naturally present in the cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON thatis opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine, wherein the target RNA nucleic acid molecule is a transcript molecule encoding the human Asparagine-Linked Glycosylation 12 homolog (ALG12) glycosyltransferase. The human ALG12 gene (NCBI: 79087; Ensembl: ENSG00000182858; OMIM: 607144; UniProt: Q9BV10) and its encoded protein are known in the art under a variety of aliases, but ALG12 is what is being used herein.

[0052] In a preferred embodiment, the deamination of the target adenosine within the transcript molecule results in a loss-of-function variant of the encoded ALG12 protein, subsequently resulting in increased levels of LDL-receptors on the surface of hepatocytes, and / or reduced LDL-C and Apo-B plasma levels.

[0053] In a preferred embodiment, the deamination results in a change from serine to glycine at position 275 (S275G) in the human ALG12 amino acid sequence. The functionality of the resulting variant (S275G) preferably resembles the functionality of the variant S275N that occurs in nature and that has been observed in homozygous CDG-1G patients and in their heterozygous (non-affected; non-diseased) carrier parents.

[0054] In a preferred embodiment, the transcript molecule is a pre-mRNA or an mRNA molecule.

[0055] In a preferred embodiment, the ALG12 loss-of-function mutant is generated in cells that express LDL receptors, such as hepatocytes. Preferably, the target cell of interest, in which the AON preferably acts, is a human liver cell, more preferably a hepatocyte.

[0056] In one aspect, the disclosure relates to an AON in which at least one nucleotide comprises one or more non-naturally occurring chemical modifications, or one or more additional non- naturally occurring chemical modifications, in the ribose, linkage, or base moiety.

[0057] Preferably, the orphan nucleotide is a cytidine, a cytidine analog, a uridine, or a uridine analog, preferably wherein the orphan nucleotide is a deoxynucleotide. A preferred cytidine analog is a nucleotide carrying a 6-amino-5-nitro-3-yl-2(1 H)-pyridone nucleobase, herein and elsewhere also referred to as a ‘Benner’s base’. A preferred uridine analog is a nucleotide carrying an iso-uracil base.

[0058] Preferably, the orphan nucleotide is not a cytidine comprising a 2’-OMe ribose substitution. In one aspect, the nucleotide numbering in the AON is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end, and preferably the nucleotide at position -1 is a deoxyinosine (Id). This Id at position -1 in the AON is opposite the cytidine in the target sequence, which is 5’ from the target adenosine, see Fig. 1.

[0059] In a preferred embodiment, the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end, and wherein linkage position -2 is a methylphosphonate (MP) or a mesyl phosphoramidate (PNms) linkage. Since manufacturing AONs with MP linkages is sometimeschallenging, it is preferred, for ease of manufacturing amongst others, to introduce a PNms linkage at linkage position -2 in the AON.

[0060] In another preferred embodiment, the linkage between the terminal 5’ nucleotide and its neighbour nucleotide in the AON, as well as the linkage between the terminal 3’ nucleotide and its neighbour nucleotide in the AON are each independently or both a (1 ,3-dimethylimidazolidin- 2-ylidene) phosphoramidate (PNdmi) linkage or a PNms linkage.

[0061] Other linkage modifications that are preferably used in the AONs as disclosed herein are phosphorothioate (PS), phosphonoacetate, phosphorodithioate, and sulfonylphosphoramidate linkages.

[0062] In a preferred embodiment, the AON comprises or consists of the sequence of any one of SEQ ID NO:2 to 51.

[0063] In a preferred embodiment, the AON is directly or indirectly bound, preferably conjugated, to a cell-targeting moiety and / or an endosomal release moiety. Multiple cell-targeting moieties are known in the field. To target liver cells, and more specifically hepatocytes, it is preferred to use / V-Acetylgalactosamine (GalNAc) structures for specific interactions with asialoglycoprotein receptors that are highly expressed on hepatocytes. Hence, a preferred cell-targeting moiety is a GalNAc molecule, more preferably a tri-antennary GalNAc molecule, even more preferably the GalNAc structures disclosed in WO2022 / 271806.

[0064] To allow the AON to escape an endosome as soon as it has entered a target cell, it is preferred to apply endosomal release moieties that enable such escape. The endosomal release moiety may be administered separately from the AON, either in a same composition during administration or pre- or post-administration of the AON. However, it is preferred that the endosomal release moiety is present exactly where the AON has been trapped in an endosome.

[0065] One way to achieve such 1 :1 activity is by having the AON attached to the endosomal release moiety. In one aspect, the binding of the endosomal release moiety is by conjugating it to the AON itself, either directly or indirectly, either by linkers or by the GalNAc structure(s) discussed above). The AON conjugated to a GalNAc moiety may be a first formulation in a kit- of-parts, in which a second formulation is an endosomal release enhancer that is also conjugated to a GalNAc moiety.

[0066] A preferred endosomal release moiety is a triterpene glycoside, such as a saponin, preferably AG1856. Conjugation of saponins, such as AG1856, to AONs has been described (see WO2021 / 122998 and WO2024 / 153801).

[0067] In one aspect, the disclosure relates to a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON as disclosed herein.

[0068] In one aspect, the disclosure relates to a nanoparticle delivery vehicle formulation comprising an AON as disclosed herein. Preferably, the nanoparticle delivery vehicle is a LipidNanoparticle (LNP). The LNP may comprise an AON as disclosed herein, which is conjugated to a saponin such as AG1856, for delivery in vivo, with subsequent endosomal release when appropriate.

[0069] In one aspect, the disclosure relates to a pharmaceutical composition comprising an AON, a vector, or a nanoparticle delivery vehicle formulation as disclosed herein, and a pharmaceutically acceptable carrier.

[0070] In one aspect, the disclosure relates to an AON, a vector, a nanoparticle delivery vehicle formulation, or a pharmaceutical composition as disclosed herein, for use in the treatment, prevention, slowing down or amelioration of cardiovascular disease (OVD). It is to be understood that the current disclosure does not relate to compounds and methods of using such compounds in the treatment of CDG disorders, such as CDG-1G. The knowledge obtained from the assessment of patients suffering from CDG-1G and their heterozygous parents (carriers of the ALG12 p.S275N mutation; see Van den Boogert et al. 2019) served to realize that the mutation referred to as S275G would provide a potential treatment for CVD.

[0071] In one aspect, the disclosure relates to a use of an AON, a vector, a nanoparticle delivery vehicle formulation, or a pharmaceutical composition as disclosed herein, in the manufacture of a medicament for the treatment of CVD.

[0072] In one aspect, the disclosure relates to an in vitro, ex vivo, or in vivo method of editing a human ALG12 polynucleotide, the method comprising contacting the ALG12 polynucleotide with an AON as disclosed herein, thereby editing the ALG12 polynucleotide.

[0073] In one aspect, the disclosure relates to a method of treating, slowing down, preventing, or ameliorating CVD in a patient in need thereof, the method comprising contacting an ALG12 polynucleotide in a cell of the subject with an AON as disclosed herein, thereby treating the patient.

[0074] In one aspect, the disclosure relates to a method for the deamination of a target adenosine in a human ALG12 pre-mRNA or mRNA molecule in a cell, wherein the target adenosine is at position 823 of the human ALG12 mRNA sequence, the method comprising the steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the ALG12 pre-mRNA or mRNA molecule; (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine; and optionally (v) identifying the presence of the inosine in the target RNA molecule. Preferably, step (v) comprises: a) determining the sequence of the ALG12 pre-mRNA or mRNA molecule; b) assessing the presence of an ALG12 protein variant with a glycine at position 275 in the amino acid sequence; or using a functional read-out, preferably assessing a reduced plasma concentration of LDL-C.

[0075] Definitions

[0076] The term ‘nucleoside’ refers to the nucleobase linked to the (deoxy)ribosyl sugar, without phosphate groups. A ‘nucleotide’ is composed of a nucleoside and one or more phosphate groups. The term ‘nucleotide’ thus refers to the respective nucleobase-(deoxy)ribosyl- phospholinker, as well as any chemical modifications of the ribose moiety or the phospho group. Thus, the term would include a nucleotide including a locked ribosyl moiety (comprising a 2’-4’ bridge, comprising a methylene group or any other group), an unlocked nucleic acid (UNA), a threose nucleic acid (TNA), a nucleotide including a linker comprising a phosphodiester, phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP, methyl thiophosphonate, phosphoramidate linkages, and the like.

[0077] Sometimes the terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, inosine, and hypoxanthine, are used interchangeably to refer to the corresponding nucleobase on the one hand, and the nucleoside or nucleotide on the other. Thymine (T) is also known as 5-methyluracil (m5U) and is a uracil (U) derivative; thymine, 5-methyluracil and uracil can be interchanged throughout the disclosure. Likewise, thymidine is also known as 5-methyluridine and is a uridine derivative; thymidine, 5-methyluridine and uridine can be interchanged throughout the disclosure.

[0078] Sometimes the terms nucleobase, nucleoside and nucleotide are used interchangeably, unless the context clearly requires differently, for instance when a nucleoside is linked to a neighbouring nucleoside and the linkage between these nucleosides is modified. As stated herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms ‘ribonucleoside’ and ‘deoxyribonucleoside’, or ‘ribose’ and ‘deoxyribose’ are as used in the art.

[0079] Whenever reference is made to an oligonucleotide, oligo, ON, ASO, oligonucleotide composition, antisense oligonucleotide, AON, (RNA) editing oligonucleotide, EON, and RNA (antisense) oligonucleotide, both oligoribonucleotides and deoxyoligoribonucleotides are meant unless the context dictates otherwise. Potentially the oligonucleotide may completely lack RNA or DNA nucleotides (as they appear in nature) and may consist completely of modified nucleotides.

