Antisense oligonucleotides for the treatment of neurological disorders

Abstract

The disclosure relates to the field of diseases caused by a lowered synaptic inhibition, preferably those that are caused by a diminished activity of the potassium (K) / chloride (Cl) Cotransporter 2 (KCC2). The disclosure involves oligonucleotides and the use thereof in RNA editing methods in targeting a variety of target adenosines in the human SLC12A5 transcript molecule that encodes KCC2. The transcript molecule is edited such that the resulting KCC2 protein has a gain-of-function and / or different function, for example reduced autoinhibition. The disclosure relates to oligonucleotides and their use in the treatment of neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and / or epilepsy.

Description

ANTISENSE OLIGONUCLEOTIDES FOR THE TREATMENT OF NEUROLOGICAL DISORDERS CROSS-REFERENCE TO RELATED APPLICATIONS AND INCORPORATION BY REFERENCE

[0001] This PCT application claims the priority benefit of U.S. Provisional Application No. 63 / 695,237, filed on September 16, 2024, U.S. Provisional Application No. 63 / 783,142, filed on April 3, 2025, U.S. Provisional Application No.63 / 841,147, filed July 9, 2025, and U.S. Provisional Application No.63 / 859,721, filed August 7, 2025, which are herein incorporated by reference in their entirety. REFERENCE TO SEQUENCE LISTING SUBMITTED ELECTRONICALLY

[0002] The content of the electronically submitted ST.26 sequence listing in xml format (name 4430_024PC05_SequenceListing_ST26.xml; size is 2,919,725 bytes; and date of creation: September 8, 2025) filed with the application is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] This disclosure relates to the field of medicine, and in particular to the field of neurological disorders. The disclosure describes antisense oligonucleotides that mediate nucleotide-specific RNA editing in the human SLC12A5 gene transcript to bring about amino acid changes of the encoded potassium (K+) / chloride (Cl-) cotransporter 2 (KCC2) protein that influence its activity. BACKGROUND

[0004] Cation-chloride cotransporters (CCC) are central to the regulation of intracellular Cl- concentration and have key physiological roles in neuronal excitability, trans-epithelial salt and water movement, and regulation of cell volume. In humans, there are two broadly distributed subfamilies: the Na+(K+)-coupled transporters NKCC1, 2 and NCC (encoded by the SLC12A1, SLC12A2, and SLC12A3 genes, respectively) and the K+-coupled KCC1, KCC2, KCC3, and KCC4 proteins (encoded by the SLC12A4, SLC12A5, SLC12A6, andSLC12A7 genes, respectively). The NKCCs and NCC utilize the energy of the Na+ gradient generated by Na+ / K+ ATPase to accumulate cellular Cl- above its electrochemical equilibrium while KCCs utilize the outwardly directed K+ gradient to extrude Cl-. These transporters generate electrochemical Cl- gradients across the neuronal plasma membrane while channels dissipate them, and these interactions set the polarity and driving force of GABAA receptor-mediated chloride currents. The CCCs are strictly electroneutral transporters, NKCCs transporting two Cl- ions coupled to the movement of one Na+ and one K+, whereas NCC and KCC couple the movement of one Cl- ion and either Na+ or K+ respectively. Fast synaptic transmission relies on ion fluxes through these ligand-gated channels. Therefore, maintaining transmembrane ionic gradients is critical to preserve synaptic efficacy.

[0005] In the adult vertebrate Central Nervous System (CNS), KCC2 is highly enriched in neurons and continuously extrudes Cl ions (Cl-), thus ensuring that intracellular levels of Cl- remain low, as required for inhibitory γ-aminobutyric acid (GABA)ergic and glycinergic neurotransmission. The Cl- gradient determines whether GABA receptor and glycine receptor (GlyR) generate inhibitory or excitatory signals within the brain. In the developed CNS, maintaining the inhibitory tone of GABA receptors and GlyR is critical for normal neuronal function. Disruption of inhibitory signalling is associated with a wide range of neurological and psychiatric disorders. A key mechanism that has emerged in the understanding of the underlying mechanism of disinhibition involves the disruption of Cl- homeostasis resulting from the loss of activity of KCC2. This mechanism appears across several pathological pain syndromes with diverse etiologies, including spinal cord injury (Coull JA et al. 2003. Nature 424(6951):938-942), inflammation (Lu Y et al. 2008. J Physiol (Lond.) 586(Pt 23):5701-5715), painful diabetic neuropathy (Jolivalt CG et al. 2008. Pain 140(1):48-57), trigeminal pain (Wei B et al.2013. Neuroscience 228:334-348), morphine-induced hyperalgesia (Ferrini F et al. 2013. Nat Neurosci. 16(2):183-192), and epilepsy (Cohen I et al.2002. Science 298:1418-1421; Huberfeld G et al.2007. J Neurosci. 27(37):9866-9873). Since KCC2 is responsible for Cl- extrusion, disruption of its function causes a collapse of the transmembrane Cl- gradient and a depolarizing shift in GABAA reversal potential (EGABA). This in turn leads to a decrease in inhibitory efficacy. In chronic pain, GABAergic transmission is compromised, causing circuit malfunction, and disrupting inhibitory neural networks. Additionally, it is found that in chronic pathologic pain KCC2expression is attenuated in the primary sensory gate in spinal cord dorsal horn (SCDH) neurons. It is therefore well established that this key pathophysiological mechanism contributes to an imbalance of excitation / inhibition because it corrupts inhibitory neurotransmission, leading to inhibitory circuit malfunction. Notably, there is no ‘back-up’ protein that can rescue the KCC2 expression deficit.

[0006] There are two isoforms of KCC2: KCC2a and KCC2b that arise from alternative transcriptional start sites within the human SLC12A5 gene. These transcripts translate to two protein isoforms that differ in their N-termini, with the KCC2a form constituting the larger of the two splice variants. KCC2a levels remain relatively constant during pre- and postnatal development, whereas KCC2b, on the other hand, is scarcely present during prenatal development and is strongly upregulated during postnatal development. The upregulation of KCC2b expression is thought to be responsible for the ‘developmental shift’ observed in mammals from depolarizing postsynaptic effects of inhibitory synapses in early neural networks to hyperpolarizing effects in mature neural networks. It has been established that besides its function in regulating intraneuronal Cl- homeostasis, the activity of KCC2 is also associated with transmembrane water fluxes that compensate solute fluxes associated with synaptic activity. Moreover, KCC2 interaction with the actin cytoskeleton appears critical both for dendritic spine morphogenesis and the maintenance of glutamatergic synapses (Chamma I et al.2012. Front Cell Neurosci.6:5). KCC2b knockout mice can survive up to postnatal day 17 due to the presence of functional KCC2a alone, but they exhibit low body weight, motor deficits and generalized seizures. Complete KCC2 knockouts, in which both KCC2a and KCC2b are absent, die after birth due to respiratory failure.

[0007] Enhancing KCC2 activity can potentially be used as treatment of a wide variety neurological disorders where a lower inhibitory tone exacerbates or is the underlying cause of the disease. For example, increasing KCC2 function has been proposed for treatment of pathogenic pain (Doyon N et al. 2013. Expert Rev Neurother. 13(5):469-471). Here, enhancing inhibitory signalling through increasing KCC2 function alleviates deficits in GABAA and glycine inhibitory signalling observed in neuropathic pain (Lorenzo L-E et al. 2020. Nature Communications 11:869). In epilepsy where seizures are induced by increased excitatory neuronal activity or by a reduction in inhibitory tone, increasing KCC2 has been proposed as therapeutic strategy (Moore YE et al. 2017. Trends Neurosci.40(9):555-571). For example, genetic removal of KCC2 activity impairing phosphorylation sites, was shown to be sufficient to limit the onset and severity of seizures (Moore YE et al. 2018. Proc Natl Acad Sci USA. 115(40):10166-10171). Moreover, in the neurodevelopmental disorders such as autism spectrum disorder (ASD) and Rett syndrome (a severe form of ASD), impaired GABAergic inhibitory function has been observed (Tyzio R et al. 2014. Science 343(6171):675-679; Tang X et al. 2016. Proc Natl Acad Sci USA. 113(3):751-756). In post-mortem brain samples from Rett syndrome patients, a significantly lower expression of KCC2 was observed (Hinz L et al. 2019. Acta Neuropathol Commun. 7(1):196) and restoring KCC2 function in neurons derived from patients with Rett syndrome was shown to restore functional deficits observed in these neurons (Tang et al.2016).

[0008] Despite the use of the small molecule and KCC2 ‘activator’ CLP257, which appeared not to act on KCC2 itself, and despite the numerous attempts to increase KCC2 activity in individuals suffering from neurological diseases such as neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and / or epilepsy, there remains a need for alternatives to treat these severe quality-of-life related disorders. The present disclosure aims to provide such alternative, and / or improved, compounds and compositions for use in the treatment of neuronal disorders in which an increase in KCC2 activity is beneficial. BRIEF SUMMARY

[0009] Disclosed herein are oligonucleotides that mediate nucleotide-specific RNA editing of the human SLC12A5 transcript, RNA editing oligonucleotides (EONs), which form a double-stranded complex with a region of an endogenous human SLC12A5 transcript molecule in a cell, wherein the transcript molecule is a pre-mRNA or an mRNA molecule, wherein the region of the SLC12A5 transcript molecule comprises a target adenosine, and wherein the double-stranded complex can recruit an endogenous ADAR enzyme to deaminate the target adenosine (A) into an inosine (I), thereby editing the SLC12A5 transcript molecule, wherein the target A is any one of: i) the A in the UAC codon encoding tyrosine (Y) at position 68 of the SLC12A5- encoded KCC2b isoform;ii) the A in the ACC codon encoding threonine (T) at position 69 of the SLC12A5- encoded KCC2b isoform; iii) the first A in the AAC codon encoding asparagine (N) at position 70 of the SLC12A5-encoded KCC2b isoform; iv) the second A in the AAC codon encoding asparagine (N) at position 70 of the SLC12A5-encoded KCC2b isoform; v) the A in the CAG codon encoding glutamine (Q) at position 73 of the SLC12A5- encoded KCC2b isoform; vi) the A in the CAU codon encoding histidine (H) at position 78 of the SLC12A5- encoded KCC2b isoform; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79 of the SLC12A5-encoded KCC2b isoform; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82 of the SLC12A5-encoded KCC2b isoform; ix) the A in the UAC codon encoding tyrosine (Y) at position 446 of the SLC12A5- encoded KCC2b isoform; x) the A in the AGU codon encoding serine (S) at position 932 of the SLC12A5- encoded KCC2b isoform; or xi) the A in the ACG codon encoding threonine (T) at position 906 of the SLC12A5- encoded KCC2b isoform.

[0010] In one aspect, the SLC12A5-encoded KCC2b isoform is according to the consensus sequence referenced by NCBI Ref. Seq. No. NP_065759 (NM_020708). In one aspect, the target SLC12A5 transcript molecule is mutated, and the mutation may be the cause of one or more neurological disorders. Mutations that have been identified that cause a decrease in KCC2 activity are for example A191V, M415V, L288H, S376L, L403P, G528D, and R952H (KCC2b isoform). It is the purpose of the present disclosure to provide methods and means to increase the KCC2 activity.

[0011] The eleven codon positions listed above, when editing of the target A has taken place, refer to the following amino acid changes in the KCC2b isoform: i) Y68C, ii) T69A, iii) N70D, iv) N70S, v) Q73R, vi) H78R, vii) E79G, viii) E82G, ix) Y446C, x) S932G, and xi) T906A, respectively. It is to be understood that the disclosed EONs can bring about editing of the equivalent target adenosines in the KCC2a isoform, which is 23 amino acidslonger than KCC2b. Hence, for all target sites, the target A is at the equivalent position within the SLC12A5-encoded KCC2a isoform as referenced by NCBI Reference Sequence: NM_001134771.2 (mRNA) and NCBI Ref. Seq. No. NP_001128243 (protein), and is offset by exactly 23 positions in the amino acid sequence and such as the amino acid changes for KCC2a are referred to as: i) Y91C, ii) T92A, iii) N93D, iv) N93S, v) Q96R, vi) H101R, vii) E102G, viii) E105G, ix) Y469C, x) S955G, and xi) T929A, respectively.

[0012] Disclosed herein is also an EON according to the disclosure for use in the treatment of a disorder caused by a diminished GABAergic inhibition, that may be caused by a lowered, hampered, diminished, or even absent KCC2 activity, including without limitation neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and epilepsy.

[0013] Disclosed herein is also a method of editing a SLC12A5 polynucleotide, the method comprising contacting the SLC12A5 polynucleotide with an EON capable of effecting an ADAR-mediated A to I editing of a target A in a codon encoding an amino acid that is associated with regulating intraneuronal Cl- homeostasis by the SLC12A5-encoded protein KCC2, thereby editing the SLC12A5 polynucleotide. Preferably, the editing results in a KCC2 protein with a gain-of-function, or in a restoration of function when the KCC2 activity was lowered before editing took place. Disclosed herein is also a method of treating a disorder caused by a diminished GABAergic inhibition, such as those caused by a diminished KCC2 activity, in an individual in need thereof, the method comprising contacting a SLC12A5 polynucleotide in a cell of the subject with an EON capable of effecting an ADAR-mediated A to I editing of a target A in a codon encoding an amino acid that is associated with regulating intraneuronal Cl- homeostasis by the SLC12A5- encoded protein KCC2, thereby treating the individual. Disclosed is also a method of deaminating a target A in an SLC12A5 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: (i) providing the cell with an EON as disclosed herein, (ii) allowing uptake by the cell of the EON, (iii) allowing annealing of the EON to the SLC12A5 pre- mRNA or mRNA molecule, (iv) allowing an endogenous ADAR enzyme to deaminate the target A in the target RNA molecule to an I; and optionally (v) identifying the presence of the I in the target RNA molecule.BRIEF DESCRIPTION OF THE DRAWINGS

[0014] One or more embodiments will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0015] FIG. 1A shows part (positions 172 to 276) of the 5’ to 3’ target sequence of the transcript of the human KCC2b isoform (SEQ ID NO:1) with positions of the codons encoding tyrosine 68 (UAC), threonine 69 (ACC), asparagine 70 (AAC), glutamine 73 (CAG), histidine 78 (CAU), glutamic acid 79 (GAA), and glutamic acid 82 (GAA) in bold. The target adenosines within these codons, as outlined herein, are underlined. The boxed area represents the transcript sequence encoding the 27 amino acid containing N-terminal peptide (see below) involved in autoinhibition of the KCC2 protein (provided as SEQ ID NO:27). On the second line, directly below the transcript sequence, the encoded amino acid sequence is provided (SEQ ID NO:2). The third line shows the position number of the amino acid which is edited in a target codon. The lines below show respectively, for each identified target codon and its respective encoded amino acid, the edited three-letter codon and their encoded amino acid. Position N70 is edited to N70D and / or N70S. C is cysteine, A is alanine, D is aspartic acid, S is serine, R is arginine, and G is glycine, as known to the person skilled in the art.

[0016] FIG.1B is according to FIG.1A and shows part of the 5’ to 3’ target sequence of the transcript of the human KCC2a isoform (also SEQ ID NO:1) with the codons encoding tyrosine 91 (UAC), threonine 92 (ACC), asparagine 93 (AAC), glutamine 96 (CAG), histidine 101 (CAU), glutamic acid 102 (GAA), and glutamic acid 105 (GAA) in bold.

[0017] FIG.1C shows: i) on top, part of the 5’ to 3’ target sequence of the mRNA transcript of the human KCC2b isoform (SEQ ID NO:3) surrounding the UAC codon for tyrosine at position 446, followed by the encoded amino acid sequence (SEQ ID NO:4); ii) in the middle, part of the 5’ to 3’ target sequence of the mRNA transcript of the human KCC2b isoform (SEQ ID NO:5) surrounding the AGU codon for serine at position 932, followed by the encoded amino acid sequence (SEQ ID NO:6); and iii) at the bottom, part of the 5’ to 3’ target sequence of the mRNA transcript of the human KCC2b isoform (SEQ ID NO:7) surrounding the ACG codon for threonine at position 906, followed by the encoded amino acid sequence (SEQ ID NO:8). The target codons are in bold, and the target adenosine residues within these codons are underlined. The three-letter codons that occur after editingas outlined further herein are provided below the amino acid sequences, followed by the encoded amino acids in which C, G, and A are as mentioned in FIG.1A.

[0018] FIG. 1D shows SEQ ID NO:474, which is the human KCC2a mRNA coding sequence according to NCBI Reference Sequence: NM_001134771.2.

[0019] FIG.1E shows the encoded protein sequence (SEQ ID NO:475) corresponding to human KCC2a. The 27 amino acid containing N-terminal peptide of SEQ ID NO:28 is underlined.

[0020] FIG. 1F shows SEQ ID NO:476, which is the human KCC2b mRNA coding sequence according to NCBI Reference Sequence: NM_020708.5

[0021] FIG.1G shows the encoded protein sequence (SEQ ID NO:477) corresponding to human KCC2b. The 27 amino acid containing N-terminal peptide of SEQ ID NO:28 is underlined. The 11 target positions as outlined herein are provided in bold face.

[0022] FIG. 2 shows a schematic model (from Zhang S et al. 2021. Communications Biology 4:226) indicating the position of the N-terminal peptide (left) locking the KCC2 dimer in an autoinhibition state when the KCC2 protein is phosphorylated at positions T906 and T1007 (KCC2b), and (right) when it displaces from the cytosolic facing cavity when T906 and T1007 are de-phosphorylated, which situation represents the active state of the KCC2 dimer.

[0023] FIG.3A shows the K+ influx rate generated by ten KCC2b variants (as indicated) after transfection of expression plasmids expressing these ten variants in HEK293 cells, after 72 hrs, using FLIPRTMassay. WT is the control version of KCC2b.

[0024] FIG.3B shows the fold increase calculated from the control (dotted line) within the same ten samples.

[0025] FIG. 4 shows the normalized KCC2 activity (K+ influx rate) in HEK293 cells generated by the indicated KCC2b variants, after 48 hrs, upon transfection of expression plasmids expressing the respective variants, using a 30 min treatment with WNK1 inhibitor WNK463 and KCC2 inhibitor VU0240551, in comparison to no treatment (control) and WT.

[0026] FIG. 5 shows the 5’ to 3’ sequences and chemical modifications of an initial set comprising 14 editing oligonucleotides (EONs T906A-1 to 14) that were designed to bring about editing of the target adenosine in the ACG codon encoding threonine at position 906 in the human KCC2b isoform (T906). The SEQ ID NO and RM reference number of eachof the modified EONs is given. The chemical modifications in the EONs are as follows: m5Ce is 2’-MOE modified 5-methylcytidine; m5Ue is 2’-MOE modified 5-methyluridine (Te; 2’-MOE modified thymidine); Ge is 2’-MOE modified guanosine; Gm, Am, Um, and Cm are 2’-OMe modified guanosine, adenosine, uridine, and cytidine, respectively; Af, Uf, Gf, and Cf are 2’-F modified adenosine, uridine, guanosine, and cytosine, respectively; Zd is a cytidine analog that is also referred to as a nucleoside carrying a Benner’s base (as further outlined herein), with a deoxy moiety (= DNA) at the 2’ ribose position; Id is deoxyinosine; Cd is deoxycytidine; 7Ad is 7-deaza-2’-deoxyadenosine; ^ refers to a PO linkage; * refers to a phosphorothioate (PS) linkage; ! refers to a (1,3-dimethylimidazolidin- 2-ylidene) phosphoramidate (PNdmi) linkage; ^ refers to a methylphosphonate (MP) linkage; # refers to a mesyl phosphoramidate (PNms) linkage.

