Antisense oligonucleotides for the treatment of classic galactosemia

Abstract

The present invention relates to antisense oligonucleotides (AONs) that can mediate RNA editing by binding to a target RNA nucleic acid molecule, preferably an RNA transcript molecule, in a cell and recruiting an endogenous deaminating enzyme in the cell to deaminate one or more target adenosine nucleotides in the target RNA molecule to an inosine. The target RNA molecule is a transcript molecule from the human galactokinase (GALK1) gene that preferably encodes a wildtype GALK1 protein. The RNA editing of the target adenosine, changing an aspartic acid (D) residue to a glycine (G) residue at position 186 in the protein, results in a GALK1 protein with a loss-of-function in its ability to phosphorylate galactose, thereby reducing the amount of galactose-1-phosphate (Gal-1-P), which is one of the major accumulating metabolites in classic galactosemia caused by a galactose-1-phosphate uridylyltransferase (GALT) deficiency, providing a potential treatment for classic galactosemia.

Description

[0001] P5117PC00 ANTISENSE OLIGONUCLEOTIDES FOR THE TREATMENT OF CLASSIC GALACTOSEMIA TECHNICAL FIELD This disclosure relates to the field of medicine, particularly to the field of galactosemia,and more particularly to classic galactosemia, an autosomal recessive disorder caused by thedeficiency of galactose-1-phosphate uridylyltransferase (GALT). The disclosure describes antisense oligonucleotides that mediate nucleotide-specific RNA editing in transcripts of the gene encoding galactokinase (GALK1), another protein involved in the Leloir pathway of galactose metabolism, to bring about amino acid changes of the encoded protein thatinfluence its activity.BACKGROUND Galactose is the C-4 epimer of glucose, with an identical molecular formula, but a distinct structural formula. Despite its strong structural similarity to glucose, the conversionfrom galactose into glucose requires a few evolutionary-conserved enzymatic steps, allresiding in the cytoplasm, known as the ‘Leloir’ pathway of galactose metabolism. The main source of galactose in humans is dietary, mainly from dairy products containing lactose, but other non-dairy foodstuffs can also contain galactose moieties. In humans, galactose can also be produced endogenously, mostly through the enzymatic conversion between uridinediphosphate-glucose (UDP-glucose) and uridine diphosphate-galactose (UDP-Gal), as wellas the turnover of glycoprotein and glycolipids. Upon entry into the Leloir pathway, galactose is first phosphorylated by galactokinase (GALK1) to form galactose-1-phosphate (Gal-1-P). Together with the second substrate UDP-glucose, Gal-1-P is converted by galactose-1- phosphate uridylyltransferase (GALT) to form UDP-galactose and glucose-1-phosphate (Glc-1-P). The Leloir pathway is completed by reversibly forming UDP-glucose from UDP-galactoseby UDP-galactose-4-epimerase (GALE) and the formation of glucose from Glc-1-P through gluconeogenesis. Galactosemias are a rare group of hereditary disorders of galactose metabolism. To date, four types have been described, each affecting a different step in the main route ofgalactose disposal: type I: (classic) galactosemia caused by a GALT-deficiency; type II:GALK1-deficiency; type III: GALE-deficiency; and type IV: galactose mutarotase (GALM)- deficiency. In type I, generally referred to as classic galactosemia, the conversion of Gal-1-P to Glc-1-P and UDP-Gal is hampered by the severe GALT deficiency. It is an autosomal recessive disorder with a prevalence of 1:16,000 to 1:50,000 live births in Western countries.Most affected patients develop brain impairments (85.0%), primary ovarian insufficiency(79.9%) and a diminished bone mineral density (26.5%). Additionally, data showed that a more P5117PC00 favorable outcome among patients is achieved by onset of a galactose-restricted diet in the first week of life and detection by newborn screening. Infants born with classic galactosemia usually become ill within days after birth if exposed to breast milk or lactose-containing formula. Initially, the infant develops jaundice, and if lactose exposure continues,complications such as liver failure, Escherichia coli (E. coli) sepsis, coma, and death followshortly after. Many countries in the Western world have included classic galactosemia as one of the conditions screened for in the newborn period, ensuring that most infants survive without becoming ill. The diet (for instance by replacing lactose / galactose using soy-based formulas) resolves the neonatal syndrome but fails to prevent burdensome chronic impairments,because most patients with classic galactosemia, despite early screening and dietaryrestrictions, continue to accumulate significant amounts of galactose, galactitol (formed from galactose through the enzyme aldose reductase) and Gal-1-P in their cells. The most common complications are speech dyspraxia, ataxia, and premature ovarian insufficiency. To date, the pathophysiology of the acute toxicity syndrome and the chronic complications remains largelyunknown, but it is reasonable to assume that any blockage in a metabolic pathway will lead to(i) accumulating precursor(s), (ii) alternate metabolites normally not encountered, or (iii) absent metabolites past the enzymatic block. Any, or a combination of these possibilities, could be responsible for the phenotypes associated with the enzymatic blockage. As to classic galactosemia, it is apparent that galactose and Gal-1-P accumulate in patients, with galactosebeing further metabolized through alternative pathways to form galactitol and galactonate. Inany case, because patients with classic galactosemia experience such very severe complications, despite dietary restrictions, there is a clear and unmet need for more adequate treatment to prevent complications. Based on important lines of evidence that strongly suggest that Gal-1-P and / orgalactitol are the most important and toxic metabolites for developing long-term complicationsin classic galactosemia, it was proposed that pharmacological inhibition of GALK1, in conjunction with dietary galactose / lactose restriction, may provide a treatment for the disorder by preventing Gal-1-P and / or galactitol accumulation (Bosch AM.2006. J Inherit Metab Dis 29(4):516-525; Timson DJ.2007. Current Enzyme Inhibition 3:77-94; Wiernega KJ et al.2008.J Biomol Screen 13(5):415-423; Tang M et al. 2012. Mol Genet Metab 105(1):44-55; DelnoyB et al. 2021. J Pers Med 11:75). Several suggestions and attempts have been made to develop treatments for classic galactosemia, such as the development of a compound referred to as govorestat, which is an aldose reductase inhibitor, currently in clinical phase, directed at preventing the accumulation of galactitol, which is produced from galactose through theconversion by the aldose reductase enzyme. Other approaches include siRNA to down-regulate expression of GALK1, exon-skipping during splicing of the pre-mRNA of GALT thereby removing exons containing mutations, using a small molecule approach, using aldose P5117PC00 reductase inhibitors, using Salubrinal to stimulate cellular stress response, and pre-clinical work using mRNA encoding wildtype GALT (Lai K et al.2014. Future Med Chem 6(9):1003- 1015; Chiappori F et al.2013. Eur J Med Chem 63:423-434; Tang et al.2012; Balakrishnan B et al.2019. Mol Ther 28(1):304-312; WO2019 / 023648; WO2018 / 232317; WO2017 / 201348).None of these have thus far led to an available medicament to treat classic galactosemiapatients in need. The present disclosure relates to a completely different approach of targeting GALK1 and through the methods and means disclosed herein, inhibit its activity. The goal herein is to provide a GALK1 protein with a loss-of-function (LOF), which will prevent accumulation of Gal-1-P, which as outlined above is together with galactitol among the most toxic metabolites inthe occurrence of classic galactosemia, caused by GALT-deficiency. The disclosure relates to targeting the transcript of human wildtype GALK1, namely by using antisense oligonucleotides (AONs) and the cell’s own nucleic acid post-transcriptional modification machinery to specifically amend a nucleotide (deamination of an adenosine) in the GALK1 (pre-) mRNAtranscript. The present disclosure relates to changing a codon in the (pre-) mRNA of thewildtype transcript of human GALK1 to a codon encoding a different amino acid, thereby removing (or changing) an active site residue involved in the phosphorylation of galactose. This disturbance leads to a lower / diminished / absent ability of the LOF protein towards accumulating Gal-1-P, which in fact, as outlined above, is the result from a GALT-deficiency.Hence, the present disclosure relates to amending the wildtype transcript of human GALK1 toa transcript encoding a LOF mutant of GALK1, and thereby treat classic galactosemia caused by the deficiency in the downstream enzyme GALT. The technology that the present disclosure relates to is generally referred to as ‘RNA editing’. RNA editing is a natural process through which eukaryotic cells alter the sequence oftheir RNA molecules, often in a site-specific and precise way, thereby increasing the repertoireof 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 areadenosine (A) to inosine (I) conversions and cytidine (C) to uridine (U) conversions, whichoccur through enzymes called Adenosine Deaminases acting on RNA (ADAR) and Apolipoprotein B mRNA Editing Enzyme, Catalytic Polypeptide / Activation-Induced Cytidine Deaminase (APOBEC / AID), respectively. ADAR is a multi-domain protein, comprising of a catalytic domain and two to threedouble-stranded RNA (dsRNA) recognition domains, depending on the enzyme in question.Each recognition domain recognizes a specific dsRNA sequence and / or conformation. The P5117PC00 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 an 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 isin a coding region of an mRNA or pre-mRNA, it can recode the amino acid sequence. A-to-Iconversions may also occur in the 5’ untranslated region (UTR) 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 placein splice elements in introns or exons in pre-mRNAs, thereby altering the pattern of splicing.As a result, exons may be (partially) included or excluded. The enzymes catalysing adenosine deamination are within an enzyme family of ADARs, which include human deaminases hADAR1 and hADAR2, as well as hADAR3. However, to date, for hADAR3 no deaminase activity has been demonstrated. The use of oligonucleotides to edit a target RNA, applying adenosine deaminase, hasbeen described (e.g., Woolf et al. 1995. Proc Natl Acad Sci USA 92:8298-8302; Montiel- Gonzalez et al. Proc Natl Acad Sci USA 2013, 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), is the need for a fusion protein consisting of the boxB recognitiondomain of bacteriophage Lambda N-protein, fused to the adenosine deaminase domain of atruncated 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) suffers from similar drawbacks, in that it is not clear how toapply the system without having to genetically modify the ADAR first and subsequentlytransfect 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 RNA oligonucleotides of Woolf et al. (1995) that were 100% complementary to the target RNA sequences, suffered from severe lack of specificity: nearly all adenosines in the target RNAstrand that were complementary to the AON were edited.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) showed that such off-target editing can be suppressed by using 2’-O-methyl (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 P5117PC00 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 whilemaintaining a specificity in which a single adenosine within a target RNA molecule can betargeted 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 / hairpin structure (therein referred to as the ‘recruitment portion’), which ispreferably non-complementary to the target RNA. Such oligonucleotides are referred to as‘self-looping AONs’. The recruitment portion is thought to act in recruiting a natural ADAR enzyme present in the cell (endogenously present) 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 mimickingeither 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 nucleicacid strand. The stem-loop structure of the recruitment portion as described is anintramolecular 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 since then been described in WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, US11,390,865, WO2020 / 246560, and WO2022 / 078995. WO2017 / 220751 and WO2018 / 041973 describe a next generation type of AONs thatdo not comprise such a stem-loop structure but that are (almost fully) complementary to the targeted area, and that appeared still capable of attracting endogenous ADAR enzymes. 