[0080] Whenever reference is made to an ‘oligoribonucleotide’ it may comprise the bases A, G, C, U, or I. Whenever reference is made to a ‘deoxyoligoribonucleotide’ it may comprise the bases A, G, C, T, or I. However, an oligonucleotide of the present disclosure may comprise a mix of ribonucleosides and deoxyribonucleosides. When a deoxyribonucleotide is used, hence without a modification at the 2’ position of the sugar, the nucleotide is often abbreviated to dA. dC, dG or T in which the ‘d’ represents the deoxy nature of the nucleoside, while a ribonucleoside that is either normal RNA or modified at the 2’ position is often abbreviated without the ‘d’ and often abbreviated with their respective modifications and as explained herein.

[0081] Whenever reference is made to nucleotides in the oligonucleotide, such as cytosine, 5- methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5- hydroxycytosine, and p-D-glucosyl-5-hydroxymethylcytosine are included. Whenever referenceis made to adenine, N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine and 7-methyladenine are included. Whenever reference is made to uracil, dihydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1 -methylpseudouracil, 4-thiouracil and 5- hydroxymethyluracil are included. Whenever reference is made to guanine, 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7-trimethylguanosine and N2,7- dimethylguanosine are included. Whenever reference is made to nucleosides or nucleotides, ribofuranose derivatives, such as 2’-deoxy, 2’-hydroxy, and 2’-O-substituted variants, such as 2’-0Me, are included, as well as other modifications, including 2’-4’ bridged variants.

[0082] Whenever reference is made to oligonucleotides, linkages between two mononucleotides may be phosphodiester linkages as well as modifications thereof, including, phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP, phosphoramidate linkers, phosphoryl guanidine, thiophosphoryl guanidine, sulfono phosphoramidate and the like.

[0083] The term ‘comprising’ encompasses ‘including’ as well as ‘consisting of’, e.g., a composition ‘comprising X’ may consist exclusively of X or may include something additional, e.g., X + Y. The term ‘about’ in relation to a numerical value x is optional and means, e.g., x+10%.

[0084] The word ‘substantially’ does not exclude ‘completely’, e.g., a composition which is ‘substantially free from Y’ may be completely free from Y. Where relevant, the word ‘substantially’ may be omitted from the definition of the disclosure.

[0085] The term ‘complementary’ as used herein refers to the fact that the AON hybridizes under physiological conditions to a second nucleic acid strand (for instance when the oligonucleotide as a first nucleic acid strand (= guide oligonucleotide) forms a heteroduplex RNA editing oligonucleotide complex, or HEON, with another complementary nucleic acid strand), or when it forms a double stranded complex with the target RNA sequence. The term does not necessarily mean that each nucleotide in a nucleic acid strand has a perfect pairing with its opposite nucleotide in the opposite sequence. In other words, while an AON may be complementary to a target sequence, there may be mismatches, wobbles and / or bulges between the oligonucleotide and the target sequence, while under physiological conditions that AON still hybridizes to the target sequence such that the cellular RNA editing enzymes can edit the target adenosine. The term ‘substantially complementary’ therefore also means that despite the presence of the mismatches, wobbles, and / or bulges, the AON has enough matching nucleotides between the AON and target sequence that under physiological conditions the AON hybridizes to the target RNA. As shown herein, an AON may be complementary, but may also comprise one or more mismatches, wobbles and / or bulges with the target sequence, if under physiological conditions the AON is able to hybridize to its target.

[0086] The term ‘downstream’ in relation to a nucleic acid sequence means further along the sequence in the 3' direction; the term ‘upstream’ means the converse. Thus, in any sequenceencoding a polypeptide, the start codon is upstream of the stop codon in the sense strand but is downstream of the stop codon in the antisense strand.

[0087] References to ‘hybridisation’ typically refer to specific hybridisation and exclude non-specific hybridisation. Specific hybridisation can occur under experimental conditions chosen, using techniques well known in the art, to ensure that most stable interactions between probe and target are where the probe and target have at least 70%, preferably at least 80%, more preferably at least 90% sequence identity.

[0088] The term ‘mismatch’ is used herein to refer to opposing nucleotides in a double stranded RNA complex which do not form perfect base pairs according to the Watson-Crick base pairing rules. In the historical sense, mismatched nucleotides are G-A, C-A, ll-C, A-A, G-G, C-C, Il-Il pairs. In some embodiments the AON of the present disclosure comprises fewer than four mismatches with the target sequence, for example 0, 1 or 2 mismatches. ‘Wobble’ base pairs are G-ll, l-ll, l-A, and l-C base pairs. In one aspect of the present disclosure, a wobble base pair is present between the -1 position in the AON (that is preferably inosine, more preferably deoxyinosine) and the opposite cytidine in the target sequence, in combination with a preferred mismatch between the target adenosine and the orphan nucleotide, which preferably is a deoxynucleotide carrying a Benner’s base. Although a G:G pairing would be considered a mismatch, that does not necessarily mean that the interaction is unstable, which means that the term ‘mismatch’ may be somewhat outdated based on the current disclosure where a Hoogsteen base-pairing may be seen as a mismatch based on the origin of the nucleotide but still be relatively stable. An isolated G:G pairing in duplex RNA can for instance be quite stable but still be defined as a mismatch.

[0089] The term ‘splice mutation’ relates to a mutation in a gene that encodes for a pre-mRNA, wherein the splicing machinery is dysfunctional in the sense that splicing of introns from exons is disturbed and due to the aberrant splicing, the subsequent translation is out of frame resulting in premature termination of the encoded protein. Often such shortened proteins degrade rapidly and do not have any functional activity.

[0090] An AON (and the complementary nucleic acid strand when two oligonucleotides form a HEON) according to the present disclosure may be chemically modified almost in its entirety, for example by providing nucleotides with a ribose sugar moiety carrying a 2’-OMe substitution, a 2’-F substitution, or a 2’-O-methoxyethyl (2’-MOE) substitution. The orphan nucleotide in the AON is preferably a cytidine or analog thereof (such as a nucleotide carrying a Benner’s base), or a uridine or analog thereof (such as iso-uridine), and / or in one embodiment comprises a diF modification at the 2’ position of the sugar, in another embodiment comprises a deoxyribose (2’- H, DNA), and in yet a further embodiment, at least one and in another embodiment both the two neighbouring nucleotides flanking the orphan nucleotide do not comprise a 2’-OMe modification. Complete modification wherein all nucleotides of the oligonucleotide hold a 2’-OMe modification, with natural bases, results in a non-functional oligonucleotide as far as RNA editinggoes (known in the art), presumably because it hinders the ADAR activity at the targeted position. In general, an adenosine in a target RNA can be protected from editing by providing an opposing nucleotide with a 2'-0Me group (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing a guanine or adenine as opposing base, as these two nucleobases are also able to reduce editing of the opposing adenosine.

[0091] Various chemistries and modifications are known in the field of oligonucleotides that can be readily used in accordance with the disclosure. The regular internucleoside linkages between the nucleotides may be altered by mono- or di-thioation of the phosphodiester bonds to yield PS esters or phosphorodithioate esters, respectively. Other modifications of the internucleosidic linkages are possible, including amidation and peptide linkers.

[0092] In an embodiment, the AON as disclosed herein comprises 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides.

[0093] It is known in the art that RNA editing entities (such as human ADAR enzymes) edit dsRNA structures with varying specificity, depending on several factors. One important factor is the degree of complementarity of the two strands making up the dsRNA sequence. Perfect complementarity of the two strands usually causes the catalytic domain of human ADAR to deaminate adenosines in a non-discriminative manner, reacting with any adenosine it encounters. The specificity of hADARI and 2 can be increased by introducing chemical modifications and / or ensuring several mismatches in the dsRNA, which presumably helps to position the dsRNA binding domains in a way that has not been clearly defined yet. Additionally, the deamination reaction itself can be enhanced by providing an oligonucleotide that comprises a mismatch opposite the adenosine to be edited. Following the instructions in the present application, those of skill in the art will be capable of designing the complementary portion of the oligonucleotide according to their needs.

[0094] The RNA editing protein present in the cell that is of most interest to be used with an AON disclosed herein is human ADAR2. It will be understood by a person having ordinary skill in the art that the extent to which the editing entities inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing molecule. The exact modification may be determined through some trial and error and / or through computational methods based on structural interactions between the AON and the recognition domain of the editing molecule. In addition, or alternatively, the degree of recruiting and redirecting the editing entity resident in the cell may be regulated by the dosing and the dosing regimen of the AON. This is something to be determined by the experimenter in vitro) or the clinician, usually in phase I and / or II clinical trials.

[0095] The target cell can be located in vitro, ex vivo or in vivo. One advantage of the disclosure is that it can be used with cells in situ in a living organism, but it can also be used with cells inculture. In some embodiments cells are treated ex vivo and are then introduced into a living organism e.g., re-introduced into an organism from whom they were originally derived). The disclosure also provides methods to edit target RNA sequences in cells from a transplant or within a so-called organoid, e.g., a liver tissue organoid. Organoids can be thought of as three- dimensional in v / tro-derived tissues but are driven using specific conditions to generate individual, isolated tissues. In a therapeutic setting they are useful because they can be derived in vitro from a patient’s cells, and the organoids can then be re-introduced to the patient as autologous material which is less likely to be rejected than a normal transplant.

[0096] Without wishing to be bound by theory, the RNA editing through hADAR2 is thought to take place on primary transcripts in the nucleus, during transcription or splicing, or in the cytoplasm, where e.g., mature mRNA, miRNA or ncRNA can be edited.

