Antisense oligonucleotides for the treatment of metabolic disorders

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 melanocortin 4 receptor (MC4R) gene that encodes the MC4R protein. The RNA editing of the target adenosine, changing an isoleucine residue to a valine residue at position 317 in the protein (I317V), results in a gain-of-function of the MC4R protein, which will result in loss of body weight and lower BMI, thereby lowering the risk of suffering from disorders related to obesity.

Description

[0001] ANTISENSE OLIGONUCLEOTIDES FOR THE TREATMENT OF METABOLIC DISORDERS

[0002] TECHNICAL FIELD

[0003] This disclosure relates to the field of medicine, and in particular to the field of metabolic disorders like obesity, and related diseases such as type II diabetes and coronary artery disease. The disclosure describes antisense oligonucleotides that mediate nucleotide-specific RNA editing in the human melanocortin 4 receptor (MC4R) gene transcript to bring about amino acid changes of the encoded MC4R protein that influence its activity.

[0004] BACKGROUND

[0005] The melanocortin system controls a range of physiological functions mediated via a family of five melanocortin receptors (MC1 R to MC5R) that were identified and cloned in the early 1990s. The system, for example, regulates seemingly distinct processes such as hair colour, skin tanning, sexual function, adrenocortical steroidogenesis, exocrine gland function and energy homeostasis. The distinct physiological roles of the melanocortin peptides, encoded by the proopiomelanocortin (Pome), agouti and agouti-related peptide (Agrp) genes, result from the expression of the different melanocortin receptor subtypes in various tissues. MC1 R is expressed primarily in the skin, where it regulates pigmentation. MC2R is exclusively expressed in the adrenal gland, where it is the endogenous receptor for adrenocorticotropic hormone. MC5R is expressed in exocrine glands, whereas MC3R and MC4R are expressed predominantly in the central nervous system (CNS). Expression of MC4R has also been detected in the gastrointestinal system and liver.

[0006] Interest in the melanocortin system’s role in metabolism and energy homeostasis emerged following the generation of Mc4r knockout mice models and the pharmacological manipulation of this receptor (Butler AA and Cone RD. 2002. Neuropeptides 36:77-84; Oilman Mm et al. 1997. Science 278:135-138; Fan W et al. 1997. Nature 385:165-168; Huszar D et al. 1997. Cell 88:131-141). Mc4r knockout mice were hyperphagic and severely obese, resembling the phenotype observed in the leptin-deficient (ob / ob), leptin receptor-deficient (db / db) and Pome knockout mice. That MC4R is an important influencer of obesity was further confirmed by a study in which more than 640,000 exomes of humans were sequenced and several genetic variants in 16 different genes were found to be associated with the occurrence or the protection from obesity (Akbari P et al. 2021. Science 373:eabf8683). The authors confirmed protein loss-of-function and deleterious missense variants in MC4R with higher BMI and that of gain-of-function (GOF) variants in MC4R (Val103lle [V103I] and lle251 Leu [1251 L]) with lower BMI. These genetic findings suggest the convergence of a common neuronal pathway critical for feeding and energy homeostasis. It has been determined that the genes encoding MC4R, leptin, leptin receptor, POMC and AgRP compose the central leptin- melanocortin pathway, a critical CNS pathway that regulates feeding and body weight, a system that appears to be well-conserved from rodents to humans. Following the establishment of the critical role of this system in body weight control, intense effort has been focused on developing pharmacological compounds that target the melanocortin system for treating metabolic disorders. For instance, pharmacological compounds targeting the MC4R have been pursued as potential treatments for obesity.

[0007] The endogenous agonists of the CNS receptors MC3R and MC4R are produced by the POMC-expressing neurons in the hypothalamic arcuate nucleus (ARH) and the nucleus of the solitary tract (NTS). These neurons produce the peptide POMC, which is post- translationally processed into peptide products (a-, p-, and y-MSH) that stimulate MC3R and MC4R, with a varying affinity for each receptor. Genetic deletion of Pome in mice resulted in hyperphagic obesity, impaired glucose homeostasis and reduced energy expenditure. Consistently, humans with LOF mutations in POMC also develop early-onset obesity, associated with both hyperphagia and reduced energy expenditure, along with glucocorticoid insufficiency and red hair due to the absence of adrenocorticotropic hormone and MSH peptides. The primary role of POMC in regulating feeding and body weight is probably determined by POMC neurons in the ARH.

[0008] FDA approval was granted for a drug named lorcaserin, which is a 5-HT2C receptor agonist. The activation of the 5-HT2C receptor results in increased POMC production, which in turn stimulates MC3R and MC4R, as discussed above. Although lorcaserin was marketed and appeared effective in some patients, the average placebo-controlled weight loss following one year of treatment was approximately 5%, thus necessitating improved efficiency. Furthermore, severe side-effects such as psychiatric dependencies and high risk of cancer prompted the FDA to withdraw the drug from the market.

[0009] Early efforts to develop potent melanocortin agonists to promote skin tanning and treat obesity, via the discovery of super-potent pan-receptor agonists, revealed penile erection as an adverse effect in preclinical animal models and humans. The sexual adverse effects of melanocortin agonists are presumably due to the action of MC4R in autonomic control regions of the hindbrain that directly innervate the reproductive organs. This serendipitous discovery led to further interest in the utility of melanocortin receptor agonists for treating erectile dysfunction and other disorders of altered sexual function. However, the availability of alternative therapeutics and the potential adverse effects have largely impeded the pursuit of melanocortin-based therapeutics for sexual dysfunction in men. In contrast, a drug named bremelanotide was initially investigated as a melanocortin receptor agonist for treatment of female hypoactive sexual desire disorder. It is a non-selective melanocortin receptor agonist. It acts on MC1 R to MC5R (although predominantly on MC3R and MC4R) and is an amino- terminal deaminated metabolite of the non-selective seven-residue lactam-cycle analogue of a-MSH. In randomized, double-blind control trials the compound showed modest improvements in sexual desire, while it was not associated with severe adverse effects, although nausea was reported in nearly 40% of patients receiving the drug. Weight loss after intake of bremelanotide in obese women has been reported recently (Spana C et al. 2022. Diabetes Obese Metab 24(6): 1084-1093), but the drug has not yet been approved for this indication.

[0010] Setmelanotide is a recently FDA- and EMA-approved MC4R agonist that produces profound weight loss in genetic obesity patients with leptin receptor deficiency or POMC deficiency and shows promise in treating other forms of syndromic obesity, such as Bardet- Biedl syndrome, Alstrdm syndrome, and Prader-Willi syndrome. Setmelanotide did not show any positive effect in comparison to placebo in Ph1b trial in heterozygous individuals with defective MC4R functionality. Importantly, the mechanism governing the specific efficacy of setmelanotide in treating syndromic obesity without affecting diet-induced obesity remains unknown. It is hypothesized that increasing the dosing may influence diet-induced obesity, but that such higher dosing may surmount the safe therapeutic window that avoids undesired adverse effects, such as changes in systemic blood pressure.

[0011] Although the field related to MC4R agonists to treat metabolic disorders such as obesity (and consequently e.g., type II diabetes) and sexual dysfunction is showing significant progress, the use of peptide compounds always holds the risk of undesired adverse effects, as indicated above. Hence, there is a constant need for improved and other mechanistic efforts to target MC4R and thereby increase its activity when for instance weight loss is desired.

[0012] SUMMARY OF THE INVENTION

[0013] Disclosed herein is an antisense oligonucleotide (AON) that is capable of recruiting an endogenous ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human melanocortin 4 receptor (MC4R) gene encoding the MC4R protein. Preferably, the transcript molecule is a pre-mRNA or an mRNA molecule. Preferably, the cell is a neuronal cell, more preferably a cell in the CNS. In one aspect, the MC4R gene is wildtype, and the deamination of the target adenosine results in a GOF of the MC4R protein. Preferably, the target adenosine is the adenosine in the AUG codon coding for isoleucine (I) at position 317 of the MC4R protein, and wherein the deamination of the adenosine changes the amino acid to a valine (V). 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(1 H)- pyridone nucleobase. In one aspect, the orphan nucleotide is a uridine analog such as a deoxynucleotide comprising an iso-uracil nucleobase. In one aspect the nucleotide numbering in the AON is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and the first nucleotide 3’ from the orphan nucleotide (-1) is a deoxyinosine.

[0014] 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, PNdmi internucleotide linkage, or a linkage moiety with the structure according to formula (I), wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group.

[0015] 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 and negatively (-) incremented towards the 3’ end, and wherein linkage position -2 is an MP linkage.

[0016] 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 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.

[0017] Disclosed herein is a pharmaceutical composition comprising an AON as disclosed herein, or a vector as disclosed herein, and a pharmaceutically acceptable carrier. Disclosed herein is an AON for use in the treatment of a metabolic disease such as obesity, caused either by diet or due to a genetic mutation in a gene involved in the central melanocortin system. Disclosed herein is also an AON for use in the manufacture of a medicament for the treatment of a metabolic disease, preferably obesity or a disorder related to obesity such as type II diabetes or coronary artery disease.

[0018] Disclosed herein is a method of editing a human MC4R polynucleotide in a cell, preferably a cell in the CNS, wherein the human MC4R polynucleotide is a pre-mRNA or mRNA molecule, the method comprising contacting the MC4R polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, thereby editing the MC4R polynucleotide to encode an MC4R protein with a GOF. Preferably the method comprises the step of deaminating the adenosine in the codon AUG, codon for isoleucine at position 317 in the MC4R protein, thereby rendering the change from AUG to IUC, which is read as GUC by the translation machinery, and thereby changing the codon for isoleucine to valine (1317V).

[0019] Disclosed herein is a method of treating, ameliorating, or slowing down the progression of a metabolic disease, such as obesity, in a human subject in need thereof, the method comprising administering to said subject an AON as disclosed herein, a vector as disclosed herein, or a pharmaceutical composition as disclosed herein, thereby contacting a MC4R polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination at position 949 in the human MC4R coding sequence, thereby editing the MC4R polynucleotide to encode an MC4R protein with a GOF, thereby treating the subject.

[0020] BRIEF DESCRIPTION OF THE DRAWINGS

[0021] One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0022] Fig. 1A shows the 5’ to 3’ sequence of part of the human MC4R mRNA transcript in which the AUC codon coding for isoleucine (I) at position 317 in the MC4R protein is in bold (SEQ ID NO:1). The underlined adenosine is the target for RNA editing as disclosed herein, resulting in a codon coding for a valine (V) residue (IUC / GUC) at this position after editing. SEQ ID NO:2 represents the basic 32 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-01 to MC4R-04. SEQ ID NO:3 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-05 to MC4R-08. SEQ ID NO:4 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-09 to MC4R-12. SEQ ID NO:5 represents the basic 26 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-13 to MC4R-16. SEQ ID NO:6 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-17 to MC4R- 20. SEQ ID NO:7 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-21 to MC4R-24. SEQ ID NO:8 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AONs MC4R-25 to MC4R-29. SEQ ID NO:9 represents the basic 27 nucleotide sequence (without modifications, except for the orphan nucleotide and position -1) in AON MC4R-30. Fig. 1 B shows the sequences (5’ to 3’) of the 30 initial AONs that were designed to cause the deamination of the target adenosine in SEQ ID NO:1. AONs MC4R-01 to MC4R-30 (with modifications) are SEQ ID NO: 10 to 39, respectively. The chemical modifications are as follows: Ae and Ge are 2’-MOE modified adenosine and guanosine, respectively; Cm, Am, Um, and Gm are 2’-OMe modified cytidine, adenosine, uridine, and guanosine, respectively; Gf, Cf, Af, and Uf are 2’-F modified guanosine, cytidine, adenosine, and uridine, respectively; 5mCe is 2’-MOE modified 5-methyl-cytidine; Zd (orphan nucleotide), in bold face, is a deoxynucleotide (deoxycytidine analog) carrying a Benner’s base; Id is deoxyinosine; Te is 2’-MOE modified thymidine (identical to a 5-methyluridine with a 2’-MOE substitution); “I” refers to a PNdmi linkage; “A” refers to a MP linkage; “*” refers to a PS linkage. All other internucleoside linkages (no symbol here) are phosphodiester linkages.