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 nucleosideopposite the target adenosine, but in other embodiments AONs (or “RNA editingoligonucleotides” – even though the deamination reaction is carried out by the ADAR enzyme – and 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 ADARenzymes when the sequence of the AON was carefully selected such that it couldattract / recruit ADAR. The ‘orphan nucleotide’, which is defined as the nucleotide in the AON that is positioned directly opposite the target adenosine in the target RNA molecule, was a P5117PC00 nucleotide with an unmodified cytosine nucleobase and that did not carry a 2’-OMe modification. The orphan nucleotide can be a deoxyribonucleotide (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 nucleotide contained chemicalmodifications (such as DNA in comparison to RNA) that further improved the RNA editingefficiency and / or increased the resistance against nucleases. Such effects could even be further improved by using sense oligonucleotides (SONs) that ‘protected’ the AONs against breakdown upon delivery to the cells (described in WO2018 / 134301 and US11,274,300). The use of chemical modifications and particular structures in oligonucleotides thatcould be used in ADAR-mediated editing of specific adenosines in a target RNA have beenthe subject of numerous disclosures in the field, such as WO2019 / 111957, WO2019 / 158475, WO2020 / 165077, WO2020 / 201406, WO2020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, WO2022 / 007803, WO2022 / 018207,WO2022 / 026928, and WO2022 / 124345. The use of specific sugar moieties has beendisclosed in for instance WO2020 / 154342, WO2020 / 154343, WO2020 / 154344, WO2022 / 103839, and WO2022 / 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 targetsequences) has been described in WO2011 / 005761, WO2014 / 010250, WO2014 / 012081,WO2015 / 107425, WO2017 / 015575 (for HTT), WO2017 / 062862, WO2017 / 160741, WO2017 / 192664, WO2017 / 192679 (for DMD), WO2017 / 198775, WO2017 / 210647, WO2018 / 067973, WO2018 / 098264, WO2018 / 223073 (for APOC3), WO2018 / 237194, WO2019 / 032607 (for C9orf72), WO2019 / 055951, WO2019 / 075357 (for SMA / ALS),WO2019 / 200185 (for DM1), WO2019 / 217784 (for DM1), WO2019 / 219581, WO2020 / 118246(for DM1), WO2020 / 160336 (for HTT), WO2020 / 191252, WO2020 / 196662, WO2020 / 219981 (for USH2A), WO2020 / 219983 (for RHO), WO2020 / 227691 (for C9orf72), WO2021 / 071788 (for C9orf72), WO2021 / 071858, WO2021 / 178237 (for MAPT), WO2021 / 234459, WO2021 / 237223, WO2022 / 099159, WO2021 / 030778, WO2022 / 174053, andWO2023 / 278589. Next to these disclosures, an extensive number of publications relate to thetargeting 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 WO2020 / 157008 and WO2021 / 136404 (USH2A);WO2021 / 113270 (APP); WO2021 / 113390 (CMT1A); WO2021 / 209010 (IDUA, Hurlersyndrome); WO2021 / 231673 and WO2021 / 242903 (LRRK2); WO2021 / 231675 (ASS1); WO2021 / 231679 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 P5117PC00 and WO2021 / 231692 (OTOF, autosomal recessive non-syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); WO2021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1). The present disclosure provides one or more alternative, and / or improved, compoundsor compositions for use in the treatment of classic galactosemia, by generating a LOF mutantof human GALK1 through specifically targeting a single adenosine in the human (pre-) mRNA transcript encoding the protein, applying endogenously present ADAR enzymes in the cell. SUMMARY OF THE INVENTION Disclosed herein is an antisense oligonucleotide (AON) that is capable of forming adouble-stranded complex with a region of a target RNA nucleic acid molecule in a human cell and recruiting an endogenous (= naturally present) ADAR enzyme 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 thetarget adenosine into an inosine, and wherein the target RNA nucleic acid molecule is atranscript molecule of the human galactokinase 1 (GALK1) gene, that encodes the GALK1 protein. Preferably, the human cell is a liver cell. In one aspect, the human GALK1 gene is according to NCBI Gene ID No. 2584 (NCBI reference sequence: NP_000145), generally herein regarded as ‘wildtype’, and the deamination of the target adenosine results in a loss-of-function (LOF) of the GALK1 protein, which is a (partial or complete) loss in the function tophosphorylate galactose into galactose-1-phosphate (Gal-1-P), which is the substrate for galactose-1-phosphate uridylyltransferase (GALT). Preferably, the target adenosine is the adenosine in the middle of the GAC codon coding for aspartic acid (D, Asp) at position 186 of the human GALK1 protein amino acid sequence, and wherein the deamination of theadenosine changes the amino acid to a glycine (G, Gly) because of the resulting GIC (GGC)codon. In one aspect, the orphan nucleotide is a deoxycytidine or a deoxyuridine. In one aspect, the orphan nucleotide is a cytidine analog such as a deoxynucleotide comprising a 6- amino-5-nitro-3-yl-2(1H)-pyridone nucleobase (also known as Benner’s base). In one aspect, the orphan nucleotide is a uridine analog such as a deoxynucleotide comprising an iso-uracilnucleobase. In one aspect the nucleotide numbering in the AON is such that the orphannucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’- end and negatively (-) incremented towards the 3’-end, and the first nucleotide 3’ from the orphan nucleotide (-1) in the AON (which is positioned opposite the guanosine that is 5’ of the target adenosine in the target sequence) is a nucleotide analog that can induce a syn-conformation of the guanosine, preferably a modified purine nucleobase, more preferablywherein this modified purine nucleobase is selected from the group consisting of: 7-deaza-2’- deoxyadenosine (7-deaza Ad); 7-deaza-2’-adenosine (7-deaza A); 7-deaza-2’-deoxy-2’- P5117PC00 fluoroadenosine (7-deaza Af); 7-deaza-2’-deoxy-2’-ara-fluoroadenosine; 7-deaza-2’-deoxy- 2’,2’-difluoroadenosine; 3-deaza-2’-deoxyadenosine (3-deaza Ad); 3-deaza-2’-adenosine (3- deaza A); 3-deaza-2’-deoxy-2’-fluoroadenosine (3-deaza Af); 3-deaza-2’-deoxy-2’-ara- fluoroadenosine; 3-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3,7-dideaza-2’-deoxyadenosine(3,7-dideaza Ad); 3,7-dideaza-2’-adenosine (3,7-dideaza A); 3,7-dideaza-2’-deoxy-2’-fluoroadenosine (3,7-dideaza Af); 3,7-dideaza-2’-deoxy-2’-ara-fluoroadenosine; 3,7-dideaza- 2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-O-[2-(methoxy)ethyl] adenosine; 3-deaza-2’-O- [2-methylamino-2-oxoethyl] adenosine; 2’-deoxy-2’-fluoroguanosine; 2’-ara-fluoro guanosine (FANA G); 2’,2’-difluoro guanosine; 2’-deoxyinosine (Id); 2’-OH-inosine; 2’-fluoroinosine (If);2‘-ara-fluoro inosine (FANA I); 2’,2’-difluoro inosine; 5-formylindole-2’-deoxyriboside; 5-formyl-2’-fluoro-2’-deoxyriboside; 5-formylindole-2’-ara-fluoro-2’-deoxyriboside; 5-formylindole-2’,2’- difluoro-2’-deoxyriboside; 5-formylindole-2’-O-methylriboside; 5-formylindole-2’-O-[2- (methoxy)ethyl]riboside; 5-formylindole-2’-O-[2-methylamino-2-oxoethyl]riboside; beta-(4- amidino-1H-imidazol-1-yl) riboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-deoxyriboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-ara-fluoro-2’-deoxyriboside; and beta-(4-amidino-1H-imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside. Disclosed is an AON, wherein the AON comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP), sulfonylphosphoramidate,a (1,3-dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi), or a mesylphosphoramidate (PNms) internucleotide linkage. Disclosed is an AON, 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 andnegatively (-) incremented towards the 3’-end, and wherein linkage position -2 is an MP or aPNms linkage. Disclosed is an AON, wherein the AON comprises one or more nucleotides comprising a mono- or di-substitution 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 branchedlower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, that may be interrupted by one ormore 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. Disclosed herein is an AON, wherein the AON is covalently or non-covalently, directlyor through a linker, bound to a triterpene glycoside, preferably AG1856, following the teachingof WO2024 / 153801, that shows that it is very efficient to increase RNA editing, especially P5117PC00 when an AON is connected (conjugated) 1:1 with a saponin, especially when the saponin is AG1856, which is known from WO2021 / 122998. Hence, to increase the endosomal release (intracellularly) of the AON and make it available for RNA target hybridization, the AON can be attached / conjugated (non-covalently, but preferably covalently) to AG1856 beforeadministration to the cell or the subject to be treated. In one aspect, the AON is covalently ornon-covalently, directly or through a linker, bound to a GalNAc moiety. The saponin may also be linked to the AON via the GalNAc moiety, or the GalNAc moiety may be linked to the AON via the saponin. Such conjugates are known to the person skilled in the art. Disclosed herein is a pharmaceutical composition comprising an AON as disclosedherein, a vector as disclosed herein, or a nanoparticle delivery vehicle formulation as disclosedherein, and a pharmaceutically acceptable carrier. Disclosed herein is an AON for use in the treatment of classic galactosemia caused by a deficiency in GALT, by changing a wildtype human GALK1 protein into a GALK1 protein with a LOF. Disclosed herein is a use of an AON for the manufacture of a medicament for thetreatment of classic galactosemia caused by a deficiency in GALT, by changing a wildtype human GALK1 protein into a GALK1 protein with a LOF. Disclosed herein is a method of editing a human GALK1 polynucleotide in a cell, wherein the human GALK1 polynucleotide is a GALK1 pre-mRNA or mRNA molecule, themethod comprising contacting the GALK1 polynucleotide with an AON capable of triggeringan ADAR-mediated adenosine to inosine deamination, thereby editing the GALK1 polynucleotide to encode a GALK1 protein with a LOF, preferably a loss in the functionality of phosphorylating galactose to Gal-1-P. Preferably the method comprises the step of deaminating the adenosine in the codon GAC coding for aspartic acid at position 186 in thehuman GALK1 protein amino acid sequence, thereby rendering the change from GAC to GIC,which is read as GGC by the translation machinery, and thereby changing the codon for aspartic acid (D) to a codon for glycine (G). The change is herein also referred to as D186G, or as p.Asp186Gly. Disclosed herein is a method of treating, ameliorating, or slowing down the progressionof classic galactosemia, in a human subject in need thereof, the method comprisingadministering to said subject an AON as disclosed herein, a vector as disclosed herein, a nanoparticle delivery vehicle formulation, as disclosed herein, or a pharmaceutical composition as disclosed herein, thereby contacting a GALK1 polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosinedeamination at position 186 in the human GALK1 coding sequence, thereby editing the GALK1polynucleotide to encode a GALK1 protein with a diminished or lost functionality, thereby treating the subject. P5117PC00 BRIEF DESCRIPTION OF THE DRAWINGS One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Fig. 1 shows the 5’ to 3’ sequence of part of the wildtype human GALK1 mRNAtranscript in which the GAC codon coding for aspartic acid (D) at position 186 in the GALK1 protein is in bold (SEQ ID NO:1011). The underlined adenosine is the target for RNA editing as disclosed herein, resulting in a codon coding for a glycine (G) residue (GIC / GGC) at this position 186 after editing. Below the target sequence the 5’ to 3’ sequences are provided ofthe 1010 initial AONs (SEQ ID NO:1 to 1010, as indicated) that were designed to target thetarget adenosine in SEQ ID NO:1011. The chemical modifications are as follows: Um, Am, Gm, and Cm are 2’-OMe modified uridine, adenosine, guanosine, and cytidine, respectively; m5Ue is 2’-MOE modified 5-methyl-uridine (identical to a thymidine with a 2’-MOE substitution, or ‘Te’); m5Ce is 2’-MOE modified 5-methyl-cytidine; Ae and Ge 2’-MOE modified adenosineand guanosine, respectively; Gf, Cf, Af, and Uf are 2’-F modified guanosine, cytidine,adenosine, and uridine, respectively; Zd is a deoxynucleotide (deoxycytidine analog) carrying a Benner’s base; Gd and Cd are deoxyguanosine and deoxycytidine, respectively; 8d is a deoxynucleotide carrying an iso-uracil base; “^” refers to a PO linkage; “!” refers to a PNdmi linkage; “^” refers to a MP linkage; “*” refers to a PS linkage; L001 is a tri-antennary GalNAcmoiety (OP-042; Hongene Biotech).Fig. 2 shows the editing percentages obtained after transfection of 100nM of the indicated AONs in Primary Human Hepatocytes (PHH’s), using ddPCR. Fig. 3 shows the editing percentages obtained after transfection of 200nM of the indicated AONs in PHH’s, using ddPCR. DETAILED DESCRIPTION Galactosemia is a rare, slowly progressive disease caused by a genetic inability to metabolize the sugar galactose. There are several subtypes of galactosemia, of which type I, or classic galactosemia, is caused by a galactose-1-phosphate uridylyltransferase (GALT)deficiency and of which type II is caused by a galactokinase (GALK1) deficiency.Galactosemia can greatly affect development and quality of life, resulting in CNS complications, including deficiencies in speech, cognition, behaviour, and motor skills. The CNS phenotype present in galactosemia patients has been shown to progressively worsen over time with age. CNS deficiencies may emerge in early childhood as mild to moderate, butthey often progress to severe deficiencies in later childhood and adulthood. Galactosemia alsocauses ovarian insufficiency in females, and it often results in ophthalmic complications, such as cataracts. The fact that type II galactosemia patients display milder symptoms in P5117PC00 comparison to type I patients, has suggested that the accumulation of the galactose-1- phosphate (Gal-1-P) metabolite that is produced (normally) by GALK1 and that serves (normally) as a substrate for GALT to convert it to glucose-1-phosphate (Glc-1-P) is the factor that contributes most to the severeness of classic galactosemia. Besides a dietary adjustment(intake of food avoiding galactose / lactose) no real treatment for the disease is currentlyavailable. The inventors of the current invention envisioned that introducing a mutation in the transcript of the wildtype GALK1 gene, to (partially or completely) knockout the activity / functionality of GALK1 protein would lower the concentration of the GALT substrate Gal-1-P and would thereby relieve or diminish (treat) galactosemia caused by a deficiency inGALT.The function and structure of human GALK1 has been studied in detail. It has been suggested that through an active base mechanism in which aspartic acid at position 186 abstracts a proton from galactose the protein executes its phosphorylating functionality. Aspartic acid is a hydrogen acceptor according to the rules set by Brønsted and Lowry. Theseanionic carboxylate groups behave as Brønsted bases at normal pH circumstances. Thisallows the proton to be moved from galactose leading to the subsequent phosphorylation. Indeed, changing the aspartic acid (D) at position 186 to alanine (A) or asparagine (N) resulted in variants with no detectable galactokinase activity (Megarity CF et al.2011. Bioorg Chem 39(3):120-126). The inventors of the present disclosure reasoned that changing the aspartic acid atposition 186 to glycine should also result in a loss-of-function (LOF) of the GALK1 protein, through which the amount of Gal-1-P should be lowered in patients with classic galactosemia. The change in the amino acid is on the level of the transcript, the (pre-) mRNA, not the DNA. The inventors of the present invention realized that the codon for aspartic acid atposition 186, being GAC, is eligible for RNA editing, and that RNA editing of the adenosine inthis codon to an inosine (read as a guanosine by the translation machinery) would result in a LOF version of the encoded GALK1 protein, thereby elevating disease. Hence, RNA editing using editing antisense oligonucleotides would be a first-in-class treatment for patients suffering from classic galactosemia. Accordingly, herein disclosed is an AON (as well asmethods of treatment as outlined herein) that can target the transcripts encoding the wildtypeGALK1. In addition to AON delivery methods comprising injections (e.g., IV, SC) delivery methods based on AON inhalation, nebulization and intranasal administration have proven beneficial and such methods are also contemplated to treat classic galactosemia. RNA editing is not gene therapy, because it is not irreversible and does not target thepatient’s DNA. The edited RNA disappears from the system after serving as a template fortranslation. Hence, RNA editing through the therapy disclosed herein can be temporary but can also be maintained for prolonged periods of time, as long as needed. Since the targeting P5117PC00 of the GALK1 transcript is highly specific and will not affect other (pre-) mRNA molecules, the risk of adverse side effects as often observed with chaperone compounds or agonists is low. The AONs as disclosed herein can recruit deaminating enzymes, such as ADAR1 and / or ADAR2 that are endogenously present in a cell. An AON as disclosed herein canmediate RNA editing of a target adenosine present in a target RNA molecule after it is boundto the target RNA molecule, since the deaminating enzymes are recruited to the double- stranded AON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine. The oligonucleotides are herein abbreviated to “AONs”, but sometimes also referredto as ‘editing oligonucleotides’, or ‘EONs’, even though the RNA editing event is performed bythe deamination enzyme and the action of the oligonucleotide only triggers the RNA editing to take place. There is a constant need for improving the pharmacokinetic properties of the AONs without negatively affecting the efficiency in which the target adenosine is edited in the target RNA, and / or without negatively affecting the stability of the AON itself, which is constantlyprone to breakdown because of nucleases present in a natural cell. Many chemicalmodifications are available for the generation of AONs (and many have been applied in the art). However, many of these properties are not always compatible with the desire of achieving efficient RNA editing. In the search for better pharmacokinetic properties, it was found earlier that a 2’-O-methoxyethyl (or 2’-methoxyethoxy, or 2’-MOE) modification of the ribose of some,but not all, nucleotides surprisingly appeared compatible with efficient ADAR engagement andediting (WO2019 / 158475). In a similar fashion, it was found earlier that a PS linkage at some, but not all, internucleoside linkages surprisingly appeared compatible with efficient ADAR engagement and editing (WO2019 / 219581). Also, it was found earlier that phosphonoacetate linkage modifications and / or unlocked nucleic acid (UNA) ribose modifications of some, butnot all, positions in the AON appeared compatible with efficient engagement of an enzymewith nucleotide deamination activity and with subsequent deamination (WO2020 / 165077). Whereas the properties of phosphonoacetate and UNA modifications were known as such, the compatibility thereof with engagement of enzymes with nucleotide deamination activity and with the deamination reaction was not known. Disclosed herein are AONs that can provide (mediate, cause, or trigger) RNA editingof a target adenosine in a target transcript molecule, such as pre-mRNA and / or mRNA. The target transcript molecule (human GALK1 (pre-) mRNA) is preferably wildtype and the editing results in a transcript that encodes a protein with a LOF. RNA editing is often applied to correct G>A mutations that cause a disease. This is notthe case in the present disclosure. Here, RNA editing is applied to introduce a mutation thatcauses the resulting protein to act less effectively, or more preferably with a complete loss in functionality. Non-limiting examples of transcript molecules (as disclosed in the art) that are P5117PC00 targeted using RNA editing for a variety of treatments are SERPINA1 (for the treatment of alpha1-antitrypsin (A1AT) deficiency; see e.g., WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, and WO2021 / 243023), IDUA (for the treatment of Hurler syndrome; see e.g., WO2017 / 220751, WO2018 / 041973, and WO2021 / 209010), LRRK2 (for the treatment ofParkinson’s disease; see e.g., WO2016 / 097212, WO2017 / 220751, WO2018 / 041973,WO2021 / 231673 and WO2021 / 242903), ABCA4 (for the treatment of Stargardt disease; see e.g., WO2021 / 130313 and WO2021 / 231830), USH2A (for the treatment of Usher syndrome; see e.g., WO2020 / 157008, WO2020 / 219981 and WO2021 / 136404), APP (see e.g., WO2021 / 113270), CMT1A (see e.g., WO2021 / 113390), ASS1 (see e.g., WO2021 / 231675),GJB2 (see e.g., WO2021 / 231679), MECP2 (for the treatment of Rett syndrome; see e.g.,WO2019 / 071274 and WO2021 / 231680), OTOF (for the treatment of autosomal recessive non-syndromic hearing loss; see e.g., WO2021 / 231685 and WO2021 / 231692), XLRS (see e.g., WO2021 / 231691), and PCSK9 (for the treatment of hypercholesterolemia; see e.g., WO2023 / 152371). The present disclosure relates to AONs that mediate RNA editing, using endogenous(naturally present) ADAR enzymes in the host cell (preferably liver cells including hepatocytes) of an adenosine present in the transcript of the human GALK1 gene. As described, the GALK1 enzyme is involved in metabolism predominantly in the liver although protein and some mRNA can be found in several organ systems. An AON as disclosed herein aims to lower the activityof the GALK1 protein and its ability to phosphorylate galactose. Targeting the adenosine inthe codon for aspartic acid at position 186 in the GALK1 protein amino acid sequence and changing it to an inosine resulting in a glycine at that position is a preferred example of a wildtype GALK1 transcript that is turned into a mutated variant.DefinitionsWhenever 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 completelylack RNA and DNA nucleotides (as they appear in nature) and may consist completely ofmodified nucleotides. 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 AON as disclosed herein may comprise a mix of ribonucleotides and deoxyribonucleotides. When a deoxyribonucleotide is used, hencewithout a modification at the 2’ position of the sugar, the nucleotide is often abbreviated to dA(or Ad), dC (or Cd), dG (or Gd) 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 P5117PC00 often abbreviated without the ‘d’, and often abbreviated with their respective modifications and as explained herein. 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 morephosphate 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 aphosphodiester, phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP (or MeP),methyl thiophosphonate, phosphoramidate linkages, PNdmi according to the structure of formula (IV) as described herein, and a linkage according to the structure of formula (I) as described herein. Sometimes the terms nucleobase, nucleoside and nucleotide are used interchangeably, unless the context clearly requires differently, for instance when a nucleosideis 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. Sometimes the terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, inosine, and hypoxanthine, areused interchangeably to refer to the corresponding nucleobase on the one hand, and thenucleoside or nucleotide on the other. The nucleobase thymine (T) is also known as 5- methyluracil (m5U) and is an uracil (U) derivative; thymine and 5-methyluracil can be interchanged throughout the document text. Likewise, the nucleotide thymidine is also known as 5-methyluridine and is a uridine derivative; thymidine and 5-methyluridine can beinterchanged throughout the document text.Whenever reference is made to nucleotides in the oligonucleotide, such as cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5- hydroxycytosine, and β-D-glucosyl-5-hydroxymethylcytosine are included. Whenever reference is made to adenine, N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine and7-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 madeto nucleosides or nucleotides, ribofuranose derivatives, such as 2’-deoxy, 2’-hydroxy, and 2’-O–substituted variants, such as 2’-O-methyl (2’-OMe), are included, as well as other modifications, including 2’-4’ bridged variants. Whenever reference is made to P5117PC00 oligonucleotides, one or more linkages may be a naturally occurring phosphodieaster linkage, whereas the remaining linkages between two mononucleotides may be a modified linkage. Examples of such modified linkages are phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP, phosphoramidate linkages, phosphoryl guanidine, thiophosphorylguanidine, sulfono phosphoramidate, PNdmi according to the formula (IV) as further outlinedbelow, and the linkage structure according to formula (I) as further outlined in detail below. 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%.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 invention. The term ‘conducive to’ or ‘mediate’ can be used interchangeably with ‘capable offacilitating’. When used in the context of an AON that is conducive to ADAR editing (or canmediate ADAR editing), this means that the AON, after entry into the cell, interacts with the target RNA sequence, thereby forming a double stranded structure which is recognized by the ADAR enzyme, which can then deaminate the target adenosine into an inosine. Hence, the AON itself does not have the enzymatic function (the ADAR enzyme has), but it can trigger,induce, cause, organize, mediate, provide, give, produce, facilitate, result in RNA editing afterbinding to the target RNA molecule. 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, U-C, A-A,G-G, C-C, U-U pairs. In some embodiments AONs as disclosed herein comprise fewer thanfour mismatches with the target sequence, for example 0, 1 or 2 mismatches. ‘Wobble’ base pairs are G-U, I-U, I-A, and I-C base pairs. When a U is placed opposite the target A, there is no mismatch, and the AON may be 100% complementary, although an iso-uridine (iso-U) opposite the target adenosine qualifies as a mismatch, since it does not pair according to theWatson-Crick rules of base pairing. When a C is placed opposite the target A, there is at least1 mismatch between the AON and the target sequence. 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 thenucleotide but still be relatively stable. An isolated G:G pairing in duplex RNA can for instancebe quite stable but will still be defined as a mismatch. Analysis of natural targets of ADAR enzymes has indicated that these generally include mismatches between the two strands that P5117PC00 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 Acids Res 39(13):5669-5681). Characterization of optimal patterns of paired / mismatched nucleotides between the AONs and the target RNA also appears importantto the development of efficient ADAR-based AON therapy.The term ‘complementary’ as used herein refers to the fact that the AON hybridizes under physiological conditions to a second nucleic acid strand. Examples are (i) when the AON as a first nucleic acid strand (= guide oligonucleotide) forms a heteroduplex RNA editing oligonucleotide complex with second complementary nucleic acid strand (in vitro), or (ii) whenit forms a double stranded complex with the target RNA molecule. The term does notnecessarily 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 AON and the target sequence, while under physiological conditions that AON stillhybridizes to the target sequence such that the cellular RNA editing enzymes can deaminatethe target adenosine to an inosine. The term ‘substantially complementary’ therefore also means that despite the presence of the mismatches, wobbles, and / or bulges, the AON has enough matching nucleotides with the target sequence that under physiological conditions the AON hybridizes to the target RNA molecule. As shown herein, an AON may becomplementary, but may also comprise one or more mismatches, wobbles and / or bulges withthe target sequence, if under physiological conditions the AON is able to hybridize to its target. ADARs are multidomain proteins with N-terminal double stranded RNA binding domains (dsRBDs) and C-terminal deaminase domains. Two ADAR genes encode catalytically active ADARs in humans (ADAR encoding ADAR1 proteins and ADARB1encoding the ADAR2 protein). ADAR1 is expressed in two protein isoforms (p110 and p150)that differ in their N-terminal structures. Since the substrate for ADARs is an RNA duplex, the enzymes access the reactive adenosine using a base flipping mechanism (Stephens O.M. et al. Biochemistry.2000.39(40): 12243-12251). Also, because ADARs require duplex RNA for activity, their reaction can be directed to specific adenosines in different transcripts usingcomplementary guide strands for duplex formation at the target sites. This approach iscurrently being pursued to develop therapeutic guide strands that recruit ADARs to correct disease-causing mutations in RNA (Qu L. et al. Nat. Biotechnol. 2019. 37(9):1059-1069; Merkle T. et al. Nat. Biotechnol.2019.37(2):133-138; Katrekar D. et al. Nat. Methods 2019. 16(3):239-242; Monian P. et al. Nat. Biotechnol. 2022, DOI:10.1038 / s41587-022-01225-1).While this approach is promising, ADARs have sequence preferences that make certainadenosines disfavored for reaction, limiting the current scope of this approach. For instance, the nearest neighbor nucleotide preferences for ADARs show a strong bias against reaction P5117PC00 at adenosines in 5’-GA sites (Eggington J.M. et al. Nat. Commun. 2011. 2(319):DOI:10.1038 / ncomms1324). This preference is explained by structural studies of ADAR2 bound to transition state analog-containing RNA that suggest a clash between the 2- amino group of the 5’-G and G489 of the ADAR2 loop involved in stabilizing the flipped-outconformation required for the adenosine deamination reaction (Matthews et al. Nat Struct MolBiol 2016. 