[0027] FIG.6 shows the percentage editing obtained after gymnotic uptake of the 14 EONs of FIG. 5 in forebrain neurons cultivated from human induced pluripotent stem cells (‘hiPSC neurons’). Shown are the five EONs comprising a deoxyinosine (Id) at the -1 position in the EON (opposite the 5’-G in the target sequence), the three EONs comprising a deoxycytidine (Cd) at the -1 position in the EON, and the six EONs comprising a 7-deaza- 2’-deoxyadenosine (7Ad) at the -1 position in the EON. NT was a non-transfected control.

[0028] FIGS. 7A, 7B, 7C, 7D, 7E, 7F, 7G, 7H, 7I, 7J, 7K, and 7L show the 5’ to 3’ sequences and chemical modifications of 190 EONs that were designed to bring about editing of the target adenosines in the target transcript molecule of SEQ ID NO:27 that encodes the N-terminal peptide of SEQ ID NO:28, to bring about the Y68C, T69A, N70D, N70S, Q73R, H78R, E79G, and E82G changes. The alternative (Alt.) names of the EONs refer to these changes, respectively. SEQ ID NO’s and RM reference numbers are given. The chemical modifications are as provided in FIG. 5, wherein Ae is 2’-MOE modified adenosine, Gd is deoxyguanosine, m5Ud is a deoxynucleotide with a 5-methyluridine, and Ad is deoxyadenosine.

[0029] FIG. 8 shows the editing percentage, determined by Sanger sequencing, of the target adenosines leading to the eight amino acid changes in the N-terminal peptide of KCC2b: Y68C, T69A, N70D, N70S, Q73R, H78R, E79G, and E82G, using six EONs for each target. The names of the specific EONs that were initially tested are according to the alternative names listed in FIGS.7A-7L.

[0030] FIG.9 shows the editing percentage by ddPCR of the target adenosine leading to the Q73R and H78R changes in the N-terminal peptide of KCC2b, using Q73R-1 to Q73R- 6 and H78R-1 to H78R-6, respectively, as indicated.

[0031] FIGS.10A, 10B, and 10C show the 5’ to 3’ sequences and chemical modifications of EONs that were designed to bring about editing of the target adenosine in the CAG codon for glutamine at position 96 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (Q96R). The names, RM reference number codes, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG.5 and FIGS.7A-7L.

[0032] FIGS.11A, 11B, and 11C show the 5’ to 3’ sequences and chemical modifications of EONs that were designed to bring about editing of the target adenosine in the CAU codon for histidine at position 101 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (H101R). The names, RM reference number codes, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG.5 and FIGS.7A-7L.

[0033] FIG.12 shows the 5’ to 3’ sequences and chemical modifications of 9 EONs that were designed to target the eight different target adenosines in wildtype mice. mB91-4 targets the Y91 site. mB92-4 targets the T92 site. mB93A-4 targets the N93 site to change the codon to encode aspartic acid (D). mB93B-4 targets the N93 site to change the codon to encode serine (S). mB96-6 targets the Q96 site (KCC2a), which is the Q73 site in KCC2b, and a stabilized version of this EON, which is provided here as mB96-29. mB101- 4 targets the H101 site. mB102-4 targets the E102 site. mB105-4 targets the E105 site.

[0034] FIG.13A, FIG.13B, FIG.13C, FIG.13D, FIG.13E, FIG.13F, FIG.13G, FIG. 13H, and FIG. 13J show the editing percentages in cortex, hippocampus, midbrain, cerebellum, spinal cord (lumbar), spinal cord (cervical), and spinal cord (thoracic) in mice after 14 days upon ICV treatment with 100 μg of the following EONs: mB91-4 (FIG.13A), mB92-4 (FIG.13B), mB93A-4 (FIG.13C), mB93B-4 (FIG.13D), mB96-6 (FIG.13E), mB96-29 (FIG.13F), mB101-4 (FIG. 13G), mB102-4 (FIG.13H), and mB105-4 (FIG. 13J).

[0035] FIG. 14A, FIG. 14B, FIG. 14C, FIG. 14D, and FIG. 14E shows the editing percentages in cortex, hippocampus, midbrain, cerebellum, spinal cord (lumbar), spinal cord (cervical), and spinal cord (thoracic) in mice after 14 days upon ICV treatment with200 μg of the following EONs: mB91-4 (FIG.14A), mB92-4 (FIG.14B), mB93B-4 (FIG. 14C), mB96-6 (FIG.14D), and mB102-4 (FIG.14E).

[0036] FIG.15 shows two cytosine analogs as further outlined herein, with on the left the 6-amino-5-nitro-3-yl-2(1H)-pyridone (also referred to herein as Benner’s base or Z base), and on the right the 5-aza-5,6-dihydro cytosine (also referred to herein as E base).

[0037] FIG.16 shows the 5’ to 3’ sequences of the EONs (herein referred to as ‘AONs’) used in the experiments to determine the editing efficiency using a Z nucleobase or an E nucleobase at the orphan position. The chemical modifications are as provided in FIG.5, wherein Ed (orphan nucleotide) is a deoxyribonucleoside comprising the deoxycytidine analog 5-aza-5,6-dihydro cytosine (E base). EON1 (= B96-111; RM122584; SEQ ID NO:266; AON1 in FIG. 17A) targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a and comprises a deoxynucleotide at the orphan position with a Z nucleobase (Zd). EON2 (= B96-155; RM123434; SEQ ID NO:335; AON2 in FIG.17A) is identical to EON1 and targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a but instead comprises a deoxynucleotide at the orphan position with an E nucleobase (Ed). EON3 (RM123359; SEQ ID NO:336; AON3 in FIG.17A) is identical to EON1 and targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a but is an on-target negative control since it comprises a 2’-OMe modified nucleotide at the orphan position with an uracil nucleobase (Um) and should be inactive for RNA editing. EON4 (RM123361; SEQ ID NO:337; AON4 in FIG. 17A) comprises a Zd nucleotide at the same position as EON1 but is a further scrambled control of EON1, also serving as a negative control. EON5 (= B101-68; RM122501; SEQ ID NO:283; AON5 in FIG. 17B) targets the adenosine in the CAU codon for histidine at position 101 in KCC2a and comprises a deoxynucleotide at the orphan position with a Z nucleobase (Zd). EON6 (= B101-148; RM123435; SEQ ID NO:338; AON6 in FIG. 17B) is identical to EON5 and targets the adenosine in the CAU codon for histidine at position 101 in KCC2a but instead comprises a deoxynucleotide at the orphan position with an E nucleobase (Ed). EON7 (RM123360; SEQ ID NO:339; AON7 in FIG. 17B) is identical to EON5 and targets the adenosine in the CAU codon for histidine at position 101 in KCC2a but is an on-target negative control since it comprises a 2’-OMe modified nucleotide at the orphan position with an uracil nucleobase (Um) and should be inactive for RNA editing. EON8 (RM123362; SEQ ID NO:340; AON8 in FIG.17B) comprises a Zd nucleotide at the sameposition as EON5 but is a further scrambled control of EON5 also serving as a negative control. n.a. = not applicable.

[0038] FIG. 17A and FIG. 17B show the editing percentages of two gymnotic uptake experiments conducted on the adenosines of the Q96 and H101 target sites (KCC2a), respectively. FIG. 17A shows the editing percentages after administration of EONs targeting the adenosine of the CAG codon for glutamine at position 96 in KCC2a. FIG. 17B shows the editing percentages after administration of EONs targeting the adenosine in the CAU codon for histidine at position 101 in KCC2a. To highlight the low batch-to-batch variability, data is shown from two independent iPSC forebrain batches (referred to as cell batch A and B, respectively) for AON1 (FIG. 17A) and AON5 (FIG.17B), provided by the first 2 bars. AON2, AON3, AON4, AON6, AON7, and AON8 were tested on cell batch B only. NT represents a non-treated negative control.

[0039] FIGS.18A, 18B and 18C show the 5’ to 3’ sequences and chemical modifications of 34 EONs that were designed to bring about editing of the target adenosine in the CAG codon for glutamine at position 96 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (Q96R). The names, RM reference number codes, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG. 5 and FIGS. 7A-7L, wherein Im is a 2’-OMe modified nucleotide comprising a hypoxanthine nucleobase, wherein If is a 2’-F modified nucleotide comprising a hypoxanthine nucleobase, and wherein Ie is a 2’-MOE modified nucleotide comprising a hypoxanthine nucleobase.

[0040] FIGS.19A, 19B and 19C show the 5’ to 3’ sequences and chemical modifications of 37 EONs that were designed to bring about editing of the target adenosine in the CAU codon for histidine at position 101 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (H101R). The names, RM reference number codes, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG.5, FIGS.7A-7L, and FIGS.18A-18C.

[0041] FIGS.20A, 20B and 20C show the 5’ to 3’ sequences and chemical modifications of two sets of EONs that were designed: i) to bring about editing of the target adenosine in the CAG codon for glutamine at position 96 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (Q96R) (B96-155 to B96-180; SEQ ID NO:335, 439 to 463, respectively); and ii) to bring about editing of the target adenosine inthe CAU codon for histidine at position 101 in the KCC2a isoform, to change the codon such that it encodes arginine at that position (H101R) (B101-148 to B101-153; SEQ ID NO:338, 464 to 468, respectively). The names, RM reference numbers, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG.5, FIG.7A-7L, FIGS.18A-18C, and FIGS.19A-19C.

[0042] FIGS.21A and 21B show the 5’ to 3’ sequences and chemical modifications of 12 EONs targeting the adenosine in the CAG codon in the mouse Slc12a5 transcript for glutamine at position 96 (Q96) in the mouse Kcc2a isoform, and of 6 EONs targeting the adenosine in the CAU codon in the mouse Slc12a5 transcript for histidine at position 101 (H101) in the mouse Kcc2a isoform. The names, RM reference numbers, SEQ ID NO’s and sequences with chemical modifications are provided. The chemical modifications are as provided in FIG. 5, FIG.7A-7L, FIGS. 18A-18C, FIGS.19A-19C, and FIGS. 20A- 20C.

[0043] FIGS.22A and 22B show the chemical structure (formula (VII)) of the RM122556 EON, herein also referred to as B96-83 (SEQ ID NO:238).

[0044] FIGS.23A and 23B show the chemical structure (formula (VIII)) of the RM122584 EON, herein also referred to as B96-111 (SEQ ID NO:266), as EON1, and as AON1 in FIG.17A.

[0045] FIGS.24A and 24B show the chemical structure (formula (IX)) of the RM123389 EON, herein also referred to as B96-147 (SEQ ID NO:394).

[0046] FIGS. 25A and 25B show the chemical structure (formula X) of the RM129114 EON, herein also referred to as B96-179 (SEQ ID NO:462). DETAILED DESCRIPTION

[0047] The human KCC2 protein in the cells of the CNS is extensively post- transcriptionally modified, and the functional properties as a K+ / Cl- cotransporter of KCC2 is reciprocally regulated by serine / threonine phosphorylation. One site that is post- translationally phosphorylated is the threonine residue at position 1007 in the human KCC2b isoform (see, NCBI Ref. Seq. No. NP_065759.1) that is equivalent to the threonine at position 1030 in the human KCC2a isoform (see, NCBI Ref. Seq. No. NP_001128243.1). Hereinafter, because the targeting discussed is predominantly for transcripts present in mature neuronal cells, it is noted that the target sites are generally referred to as being atpositions within KCC2b, but it is to be understood that the equivalent positions in KCC2a are also encompassed by the present disclosure and can also be changed with the compounds and compositions as disclosed herein. When the disclosure refers to targeting any positions in KCC2b, in fact the equivalent sites in KCC2a are intrinsically also meant and both isoforms are included.

[0048] It has been demonstrated that phosphorylation of this T1007 site leads to decreased activity of the KCC2 channel resulting in decreased inhibitory tone (Pisella LI et al.2019. Sci Signal.12(603):eaay0300). The same holds true for the phosphorylation site at position 906 (T906). Phosphorylation of the T906 and T1007 residues brings the KCC2b dimer in an autoinhibition state. Inversely, it was shown that disruption of the 1007 phosphorylation site enhances KCC2 function and in turn increases inhibitory signalling (Weber M et al. 2014. J Biol Chem 289(27):18668-18679; Moore et al.2018., supra), which together lead to the realization that post-transcriptionally changing the threonine at this position to a different residue (that could not be phosphorylated) would in principle make that the KCC2 protein more active in its inhibitory effect and would therethrough allow a potential treatment of disorders in which KCC2 activity is diminished, either through lowered expression, loss-of-function mutations or through post-translational modification processes, such as an increased phosphorylation rate of the threonine at position 1007.

[0049] Because of the plethora of disorders caused by a lowered KCC2 activity, as discussed above, any of such disorders could potentially be treated when the KCC2 activity could (transiently) be upregulated to yield a higher inhibitory effect, even when the human SLC12A5 gene, encoding KCC2, is the consensus sequence as mentioned above. Editing the T1007 site, encoded by the ACC codon, to an alanine (T1007A) encoded by a GCC codon, is the subject of Int. Patent Application Publication No. WO2024 / 206175, which is herein incorporated by reference in its entirety. The technology that the inventors thereof envisioned is generally referred to as ‘RNA editing’, in which a specific adenosine present in a transcript molecule, such as a pre-mRNA or a mRNA molecule, is deaminated to an inosine, which is seen by the translation machinery as a guanosine. Hence, by editing the ACC codon for the threonine residue at position 1007 in KCC2b (or position 1030 in KCC2a) to an ICC codon (= GCC) in the transcript, the resulting protein would comprise an alanine residue at this position instead of a threonine and the protein can no longer be phosphorylated at this site. The inventors realized that the change should reflect a gain-of-function alteration of the protein, which technically cannot be induced in human subjects through knock-in procedures as outlined in Moore et al. (2018), supra, or by gene therapy.

[0050] KCC2, as well as KCC4, are locked in their autoinhibition state (for which in KCC2 phosphorylation of the residues at T906 and T1007 are required) at least through an N- terminal peptide that binds at the cytosolic facing cavity (Xie Y et al. 2020. Sci Adv 6:eabc5883; see FIG. 2 for the envisioned model, proposed by Zhang et al. 2021). Dephosphorylation of the T906 and T1007 sites makes that the N-terminal peptide dissociates from the cytosolic facing cavity and subsequently associates with residues in the C-terminal domain, which dislocation allows the protein dimer to get into its active state (Xie et al.2020). The 27 amino acid containing N-terminal peptide in KCC2a has the following sequence: 81-VSSLLSGLAN-YTNLPQG-SREHEEAENN-107 (SEQ ID NO:28), in which the N-terminal part is referred to as N1, the central part is referred to as L12, and the C-terminal part is referred to as N2. Starting position is residue 81 and the last position is residue 107 in KCC2a. In KCC2b these positions are 58 and 84, respectively. The transcript molecule that encodes the 27 amino acids of the N-terminal peptide of SEQ ID NO:28 is boxed in FIG.1A and FIG.1B and is provided as SEQ ID NO:27, which is present in the human KCC2 transcript according to NCBI Reference Sequence: NM_020708.5 that represents the NCBI Ref. Seq. No. NP_065759 (NM_020708).

[0051] Xie et al. (2020) generated several mutants within the N-terminal peptide to determine the rate of K+ influx in comparison to the unmutated version protein. Changes in several positions (Y91A, T92E, N93A, L94A, Q96A, H101A, E102A / E105A; all for KCC2a) indeed showed an increase in K+ influx, indicating the importance of the interaction of the N-terminal peptide with its respective binding partners elsewhere in the protein to lock the KCC2 dimer in an autoinhibition state. The exact mutations that were generated by Xie et al. (2020) cannot be made by RNA editing, but the inventors envisioned that other amino acids changes within the N-terminal peptide caused by editing could provide a similar effect. Table 1 shows 11 edited sites in the KCC2 protein, of which the first 8 are within the N-terminal peptide. The codon encoding leucine at position 94 (L94) cannot be edited through RNA editing, as disclosed herein, because this residue is encoded by a CUG codon. Two further positions listed in Table 1 (Y446 and S932 in KCC2b) are located elsewhere in the KCC2 protein (see below), whereas T906 (in KCC2b) represents the other phosphorylation site besides the T1007 position referred to above. Table 1 showsthe different numbered positions in KCC2a and KCC2b, the unedited codon, the edited codon, the codon how it translates and the encoded amino acid (single letters), as well as the amino acid after RNA editing.

[0052] Table 1. KCC2 editing sites as disclosed herein. The positions in the two KCC2 isoforms are provided with their respective one-letter amino acid codes. After RNA editing the codon in the transcript molecule changes to encode a different amino acid. These changes are provided for both KCC2a and KCC2b. The coding sequence (mRNA transcript, 3420 nt including ATG start codon and TGA stop codon; see SEQ ID NO:474) and encoded protein sequence (1139 amino acids; see SEQ ID NO:475) of the KCC2a isoform are according to NCBI Reference Sequence: NM_001134771.2. The coding sequence (mRNA transcript, 3351 nt including ATG start codon and TGA stop codon; see SEQ ID NO:476) and encoded protein sequence (1116 amino acids; see SEQ ID NO:477) of the KCC2b isoform are according to NCBI Reference Sequence: NM_020708.5.

[0053] The RNA editing technology provides a unique transient method of altering the KCC2 protein in the CNS of human individuals in need thereof, preferably in the treatment of (chronic) pain and / or seizure (epilepsy) disorders, without altering the individual’s genome. It also allows the treatment of CNS disorders in which the KCC2 protein is attenuated in its expression or mutated towards a loss-of-function mutant, such as those identified by Stödberg and colleagues (2015. Nat Commun 6:8038) and Saitsu and colleagues (2016. Sci Rep 6:30072).

[0054] In one embodiment, the disclosure relates to EONs that are used to specifically cause the deamination of a specific target adenosine in the transcript of the (human) mutant SLC12A5 transcript (pre-mRNA and / or mRNA) in vivo, using endogenous deaminating enzymes (see below), to produce a KCC2 protein that: i) can no longer be locked in its autoinhibitory state through the position of its N-terminal peptide that – in the autoinhibitory state – is bound to the cytosolic facing cavity; ii) has an increased K+ transport activity; iii) cannot be phosphorylated at the T906 position encoded by the codon in which the adenosine was present. In some embodiments, after RNA editing the resulting KCC2 protein (be it the KCC2a and / or the KCC2b isoform) is enhanced in its transport function.