[0023] Fig. 2 shows a further set of AONs (SEQ ID NO:44 to 93 and SEQ ID NO:97 to 158, with their respective RM names) that was designed and tested to cause the deamination of the target adenosine depicted in Fig. 1 (SEQ ID NO:1). The chemical modifications are as provided in Fig. 1 B, wherein m5Ue is 2’-MOE modified 5-methyluridine (Te in Fig. 1 B); m5CI, Al, and m5UI are locked nucleic acids with a 5-methylcytidine, adenine and 5-methyluridine nucleobase, respectively; Ad is deoxyadenosine; and “e” represents a phosphodiester linkage.

[0024] Fig. 3A and Fig. 3B show the editing percentages of the human MC4R transcript at the adenosine underlined in SEQ ID NO:1 in Fig. 1 , after transfection of the AONs provided in Fig. 2 with SEQ ID NO:44 to 93, in human HeLa cells that overexpress human wt MC4R. The (negative) controls were the use of two unrelated AONs (RM4266 and RM4777), a mock transfection control (no AON), a non-treated (NT) sample with only PBS and with cells only and two gBIock controls, that only contained the wt MC4R DNA to check for the PCR assay.

[0025] Fig. 4A and Fig. 4B show the editing percentages of the human MC4R transcript at the adenosine underlined in SEQ ID NO:1 in Fig. 1 , after gymnotic treatment with the AONs provided in Fig. 2 with SEQ ID NO:44 to 93, in human HeLa cells that overexpress human wt MC4R, using a 4-hr pulse treatment with the AG1856 saponin, as detailed in the accompanying example. The (negative) controls were the use of two unrelated AONs (RM4266 and RM4777), a mock control (no AON), a non-treated (NT) sample with cells only, and two gBIock controls, that only contained the wt MC4R DNA to check for the PCR assay.

[0026] Fig. 5A and Fig. 5B show the editing percentages of the human MC4R transcript at the adenosine underlined in SEQ ID NO:1 in Fig. 1 , after transfection of the AONs provided in Fig. 2 with SEQ ID NO:97 to 158, in human HeLa cells that overexpress human wt MC4R. Both figures are of the same experiment, displaying each a subset of the tested AONs in comparison to RM 117697 (dotted line) and the other controls. The (negative) controls were the use of two unrelated AONs (RM4266 and RM4777), a mock transfection control (no AON), a non-treated (NT) sample, and a sample with PBS only.

[0027] Fig. 6 shows the editing percentages of the human MC4R transcript at the adenosine underlined in SEQ ID NO:1 in Fig. 1 , after transfection of the specified AONs (which is a subset of the AON set of SEQ ID NO:97 to 158, and as shown in Fig. 5A and Fig. 5B) in human HeLa cells that overexpress human wt MC4R. The controls are as provided in Fig. 5A and Fig. 5B. DETAILED DESCRIPTION

[0028] The present invention aims to provide one or more alternative, and / or improved, compounds or compositions for use in the treatment of metabolic disease, such as obesity, resulting from: a diet, a genetic mutation in the relevant MC4R metabolic pathway, or any other cause.

[0029] The present disclosure relates to a completely different approach of increasing the activity of MC4R, namely by using antisense oligonucleotides (AONs) and the cell’s own nucleic acid post-transcriptional modification machinery to specifically target and amend one or more nucleotides in the MC4R (pre-) mRNA transcript, thereby providing an MC4R protein with a gain-of-function (GOF). The technology that the present disclosure relates to is generally referred to as ‘RNA editing’. Interestingly, the invention is also useful for treating other eating disorders such anorexia nervosa and bulimia nervosa, wherein the introduction of a loss-of-function (LOF) mutation in the (wildtype) MC4R (pre-) mRNA causes a decrease in MC4R activity, which would be beneficial in ultimately providing weight gain, or providing a properly functioning metabolism and energy homeostasis system.

[0030] RNA editing is a natural process through which eukaryotic cells alter the sequence of their RNA molecules, often in a site-specific and precise way, thereby increasing the repertoire of genome encoded RNAs by several orders of magnitude. RNA editing enzymes have been described for eukaryotic species throughout the animal and plant kingdoms, and these processes play an important role in managing cellular homeostasis in metazoans from the simplest life forms (such as Caenorhabditis elegans) to humans. Examples of RNA editing are adenosine (A) to inosine (I) conversions and cytidine (C) to uridine (U) conversions, which occur through enzymes called Adenosine Deaminases acting on RNA (ADAR) and Apolipoprotein B mRNA Editing Enzyme, Catalytic Polypeptide I Activation-Induced Cytidine Deaminase (APOBEC / AID), respectively.

[0031] ADAR is a multi-domain protein, comprising of a catalytic domain and two to three double-stranded RNA (dsRNA) recognition domains, depending on the enzyme in question. Each recognition domain recognizes a specific dsRNA sequence and / or conformation. The catalytic domain does also play a role in recognizing and binding a part of the dsRNA helix, although the key function of the catalytic domain is to convert an A into 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 is in a coding region of an mRNA or pre-mRNA, it can recode the amino acid sequence. A-to-l conversions 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-l conversions may take place in splice elements in introns or exons in pre-mRNAs, thereby altering the pattern of splicing. As a result, exons may be (partially) included or excluded. The enzymes catalysing adenosine deamination are within an enzyme family of ADARs, which include human deaminases hADARI and hADAR2, as well as hADAR3. However, to date, for hADAR3 no deaminase activity has been demonstrated.

[0032] The use of oligonucleotides to edit a target RNA, applying adenosine deaminase, has been 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 recognition domain of bacteriophage Lambda N-protein, fused to the adenosine deaminase domain of a truncated natural ADAR protein. It requires target cells to be either transduced with the fusion protein, which is a major hurdle, or that target cells are transfected with a nucleic acid construct encoding the engineered adenosine deaminase fusion protein for expression. The system described by Vogel et al. (2014) suffers from similar drawbacks, in that it is not clear how to apply the system without having to genetically modify the ADAR first and subsequently transfect or transform the cells harboring the target RNA, to provide the cells with this genetically engineered protein. US 9,650,627 describes a similar system. The oligonucleotides of Woolf et al. (1995) that were 100% complementary to the target RNA sequences, suffered from severe lack of specificity: nearly all adenosines in the target RNA strand that were complementary to the AON were edited.

[0033] 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 adenosine on the target RNA. However, the specific editing effect at the target nucleotide has not been shown to take place without the use of recombinant ADAR enzymes having covalent bonds with the AON. Several publications have now shown that the recruitment of endogenous ADAR (hence without the need for an exogenous and / or recombinant source) is feasible while maintaining a specificity in which a single adenosine within a target RNA molecule can be targeted and deaminated to an inosine. WO2016 / 097212 discloses AONs for the targeted editing of RNA, wherein the AONs are characterized by a sequence that is complementary to a target RNA sequence (therein referred to as the ‘targeting portion’) and by the presence of a stem-loop I hairpin structure (therein referred to as the ‘recruitment portion’), which is preferably non-complementary to the target RNA. Such oligonucleotides are referred to as ‘self-looping AONs’. The recruitment portion 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 mimicking either a natural substrate (e.g., the GluB receptor) or a Z-DNA structure known to be recognized by the dsRNA binding domains, or Z-DNA binding domains, of ADAR enzymes. A stem-loop structure can be an intermolecular stem-loop structure, formed by two separate nucleic acid strands, or an intramolecular stem loop structure, formed within a single nucleic acid strand. The stem-loop structure of the recruitment portion as described is an intramolecular stem-loop structure, formed within the AON itself, and are thought to attract (endogenous) ADAR. Similar stem-loop structure-comprising systems for RNA editing have since then been described in WO2017 / 050306, W02020 / 001793, WO2017 / 010556, US11 ,390, 865, W02020 / 246560, and WO2022 / 078995.

[0034] WO2017 / 220751 and WO2018 / 041973 describe a next generation type of AONs that do not comprise such a stem-loop structure but that are (almost fully) complementary to the targeted area, 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 nucleoside opposite the target adenosine, but in other embodiments AONs (or “RNA editing oligonucleotides” - 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 ADAR enzymes when the sequence of the AON was carefully selected such that it could attract / recruit ADAR. The ‘orphan nucleoside’, which is defined as the nucleoside in the AON that is positioned directly opposite the target adenosine in the target RNA molecule, was a nucleotide with an unmodified cytosine nucleobase and that did not carry a 2’-OMe modification. The orphan nucleoside can be a deoxyribonucleoside (DNA), wherein the remainder of the AON could still carry 2’-O-alkyl modifications at the sugar entity (such as 2’- OMe), or the nucleotides directly surrounding the orphan nucleoside contained chemical modifications (such as DNA in comparison to RNA) that further improved the RNA editing efficiency and / or increased the resistance against nucleases. 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 that could be used in ADAR-mediated editing of specific adenosines in a target RNA have been the subject of numerous disclosures in the field, such as WO2019 / 111957, WO2019 / 158475, W02020 / 165077, W02020 / 201406, W02020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, W02022 / 007803, W02022 / 018207, WO2022 / 026928, and WO2022 / 124345. The use of specific sugar moieties has been disclosed in for instance W02020 / 154342, W02020 / 154343, W02020 / 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 target sequences) has been described in WO2011 / 005761 , W02014 / 010250, WO2014 / 012081 , WO2015 / 107425, WO2017 / 015575 (for HTT), WO2017 / 062862, W02017 / 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 , W02020 / 118246 (for DM1), W02020 / 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, and WO2023 / 278589. Next to these disclosures, an extensive number of publications relate to the targeting of specific RNA target molecules, or specific adenosines within such RNA target molecules, be it to repair a mutation that resulted in a premature stop codon, or other mutation causing disease. Examples of such disclosures in which adenosines are targeted within specified target RNA molecules are W02020 / 157008 and WO2021 / 136404 (USH2A); WO2021 / 113270 (APP); WO2021 / 113390 (CMT1A); W02021 / 209010 (IDUA, Hurler syndrome); WO2021 / 231673 and WO2021 / 242903 (LRRK2); WO2021 / 231675 (ASS1); WO2021 / 231679 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 and WO2021 / 231692 (OTOF, autosomal recessive non-syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); W02021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1).

[0035] Obesity is associated with type II diabetes and coronary artery disease, which together account for significant morbidity, mortality, and substantial healthcare costs globally. While advances in the understanding of the molecular mechanisms involved in weight regulation have informed the development of new weight-loss therapies, some drugs lack target specificity, while others affect multiple signalling pathways downstream of their intended target, leading to adverse effects that limit their long-term use. Therefore, there is a substantial unmet need for safe and effective weight-loss therapies.