23(5):426-433). Earlier work with fusion proteins bearing ADAR deaminase domains indicated that editing efficiency at 5’-GA sites could be improved with a G-A or G-G pair at the 5’ nearest neighbor (Schneider M.F. et al. Nucleic Acids Res.2014.42(10):p.e87). However, the basis for this effect has not been reported nor has this effect been establishedfor full length ADARs bearing native dsRBD RNA binding domains. In WO2024 / 013361 it isshown that G-A and G-G pairs on the 5’ side of an editing site improve editing efficiency compared to a 5’ G-C pair for full length ADAR2 and ADAR1 p110. Using X-ray crystallography, the structure of an active fragment of human ADAR2 bound to duplex RNA bearing a G:G pair adjacent to an editing site was determined. In the present disclosure relatedto editing of the adenosine in the GAU codon at position 843 also a 5’-G is present next to thetarget adenosine in the target sequence. WO2024 / 013361 discloses that the ADAR enzyme prefers the Gsyn:Gantipairing since the 2-amino group of the 5’-G in the syn conformation does not clash with the minor groove of the enzyme. It was shown that the use of nucleosides capable of stable pairing with the 5’-G in the syn conformation enables more efficient editingwithin 5’-GA target sites, providing a solution to the problem in deaminating these unfavoredediting sites. In the present disclosure, editing of the adenosine in the GAC codon at position 186 also means that there is a 5’-G from the target A. This means that the design rules as outlined in WO2024 / 013361 also apply here. Hence, it is preferred that the nucleotide at position -1 in the AON is a nucleotide analog that can induce a syn conformation of the 5’-Gin the target sequence, preferably a modified purine nucleobase, more preferably wherein thismodified purine nucleobase is selected from the group consisting of: 7-deaza-2’- deoxyadenosine (7-deaza Ad); 7-deaza-2’-adenosine (7-deaza A); 7-deaza-2’-deoxy-2’- fluoroadenosine (7-deaza Af); 7-deaza-2’-deoxy-2’-ara-fluoroadenosine; 7-deaza-2’-deoxy- 2’,2’-difluoroadenosine; 3-deaza-2’-deoxyadenosine (3-deaza Ad); 3-deaza-2’-adenosine (3-deaza A); 3-deaza-2’-deoxy-2’-fluoroadenosine (3-deaza Af); 3-deaza-2’-deoxy-2’-ara-fluoroadenosine; 3-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3,7-dideaza-2’-deoxyadenosine (3,7-dideaza Ad); 3,7-dideaza-2’-adenosine (3,7-dideaza A); 3,7-dideaza-2’-deoxy-2’- fluoroadenosine (3,7-dideaza Af); 3,7-dideaza-2’-deoxy-2’-ara-fluoroadenosine; 3,7-dideaza- 2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-O-[2-(methoxy)ethyl] adenosine; 3-deaza-2’-O-[2-methylamino-2-oxoethyl] adenosine; 2’-deoxy-2’-fluoroguanosine; 2’-ara-fluoro guanosine(FANA G); 2’,2’-difluoro guanosine; 2’-deoxyinosine (Id); 2’-OH-inosine; 2’-fluoroinosine (If); 2‘-ara-fluoro inosine (FANA I); 2’,2’-difluoro inosine; 5-formylindole-2’-deoxyriboside; 5-formyl- P5117PC00 2’-fluoro-2’-deoxyriboside; 5-formylindole-2’-ara-fluoro-2’-deoxyriboside; 5-formylindole-2’,2’- difluoro-2’-deoxyriboside; 5-formylindole-2’-O-methylriboside; 5-formylindole-2’-O-[2- (methoxy)ethyl]riboside; 5-formylindole-2’-O-[2-methylamino-2-oxoethyl]riboside; beta-(4- amidino-1H-imidazol-1-yl) riboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-deoxyriboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-ara-fluoro-2’-deoxyriboside; and beta-(4-amidino-1H-imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside. This means that the nucleotide at position -1 in SEQ ID NO:1 to 960 (Cd) is preferably replaced with a nucleotide analog that can induce a syn conformation of the 5’-G in the target sequence, as disclosed above. Moreover, this also means that the nucleotide at position -1 in SEQ ID NO:961 to 1010 (Gd) is preferably replacedwith a nucleotide analog that can induce a syn conformation of the 5’-G in the target sequence,as disclosed above. The term ‘orphan nucleotide’ relates to the nucleotide in the AON that is directly opposite the target adenosine, which is the adenosine that is deaminated by the deaminating enzyme. The orphan nucleotide may be a natural cytidine, a deoxycytidine, a uridine, or adeoxyuridine. It may also be a chemically modified nucleotide, as further described in detailbelow, or a known or chemically modified analog of a natural (deoxy)cytidine, such as a nucleotide carrying a Benner’s base, or a known or chemically modified analog of a natural (deoxy)uridine, such as iso-uridine, as further outlined in detail below. A ‘nucleotide analog’ refers to an analog of a nucleic acid nucleotide. The nucleotideanalog is an analog of adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine,deoxyguanosine, deoxycytidine, deoxythymidine or deoxyuridine. 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 sequence encoding a polypeptide, the start codon is upstream of the stop codon in the sense strand butis downstream of the stop codon in the antisense strand. The same holds true for the AONsas disclosed herein. Nucleotides that are upstream of the orphan nucleotide in the antisense oligonucleotide are located towards the 5’ terminus, and nucleotides that are downstream of the orphan nucleotide are located towards the 3’ terminus. The nucleotide ‘numbering’ in an AON as disclosed herein is such that the orphannucleotide is number 0 and the nucleotide 5’ from the orphan nucleotide is number +1.Counting is further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, wherein the first nucleotide 3’ from the orphan nucleotide is number -1. 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 oligonucleotideare positively (+) incremented towards the 5’ end and negatively (-) incremented towards the3’ end. P5117PC00 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 atleast 80%, more preferably at least 90% sequence identity.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 aredegraded rapidly and do not have any functional activity.Whenever a ‘naked’ form in relation to the AON as disclosed herein is referred to, it means that the AON is manufactured in a laboratory or manufacturing facility, through which it is generally chemically modified to prevent it from rapid degradation after it enters the mammalian body or a tissue, or cell, upon administration. A naked form of an AON is thereforedifferent from a form in which the AON is encoded (and delivered) by a viral genome or withina plasmid vector. When such viral vectors or plasmid vectors are administered, the encoded AON is expressed from the viral vector genome or from the plasmid in the cell to which the viral vector or plasmid vector is delivered. Consequently, the AON is then not chemically modified and comprises solely naturally occurring RNA nucleotides. The length of the AON as disclosed herein, and when delivered in a naked form ispreferably 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. However, when the AON as disclosed herein is to be delivered through the expression of a viral vector, then the AON may be longer, such as 70, 80, 90, 100, 150, or200 or more nucleotides in length.The term ‘HEON’ refers to a heteroduplex double-stranded complex molecule wherein an AON as disclosed herein is hybridized to a partially or fully complementary, partially of fully overlapping sense oligonucleotide. Because the AON as disclosed herein often has specified chemical modifications that are different from the chemical modifications in the sense strand,the two strands form such a heteroduplex RNA editing oligonucleotide complex. The sensestrand may be chemically modified almost in its entirety, similar or different to what is performed in the AON as disclosed herein, for example by providing nucleotides with a ribose sugar moiety carrying a 2’-OMe substitution, a 2’-F substitution, or a 2’-MOE substitution. It is to be understood that the sense strand present in the HEON is a different entity in comparisonto the target RNA molecule in the cell. The sense strand in an HEON is preferably 15, 16, 17,18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. P5117PC00 The HEON is often generated in vitro and used as a delivery tool to protect the AON from degradation when administered to the cell. In other words, the HEON is preferably formed before the AON is administered to the cell. Preferred aspects of HEONs that may be used for AONs as disclosed herein are discussed in PCT / EP2023 / 079290 (not published). Embodiments Disclosed herein is an AON that is capable of forming a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell and recruiting an endogenous ADAR enzyme in the cell, wherein the region comprises a target adenosine, wherein thenucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, whereinthe ADAR enzyme can deaminate the target adenosine into an inosine, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human GALK1 gene. Endogenous ADAR enzyme is an ADAR enzyme that is naturally present in the cell, without the need to over-express or administer exogenous ADAR or have it stably expressed from non-naturalsources. In one aspect, the transcript molecule is a pre-mRNA or an mRNA molecule. In oneaspect, the human GALK1 gene is according to NCBI Gene ID No. 2584 (NCBI reference sequence: NP_000145), generally herein regarded as ‘wildtype, and the target adenosine is the middle adenosine in the GAC codon coding for aspartic acid (D, Asp) at position 186 of the human GALK1 protein, and wherein the deamination of the adenosine changes the aminoacid to a glycine (G, Gly), and wherein the deamination of the target adenosine results in aGALK1 protein with a LOF modality. The target adenosine in the GAC codon is nucleotide 557 in the mRNA sequence, and the change, brought about by the RNA editing as disclosed herein, may also be referred to as c.557A>G. In one aspect, the oligonucleotide as disclosed herein comprises or consists of thesequence of any one of SEQ ID NO:1 to 1010, preferably of the sequence selected from thegroup consisting of: SEQ ID NO:985, 986, 990, 991, and 995. In a preferred aspect, the nucleotide at position -1 in these AONs is replaced by a nucleotide analog that can induce a syn conformation of the 5’-G in the target sequence, as disclosed above. In one aspect, an AON as disclosed herein comprises or consists of any of the sequences provided in Fig.1,optionally including the indicated chemical modifications to the nucleobase, sugar moietyand / or linkage, or any combination thereof. In one aspect, the AON as disclosed herein comprises or consists of the sequence of any one of SEQ ID NO:1 to 1010, preferably of the sequence selected from the group consisting of: SEQ ID NO:985, 986, 990, 991, and 995, but wherein the nucleotide at position -1 in the oligonucleotide is not as provided in Fig.1, but ismodified according to what has been outlined above, to accommodate the most efficientediting in view of the 5’-G that is next to the target A. P5117PC00 In one aspect, the orphan nucleotide in an AON as disclosed herein does not comprise a natural cytosine nucleobase. In one aspect, the orphan nucleotide in an AON as disclosed herein does not comprise a 2’-OMe substituted ribose. In one aspect, the orphan nucleotide in an AON as disclosed herein is a deoxycytidine or a deoxyuridine. In one aspect, the orphannucleotide in an AON as disclosed herein is a cytidine analog or a uridine analog. In oneaspect, a cytidine analog is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1H)- pyridone nucleobase. In one aspect, a uridine analog is a deoxynucleotide comprising an iso- uracil nucleobase. In one aspect, at least one nucleotide or nucleotide analog in the AON comprises asubstitution at the 2' position of the ribose, wherein the substitution is selected from the groupconsisting of: H (DNA); OH (RNA); F; ara-F; diF; 2’-C-methyl’-2’-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; with the provisothat the nucleotide that is opposite the target adenosine does not comprise a 2’-O-methyl or a 2’-(2-methoxy)ethyl ribose modification. In one aspect, at least one nucleotide or nucleotide analog in the AON is an arabinonucleic acid. In one aspect, the nucleotide analog comprises a modified nucleobase. In one aspect, the AON is covalently or non-covalently, directly orthrough a linker, bound to a GalNAc moiety. The skilled person can select the right linker andneed for covalent or non-covalent binding of a GalNAc moiety, when the AON needs to be delivered to liver cells, especially hepatocytes. A preferred GalNAc moiety to be used in the AONs as disclosed herein, is disclosed in WO2022 / 271806. In one aspect, the AON as disclosed herein is 15, 16, 17, 18, 19, 20, 21, 22, 23, 24,25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49,50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. In one aspect, an AON as disclosed herein is in a naked form. In one aspect, an AON as disclosed herein is in a circular format. In one aspect, an AON as disclosed herein is not in a naked form but is expressed from the genome of a viral vector. In one aspect, an AON as disclosed herein is not in a nakedform but is expressed from an expression vector such as a plasmid. In one aspect, when theAON as disclosed herein is not in a naked form, the AON is 15 to 60 nucleotides in length as indicated above, or in another embodiment, from 61 to 300 nucleotides in length. In one aspect, an AON as disclosed herein comprises one or more modifications in the linkage moiety, which is each independently selected from a PS, phosphonoacetate,phosphorodithioate, MP, sulfonylphosphoramidate, PNdmi internucleotide linkage (accordingto the structure of formula (IV) as outlined below), or a linkage moiety with the structure according to formula (I) as outlined below. P5117PC00 Disclosed herein is an AON, 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 an MP or aPNms linkage. In one aspect, the linkage between the most terminal two nucleotides on the5’ and / or 3’ terminus of the AON as disclosed herein is a PNdmi or a PNms linkage. Disclosed herein is an AON, wherein the AON comprises one or more nucleotides comprising a mono- or di-substitution 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 beinterrupted 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. Disclosed herein is a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding a chemicallyunmodified AON as disclosed herein. The skilled person understands that any of the AON sequences provided in Fig.1 can also be generated without chemical modifications (wherein the orphan nucleotide is either a C or a U) and that such unmodified AONs can be expressed from a vector as pure RNA, after entry of the vector into the cell. Disclosed herein is a pharmaceutical composition comprising an AON as disclosedherein, or a vector as disclosed herein, and a pharmaceutically acceptable carrier. Disclosed herein is an AON for use in the treatment of classic galactosemia. Disclosed herein is a use of an AON as disclosed herein, for the manufacture of a medicament for the treatment of classic galactosemia. Disclosed herein is a method of editing a human GALK1 polynucleotide in a cell,wherein the human GALK1 polynucleotide is a pre-mRNA or mRNA molecule, the method comprising contacting the GALK1 polynucleotide with an AON capable of triggering an ADAR- mediated adenosine to inosine deamination, thereby editing the GALK1 polynucleotide to encode a GALK1 protein with a LOF. Disclosed is a method of treating, ameliorating, or slowing down classic galactosemiacaused by a GALT deficiency in a human subject in need thereof, the method comprising administering to said subject an AON as disclosed herein, a nanoparticle delivery vehicle formulation as disclosed herein, or a vector as disclosed herein, or a pharmaceutical composition as disclosed herein, thereby contacting a GALK1 polynucleotide, preferably awildtype pre-mRNA or an mRNA transcript molecule, in a cell of the subject with an AONcapable of effecting an ADAR-mediated adenosine to inosine deamination, thereby editing the P5117PC00 GALK1 polynucleotide to encode a GALK1 protein with a lowered, diminished, or absent ability to phosphorylate galactose. Disclosed herein is an in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in a human GALK1 pre-mRNA or mRNA molecule in a cell, the method comprisingthe steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by thecell of the AON; (iii) allowing annealing of the AON to the GALK1 transcribed pre-mRNA or mRNA target molecule; and (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine. In one aspect, the in vivo method as disclosed herein comprises step (v) of using a functional read-out to identify the presenceof the inosine in the target RNA molecule. Bodily fluid available biomarkers known to theperson skilled in the art, especially in the art of diagnosing and treating classic galactosemia, such as measuring levels of galactose in blood and / or urine, measuring levels of N-Glycan and / or Gal-1-P in serum, and / or measuring levels of galactitol in urine samples are examples can potentially be used as biomarkers. Chemical modifications Various chemistries and modifications are known in the field of oligonucleotides that can be readily used in accordance with the invention. All chemical modifications listed herein that may be used in the AON as disclosed herein may also be used for a sense strand that iscomplementary to the AON, when the AON and the complementary strand form a HEONcomplex, such as described in WO2024 / 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 beused together with the AON in a pharmaceutical product. This includes the use of hydrophobicmoieties (such as tocopherol and cholesterol) and cell-specific ligands, that have also been described herein, and in detail in WO2024 / 084048, which may either be bound to the AON or its opposite strand, or both. 