[0055] RNA editing is a natural process through which eukaryotic cells alter the sequence of their RNA molecules, often in a site-specific and precise way, thereby increasing the repertoire of genome encoded RNAs by several orders of magnitude. RNA editing enzymes have been described for eukaryotic species throughout the animal and plant kingdoms, and these processes play an important role in managing cellular homeostasis in metazoans from the simplest life forms (such as Caenorhabditis elegans) to humans. Examples of RNA editing are adenosine (A)-to-inosine (I) conversions and cytidine (C)-to-uridine (U) conversions, which occur through enzymes called Adenosine Deaminases acting on RNA (ADAR) and APOBEC / AID (cytidine deaminases that act on RNA), respectively.

[0056] ADAR is a multi-domain protein, comprising a catalytic domain, and two to three double-stranded (ds) RNA 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 I in a nearby, predefined, position in the target RNA, by deamination of the nucleobase. As mentioned above, inosine is read as guanosine by the translational machinery of the cell, meaning that, if an edited adenosine is in a coding region of an mRNA or pre-mRNA, it can recode the protein sequence. A-to-I conversions may also occur in 5’ non-coding sequences of a target mRNA, creating new translational start sites upstream of the original start site, which gives rise to N-terminally extended proteins, or in the 3’ UTR or other non-coding parts of the transcript, which may affect the processing and / or stability of the RNA. In addition, A-to- I conversions may take place in splice elements in introns or exons in pre-mRNAs, therebyaltering the pattern of splicing. As a result, exons may be included or skipped. The enzymes catalysing adenosine deamination are within an enzyme family of ADARs, which include human deaminases hADAR1 and hADAR2, as well as hADAR3. However, for hADAR3 no deaminase activity has been demonstrated.

[0057] The use of antisense 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.2013. Proc Natl Acad Sci USA 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, genetically fused to the adenosine deaminase domain of a truncated natural ADAR protein. It requires target cells to be either transduced with the fusion protein, which is a major hurdle, or that target cells are transfected with a nucleic acid construct encoding the engineered adenosine deaminase fusion protein for expression. The system described by Vogel et al. (2014) 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 Patent No. 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 complementary to the oligonucleotide were edited.

[0058] It is known that ADAR may act on any dsRNA. Through a process sometimes referred to as ‘promiscuous editing’, the enzyme will edit multiple adenosines 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 oligonucleotides.

[0059] 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. Intl. Patent Application Publication No. WO2016 / 097212, which is herein incorporated by reference in its entirety, discloses oligonucleotides for the targeted editing of RNA, wherein the oligonucleotides 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 (or 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 oligonucleotides”. The recruitment portion acts in recruiting a natural ADAR enzyme present in the cell to the dsRNA formed by hybridization of the target sequence with the targeting portion. Due to the recruitment portion, there is no need for conjugated entities or presence of modified recombinant ADAR enzymes. Intl. Patent Application Publication No. WO2016 / 097212, which is herein incorporated by reference in its entirety, 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 oligonucleotide itself, and are thought to attract (endogenous) ADAR. Similar stem-loop structure-comprising systems for RNA editing have been described in Intl. Patent Application Publication Nos. WO2017 / 050306, WO2020 / 001793, WO2017 / 010556, WO2020 / 246560, and WO2022 / 078995, all of which are herein incorporated by reference in their entireties.

[0060] Intl. Patent Application Publication Nos. WO2017 / 220751 and WO2018 / 041973, which are herein incorporated by reference in their entireties, describe a next generation type of oligonucleotides that do not comprise such a stem-loop structure but that are (almost fully) complementary to the targeted area. In one embodiment, one or more mismatching nucleotides, wobbles, or bulges exist between the oligonucleotide and the target sequence. A sole mismatch may be at the site of the nucleoside opposite the target adenosine, but inother embodiments oligonucleotides (often referred to as “RNA editing oligonucleotides”, abbreviated to ‘EONs’, although they do not have the enzymatic deamination or editing activity themselves) 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 EONs lacking a stem-loop structure and with endogenous ADAR enzymes when the sequence of the EON was carefully selected such that it could attract / recruit ADAR.

[0061] The “orphan nucleoside”, which is defined as the nucleoside in the EON that is positioned directly opposite the target adenosine in the target RNA molecule, did not carry a 2’-OMe modification. The orphan nucleoside can be a deoxyribonucleoside (DNA) without any substitution at the 2’ position of the ribose sugar moiety, wherein the remainder of the EON could still carry 2’-O-alkyl modifications (such as 2’-OMe) at their ribose sugars.

[0062] The nucleotides directly surrounding the orphan nucleoside contained chemical modifications (including being DNA and not 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 EONs against breakdown (described in Int. Patent Application Publ. No. WO2018 / 134301). 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 publications in the field, such as Intl. Patent Application Publication Nos. WO2019 / 111957, WO2019 / 158475, WO2020 / 165077, WO2020 / 201406, WO2020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, WO2022 / 007803, WO2022 / 018207, WO2022 / 026928, and WO2022 / 124345, all of which are herein incorporated by reference in their entireties.

[0063] The use of specific sugar moieties has been disclosed in for instance Intl. Patent Application Publication Nos. WO2020 / 154342, WO2020 / 154343, WO2020 / 154344, WO2022 / 103839, and WO2022 / 103852, which are herein incorporated by reference in their entireties, 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, orspecifically for RNA-editing oligonucleotides, related to a wide variety of target sequences) has been described in Intl. Patent Application Publication Nos. WO2011 / 005761, WO2014 / 010250, WO2014 / 012081, WO2015 / 107425, WO2017 / 015575 (HTT), WO2017 / 062862, WO2017 / 160741, WO2017 / 192664, WO2017 / 192679 (DMD), WO2017 / 198775, WO2017 / 210647, WO2018 / 067973, WO2018 / 098264, WO2018 / 223056 (PNPLA3), WO2018 / 223073 (APOC3), WO2018 / 223081 (PNPLA3), WO2018 / 237194, WO2019 / 032607 (C9orf72), WO2019 / 055951, WO2019 / 075357 (SMA / ALS), WO2019 / 200185 (DM1), WO2019 / 217784 (DM1), WO2019 / 219581, WO2020 / 118246 (DM1), WO2020 / 160336 (HTT), WO2020 / 191252, WO2020 / 196662, WO2020 / 219981 (USH2A), WO2020 / 219983 (RHO), WO2020 / 227691 (C9orf72), WO2021 / 071788 (C9orf72), WO2021 / 071858, WO2021 / 178237 (MAPT), WO2021 / 234459, WO2021 / 237223, and WO2022 / 099159, which are herein incorporated by reference in their entireties. 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 Intl. Patent Application Publication Nos. WO2020 / 157008 and WO2021 / 136404 (USH2A); WO2021 / 113270 (APP); WO2021 / 113390 (CMT1A); WO2021 / 209010 (IDUA, Hurler syndrome); WO2021 / 231673 and WO2021 / 242903 (LRRK2); WO2021 / 231675 (ASS1); WO2021 / 231679 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2); WO2021 / 231685 and WO2021 / 231692 (OTOF, autosomal recessive non-syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); WO2021 / 130313 and WO2021 / 231830 (ABCA4); and WO2021 / 243023 (SERPINA1), which are herein incorporated by reference in their entireties.

[0064] The EONs as disclosed herein can recruit deaminating enzymes, such as ADAR1 and / or ADAR2 that are endogenously present (= naturally present) in a cell. An EON 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 double-stranded EON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine. The EONs of the present disclosure may be conjugated to a saponin either directly or indirectly via linkers. These conjugations providean improved endosomal release after the EON has entered the target cell. In some embodiments the present disclosure provides EONs that are conjugated to a saponin and subsequently ‘packaged’ in lipid nanoparticles (LNPs) through which the EON can be delivered very efficiently to the target cells and subsequently released from the endosome as soon as the EON has entered the cell.

[0065] 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 after serving as a template for translation. 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 SLC12A5 transcript is highly specific and will not affect other (pre-) mRNA molecules, the risk of adverse side effects as often observed with chaperone compounds or agonists is low.

[0066] The EONs as disclosed herein can recruit deaminating enzymes, such as ADAR1 and / or ADAR2 that are endogenously present in a cell. An EON 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 double-stranded EON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine.

[0067] The oligonucleotides are herein abbreviated to “EONs”, as ‘editing oligonucleotides’, 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 EONs 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 EON itself, which is constantly prone to breakdown because of nucleases present in a natural cell.

[0068] Many chemical modifications are available for the generation of EONs (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 (Int. Patent Application Publ. No. WO2019 / 158475). In a similar fashion, it was found earlier that a PS linkage at some, butnot all, internucleoside linkages surprisingly appeared compatible with efficient ADAR engagement and editing (Int. Patent Application Publ. No. 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 EON appeared compatible with efficient engagement of an enzyme with nucleotide deamination activity and with subsequent deamination (Int. Patent Application Publ. No. WO2020 / 165077). Whereas the properties of phosphonoacetate and UNA modifications were known as such, the compatibility thereof with engagement of enzymes with nucleotide deamination activity and with the deamination reaction was not known.

[0069] RNA editing is often applied to correct G>A mutations that cause a disease. This is not the case in the present disclosure. Without being bound by theory, RNA editing introduces a mutation that causes the resulting protein to have an altered and / or an increased activity, e.g. defined as K+ influx rate, in comparison to the unedited version, and / or that is less inhibited by autoinhibition. The increased activity of KCC2 can be induced by changing single amino acids that disrupt the N-terminus sites of KCC2 (Xie et al.2020), alternative potassium bindings sites (Becker L et al.2023. J Biol Chem.299(10):105190), and C-terminus phosphorylation sites (Cordshagen A et al. 2018. J Biol Chem. 293(44):16984-16993). The EONs and editing methods described herein, lead to specific amino acids alterations at such sites, where changes in the protein-coding sequence of KCC2 are the consequences of RNA editing (e.g. A-to-G mutations).

[0070] Non-limiting 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 alpha1-antitrypsin (A1AT) deficiency; see e.g., Int. Patent Application Publ. Nos. WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, and WO2021 / 243023), IDUA (for the treatment of Hurler syndrome; see e.g., Int. Patent Application Publication Nos. WO2017 / 220751, WO2018 / 041973, and WO2021 / 209010), LRRK2 (for the treatment of Parkinson’s disease; see e.g., Int. Patent Application Publication Nos. WO2016 / 097212, WO2017 / 220751, WO2018 / 041973, WO2021 / 231673 and WO2021 / 242903), ABCA4 (for the treatment of Stargardt disease; see e.g., Int. Patent Application Publication Nos. WO2021 / 130313 and WO2021 / 231830), USH2A (for the treatment of Usher syndrome; see e.g., Int. Patent Application Publication Nos. WO2020 / 157008, WO2020 / 219981 and WO2021 / 136404), APP (see e.g., Int. Patent Application Publication No.WO2021 / 113270), CMT1A (see e.g., Int. Patent Application Publication No. WO2021 / 113390), ASS1 (see e.g., Int. Patent Application Publication No. WO2021 / 231675), GJB2 (see e.g., Int. Patent Application Publication No. WO2021 / 231679), MECP2 (for the treatment of Rett syndrome; see e.g., Int. Patent Application Publication Nos. WO2019 / 071274 and WO2021 / 231680), OTOF (for the treatment of autosomal recessive non-syndromic hearing loss; see e.g., Int. Patent Application Publication Nos. WO2021 / 231685 and WO2021 / 231692), XLRS (see e.g., Int. Patent Application Publication No. WO2021 / 231691), and PCSK9 (for the treatment of hypercholesterolemia; see e.g., Int. Patent Application Publication No. WO2023 / 152371). Embodiments

[0071] Disclosed herein is an RNA editing oligonucleotide (EON) forming a double- stranded complex with a human SLC12A5 RNA molecule in a cell, said SLC12A5 RNA molecule comprising a target adenosine, wherein the target adenosine is: i) the A in the UAC codon encoding tyrosine (Y) at position 68; ii) the A in the ACC codon encoding threonine (T) at position 69; iii) the first A in the AAC codon encoding asparagine (N) at position 70; iv) the second A in the AAC codon encoding asparagine (N) at position 70; v) the A in the CAG codon encoding glutamine (Q) at position 73; vi) the A in the CAU codon encoding histidine (H) at position 78; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82; ix) the A in the UAC codon encoding tyrosine (Y) at position 446; x) the A in the AGU codon encoding serine (S) at position 932; and / or xi) the A in the ACG codon encoding threonine (T) at position 906; wherein each amino acid position is according to the amino acid sequence of SEQ ID NO:477, wherein a nucleotide in the EON that is directly opposite the target adenosine is the orphan nucleotide, wherein counting of the nucleotides in the EON is such that the orphan nucleotide is number 0 and the nucleotides 5’ from the orphan nucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, optionally wherein the double-stranded complex can recruit an endogenous ADAR enzyme in the cell to deaminate the target adenosine into an inosine, thereby editing the SLC12A5 RNA molecule.

[0072] In one embodiment, the deamination of the target adenosine results in a SLC12A5- encoded KCC2 protein with an increased activity, optionally resulting in an increased K+ flux, optionally wherein the increased activity results in a higher GABAergic inhibition.

[0073] In one embodiment, the cell is a neuron, preferably a brain cell, and the RNA molecule is a pre-mRNA or an mRNA molecule.

[0074] In one embodiment, disclosed herein is an EON with at least one non-naturally occurring chemical modification, and / or comprising one or more additional non-naturally occurring chemical modifications in the ribose, linkage, or base moiety.

[0075] In one embodiment, the orphan nucleotide is a cytidine, a cytidine analog, a uridine, or a uridine analog.

[0076] In one embodiment, the orphan nucleotide is a deoxynucleotide.

[0077] In one embodiment, the orphan nucleotide is a cytidine analog. One preferred cytidine analog is a deoxyribonucleoside comprising a 6-amino-5-nitro-3-yl-2(1H)- pyridone nucleobase (also known as Benner’s base, Z base, Z nucleobase or simply as ‘Z’).

[0078] Another preferred cytidine analog is a deoxyribonucleoside comprising a cytosine analog nucleobase according to formula (I):or any of its tautomeric forms, wherein: R1, R2, R4and / or R5is H; OH; SH; =O; NH2; a halogen; a linear or branched lower (C1-C10) alkyl; or a C1-C6 cycloalkyl, wherein the alkyl and / or cycloalkyl is optionally interrupted by one or more heteroatoms; and R3is H; OH; SH; =O; NH2; or a halogen. If R1, R2, R3, R4and R5are H, which is a preferred embodiment, the cytosine analog is referred to as a 5-aza-5,6-dihydro cytosine nucleobase (also known as E base, E nucleobase, or simply as ‘E’), or any of its enol or keto tautomeric forms thereof. Examples of enol tautomers of the E base are enol-amino and enol-imino tautomers. Examples of keto tautomers of the E base are keto-amino and keto-imino tautomers.

[0079] The field of nucleoside analogs has long been instrumental in developing therapeutics targeting viral infections and epigenetic modifications. Among these, 5-aza- 5,6-dihydro cytosine represents a structurally novel cytosine (= nucleobase) analog with distinct chemical and biological properties. Unlike as observed in conventionalnucleosides, this base features a saturated 5,6 bond, significantly altering its tautomeric equilibrium and base-pairing characteristics. This unique tautomeric flexibility allows it to adopt multiple conformations in solution, a property linked to its capacity for inducing targeted mutagenesis in RNA-dependent replication systems (Li D et al. Proc Natl Acad Sci USA 2014, 111(32):E3252-E3259). One of the most well-characterized applications of 5-aza-5,6-dihydro cytosine, particularly when used in a deoxyribonucleoside form (2’- deoxy-5-aza-5,6-dihydro cytidine), has been in the context of viral lethal mutagenesis. This mechanism relies on the incorporation of error-prone nucleotides into the genome of rapidly mutating viruses, such as HIV-1, where it causes a progressive increase in mutation frequency beyond the error threshold necessary for viable replication. It has been shown that 5-aza-5,6-dihydro cytosine forms a mismatch with both adenine and guanine, a direct consequence of its ability to interconvert between distinct tautomeric states. It was further confirmed that the E nucleobase exists in a mixture of enol and keto tautomers, with the enol form being predominant under physiological conditions. This property is crucial in explaining its base-pairing promiscuity, which underlies its mutagenic effects (Li et al. 2014).

[0080] Beyond its role in antiviral mutagenesis, the E base has also demonstrated potential in epigenetic applications. DNA methylation, a key regulatory process in gene expression, is commonly dysregulated in malignancies. The structurally related nucleoside 2'-deoxy-5- azacytidine (also known as decitabine) is widely used for DNA demethylation in cancer therapy by covalently trapping DNA methyltransferases (DNMTs; Matoušová M et al. Epigenetics 2011, 6(6):769-776). However, this compound is associated with cytotoxicity and instability in aqueous environments. In contrast, 2’-deoxy-5-aza-5,6-dihydro cytidine retains DNMT inhibitory activity while exhibiting greater hydrolytic stability due to the absence of a reactive C5-C6 double bond. As a result, 2’-deoxy-5-aza-5,6-dihydro cytidine has been shown to induce hypomethylation with reduced cellular toxicity, making it a promising candidate for therapeutic applications requiring prolonged treatment. Importantly, preclinical and clinical investigations have reported minimal cytotoxic effects of 2’-deoxy-5-aza-5,6-dihydro cytidine in mammalian systems. Unlike many nucleoside analogs that indiscriminately disrupt host polymerases, this compound displays selectivity in incorporation, reducing the likelihood of off-target effects. Furthermore, its stability profile allows for alternative modes of administration, including potential oralformulations, which would overcome the limitations of current hypomethylating agents requiring parenteral delivery (Matoušová et al.2011).

[0081] In one aspect, the orphan nucleotide is a uridine analog such as a deoxynucleotide comprising an iso-uracil nucleobase.

[0082] In one embodiment, disclosed is an EON as disclosed herein, wherein the target A is: the first A in the GAA codon encoding glutamic acid (E) at position 79; the first A in the GAA codon encoding glutamic acid (E) at position 82; the A in the AGU codon encoding serine (S) at position 932; and / or the A in the ACG codon encoding threonine (T) at position 906, wherein the nucleotide on the -1 position in the EON can induce a syn conformation of the guanosine that is located directly 5’ from the target A, preferably wherein the nucleotide on the -1 position is a 7-deaza-2’-deoxyadenosine.

[0083] In one embodiment, disclosed is an EON as disclosed herein, wherein the target A is: the first A in the GAA codon encoding glutamic acid (E) at position 82; wherein the nucleotide on the -1 position in the EON can induce a syn conformation of the guanosine that is located directly 5’ from the target A, preferably wherein the nucleotide on the -1 position is a 7-deaza-2’-deoxyadenosine.