[0036] The melanocortin 4 receptor (MC4R) is a brain-expressed G protein coupled receptor involved in weight regulation. Feeding-induced release of the melanocortin peptides a- and p- melanocyte-stimulating hormone (MSH) leads to activation of MC4R-expressing neurons, resulting in reduced food intake. Targeted deletion of Mc4nr\ rodents causes weight gain in a gene-dosage-dependent manner (Huszar et al. 1997). In humans, rare heterozygous MC4R variants have been identified in obese children and adults in many populations. MC4R deficiency in rodents and humans is characterized by low blood pressure due to impaired sympathetic nervous system activation. As predicted by these genetic findings, first- generation MC4R agonists caused weight loss but increased blood pressure, which halted their development. A second-generation MC4R agonist reduced weight in rare patients with obesity due to genetic disruption of the melanocortin pathway without affecting blood pressure. However, off-target effects, such as on MC1 R (causing skin pigmentation) may limit wider use. The inventors of the present invention realized that a highly specific targeting of the MC4R transcript (and thereby amending the translated protein) should lower the risk of such adverse side effects, because it would only target the (pre-) mRNA stemming from the MC4R gene, while not targeting any other player involved in human energy homeostasis.

[0037] Recently, a large study was published revealing 61 nonsynonymous variants in human MC4R, using data from 452,300 European ancestry participants aged 40-69 years (Lotta LA et al. 2019. Cell 177:597-607). Of these 61 variants, many appeared to have no influence on the function of MC4R and acted as wildtype, while many were considered LOF variants. Interestingly, a subset of individuals (6%, n = 28,161) were carriers of GOF alleles. The functionality was investigated by measuring canonical Gas-mediated signalling by quantifying the maximal efficacy of ligand (a-MSH)-induced cAMP production in a time-resolved assay in HEK293 cells transiently transfected with constructs encoding the variants, and by quantifying the interaction between wildtype / mutant MC4R and p-arrestin-2 using a time-resolved enzyme complementation assay. From genetic association studies it was found that individuals with the GOF variants had significantly lower BMI and up to 50% lower risk of obesity, type II diabetes, and coronary artery disease. Furthermore, in contrast to the LOF variants, the GOF variants were associated with lower diastolic blood pressure and lower resting heart rate. The inventors of the present invention realized that if any of these variants involved in GOF of MC4R would be eligible for RNA editing, that such would be a preferred target for the treatment of obesity, by very specifically altering the transcript for the wildtype MC4R protein in obese individuals, without affecting any other compound involved in human energy homeostasis. In fact, one of the GOF alleles that was identified is a mutation from isoleucine (encoded by an AUG codon in the mRNA) to valine (encoded by a GUC codon) at position 317 in the MC4R protein (1317V). Clearly, such RNA editing of this specific target adenosine could potentially treat obese patients that suffer from weight gain through their diet, as well as syndromic obesity, where a genetic background is the cause of the disorder. This means that the population that can potentially be treated through this RNA editing therapy is significantly large. For most if not all obese patients, it would mean that their wildtype MC4R transcripts are targeted (and edited) to yield a MC4R protein that has increased activity, thereby lowering the (need or urge of) food intake through the signalling pathways in their melanocortin system, ultimately resulting in weight-loss. In concert with this, introducing a LOF mutation in the MC4R (pre-) mRNA of patients suffering from anorexia nervosa or bulimia nervosa could also be proven to be therapeutically relevant to treat their metabolic disorder. One example of a LOF variant that can be introduced in the wildtype mRNA of MC4R by RNA editing is the His76Arg (H76R; c.227A>G) variant identified by Lotta et al. (2019).

[0038] RNA editing is not gene therapy, because it is not irreversible and does not target the patient’s DNA. The edited RNA disappears from the system, while the wildtype MC4R gene itself remains intact. 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 of the MC4R transcript is highly specific and will not affect other (pre-) mRNA molecules, the risk of adverse side effects as observed with MC4R agonists (discussed above) is low. For instance, transcripts of MC1R, MC2R, MC3R, and MC5R are not edited through the compounds as disclosed herein.

[0039] 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 can mediate RNA editing of a target adenosine present in a target RNA molecule after it is bound to the target RNA molecule, since the deaminating enzymes are recruited to the doublestranded AON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine.

[0040] The oligonucleotides are herein abbreviated to “AONs”, but sometimes also referred to as ‘editing oligonucleotides’, or ‘EONs’, even though the RNA editing event is performed by the 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 constantly prone to breakdown because of nucleases present in a natural cell. Many chemical modifications 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 and editing (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, but not all, positions in the AON appeared compatible with efficient engagement of an enzyme with nucleotide deamination activity and with subsequent deamination (W02020 / 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.

[0041] Disclosed herein are AONs that can provide (mediate, cause, or trigger) RNA editing of a target adenosine in a target transcript molecule, such as pre-mRNA and / or mRNA. The target transcript molecule may be encoded by a mutated gene, although in most cases regarding MC4R the originating gene is wildtype and the editing results in a GOF of the encoded protein. As disclosed in more detail herein, in one aspect, the target transcript molecule is encoded by a wildtype gene, wherein the target nucleic acid molecule is a transcript from a wildtype human MC4R gene as shown in the present disclosure, wherein the RNA editing makes that the encoded MC4R protein obtains an increased / improved activity, which ultimately results in weight loss of the treated subject. In another aspect, the MC4R gene may be mutated and harbouring a LOF mutation (elsewhere in the coding sequence), wherein the AON targets the adenosine in the isoleucine codon of position 317 to switch the MC4R protein from a less-functional state to a wildtype activity or, even more preferred to a level that is a GOF above the wildtype level.

[0042] RNA editing is often applied to correct G>A mutations that cause a disease. Nonlimiting examples of transcript molecules (as disclosed in the art) that are targeted using RNA editing for a variety of treatments are SERPINA1 (for the treatment of alphal -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 of Parkinson’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., W02021 / 130313 and WO2021 / 231830), USH2A (for the treatment of Usher syndrome; see e.g., WG2020 / 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 RNA editing to target an otherwise wildtype transcript molecule to a mutated version, which results in a GOF of the encoded protein, which is a next-level approach in the field of RNA modulation.

[0043] The present disclosure relates to AONs that mediate RNA editing, using endogenous (naturally present) ADAR enzymes in the host cell, preferably cells in the CNS that express MC4R, of one or more adenosines present in the transcript of the MC4R gene. An AON as disclosed herein aims to increase the activity of the MC4R protein in its function in the central leptin-melanocortin pathway, a critical CNS pathway that regulates feeding and body weight. Targeting the adenosine in the codon for isoleucine at position 317 in the MC4R protein and changing it to an inosine resulting in a valine at that position is a preferred example of a GOF variant. It cannot be excluded that other GOF variants in MC4R will be identified that may also be introduced through RNA editing. Hence, the present disclosure also relates to compositions comprising multiple AONs that each target a specific adenosine in the MC4R transcript and bring about mutations leading to a variety of GOF variants. Hence, it is not excluded that two or more adenosines may be targeted for deamination in a single treatment. Without wishing to be bound by theory a synergistic or additive effect may be obtained by combining AONs as disclosed herein for targeting a multitude of adenosines, and thereby a multitude of amino acids within a single MC4R protein, to increase the therapeutic effect.

[0044] Definitions

[0045] Whenever reference is made to an oligonucleotide, oligo, ON, ASO, oligonucleotide composition, antisense oligonucleotide, AON, (RNA) editing oligonucleotide, EON, and RNA (antisense) oligonucleotide, both oligoribonucleotides and deoxyoligoribonucleotides are meant unless the context dictates otherwise. Potentially the oligonucleotide may completely lack RNA and DNA nucleotides (as they appear in nature) and may consist completely of modified nucleotides. Whenever reference is made to an ‘oligoribonucleotide’ it may comprise the bases A, G, C, II, 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, hence without 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 often abbreviated without the ‘d’, and often abbreviated with their respective modifications and as explained herein.

[0046] The term ‘nucleoside’ refers to the nucleobase linked to the (deoxy)ribosyl sugar, without phosphate groups. A ‘nucleotide’ is composed of a nucleoside and one or more phosphate groups. The term ‘nucleotide’ thus refers to the respective nucleobase- (deoxy)ribosyl-phospholinker, as well as any chemical modifications of the ribose moiety or the phospho group. Thus, the term would include a nucleotide including a locked ribosyl moiety (comprising a 2’-4’ bridge, comprising a methylene group or any other group), an unlocked nucleic acid (UNA), a threose nucleic acid (TNA), a nucleotide including a linker comprising a phosphodiester, phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP (or MeP), methyl thiophosphonate, phosphoramidate linkages, PNdmi, 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 nucleoside is linked to a neighbouring nucleoside and the linkage between these nucleosides is modified. As stated herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms ‘ribonucleoside’ and ‘deoxyribonucleoside’, or ‘ribose’ and ‘deoxyribose’ are as used in the art.

[0047] Sometimes the terms adenosine and adenine, guanosine and guanine, cytidine and cytosine, uracil and uridine, thymine and thymidine / uridine, inosine, and hypoxanthine, are used interchangeably to refer to the corresponding nucleobase on the one hand, and the nucleoside or nucleotide on the other. The nucleobase thymine (T) is also known as 5- methyluracil (m5U) and is a 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 be interchanged throughout the document text.

[0048] Whenever reference is made to nucleotides in the oligonucleotide, such as cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5- hydroxycytosine, and p-D-glucosyl-5-hydroxymethylcytosine are included. Whenever reference is made to adenine, N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine and 7-methyladenine are included. Whenever reference is made to uracil, di hydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1-methylpseudouracil, 4-thiouracil and 5-hydroxymethyluracil are included. Whenever reference is made to guanine, 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7- trimethylguanosine and N2,7-dimethylguanosine are included. Whenever reference is made to nucleosides or nucleotides, ribofuranose derivatives, such as 2’-deoxy, 2’-hydroxy, and 2’- O-substituted variants, such as 2’-O-methyl (2’-0Me), are included, as well as other modifications, including 2’-4’ bridged variants. Whenever reference is made to 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, thiophosphoryl guanidine, sulfono phosphoramidate, PNdmi and the linkage structure according to formula (I), further outlined in detail below.

[0049] 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%.

[0050] 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.

[0051] The term ‘conducive to’ or ‘mediate’ can be used interchangeably with ‘capable of facilitating’. When used in the context of an AON that is conducive to ADAR editing (or can mediate 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 after binding to the target RNA molecule.

[0052] The term ‘mismatch’ is used herein to refer to opposing nucleotides in a double stranded RNA complex which do not form perfect base pairs according to the Watson-Crick base pairing rules. In the historical sense, mismatched nucleotides are G-A, C-A, ll-C, A-A, G-G, C-C, Il-Il pairs. In some embodiments AONs as disclosed herein comprise fewer than four mismatches with the target sequence, for example 0, 1 or 2 mismatches. ‘Wobble’ base pairs are G-ll, l-ll, l-A, and l-C base pairs. When a II is placed opposite the target A, there is no mismatch, and the AON may be 100% complementary, although an iso-uridine (iso-ll) opposite the target adenosine qualifies as a mismatch, since it does not pair according to the Watson-Crick rules of base pairing. When a C is placed opposite the target A, there is at least 1 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 the nucleotide but still be relatively stable. An isolated G:G pairing in duplex RNA can for instance be 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 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 important to the development of efficient ADAR-based AON therapy.