[0097] As outlined herein, it is preferred to target the adenosine at position 823 in the wild-type ALG12 transcript product to yield a change from a AGC codon (encoding serine) to IGC (or GGC, encoding glycine). Generally spoken, RNA editing may be used to create RNA sequences with different properties. Such properties may be coding properties (creating proteins with different sequences or length, leading to altered protein properties or functions), or binding properties (causing inhibition or over-expression of the RNA itself or a target or binding partner; entire expression pathways may be altered by recoding miRNAs or their cognate sequences on target RNAs). Protein function or localization may be changed at will, by functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co- or post-translational modification, catalytic sites of enzymes, binding sites for binding partners, signals for degradation or activation and so on. These and other forms of RNA and protein “engineering”, whether to prevent, delay or treat disease or for any other purpose, in medicine or biotechnology, as diagnostic, prophylactic, therapeutic, research tool or otherwise, are encompassed by the present disclosure. Hence, any RNA editing of a target adenosine in the ALG 12 transcript that results in reduction of ALG12 protein function (with eventually reduction in plasma LDL-C levels) is encompassed by the present disclosure.

[0098] The present disclosure opens a whole new field of treating CVD using genetic editing techniques. The genetic editing technique is not particularly limited. Suitable techniques include known gene therapy techniques, which include DNA editing techniques such as CRISPR / Cas, ZFNs, TALENs, and meganucleases, and preferably RNA editing techniques such as ADAR- mediated editing techniques, as further outlined in detail herein.

[0099] The amount of AON to be administered, the dosage and the dosing regimen can vary from cell type to cell type, the disease to be treated, the target population, the mode of administration {e.g., systemic versus local), the severity of disease and the acceptable level of side activity, but these can and should be assessed by trial and error during in vitro research, in pre-clinical and clinical trials. The trials are particularly straightforward when the modified sequence leadsto an easily detected phenotypic change, or a change in (the level of, or activity of) a specified biomarker. It is possible that higher doses of AONs could compete for binding to an ADAR within a cell, thereby depleting the amount of the entity, which is free to take part in RNA editing, but routine dosing trials will reveal any such effects for a given AON and a given target.

[0100] One suitable trial technique involves delivering the AON to cell lines, or a test organism and then taking biopsy samples at various time points thereafter. The sequence of the target RNA can be assessed in the biopsy sample and the proportion of cells having the modification can easily be followed. After this trial has been performed once then the knowledge can be retained, and future delivery can be performed without needing to take biopsy samples. A method of the disclosure can thus include a step of identifying the presence of the desired change in the cell’s target RNA sequence, thereby verifying that the target RNA sequence has been modified. This step will typically involve sequencing of the relevant part of the target RNA, or a cDNA copy thereof (or a cDNA copy of a splicing product thereof, in case the target RNA is a pre-mRNA), as discussed above, and the sequence change can thus be easily verified. Alternatively, the change may be assessed on the function of the protein, for instance by measuring the reduction rate of UDP-Gal, or assessing glycosylation levels of transferrin in serum, before and after treatment, or any other potential marker, which measurements are preferably performed in vitro on samples obtained from the treated subject.

[0101] After RNA editing has occurred in a cell, the modified RNA can become diluted over time, for example due to cell division, limited half-life of the edited RNAs, etc. Thus, in practical therapeutic terms a method of the disclosure may involve repeated delivery of an AON until enough target RNAs have been modified to provide a tangible benefit to the patient and / or to maintain the benefits over time.

[0102] AONs of the disclosure are particularly suitable for therapeutic use, and so the disclosure also provides a pharmaceutical composition comprising an AON of the disclosure, or a vector or plasmid encoding the AON of the disclosure, and a pharmaceutically acceptable carrier. In some embodiments of the disclosure the pharmaceutically acceptable carrier can simply be a saline solution. This can usefully be isotonic or hypotonic, particularly for pulmonary delivery. The disclosure also provides a delivery device (e.g., syringe, inhaler, nebuliser) which includes a pharmaceutical composition of the disclosure.

[0103] The disclosure also provides an AON of the disclosure for use in the treatment of CVD. This treatment can be achieved through making a change in a target ALG12 RNA sequence in a mammalian, such as a human liver cell. Similarly, the disclosure provides the use of an AON of the disclosure in the manufacture of a medicament for making a change in a target ALG12 RNA sequence in a mammalian, preferably a human liver cell, as described herein, and thereby treating, preventing, or ameliorating CVD.

[0104] The disclosure also provides a method for the deamination of at least one specific target adenosine present in a target ALG12 RNA sequence in a cell, the method comprising the stepsof: providing the cell with an AON according to the disclosure; allowing uptake by the cell of the AON; allowing annealing of the AON to the target RNA molecule; allowing a mammalian ADAR enzyme comprising a natural dsRNA binding domain as found in the wild type enzyme to deaminate the target adenosine (preferably the adenosine at position 823 in the ALG12 transcript product) in the target RNA molecule to an inosine; and optionally identifying the presence of the inosine in the RNA sequence, or guanosine in the resulting cDNA sequence.

[0105] The term cardiovascular disease, or OVD, includes conditions such as CAD, sometimes known as coronary heart disease, strokes and transient ischaemic attack (TIA; or mini stroke), peripheral arterial disease, and / or aortic disease. The AONs of the disclosure may be used in the treatment, prevention, slowing down, or amelioration of any or all these conditions. In a preferred embodiment, the CVD for treatment according to the disclosure is CAD.

[0106] The disclosure also provides a method for the deamination of at least one specific target adenosine present in a target ALG12 RNA sequence in a cell, the method comprising the steps of: providing the cell with a vector or plasmid encoding the AON as disclosed herein; allowing uptake by the cell of the vector or plasmid; allowing annealing of the AON to the target RNA molecule; allowing a mammalian ADAR enzyme comprising a natural dsRNA binding domain as found in the wild type enzyme to deaminate the target adenosine (preferably the adenosine at position 823 in the ALG12 transcript product) in the target RNA molecule to an inosine; and optionally identifying the presence of the inosine in the RNA sequence.

[0107] In a preferred aspect, depending on the ultimate effect of A to I conversion, the identification step comprises the following steps: sequencing the target RNA; assessing the presence or absence of an A to G conversion in target RNA derived cDNA; assessing the presence or absence of a functional protein; assessing whether splicing of the pre-mRNA was altered by the deamination; or using a functional read-out, where the target RNA after the deamination should encode an enzyme with a reduced functionality. For instance, a reduction in ALG12 function or activity can be detected through a reduction in fibrinogen in plasma, a decrease of LDL-C in serum, or through a decrease in serum of tetrasialylated transferrin levels with a corresponding increase in lower sialylation levels (such as an increase in trisialylated transferrin levels).

[0108] A very suitable manner to identify the presence of an inosine after deamination of the target adenosine is of course dPCR or even sequencing, using methods that are well-known to the person skilled in the art. However, the person skilled in the art of liver disease may apply tests to monitor certain biomarkers related to LDL-C and or fibrinogen levels, as discussed above.

[0109] The AON according to the disclosure is suitably administrated in aqueous solution, e.g. saline, or in suspension, optionally comprising additives, excipients and other ingredients, compatible with pharmaceutical use, at concentrations ranging from 1 ng / ml to 1 g / ml, preferably from 10 ng / ml to 500 mg / ml, more preferably from 100 ng / ml to 100 mg / ml. Dosage may suitablyrange from between about 1 pg / kg to about 100 mg / kg, preferably from about 10 pg / kg to about 10 mg / kg, more preferably from about 100 pg / kg to about 1 mg / kg. Administration may be by inhalation (e.g., through nebulization), intranasally, orally, by injection or infusion, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra-peritoneally, intrarectally, intrathecally, intra-cisterna magna, parenterally, and the like. Administration may be in solid form, in the form of a powder, a pill, a gel, a solution, a slow-release formulation, or in any other form compatible with pharmaceutical use in humans. As outlined herein, the AONs of the present disclosure may also be delivered through a delivery vehicle such as an LNP. Amounts of LNP carrying AONs as disclosed herein can and will also be determined in (pre-) clinical phases.

[0110] In one embodiment, a method according to the disclosure comprises the steps of administering to the subject an AON or pharmaceutical composition according to the disclosure, allowing the formation of a double stranded nucleic acid complex of the AON with its specific complementary target nucleic acid molecule in a cell in the subject; allowing the engagement of an endogenous present adenosine deaminating enzyme, such as ADAR2; and allowing the enzyme to deaminate the target adenosine in the target nucleic target molecule to an inosine, thereby alleviating, preventing, slowing down, or ameliorating OVD.

[0111] RNA editing molecules present in the cell will usually be proteinaceous 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, still more preferably an adenosine deaminase. These are enzymes with ADAR activity. The ones of most interest are the human ADARs, hADARI and hADAR2, including any isoforms thereof. RNA editing enzymes known in the art, for which oligonucleotide constructs according to the disclosure may conveniently be designed, include the adenosine deaminases acting on RNA (ADARs), such as hADARI and hADAR2 in humans or human cells and cytidine deaminases.

[0112] It is known that hADARI exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 110 kDa version, that is produced through alternative splicing from a common pre-mRNA. Consequently, the level of the 150 kDa isoform available in the cell may be influenced by interferon, particularly interferon-gamma (IFN-y). hADARI is also inducible by TNF-a. This provides an opportunity to develop combination therapy, whereby IFN-y or TNF-a and AONs according to the disclosure are administered to a patient either as a combination product, or as separate products, either simultaneously or subsequently, in any order. Certain disease conditions may already coincide with increased IFN-y or TNF-a levels in certain tissues of a patient, creating further opportunities to make editing more specific for diseased tissues. It will be understood by a person having ordinary skill in the art that the extent to which the editing entities inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0113] Chemical modifications

[0114] Various chemistries and modifications are known in the field of oligonucleotides that can be readily used in accordance with the disclosure. All chemical modifications listed herein that may be used in the AON as disclosed herein may also be used for a sense strand that is complementary to the AON, when the AON and the complementary strand form a HEON complex, such as described in W02024 / 084048, and as disclosed above, except that the opposite sense strand does not have an orphan nucleotide. Hence, the modification related to the orphan nucleotide relate only to the AON as disclosed herein, but all other modifications relate to the AON as disclosed herein and any (protecting) sense oligonucleotide that may be used together with the AON in a pharmaceutical product. This includes the use of hydrophobic moieties (such as tocopherol and cholesterol) and cell-specific ligands, that have also been described herein, and in detail in W02024 / 084048, which may either be bound to the AON or its opposite strand, or both.