[0084] In one embodiment, disclosed is an EON as disclosed herein, wherein the target A is: the A in the ACC codon encoding threonine (T) at position 69; the first A in the AAC codon encoding asparagine (N) at position 70; the A in the CAG codon encoding glutamine (Q) at position 73; the A in the CAU codon encoding histidine (H) at position 78; and / or wherein the nucleotide on the -1 position in the EON is a deoxyinosine. In one embodiment, disclosed is an EON as disclosed herein, wherein the target A is: the A in the CAG codon encoding glutamine (Q) at position 73; and / or the A in the CAU codon encoding histidine (H) at position 78; wherein the nucleotide on the -1 position in the EON is a deoxyinosine.

[0085] In one embodiment, the one or more additional modifications in the linkage moiety is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methylphosphonate (MP; or MeP), sulfonylphosphoramidate, mesyl phosphoramidate (PNms), or a (1,3-dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi) internucleotide linkage. In one embodiment, the one or more additional modifications in the ribose moiety is 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.

[0086] In one aspect, the EON as disclosed herein, is covalently or non-covalently, directly or through a linker, bound to a triterpene glycoside saponin, preferably AG1856.

[0087] In one aspect, the EON as disclosed herein comprises the sequence 5’-(X)n- Uf*Gd*Zd*Ad˄Gm^Uf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the Y68C change. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the UAC codon encoding tyrosine (Y) at position 68 and is selected from the group consisting of SEQ ID NO:29-34.

[0088] In one aspect, the EON as disclosed herein comprises the sequence 5’-(X)n- Gf*Gd*Zd*Id˄Um^Af-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the T69A change. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the ACC codon encoding threonine (T) at position 69 and is selected from the group consisting of SEQ ID NO:35-40.

[0089] In one aspect, the EON as disclosed herein comprises the sequence 5’-(X)n- Gf*m5Ud*Zd*Id˄Gm^Uf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, andwherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the N70D change. In one aspect, the EON as disclosed herein is directed at deamination of the first adenosine in the AAC codon encoding asparagine (N) at position 70 and is selected from the group consisting of SEQ ID NO:41-46.

[0090] In one aspect, the EON as disclosed herein comprises the sequence 5’-(X)n- Gf*Gd*Zd*m5Ud˄Gm^Gf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the N70S change. In one aspect, the EON as disclosed herein is directed at deamination of the second adenosine in the AAC codon encoding asparagine (N) at position 70 and is selected from the group consisting of SEQ ID NO:47-52.

[0091] In one aspect, the EON as disclosed herein comprises the sequence: 5’-(X)n-Cf*Cd*Zd*Id˄Gm^Gf-(Y)m-3’, 5’-(X)n-Cf*Cd*Zd*Id˄Gm*Gf-(Y)m-3’, 5’-(X)n-Cf*m5Ce*Zd*Id˄Gm*Gf-(Y)m-3’, 5’-(X)n-Cf*Cd*Zd#Id˄Gm*Gf-(Y)m-3’, 5’-(X)n-Cf*Cd*Zd#Id*Gm*Gf-(Y)m-3’, 5’-(X)n-Cf*Cd*Zd*Id^Im^Gf-(Y)m-3’, 5’-(X)n-Cf*Cd*Ed*Id^Im^Gf-(Y)m-3’, or 5’-(X)n-Cf*Cd*Ed*Id^Gm^Gf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the Q73R change.

[0092] In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the CAG codon encoding glutamine (Q) at position 73, wherein the EON comprises the sequence: 5’-…N5*N4N3N2*N1*O*M1^M2M3…-3’ wherein: - N5 is a 2’-F modified nucleotide comprising an uracil nucleobase (Uf); - N4 is Uf, or a 2’-OMe modified nucleotide comprising an uracil nucleobase (Um); - N3 is a 2’-OMe modified cytidine nucleotide (Cm), or a 2’-F modified cytidine nucleotide Cf; - N2 is Cf; - N1 is a deoxycytidine (Cd); - O is the orphan nucleotide that is directly opposite the target adenosine, wherein O is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase (Zd), or a deoxynucleotide comprising a 5-aza-5,6-dihydro cytosine nucleobase (Ed); - M1 is deoxynucleotide comprising a hypoxanthine nucleobase (Id); - M2 is a 2’-OMe modified nucleotide comprising a guanine nucleobase (Gm), or a hypoxanthine nucleobase (Im); - M3 is a 2’-F modified nucleotide comprising a guanine nucleobase (Gf), or a hypoxanthine nucleobase (If); - * is a phosphorothioate linkage; - ^ is a methylphosphonate linkage; - the linkage between N4 and N3 is a phosphorothioate linkage, or a (1,3- dimethylimidazolidin-2-ylidene) phosphoramidate linkage; - the linkage between N3 and N2 is a phosphodiester linkage, or a phosphorothioate linkage; and - the linkage between M2 and M3 is a phosphodiester linkage, or a phosphorothioate linkage.

[0093] In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the CAG codon encoding glutamine (Q) at position 73 and is selected from the group consisting of SEQ ID NO:53-122, 226-275, 335, 368-401, and 439-463. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the CAG codon encoding glutamine (Q) at position 73 and is selected from the group consisting ofSEQ ID NO:238, 266, 394, 462, 335, 395, 377, 379, 400, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 463, 58, 73, 84, 98, 111, 112, 113, 114, 115, 120, 234, 236, 248, 267, and 271.

[0094] In one aspect, the EON as disclosed herein comprises the sequence: 5’-(X)n-Cf*Ad*Zd*Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf*Ae*Zd*Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf*Ad#Zd*Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf*Ad*Zd#Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf#Ad*Zd*Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf#Ad*Zd#Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cm*Ad*Zd*Id˄Cm^Uf-(Y)m-3’, 5’-(X)n-Cf!Ad*Zd*Id˄Cm^Uf-(Y)m-3’, or 5’-(X)n-Cf*Ad*Ed*Id˄Cm^Uf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or 22, wherein m = 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the H78R change. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the CAU codon encoding histidine (H) at position 78 and is selected from the group consisting of SEQ ID NO:123-182, 276-325, 338, 402-438, and 464-468. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the CAU codon encoding histidine (H) at position 78 and is selected from the group consisting of SEQ ID NO:338, 418, 283, 404, 405, 411, 413, 420, 427, 428, 436, 464, 465, 466, 467, 468, 126, 125, 127, 130, 131, 137, 138, 140, 152, 179, 286, 292, 293, and 294.

[0095] In one aspect, the EON as disclosed herein comprises the sequence: 5’-(X)n-Cf*m5Ud*Zd*Id˄Am^Uf-(Y)m-3’, 5’-(X)n-Cf*m5Ud*Zd*7Ad#Am^Uf-(Y)m-3’, or 5’-(X)n-Cf*m5Ud*Zd#7Ad*Am^Uf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobasemodifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or 22, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the E79G change. In one aspect, the EON as disclosed herein is directed at deamination of the first adenosine in the GAA codon encoding glutamic acid (E) at position 79 and is selected from the group consisting of SEQ ID NO:183-200.

[0096] In one aspect, the EON as disclosed herein comprises the sequence: 5’-(X)n-Uf*m5Ud*Zd*Id˄Um^Gf-(Y)m-3’, 5’-(X)n-Uf*m5Ud*Zd*7Ad#Um^Gf-(Y)m-3’, or 5’-(X)n-Uf*m5Ud*Zd#7Ad*Um^Gf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to the linkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or 22, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:27 to bring about the E82G change. In one aspect, the EON as disclosed herein is directed at deamination of the first adenosine in the GAA codon encoding glutamic acid (E) at position 82 and is selected from the group consisting of SEQ ID NO:201-218.

[0097] In one aspect, the EON as disclosed herein comprises the sequence: 5’-(X)n-Cf*Ge*Zd*Id˄Um^Uf-(Y)m-3’, 5’-(X)n-Cm*Ge*Zd*Id˄Um^Uf-(Y)m-3’, 5’-(X)n-Cf*Ge*Zd*Cd˄Um^Uf-(Y)m-3’, 5’-(X)n-Cm*Ge*Zd*Cd˄Um^Uf-(Y)m-3’, 5’-(X)n-Cf*Ge*Zd*7Ad!Um^Uf-(Y)m-3’, 5’-(X)n-Cm*Ge*Zd*7Ad!Um^Uf-(Y)m-3’, 5’-(X)n-Cf*Ge*Zd*7Ad#Um^Uf-(Y)m-3’, or 5’-(X)n-Cm*Ge*Zd*7Ad#Um^Uf-(Y)m-3’, wherein the modifications are according to the accompanying drawings, wherein X and Y are nucleotides that are preferably modified and that may comprise sugar and nucleobase modifications as disclosed herein, wherein the internucleoside linkages are according to thelinkages as disclosed herein, wherein n = 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 21, wherein m = 3, 4, 5, 6, 7, 8, 9, or 10, and wherein the EON is sufficiently complementary to the target sequence of SEQ ID NO:7 to bring about the T906A change. In one aspect, the EON as disclosed herein is directed at deamination of the adenosine in the ACG codon encoding threonine (T) at position 906 and is selected from the group consisting of SEQ ID NO:9-22.

[0098] The disclosure also relates to a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an EON comprising a nucleotide sequence of an EON according to any one of SEQ ID NO:9 to 22, 29 to 218, 226 to 335, 338, and 368 to 468, wherein the orphan nucleotide is a cytidine or a uridine.

[0099] Disclosed herein is also a pharmaceutical composition comprising an EON or a vector according as disclosed herein, and a pharmaceutically acceptable carrier.

[0100] Disclosed herein is also an EON, a vector, or a pharmaceutical composition as disclosed herein for use as a medicament or for use in therapy.

[0101] The disclosure also relates to an EON, or a vector, as disclosed herein, for use in the treatment of a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity. In one embodiment, the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, and / or epilepsy.

[0102] The disclosure also relates to a use of an EON, or a vector, as disclosed herein, in the manufacture of a medicament for the treatment of a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity. In one embodiment, the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, and / or epilepsy.

[0103] The disclosure also relates to a method of editing a human SLC12A5 polynucleotide, the method comprising contacting the SLC12A5 polynucleotide with an EON capable of effecting an ADAR-mediated A to I editing of a target A in a codon encoding an amino acid that is associated with: i) an autoinhibitory folding of the N-terminal peptide of SEQ ID NO:28 in the SLC12A5-encoded protein KCC2 protein; ii) K+ binding and / or transport, and / or Cl- binding and / or transport; or iii) phosphorylation of the SLC12A5-encoded protein KCC2, thereby editing the SLC12A5 polynucleotide.

[0104] The disclosure also relates to a method of treating a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity, in an individual in need thereof, the method comprising contacting a SLC12A5 polynucleotide in a cell of the subject with an EON capable of effecting an ADAR-mediated A to I editing of a target A in a codon encoding an amino acid that is associated with: i) an autoinhibitory folding of the N-terminal peptide of SEQ ID NO:28 in the SLC12A5-encoded protein KCC2 protein; ii) K+ binding and / or transport; or iii) phosphorylation of the SLC12A5-encoded protein KCC2, thereby treating the individual.

[0105] In one aspect, the target A is: i) the A in the UAC codon encoding tyrosine (Y) at position 68; ii) the A in the ACC codon encoding threonine (T) at position 69; iii) the first A in the AAC codon encoding asparagine (N) at position 70; iv) the second A in the AAC codon encoding asparagine (N) at position 70; v) the A in the CAG codon encoding glutamine (Q) at position 73; vi) the A in the CAU codon encoding histidine (H) at position 78; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82; ix) the A in the UAC codon encoding tyrosine (Y) at position 446; x) the A in the AGU codon encoding serine (S) at position 932; and / or xi) the A in the ACG codon encoding threonine (T) at position 906. The disclosure also relates to a method of treating a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity, the method comprising administering to an individual in need thereof a therapeutically effective amount of an EON, a vector, or a pharmaceutical composition as disclosed herein. Preferably, the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, and / or epilepsy.

[0106] In one aspect, the disclosure relates to a method of deaminating a target adenosine in an SLC12A5 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: i) providing the cell with an EON as disclosed herein; ii) allowing uptake by the cell of the EON; iii) allowing annealing of the EON to the SLC12A5 pre-mRNA or mRNA molecule; iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine; and optionally v) identifying the presence of the inosine in the target RNA molecule. Preferably step v) comprises: a) determining the sequence of the SLC12A5 pre-mRNA or mRNA molecule; b) assessing the presence of an SLC12A5-encoded KCC2 protein with a lower phosphorylation rate or a gain-of-functionin K+ binding and / or transport; or c) using a functional read-out, preferably assessing the level of GABAergic inhibition in the cell.

[0107] The present disclosure also relates to a nucleic acid molecule for editing a target adenosine in a human SLC12A5 pre-mRNA or mRNA molecule, wherein the target region is selected from the group consisting of SEQ ID NO:1, 3, 5, 7, and 27. The target A is preferably selected from the group of 11 target adenosine residues as outlined in detail herein, e.g. without limitation Table 1, FIG.1A, FIG.1B, and FIG.1C.

[0108] Disclosed herein are EONs that can mediate RNA editing of a target adenosine in the human SLC12A5 transcript (pre-mRNA and / or mRNA), through which the resulting KCC2 protein is no longer inhibited in functioning as a K+ and / or Cl- transporter; and more preferably has a gain-of-function. The absence of phosphorylation at the T906 site (as well as the T1007 site) increases the activity of the resulting (mutant) KCC2 protein. The absence of an interaction of the N-terminal peptide of SEQ ID NO:28 with its counterpart residues that cause the KCC2 dimer to be in an autoinhibitory state, also increases the protein activity.

[0109] In one embodiment, the EON herein is a single-stranded oligonucleotide comprising an orphan nucleotide as defined above, wherein the orphan nucleotide is chemically modified as disclosed herein, and wherein the remainder of the oligonucleotide is also chemically modified to prevent it from nuclease breakdown also as disclosed herein. In one embodiment, the disclosure relates to any kind of oligonucleotide or heteroduplex oligonucleotide complex, that may or may not be bound to hairpin structures (internally or at the terminal end(s)), that may be bound to ADAR or catalytic domains thereof, or wherein the oligonucleotide is expressed through a vector, such as an adeno-associated virus (AAV), or wherein the oligonucleotide is in a circular format. It is to be understood that any kind of oligonucleotide-based RNA editing is encompassed by the disclosure if it relates to the deamination of an adenosine in the SLC12A5 transcript, preferably: - the second A in the AAC codon encoding asparagine (N) at position 70; - the A in the UAC codon encoding tyrosine (Y) at position 68; - the A in the ACC codon encoding threonine (T) at position 69; - the first A in the AAC codon encoding asparagine (N) at position 70; - the A in the CAG codon encoding glutamine (Q) at position 73; - the A in the CAU codon encoding histidine (H) at position 78;- the first A in the GAA codon encoding glutamic acid (E) at position 79; - the first A in the GAA codon encoding glutamic acid (E) at position 82; - the A in the UAC codon encoding tyrosine (Y) at position 446; - the A in the AGU codon encoding serine (S) at position 932; and / or - the A in the ACG codon encoding threonine (T) at position 906, wherein the position is according to the SLC12A5-encoded human KCC2b isoform.

[0110] In one embodiment, an EON herein is a ‘naked’ oligonucleotide, comprising a variety of chemical modifications in the ribose sugar, the base, and / or the internucleoside linkage of one or more of the nucleotides within the sequence, that can hybridize to the SLC12A5 transcript or a part thereof that includes the target adenosine, and can recruit endogenous ADAR for the deamination of the target adenosine. The endogenous ADAR enzyme is preferably human ADAR1 or ADAR2. The cell is preferably a human neuronal cell. The SLC12A5 transcript molecule is preferably a pre-mRNA or an mRNA molecule. The EON herein preferably targets an adenosine for deamination that causes a gain-of- function of the KCC2 protein. Although several mutations are known that cause a dysfunction of the KCC2 protein, an embodiment of an adenosine that is targeted through the EONs as disclosed herein is an adenosine that is in a codon for an amino acid that plays a role in the autoinhibitory activity brought about by the N-terminal peptide of SEQ ID NO:28 in binding to KCC2 itself. In one embodiment the adenosine is in a codon that encodes a phosphorylation site in KCC2, more preferably T906, wherein the resulting codon (after deamination of the adenosine) is no longer a phosphorylation site (T906A). Loss of phosphorylation of this site in KCC2 increases its GABAergic inhibitory activity, thereby lowering abnormal neuronal activation (e.g., causing chronic neurological pain) and synchronization that underlies seizures.

[0111] In one embodiment, the EON comprises at least one mismatch with the (overlapping) sequence of the target transcript molecule. When the orphan nucleotide is uridine, then the EON does not necessarily comprise a mismatch. Mismatches may be introduced in other parts of the EON, where required, if the EON is capable of hybridizing under natural conditions to the target transcript. In an embodiment, the EON herein comprises one or more mismatches, wobbles, or bulges, wherein a single mismatch may be present when the target adenosine has an opposite cytidine, or a uridine analog (that not fully matches in comparison to a uridine) in the EON. If the orphan nucleotide is a cytidine,that cytidine preferably does not comprise a 2’-OMe ribose substitution. Preferably, also when the orphan nucleotide is different from a cytidine, it also does not comprise a 2’-OMe ribose substitution if it hinders deamination by the ADAR enzyme.

[0112] Disclosed herein is a vector, preferably a viral vector, more preferably an adeno- associated virus (AAV) vector, comprising a nucleic acid molecule encoding an EON herein. When the EON is delivered through the means of a viral vector or a plasmid vector, the produced EON in the cell does not have chemical modifications. Further disclosed herein is a pharmaceutical composition comprising an EON as disclosed herein, or a viral vector or plasmid vector as disclosed herein, and a pharmaceutically acceptable carrier.

[0113] In an embodiment, disclosed herein is an EON, a vector, or a pharmaceutical composition for use in the treatment of a subject in need thereof, wherein the subject suffers from a disorder, wherein the KCC2 inhibitory activity is lowered or absent, either through a loss-of-function mutation, lowered expression of the transcript and / or protein, or through (potentially increased rates of) post-translational modifications such as activity-inhibiting phosphorylation of certain sites in the (wild-type) protein. In an embodiment, disclosed herein is an EON or a vector in the manufacture of a medicament for the treatment of a disorder in a subject, wherein the KCC2 inhibitory activity in neuronal cells is lowered or absent, either through a loss-of-function mutation, lowered expression of the transcript and / or protein, or through (potentially increased rates of) post-translational modifications such as activity-inhibiting phosphorylation of certain sites in the (wild-type) protein. Without wishing to be bound by theory, RNA editing through human ADAR is thought to take place on primary transcripts in the nucleus, where during transcription or splicing, or in the cytoplasm for example mature mRNA, miRNA or ncRNA can be edited.

[0114] It should be clear, that targeted editing as described herein can be applied to any adenosine within the SLC12A5 transcript if the deamination of the adenosine results in an increase or restoration of KCC2 protein function. As outlined herein, however, it is preferred to target the first adenosine that is present in the codon encoding threonine at position 1007 of the mature KCC2b splice variant.