[0053] 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) when it forms a double stranded complex with the target RNA molecule. The term does not necessarily mean that each nucleotide in a nucleic acid strand has a perfect pairing with its opposite nucleotide in the opposite sequence. In other words, while an AON may be complementary to a target sequence, there may be mismatches, wobbles and / or bulges between the AON and the target sequence, while under physiological conditions that AON still hybridizes to the target sequence such that the cellular RNA editing enzymes can deaminate the 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 be complementary, but may also comprise one or more mismatches, wobbles and / or bulges with the target sequence, if under physiological conditions the AON is able to hybridize to its target.

[0054] 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 a deoxyuridine. It may also be a chemically modified nucleotide, as further described in detail below, 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.

[0055] A ‘nucleotide analog’ refers to an analog of a nucleic acid nucleotide. The nucleotide analog is an analog of adenosine, guanosine, cytidine, thymidine, uridine, deoxyadenosine, deoxyguanosine, deoxycytidine, deoxythymidine or deoxyuridine.

[0056] 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 but is downstream of the stop codon in the antisense strand. The same holds true for the AONs as 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.

[0057] The nucleotide ‘numbering’ in an AON as disclosed herein is such that the orphan nucleotide 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 oligonucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end.

[0058] References to ‘hybridisation’ typically refer to specific hybridisation and exclude non-specific hybridisation. Specific hybridisation can occur under experimental conditions chosen, using techniques well known in the art, to ensure that most stable interactions between probe and target are where the probe and target have at least 70%, preferably at least 80%, more preferably at least 90% sequence identity.

[0059] The term ‘splice mutation’ relates to a mutation in a gene that encodes fora 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 are degraded rapidly and do not have any functional activity.

[0060] 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 therefore different from a form in which the AON is encoded (and delivered) by a viral genome or within a 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.

[0061] The length of the AON as disclosed herein, and when delivered in a naked form is preferably 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides 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, or 200 or more nucleotides in length.

[0062] 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 sense strand 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’-0Me substitution, a 2’-F substitution, or a 2’-M0E substitution. It is to be understood that the sense strand present in the HEON is a different entity in comparison to 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. 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.

[0063] Embodiments

[0064] Disclosed here is an AON that is capable of recruiting an endogenous ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human MC4R gene encoding the MC4R protein. In one aspect, the transcript molecule is a pre-mRNA or an mRNA molecule. In one aspect, the cell is a neuronal cell, preferably a cell in (or “from” when RNA editing is performed in in vitro cell cultures) the CNS. In one aspect, the MC4R gene is wildtype, and the target adenosine is the adenosine in the AUG codon coding for isoleucine (I) at position 317 of the MC4R protein, and wherein the deamination of the adenosine changes the amino acid to a valine (V), and wherein the deamination of the target adenosine results in an MC4R protein that is stimulated in its function (gain-of-function; GOF). In one aspect, one or more AONs are combined to target different adenosines, preferably the adenosines referred to herein, each resulting in an amino acid change in the resulting MC4R protein and each independently (but potentially synergistically or additively) contributing to the GOF of the MC4R protein.

[0065] In one aspect, the oligonucleotide as disclosed herein comprises or consists of the sequence of any one of SEQ ID NO:8, 129, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 99, 101 , 103, 105, 115, 116, 117, 118, 119, 120, 121 , 122, 127, 128, 130, 131 , 132, and 133. In a preferred aspect, the oligonucleotide as disclosed herein comprises the nucleotide sequence of SEQ ID NO:8, in which the nucleotide comprising the Benner’s base and the inosine at -1 are deoxynucleotides, and preferably wherein the adenosine at position +1 is also a deoxynucleotide. In one aspect, an AON as disclosed herein comprises or consists of any of the sequences provided in Figs. 1A, 1 B, and 2, optionally including the indicated chemical modifications to the nucleobase, sugar moiety and / or linkage, or any combination thereof. In one aspect, the orphan nucleotide in an AON as disclosed herein does not comprise a natural cytosine nucleobase and does not comprise a 2’-0Me substituted ribose. In one aspect, the orphan nucleotide in an AON as disclosed herein is a deoxycytidine or a deoxyuridine. In one aspect, the orphan nucleotide in an AON as disclosed herein is a cytidine analog or a uridine analog. In one aspect, a cytidine analog is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1 H)-pyridone nucleobase. In one aspect, a uridine analog is a deoxynucleotide comprising an iso-uracil nucleobase.

[0066] In one aspect, the nucleotide numbering in an AON as disclosed herein is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein the first nucleotide 3’ from the orphan nucleotide (-1) is a deoxyinosine. The nucleotide opposite this position is a cytidine in the target RNA nucleic acid molecule of MC4R.

[0067] 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 naked form but is expressed from an expression vector such as a plasmid. In one aspect, when the AON 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.

[0068] In one aspect, an AON as disclosed herein comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP), sulfonylphosphoramidate, PNdmi internucleotide linkage, or a linkage moiety with the structure according to formula (I) wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group.

[0069] 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 linkage or a PNms linkage. In one aspect, the linkage between the most terminal two nucleotides on the 5’ and / or 3’ terminus of the AON as disclosed herein is a PNdmi linkage or a PNms linkage.

[0070] 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 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.

[0071] Disclosed herein is a vector, preferably a viral vector, more preferably an adeno- associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON as disclosed herein.

[0072] Disclosed herein is a pharmaceutical composition comprising an AON as disclosed herein, or a vector as disclosed herein, and a pharmaceutically acceptable carrier.

[0073] Disclosed herein is an AON for use in the treatment of a metabolic disease such as obesity, or a disease that is related to the obesity, such as type II diabetes and / or coronary artery disease.

[0074] Disclosed herein is a use of an AON as disclosed herein, for use in the manufacture of a medicament for the treatment of a metabolic disease such as obesity, or a disease that is related to the obesity, such as type II diabetes and / or coronary artery disease.

[0075] Disclosed herein is a method of editing a human MC4R polynucleotide in a cell, preferably a neuronal cell of the CNS, wherein the human MC4R polynucleotide is a pre- mRNA or mRNA molecule, the method comprising contacting the MC4R polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, thereby editing the MC4R polynucleotide to encode an MC4R protein with a GOF.

[0076] Disclosed is a method of treating, ameliorating, or slowing down obesity in a human subject in need thereof, the method comprising administering to said subject an AON as disclosed herein, or a vector as disclosed herein, or a pharmaceutical composition as disclosed herein, thereby contacting a MC4R polynucleotide, preferably a pre-mRNA or an mRNA transcript molecule, such as transcribed from a wildtype MC4R gene, in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination, thereby editing the MC4R polynucleotide to encode an MC4R protein with a GOF.

[0077] Disclosed herein is an in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in an MC4R pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the MC4R 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 presence of the inosine in the target RNA molecule. The most preferred read-out is significant weight loss after treatment. In a preferred aspect, the cell is a neuronal cell, more preferably a cell of the CNS. In one aspect, a method as disclosed herein comprises the step of administering a triterpene glycoside before, after or simultaneously with administering the AON, wherein in a preferred aspect, the triterpene glycoside is AG 1856. The triterpene glycoside may also be conjugated to the AON using a hydrazone linker or any other suitable linker.

[0078] Chemical modifications

[0079] 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 is complementary to the AON, when the AON and the complementary strand form a HEON complex, such as described in W02024 / 084048 and as disclosed above, except that the opposite sense strand does not have an orphan nucleotide. Hence, the modification related to the orphan nucleotide relate only to the AON as disclosed herein, but all other modifications relate to the AON as disclosed herein and any (protecting) sense oligonucleotide that may be used together with the AON in a pharmaceutical product. This includes the use of hydrophobic moieties (such as tocopherol and cholesterol) and cell-specific ligands, that have also been described herein, and in detail in W02024 / 084048, which may either be bound to the AON or its opposite strand, or both.

[0080] The skilled person knows that an oligonucleotide, such as an AON as outlined herein, generally consists of repeating monomers. Such a monomer is most often a nucleotide or a chemically modified nucleotide. The most common naturally occurring nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (II). These consist of a pentose sugar, a ribose, a 5’-linked phosphate group which is linked via a phosphate ester, and a T-linked base. The sugar connects the base and the phosphate and is therefore often referred to as the “scaffold” of the nucleotide. A modification in the pentose sugar is therefore often referred to as a ‘scaffold modification’. The original pentose sugar may be replaced in its entirety by another moiety that similarly connects the base and the phosphate. It is therefore understood that while a pentose sugar is often a scaffold, a scaffold is not necessarily a pentose sugar. Examples of scaffold modifications that may be applied in the monomers of the AON as disclosed herein are disclosed in W02020 / 154342, W02020 / 154343, and W02020 / 154344.

[0081] A nucleoside in the AON as disclosed herein may be a natural nucleoside (deoxyribonucleoside or ribonucleoside) or a non-natural nucleoside. It is noted that for RNA editing, in which double-stranded RNA is generally the substrate for enzymes with deamination activity (such as ADARs), ribonucleosides are considered ‘natural’, while deoxyribonucleosides may then be, for the sake of argument, considered as non-natural, or considered as modified, simply because DNA is not present in the RNA-RNA double stranded (natural) substrate configurations. The skilled person appreciates that when the nucleotide has a natural ribose moiety, it may still be non-naturally modified in the base and / or the linkage.

[0082] 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 delivery vehicle such as a viral vector or plasmid), the biodistribution and the resistance to nuclease-mediated breakdown. The skilled person is aware, and it has been described in detail in the art, that a variety of chemical modifications can assist in overcoming such limitations. Examples of such now commonly used chemical modifications are the 2’-0Me, 2’-F and 2’-MOE modifications of the sugar and the use of PS linkages between nucleosides, as described herein.