[0115] The skilled person knows that an oligonucleotide, such as an AON as outlined herein, generally consists of repeating monomers. Such a monomer is most often a nucleotide or a chemically modified nucleotide. The most common naturally occurring nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (II). These consist of a pentose sugar, a ribose, a 5’-linked phosphate group which is linked via a phosphate ester, and a T-linked base. The sugar connects the base and the phosphate and is therefore often referred to as the “scaffold” of the nucleotide. A modification in the pentose sugar is therefore often referred to as a ‘scaffold modification’. The original pentose sugar may be replaced in its entirety by another moiety that similarly connects the base and the phosphate. It is therefore understood that while a pentose sugar is often a scaffold, a scaffold is not necessarily a pentose sugar. Examples of scaffold modifications that may be applied in the monomers of the AON as disclosed herein are disclosed in W02020 / 154342, W02020 / 154343, and WO2020 / 154344.

[0116] A nucleoside in the AON as disclosed herein may be a natural nucleoside (deoxyribonucleoside or ribonucleoside) or a non-natural nucleoside. It is noted that for RNA editing, in which double-stranded RNA is generally the substrate for enzymes with deamination activity (such as ADARs), ribonucleosides are considered ‘natural’, while deoxyribonucleosides may then be, for the sake of argument, considered as non-natural, or considered as modified, simply because DNA is not present in the RNA-RNA double stranded (natural) substrate configurations. The skilled person appreciates that when the nucleotide has a natural ribose moiety, it may still be non-naturally modified in the base and / or the linkage.

[0117] It is recognized in the art that common limiting factors in oligonucleotide-based therapies are the oligonucleotide’s ability to be taken up by the cell, when delivered per se, or ‘naked’ (= without applying a viral vector or plasmid), the biodistribution and the resistance to nuclease- mediated breakdown. The skilled person is aware, and it has been described in detail in the art,that a variety of chemical modifications can assist in overcoming such limitations. Examples of such now commonly used chemical modifications are the 2’-OMe, 2’-F, 2’,2’-diF, and 2’-MOE modifications of the sugar and the use of PS linkages between nucleosides, as described herein.

[0118] Scaffold modifications (ribose)

[0119] The ribose 2’ groups in all nucleotides of the AON as disclosed herein, except for the ribose sugar moiety of the orphan nucleotide that has certain limitations in respect of compatibility with RNA editing, can be independently selected from 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-OMe, 2’-MOE, 2’-F, or 2’-4’-linked (for instance a locked nucleic acid (LNA)), or other ribosyl T-substitutions, 2’ substitutions, 3’ substitutions, 4’ substitutions or 5’ substitutions. The orphan nucleotide in the AON that comprises no other chemical modifications to the ribose sugar, the base, or the linkage preferably does not carry a 2’-OMe or 2’-MOE substitution when the nucleobase is a naturally occurring cytosine, but may carry a 2’-F, a 2’,2’-difluoro (diF), or 2’-ara-F (FANA) substitution or may be DNA. WO2024 / 013360 describes the modification of the 2’ position of the ribose sugar moiety of the orphan nucleotide by a 2’,2’-disubstituted substitution such as diF, which is also applicable to what is disclosed here. The 2’-4’ linkage can be selected from many linkers known in the art such as a methylene linker, amide linker, or constrained ethyl linker (cEt).

[0120] An AON as disclosed herein may comprise one or more nucleotides carrying a 2’-MOE ribose modification. Also, an AON as disclosed herein may comprise one or more nucleotides not carrying a 2’-MOE ribose modification, or wherein the 2’-MOE ribose modifications are at positions that do not prevent the enzyme with adenosine deaminase activity from deaminating the target adenosine. An AON as disclosed herein may comprise a 2’-OMe ribose modification at a position that does not comprise a 2’-MOE ribose modification. An AON as disclosed herein may comprise deoxynucleotides at positions that do not comprise a 2’-MOE or a 2’-OMe ribose modification, or other 2’ ribose substitution. An AON as disclosed herein may comprise one or more nucleotides comprising a 2’ substitution comprising a 2’-MOE, 2’-OMe, 2’-OH, 2’-deoxy, threose nucleic acid (TNA), 2’-fluoro (2’-F), 2’,2’-difluoro (diF) modification, 2’-fluoro-2’-C-methyl modification, or a 2’-4’-linkage (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA or examples mentioned in e.g. WO2018 / 007475)).

[0121] Other nucleic acid monomers that may be used in an AON as disclosed herein are arabinonucleic acids and 2’-deoxy-2’-fluoroarabinonucleic acid (FANA), for instance for improved affinity purposes. The 2’-4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker. A wide variety of 2’ modifications that may present in an AON as disclosed herein are known in the art, including but not limited to the modifications outlined in detail in WO2016 / 097212, WO2017 / 220751 , WO2018 / 041973, WO2018 / 134301 , WO2019 / 219581 , WO2019 / 158475, and WO2022 / 099159.

[0122] In all cases, the modifications should be compatible with RNA editing such that the AON fulfils its role as an oligonucleotide that can form a double stranded complex with the target RNA and by generating this double-stranded nucleic acid complex, recruit a deaminating enzyme, which can subsequently deaminate the target adenosine.

[0123] Where a monomer in an AON as disclosed herein comprises an unlocked nucleic acid (UNA) ribose modification, that monomer can have a 2’ position comprising the same modifications discussed above, such as a 2’-MOE, a 2’-OMe, a 2’-OH, a 2’-deoxy, a 2’-F, a 2’,2’-diF, a 2’-fluoro-2’-C-methyl, an arabinonucleic acid, a FANA, or a 2’-4’-linkage (i.e., a bridged nucleic acids such as an LNA).

[0124] In one aspect, the AON as disclosed herein comprises at least one nucleotide comprising a TNA ribose modification. In one aspect, the AON as disclosed herein comprises at least one nucleotide with a sugar moiety that comprises a 2’-fluoro (2’-F) modification. A preferred position for the nucleotide that carries a 2’-F modification is position -3 in the AON, which may be but is not necessarily present together with an identical 2’ modification in the orphan nucleotide as discussed above.

[0125] Base modifications

[0126] A base, sometimes called a nucleobase, is generally adenine, cytosine, guanine, thymine or uracil, or a derivative thereof. A nucleobase is defined as a moiety that can bond to another nucleobase through H-bonds, polarized bonds (such as through OF moieties) or aromatic electronic interactions.

[0127] Cytosine, thymine, and uracil are pyrimidine bases, and are generally linked to the scaffold through their 1 -nitrogen.

[0128] Adenine and guanine are purine bases and are generally linked to the scaffold through their 9-nitrogen.

[0129] The terms ‘adenine’, ‘guanine’, ‘cytosine’, ‘thymine’, ‘uracil’ and ‘hypoxanthine’ as used herein refer to the nucleobases as such. The terms ‘adenosine’, ‘guanosine’, ‘cytidine’, ‘thymidine’, ‘uridine’ and ‘inosine’ refer to the nucleobases linked to the (deoxy)ribosyl sugar. The nucleobases in an AON as disclosed herein can be adenine, cytosine, guanine, thymine, or uracil or any other moiety able to interact with another nucleobase through H-bonds, polarized bonds (such as CF) or aromatic electronic interactions.

[0130] The nucleobases at any position in the AON as disclosed herein can be a modified form of adenine, cytosine, guanine, or uracil, such as hypoxanthine (the nucleobase in inosine), pseudouracil, pseudocytosine, isouracil, N3-glycosylated uracil, 1 -methylpseudouracil, orotic acid, agmatidine, lysidine, 2-thiouracil, 2-thiothymine, 5-substituted pyrimidine (e.g., 5- halouracil, 5-halomethyluracil, 5-trifluoromethyluracil, 5-propynyluracil, 5-propynylcytosine, 5- aminomethyluracil, 5-hydroxymethyluracil, 5-formyl uracil, 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-oxo-adenine, 3-deazapurine (such as a 3-deaza-adenosine), pseudoisocytosine, N4-ethylcytosine, N2-cyclopentylguanine, N2-cyclopentyl-2-aminopurine, N2-propyl-2-aminopurine, 2,6-diaminopurine, 2-aminopurine, G-clamp and its derivatives, Super A, Super T, Super G, amino-modified nucleobases or derivatives thereof; and degenerate or universal bases, like 2,6-difluorotoluene, or absent like abasic sites {e.g. 1 -deoxyribose, 1 ,2- dideoxyribose, 1-deoxy-2-O-methylribose, azaribose).

[0131] Modified bases comprise 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 that are or will be known in the art. Purine nucleobases and / or pyrimidine nucleobases may be modified to alter their properties, for example by amination or deamination of the heterocyclic rings. The exact chemistry and formats may vary from oligonucleotide construct to oligonucleotide construct and from application to application, and may be worked out in accordance with the wishes and preferences of those of skill in the art.