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 achemically modified nucleotide. The most common naturally occurring nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (U). These consist of a pentose sugar, a ribose, a 5’-linked phosphate group which is linked via a phosphate ester, and a 1’-linked base. The sugarconnects the base and the phosphate and is therefore often referred to as the “scaffold” of thenucleotide. 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 P5117PC00 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 WO2020 / 154342, WO2020 / 154343, and WO2020 / 154344. Anucleoside 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, orconsidered 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. 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 tonuclease-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 betweennucleosides, as described herein.Scaffold modifications (ribose) 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 ofcompatibility 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 1’-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’-OMeor 2’-MOE substitution when the nucleobase is a naturally occurring cytosine, but may carry a2’-F, a 2’,2’-difluoro (diF), or 2’-ara-F (FANA) substitution or may be DNA. PCT / EP2023 / 069609 (unpublished) 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 manylinkers known in the art such as a methylene linker, amide linker, or constrained ethyl linker(cEt). P5117PC00 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 deaminatingthe target adenosine. An AON as disclosed herein may comprise a 2’-OMe ribose modificationat 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,TNA, 2’-fluoro (2’-F), 2’,2’-difluoro (diF) modification, 2’-fluoro-2’-C-methyl modification, or a2’-4’-linkage (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA or examples mentioned in e.g. WO2018 / 007475)). 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 fromlinkers known in the art, such as a methylene linker or constrained ethyl linker. A wide varietyof 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. In all cases, the modifications should be compatible with RNA editingsuch that the AON fulfils its role as an oligonucleotide that can form a double stranded complexwith the target RNA and by generating this double-stranded nucleic acid complex, recruit a deaminating enzyme, which can subsequently deaminate the target adenosine. 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 modificationsdiscussed 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 a locked nucleic acid (LNA)). In one aspect, the AON as disclosed herein comprises at least one nucleotide comprising a threose nucleic acid (TNA) ribose modification. In one aspect, the AON as disclosed herein comprises at least one nucleotide with a sugarmoiety that comprises a 2’-fluoro (2’-F) modification. A preferred position for the nucleotidethat carries a 2’-F modification is position -3 in the AON, which may be present together with an identical 2’ modification in the orphan nucleotide as discussed above. Base modifications Abase, 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 CF moieties) or P5117PC00 aromatic electronic interactions. Cytosine, thymine, and uracil are pyrimidine bases, and are generally linked to the scaffold through their 1-nitrogen. Adenine and guanine are purine bases and are generally linked to the scaffold through their 9-nitrogen. The terms ‘adenine’, ‘guanine’, ‘cytosine’, ‘thymine’, ‘uracil’ and ‘hypoxanthine’ as used herein refer to thenucleobases 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. The nucleobases at any position in the AON asdisclosed herein can be a modified form of adenine, cytosine, guanine, or uracil, such ashypoxanthine (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-formyluracil, 5-aminomethylcytosine, 5-formylcytosine), 5-hydroxymethylcytosine, 7-deazaguanine, 7-deazaadenine, 7-deaza-2,6-diaminopurine, 8- aza-7-deazaguanine, 8-aza-7-deazaadenine, 8-aza-7-deaza-2,6-diaminopurine, 8-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, SuperT, 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). 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 thatare 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 chemistries and formats may vary from oligonucleotide construct to oligonucleotide construct and from application to application, and may be worked out inaccordance with the wishes and preferences of those of skill in the art.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 notlimited to 2’-O-modified RNA monomers, such as 2’-O-alkyl or 2’-O-(substituted)alkyl such as2’-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- P5117PC00 (dimethylamino)ethyl); 2’-deoxy (DNA); 2’-O-(haloalkyl)methyl such as 2’-O-(2- chloroethoxy)methyl (MCEM), 2’-O-(2,2-dichloroethoxy)methyl (DCEM); 2’-O-alkoxycarbonyl such as 2’-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2’-O-[2-N-methylcarbamoyl)ethyl] (MCE), 2’-O-[2-(N,N-dimethylcarbamoyl)ethyl] (DCME); 2’-halo e.g. 2’-F, FANA; 2'-O-[2-(methylamino)-2-oxoethyl] (NMA); a bicyclic or bridged nucleic acid (BNA) scaffoldmodification such as a conformationally restricted nucleotide (CRN) monomer, a locked nucleic acid (LNA) monomer, a xylo-LNA monomer, an α-LNA monomer, an α-l-LNA monomer, a β-d-LNA monomer, a 2’-amino-LNA monomer, a 2’-(alkylamino)-LNA monomer, a 2’-(acylamino)-LNA monomer, a 2’-N-substituted 2’-amino-LNA monomer, a 2’-thio-LNAmonomer, a (2’-O,4’-C) constrained ethyl (cEt) BNA monomer, a (2’-O,4’-C) constrainedmethoxyethyl (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-2H-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 BNAmonomer (such as AmNA), an urea-bridged BNA monomer, a sulfonamide-bridged BNA monomer, a bicyclic carbocyclic nucleotide monomer, a TriNA monomer, an α-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, anoxetane nucleotide monomer, a locked PMO monomer derived from 2’-amino LNA, aguanidine-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 thesemodifications are known to the person skilled in the art. The orphan nucleotide 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 deaminationby 60-fold when compared to the wild-type enzyme (Kuttan and Bass. Proc Natl Acad Sci USA2012.109(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). 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 maybe), as indicated above. The crystal structure of ADAR2 E488Q bound to double strandedRNA (dsRNA) revealed that the glutamine (Gln; Q) side chain at position 488 can donate an P5117PC00 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 (Gln) the amide group of the glutamine is absent and is instead a carboxylic acid. To obtain the same contact of the orphan cytidine with theE488Q mutant would then, for the wild-type situation, require protonation for this contact tooccur. 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 E488QADAR2 mutant. By replacing the nucleotide opposite the target adenosine in the AON withcytidine 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. J Org Chem 2009. 74(21):8021-8030; Burchenal et al. (1976)Cancer Res 36:1520-1523) and Benner’s base Z (also referred to as ‘dZ’ or ‘Zd’; Yang et al.Nucl Acid Res 2006. 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 also referred to with its chemical name 6-amino-5-nitro-3-yl-2(1H)-pyridone. The presence of the cytidine analog in the AON may exist in addition to modifications to theribose 2’ group. The ribose 2’ groups in the orphan nucleotide can be independently selectedfrom 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-OMe, 2’-MOE, 2’-F, or 2’-4’-linked (i.e., a bridged nucleic acid such as a locked nucleic acid (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. The orphan nucleotide in the AON as disclosed herein is preferably a cytidine or analogthereof (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 anotheraspect both the neighbouring (directly adjacent) nucleotides flanking the orphan nucleotide donot 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, anadenosine in a target RNA can be protected from editing by providing an opposing nucleotidewith a 2'-OMe group (at least when there are no other chemical substitutions or modifications P5117PC00 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. Linkage modifications Anucleoside is generally connected to neighboring nucleosides through condensationof 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 scaffoldmoieties. Because of this characteristic, the alternating copolymer formed by linked scaffoldsof 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 phosphorothioate, such a moiety is still referred to asthe 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. 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 asdisclosed 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 phosphorothioate (PS), chirally pure PS, (R)-PS, (S)-PS, methyl phosphonate (MP, also referred to as MeP), chirallypure MP, (R)-MP, (S)-MP, phosphoryl guanidine (such as PNdmi), chirally pure phosphorylguanidine, (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, metylboranophosphate, methyl borano PS, methyl boranophosphonate, methylboranophosphothioate, 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), andthioacetamide 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’ P5117PC00 and 2’^5’ linkages. An AON as disclosed herein may also comprise one or more linkage modifications according to the structure of formula (I) wherein:X = O or S; andR = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a C1-C6alkoxy, a substituted C1-C6alkoxy, a C1-C20alkyl, a substituted C1-C20alkyl, a C1-C6alkenyl, a C1-C6substituted alkenyl, a C1-C6alkynyl, a substituted C1-C6alkynyl, or a conjugate group. In a preferred embodiment, X = Oand R = methyl and the linkage modification is referred to as mesyl phosphoramidate, MsPAor PNms. 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, Requals one of the following structures (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0002] P5117PC00 Disclosed herein is also an AO that is able to mediate deamination by recruitment of a deaminating enzyme in a cell after the AON has formed a double-strandedcomplex with a region of a target RNA nucleic acid molecule in a cell, wherein the regioncomprises a target adenosine, wherein the deaminating enzyme can deaminate the target adenosine into an inosine, and whe rein the AON comprises a moiety with a structure according to formula (II): 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 C1-C6alkoxy, a substituted C1-C6alkoxy, aC1-C20 alkyl, a substituted C1-C20 alkyl, a C1-C6 alkenyl, a C1-C6 substituted alkenyl, a C1-C6 P5117PC00 alkynyl, a substituted C1-C6alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl. 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 butadds significant resistance to nuclease degradation. A preferred nucleotide analogue orequivalent 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,selenophosphate or 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 nucleotidesmay be altered by mono- or di-thioation of the phosphodiester bonds to yield PS esters orphosphorodithioate 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 moststable oligonucleotide compound.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 anRp or Sp configuration at a specified linkage position may depend on the target sequence andthe 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 theconfigurations, while for other positions such does not matter. In one aspect, the AON asdisclosed herein comprises one or more (chirally pure or chirally mixed) PS linkages. 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 connectsthe terminal two nucleotides on each end of the AON. AONs as disclosed herein may alsocomprise 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 P5117PC00 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 as pect, the AON as disclosed herein comprises atleast one MP internucleoside linkage according to the structure of formula (III): 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 modificationaccording to the structure of formula (I), instead of an MP linkage. WO2020 / 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 ofoligonucleotides is challenging in view of additional manufacturing (purification) steps in thecoupling and decoupling process. In one aspect, the AON does not comprise an MP linkage. 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) P5117PC00 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. In one aspect, the AON as disclosed herein comprises at least one phosphonoacetateand / or at least one phosphonoacetamide internucleoside linkage.Conjugate chemistries 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 ahydrophobic moiety, such as palmityl or an analog thereof, cholesterol or analog thereof, ortocopherol 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 hydrogenbond, 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. 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 anendogenous enzyme such as a nuclease, or by physiological circumstances specific to partsof 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. An uncleavablelinker refers to a linker that is not cleaved under physiological conditions, or very slowlycompared 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 maybe usually from 2 to 20 bases in length, from 3 to 10 bases in length, or from 4 to 6 bases inlength. 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. 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.General P5117PC00 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 oneaspect, the AON as disclosed herein comprises at least one internucleoside linkage accordingto 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’-Fmodification, and / or the AON comprises an orphan nucleotide that carries a 2’-H in the sugarmoiety 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 diF substitution in the sugar moiety. In one aspect, the orphan nucleotide carries a 2’-F and a2’-C-methyl in the sugar moiety. In one aspect, the orphan nucleotide comprises a 2’-F in thearabinose configuration (FANA) in the sugar moiety. 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 adenosinein the target RNA molecule, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, a nd wherein the orphan nucleotide has the structure of formula (V): wherein: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from the groupconsisting of: cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxo-adenine, and 6-amino-5-nitro-3-yl-2(1H)-pyridone; R1and R2are both selected, independently, from H, OH, F or CH3; R3is the part of the AON that is 5’ of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4is 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 theorphan nucleotide may be DNA, more preferably the nucleotide at the 3’ (position -1).