[0083] Scaffold modifications (ribose)

[0084] The ribose 2’ groups in all nucleotides of the AON as disclosed herein, except for the ribose sugar moiety of the orphan nucleotide that has certain limitations in respect of compatibility with RNA editing, can be independently selected from 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-0Me, 2’-MOE, 2’-F, or 2’-4’-linked (for instance a locked nucleic acid (LNA)), or other ribosyl T-substitutions, 2’ substitutions, 3’ substitutions, 4’ substitutions or 5’ substitutions. The orphan nucleotide in the AON that comprises no other chemical modifications to the ribose sugar, the base, or the linkage preferably does not carry a 2’-0Me or 2’-MOE substitution when the nucleobase is a naturally occurring cytosine, but may carry a 2’-F, a 2’,2’-difluoro (diF), or 2’-ara-F (FANA) substitution or may be DNA. WO2024 / 013360 describes the modification of the 2’ position of the ribose sugar moiety of the orphan nucleotide by a 2’,2’-disubstituted substitution such as diF, which is also applicable to what is disclosed here. The 2’-4’ linkage can be selected from many linkers known in the art, such as a methylene linker, amide linker, or constrained ethyl linker (cEt). An AON as disclosed herein may comprise one or more nucleotides carrying a 2’-MOE ribose modification. Also, an AON as disclosed herein may comprise one or more nucleotides not carrying a 2’-MOE ribose modification, or wherein the 2’-MOE ribose modifications are at positions that do not prevent the enzyme with adenosine deaminase activity from deaminating the target adenosine. An AON as disclosed herein may comprise a 2’-0Me ribose modification at a position that does not comprise a 2’-MOE ribose modification. An AON as disclosed herein may comprise deoxynucleotides at positions that do not comprise a 2’-MOE or a 2’-0Me 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’-0Me, 2’-OH, 2’-deoxy, TNA, 2’-fluoro (2’-F), 2’,2’-difluoro (diF) modification, 2’-fluoro-2’-C-methyl modification, or a 2’-4’-linkage (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA or examples mentioned in e.g. WO2018 / 007475)). Other nucleic acid monomers that may be used in an AON as disclosed herein are arabinonucleic acids and 2’-deoxy-2’-fluoroarabinonucleic acid (FANA), for instance for improved affinity purposes. The 2’-4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker. A wide variety of 2’ modifications that may present in an AON as disclosed herein are known in the art, including but not limited to the modifications outlined in detail in WO2016 / 097212, WO2017 / 220751 , WO2018 / 041973, WO2018 / 134301 , WO2019 / 219581 , WO2019 / 158475, and WO2022 / 099159. In all cases, the modifications should be compatible with RNA editing such that the AON fulfils its role as an oligonucleotide that can form a double stranded complex with the target RNA and by generating this double-stranded nucleic acid complex, recruit a deaminating enzyme, which can subsequently deaminate the target adenosine. Where a monomer in an AON as disclosed herein comprises an unlocked nucleic acid (UNA) ribose modification, that monomer can have a 2’ position comprising the same modifications discussed above, such as a 2’-MOE, a 2’-OMe, a 2’-OH, a 2’-deoxy, a 2’-F, a 2’,2’-diF, a 2’- fluoro-2’-C-methyl, an arabinonucleic acid, a FANA, or a 2’-4’-linkage (i.e., a bridged nucleic acids such as a locked nucleic acid (LNA)). In an 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 sugar moiety that comprises a 2’-fluoro (2’-F) modification. A preferred position for the nucleotide that carries a 2’-F modification is position -3 in the AON, which may be present together with an identical 2’ modification in the orphan nucleotide as discussed above.

[0085] Base modifications

[0086] A base, sometimes called a nucleobase, is generally adenine, cytosine, guanine, thymine or uracil, or a derivative thereof. A nucleobase is defined as a moiety that can bond to another nucleobase through H-bonds, polarized bonds (such as through OF moieties) or aromatic electronic interactions. 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 the nucleobases as such. The terms ‘adenosine’, ‘guanosine’, ‘cytidine’, ‘thymidine’, ‘uridine’ and ‘inosine’ refer to the nucleobases linked to the (deoxy)ribosyl sugar. The nucleobases in an AON as disclosed herein can be adenine, cytosine, guanine, thymine, or uracil or any other moiety able to interact with another nucleobase through H-bonds, polarized bonds (such as CF) or aromatic electronic interactions. The nucleobases at any position in the AON as disclosed herein can be a modified form of adenine, cytosine, guanine, or uracil, such as hypoxanthine (the nucleobase in inosine), pseudouracil, pseudocytosine, isouracil, N3- glycosylated uracil, 1 -methylpseudouracil, orotic acid, agmatidine, lysidine, 2-thiouracil, 2- thiothymine, 5-substituted pyrimidine (e.g., 5-halouracil, 5-halomethyluracil, 5- trifluoromethyluracil, 5-propynyluracil, 5-propynylcytosine, 5-aminomethyluracil, 5- hydroxymethyluracil, 5-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, Super T, Super G, amino-modified nucleobases or derivatives thereof; and degenerate or universal bases, like 2,6-difluorotoluene, or absent like abasic sites (e.g. 1 -deoxyribose, 1 ,2- dideoxyribose, 1-deoxy-2-O-methylribose, azaribose). Modified bases comprise synthetic and natural bases such as inosine, xanthine, hypoxanthine and other -aza, deaza, -hydroxy, -halo, -thio, thiol, -alkyl, -alkenyl, -alkynyl, thioalkyl derivatives of pyrimidine and purine bases that are or will be known in the art. Purine nucleobases and / or pyrimidine nucleobases may be modified to alter their properties, for example by amination or deamination of the heterocyclic rings. The exact chemistries and formats may vary from oligonucleotide construct to oligonucleotide construct and from application to application, and may be worked out in accordance with the wishes and preferences of those of skill in the art.

[0087] A scaffold modification indicates the presence of a modified version of the ribosyl moiety as naturally occurring in RNA (i.e., the pentose moiety), such as bicyclic sugars, tetrahydropyrans, hexoses, morpholinos, 2’-modified sugars, 4’-modified sugar, 5’-modified sugars and 4’-substituted sugars. Examples of suitable modifications include, but are not limited to 2’-O-modified RNA monomers, such as 2’-O-alkyl or 2’-O-(substituted)alkyl such as 2’-0Me, 2’-O-(2-cyanoethyl), 2’-M0E, 2’-O-(2-thiomethyl)ethyl, 2’-O-butyryl, 2’-O-propargyl, 2’-O-allyl, 2’-O-(2-aminopropyl), 2’-O-(2-(dimethylamino)propyl), 2’-O-(2-amino)ethyl, 2’-O-(2- (dimethylamino)ethyl); 2’-deoxy (DNA); 2’-O-(haloalkyl)methyl such as 2’-O-(2- chloroethoxy)methyl (MCEM), 2’-O-(2,2-dichloroethoxy)methyl (DCEM); 2’-O-alkoxycarbonyl such as 2’-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2’-O-[2- / V-methylcarbamoyl)ethyl] (MCE), 2’-O-[2-( / V, / V-dimethylcarbamoyl)ethyl] (DCME); 2’-halo e.g. 2’-F, FANA; 2'-O-[2- (methylamino)-2-oxoethyl] (NMA); a bicyclic or bridged nucleic acid (BNA) scaffold modification such as a conformationally restricted nucleotide (CRN) monomer, a locked nucleic acid (LNA) monomer, a xy / o-LNA monomer, an a-LNA monomer, an a-l-LNA monomer, a p-d-LNA monomer, a 2’-amino-LNA monomer, a 2’-(alkylamino)-LNA monomer, a 2’-(acylamino)-LNA monomer, a 2’- / V-substituted 2’-amino-LNA monomer, a 2’-thio-LNA monomer, a (2’-O,4’-C) constrained ethyl (cEt) BNA monomer, a (2’-O,4’-C) constrained methoxyethyl (cMOE) BNA monomer, a 2’,4’-BNANC(NH) monomer, a 2’,4’-BNANC(NMe) monomer, a 2’,4’-BNANC(NBn) monomer, an ethylene-bridged nucleic acid (ENA) monomer, a carba-LNA (cLNA) monomer, a 3,4-dihydro-2 / 7-pyran nucleic acid (DpNA) monomer, a 2’- C-bridged bicyclic nucleotide (CBBN) monomer, an oxo-CBBN monomer, a heterocyclic- bridged BNA monomer (such as triazolyl or tetrazolyl-linked), an amido-bridged BNA monomer (such as AmNA), an urea-bridged BNA monomer, a sulfonamide-bridged BNA monomer, a bicyclic carbocyclic nucleotide monomer, a TriNA monomer, an a-l-TriNA monomer, a bicyclo DNA (bcDNA) monomer, an F-bcDNA monomer, a tricyclo DNA (tcDNA) monomer, an F-tcDNA monomer, an alpha anomeric bicyclo DNA (abcDNA) monomer, an oxetane nucleotide monomer, a locked PMO monomer derived from 2’-amino LNA, a guanidine-bridged nucleic acid (GuNA) monomer, a spirocyclopropylene-bridged nucleic acid (scpBNA) monomer, and derivatives thereof; cyclohexenyl nucleic acid (CeNA) monomer, altriol nucleic acid (ANA) monomer, hexitol nucleic acid (HNA) monomer, fluorinated HNA (F- HNA) monomer, pyranosyl-RNA (p-RNA) monomer, 3’-deoxypyranosyl DNA (p-DNA), unlocked nucleic acid UNA); an inverted version of any of the monomers above. All these modifications are known to the person skilled in the art.

[0088] The orphan nucleotide

[0089] Mutagenesis studies of human ADAR2 revealed that a single mutation at residue 488 from glutamate to glutamine (E488Q), gave an increase in the rate constant of deamination by 60-fold when compared to the wild-type enzyme (Kuttan and Bass. Proc Natl Acad Sci USA 2012. 109(48): 3295-3304). During the deamination reaction, ADAR flips the edited base out of its RNA duplex, and into the enzyme active site (Matthews et al. 2016). When ADAR2 edits adenosines in the preferred context (an A:C mismatch) the nucleotide opposite the target adenosine is often referred to as the ‘orphan nucleotide’ (or ‘orphan cytidine’ as the case may be), as indicated above. The crystal structure of ADAR2 E488Q bound to double stranded RNA (dsRNA) revealed that the glutamine (Gin; Q) side chain at position 488 can donate an H-bond to the N3 position of the orphan cytidine, which leads to the increased catalytic rate of ADAR2 E488Q. In the wild-type enzyme, wherein a glutamate (or glutamic acid; Glu; E) is present at position 488 instead of a glutamine (Gin) the amide group of the glutamine is absent and is instead a carboxylic acid. To obtain the same contact of the orphan cytidine with the E488Q mutant would then, for the wild-type situation, require protonation for this contact to occur. T o make use of endogenously expressed ADAR2 to correct disease relevant mutations, it is essential to maximize the editing efficiency of the wild type ADAR2 enzyme present in the cell. WO2020 / 252376 discloses the use of AONs with modified RNA bases, especially at the position of the orphan cytidine to mimic the hydrogen-bonding pattern observed by the E488Q ADAR2 mutant. By replacing the nucleotide opposite the target adenosine in the AON with cytidine analogs that serve as H-bond donors at N3, it was envisioned that it would be possible to stabilize the same contact that is believed to provide the increase in catalytic rate for the mutant enzyme. Two cytidine analogs were of particular interest: pseudoisocytidine (also referred to as ‘piC’; Lu et al. 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’, when it is a deoxynucleotide; 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 as 6-amino-5-nitro-3-yl-2(1 H)- pyridone. The presence of the cytidine analog in the AON may exist in addition to modifications to the ribose 2’ group. The ribose 2’ groups in the orphan nucleotide can be independently selected from 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-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.

[0090] The orphan nucleotide in the AON as disclosed herein is preferably a cytidine or analog thereof (such as a nucleotide carrying a Benner’s base) or a uridine or analog thereof (such as iso-uridine). The orphan nucleotide, whether it is a cytidine or analog thereof, or a uridine or analog thereof, preferably comprises a deoxyribose (2’-H; = DNA) but may also comprise a di F modification at the 2’ position of the sugar. In an embodiment at least one and in another embodiment both the neighbouring (directly adjacent) nucleotides flanking the orphan nucleotide do not comprise a 2’-OMe modification. Complete modification wherein all nucleotides of the oligonucleotide hold a 2’-OMe modification (including the orphan nucleotide), with natural bases, results in a non-functional oligonucleotide as far as RNA editing goes (known in the art), presumably because it hinders the ADAR activity at the targeted position. In general, an adenosine in a target RNA can be protected from editing by providing an opposing nucleotide with a 2'-OMe group (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing a guanine or adenine as opposing base, as these two nucleobases are also able to reduce editing of the opposing adenosine.