[0132] A scaffold modification indicates the presence of a modified version of the ribosyl moiety as naturally occurring in RNA (i.e., the pentose moiety), such as bicyclic sugars, tetrahydropyrans, hexoses, morpholinos, 2’-modified sugars, 4’-modified sugar, 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 such as 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 such as 2’-O-(2- chloroethoxy)methyl (MCEM), 2’-O-(2,2-dichloroethoxy)methyl (DCEM); 2’-O-alkoxycarbonyl such as 2’-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2’-O-[2- / V-methylcarbamoyl)ethyl] (MCE), 2’- O-[2-( / V, / V-dimethylcarbamoyl)ethyl] (DCME); 2’-halo e.g. 2’-F, FANA; 2'-O-[2-(methylamino)-2- oxoethyl] (NMA); a bicyclic or bridged nucleic acid (BNA) scaffold modification such as a conformationally restricted nucleotide (CRN) monomer, a locked nucleic acid (LNA) monomer, a xy / o-LNA monomer, an a-LNA monomer, an a-l-LNA monomer, a p-d-LNA monomer, a 2’- amino-LNA monomer, a 2’-(alkylamino)-LNA monomer, a 2’-(acylamino)-LNA monomer, a 2’- / V-substituted 2’-amino-LNA monomer, a 2’-thio-LNA monomer, a (2’-O,4’-C) constrained ethyl (cEt) BNA monomer, a (2’-O,4’-C) constrained methoxyethyl (cMOE) BNA monomer, a 2’,4’- BNANC(NH) monomer, a 2’,4’-BNANC(NMe) monomer, a 2’,4’-BNANC(NBn) monomer, an ethylene-bridged nucleic acid (ENA) monomer, a carba-LNA (cLNA) monomer, a 3,4-dihydro- 2 / 7-pyran nucleic acid (DpNA) monomer, a 2’-C-bridged bicyclic nucleotide (CBBN) monomer, an oxo-CBBN monomer, a heterocyclic-bridged BNA monomer (such as triazolyl or tetrazolyl- linked), an amido-bridged BNA monomer (such as AmNA), an urea-bridged BNA monomer, asulfonamide-bridged BNA monomer, a bicyclic carbocyclic nucleotide monomer, a TriNA monomer, an a-l-TriNA monomer, a bicyclo DNA (bcDNA) monomer, an F-bcDNA monomer, a tricyclo DNA (tcDNA) monomer, an F-tcDNA monomer, an alpha anomeric bicyclo DNA (abcDNA) monomer, an oxetane nucleotide monomer, a locked PMO monomer derived from 2’-amino LNA, a guanidine-bridged nucleic acid (GuNA) monomer, a spirocyclopropylene- bridged nucleic acid (scpBNA) monomer, and derivatives thereof; cyclohexenyl nucleic acid (CeNA) monomer, altriol nucleic acid (ANA) monomer, hexitol nucleic acid (HNA) monomer, fluorinated HNA (F-HNA) monomer, pyranosyl-RNA (p-RNA) monomer, 3’-deoxypyranosyl DNA (p-DNA), unlocked nucleic acid UNA); an inverted version of any of the monomers above. All these modifications are known to the person skilled in the art.

[0133] The orphan nucleotide

[0134] Mutagenesis studies of human ADAR2 revealed that a single mutation at residue 488 from glutamate to glutamine (E488Q), gave an increase in the rate constant of deamination by 60- fold when compared to the wild-type enzyme (Kuttan and Bass. Proc Natl Acad Sci USA 2012. 109(48): 3295-3304). During the deamination reaction, ADAR flips the edited base out of its RNA duplex, and into the enzyme active site (Matthews et al. 2016. Nat Struct Mol Biol. 23(5) :426- 433). When ADAR2 edits adenosines in the preferred context (an A:C mismatch) the nucleotide opposite the target adenosine is often referred to as the ‘orphan nucleotide’ (or ‘orphan cytidine’ as the case may be), as indicated above. The crystal structure of ADAR2 E488Q bound to double stranded RNA (dsRNA) revealed that the glutamine (Gin; Q) side chain at position 488 can donate an H-bond to the N3 position of the orphan cytidine, which leads to the increased catalytic rate of ADAR2 E488Q. In the wild-type enzyme, wherein a glutamate (or glutamic acid; Glu; E) is present at position 488 instead of a glutamine the amide group of the glutamine is absent and is instead a carboxylic acid. To obtain the same contact of the orphan cytidine with the E488Q mutant would then, for the wild-type situation, require protonation for this contact to occur. To make use of endogenously expressed ADAR2 to correct disease relevant mutations, it is essential to maximize the editing efficiency of the wild type ADAR2 enzyme present in the cell. WO2020 / 252376 discloses the use of AONs with modified RNA bases, especially at the position of the orphan cytidine to mimic the hydrogen-bonding pattern observed by the E488Q ADAR2 mutant. By replacing the nucleotide opposite the target adenosine in the AON with cytidine analogs that serve as H-bond donors at N3, it was envisioned that it would be possible to stabilize the same contact that is believed to provide the increase in catalytic rate for the mutant enzyme. Two cytidine analogs were of particular interest: pseudoisocytidine (also referred to 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 referred to as ‘dZ’; Yang et al. 2006. NuclAcid Res. 34(21):6095-6101) that were initially selected because they offer hydrogen-bond donation at N3 with minimal perturbation to the shape of the nucleobase. Benner’s base is alsoreferred to with its chemical name 6-amino-5-nitro-3-yl-2(1 H)-pyridone. The presence of the cytidine analog in the AON may exist in addition to modifications to the ribose 2’ group. The ribose 2’ groups in the orphan nucleotide can be independently selected from 2’-H (i.e. , DNA), 2’-OH (i.e., RNA), 2’-0Me, 2’-M0E, 2’-F, or 2’-4’-linked (i.e., a bridged nucleic acid such as an LNA), or other 2’ substitutions. The 2’-4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker.

[0135] The orphan nucleotide in the AON as disclosed herein is preferably a cytidine or analog thereof (such as a nucleotide carrying a Benner’s base) or a uridine or analog thereof (such as iso-uridine). The orphan nucleotide, whether it is a cytidine or analog thereof, or a uridine or analog thereof, preferably comprises a deoxyribose (2’-H; = DNA) but may also comprise a diF modification at the 2’ position of the sugar. In one aspect at least one and in another aspect both the neighbouring (directly adjacent) nucleotides flanking the orphan nucleotide do not comprise a 2’-OMe modification. Complete modification wherein all nucleotides of the oligonucleotide hold a 2’-OMe modification (including the orphan nucleotide), with natural bases, results in a non-functional oligonucleotide as far as RNA editing goes (known in the art), presumably because it hinders the ADAR activity at the targeted position. In general, an adenosine in a target RNA can be protected from editing by providing an opposing nucleotide with a 2'-OMe group (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing a guanine or adenine as opposing base, as these two nucleobases are also able to reduce editing of the opposing adenosine.

[0136] Linkage modifications

[0137] A nucleoside is generally connected to neighboring nucleosides through condensation of its 5’-phosphate moiety to the 3’-hydroxyl moiety of the neighboring nucleotide monomer. Similarly, its 3’-hydroxyl moiety is generally connected to the 5’-phosphate of a neighboring nucleotide monomer. This forms phosphodiester bonds. The phosphodiesters and the scaffold form an alternating copolymer. The bases are grafted on this copolymer, namely to the scaffold moieties. Because of this characteristic, the alternating copolymer formed by linked scaffolds of an oligonucleotide is often called the ‘backbone’ of the oligonucleotide. Because phosphodiester bonds connect neighboring monomers together, they are often referred to as “backbone linkages”. It is understood that when a phosphate group is modified so that it is instead an analogous moiety such as a PS, such a moiety is still referred to as the backbone linkage of the monomer. This is referred to as a “backbone linkage modification”. In general terms, the backbone of an oligonucleotide comprises alternating scaffolds and backbone linkages.

[0138] As outlined in detail herein, naked AONs as disclosed herein comprise at least one, preferably multiple linkage modifications. It is generally more preferred that the AON as disclosed herein comprises linkage modifications at most, and potentially all positions if theAON is capable of mediating RNA editing through the deamination enzyme when the AON is bound to the target RNA nucleic acid molecule. A linkage modification can be, but is not limited to, a modified version of the phosphodiester present in RNA, such as PS, chirally pure PS, (R)- PS, (S)-PS, methyl phosphonate (MP, also referred to as MeP), chirally pure MP, (R)-MP, (S)- MP, phosphoryl guanidine (such as PNdmi), chirally pure phosphoryl guanidine, (R)-phosphoryl guanidine, (S)-phosphoryl guanidine, phosphorodithioate (PS2), phosphonacetate (PACE), phosphonoacetamide (PACA), thiophosphonoacetate, thiophosphonoacetamide, methyl phosphorohioate, methyl thiophosphonate, PS prodrug, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, boranophosphate, borano PS, metyl boranophosphate, methyl borano PS, methyl boranophosphonate, methyl boranophosphothioate, phosphate, phosphotriester, aminoalkylphosphotriester, and their derivatives. Another modification includes phosphoramidite, phosphoramidate, N3’->P5’ phosphoramidate, phosphorodiamidate, phosphorothiodiamidate, sulfamate, diethylenesulfoxide, amide, sulfonate, siloxane, sulfide, sulfone, formacetyl, alkenyl, methylenehydrazino, sulfonamide, triazole, oxalyl, carbamate, methyleneimino (MMI), and thioacetamide nucleic acid (TANA); and their derivatives. Various salts, mixed salts, deprotonated, protonated, tautomeric, and free acid forms are also included, as well as 3’->3’ and 2’->5’ linkages. An AON as disclosed herein may also comprise one or more linkage modifications according to the structure of formula (I):

[0139] wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl and the linkage modification is referred to as mesyl phosphoramidate, MsPA or PNms.