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 P5117PC00 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 thedisclosure, the enzyme with adenosine deaminase activity is preferably ADAR1, ADAR2, orADAT. 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 disclosedherein preferably does not comprise a boxB RNA hairpin sequence. In one aspect, an AONas 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 iso-uridine opposite the target adenosine, which likely does not pair like G pairs with U. 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. 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 andactivity. 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 asthose in the deaminase domain. For example, PS linkages between nucleotides or 2’-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 ofthe nucleotides that are opposite to the target trinucleotide. For example, the 5’-CAA-3’ targetsequence, 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 P5117PC00 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), more preferably a deoxynucleotide carrying a hypoxanthine nucleobase. The AON as disclosed herein, in contrast to what has been described for siRNA, orgapmers and their relation towards RNase breakdown and the use of such gapmers in double-stranded 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. It is not desired that the target transcript molecule is degraded through the binding of the AON to the transcript molecule. In one embodiment, the AON doesnot comprise four or more consecutive DNA nucleotides anywhere within its sequence. In anembodiment, 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 other nucleotides within the AON may be DNA, but also that there isno stretch of four or more consecutive DNA nucleotides within the AON. Hence, the AON asdisclosed herein is not a gapmer. A gapmer reduces the expression of a target transcript but does not produce RNA editing of a specified adenosine within the target transcript. 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 atthe 5’ end (5’ wing region) and the 3’ end (3’ wing region) thereof. In contrast, the AON asdisclosed herein 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 and to allow the mRNA transcript being translated into a protein. The AONs as disclosed herein may also be administered in the context of aids that willincrease 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”, as outlined infra. A saponin that can be used in the methods as disclosed hereinis AG1856, disclosed in WO2021 / 122998 and further described for use with RNA editingproducing oligonucleotides in WO2024 / 153801. Disclosed herein is also a pharmaceutical composition comprising the AON as disclosed herein, and further comprising a pharmaceutically acceptable carrier, solvent, diluent, and / or other additive (such as a saponin or triterpene glycoside like AG1856 (asdiscussed above), which in fact may be conjugated to the AON, and may also be administeredseparately from the AON) and may be dissolved in a pharmaceutically acceptable organic solvent, or the like. Dosage forms in which the AON or the pharmaceutical composition are P5117PC00 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 bedetermined using common general knowledge in the field and may be adjusted based on thedisorder and the efficacy of the active ingredient. Although in a preferred embodiment, the AON as disclosed herein is a single-stranded oligonucleotide comprising an orphan nucleotide opposite the target adenosine, wherein the orphan nucleotide is chemically modified as disclosed herein, and wherein the remainder ofthe oligonucleotide is chemically modified to prevent it from nuclease breakdown also asdisclosed herein, in another embodiment, disclosed is 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 in a circular format. In a preferred aspect, the AON asdisclosed herein is a ‘naked’ oligonucleotide, comprising a variety of chemical modificationsin the ribose sugar and / or the base of one or more of the nucleotides within the sequence, that preferably comprises at least one linkage according to the structure of formula (I) as disclosed herein, that can hybridize to the target transcript or a part thereof that includes the target adenosine, and can recruit endogenous (naturally present) ADAR in the target cell for thedeamination of the target adenosine. In another aspect, the AON as disclosed herein, that isdelivered in a ‘naked’ form, does not comprise a stem-loop structure for recruitment of the deaminating enzyme, which allows for a shorter AON and improved cellular delivery and trafficking. Notably, when the AON comprises chemical modifications, as detailed herein, it maystill be delivered through the means of a delivery vehicle. Suitable delivery vehicles arenanoparticle delivery vehicles such as polymeric nanoparticles, dendrimers, inorganic nanoparticles and nanocrystals, organic nanocrystals, and liposomes. Preferred nanoparticles are Lipid Nanoparticles (LNP’s) that are nano-sized lipid vesicles that carry the AON of the present invention and aid to the delivery of target cells. In the event that an LNP is applied orany other similar type of carrier, the AON is still considered naked because it is not transcribedfrom 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 theperson 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:1 to 1010. The person skilled in the art P5117PC00 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. 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 factoris the degree of complementarity of the two strands making up the dsRNA sequence. Perfectcomplementarity 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 hADAR1 and 2 can be increased by introducing chemical modifications and / or ensuring several mismatches in the dsRNA, which presumably helps toposition 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 skilled in the art will be capable of designing the complementary portion of the oligonucleotide according to their needs. It will be understood by a person having ordinary skill in the art that the extent to whichthe editing enzymes 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 enzyme. The exact modification may be determined through some trial and error and / or through computational methods based on structural interactions between the AON and therecognition domain of the editing enzyme. In addition, or alternatively, the degree of recruitingand redirecting the editing enzyme 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. Disclosed herein is the site-specific editing of target adenosines in RNA sequences ineukaryotic, preferably metazoan, more preferably mammalian, more preferably human cells,more preferably human liver cells, more preferably hepatocytes. The target cell can be located in vitro, ex vivo or in vivo. One advantage of the AON as disclosed herein is that it can be used with cells in situ in a living organism, but it can also be used with cells in culture. 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 AON asdisclosed herein can also be used to edit target RNA sequences in cells from a transplant or within a so-called organoid, e.g., a brain tissue organoid, or liver spheroid. Organoids can be thought of as three-dimensional in vitro–derived tissues but are driven using specific conditions to generate individual, isolated tissues. Without wishing to be bound by theory, the RNA editing through human ADAR2 forexample 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. P5117PC00 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; entireexpression pathways may be altered by recoding miRNAs or their cognate sequences ontarget 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 oractivation and so on. These and other forms of RNA and protein “engineering”, whether toprevent, 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. The amount of AON to be administered, the dosage and the dosing regimen can varyfrom cell type to cell type, the disease to be treated, age, weight, gender, 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 leads to 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 couldcompete for binding to an ADAR enzyme within a cell, thereby depleting the amount of the enzyme, 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. One suitable trial technique involves delivering the AON to cell lines, or a test organismand then taking biopsy samples at various time points thereafter. The sequence of the targetRNA can be assessed in the biopsy sample and the proportion of cells having the modification can easily be followed. A method as disclosed herein can 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 therelevant part of the target RNA, or a cDNA copy thereof (or a cDNA copy of a splicing productthereof, in case the target RNA is a pre-mRNA), as discussed above, and the sequence change can thus be easily verified. Alternatively, as indicated above, the change may be assessed on the function of the protein before, during, and / or after treatment or assessing any other potential marker, which measurements are preferably performed in vitro on samplesobtained from the treated subject.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 P5117PC00 therapeutic terms a method as disclosed herein 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. AONs as disclosed herein are particularly suitable for therapeutic use, and sodisclosed is also a pharmaceutical composition comprising an AON as disclosed herein anda pharmaceutically acceptable carrier, solvent, or diluent. In some embodiments the pharmaceutically acceptable carrier can simply be a saline solution. This can usefully be isotonic or hypotonic, particularly for pulmonary delivery. The AON as disclosed herein is suitably administrated in aqueous solution, e.g. saline, or in suspension, optionally comprisingadditives, excipients and other ingredients, compatible with pharmaceutical use, atconcentrations 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 suitably range from between about 1 µg / kg to about 100 mg / kg, preferably from about 10 µg / kg to about 10 mg / kg, more preferably from about 100 µg / kg to about 1 mg / kg. As outlined above, the AONs of the present disclosuremay also be delivered through a delivery vehicle such as an LNP. Amounts of LNP carryingAONs as disclosed herein can and will also be determined in (pre-) clinical phases. Administration may be by injection or infusion, intracranially, intrathecally, intranasally, orally, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra- peritoneally, intrarectally, intra-cisterna magna, parenterally, and the like. Administration maybe in solid form, in the form of a powder, a pill, a gel, a solution, a slow-release formulation, orin any other form compatible with pharmaceutical use in humans. In one embodiment, depending on the ultimate deamination effect of A to I conversion, the identification step of whether the editing has taken place, comprises the following steps: sequencing the target RNA; assessing the presence or absence of a non-, or less-functionalprotein; assessing whether splicing of the pre-mRNA was altered by the deamination;assessing the concentration and / or presence / absence of a biomarker; or using a functional read-out. A functional assessment will generally be according to methods known to the skilled person. A 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 theperson skilled in the art. However, the person skilled in the art of diseases such as classicgalactosemia will preferably apply tests to monitor certain biomarkers related to (metabolic) function(s). Examples are assessing the concentration of galactose in blood and / or urine samples, assessing the concentration of Gal-1-P and / or N-glycan in serum samples, and assessing the concentration of galactitol in urine samples, before and after administering anAON as disclosed herein.In one embodiment, a method as disclosed herein comprises the steps of administering to the subject an AON, a nanoparticle delivery vehicle formulation as disclosed P5117PC00 herein, or a pharmaceutical composition as disclosed herein, 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 ADAR 1 or ADAR2; and allowing theenzyme to deaminate the target adenosine in the target nucleic target molecule to an inosine,thereby alleviating, treating, ameliorating, or slowing down progression of classic galactosemia. RNA editing molecules present in the cell will usually be proteinaceous in nature, such as the ADAR enzymes found in metazoans, including mammals. The ones of most interestare the human ADARs, hADAR1 and hADAR2, including any isoforms thereof. RNA editingenzymes known in the art, for which oligonucleotide constructs as disclosed herein may conveniently be designed, include the adenosine deaminases acting on RNA (ADARs), such as hADAR1 and hADAR2 in humans or human cells and cytidine deaminases. It is known that hADAR1 exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 110kDa 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-γ). hADAR1 is also inducible by TNF-α. This provides an opportunity to develop combination therapy, whereby IFN-γ or TNF-α and AONs as disclosed herein are administered to a patient either as a combination product, or asseparate products, either simultaneously or subsequently, in any order. Certain diseaseconditions may already coincide with increased IFN-γ or TNF-α 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 theaffinity of the first nucleic acid strand for the recognition domain of the editing molecule.An AON as disclosed herein can utilise endogenous cellular pathways and naturally available ADAR enzymes to specifically edit a target adenosine in the target RNA sequence. An AON as disclosed herein is capable of recruiting ADAR and complex with it and then facilitates the deamination of a (single) specific target adenosine nucleotide in a target RNAsequence to which it is bound. Ideally, only one adenosine is deaminated. An AON asdisclosed herein, when complexed to ADAR, preferably brings about the deamination of a single target adenosine. An AON as disclosed herein, especially when it is in a naked form, is normally longer than 10 nucleotides, preferably more than 11, 12, 13, 14, 15, 16, still more preferably morethan 17 nucleotides. In one aspect the AON as disclosed herein is longer than 20 nucleotides.The AON as disclosed herein is preferably shorter than 100 nucleotides, still more preferably shorter than 60 nucleotides, still more preferably shorter than 50 nucleotides. In a preferred P5117PC00 aspect, the AON as disclosed herein 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 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. In oneembodiment, the AON is 27, 28, 29, or 30 nucleotides in length.