[0115] 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 bindingpartner; entire expression pathways may be altered by recoding miRNAs or their cognate sequences on target RNAs). Protein function or localization may be changed at will, by functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co- or post- translational modification, catalytic sites of enzymes, binding sites for binding partners, signals for degradation or activation and so on. These and other forms of RNA and protein “engineering”, whether to prevent, delay or treat disease or for any other purpose, in medicine or biotechnology, as diagnostic, prophylactic, therapeutic, research tool or otherwise, are encompassed by the present disclosure. Hence, an EON as disclosed herein may mediate the RNA editing of any target adenosine in the SLC12A5 transcript which results in improvement or restoration of the KCC2 protein function. The disclosure opens a whole new field of treating pathologies (such as neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and epilepsy), using genetic editing techniques.

[0116] The amount of EON 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 EONs could compete for binding to an ADAR within a cell, thereby depleting the amount of the entity, which is free to take part in RNA editing, but routine dosing trials will reveal any such effects for a given EON and a given target.

[0117] One suitable trial technique involves delivering the EON 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. Also, a suitable biomarker that can be used following the present disclosure is to detect phosphorylation or to determine K+ influx, and / or by assessing the function / activity of the KCC2 protein in a particular subject, before and after treatment, or with or without treating the subject with an EON or vector as disclosed herein. After this trial has been performed once then the knowledge can beretained, and future delivery can be performed without needing to take biopsy samples. A method as disclosed herein can thus include a step of identifying the presence of the desired change in the cell’s target RNA sequence, thereby verifying that the target RNA sequence has been modified. This step will typically involve sequencing of the relevant part of the target RNA, or a cDNA copy thereof (or a cDNA copy of a splicing product thereof, in case the target RNA is a pre-mRNA), as discussed above, and the sequence change can thus be easily verified. Alternatively, as indicated above, the change may be assessed on the function of the protein, or instance by measuring thallium, K+ and / or Cl- transport capacity of KCC2. The transport of thallium, a surrogate of potassium, is directly proportional to the number of active KCC2 potassium transporters. Thallium transport can then be detected by introducing a highly sensitive thallium indicator dye, 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.

[0118] 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 EON until enough target RNAs have been modified to provide a tangible benefit to the patient and / or to maintain the benefits over time.

[0119] EONs herein are particularly suitable for therapeutic use, and so the disclosure also relates to a pharmaceutical composition comprising an EON herein, or a vector or plasmid encoding an EON herein, and a pharmaceutically acceptable carrier. 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 disclosure also provides a delivery device (e.g., a syringe) that includes a pharmaceutical composition herein.

[0120] The disclosure also provides an EON herein for use in a method for introducing a mutation in a target SLC12A5 RNA sequence in a mammalian, preferably a human neuronal cell, as described herein. Similarly, the disclosure provides the use of an EON herein in the manufacture of a medicament for making a change in a target SLC12A5 RNA sequence in a mammalian, preferably a human neuronal cell, as described herein, and thereby treating, preventing, or ameliorating diseases related to diminished GABAergic inhibition, such as those resulting from lowered KCC2 activity.

[0121] The EONs herein are suitably administrated in an aqueous solution, e.g. saline, artificial cerebrospinal fluid, or in suspension, optionally comprising additives, excipientsand 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 µg / kg to about 100 mg / kg, preferably from about 10 µg / kg to about 10 mg / kg, more preferably from about 100 µg / kg to about 1 mg / kg. Administration may be by inhalation (e.g., through nebulization), intranasally, orally, by injection or infusion, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra-peritoneally, intrarectally, intrathecally, intra- cisterna magna, parenterally, and the like. Administration may be in solid form, in the form of a powder, a pill, a gel, a solution, a slow-release formulation, or in any other form compatible with pharmaceutical use in humans.

[0122] In one embodiment, a method herein comprises the steps of administering to the subject an EON or pharmaceutical composition herein, allowing the formation of a double stranded nucleic acid complex of the EON with its specific complementary target nucleic acid molecule in a cell in the subject; allowing the engagement of an endogenous present adenosine deaminating enzyme, such as ADAR2; and allowing the enzyme to deaminate the target adenosine in the target nucleic target molecule to an inosine, thereby alleviating, preventing or ameliorating the disease related to lowered GABAergic inhibition. The diseases that may be treated according to this method are preferably, but not limited to, the CNS diseases listed herein, and any other disease in which deamination of an adenosine in SLC12A5 transcripts would restore the KCC2 protein’s function in an individual in need thereof.

[0123] RNA editing molecules present in the cell will usually be proteinaceous in nature, such as the ADAR enzymes found in metazoans, including mammals. Preferably, the cellular editing entity is an enzyme, more preferably an adenosine deaminase or a cytidine deaminase, still more preferably an adenosine deaminase. These are enzymes with ADAR activity. The ones of most interest are the human ADARs, hADAR1 and hADAR2, including any isoforms thereof. RNA editing enzymes known in the art, for which oligonucleotide constructs according to the present disclosure may conveniently be designed, include the adenosine deaminases acting on RNA (ADARs), such as hADAR1 and hADAR2 in humans or human cells and cytidine deaminases. It is known that hADAR1 exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 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-γ). hADAR1 is also inducible by TNF-α. This provides an opportunity to develop combination therapy, whereby IFN-γ or TNF-α and EONs 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-γ or TNF-α levels in certain tissues of a patient, creating further opportunities to make editing more specific for diseased tissues. It will be understood by a person having ordinary skill in the art that the extent to which the editing entities inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0124] ADARs are multidomain proteins with N-terminal double stranded RNA binding domains (dsRBDs) and C-terminal deaminase domains. Two ADAR genes encode catalytically active ADARs in humans (ADAR encoding ADAR1 proteins and ADARB1 encoding the ADAR2 protein). ADAR1 is expressed in two protein isoforms (p110 and p150) that differ in their N-terminal structures. Since the substrate for ADARs is an RNA duplex, the enzymes access the reactive adenosine using a base flipping mechanism (Stephens OM et al. Biochemistry. 2000. 39(40): 12243-12251). Also, because ADARs require duplex RNA for activity, their reaction can be directed to specific adenosines in different transcripts using complementary guide strands for duplex formation at the target sites. This approach is currently being pursued to develop therapeutic guide strands that recruit ADARs to correct disease-causing mutations in RNA (Qu L. et al. Nat. Biotechnol. 2019. 37(9):1059-1069; Merkle T. Et al. Nat. Biotechnol. 2019. 37(2):133-138; Katrekar D. Et al. Nat. Methods 2019. 16(3):239-242; Monian P. Et al. Nat. Biotechnol. 2022: p. Doi: 10.1038 / s41587-022-01225-1).

[0125] While this approach is promising, ADARs have sequence preferences that make certain adenosines disfavored for reaction, limiting the current scope of this approach. For instance, the nearest neighbor nucleotide preferences for ADARs show a strong bias against reaction at adenosines in 5’-GA sites (Eggington J.M. et al. Nat. Commun. 2011. 2(319):DOI:10.1038 / ncomms1324). This preference is explained by structural studies of ADAR2 bound to transition state analog-containing RNA that suggest a clash between the 2-amino group of the 5’-G and G489 of the ADAR2 loop involved in stabilizing the flipped-out conformation required for the adenosine deamination reaction (Matthews et al. Nat Struct Mol Biol 2016. 23(5):426-433). Earlier work with fusion proteins bearing ADAR deaminase domains indicated that editing efficiency at 5’-GA sites could be improved with a G-A or G-G pair at the 5’ nearest neighbor (Schneider M.F. et al. Nucleic Acids Res.2014. 42(10):p.e87). However, the basis for this effect has not been reported nor has this effect been established for full length ADARs bearing native dsRBD RNA binding domains. In Int. Patent Application Publication No. WO2024 / 013361, which is incorporated herein in its entirety, it is shown that G-A and G-G pairs on the 5’ side of an editing site improve editing efficiency compared to a 5’ G-C pair for full length ADAR2 and ADAR1 p110. Using X-ray crystallography, the structure of an active fragment of human ADAR2 bound to duplex RNA bearing a G:G pair adjacent to an editing site was determined. In the present disclosure related to editing of the adenosine in the GAU codon at position 843 also a 5’- G is present next to the target adenosine in the target sequence. It discloses that the ADAR enzyme prefers the Gsyn:Ganti pairing since the 2-amino group of the 5’-G in the syn conformation does not clash with the minor groove of the enzyme. It was shown that the use of nucleosides capable of stable pairing with the 5’-G in the syn conformation enables more efficient editing within 5’-GA target sites, providing a solution to the problem in deaminating these unfavored editing sites.

[0126] The design rules as outlined in Int. Patent Application Publication No. WO2024 / 013361 also apply here. In the present disclosure, editing of: i) the first A in the GAA codon encoding glutamic acid (E) at position 79; ii) the first A in the GAA codon encoding glutamic acid (E) at position 82; iii) the A in the AGU codon encoding serine (S) at position 932; and iv) the A in the ACG codon encoding threonine (T) at position 906, are all preceded by a 5’-G in the target sequence, which means that the nucleotide on the - 1 position in the EON preferably induces a syn conformation of the 5’-G. The nucleotide at the -1 position is therefore preferably selected from the group consisting of: 7-deaza-2’-deoxyadenosine (7-deaza Ad); 7-deaza-2’-adenosine (7-deaza A); 7-deaza-2’-deoxy-2’-fluoroadenosine (7-deaza Af); 7-deaza-2’-deoxy-2’-ara-fluoroadenosine; 7-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-deoxyadenosine (3-deaza Ad);3-deaza-2’-adenosine (3-deaza A); 3-deaza-2’-deoxy-2’-fluoroadenosine (3-deaza Af); 3-deaza-2’-deoxy-2’-ara-fluoroadenosine; 3-deaza-2’-deoxy-2’,2’-difluoroadenosine; 3,7-dideaza-2’-deoxyadenosine (3,7-dideaza Ad); 3,7-dideaza-2’-adenosine (3,7-dideaza A); 3,7-dideaza-2’-deoxy-2’-fluoroadenosine (3,7-dideaza Af); 3,7-dideaza-2’-deoxy-2’-ara-fluoroadenosine; 3,7-dideaza-2’-deoxy-2’,2’-difluoroadenosine; 3-deaza-2’-O-[2-(methoxy)ethyl] adenosine; 3-deaza-2’-O-[2-methylamino-2-oxoethyl] adenosine; 2’-deoxy-2’-fluoroguanosine; 2’-ara-fluoro guanosine (FANA G); 2’,2’-difluoro guanosine; 2’-deoxyinosine (Id); 2’-OH-inosine; 2’-fluoroinosine (If); 2‘-ara-fluoro inosine (FANA I); 2’,2’-difluoro inosine; 5-formylindole-2’-deoxyriboside; 5-formyl-2’-fluoro-2’-deoxyriboside 5-formylindole-2’-ara-fluoro-2’-deoxyriboside; 5-formylindole-2’,2’-difluoro-2’-deoxyriboside; 5-formylindole-2’-O-methylriboside; 5-formylindole-2’-O-[2-(methoxy)ethyl]riboside; 5-formylindole-2’-O-[2-methylamino-2-oxoethyl]riboside beta-(4-amidino-1H-imidazol-1-yl) riboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-deoxyriboside; beta-(4-amidino-1H-imidazol-1-yl) 2’-ara-fluoro-2’-deoxyriboside; and beta-(4-amidino-1H-imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside.

[0127] In one embodiment, when a 5’-G is present directly adjacent to the target A, the nucleotide at position -1 in the EON is a 7-deaza-2’-deoxyadenosine. In one embodiment,the linkage at position -2 in the EON, which then links the nucleotide at position -1 with the one on position -2 is a PNdmi or a PNms linkage. Definitions

[0128] 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, U, or I. Whenever reference is made to a ‘deoxyoligoribonucleotide’ it may comprise the bases A, G, C, T, or I. However, an EON 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.

[0129] 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 UNA, a threose nucleic acid (TNA), a nucleotide including a linker comprising a phosphodiester (PO), phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP (or MeP), methyl thiophosphonate, phosphoramidate linkages, PNdmi according to the structure of formula (V) as described herein, and a linkage according to the structure of formula (II) 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 ormore phosphate groups. The terms ‘ribonucleoside’ and ‘deoxyribonucleoside’, or ‘ribose’ and ‘deoxyribose’ are as used in the art.

[0130] 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.

[0131] Whenever reference is made to nucleotides in the oligonucleotide, such as cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5- hydroxycytosine, and β-D-glucosyl-5-hydroxymethylcytosine are included. Whenever reference is made to adenine, N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine and 7-methyladenine are included. Whenever reference is made to uracil, dihydrouracil, isouracil, N3-glycosylated uracil, pseudouracil, 5-methyluracil, N1-methylpseudouracil, 4- thiouracil and 5-hydroxymethyluracil are included. Whenever reference is made to guanine, 1-methylguanine, 7-methylguanosine, N2,N2-dimethylguanosine, N2,N2,7- trimethylguanosine and N2,7-dimethylguanosine are included. Whenever reference is made to nucleosides or nucleotides, ribofuranose derivatives, such as 2’-deoxy, 2’-hydroxy, and 2’-O–substituted variants, such as 2’-O-methyl (2’-OMe), 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 according to the formula (V) as further outlined below, and the linkage structure according to formula (II) as further outlined in detail below.

[0132] The term ‘comprising’ encompasses ‘including’ as well as ‘consisting of’, e.g., a composition ‘comprising X’ may consist exclusively of X or may include somethingadditional, e.g., X + Y. The term ‘about’ in relation to a numerical value x is optional and means, e.g., x+10%.

[0133] 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.

[0134] The term ‘conducive to’ or ‘mediate’ can be used interchangeably with ‘capable of facilitating’. When used in the context of an EON that is conducive to ADAR editing (or can mediate ADAR editing), this means that the EON, 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 EON 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.

[0135] The term ‘mismatch’ is used herein to refer to opposing nucleotides in a double stranded RNA complex which do not form perfect base pairs according to the Watson- Crick base pairing rules. In the historical sense, mismatched nucleotides are G-A, C-A, U- C, A-A, G-G, C-C, U-U pairs. In some embodiments EONs as disclosed herein comprise fewer than four mismatches with the target sequence, for example 0, 1 or 2 mismatches. ‘Wobble’ base pairs are G-U, I-U, I-A, and I-C base pairs. When a U is placed opposite the target A, there is no mismatch, and the EON may be 100% complementary, although an iso-uridine (iso-U) opposite the target adenosine qualifies as a mismatch, since it does not pair according to the Watson-Crick rules of base pairing. When a C is placed opposite the target A, there is at least 1 mismatch between the EON 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 Res39(13):5669-5681). Characterization of optimal patterns of paired / mismatched nucleotides between the EONs and the target RNA also appears important to the development of efficient ADAR-based EON therapy.

[0136] The term ‘complementary’ as used herein refers to the fact that the EON hybridizes under physiological conditions to a second nucleic acid strand. Examples are (i) when the EON 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 EON may be complementary to a target sequence, there may be mismatches, wobbles and / or bulges between the EON and the target sequence, while under physiological conditions that EON 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 EON has enough matching nucleotides with the target sequence that under physiological conditions the EON hybridizes to the target RNA molecule. As shown herein, an EON may be complementary, but may also comprise one or more mismatches, wobbles and / or bulges with the target sequence, if under physiological conditions the EON is able to hybridize to its target.

[0137] The term ‘orphan nucleotide’ refers to the nucleotide in the EON 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.

[0138] 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.

[0139] 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 anysequence 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 EONs 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.

[0140] The nucleotide ‘numbering’ in an EON 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 EON 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.

[0141] 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.

[0142] The term ‘splice mutation’ relates to a mutation in a gene that encodes for a pre- mRNA, wherein the splicing machinery is dysfunctional in the sense that splicing of introns from exons is disturbed and due to the aberrant splicing, the subsequent translation is out of frame resulting in premature termination of the encoded protein. Often such shortened proteins degrade rapidly and do not have any functional activity.

[0143] Whenever a ‘naked’ form in relation to the EON as disclosed herein is referred to, it means that the EON 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 EON is therefore different from a form in which the EON is encoded (and delivered) by a viral genome or within a plasmid vector. When such viral vectors or plasmid vectors are administered, the encoded EON 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 EON is then not chemically modified and comprises solely naturally occurring RNA nucleotides.

[0144] The length of the EON 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 EON as disclosed herein is to be delivered through the expression of a viral vector, then the EON may be longer, such as 70, 80, 90, 100, 150, or 200 or more nucleotides in length.

[0145] The term ‘HEON’ refers to a heteroduplex double-stranded complex molecule wherein an EON as disclosed herein is hybridized to a partially or fully complementary, partially of fully overlapping sense oligonucleotide. Because the EON 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 EON as disclosed herein, for example by providing nucleotides with a ribose sugar moiety carrying a 2’-OMe substitution, a 2’-F substitution, or a 2’-MOE substitution. It is to be understood that the sense strand present in the HEON is a different entity in 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 EON from degradation when administered to the cell. In other words, the HEON is preferably formed before the EON is administered to the cell. Preferred aspects of HEONs that may be used for EONs as disclosed herein are discussed in Int. Patent Application Publication No. WO2024 / 084048, which is incorporated in its entirety herein. Chemical modifications

[0146] 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 EON as disclosed herein may also be used for a sense strand that is complementary to the EON, when the EON and the complementary strand form a HEON complex, such as described in Int. Patent Application Publication No.WO2024 / 084048, which is incorporated in its entirety herein, 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 EON as disclosed herein, but all other modifications relate to the EON as disclosed herein and any (protecting) sense oligonucleotide that may be used together with the EON 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 Int. Patent Application Publication No. WO2024 / 084048, which is incorporated in its entirety herein. These moieties and ligands may either be bound to the EON or its opposite strand, or both.

[0147] The skilled person knows that an oligonucleotide, such as an EON 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 (U). These consist of a pentose sugar, a ribose, a 5’-linked phosphate group which is linked via a phosphate ester, and a 1’- linked base. The 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 EON as disclosed herein are disclosed in Int. Patent Application Publication Nos. WO2020 / 154342, WO2020 / 154343, and WO2020 / 154344, which are incorporated herein in their entirety.

[0148] A nucleoside in the EON 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 thenucleotide has a natural ribose moiety, it may still be non-naturally modified in the base and / or the linkage.

[0149] It is recognized in the art that common limiting factors in oligonucleotide-based therapies are the oligonucleotide’s ability to be taken up by the cell, when delivered per se, or ‘naked’ (= without applying a viral vector or plasmid), the biodistribution and the resistance to nuclease-mediated breakdown. The skilled person is aware, and it has been described in detail in the art, that a variety of chemical modifications can assist in overcoming such limitations. Examples of such now commonly used chemical modifications are the 2’-OMe, 2’-F, 2’,2’-diF, and 2’-MOE modifications of the sugar and the use of PS linkages between nucleosides, as described herein. Scaffold modifications (ribose)

[0150] The ribose 2’ groups in all nucleotides of the EON as disclosed herein, except for the ribose sugar moiety of the orphan nucleotide that has certain limitations in respect of compatibility with RNA editing, can be independently selected from 2’-H (i.e., DNA), 2’- OH (i.e., RNA), 2’-OMe, 2’-MOE, 2’-F, or 2’-4’-linked (for instance a locked nucleic acid (LNA)), or other ribosyl 1’-substitutions, 2’ substitutions, 3’ substitutions, 4’ substitutions or 5’ substitutions. The orphan nucleotide in the EON that comprises no other chemical modifications to the ribose sugar, the base, or the linkage preferably does not carry a 2’- OMe or 2’-MOE substitution when the nucleobase is a naturally occurring cytosine, but may carry a 2’-F, a 2’,2’-difluoro (diF), or 2’-ara-F (FANA) substitution or may be DNA. Int. Patent Application Publication No. 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).