[0091] Linkage modifications

[0092] A nucleoside is generally connected to neighboring nucleosides through condensation of its 5’-phosphate moiety to the 3’-hydroxyl moiety of the neighboring nucleotide monomer. Similarly, its 3’-hydroxyl moiety is generally connected to the 5’-phosphate of a neighboring nucleotide monomer. This forms phosphodiester bonds. The phosphodiesters and the scaffold form an alternating copolymer. The bases are grafted on this copolymer, namely to the scaffold moieties. Because of this characteristic, the alternating copolymer formed by linked scaffolds of an oligonucleotide is often called the ‘backbone’ of the oligonucleotide. Because phosphodiester bonds connect neighboring monomers together, they are often referred to as backbone linkages. It is understood that when a phosphate group is modified so that it is instead an analogous moiety such as a phosphorothioate, such a moiety is still referred to as the backbone linkage of the monomer. This is referred to as a ‘backbone linkage modification’. In general terms, the backbone of an oligonucleotide comprises alternating scaffolds and backbone linkages.

[0093] As outlined in detail herein, naked AONs as disclosed herein comprise at least one, preferably multiple linkage modifications. It is generally more preferred that the AON as disclosed herein comprises linkage modifications at most, and potentially all positions if the AON 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 or MeP), chirally pure MP, (R)-MP, (S)-MP, phosphoryl guanidine (such as PNdmi), chirally pure phosphoryl guanidine, (R)- phosphoryl guanidine, (S)-phosphoryl guanidine, phosphorodithioate (PS2), phosphonacetate (PACE), phosphonoacetamide (PACA), thiophosphonoacetate, thiophosphonoacetamide, methyl phosphorohioate, methyl thiophosphonate, PS prodrug, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, boranophosphate, borano PS, metyl boranophosphate, methyl borano PS, methyl boranophosphonate, methyl boranophosphothioate, phosphate, phosphotriester, aminoalkylphosphotriester, and their derivatives. Another modification includes phosphoramidite, phosphoramidate, N3’->P5’ phosphoramidate, phosphorodiamidate, phosphorothiodiamidate, sulfamate, diethylenesulfoxide, amide, sulfonate, siloxane, sulfide, sulfone, formacetyl, alkenyl, methylenehydrazino, sulfonamide, triazole, oxalyl, carbamate, methyleneimino (MMI), and thioacetamide nucleic acid (TANA); and their derivatives. Various salts, mixed salts, deprotonated, protonated, tautomeric, and free acid forms are also included, as well as 3’->3’ and 2’->5’ linkages. An AON as disclosed herein may also comprise one or more linkage modifications according to the structure of formula (I) wherein: X = O or S; and

[0094] R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl and the linkage modification is referred to as mesyl phosphoramidate, MsPA or PNms.

[0095] In a preferred aspect, the AON as disclosed herein comprises an internucleoside linkage of the structure of formula (I), wherein X = O and R = CH3, which linkage is generally referred to herein as a PNms linkage (mesyl phosphoramidate). In other preferred aspects, R equals one of the following structures (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0096]

[0097] Disclosed herein is also an AON that is able to mediate adenosine deamination by recruitment of a deaminating enzyme in a cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the deaminating enzyme can deaminate the target adenosine into an inosine, and wherein the AON comprises a moiety at one and / or both termini with a structure according to formula (II): wherein: X = O or S;

[0098] Y = O' or S'; and

[0099] R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl.

[0100] An AON as disclosed herein may comprise a substitution of one of the non-bridging oxygens in the phosphodiester linkage. This modification slightly destabilizes base pairing but adds significant resistance to nuclease degradation. A preferred nucleotide analogue or equivalent comprises PS, phosphonoacetate, phosphorodithioate, phosphotriester, aminoalkylphosphotriester, H-phosphonate, methyl and other alkyl phosphonate including 3'- alkylene phosphonate, 5'-alkylene phosphonate and chiral phosphonate, phosphinate, phosphoramidate including 3'-amino phosphoramidate and aminoalkylphosphoramidate, thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, 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 nucleotides may be altered by mono- or di-thioation of the phosphodiester bonds to yield PS esters or phosphorodithioate esters, respectively. Other modifications of the internucleosidic linkages are possible, including amidation and peptide linkers. The skilled person can determine for what target RNA nucleic acid molecule the AON comprises a certain linkage modification at each linkage position of the AON as disclosed herein to generate the most effective and most stable oligonucleotide compound.

[0101] Many of the non-naturally occurring modifications of the linkage, such as PS, are chiral. This means that there are Rp and Sp configurations, known to the person skilled in the art. In one embodiment, the chirality of the PS linkages is controlled, which means that each of the linkages is either in the Rp or in the Sp configuration, whichever is preferred. The choice of an Rp or Sp configuration at a specified linkage position may depend on the target sequence and the efficiency of binding and induction of causing RNA editing of the target adenosine. However, if such is not specifically desired, a composition may comprise AONs as active compounds with both Rp and Sp configurations at a certain specified linkage position. Mixtures of such AONs are also feasible, wherein certain positions preferably have either one of the configurations, while for other positions such does not matter. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PS linkages. In one aspect, the AON as disclosed herein comprises one of more (chirally pure or chirally mixed) phosphoramidate (PN) linkages. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PNms linkages. In one aspect, a PN linkage connects the terminal two nucleotides on each end of the AON. AONs as disclosed herein may also comprise linkage modifications at all positions that are not chirally controlled. The AON as disclosed herein may also comprise one or more naturally occurring internucleoside linkages. The choice and number of modified linkages may depend on the specific target, the sequence, the length, and the stability of the AON observed in a particular cell type of interest, which can be assessed by methods known to the person skilled in the art. In one aspect, at least one, at least two, at least three, or at least four internucleoside linkages between the 5’ and / or the 3’ terminal two, three, four, or five nucleosides respectively of the AON as disclosed herein are modified internucleoside linkages. In one aspect, the AON as disclosed herein comprises at least one MP internucleoside linkage according to the structure of formula (III):

[0102] 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 a preferred embodiment, this position, in an AON as disclosed herein, comprises a linkage modification according to the structure of formula (I), instead of an MP linkage. W02020 / 201406 discloses the use of MP linkage modifications at certain positions surrounding the orphan nucleotide in the first nucleic acid strand. Although the presence of MP linkages is compatible with RNA editing by human ADAR enzymes, introducing MP linkages during the manufacturing of oligonucleotides is challenging in view of additional manufacturing (purification) steps in the coupling and decoupling process. In one aspect, the AON does not comprise an MP linkage.

[0103] 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)

[0104] PNdmi linkage 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.

[0105] In one aspect, the AON as disclosed herein comprises at least one phosphonoacetate and / or at least one phosphonoacetamide internucleoside linkage.

[0106] Conjugate chemistries

[0107] In one aspect, the AON as disclosed herein, or the sense strand to which it may be annealed before entering a target cell (in an HEON as disclosed herein), is bound to a hydrophobic moiety, such as palmityl or an analog thereof, cholesterol or analog thereof, or tocopherol or analog thereof. It is preferably bound to the 5’ terminus. In case a hydrophobic moiety is bound to the 5’ terminus as well as to the 3’ terminus, such hydrophobic moieties may the same or different. The hydrophobic moiety bound to the oligonucleotide may be bound directly, or indirectly mediated by another substance. When the hydrophobic moiety is bound directly, it is sufficient if the moiety is bound via a covalent bond, an ionic bond, a hydrogen bond, or the like. When the hydrophobic moiety is bound indirectly, it may be bound via a linking group (a linker). The linker may be a cleavable or an uncleavable linker. A cleavable linker refers to a linker that can be cleaved under physiological conditions, for example, in a cell or an animal body (e.g., a human body). A cleavable linker is selectively cleaved by an endogenous enzyme such as a nuclease, or by physiological circumstances specific to parts of the body or cell, such as pH or reducing environment (such as glutathione concentrations). Examples of a cleavable linker comprise, but is not limited to, an amide, an ester, one or both esters of a phosphodiester, a phosphoester, a carbamate, and a disulfide bond, as well as a natural DNA linker. Cleavable linkers also include self-immolative linkers. An uncleavable linker refers to a linker that is not cleaved under physiological conditions, or very slowly compared to a cleavable linker, for example, in a PS linkage, modified or unmodified deoxyribonucleosides linked by a PS linkage, a spacer connected through a PS bond and a linker consisting of modified or unmodified ribonucleosides. There is no restriction on the chain length, when a linker is a nucleic acid such as DNA, or an oligonucleotide. However, it may be usually from 2 to 20 bases in length, from 3 to 10 bases in length, or from 4 to 6 bases in length. There is no restriction on the length or composition of a spacer that is connects the ligand and the oligonucleotide, and may include for example ethylene glycol, triethylene glycol (TEG), HEG, alkyl chains, propyl, 6-aminohexyl, or dodecyl. 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

[0108] In addition to the specific preferred chemical modifications at certain positions in compounds as disclosed herein, AONs as disclosed herein may comprise one or more (additional) modifications to the nucleobase, scaffold and / or backbone linkage, which may or may not be present in the same monomer, for instance at the 3’ and / or 5’ position. In one aspect, the AON as disclosed herein comprises at least one internucleoside linkage according to the structure of formula (I), and / or the AON further comprises at least one nucleotide with a sugar moiety that comprises a 2’-0Me modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’-MOE modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’-F modification, and / or the AON comprises an orphan nucleotide that carries a 2’-H in the sugar moiety and is therefore referred to as a DNA nucleotide, even though additional modifications may exist in its base and / or linkage to its neighbouring nucleosides. In one aspect, the orphan nucleotide carries a 2’-F in the sugar moiety. In one aspect, the orphan nucleotide carries a diF substitution in the sugar moiety. In one aspect, the orphan nucleotide carries a 2’-F and a 2’- C-methyl in the sugar moiety. In one aspect, the orphan nucleotide comprises a 2’-F in the arabinose configuration (FANA) in the sugar moiety.

[0109] In one aspect, the AON is an antisense oligonucleotide that can form a double stranded nucleic acid complex with a target RNA molecule, wherein the double stranded nucleic acid complex can recruit an adenosine deaminating enzyme for deamination of a target adenosine in the target RNA molecule, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, and wherein the orphan nucleotide has the structure of formula (V): wherein: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from the group consisting of: cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxo- adenine, and 6-amino-5-nitro-3-yl-2(1 H)-pyridone; R1 and R2 are both selected, independently, from H, OH, F or CH3; R3 is the part of the AON that is 5’ of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4 is the part of the AON that is 3’ of the orphan nucleotide, consisting of 4 to 25 nucleotides. The nucleotide 3’ and / or 5’ from the orphan nucleotide may be DNA, more preferably the nucleotide at the 3’ (position -1). Other chemical modifications of the AON as disclosed herein include the substitution of one or more than one of any of the hydrogen atoms with deuterium or tritium, examples of which can be found in e.g., WO2014 / 022566 or WO2015 / 011694. Again, in all cases, the modifications should be compatible with editing such that the AON fulfils its role as an oligonucleotide that can, after binding to its target sequence, recruit an adenosine deaminase enzyme because of the double-stranded nucleic acid entity that arises. In all aspects of the disclosure, the enzyme with adenosine deaminase activity is preferably ADAR1 , ADAR2, or ADAT.