[0140] In a preferred aspect, the AON as disclosed herein comprises an internucleoside linkage of the structure of formula (I), wherein X = O and R = CH3, which linkage is generally referred to herein as a PNms linkage (mesyl phosphoramidate). In other preferred aspects, R equals one of the following structures (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0141] Disclosed herein is also an AON that is able to mediate adenosine deamination by recruitment of a deaminating enzyme in a cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the deaminating enzyme can deaminate the target adenosine into an inosine, and wherein the AON comprises a moiety with a structure according to formula (II):

[0142] wherein: X = O or S; Y = O' or S'; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl.

[0143] An AON as disclosed herein may comprise a substitution of one of the non-bridging oxygens in the phosphodiester linkage. This modification slightly destabilizes base pairing but adds significant resistance to nuclease degradation. A preferred nucleotide analogue or equivalent comprises PS, phosphonoacetate, phosphorodithioate, phosphotriester, aminoalkylphosphotriester, H-phosphonate, methyl and other alkyl phosphonate including 3'- alkylene phosphonate, 5'-alkylene phosphonate and chiral phosphonate, phosphinate, phosphoramidate including 3'-amino phosphoramidate and aminoalkylphosphoramidate, thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, selenophosphateor boranophosphate. Particularly preferred are internucleoside linkages that are modified to contain a PS. Particularly preferred are internucleoside linkages that are modified to contain a PNms. Particularly preferred are internucleoside linkages that are modified to contain a PNdmi. The regular internucleosidic linkages between the nucleotides may be altered by mono- or di- thioation of the phosphodiester bonds to yield PS esters or phosphorodithioate esters, respectively. Other modifications of the internucleosidic linkages are possible, including amidation and peptide linkers. The skilled person can determine for what target RNA nucleic acid molecule the AON comprises a certain linkage modification at each linkage position of the AON as disclosed herein to generate the most effective and stable oligonucleotide compound.

[0144] Many of the non-naturally occurring modifications of the linkage, such as PS, are chiral. This means that there are Rp and Sp configurations, known to the person skilled in the art. In one embodiment, the chirality of the PS linkages is controlled, which means that each of the linkages is either in the Rp or in the Sp configuration, whichever is preferred. The choice of an Rp or Sp configuration at a specified linkage position may depend on the target sequence and the efficiency of binding and induction of causing RNA editing of the target adenosine. However, if such is not specifically desired, a composition may comprise AONs as active compounds with both Rp and Sp configurations at a certain specified linkage position. Mixtures of such AONs are also feasible, wherein certain positions preferably have either one of the configurations, while for other positions such does not matter. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PS linkages.

[0145] In one aspect, the AON as disclosed herein comprises one of more (chirally pure or chirally mixed) phosphoramidate (PN) linkages. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PNms linkages. In one aspect, a PN linkage connects the terminal two nucleotides on each end of the AON. AONs as disclosed herein may also comprise linkage modifications at all positions that are not chirally controlled. The AON as disclosed herein may also comprise one or more naturally occurring internucleoside linkages. The choice and number of modified linkages may depend on the specific target, the sequence, the length, and the stability of the AON observed in a particular cell type of interest, which can be assessed by methods known to the person skilled in the art. In one aspect, at least one, at least two, at least three, or at least four internucleoside linkages between the 5’ and / or the 3’ terminal two, three, four, or five nucleosides respectively of the AON as disclosed herein are modified internucleoside linkages. In one aspect, the AON as disclosed herein comprises at least one MP internucleoside linkage according to the structure of formula (III):

[0146] As was noted in the art, a preferred position for an MP linkage in an AON is linkage position -2, thereby connecting the nucleoside at position -1 with the nucleoside at position -2. In one aspect, this position, in an AON as disclosed herein, comprises a linkage modification according to the structure of formula (I), instead of an MP linkage. W02020 / 201406 discloses the use of MP linkage modifications at certain positions surrounding the orphan nucleotide in the first nucleic acid strand. Although the presence of MP linkages is compatible with RNA editing by human ADAR enzymes, introducing MP linkages during the manufacturing of oligonucleotides is challenging in view of additional manufacturing (purification) steps in the coupling and decoupling process. In one aspect, the AON does not comprise an MP linkage.

[0147] In one aspect, the AON as disclosed herein comprises at least one PNdmi linkage, preferably linking the most terminal two nucleosides at the 5’ and / or 3’ end of the AON. A PNdmi linkage as preferably used in the AONs as disclosed herein has the structure of formula (IV):O iN-P=ON— ( .Z .N-—PNdmi(|V)linkage

[0148] In one aspect, at either end or both termini of an AON as disclosed herein, inverted deoxyT or dideoxyT nucleotides are incorporated. Other internucleoside linkages that may be used in the AONs as disclosed herein are those that are disclosed in WO2023 / 278589.

[0149] In one aspect, the AON as disclosed herein comprises at least one phosphonoacetate and / or at least one phosphonoacetamide internucleoside linkage.

[0150] Conjugate chemistries

[0151] In one aspect, the AON as disclosed herein, or the sense strand to which it may be annealed before entering a target cell (in an HEON as disclosed herein), is bound to a hydrophobic moiety, such as palmityl or an analog thereof, cholesterol or analog thereof, or tocopherol or analog thereof. It is preferably bound to the 5’ terminus. In case a hydrophobic moiety is bound to the 5’ terminus as well as to the 3’ terminus, such hydrophobic moieties may the same or different. The hydrophobic moiety bound to the oligonucleotide may be bound directly or indirectly mediated by another substance. When the hydrophobic moiety is bounddirectly, it is sufficient if the moiety is bound via a covalent bond, an ionic bond, a hydrogen bond, or the like. When the hydrophobic moiety is bound indirectly, it may be bound via a linking group (a linker). The linker may be a cleavable or an uncleavable linker.

[0152] A cleavable linker refers to a linker that can be cleaved under physiological conditions, for example, in a cell or an animal body (e.g., a human body). A cleavable linker is selectively cleaved by an endogenous enzyme such as a nuclease, or by physiological circumstances specific to parts of the body or cell, such as pH or reducing environment (such as glutathione concentrations). Examples of a cleavable linker comprise, but is not limited to, an amide, an ester, one or both esters of a phosphodiester, a phosphoester, a carbamate, and a disulfide bond, as well as a natural DNA linker. Cleavable linkers also include self-immolative linkers.

[0153] An uncleavable linker refers to a linker that is not cleaved under physiological conditions, or very slowly compared to a cleavable linker, for example, in a PS linkage, modified or unmodified deoxyribonucleosides linked by a PS linkage, a spacer connected through a PS bond and a linker consisting of modified or unmodified ribonucleosides. There is no restriction on the chain length, when a linker is a nucleic acid such as DNA, or an oligonucleotide. However, it may be usually from 2 to 20 bases in length, from 3 to 10 bases in length, or from 4 to 6 bases in length. There is no restriction on the length or composition of a spacer that is connects the ligand and the oligonucleotide, and may include for example ethylene glycol, triethylene glycol (TEG), HEG, alkyl chains, propyl, 6-aminohexyl, or dodecyl.

[0154] One or more other types of molecules may be bound to the AON through one or more linkers, including peptides, sugars, vitamins, polymers, aptamers, (fragments of) antibodies, small molecules, and the like.

[0155] General

[0156] In addition to the specific preferred chemical modifications at certain positions in compounds as disclosed herein, AONs as disclosed herein may comprise one or more (additional) modifications to the nucleobase, scaffold and / or backbone linkage, which may or may not be present in the same monomer, for instance at the 3’ and / or 5’ position. In one aspect, the AON as disclosed herein comprises at least one internucleoside linkage according to the structure of formula (I), (II), (III), and / or (IV), and / or the AON further comprises at least one nucleotide with a sugar moiety that comprises a 2’-OMe modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’-MOE modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’-F modification, and / or the AON comprises an orphan nucleotide that carries a 2’-H in the sugar moiety and is therefore referred to as a DNA nucleotide, even though additional modifications may exist in its base and / or linkage to its neighbouring nucleosides. In one aspect, the orphan nucleotide carries a 2’-F in the sugar moiety. In one aspect, the orphan nucleotide carries a di F substitution in the sugar moiety. In one aspect, the orphan nucleotide carries a 2’-F and a 2’-C-methyl in thesugar moiety. In one aspect, the orphan nucleotide comprises a 2’-F in the arabinose configuration (FANA) in the sugar moiety.

[0157] In one aspect, the AON is an antisense oligonucleotide that can form a double stranded nucleic acid complex with a target RNA molecule, wherein the double stranded nucleic acid complex can recruit an adenosine deaminating enzyme for deamination of a target adenosine in the target RNA molecule, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, and wherein the orphan nucleotide has the structure of formula (V):

[0158] wherein: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from the group consisting of: cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxo- adenine, and 6-amino-5-nitro-3-yl-2(1 H)-pyridone; R1 and R2 are both selected, independently, from H, OH, F or CH3; R3 is the part of the AON that is 5’ of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4 is the part of the AON that is 3’ of the orphan nucleotide, consisting of 4 to 25 nucleotides. The nucleotide 3’ and / or 5’ from the orphan nucleotide may be DNA, more preferably the nucleotide at the 3’ (position -1).

[0159] Other chemical modifications of the AON as disclosed herein include the substitution of one or more than one of any of the hydrogen atoms with deuterium or tritium, examples of which can be found in e.g., WO2014 / 022566 or WO2015 / 011694. Again, in all cases, the modifications should be compatible with editing such that the AON fulfils its role as an oligonucleotide that can, after binding to its target sequence, recruit an adenosine deaminase enzyme because of the double-stranded nucleic acid entity that arises. In all aspects of the disclosure, the enzyme with adenosine deaminase activity is preferably ADAR1 , ADAR2, or ADAT.