[0151] An EON as disclosed herein may comprise one or more nucleotides carrying a 2’- MOE ribose modification. Also, an EON 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 EON as disclosed herein may comprise a 2’-OMe ribose modification at a position that does not comprise a 2’-MOE ribose modification. An EON as disclosed herein may comprise deoxynucleotides at positions thatdo not comprise a 2’-MOE or a 2’-OMe ribose modification, or other 2’ ribose substitution. An EON as disclosed herein may comprise one or more nucleotides comprising a 2’ substitution comprising a 2’-MOE, 2’-OMe, 2’-OH, 2’-deoxy, TNA, 2’-fluoro (2’-F), 2’,2’- difluoro (diF) modification, 2’-fluoro-2’-C-methyl modification, or a 2’-4’-linkage (i.e., a bridged nucleic acid such as an LNA, or examples mentioned in e.g. Int. Patent Application Publication No. WO2018 / 007475). Other nucleic acid monomers that may be used in an EON 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.

[0152] A wide variety of 2’ modifications that may present in an EON as disclosed herein are known in the art, including but not limited to the modifications outlined in detail in Int. Patent Application Publication Nos. 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 EON 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 EON as disclosed herein comprises an 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 LNA). In one aspect, the EON as disclosed herein comprises at least one nucleotide comprising a TNA ribose modification. In one aspect, the EON 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 EON, which may be present together with an identical 2’ modification in the orphan nucleotide as discussed above. Base modifications

[0153] 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 CF moieties) or aromatic electronic interactions. Cytosine, thymine, and uracil are pyrimidine bases, andare 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 EON 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.

[0154] The nucleobases at any position in the EON 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 can be absent (also referred to as abasic nucleotides), such as 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.

[0155] A scaffold modification indicates the presence of a modified version of the ribosyl moiety as naturally occurring in RNA (i.e., the pentose moiety), such as bicyclic sugars, tetrahydropyrans, hexoses, morpholinos, 2’-modified sugars, 4’-modified sugar, 5’- modified sugars and 4’-substituted sugars. Examples of suitable modifications include, but are not limited to 2’-O-modified RNA monomers, such as 2’-O-alkyl or 2’-O- (substituted)alkyl such as 2’-OMe, 2’-O-(2-cyanoethyl), 2’-MOE, 2’-O-(2- thiomethyl)ethyl, 2’-O-butyryl, 2’-O-propargyl, 2’-O-allyl, 2’-O-(2-aminopropyl), 2’-O- (2-(dimethylamino)propyl), 2’-O-(2-amino)ethyl, 2’-O-(2-(dimethylamino)ethyl); 2’- deoxy (DNA); 2’-O-(haloalkyl)methyl such as 2’-O-(2-chloroethoxy)methyl (MCEM), 2’- O-(2,2-dichloroethoxy)methyl (DCEM); 2’-O-alkoxycarbonyl such as 2’-O-[2- (methoxycarbonyl)ethyl] (MOCE), 2’-O-[2-N-methylcarbamoyl)ethyl] (MCE), 2’-O-[2- (N,N-dimethylcarbamoyl)ethyl] (DCME); 2’-halo e.g. 2’-F, FANA; 2'-O-[2- (methylamino)-2-oxoethyl] (NMA); a bicyclic or bridged nucleic acid (BNA) scaffold modification such as a conformationally restricted nucleotide (CRN) monomer, a LNA monomer, a xylo-LNA monomer, an α-LNA monomer, an α-l-LNA monomer, a β-d-LNA monomer, a 2’-amino-LNA monomer, a 2’-(alkylamino)-LNA monomer, a 2’- (acylamino)-LNA monomer, a 2’-N-substituted 2’-amino-LNA monomer, a 2’-thio-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-2H-pyran nucleic acid (DpNA) monomer, a 2’-C-bridged bicyclic nucleotide (CBBN) monomer, an oxo-CBBN monomer, a heterocyclic-bridged BNA monomer (such as triazolyl or tetrazolyl-linked), an amido-bridged BNA monomer (such as AmNA), an urea-bridged BNA monomer, a sulfonamide-bridged BNA monomer, a bicyclic carbocyclic nucleotide monomer, a TriNA monomer, an α-l-TriNA monomer, a bicyclo DNA (bcDNA) monomer, an F-bcDNA monomer, a tricyclo DNA (tcDNA) monomer, an F-tcDNA monomer, an alpha anomeric bicyclo DNA (abcDNA) monomer, 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), UNA; an inverted version of any of the monomers above. All these modifications are known to the person skilled in the art. The orphan nucleotide

[0156] 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.

[0157] The crystal structure of ADAR2 E488Q bound to double stranded RNA (dsRNA) revealed that the glutamine (Gln; 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 (Gln) the amide group of the glutamine is absent and is instead a carboxylic acid. To obtain the same contact of the orphan cytidine with the E488Q mutant would then, for the wild-type situation, require protonation for this contact to occur.

[0158] To make use of endogenously expressed ADAR2 to correct disease relevant mutations, it is essential to maximize the editing efficiency of the ADAR2 enzyme that is naturally present in the cell. Int. Patent Application Publication No. WO2020 / 252376 discloses the use of EONs 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 EON 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’; Yang et al. Nucl Acid Res 2006.34(21):6095-6101) that were initially selected because they offerhydrogen-bond donation at N3 with minimal perturbation to the shape of the nucleobase. Benner’s base is also referred to with its chemical name 6-amino-5-nitro-3-yl-2(1H)- pyridone. The presence of the cytidine analog in the EON 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 an LNA), or other 2’ substitutions. The 2’-4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker.

[0159] The orphan nucleotide in the EON as disclosed herein is preferably a cytidine or analog thereof (such as a nucleotide carrying a Benner’s base) or a uridine or analog thereof (such as iso-uridine). The orphan nucleotide, whether it is a cytidine or analog thereof, or a uridine or analog thereof, preferably comprises a deoxyribose (2’-H; = DNA) but may also comprise a diF modification at the 2’ position of the sugar. In one aspect at least one and in another aspect both the neighbouring (directly adjacent) nucleotides flanking the orphan nucleotide do not comprise a 2’-OMe modification. Complete modification wherein all nucleotides of the oligonucleotide hold a 2’-OMe modification (including the orphan nucleotide), with natural bases, results in a non-functional oligonucleotide as far as RNA editing goes (known in the art), presumably because it hinders the ADAR activity at the targeted position. In general, an adenosine in a target RNA can be protected from editing by providing an opposing nucleotide with a 2'-OMe group (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing a guanine or adenine as opposing base, as these two nucleobases are also able to reduce editing of the opposing adenosine. Linkage modifications

[0160] 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 neighboringmonomers 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.

[0161] As outlined in detail herein, naked EONs as disclosed herein comprise at least one, preferably multiple linkage modifications. In some embodiments, the EON as disclosed herein comprises linkage modifications at most, and potentially all positions if the EON is capable of mediating RNA editing through the deamination enzyme when the EON is bound to the target RNA nucleic acid molecule. A linkage modification can be, but is not limited to, a modified version of the phosphodiester present in RNA, such as phosphorothioate (PS), chirally pure PS, (R)-PS, (S)-PS, methyl phosphonate (MP, also referred to as MeP), 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 EON as disclosed herein may also comprise one or more linkage modifications according to the structure of formula (II)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 C1-C6 alkoxy, a substituted C1-C6 alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a C1-C6 alkenyl, a C1-C6 substituted alkenyl, a C1-C6 alkynyl, a substituted C1-C6 alkynyl, or a conjugate group. In one embodiment, X = O and R = methyl and the linkage modification is referred to as mesyl phosphoramidate, MsPA or PNms.

[0162] In one embodiment, the EON as disclosed herein comprises an internucleoside linkage of the structure of formula (II), wherein X = O and R = CH3, which linkage is generally referred to herein as a PNms linkage (mesyl phosphoramidate). In other embodiments, R equals one of the following structures (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0163] Disclosed herein is also an EON that is able to mediate adenosine deamination by recruitment of a deaminating enzyme in a cell after the EON 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 EON comprises a moiety with a structure according to formula (III):wherein: X = O or S; Y = O- or S-; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a C1-C6 alkoxy, a substituted C1-C6 alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a C1-C6 alkenyl, a C1-C6 substituted alkenyl, aC1-C6alkynyl, a substituted C1-C6alkynyl, or a conjugate group. In one embodiment, X = O and R = methyl.

[0164] An EON 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 EON comprises a certain linkage modification at each linkage position of the EON as disclosed herein to generate the most effective and stable oligonucleotide compound.

[0165] 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 EONs as active compounds with both Rp and Sp configurations at a certain specified linkage position. Mixtures of such EONs 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 EON as disclosed herein comprises one or more (chirally pure or chirally mixed) PS linkages. In one aspect, the EON as disclosed herein comprises one ofmore (chirally pure or chirally mixed) phosphoramidate (PN) linkages. In one aspect, the EON 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 EON. EONs as disclosed herein may also comprise linkage modifications at all positions that are not chirally controlled. The EON 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 EON 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 EON as disclosed herein are modified internucleoside linkages. In one aspect, the EON as disclosed herein comprises at least one MP internucleoside linkage according to the structure of formula (IV):

[0166] As was noted in the art, a preferred position for an MP linkage in an EON is linkage position -2, thereby connecting the nucleoside at position -1 with the nucleoside at position -2. In one aspect, this position, in an EON as disclosed herein, comprises a linkage modification according to the structure of formula (II), instead of an MP linkage. Int. Patent Application Publication No. WO2020 / 201406 discloses the use of MP linkage modifications at certain positions surrounding the orphan nucleotide in the first nucleic acid strand. Although the presence of MP linkages is compatible with RNA editing by human ADAR enzymes, introducing MP linkages during the manufacturing of oligonucleotides is challenging in view of additional manufacturing (purification) steps in the coupling and decoupling process. In one aspect, the EON does not comprise an MP linkage.

[0167] In one aspect, the EON 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 EON. A PNdmi linkage as preferably used in the EONs as disclosed herein has the structure of formula (V)(V)

[0168] In one aspect, at either end or both termini of an EON as disclosed herein, inverted deoxyT or dideoxyT nucleotides are incorporated. Other internucleoside linkages that may be used in the EONs as disclosed herein are those that are disclosed in Int. Patent Application Publication No. WO2023 / 278589.

[0169] In one aspect, the EON as disclosed herein comprises at least one phosphonoacetate and / or at least one phosphonoacetamide internucleoside linkage. Conjugate chemistries

[0170] In one aspect, the EON 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 slowlycompared to a cleavable linker, for example, in a PS linkage, modified or unmodified deoxyribonucleosides linked by a PS linkage, a spacer connected through a PS bond and a linker consisting of modified or unmodified ribonucleosides. There is no restriction on the chain length, when a linker is a nucleic acid such as DNA, or an oligonucleotide. However, it 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 EON through one or more linkers, including peptides, sugars, vitamins, polymers, aptamers, (fragments of) antibodies, small molecules, and the like. General

[0171] In addition to the specific preferred chemical modifications at certain positions in compounds as disclosed herein, EONs 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 EON as disclosed herein comprises at least one internucleoside linkage according to the structure of formula (II), (III), (IV), and / or (V), and / or the EON further comprises at least one nucleotide with a sugar moiety that comprises a 2’-OMe modification, and / or the EON comprises at least one nucleotide with a sugar moiety that comprises a 2’-MOE modification, and / or the EON comprises at least one nucleotide with a sugar moiety that comprises a 2’-F modification, and / or the EON 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.

[0172] In one aspect, the EON 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 atarget adenosine in the target RNA molecule, wherein the nucleotide in the EON that is opposite the target adenosine is the orphan nucleotide, and wherein the orphan nucleotide has the structure of formula (VI):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(1H)-pyridone; R1 and R2 are both selected, independently, from H, OH, F or CH3; R3 is the part of the EON that is 5’ of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4 is the part of the EON 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).

[0173] Other chemical modifications of the EON 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., Int. Patent Application Publication Nos. WO2014 / 022566 or WO2015 / 011694. Again, in all cases, the modifications should be compatible with editing such that the EON 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.

[0174] EONs 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 EON as disclosed herein preferably does not comprise a boxB RNA hairpin sequence. In one aspect, an EON 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 iso-uridine opposite the target adenosine, which likely does not pair like G pairs with U. Preferably, the target adenosine in the targetsequence forms a mismatch base pair with the nucleoside in the EON that is directly opposite the target adenosine.

[0175] As outlined above, an EON 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 EON, 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 EON. 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 EON, 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 EON), more preferably a deoxyinosine.

[0176] The EON 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 double-stranded complexes (see for instance European Patent Application Publication No. 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 EONto the transcript molecule. In one embodiment, the EON does not comprise four or more consecutive DNA nucleotides anywhere within its sequence. In an embodiment, the EON 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 EON may be DNA, but also that there is no stretch of four or more consecutive DNA nucleotides within the EON. Hence, the EON 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 single-stranded nucleic acid consisting of a central region (DNA gap region with at least four consecutive deoxyribonucleotides) and wing regions positioned directly at the 5’ end (5’ wing region) and the 3’ end (3’ wing region) thereof. In contrast, the EON 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.

[0177] The EONs as disclosed herein may also be administered in the context of aids that will increase the entry of the EON 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 Int. Patent Application Publication No. WO2021 / 122998 and further described for use with RNA editing producing oligonucleotides in Int. Patent Application Publication No. WO2024 / 153801.

[0178] Disclosed herein is also a pharmaceutical composition comprising the EON 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 EON) and may be dissolved in a pharmaceutically acceptable organic solvent, or the like. Dosage forms in which the EON 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 beadministered 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.

[0179] In one embodiment, the EON 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 one embodiment, the EON 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 (II) 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 EON 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 EON and improved cellular delivery and trafficking.

[0180] When the EON comprises chemical modifications, as detailed herein, it may still be delivered through the means of a delivery vehicle. Suitable delivery vehicles are nanoparticle delivery vehicles such as polymeric nanoparticles, dendrimers, inorganic nanoparticles and nanocrystals, organic nanocrystals, and liposomes. In some embodiments, nanoparticles are Lipid Nanoparticles (LNP’s) that are nano-sized lipid vesicles that carry the EON of the present invention and aid to the delivery of target cells. In the event that an LNP is applied or any other similar type of carrier, the EON is still considered naked because it is not transcribed from an encoding polynucleotide (such as in the case of a plasmid or a vector, in which the EON is not regarded as ‘naked’). So, even though a chemically modified EON is encapsulated by a carrier, preferably an LNP, it is still seen as naked, as it has been manufactured as such in a laboratory setting andencapsulated thereafter in the carrier using methods known to the person skilled in the art. The disclosure also relates to a delivery vehicle, preferably an LNP, which comprises a ‘naked’ and chemically modified EON as disclosed herein. The person skilled in the art understands that when a delivery moiety, or attachment to the EON is used (such a GalNAc moiety to target hepatocytes in the liver) that the EON is still seen as naked as well, also when a GalNAc-EON is encapsulated in a delivery vehicle such as an LNP.

[0181] 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 hADAR1 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 skilled in the art will be capable of designing the complementary portion of the oligonucleotide according to their needs.

[0182] It is 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 EON 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 EON. This is something to be determined by the experimenter (in vitro) or the clinician, usually in phase I and / or II clinical trials.

[0183] 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 liver cells, more preferably hepatocytes. The target cell can be located in vitro, ex vivo or in vivo. One advantage of the EON as disclosed herein is thatit 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 EON 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, or liver spheroid. Organoids can be thought of as three-dimensional in vitro–derived tissues but are driven using specific conditions to generate individual, isolated tissues.

[0184] 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, where during transcription or splicing, or in the cytoplasm, for example mature mRNA, miRNA or ncRNA can be edited. Generally, 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.

[0185] The amount of EON to be administered, the dosage and the dosing regimen can vary from cell type to cell type, the disease to be treated, age, weight, gender, the target population, the mode of administration (e.g., systemic versus local), the severity of disease and the acceptable level of side activity, but these can and should be assessed by trial and error during in vitro research, in pre-clinical and clinical trials. The trials are particularly straightforward when the modified sequence leads to an easily detected phenotypic change, or a change in (the level of, or activity of) a specified biomarker. It is possible that higher doses of EONs could compete for binding to an ADAR enzyme within a cell, therebydepleting 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 EON and a given target.

[0186] One suitable trial technique involves delivering the EON 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 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.

[0187] 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 EON until enough target RNAs have been modified to provide a tangible benefit to the patient and / or to maintain the benefits over time.

[0188] EONs as disclosed herein are particularly suitable for therapeutic use, and so disclosed is also a pharmaceutical composition comprising an EON 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 EON 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 µg / kg to about 100 mg / kg, preferably from about 10 µg / kg to about 10 mg / kg, more preferably from about 100 µg / kg to about 1 mg / kg. As outlined above, the EONs of the present disclosure may also be delivered through a delivery vehicle such as an LNP. Amounts of LNP carrying EONs as disclosed herein can and will also be determinedin (pre-) clinical phases. Administration may be by injection or infusion, intracranially, intrathecally, intranasally, orally, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra-peritoneally, intrarectally, intra-cisterna magna, parenterally, and the like. Administration 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.

[0189] 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; assessing the concentration and / or presence / absence of a biomarker; or using a functional read-out. A functional assessment will generally be according to methods known to the skilled person. A suitable manner to identify the presence of an inosine after deamination of the target adenosine is of course dPCR or even sequencing, using methods that are well-known to the person skilled in the art.

[0190] 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, hADAR1 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 hADAR1 and hADAR2 in humans or human cells and cytidine deaminases. It is known that hADAR1 exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 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- γ). hADAR1 is also inducible by TNF-α. This provides an opportunity to develop combination therapy, whereby IFN-γ or TNF-α and EONs 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-γ or TNF-α levels in certain tissues of a patient, creating further opportunities to make editing more specific for diseased tissues. It will be understood by a person having ordinary skill in the art that the extent to which the editingentities 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.

[0191] An EON 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 EON 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 EON as disclosed herein, when complexed to ADAR, preferably brings about the deamination of a single target adenosine.

[0192] An EON 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 EON as disclosed herein is longer than 20 nucleotides. The EON as disclosed herein is preferably shorter than 100 nucleotides, still more preferably shorter than 60 nucleotides, still more preferably shorter than 50 nucleotides. In some embodiments, the EON 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 some embodiments, the EON 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 EON is 27, 28, 29, or 30 nucleotides in length.