[0003] P5117PC00 EXAMPLES Example 1. RNA editing of the wildtype human GALK1 transcript using a variety of AONs. An initial set of 1010 AONs was designed to target the middle adenosine in the GAC codon encoding aspartic acid at position 186 in the human GALK1 protein. The design and chemical modifications of these AONs are provided in Fig.1. For the initial screen of the AONs with SEQ ID NO:1 to 960, the following was performed. On day 0, Primary Human Hepatocytes (PHH’s; 5.0x104cells / well) were transfected with 100 nM AONs, in triplicates, using Lipofectamine®RNAiMAX Reagent at the same time of seeding, following the protocol of the manufacturer. The plates containing cells, medium and AON were held at 37 °C, 5% CO2, and the medium was refreshed 24 hrs after transfection / plating. On day 3 (72 hrs post transfection / plating) the supernatants were discarded, and subsequent analysis was performed as follows. Cells were collected and used for RNA isolation using a RNeasy 96 Kit (Qiagen-74182) according to the manufacturer’s instructions. Extracted RNA was treated with DNase I (ThermoFisher-EN0521) according to manufacturer’s protocol. Samples were incubated at 37 °C for 30 min and then 1 µL 50 mM EDTA was added and further incubated at 60 °C for 2 min. The total RNAs were then reverse- transcribed using the Maxima Reverse Transcriptase (Thermo-EP0742) kit with oligo-dT primer, random Hexamer Primer, and dNTP Mix (10 mM each). A quantitative PCR was then performed with the Digital PCR System (Bio-Rad, QX200) in 22 μl aliquots of reaction mixtures containing cDNA, appropriate pairs of primers and ddPCR Supermix for Probes (no dUTP) (Bio-Rad-1863024). The primers given in Table 1 were used with a PCR program that was as follows: 10 min at 95 °C; 40 cycles for 30 sec at 94 °C and 60 sec at 63 °C, 10 min at 98 °C and a hold step at 4°C. Then the plate was placed into the droplet QX200 reader to measure the number of positive droplets. The editing percentage was calculated by pooling the three replicates for each transfection for all A and G counts and then scored according to the formula: score = SUM(G) / (SUM(A+G) * 100 The p-value (p=0.05) represents the probability of 3 treated replicates to be different from the 3 non-treated replicates. Editing scores of the AONs are ranked from the highest editing percentage to the lowest. From the 960 AONs that were tested, 30 AONs scored above 0 % editing. These editing percentages are provided in Table 2. The other 930 EONs (giving 0% editing) are not listed in Table 2. The best scoring AONs are RM114625 (SEQ ID NO:886), RM114064 (SEQ ID NO:325), and RM114579 (SEQ ID NO:840), with percentages above 1%. P5117PC00 Table 1. Primers and probes for the dPCR of D186G editing in human GALK1 transcripts. The “+” symbol represents a Locked Nucleic Acid (LNA) at the 3’ side of the symbol. Primers and probes Sequence (5’-3’) human GALK1 FW primer AGACTCGGGCACAATAGCTG (SEQ ID NO:1012) human GALK1 REV primer GGACCTGCAGTCAATGAGCA (SEQ ID NO:1013) GALK1 probe target A HEX ATC+ATG+G+A+CCAGTTCATCTCACT (SEQ ID NO:1014) GALK1 probe target G FAM ATC+ATGG+G+CCAGTTCATCTCACT (SEQ ID NO:1015) Table 2. Editing percentages of the AONs (in a high throughput screen using 960 AONs) that provide D186G editing in human GALK1 transcripts in PHH’s, after transfection. AON Editing % RM114625 2,33 RM114064 1,19 RM114579 1,08 RM114219 0,87 RM114611 0,86 RM114057 0,79 RM114574 0,75 RM114575 0,65 RM114565 0,60 RM114577 0,58 RM114619 0,56 RM113962 0,48 RM114618 0,47 RM114544 0,42 RM113902 0,42 RM114265 0,34 RM114567 0,32 RM114329 0,30 RM114626 0,30 RM114557 0,29 RM113881 0,24 RM114665 0,23 RM114571 0,22 RM114564 0,20 RM114559 0,19 RM114670 0,17 RM113909 0,13 RM114302 0,13 RM113882 0,11 RM113874 0,11 P5117PC00 Example 2. RNA editing of the wildtype human GALK1 transcript using a further set of AONs. In a next screen, RM107589 to RM107638 (SEQ ID NO:961 to 1010, respectively), all bound to a tri-antennary GalNAc moiety on the 5’ terminus (L001 in Fig.1), were transfected in PHH’s using 100nM and 200nM AON. The same procedure was followed as outlined above and RNA was isolated 72 hrs post-transfection. Editing percentages were determined using a ddPCR as outlined above. The results of the 100nM transfections are given in Fig.2 and the results of the 200nM transfections are given in Fig.3. This shows that in this experiment the editing percentages were often higher than found in the high-throughput screen and that RM107613 (SEQ ID NO:985), RM107614 (SEQ ID NO:986), RM107618 (SEQ ID NO:990), RM107619 (SEQ ID NO:991) and RM107623 (SEQ ID NO:995) performed best, reaching editing percentages above 10%.