[0110] AONs as disclosed herein preferably do not include a 5’-terminal O6-benzylguanosine or a 5’-terminal amino modification and preferably are not covalently linked to a SNAP-tag domain (an engineered O6-alkylguanosine-DNA-alkyl transferase). An AON as disclosed herein preferably does not comprise a boxB RNA hairpin sequence. In one aspect, an AON as disclosed herein comprises 0, 1 , 2 or 3 wobble base pairs with the target sequence, and / or 0, 1 , 2, 3, 4, 5, 6, 7, or 8 mismatching base pairs with the target RNA sequence. No mismatch exists when the orphan nucleotide is uridine, which may be defined differently when the orphan nucleotide is a uridine analog or derivative. One alternative for uridine is positioning an isouridine opposite the target adenosine, which likely does not pair like G pairs with II. Preferably, the target adenosine in the target sequence forms a mismatch base pair with the nucleoside in the AON that is directly opposite the target adenosine.

[0111] As outlined above, an AON as disclosed herein makes use of specific nucleotide modifications at predefined spots to ensure stability as well as proper ADAR binding and activity. These changes may vary and may include modifications in the backbone of the AON, in the sugar moiety of the nucleotides as well as in the nucleobases or the phosphodiester linkages, as outlined in detail herein. They may also be variably distributed throughout the sequence of the AON. Specific modifications may be needed to support interactions of different amino acid residues within the RNA-binding domains of ADAR enzymes, as well as those in the deaminase domain. For example, PS linkages between nucleotides or 2’-OMe or 2’-MOE modifications may be tolerated in some parts of the AON, while in other parts they should be avoided so as not to disrupt crucial interactions of the enzyme with the phosphate and 2’-OH groups. Specific nucleotide modifications may also be necessary to enhance the editing activity on substrate RNAs where the target sequence is not optimal for ADAR editing. Previous work has established that certain sequence contexts are more amenable to editing. For example, a target sequence 5’-UAG-3’ (with the target A in the middle) contains the most preferred nearest-neighbor nucleotides for ADAR2, whereas a 5’-CAA-3’ target sequence is disfavored (Schneider et al. 2014. Nucleic Acids Res 42(10):e87). The structural analysis of ADAR2 deaminase domain hints at the possibility of enhancing editing by careful selection of the nucleotides that are opposite to the target trinucleotide. For example, the 5’-CAA-3’ target sequence, paired to a 3’-GCU-5’ sequence on the opposing strand (with the A-C mismatch formed in the middle), is disfavored because the guanosine base sterically clashes with an amino acid side chain of ADAR2. The guanosine opposite the C in such circumstances is preferably replaced by an inosine (hence, at the -1 position within the AON), more preferably a deoxyinosine.

[0112] The AON as disclosed herein, in contrast to what has been described for siRNA, or gapmers and their relation towards RNase breakdown and the use of such gapmers in doublestranded complexes (see for instance EP 3954395 A1), does not comprise a stretch of DNA nucleotides which would make a target sequence (or a sense nucleic acid strand) a target for RNase-mediated breakdown. It is not desired that the target transcript molecule is degraded through the binding of the AON to the transcript molecule. In fact, in respect of MC4R, it is desired that as much transcript remains with the deaminated nucleotide to yield the active protein with a GOF. In one embodiment, the AON does not comprise four or more consecutive DNA nucleotides anywhere within its sequence. In an embodiment, the AON is composed of as much (chemically) modified nucleotides as possible to enhance the resistance towards RNase-mediated breakdown, while at the same time being as efficient as possible in producing an RNA editing effect. This means that the orphan nucleotide and several other nucleotides within the AON may be DNA, but also that there is no stretch of four or more consecutive DNA nucleotides within the AON. Hence, the AON as disclosed 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 singlestranded nucleic acid consisting of a central region (DNA gap region with at least four consecutive deoxyribonucleotides) and wing regions positioned directly at the 5’ end (5’ wing region) and the 3’ end (3’ wing region) thereof. In contrast, the AON as disclosed 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.

[0113] The AONs as disclosed herein may also be administered in the context of aids that will increase the entry of the AON into the target cell and / or its endosomal escape as soon as it is in the cell. Moieties that can be applied for such applications are for example a set of chemical compounds (generally purified from nature) referred to as “saponins” or “triterpene glycosides”. A preferred saponin that can be used in the methods as disclosed herein is AG1856, disclosed in WO2021 / 122998 and further described for use with RNA editing producing oligonucleotides in WO2024 / 153801. Preferably, the AON as disclosed herein is conjugated to the AG 1856 saponin for efficient intracellular delivery. Conjugation is preferably executed through linking the AG1856 to the AON with a hydrazone linker that is disrupted after cell entry.

[0114] 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 (as discussed above), which in fact may also be administered separately 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 administered may depend on the disorder to be treated and the tissue that needs to be targeted and can be selected according to common procedures in the art. The pharmaceutical compositions may be administered by a single-dose administration or by multiple dose administration. It may be administered daily or at appropriate time intervals, which may be determined using common general knowledge in the field and may be adjusted based on the disorder, the severity thereof, and the efficacy of the active ingredient.

[0115] 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 of the oligonucleotide is chemically modified to prevent it from nuclease breakdown also as disclosed 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 as disclosed herein is a ‘naked’ oligonucleotide, comprising a variety of chemical modifications in 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 the deamination of the target adenosine. In another aspect, the AON as disclosed herein, that is delivered 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.

[0116] It is known in the art that RNA editing entities (such as human ADAR enzymes) edit dsRNA structures with varying specificity, depending on several factors. One important factor is the degree of complementarity of the two strands making up the dsRNA sequence. Perfect complementarity of the two strands usually causes the catalytic domain of human ADAR to deaminate adenosines in a non-discriminative manner, reacting with any adenosine it encounters. The specificity of hADARI and 2 can be increased by introducing chemical modifications and / or ensuring several mismatches in the dsRNA, which presumably helps to position the dsRNA binding domains in a way that has not been clearly defined yet. Additionally, the deamination reaction itself can be enhanced by providing an oligonucleotide that comprises a mismatch opposite the adenosine to be edited. Following the instructions in the present application, those of skill in the art will be capable of designing the complementary portion of the oligonucleotide according to their needs.

[0117] It will be understood by a person having ordinary skill in the art that the extent to which the 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 the recognition domain of the editing enzyme. In addition, or alternatively, the degree of recruiting and 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.

[0118] Disclosed herein is the site-specific editing of target adenosines in RNA sequences in eukaryotic, preferably metazoan, more preferably mammalian, more preferably human cells, more preferably human neuronal cells, and most preferably human cells of the CNS. 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 as disclosed 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. Organoids can be thought of as three-dimensional in v / tro-derived tissues but are driven using specific conditions to generate individual, isolated tissues.

[0119] Without wishing to be bound by theory, the RNA editing through human ADAR2 for example 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. Generally spoken, RNA editing may be used to create RNA sequences with different properties. Such properties may be coding properties (creating proteins with different sequences or length, leading to altered protein properties or functions), or binding properties (causing inhibition or over-expression of the RNA itself or a target or binding partner; entire expression pathways may be altered by recoding miRNAs or their cognate sequences on target RNAs). Protein function or localization may be changed at will, by functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co- or post-translational modification, catalytic sites of enzymes, binding sites for binding partners, signals for degradation or activation and so on. These and other forms of RNA and protein “engineering”, whether to prevent, delay or treat disease or for any other purpose, in medicine or biotechnology, as diagnostic, prophylactic, therapeutic, research tool or otherwise, are encompassed by the present disclosure.

[0120] The amount of AON to be administered, the dosage and the dosing regimen can vary from cell type to cell type, the disease to be treated, the target population, the mode of administration {e.g., systemic versus local), the severity of disease and the acceptable level of side activity, but these can and should be assessed by trial and error during in vitro research, in pre-clinical and clinical trials. The trials are particularly straightforward when the modified sequence 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 could compete 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.

[0121] One suitable trial technique involves delivering the AON to cell lines, or a test organism and then taking biopsy samples at various time points thereafter. The sequence of the target RNA can be assessed in the biopsy sample and the proportion of cells having the modification can easily be followed. A potential biomarker that can be used for determining the increase in MC4R activity is the amount or the accumulation of cAMP that is being produced in comparison to a wildtype situation. 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 the relevant part of the target RNA, or a cDNA copy thereof (or a cDNA copy of a splicing product thereof, in case the target RNA is a pre-mRNA), as discussed above, and the sequence change can thus be easily verified. Alternatively, 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 samples obtained from the treated subject. In an in vivo rodent model, obesity can be induced by several means (or alternatively genetically modified obese mice can be used) and loss of weight can easily be determined after treatment with the AON of the present disclosure (in this case an AON that is complementary to the mouse target sequence, serving as a model) in comparison to no treatment or treatment with a placebo or scrambled AON. The diagnosis of obesity and the use of obesity biomarkers in clinical medicine has been reviewed in detail (Nimptsch K et al. 2018. Metabolism. 92:61-70). Such biomarkers can be used in steps of the method as disclosed herein to assess the efficacy of the AONs as disclosed herein in (pre-) clinical settings. After RNA editing has occurred in a cell, the modified RNA can become diluted over time, for example due to cell division, limited half-life of the edited RNAs, etc. Thus, in practical therapeutic terms a method 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.

[0122] AONs as disclosed herein are particularly suitable for therapeutic use, and so disclosed is also a pharmaceutical composition comprising an AON as disclosed herein and a 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 comprising additives, excipients and other ingredients, compatible with pharmaceutical use, at concentrations ranging from 1 ng / ml to 1 g / ml, preferably from 10 ng / ml to 500 mg / ml, more preferably from 100 ng / ml to 100 mg / ml. Dosage may suitably range from between about 1 pg / kg to about 100 mg / kg, preferably from about 10 pg / kg to about 10 mg / kg, more preferably from about 100 pg / kg to about 1 mg / kg. Administration may be by injection or infusion, intracranially, intrathecally, intranasally, orally, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra-peritoneally, intrarectally, intra-cisterna magna, parenterally, and the like. Administration may be in solid form, in the form of a powder, a pill, a gel, a solution, a slow-release formulation, or in any other form compatible with pharmaceutical use in humans.

[0123] 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-functional protein; assessing whether splicing of the pre-mRNA was altered by the deamination; or using a functional read-out, because the target RNA after the deamination should encode an MC4R protein with a GOF, which means as an ultimate result weight loss, or a lower BMI of the human subject that is treated. The identification of the deamination into inosine may be a functional read-out using a suitable biomarker, such as levels of adipokines (e.g., leptin, adiponectin, and / or resistin). 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 the person skilled in the art. However, the person skilled in the art of metabolic disease will preferably apply tests to monitor certain biomarkers related to metabolic function(s) such as food intake and processing thereof.

[0124] In one embodiment, a method as disclosed herein comprises the steps of administering to the subject an AON or 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 the enzyme to deaminate the target adenosine in the target nucleic target molecule to an inosine, thereby alleviating, treating, ameliorating, or slowing down progression of the metabolic disease.

[0125] 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 interest are the human ADARs, hADARI and hADAR2, including any isoforms thereof. RNA editing enzymes 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 hADARI and hADAR2 in humans or human cells and cytidine deaminases. It is known that hADARI exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 110 kDa version, that is produced through alternative splicing from a common pre-mRNA. Consequently, the level of the 150 kDa isoform available in the cell may be influenced by interferon, particularly interferon-gamma (IFN-y). hADARI is also inducible by TNF-a. This provides an opportunity to develop combination therapy, whereby IFN-y or TNF-a and AONs as disclosed herein are administered to a patient either as a combination product, or as separate products, either simultaneously or subsequently, in any order. Certain disease conditions may already coincide with increased IFN-y or TNF-a levels in certain tissues of a patient, creating further opportunities to make editing more specific for diseased tissues. It will be understood by a person having ordinary skill in the art that the extent to which the editing entities inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0126] 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 RNA sequence to which it is bound. Ideally, only one adenosine is deaminated. An AON as disclosed herein, when complexed to ADAR, preferably brings about the deamination of a single target adenosine.