[0193] As described above, in some embodiments the disclosure provides an EON for forming a double-stranded complex with a human SLC12A5 target RNA molecule in a human neuron cell, for instance in the brain. Thus, the therapeutic effect is preferably on a human neuronal cell in vivo. Of course, the methods may also be carried out in vitro or ex vivo.

[0194] The disclosure provides an EON herein, or pharmaceutical composition herein, for use in the treatment of disease. The disclosure also provides the use of an EON herein, or pharmaceutical composition herein, in the manufacture of a medicament for the treatment of disease. The disclosure also provides a method for treating a disease in a patient, comprising administering a therapeutically effective amount of an EON herein or a pharmaceutical composition herein. Preferably the disease is a disease caused by lowered GABAergic inhibition, due to an increased concentration of chloride in the neuronal cell,generally caused by a lowered activity of the key chloride extruder KCC2. The EON is administered therapeutically or prophylactically because both types of treatment could be beneficial.

[0195] The therapeutically effective amount of EON herein 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. It is possible that higher doses of EON could compete for binding to an ADAR within a cell, thereby depleting the amount of the entity, which is free to take part in RNA editing, but routine dosing trials will reveal any such effects for a given ASO and a given target.

[0196] An “effective amount” refers to an amount necessary (for periods of time and for the means of administration) to achieve the desired therapeutic result. An effective amount of an EON may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability to elicit a desired response in the individual. An effective amount is also one in which any toxic or detrimental effects are outweighed by the therapeutically beneficial effects. A “therapeutically effective amount” is an amount of a pharmaceutical composition that is effective for producing some desired therapeutic effect, commensurate with a reasonable benefit / risk ratio, when administered in accordance with a desired treatment regimen.

[0197] 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 herein may involve repeated delivery of an EON herein until enough target RNAs have been modified to provide a tangible benefit to the patient and / or to maintain the benefits over time.

[0198] All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. Database entries and electronic publications disclosed in the present disclosure are incorporated by reference in their entireties. The version of the database entry or electronic publication incorporated by reference in the present application is the most recent version of the database entry or electronic publication that was publicly available at the time the present application was filed. The database entries corresponding to gene or protein identifiers (e.g.,genes or proteins identified by an accession number or database identifier of a public database such as Genbank, Refseq, or Uniprot) disclosed in the present application are incorporated by reference in their entireties. The gene or protein-related incorporated information is not limited to the sequence data contained in the database entry. The information incorporated by reference includes the entire contents of the database entry in the most recent version of the database that was publicly available at the time the present application was filed. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and not intended to be limiting. EXAMPLES Example 1. Determining K+ transport activity of KCC2 mutant forms.

[0199] As outlined above, and as shown in Table 1, a total of 11 editing sites were envisioned to be targeted through RNA editing to provide alternative KCC2 proteins that would have a gain-of-function in their K+ transport role. Eight of these variants are introduced into the N-terminal peptide of SEQ ID NO:28, namely Y68C, T69A, N70D, N70S, Q73R, H78R, E79G, and E82G. Two of these variants are outside this region but are thought to influence K+ transport, namely Y446C and S932G. One variant, T906A, is a phosphorylation site involved in keeping the KCC2 dimer in an autoinhibitory state (when phosphorylated) or in an active state (when de-phosphorylated), see FIG.2.

[0200] Expression plasmids were generated for the ten variants Y68C, T69A, N70D, N70S, Q73R, H78R, E79G, E82G, Y446C and S932G carrying the encoding codons as provided in Table 1.

[0201] The expression plasmids were generated by inserting the human SLC12A5 open reading frame into a pcDNA3.1(+) plasmid and variants were generated using site-directed mutagenesis. Plasmids were transfected using standard transfection protocols into HEK293 cells, seeded at ~30,000 cells per well in standard 96-well poly-d-lysine coated plates. After 72 hrs normalized KCC2 activity (influx / sec) was determined, normalized against KCC2 expression levels measured via immunostaining, using FLIPRTMassay (Molecular Devices) which was conducted essentially per manufacturer’s protocol except that two additional reagents were added to reduce the effects of other potassium channels (NKCC1 and ATPase) and enhance the specificity of the signal produced by the testing channel.

[0202] The results of this functionality assay are shown in FIG. 3A and FIG. 3B and indicate that all ten tested variants have a significant higher influx rate in comparison to the unedited control, strongly indicating that the mutant variants all provide a gain-of-function of the KCC2 protein. Certain double mutants were also generated (not shown), but these did not provide an increased or additive effect over the single mutants.

[0203] In concert with the above, 48 hrs after transfection, cells were treated for 30 min with 1 μM WNK463, which is a WNK1 inhibitor that specifically prevents KCC2 phosphorylation at T906 and T1007 sites, or 25 μM VU0240551, a specific KCC2 inhibitor, and KCC2 activity was measured via FLIPRTMto determine the effect on the KCC2b variants.

[0204] The effects of using these inhibitors are depicted in FIG. 4. WNK463 has a significant effect on the unedited KCC2b protein, with a strong increase in normalized KCC2 influx activity. This suggests strongly that preventing phosphorylation at T906 and T1007 can increase KCC2 activity significantly. The KCC2 inhibitor VU0240551 decreases the KCC2 influx activity in a very strong manner, indicating that the increased K+ flux is due to the KCC2 specific mutant variants. All mutant variants that were tested displayed a higher K+ influx activity in comparison to unedited control, which activity was – in most cases – not influenced using WNK463, indicating that these mutant variants are providing a gain-of-function to KCC2 regardless of whether the protein is phosphorylated or not. It appeared that the effect was strongest when the N-terminal peptide mutant variants were used: N70S, H78R, T69A, N70D, Q73R, E82G, and Y68C. These mutant variants appear to be insensitive to phosphorylation changes (see the absent effect when using WNK463). Without wishing to be bound by theory, N70S was even more active after using WNK463, which may be an indication of a potential inhibition of phosphorylation of the generated serine at this position (S70). E79G is tested in a next experiment. None of the N- terminal peptide mutant variants appeared to influence total protein expression (data not shown). Example 2. Editing of a target adenosine in the codon for threonine at position 906 in the KCC2b isoform (T906).

[0205] Intl. Patent Application Publ. No. WO2024 / 206175 discusses the RNA editing of the adenosine in the ACC codon encoding threonine at position 1007 of the human KCC2b isoform (T1007). RNA editing of this target adenosine results in an alanine residue insteadof a threonine, giving rise to the T1007A variant. The T1007 position is one of the two phosphorylation sites involved in keeping KCC2 in an inactive state. The other site is located at position 906 in the human KCC2b isoform (T906). However, the sequence surrounding the target adenosine in the case of T906 is less favourable for RNA editing since there is a 5’-G directly adjacent to the target adenosine (5’-AGACGUU-3’, in which the target adenosine is underlined and the codon for threonine is in bold). It was tested to see whether the target adenosine at this position could nevertheless be edited using: i) a deoxyinosine (Id) opposite the 5’-G; ii) a deoxycytidine (Cd) opposite the 5’-G; or iii) a nucleotide analog that can induce a syn conformation of the 5’-G. This was based on the disclosure of Int. Patent Application Publ. No. WO2024 / 013361, which is incorporated herein in its entirety. For this, a 7-deaza-2’-deoxyadenosine (7-deaza Ad; abbreviated to “7Ad” in FIG. 5) nucleotide was applied. This nucleotide can induce the required syn conformation of the 5’-G. FIG.5 shows a set of EONs that were designed to test whether the target adenosine in the ACG codon for threonine at position 906 in human KCC2b could efficiently be edited.

[0206] Neurons cultivated from induced pluripotent human stem cells (herein further referred to as ‘iPSC neurons’) express useful levels of human KCC2, enabling one to address RNA editing and potentially downstream effects. Human iPSC cells were cultured on matrigel hESC-Qualified Matrix (LDEV-free) coated plates and maintained in MTESRTMPlus supplemented with MTESRTMPlus 5x supplement and 1% pen / strep. The human iPSC cells were differentiated to mature forebrain neurons using materials and protocols supplied by Stem Cell Technologies. In short, the human iPSC cells were differentiated to embryonic bodies (EBs) using the STEMDIFFTMSMADi neural induction kit. The EBs were plated and differentiated to neural rosettes, and the differentiated neural rosettes were harvested using the STEMDIFFTMneural rosette selection reagent. The harvested neural rosettes were differentiated into neural progenitor cells (NPCs) and were maintained and expanded using STEMDIFFTMneural progenitor medium. The NPCs were subsequently differentiated using the STEMDIFFTMForebrain Neuron Differentiation Kit. The differentiated NPCs were passaged to their final culture plate at a density of approximately 55,000-60,000 cells / cm2. The forebrain neurons were maintained in BRAINPHYSTMneuronal medium supplemented with 200 µL SM1 neuronal supplement,100 µL N2 Supplement-A, 200 nM ascorbic acid, 1 mM dibutyryl-cAMP, 20 ng / mL BDNF, 20 ng / mL GDNF and 1% penicilin / streptomycin.

[0207] In a first experiment, the 14 EONs displayed in FIG. 5 were used in a gymnotic exposure experiment (gymnotic uptake = without transfection means; only adding EON to the culture medium). Herein, 5 μM EON was administered to the culture medium and kept on the cells for 48 hrs, after which the EON was slowly washed out through a replacement of 50% of the culture medium with fresh medium every 2 days. Editing was determined at 14 days after start of the experiment. RNA was isolated using MIRVANATMRNA isolation kit (Thermo Fisher Scientific), the manufacturer protocol was followed with a few adaptations to remove the EONs from RNA samples. After disruption of the cells, the supplied miRNA homogenate additive was added at 1 / 10 of the total volume. After phenol:chloroform (same volume) extraction the upper aqueous phase was heated for 1 min at 60°C. 100% ethanol was added to the aqueous phase at 1 / 3 of the total volume. The samples were added to the supplied filter columns, after centrifugation, the flow through was discarded. The filters containing the mRNA were washed according to the protocol and eluted using 30 µl 95°C nuclease-free water. The purified RNA was subsequently treated with TURBO DNASETM(Thermo Fisher Scientific) according to the supplied protocol. Then, 500 ng of the RNA was used as a template for cDNA synthesis using the Maxima first strand cDNA synthesis kit (Thermo Fisher Scientific) with random hexamers and OligodT according to the manufacturer’s instructions. The total reaction volume was 20 µl.

[0208] As inosines pair with cytidines during the cDNA synthesis in the reverse transcription reaction, the nucleotides incorporated in the edited positions during PCR will be guanosines. The percentage of guanosine (edited) versus adenosine (unedited) was defined by ddPCR in exon 23 of the SLC12A5 transcript.

[0209] Each ddPCR sample contained 1x ddPCR supermix for probes (no dUTP) (from Biorad), 0.9 µM forward primer “hKCC_e21_Fw”: 5’-GAGCGACATCTCAGCTTAC-3’ (SEQ ID NO:23); 0.9 µM reverse primer “hKCC_e21-22_Rev”: 5’- TGTGATACTCTGGATCTCCC-3’ (SEQ ID NO:24); 0.6 µM of each double quenched WT unedited transcript probe “hKCC2_e21_A_HEX”: / 5HEX / ATGA +GAA +G+A+C GTT G+GT +GAT G / 3IABkFQ / (SEQ ID NO:25); and mutant edited transcript probe “hKCC2_e21_G_FAM”: / 56-FAM / +ATG A+GA A+G+G +CGT +TGG TGA / 3IABkFQ / (SEQ ID NO:26) with the + sign denoting an LNA at the 3’ side; and template cDNA in a total volume of 21 µL. Droplets were made from the PCR mixes using the QX200TMdroplet generator (Biorad). Next, the droplet PCR was performed in a T100 thermal cycler (Biorad) with a heated lid of 105°C and a ramp temperature of 2°C per sec. The polymerase was heat activated at 95°C for 10 min. Each cycle the denaturation was performed at 95°C for 30 sec and the annealing / extension was performed at 59°C for 60 sec. This was repeated for 40 cycles in total. The enzymes were deactivated at 98°C for 10 min and the reaction was held at 8°C. Fluorescent signal from the droplets was measured by the QX200TMdroplet reader (Biorad). Determining the number of fluorescent positive droplets and subsequent absolute quantification was performed with QUANTASOFTTMsoftware (Bio- Rad). As an additional quality control samples were removed which significantly deviated from the rest in terms of absolute copy numbers. This was determined by calculating the average number of total SLC12A5 (A + G transcripts) copies for each condition, samples were removed if they had a five-fold lower or higher number of SLC12A5 copies compared to the average.

[0210] Each sample was measured in duplicate with the average values used to determine the editing percentage. The editing percentage was calculated by dividing the number of edited (G) SLC12A5 transcript copies per well with the total number of SLC12A5 copies per well (A + G transcripts). Results are shown in FIG.6. In concert with the dis-favoured 5’-G in the target sequence, indeed using a deoxycytidine at the -1 position in the EON does not give a high editing percentage (EONs B906-6, -7, and -8). Using a deoxyinosine at the -1 position in the EON improves it somewhat (EONs B906-1 to -5). However, when the -1 position was modified to contain a 7-deaza-2’-deoxyadenosine (EONs B906-9 to - 14) the editing percentage increased significantly, with B906-10, -11, -13, and -14 performing best. This shows that, besides editing the T1007 site as described in Int. Patent Application Publication No. WO2024 / 206175, also the second phosphorylation site at position 906 can be amended through RNA editing. It is likely that when both sites are knocked out through RNA editing an even stronger (additive or even synergistic) effect on the KCC2 influx activity can be observed, which provides a very useful tool to treat neurological disorders, such as neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and / or epilepsy, as disclosed herein.Example 3. Editing of human SLC12A5 transcripts to yield mutant variants of the N- terminal peptide.

[0211] As explained above, the N-terminal peptide (as shown in SEQ ID NO:28) is thought to be involved in the autoinhibitory state of the KCC2 dimer when the sites at positions 1007 and 906 (in KCC2b) are phosphorylated (see FIG.2). As shown in the art (Xie et al. 2020), when certain positions within the N-terminal peptide were changed, the interaction with their counterpart residues in KCC2 itself in the cytosolic facing cavity was disrupted. This brought the realization with the inventors of the present disclosure to use RNA editing to provide transient expression of KCC2 variant proteins (“editing” variants) which display increased activity. As shown in Example 1 above, expression plasmids expressing the different “editing” variants indeed showed that KCC2 function in K+ influx activity was increased. This supports the idea that changing the SLC12A5 transcript molecules at the specified positions would lead to KCC2 editing variant types in vivo and provide a tool for the treatment of the neurological disorders that would benefit from a KCC2 protein with increased activity, such as in some neurodevelopment disorders, neuropsychiatric disorders, chronic pain disorders, and epilepsy.

[0212] The envisioned editing variants are listed in Table 1 for both KCC2a and KCC2b isoforms. For each of the 8 target sites EONs were designed to determine whether editing could be achieved, using endogenous (naturally present) human ADAR in human iPSC forebrain neurons. These EONs are provided in FIGS. 7A-7L. Initially, six EONs were designed for each of these targets, followed by an additional set of EONs for Q73R, H78R, E79G, and E82G. FIGS. 7A-7L list all EONs that were initially designed for the first experiment and subsequent experiments, based on the best performing EONs. Y68C-1 to Y68C-6 (SEQ ID NO:29 to 34, respectively) are directed to deamination of the adenosine in the UAC codon for tyrosine at position 68 in KCC2b. T69A-1 to T69A-6 (SEQ ID NO:35 to 40, respectively) are directed to deamination of the adenosine in the ACC codon for threonine at position 69 in KCC2b. N70D-1 to N70D-6 (SEQ ID NO:41 to 46, respectively) are directed to deamination of the first adenosine in the AAC codon for asparagine at position 70 in KCC2b. N70S-1 to N70S-6 (SEQ ID NO:47 to 52, respectively) are directed to deamination of the second adenosine in the AAC codon for asparagine at position 70 in KCC2b. Q73R-1 to Q73R-70 (also referred to as B96-1 to B96-70, respectively; SEQ ID NO:53 to 122) are directed to deamination of the adenosine in the CAG codon for glutamine at position 73 in KCC2b. H78R-1 to H78R-60 (also referred to as H101-1 to H101-60,respectively; SEQ ID NO:123 to 182, respectively) are directed to deamination of the adenosine in the CAU codon for histidine at position 78 in KCC2b. E79G-1 to E79G-18 (SEQ ID NO:183 to 200, respectively) are directed to deamination of the first adenosine in the GAA codon for glutamic acid at position 79 in KCC2b. E82G-1 to E82G-18 (SEQ ID NO:201 to 218, respectively) are directed to deamination of the first adenosine in the GAA codon for glutamic acid at position 82 in KCC2b.

[0213] In an initial experiment, the six EONs listed in FIGS.7A-7L for each of the 8 target sites (total of 48 EONs) were tested in an editing experiment. The 48 EONs were used in a gymnotic exposure experiment using human iPSC forebrain neurons, with 6.0x105cells per treatment. Herein, 5 μM EON was administered to the culture medium and kept on the cells for 48 hrs, after which the EON was slowly washed out through a replacement of 50% of the culture medium with fresh medium every 2 days. RNA was isolated using MIRVANATMRNA isolation kit (Thermo Fisher Scientific), the manufacturer protocol was followed with a few adaptations to remove the EONs from RNA samples.

[0214] After disruption of the cells, the supplied miRNA homogenate additive was added at 1 / 10 of the total volume. After phenol:chloroform (same volume) extraction the upper aqueous phase was heated for 1 min at 60°C.100% ethanol was added to the aqueous phase at 1 / 3 of the total volume. The samples were added to the supplied filter columns, after centrifugation, the flow through was discarded. The filters containing the mRNA were washed according to the protocol and eluted using 30 µl 95°C nuclease-free water. The purified RNA was subsequently treated with TURBO DNASETM(Thermo Fisher Scientific) according to the supplied protocol. RNA concentrations were measured using the Nanodrop and 500 ng RNA was used for cDNA synthesis. cDNA synthesis was performed using the Maxima reverse transcriptase kit (Thermo Fisher Scientific) according to the manufacturer’s instructions, with a combination of random hexamer and oligo-dT primers. Then, cDNA was subjected to RT-PCR using the Amplitaq Gold 360 DNA Polymerase 1000U kit (Applied Biosystems) according to the manufacturer’s instructions to amplify the target region. Then, PCR products were analysed by Sanger sequencing to detect inhibition of A-I editing. The following primers were used: “hKCC2_e02-03_fw”: 5’-CCT TGT TTG AGG AGG AGA TG-3’ (SEQ ID NO:219) and “hKCC2_e04_rev”: 5’- GAT GAC GCC AAA GAT GTT CT-3’ (SEQ ID NO:220).