Claims

P5117PC00 CLAIMS 1. An antisense oligonucleotide (AON) that is capable of forming a double-stranded complex with a region of a target RNA nucleic acid molecule in a human cell and recruiting an endogenous ADAR enzyme 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 nucleotide numbering 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, wherein the ADAR enzyme can deaminate the target adenosine into an inosine, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human galactokinase (GALK1) gene encoding the GALK1 protein.

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

3. An AON according to claim 1 or 2, wherein the cell is a liver cell, more preferably a hepatocyte.

4. An AON according to any one of claims 1 to 3, wherein the target adenosine is the adenosine in the GAC codon coding for an aspartic acid (D) residue at position 186 of the GALK1 protein, and wherein the deamination of the adenosine results in an inosine that translates the codon to a glycine (G) residue at position 186, and wherein the deamination of the target adenosine results in a GALK1 protein that has a diminished, lowered, or absent ability to phosphorylate galactose.

5. An AON according to any one of claims 1 to 4, wherein the oligonucleotide comprises or consists of the sequence of any one of SEQ ID NO:1 to 1010, preferably SEQ ID NO:985, 986, 990, 991, and 995.

6. An AON according to any one of claims 1 to 5, wherein the nucleotide at position -1 in the AON is a nucleotide analog that can induce a syn conformation of the guanosine in the GAC codon in the target sequence, wherein the nucleotide analog is preferably a modified purine nucleobase, more preferably wherein this modified purine nucleobase is selected from the group consisting of: 7-deaza-2’-deoxyadenosine (7-deaza Ad); 7-deaza-2’-adenosine (7- deaza A); 7-deaza-2’-deoxy-2’-fluoroadenosine (7-deaza Af); 7-deaza-2’-deoxy-2’-ara- fluoroadenosine; 7-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-deoxyadenosine (3- deaza Ad); 3-deaza-2’-adenosine (3-deaza A); 3-deaza-2’-deoxy-2’-fluoroadenosine (3-deaza Af); 3-deaza-2’-deoxy-2’-ara-fluoroadenosine; 3-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3,7-P5117PC00 dideaza-2’-deoxyadenosine (3,7-dideaza Ad); 3,7-dideaza-2’-adenosine (3,7-dideaza A); 3,7- dideaza-2’-deoxy-2’-fluoroadenosine (3,7-dideaza Af); 3,7-dideaza-2’-deoxy-2’-ara- fluoroadenosine; 3,7-dideaza-2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-O-[2- (methoxy)ethyl] adenosine; 3-deaza-2’-O-[2-methylamino-2-oxoethyl] adenosine; 2’-deoxy-2’- fluoroguanosine; 2’-ara-fluoro guanosine (FANA G); 2’,2’-difluoro guanosine; 2’-deoxyinosine (Id); 2’-OH-inosine; 2’-fluoroinosine (2’-F-I); 2‘-ara-fluoro inosine (FANA I); 2’,2’-difluoro inosine; 5-formylindole-2’-deoxyriboside; 5-formyl-2’-fluoro-2’-deoxyriboside; 5-formylindole- 2’-ara-fluoro-2’-deoxyriboside; 5-formylindole-2’,2’-difluoro-2’-deoxyriboside; 5-formylindole- 2’-O-methylriboside; 5-formylindole-2’-O-[2-(methoxy)ethyl]riboside; 5-formylindole-2’-O-[2- methylamino-2-oxoethyl]riboside; beta-(4-amidino-1H-imidazol-1-yl) riboside; beta-(4- amidino-1H-imidazol-1-yl) 2’-deoxyriboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-ara-fluoro- 2’-deoxyriboside; and beta-(4-amidino-1H-imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside.

7. An AON according to any one of claims 1 to 6, wherein the orphan nucleotide does not comprise a natural cytosine nucleobase and does not comprise a 2’-OMe substituted ribose.

8. An AON according to any one of claims 1 to 7, wherein the orphan nucleotide is a deoxycytidine or a deoxyuridine.

9. An AON according to anyone of claims 1 to 7, wherein the orphan nucleotide is a deoxynucleotide and is a cytidine analog or a uridine analog.

10. An AON according to claim 9, wherein the cytidine analog is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase.

11. An AON according to claim 9, wherein the uridine analog is a deoxynucleotide comprising an iso-uracil nucleobase.

12. An AON according to any one of claims 1 to 11, wherein the AON is 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length.

13. An AON according to any one of claims 1 to 12, wherein the AON comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP), sulfonylphosphoramidate, a (1,3-dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi), or a mesyl phosphoramidate (PNms) internucleotide linkage.

14. An AON according to claim 13, wherein the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkageP5117PC00 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.

15. An AON according to claim 13 or 14, wherein the linkage between the most terminal two nucleotides on the 5’ and / or 3’ terminus of the AON is a PNdmi or a PNms linkage.

16. An AON according to any one of claims 1 to 15, wherein the AON comprises one or more nucleotides comprising a mono- or di-substitution 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.

17. 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.

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

19. A nanoparticle delivery vehicle formulation according to claim 18, wherein the nanoparticle delivery vehicle is a Lipid Nanoparticle (LNP).

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

21. An AON according to any one of claims 1 to 16, a vector according to claim 17, or a nanoparticle delivery vehicle formulation according to claim 18 or 19, for use in the treatment of classic galactosemia, preferably caused by a deficiency of galactose-1-phosphate uridylyltransferase (GALT).

22. Use of an AON according to any one of claims 1 to 16, a vector according to claim 17, or a nanoparticle delivery vehicle formulation according to claim 18 or 19, in the manufacture of a medicament for the treatment of classic galactosemia, preferably caused by a GALT deficiency.P5117PC00 23. A method of editing a human GALK1 polynucleotide in a cell, wherein the human GALK1 polynucleotide is a pre-mRNA or mRNA molecule, the method comprising contacting the GALK1 polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, thereby editing the GALK1 polynucleotide to encode a GALK1 protein with a loss-of-function (LOF), preferably a LOF in the ability to phosphorylate galactose.

24. A method of treating, ameliorating, or slowing down the progression of classic galactosemia, the method comprising administering to said subject an AON according to any one of claims 1 to 16, a vector according to claim 17, a nanoparticle delivery vehicle formulation according to claim 18 or 19, or a pharmaceutical composition according to claim 20, thereby contacting a GALK1 transcript molecule in a cell of the subject with an AON that can initiate an ADAR-mediated adenosine to inosine deamination, thereby editing the GALK1 polynucleotide to encode a GALK1 protein with a LOF, preferably by changing the aspartic acid residue at position 186 in the human GALK1 protein to a glycine residue, thereby treating the subject.

25. An in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in a GALK1 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: (i) providing the cell with an AON according to any one of claims 1 to 16; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the GALK1 pre-mRNA or mRNA molecule; and (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine.

26. An in vivo method according to claim 25, comprising step (v) of assessing a biomarker to identify the presence of the GALK1 protein with the LOF.

Images