[0127] 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 more than 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 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 one embodiment, the AON is 27, 28, 29, or 30 nucleotides in length.

[0128] EXAMPLES

[0129] Example 1. RNA editing of the MC4R transcript using a variety of AONs.

[0130] An initial set of 30 AONs was designed to target the adenosine in the AUG codon encoding isoleucine at position 317 in the human MC4R protein, which then will be a valine at that position (1317V). The design and chemical modifications of these AONs are provided in Fig. 1 B. A subsequent set of 50 AONs was designed to target the same adenosine. The design and chemical modifications of these AONs are provided in Fig. 2.

[0131] For the screen of the 50 AONs shown in Fig. 2, HEK293 cells stably expressing human wt MC4R mRNA (GenBank: AY236539.1) were used (Innoprot). Cells were cultured in Advanced MEM (Thermo Fisher), supplemented with 10% fetal bovine calf serum (Biowest), 100 ll / rnl pen / strep (Thermo Fisher) and 2 mM L-Glutamine (Thermo Fisher). To preserve the stable overexpression of MC4R, the cells were continuously cultured in the presence of 10 pg / ml Puromycin (Thermo Fisher).

[0132] For transfection experiments the cells were seeded in a collagen pre-coated 24 well plate, with a density of -75.000 cells / well. Cells were cultured O / N at 37 °C and 5% CO2. AONs were transfected at a concentration of 100 nM using RNAiMAX Lipofectamine (Thermo Fisher) according to the manufacturer's specifications, in a 1 :3 (W:W) AON Transfection charge ratio. In a further setup, the cells were gymnotically (no transfection aids) exposed to 5 pM of AON. 24 h after initiation of AON exposure the cells received a 4 h pulse treatment of the triterpene glycoside AG 1856 (saponin; 0.5 pM). Without wishing to be bound by theory, it is thought that the saponin enables the release of the AON from the endosomal entrapment. After these 4 hr the media was refreshed again. 72 hrs after start of the incubation, the medium was removed, and total RNA was isolated using the ReliaPrep™ RNA Miniprep (Promega). cDNA was synthesized from 500 ng total RNA using Maxima Reverse Transcriptase kit using a mixture of random hexamers primers (Thermo Fisher). The percentage of ADAR-mediated A-to-l (G) conversion was determined by quantitative digital PCR (dPCR, Qiaquity; Qiagen) assays designed to the different target sites, using the primers and probes provided in Table 1. The amplification settings are provided in Table 2. The percentage editing is calculated by dividing the counted guanidine partitions by the sum of the guanidine and adenine partitions, multiplied by 100. A control dPCR was performed upstream of the transcript for standardization purposes. Further negative controls were the use of two unrelated AONs (RM4266 and RM4777), a mock transfection control (no AON), a non-treated (NT) sample with only PBS and with cells only and two gBIock controls, that only contained the wt MC4R DNA to check for the PCR assay. In the case of co-treatment with the saponin, also an AG1856 alone control was taken along. Table 1. Primer and probe sequences for quantitative PCR assays. The + indicates a locked nucleic acid (LNA) ribose at the 3’ side. The SEQ ID NO is given between brackets.

[0133] Table 2. PCR amplification setting

[0134] The editing percentage results of the transfection experiment with AONs named RM 117662 to RM 117711 (SEQ ID NO:44 to 93, respectively) are given in Fig. 3A and Fig. 3B, in which the y-axis and the controls in both figures are the same. Clearly, where editing was almost absent or slightly above background in most cases, there was a significant percentage of editing observed after transfecting RM117691 (SEQ ID NO:73), RM117692 (SEQ ID NO:74), RM117693 (SEQ ID NO:75), RM117694 (SEQ ID NO:76), RM117695 (SEQ ID NO:77), RM117696 (SEQ ID NO:78), RM117697 (SEQ ID NO:79), RM117698 (SEQ ID NQ:80), and RM 117700 (SEQ ID NO:82), with RM 117697 (SEQ ID NO:79) outperforming the others with approximately 20% editing.

[0135] The results of the experiment with the saponin pulse treatment are given in Fig. 4A and Fig. 4B, in which again the y-axis and the controls in both these figures are the same. In line with what was found after transfections (see above), RM117691 (SEQ ID NO:73),

[0136] RM117692 (SEQ ID NO:74), RM117693 (SEQ ID NO:75), RM117694 (SEQ ID NO:76),

[0137] RM117695 (SEQ ID NO:77), RM117696 (SEQ ID NO:78), RM117697 (SEQ ID NO:79),

[0138] RM 117698 (SEQ ID NQ:80), and RM 117700 (SEQ ID NO:82), now together with RM 117699 (SEQ ID NO:81) performed good with editing percentages reaching approximately 4%, and with RM 117697 (SEQ ID NO:79) again outperforming the others. When RM 117699 was used in transfection, some increased cell death was observed, possibly explaining the fact that it did not provide the editing percentages that was seen with the AONs in the same range, such as RM 117698.

[0139] Example 2. RNA editing of the MC4R transcript using a further set of AONs.

[0140] In a follow up experiment, the sequence of RM117697 was used to design new AONs as this AON was the best performer of the previous experiments. A transfection experiment with 100 nM EON per transfection in HEK293 cells stably expressing human wt MC4R mRNA was performed as outlined above.

[0141] The editing percentage results of the transfection experiment with AONs named RM120128 to RM120188 (SEQ ID NO:97 to 158, respectively, see Fig. 2) are given in Fig. 5A and Fig. 5B, in which the y-axis and the controls in both figures are the same. Compared to RM117697, RM120130 (SEQ ID NO:99), RM120132 (SEQ ID NQ:101), RM120134 (SEQ ID NQ:103), RM120136 (SEQ ID NQ:105), RM120146 (SEQ ID NO:115), RM120148 (SEQ ID NO:117), RM120152 (SEQ ID NO:121), RM120160 (SEQ ID NO:129), RM120162 (SEQ ID NO:131) and RM120164 (SEQ ID NO:133) significantly outperformed RM117697 with an editing percentage of 3% compared to RM120160 (SEQ ID NO:129) with editing percentage up to 11 %.

[0142] The transfection, incubation and subsequent dPCR was repeated for a subset of good performing AONs, in comparison to RM 117697. The editing percentages obtained after the transfection of these AONs and the dPCR are provided in Fig. 6, with RM 117697 providing the standard (dotted line), and showing that RM120160 (SEQ ID NO:129) performed well again, but also that RM120152 (SEQ ID NO:121) and RM120159 (SEQ ID NO:128) reached editing levels of around 7%. Also, RM120161 (SEQ ID NQ:130), RM120162 (SEQ ID NO:131), RM120163 (SEQ ID NO:132), RM120164 (SEQ ID NO:133), RM120147 (SEQ ID NO:116), RM120148 (SEQ ID NO:117), RM120149 (SEQ ID NO:118), RM120150 (SEQ ID NO:119), RM120151 (SEQ ID NQ:120), RM120153 (SEQ ID NO:122), and RM120158 (SEQ ID NO: 127) performed relatively good.

[0143] These results show that it appeared possible to edit the adenosine in the AUG codon of the (wild-type) human MC4R transcript to an inosine in cells, using endogenously present ADAR enzymes, which allows then the change from isoleucine at position 317 to a valine residue in the human MC4R protein, thereby providing a useful tool to treat a disease resulting from obesity, such as type II diabetes or coronary artery disease, as explained in detail in the present disclosure.

Claims

CLAIMS1. An antisense oligonucleotide (AON) that is capable of recruiting an endogenous ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human melanocortin 4 receptor (MC4R) gene encoding the MC4R 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 neuronal cell, preferably a cell of the central nervous system (CNS).

4. An AON according to any one of claims 1 to 3, wherein the MC4R gene is wildtype, and wherein the target adenosine is the adenosine in the AUG codon coding for an isoleucine (I) residue at position 317 of the MC4R protein, and wherein the deamination of the adenosine results in an inosine that translates the codon to a valine (V) residue at position 317; wherein the deamination of the target adenosine results in an MC4R protein that has a gain-of-function in comparison to the wildtype MC4R protein.

5. An AON according to any one of claims 1 to 4, wherein the orphan nucleotide is a deoxycytidine, a deoxyuridine, a cytidine analog, or a uridine analog.

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

7. An AON according to any one of claims 1 to 6, 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, and wherein the first nucleotide 3’ from the orphan nucleotide (-1) is a deoxyinosine.

8. An AON according to any one of claims 1 to 7, 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.

9. An AON according to any one of claims 1 to 8, 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, PNdmi internucleotide linkage, or a linkage moiety with the structure according to formula (I)wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group.

10. An AON according to claim 9, 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 linkage.11 . An AON according to any one of claims 1 to 10, 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.

12. An AON according to any one of claims 1 to 11 , wherein the oligonucleotide comprises or consists of the sequence of any one of SEQ ID NO:8, 129, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 99, 101 , 103, 105, 115, 116, 117, 118, 119, 120, 121 , 122, 127, 128, 130, 131 , 132, and13. 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.

14. A pharmaceutical composition comprising an AON according to any one of claims 1 to 12, or a vector according to claim 13, and a pharmaceutically acceptable carrier.

15. An AON according to claim 1 to 12, for use in the treatment of a metabolic disease, such as obesity or sexual dysfunction, or a disease resulting from obesity, such as type II diabetes or coronary artery disease.

16. Use of an AON according to any one of claims 1 to 12, in the manufacture of a medicament for the treatment of a metabolic disease, such as obesity or sexual dysfunction, or a disease resulting from obesity, such as type II diabetes or coronary artery disease.

17. A method of editing a human MC4R polynucleotide in a cell, preferably a neuronal cell, wherein the human MC4R polynucleotide is a pre-mRNA or mRNA molecule, the method comprising contacting the MC4R polynucleotide with an AON capable of triggering an ADAR- mediated adenosine to inosine deamination, thereby editing the MC4R polynucleotide to encode an MC4R protein with a gain-of-function.

18. A method of treating, ameliorating, or slowing down the progression of a metabolic disease, such as obesity or sexual dysfunction, or a disease resulting from obesity, such as type II diabetes or coronary artery disease, in a human subject in need thereof, the method comprising administering to said subject an AON according to any one of claims 1 to 12, or a vector according to claim 13, or a pharmaceutical composition according to claim 14, thereby contacting a MC4R polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination, thereby editing the MC4R polynucleotide to encode an MC4R protein with a gain-of-function, thereby treating the subject.

19. An in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in an MC4R 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 12;(ii) allowing uptake by the cell of the AON;(iii) allowing annealing of the AON to the MC4R 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.

20. An in vivo method according to claim 19, comprising step (v) of using a functional readout to identify the presence of the inosine in the target RNA molecule.

21. A method according to any one of claims 17 to 20, wherein the cell is a neuronal cell, preferably a cell of the CNS.

22. A method according to any one of claims 17 to 21 , comprising the step of administering a triterpene glycoside before, after or simultaneously with administering the AON, preferably wherein the triterpene glycoside is conjugated to the AON.

23. A method according to claim 22, wherein the triterpene glycoside is AG1856.

Images