[0160] AONs as disclosed herein preferably do not include a 5’-terminal O6-benzylguanosine or a 5’-terminal amino modification and preferably are not covalently linked to a SNAP-tag domain (an engineered O6-alkylguanosine-DNA-alkyl transferase). An AON as disclosed herein preferably does not comprise a boxB RNA hairpin sequence. In one aspect, an AON as disclosed herein comprises 0, 1 , 2 or 3 wobble base pairs with the target sequence, and / or 0, 1 , 2, 3, 4, 5, 6, 7, or 8 mismatching base pairs with the target RNA sequence. No mismatch exists when the orphan nucleotide is uridine, which may be defined differently when the orphan nucleotide is a uridine analog or derivative. One alternative for uridine is positioning an isouridine opposite the target adenosine, which likely does not pair like G pairs with II. Preferably,the target adenosine in the target sequence forms a mismatch base pair with the nucleoside in the AON that is directly opposite the target adenosine.

[0161] As outlined above, an AON as disclosed herein makes use of specific nucleotide modifications at predefined spots to ensure stability as well as proper ADAR binding and activity. These changes may vary and may include modifications in the backbone of the AON, in the sugar moiety of the nucleotides as well as in the nucleobases or the phosphodiester linkages, as outlined in detail herein. They may also be variably distributed throughout the sequence of the AON. Specific modifications may be needed to support interactions of different amino acid residues within the RNA-binding domains of ADAR enzymes, as well as those in the deaminase domain. For example, PS linkages between nucleotides or2’-OMe or2’-MOE modifications may be tolerated in some parts of the AON, while in other parts they should be avoided so as not to disrupt crucial interactions of the enzyme with the phosphate and 2’-OH groups. Specific nucleotide modifications may also be necessary to enhance the editing activity on substrate RNAs where the target sequence is not optimal for ADAR editing. Previous work has established that certain sequence contexts are more amenable to editing. For example, a target sequence 5’-UAG-3’ (with the target A in the middle) contains the most preferred nearest- neighbor nucleotides for ADAR2, whereas a 5’-CAA-3’ target sequence is disfavored (Schneider et al. 2014. Nucleic Acids Res 42(10):e87). The structural analysis of ADAR2 deaminase domain hints at the possibility of enhancing editing by careful selection of the nucleotides that are opposite to the target trinucleotide. For example, the 5’-CAA-3’ target sequence, paired to a 3’-GCU-5’ sequence on the opposing strand (with the A-C mismatch formed in the middle), is disfavored because the guanosine base sterically clashes with an amino acid side chain of ADAR2. The guanosine opposite the C in such circumstances is preferably replaced by an inosine (hence, at the -1 position within the AON), preferably a deoxyinosine.

[0162] The AONs as disclosed herein may also be administered in the context of aids that will increase the entry of the AON into the target cell and / or its endosomal escape as soon as it is in the cell. Moieties that can be applied for such applications are for example a set of chemical compounds (generally purified from nature) referred to as “saponins” or “triterpene glycosides”. A preferred saponin that can be used in the methods as disclosed herein is AG 1856, disclosed in WO2021 / 122998 and further described for use with RNA editing producing oligonucleotides in WO2024 / 153801.

[0163] In one embodiment, the AON as disclosed herein comprises a delivery moiety. Preferably, for delivery to liver cells, the delivery moiety is a GalNAc moiety. Preferred GalNAc moieties are tri-antennary GalNAc moieties as shown in Formula VI, VII, VII 1(a) and VII 1(b), and IX (below). Preferred mono-antennary GalNAc moieties are as shown in Formula XVII and XVIII (below).

[0164] Formula VI is as follows and includes connection point E which conjugates to another moiety such as a linker, a saponin or an AON (see WO2022 / 271806):

[0165] Formula VII is as follows and includes a linker and a connection point F which conjugates to another moiety such as a linker, a saponin or an AON (see WO2014 / 179620):

[0166] Formula VII 1(a) and VII 1(b) include a linker and a connection point which conjugates to another moiety such as a linker, a saponin or an AON (see W02009 / 073809):(Vlll(a))(Vlll(b))

[0167] Formula IX (see WO2011 / 104169), wherein the squiggly line indicates a connection point, optionally via linker and / or a spacer, to the AON:(X)(XI)

[0168] When the GalNAc moiety of Formula VI is applied, the GalNAc moiety is preferably conjugated to the AON at its 3’ terminus, via connection point E of Formula VI, optionally via a linker and / or a spacer, for example as depicted in Formula XII (below).(XII)

[0169] When the GalNAc moiety of Formula VII is applied, the GalNAc moiety is preferably conjugated to the AON at its 5’ terminus, via connection point F of Formula VII, optionally via a linker and / or a spacer.

[0170] The base sequence of the AON herein is complementary to part of the base sequence of a target ALG12 transcription product that includes at least a target adenosine (preferably the adenosine at position 823) that is to be deaminated to an inosine and therefore can anneal (or hybridize) to the target transcription product. The complementarity of a base sequence can be determined by using a BLAST program or the like. Those skilled in the art can easily determine the conditions (temperature, salt concentration, and the like) under which two strands can be hybridized, taking into consideration the complementarity between the strands.

[0171] The AON according to the present disclosure, in contrast to what has been described for gapmers and their relation towards RNase breakdown and the use of such gapmers in doublestranded complexes (see for instance EP 3954395 A1), does not comprise a stretch of DNA nucleotides which would make a target sequence (or a sense nucleic acid strand) a target for RNase-mediated breakdown. In one embodiment, the AON does not comprise four or more consecutive DNA nucleotides anywhere within its sequence. In an embodiment, the AON is composed of as much (chemically) modified nucleotides as possible to enhance the resistance towards RNase-mediated breakdown, while at the same time being as efficient as possible in producing an RNA editing effect. This means that the orphan nucleotide and several othernucleotides within the AON may be DNA, but also that there is no stretch of four or more consecutive DNA nucleotides within the AON. Hence, the AON according to the present disclosure is not a gapmer. A gapmer is in principle a single-stranded nucleic acid consisting of a central region (DNA gap region with at least four consecutive deoxyribonucleotides) and wing regions positioned directly at the 5’ end (5’ wing region) and the 3’ end (3’ wing region) thereof. In contrast, the AON according to the disclosure may be any oligonucleotide that produces an RNA editing effect in which a target adenosine in a target RNA molecule is deaminated to an inosine and accordingly is resistant to RNase-mediated breakdown as much as possible to yield this effect.

[0172] It should be noted that when an AON is delivered through a vector, for instance an AAV vector, chemical modifications are not present in the AON that acts on the target RNA molecule. Although it is preferred to use ‘naked’ AONs that have chemical modifications as outlined herein, AONs that are delivered through other means, for instance through AAV vector expression, or editing molecules that are circular, or have hairpin structures (recruiting portions, e.g., as disclosed in WO2016 / 097212, WO2017 / 050306, W02020 / 001793, WO2017 / 010556, W02020 / 246560, and WO2022 / 078995) are also encompassed by the present disclosure because these can also be applied to edit adenosines in the target ALG12 RNA molecule to generate a ALG12 protein with reduced function. Notably, when the AON comprises chemical modifications, as detailed herein, it may still be delivered through the means of 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 LN P’s that are nano-sized lipid vesicles that carry the AON of the present disclosure and aid to the delivery of target cells. If an LNP is applied or any other similar type of carrier, the AON is still considered naked because it is not transcribed from an encoding polynucleotide (such as in the case of a plasmid or a vector, in which the AON is not regarded as ‘naked’). So, even though a chemically modified AON is encapsulated by a carrier, preferably an LNP, it is still seen as naked, as it has been manufactured as such in a laboratory setting and encapsulated thereafter in the carrier using methods known to the person skilled in the art. The disclosure also relates to a delivery vehicle, preferably an LNP, which comprises a ‘naked’ and chemically modified AON as disclosed herein, even more preferably as disclosed in any one of SEQ ID NO:2 to 51. The person skilled in the art understands that when a delivery moiety, or attachment to the AON is used (such a GalNAc moiety to target hepatocytes in the liver) that the AON is still seen as naked as well, also when a GalNAc-AON is encapsulated in a delivery vehicle such as an LNP. The same holds true when an AON is bound to a saponin as further outlined in detail herein.

[0173] The disclosure also provides a pharmaceutical composition comprising the AON according to the disclosure and further comprising a pharmaceutically acceptable carrier and / or other additive and may be dissolved in a pharmaceutically acceptable organic solvent, or thelike. Dosage forms in which the AON or the pharmaceutical composition are administered may depend on the disorder to be treated and the tissue that needs to be targeted and can be selected according to common procedures in the art. The pharmaceutical compositions may be administered by a single-dose administration or by multiple dose administration. It may be administered daily or at appropriate time intervals, which may be determined using common general knowledge in the field and may be adjusted based on the disorder and the efficacy of the active ingredient.

[0174] An AON according to the present disclosure can utilise endogenous cellular pathways and naturally available ADAR enzymes to specifically edit a target adenosine in the target RNA sequence. An AON of the disclosure is capable of recruiting ADAR and complex with it and then facilitates the deamination of a (single) specific target adenosine nucleotide in a target RNA sequence. Ideally, only one adenosine is deaminated. An AON of the disclosure, when complexed to ADAR, preferably brings about the deamination of a single target adenosine.

[0175] Analysis of natural targets of ADAR enzymes has indicated that these generally include mismatches between the two strands that form the RNA helix edited by ADAR1 or 2. It has been suggested 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 optimal patterns of paired / mismatched nucleotides between the AONs and the target RNA also appears important to the development of efficient ADAR-based AON therapy.

[0176] In one embodiment, a nucleotide analogue or equivalent within the AON comprises one or more base modifications or substitutions. Modified bases comprise 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 that are or will be known in the art. Purine nucleobases and / or pyrimidine nucleobases may be modified to alter their properties, for example by amination or deamination of the heterocyclic rings. The exact chemistry and formats may vary from oligonucleotide construct to oligonucleotide construct and from application to application, and may be worked out in accordance with the wishes and preferences of those of skill in the art.

[0177] An AON according to the disclosure is normally longer than 10 nucleotides, preferably more than 11 , 12, 13, 14, 15, 16, still more preferably more than 17 nucleotides. In one aspect the AON according to the disclosure is longer than 20 nucleotides. The AON according to the disclosure is preferably shorter than 100 nucleotides, still more preferably shorter than 60 nucleotides, still more preferably shorter than 50 nucleotides. In a preferred aspect, the AON according to the disclosure comprises 18 to 70 nucleotides, more preferably comprises 18 to 60 nucleotides, and even more preferably comprises 18 to 50 nucleotides. Hence, in a particularly preferred aspect, the AON of the present disclosure comprises 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44,45, 46, 47, 48, 49 or 50 nucleotides. In one embodiment, the AON is 27, 28, 29, or 30 nucleotides in length.

[0178] EXAMPLES

[0179] Example 1. RNA editing of the wildtype human ALG12 transcript in human Huh7 cells using a variety of AONs.

[0180] An initial set of 50 AONs was designed to target the adenosine in the AGO codon encoding serine at position 275 in the human wildtype ALG12 protein (Fig. 1). The design and chemical modifications of these AONs are also provided in Fig. 1.

[0181] In the initial screen of 30 AONs from the initially designed 50 AONs, human Huh7 cells were transfected with 100 nM AON using Lipofectamine 2000, as follows. On day 0, Huh7 cells with a density of 2.0x104cells / well were transfected with AONs using Lipofectamine®RNAiMAX Reagent at the same time of seeding. For this, Lipofectamine Reagent in Opti-MEM® Medium was made with a ratio of 97:3 (Opti-MEM: Lipofectamine Reagent) and incubated at RT for 15 min and gently vortexed. Then the AONs were diluted with Nuclease-Free Water to make 12x stock solutions. Then, 50 pl of the Opti- MEM / RNAiMAX complexes were administered into each well of a culture plate and 50 pl 12x AON stock solution was added. Cells were then treated with trypsin, centrifuged, and resuspended in 2% FBS medium, and cell density was adjusted to 200,000 cells / ml. Then 20 pl of RNAiMAX-AON complex was added to the cell plate plus 100 pl cell suspension. This mix was incubated at 37°C, 5% CO2 for 72 hr. Media was changed at 24 hr after plating / start of transfection.

[0182] After 72 h of incubation, the medium was removed, and total RNA was isolated using the RNeasy Micro kit (Qiagen). cDNA was synthesized using Maxima Reverse T ranscriptase kit using a mixture of random hexamers and oligo(dT) primers.

[0183] The percentage of ADAR-mediated A-to-l conversion was determined by quantitative digital droplet PCR (ddPCR) assay with the Digital PCR System of Bio-Rad (QX200) in 22 pl aliquots of reaction mixtures containing cDNA, appropriate pairs of primers and ddPCR™ Supermix for Probes (no dUTP) (Bio-Rad-1863024). The primers and probes used for this assay are provided in Table 1. The percentage was calculated by dividing the guanidine containing cDNA species (= related to the GGC codon that occurs after editing) by the total number of target copies, multiplied by 100. A control dPCR was performed upstream or downstream of the transcript for standardization purposes (not shown).

[0184] Table 1 : Primer and probe sequences for quantitative dPCR assays. The SEQ ID NO is given. The + is an LNA modification of the nucleotide following the + symbol.

[0185] The results for the 30 tested AONs are provided in Fig. 2 (names of the AONs are given below the graph). Although some AONs provided very low editing percentages, some AONs did yield very high levels of RNA editing, going up to 50% in some cases. AONs that performed particularly good were RM117396 (SEQ ID NO:7), RM117395 (SEQ ID NO:6), RM1 17397 (SEQ ID NO:8), RM117394 (SEQ ID NO:5), RM117399 (SEQ ID NQ:10),RM1 17392 (SEQ ID NO:3), RM117436 (SEQ ID NO:47), RM117393 (SEQ ID NO:4),RM1 17427 (SEQ ID NO:38), RM117429 (SEQ ID NQ:40), RM117424 (SEQ ID NO:35),RM1 17404 (SEQ ID NO:15), RM117426 (SEQ ID NO:37), RM117407 (SEQ ID NO:18),RM1 17408 (SEQ ID NO:19), RM117400 (SEQ ID NO:11), and RM117406 (SEQ ID NO:17).

[0186] In a follow-up experiment, AONs are tested for RNA editing after gymnotic uptake (no transfection reagents), in the context of AG1856 (either co-administered, administered separately or in a pulse-treatment), and attached to GalNAc and / or AG1856, followed by further optimization of the best performing AONs (chemical modifications, symmetry, etc.) and then followed by assessments in liver organoids and in in vivo experiments.

Claims

CLAIMS1 . An antisense oligonucleotide (AON) that can form a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell, wherein the double-stranded complex can recruit an endogenous ADAR enzyme naturally present in the cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine, wherein the target RNA nucleic acid molecule is a transcript molecule encoding the human Asparagine-Linked Glycosylation 12 homolog (ALG12) glycosyltransferase.

2. An AON according to claim 1 , wherein the deamination results in a change from serine to glycine at position 275 (S275G) in the human ALG12 amino acid sequence.

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

4. An AON according to any one of claims 1 to 3, wherein the cell is a human liver cell, preferably a hepatocyte.

5. An AON according to any one of claims 1 to 4, wherein at least one nucleotide comprises one or more non-naturally occurring chemical modifications, or one or more additional non-naturally occurring chemical modifications, in the ribose, linkage, or base moiety.

6. An AON according to claim 5, wherein the orphan nucleotide is a cytidine, a cytidine analog, a uridine, or a uridine analog, preferably wherein the orphan nucleotide is a deoxynucleotide.

7. An AON according to claim 6, wherein the orphan nucleotide is a deoxynucleotide carrying a 6-amino-5-nitro-3-yl-2(1 H)-pyridone nucleobase.

8. An AON according to any one of claims 1 to 7, wherein the nucleotide numbering in the AON is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end, and wherein the nucleotide at position -1 is a deoxyinosine.

9. An AON according to any one of claims 1 to 8, wherein the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’-end and negatively (-) incremented towards the 3’-end, and wherein linkage position -2 is a methylphosphonate (MP) or a mesyl phosphoramidate (PNms) linkage.

10. An AON according to any one of claims 1 to 9, wherein the AON comprises or consists of the sequence of any one of SEQ ID NO:2 to 51 , preferably selected from the group consisting of SEQ ID NO:7, 6, 8, 5, 10, 3, 47, 4, 38, 40, 35, 15, 37, 18, 19, 11 , and 17.

11. An AON according to any one of claims 1 to 10, wherein the AON is directly or indirectly bound, preferably conjugated, to a cell-targeting moiety and / or an endosomal release moiety.

12. An AON according to claim 11 , wherein the cell-targeting moiety is a GalNAc molecule, preferably a tri-antennary GalNAc molecule.

13. An AON according to claim 11 or 12, wherein the endosomal release moiety is a triterpene glycoside, such as a saponin, preferably AG 1856.

14. A vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON according to any one of claims 1 to 4.

15. A nanoparticle delivery vehicle formulation comprising an AON according to any one of claims 1 to 13.

16. A nanoparticle delivery vehicle formulation according to claim 15, wherein the nanoparticle delivery vehicle is a Lipid Nanoparticle (LNP).

17. A pharmaceutical composition comprising an AON according to any one of claims 1 to 13, a vector according to claim 14, or a nanoparticle delivery vehicle formulation according to claim 15 or 16, and a pharmaceutically acceptable carrier.

18. An AON according to claim 1 to 13, a vector according to claim 14, a nanoparticle delivery vehicle formulation according to claim 15 or 16, or a pharmaceutical composition according to claim 17, for use in the treatment of cardiovascular disease (OVD).

19. Use of an AON according to claim 1 to 13, a vector according to claim 14, a nanoparticle delivery vehicle formulation according to claim 15 or 16, or a pharmaceutical composition according to claim 17, in the manufacture of a medicament for the treatment of OVD.

20. An in vitro, ex vivo, or in vivo method of editing a human ALG12 polynucleotide, the method comprising contacting the ALG12 polynucleotide with an AON according to any one of claims 1 to 13, thereby editing the ALG12 polynucleotide.

21. A method of treating, slowing down, preventing, or ameliorating CVD in a patient in need thereof, the method comprising contacting an ALG12 polynucleotide in a cell of the subject with an AON according to any one of claims 1 to 13, thereby treating the patient.

22. A method for the deamination of a target adenosine in a human ALG12 pre-mRNA or mRNA molecule in a cell, wherein the target adenosine is at position 823 of the human ALG12 mRNA sequence, the method comprising the steps of:(i) providing the cell with an AON according to any one of claims 1 to 13;(ii) allowing uptake by the cell of the AON;(iii) allowing annealing of the AON to the ALG12 pre-mRNA or mRNA molecule;(iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine; and optionally(v) identifying the presence of the inosine in the target RNA molecule.

23. A method according to claim 22, wherein step (v) comprises: a) determining the sequence of the ALG12 pre-mRNA or mRNA molecule; b) assessing the presence of an ALG12 protein variant with a glycine at position 275 in the amino acid sequence; or c) using a functional read-out, preferably assessing a reduced plasma concentration of LDL-C.

Images