[0215] The result are shown in FIG.8, which shows that efficient editing was observed for the adenosine in the CAG codon for glutamine at position 73 in KCC2b (Q73) and the adenosine in the CAU codon for histidine at position 78 in KCC2b (H78), with EON Q73R- 6 (SEQ ID NO:58) providing the highest editing efficiency of more than 30% in that position and with EON H78R-5 (SEQ ID NO:127) providing the highest editing efficiency of around 18% in that position. The background signals in the case of E82G were relatively high in this first experiment, which makes it difficult to determine the exact editing levels. The design of Q73R-6 was used to design EONs Q73R-7 to Q73R-70. The design of H78R- 5 was used to design H78R-7 to H78R-60.

[0216] Editing levels were also determined for the gymnotic treatment with Q73R-1 to Q73R-6 to bring about the Q73R change and with H78R-1 to H78R-6 to bring about the H78R change, by editing the respective target nucleic acids in these human iPSC forebrain neurons using a ddPCR setup. The protocol was generally as described in example 2, with the respective primers and probes as provided in Table 2.

[0217] Table 2. Primers and probes (all 5’ to 3’) for the ddPCR assays to determine RNA editing for the Q73R and H78R target sites, as shown (and following the notation for KCC2a and KCC2b provided in Table 1, in which Q96R = Q73R and H101R = H78R). The + symbol denotes an LNA at the 3’ side of the symbol.

[0218] FIG. 9 shows the results of these ddPCR determinations, indicating that indeed proper editing was achieved for both target adenosines, with Q73R-6 (SEQ ID NO:58) providing an editing of around 25% for the adenosine > inosine deamination leading to Q73R and with H78R-4 (SEQ ID NO:126) providing an editing of around 17% for the adenosine > inosine deamination leading to H78R. These results, for both Q73R as well asfor H78R are in line with what was observed after Sanger sequencing as provided in FIG. 8.

[0219] These results, also in view of what has been described in Int. Patent Application Publication No. WO2024 / 206175, show that the inventors were able to achieve high editing percentages on alternative sites within the human SLC12A5 transcript, especially in some positions in the N-terminal peptide that were found to be crucial for KCC2 functioning. Rather than changing the phosphorylation site at position 1007, the current disclosure provides alternative changes in amino acid position(s) that are not phosphorylated but that are key in the functionality of the K+-coupled transporter KCC2. Example 4. Editing of human SLC12A5 transcripts using a further set of EONs (Q96R).

[0220] Further to the screen of EONs in the previous example, two sets of EONs (first 70, followed by an additional 50) were designed to target the codon for Q73 in KCC2b (Q96 in KCC2a, see Table 1) to determine the optimal length and chemical composition. As in the previous example, the EONs were screened in human iPSC-derived forebrain neurons, using 5 μM EONs in a 2-week wash-out experiment, followed by RNA purification and ddPCR analysis as described above, using the Q73R primers and probes of Table 2. The 50 additional EONs targeting the Q96 (KCC2a) site are provided in FIGS. 10A-10C, with their respective sequence and chemical modifications, names and numbers.

[0221] The percentage editing results obtained with B96-1 to B96-70 (Q73R-1 to Q73-70, respectively; SEQ ID NO:53 to 122, respectively) are provided in Table 3 and show that the EONs that were designed with a variety of 2’-F modification patterns (B96-55 to B96- 65) performed best in general (around 40% editing, numbers provided in bold), with B96- 59 (SEQ ID NO:111), B96-60 (SEQ ID NO:112), B96-61 (SEQ ID NO:113), and B96-62 (SEQ ID NO:114) performing best. Also a few other EONs outperformed, such as B96-21 (SEQ ID NO:73), B96-32 (SEQ ID NO:84), B96-46 (SEQ ID NO:98), B96-63 (SEQ ID NO:115), and B96-68 (SEQ ID NO:120).

[0222] Table 3. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the Q96 (KCC2a) site. NT (non-treated) and B96-1, B96-2, B96-3, B96-4, B96-5, and B96-6 were performed in 6 different wells. The further treatments were in triplicate.

[0223] In a second experiment, using an identical experimental setup, EONs B96-71 to B96-120 (see FIGS. 10A-10C) were tested for their ability to edit human SLC12A5 transcripts in human iPSC-derived forebrain neurons. B96-71 to B96-83 were designed based on the chemical modifications and length, as well as the 5’ and 3’ arm from the position of the orphan nucleotide of B96-21 (= Q73R-21; RM121246; SEQ ID NO:73). B96-84 to B96-96 were designed based on B96-6 (=Q73R-6; RM119516; SEQ ID NO:58) and B96-62 (= Q73R-62; RM121287; SEQ ID NO:114). B96-97 to B96-108 were designed based on B96-32 (= Q73R-32; RM121257; SEQ ID NO:84), and B96-109 to B96-120 were designed based on B96-37 (= Q73R-37; RM121262; SEQ ID NO:89). The editing percentages following the use of these EONs are provided in Table 4. Further to the good results obtained with B96-1 to B96-70, the levels again reached 40% to 50% editing, with B96-111 (SEQ ID NO:266), B96-116 (SEQ ID NO:271), B96-79 (SEQ ID NO:234), B96- 112 (SEQ ID NO:267), B96-93 (SEQ ID NO:248), B96-81 (SEQ ID NO:236), B96-83 (SEQ ID NO:238), and B96-113 (SEQ ID NO:268) performing best, with percentages well above 40.

[0224] Table 4. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the Q96 (KCC2a) site. Treatments were in triplicate.Example 5. Editing of human SLC12A5 transcripts using a further set of EONs (H101R).

[0225] The same experiment as described in Example 4 was performed for the H78R in KCC2b (H101R in KCC2a) site using an initial set of 60 EONs (B101-1 to B101-60), followed by a follow-up set of 50 EONs (B101-61 to B101-110) that were based on the best performers from the first 60 EONs. The additional 50 EONs to generate the H101R change are provided in FIGS. 11A-11C. Like the previous example, the EONs were screened in human iPSC-derived forebrain neurons, using 5 μM EONs in a 2-week wash- out experiment, followed by RNA purification and ddPCR analysis as described above, using the H101R primers and probes of Table 2.

[0226] The percentage editing results obtained with B101-1 to B101-60 (H78R-1 to H78R- 60, respectively; SEQ ID NO:123 to 182, respectively) are provided in Table 5 and show that there was no clearly defined pattern that outperformed another design. However, B101- 57 (SEQ ID NO:179), B101-18 (SEQ ID NO:140), B101-15 (SEQ ID NO:137), B101-8 (SEQ ID NO:130), B101-9 (SEQ ID NO:131), B101-30 (SEQ ID NO:152), and B101-16 (SEQ ID NO:138) outperformed with editing levels of 17% to more than 21%.

[0227] Table 5. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the site H101 in KCC2a (H78R in KCC2b). NT is a non-treated sample. B101-1, B101-2, B101-3, B101-4, and B96-5 were performed in 6 different wells, but not all samples provided results (NA). The further treatments were in triplicate.

[0228] In a next experiment, using an identical experimental setup, EONs B101-61 to B101-110 (see FIGS. 11A-11C) were tested for their ability to edit human SLC12A5 transcripts in human iPSC-derived forebrain neurons. B101-68 to B101-80 were designed based on B101-57 (H78R-58; SEQ ID NO:179). B101-81 to B101-98 were designed based on B101-18 (H78R-18; SEQ ID NO:140). B101-99 to B101-104 were designed based on B101-8 (H78R-8; RM121297; SEQ ID NO:130). B101-105 to B101-110 were designed based on B101-9 (H78R-9; SEQ ID NO:131). The editing percentages following the use of these EONs are provided in Table 6. Further to the good results obtained with B101-1 to B101-60, editing levels now reached almost 30%, with B101-68 (SEQ ID NO:283), B101- 78 (SEQ ID NO:293), B101-77 (SEQ ID NO:292), B101-71 (SEQ ID NO:286), and B101- 79 (SEQ ID NO:294) performing best.

[0229] Table 6. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the H101 (KCC2a) site. Treatments were in triplicate. NA indicates samples from which no results were retrieved.Example 6. In vivo editing of mouse Slc12a5 transcripts targeting all eight target sites in the N-terminal peptide.

[0230] To test editing efficiency in vivo, mouse-specific EONs were designed based on data from EONs from Example 4 and 5. These EONs were designed to edit mouse Slc12a5 transcripts at the equivalent sites. The following EONs were designed (which are shown in detail with their chemical modifications in FIG.12):

[0231] Wildtype C57BL / 6J 10-19 weeks old mice were used in the study. These were group-housed under specific pathogen–free conditions. The room was temperature- controlled (23 ± 1°C) and humidity-controlled (55 ± 10%) and on a 12-hour light-dark cycle (lights on at 7:00). Food and water were provided ad libitum. EONs were formulated with PBS without calcium and magnesium (Gibco) at a concentration of ~40mg / ml. For each EON, 3-4 male or female animals (randomly assigned) received intracerebroventricular (ICV) injection with 11.34 nmol EON, and 2 animals were injected with 22.68 nmol EON (100 μg / 200 μg, equimolar). Animals were sacrificed 14 days after injection followed by tissue collection. Cortical, hippocampal, midbrain, cerebellar, cervical, thoracic and lumbar tissue was collected and frozen for further analysis.

[0232] RNA was isolated using MIRVANATMRNA isolation kit (Thermo Fisher Scientific), the manufacturer protocol was followed with a few adaptations to remove the EONs from RNA samples. After disruption of the cells, the supplied miRNA homogenate additive was added at 1 / 10th of the total volume. After phenol:chloroform (50 / 50) extraction the upper aqueous phase was heated for 1 min at 60°C.100% ethanol was added to the aqueous phase at 1 / 3rd of the total volume. The samples were added to the supplied filter columns, after centrifugation, the flow through was discarded. The filters containing mRNA were washed according to the protocol and eluted using 30 µl 95°C nuclease-free water. The purified RNA was subsequently treated with TURBO DNASETM(Thermo Fisher Scientific) according to the supplied protocol.

[0233] RNA concentrations were measured using Nanodrop and 750 ng RNA was used for cDNA synthesis. cDNA synthesis was performed using the Maxima reverse transcriptase kit (Thermo Fisher Scientific) according to the manufacturer’s instructions, with a combination of random hexamer and oligo-dT primers.

[0234] Primers and probes targeting the mouse Slc12a5 editing region, reference area and mouse housekeeping gene were designed and purchased from Integrated DNA Technologies. Probes were labelled with FAM (or an alternative fluorophore, such as HEX, TEX, Cy5), depending on the assay requirements. Final primer and probe concentrations were optimized to ensure efficient amplification (500 nM for primers and 250 nM for probes in the final reaction). Primer and probe sequences including their respective details are provided in Table 7.

[0235] Table 7. Primers and probes (all 5’ to 3’) for the ddPCR assays to determine RNA editing for the Y91C, T92A, N93D (A), N93S (B), Q96R, H101R, E102G, and E105G sites, as given above. The number in the primer / probe name relates to the respective target sites and following the notation for KCC2a. The + symbol denotes an LNA at the 3’ side of the symbol. mHMBS is a reference gene.

[0236] Each digital PCR reaction was assembled in a total volume of 12 µL per well, including: 3.0 µL of 4x QIAcuity Probe PCR Master Mix (Qiagen), 1.2 µL of 10x primer / probe assay mix (containing forward primer, reverse primer, and probe), 1.2 μL of cDNA and nuclease-free water to bring the final volume to 12 µL. After mixing, the 12 µL reaction mixtures were transferred into the corresponding wells of a QIAcuity 8.5K Nanoplate, 96-well format (Qiagen), designed to partition each well into ~8,500 micro- chambers via microfluidic channels. The plate was sealed with a QIAcuity Nanoplate Seal in accordance with the manufacturer’s guidelines. The loaded and sealed nanoplate was placed into the QIAcuity System (Qiagen), followed by amplification and data collection using dPCR. Digital PCR data were processed using QIAcuity Software Suite (Qiagen), which sets thresholds to distinguish positive and negative partitions. Target concentration was calculated based on Poisson statistics, allowing absolute quantification.

[0237] The results for the administrations with 100 μg EON are provided in FIG. 13A, FIG. 13B, FIG. 13C, FIG. 13D, FIG. 13E, FIG. 13F, FIG. 13G, FIG. 13H, and FIG. 13J. Data shows mouse Slc12a5 mRNA editing with EONs targeting mouse editing sites. mB91-4 targets the Y91 site. mB92-4 targets the T92 site. mB93A-4 targets the N93 site to change the codon to encode aspartic acid (D). mB93B-4 targets the N93 site to change the codon to encode serine (S). mB96-6 targets the Q96 site (KCC2a), which is the Q73 site in KCC2b. mB101-4 targets the H101 site. mB102-4 targets the E102 site. mB105-4 targets the E105 site.

[0238] FIG. 13A shows the editing results with the mB91-4 EON in the different mouse tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing slightly above 1% was observed in the cerebellum. FIG.13B shows the editing results with the mB92-4 EON in the different mouse tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was not significantly above background in any of the tissues. FIG. 13C shows the editing results with the mB93A-4 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was not significantly above background in any of the tissues. FIG. 13D shows the editing results with the mB93B-4 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was relatively low in all tissues, with a strong variation observed in the hippocampus. FIG. 13E shows the editing results with the mB96-6 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was significantly above background, reaching levels of approximately 2% in all tissues. FIG. 13F also shows editing results with the mB96-29 EON, which is a more stabilized version of mB96-6, detected in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was significantly above background, reaching levels above 10%, especially in the spinal cord sections. FIG.13G shows the editing results with the mB101-4 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Modest editing was observed in midbrain and spinal cord sections, reaching levels between 2 and 3%. FIG. 13H shows the editing results with the mB102-4 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical,and SC thoracic. Editing was very modest and not significantly above background in any of the tissues. FIG.13J shows the editing results with the mB105-4 EON in the different mouse brain tissues: cortex, hippocampus, midbrain, cerebellum, SC lumbar, SC cervical, and SC thoracic. Editing was very modest and not significantly above background in any of the tissues.

[0239] The editing results obtained after administration of 200 μg (applying mB91-4, mB92-4, m93B-4, mB96-6, and mB102-4) in the different mouse neuronal tissues are provided in FIG. 14A, FIG. 14B, FIG. 14C, FIG. 14D, and FIG. 14E. An increase in editing – in comparison to the results with the 100 μg administrations – was observed in the case of mB92-4 (FIG.14B), mB93B-4 (FIG.14C), and mB96-6 (FIG.14D), with the latter providing an almost 2-fold increase. mB96-29 was not tested in the 200 μg administration setup.

[0240] The in vivo experiments demonstrate that ICV administration achieved editing of the mouse Slc12a5 transcript, albeit that some sites were edited more efficiently than others and some difference in editing exists between tissues. The highest editing was obtained with mB96-29 (SEQ ID NO:327), which is derived from mB96-6 (SEQ ID NO:326) with average editing levels reaching 11% in the spinal cord. Example 7. Editing of human SLC12A5 transcripts using Z and E nucleobases at the orphan position.

[0241] The Benner’s base (Z) is a cytosine analog depicted in FIG. 15, that also shows another cytosine analog: the 5-aza-5,6-dihydro cytosine nucleobase, which is herein and elsewhere referred to as the ‘E base.’ Throughout this disclosure it is referred to as ‘E’ (see FIG. 15). The effect on RNA editing of the E base installed at the orphan position was studied in this example. To test the difference between EONs comprising a Z base and an E base at the orphan position, EONs were manufactured that: i) target the target adenosine for the codon at position 96 in KCC2a isoform; and ii) target the target adenosine for the codon at position 101, in KCC2a isoform (see Table 1 for the corresponding positions in KCC2b). The target sequence of the human SLC12A5 transcript is as provided in FIG.1B.

[0242] In each of the two experiments targeting the different adenosines, four different EONs were used in a gymnotic exposure using human iPSC forebrain neurons, with 6.0x105cells per treatment. Two different iPSC forebrain neuron batches were used: cellbatch A and cell batch B. The sequences and specific chemical modifications of the EONs tested in this experiments are shown in FIG.16.

[0243] EON1 (= B96-111; RM122584; SEQ ID NO:266; AON1 in FIG.17A) targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a and comprises a deoxynucleotide at the orphan position with a Z nucleobase (Zd). EON2 (= B96-155; RM123434; SEQ ID NO:335; AON2 in FIG. 17A) is identical to EON1 and targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a but instead comprises a deoxynucleotide at the orphan position with an E nucleobase (Ed). EON3 (RM123359; SEQ ID NO:336; AON3 in FIG. 17A) is identical to EON1 and targets the adenosine in the CAG codon for glutamine at position 96 in KCC2a but is an on-target negative control since it comprises a 2’-OMe modified nucleotide at the orphan position with an uracil nucleobase (Um) and should be inactive for RNA editing. EON4 (RM123361; SEQ ID NO:337; AON4 in FIG.17A) comprises a Zd nucleotide at the same position as EON1 but is a further scrambled control of EON1, also serving as a negative control.

[0244] EON5 (= B101-68; RM122501; SEQ ID NO:283; AON5 in FIG.17B) targets the adenosine in the CAU codon for histidine at position 101 in KCC2a and comprises a deoxynucleotide at the orphan position with a Z nucleobase (Zd). EON6 (= B101-148; RM123435; SEQ ID NO:338; AON6 in FIG. 17B) is identical to EON5 and targets the adenosine in the CAU codon for histidine at position 101 in KCC2a but instead comprises a deoxynucleotide at the orphan position with an E nucleobase (Ed). EON7 (RM123360; SEQ ID NO:339; AON7 in FIG. 17B) is identical to EON5 and targets the adenosine in the CAU codon for histidine at position 101 in KCC2a but is an on-target negative control since it comprises a 2’-OMe modified nucleotide at the orphan position with an uracil nucleobase (Um) and should be inactive for RNA editing. EON8 (RM123362; SEQ ID NO:340; AON8 in FIG.17B) comprises a Zd nucleotide at the same position as EON5 but is a further scrambled control of EON5 also serving as a negative control.

[0245] After plating the human iPSC forebrain neurons, 5 μM EON was administered to the culture medium and kept on the cells for 48 hrs, after which the EON was slowly washed out through a replacement of 50% of the culture medium with fresh medium every 2 days. RNA isolation, cDNA synthesis, RT-PCR and ddPCR was performed as described above, whereas primers and probes for each of the specific sites were as shown in Table 2.

[0246] FIG. 17A shows the editing percentages after gymnotic uptake of the EONs targeting the adenosine in the CAG (Q96) codon and FIG. 17B shows the editing percentages after gymnotic uptake of the EONs targeting the adenosine in the CAU (H101) codon. The results showed that there was a very low batch-to-batch variability in the different assays: the first two bars in both figures represent the results obtained with the same EON for each site, but in different forebrain neuron batches (cell batch A and B, respectively). The EON comprising an E base at the orphan position for targeting the adenosine in the CAG codon for Q96 (EON2) and that was only tested in cell batch B outperformed the EON comprising a Z base at the orphan position (EON1) by 1.2-fold (~61.5% editing vs ~50% editing). Similarly, the EON comprising an E base at the orphan position for targeting the adenosine in the CAU codon for H101 (EON6) and that was also only tested in cell batch B outperformed the EON comprising a Z base at the orphan position (EON5) by 1.2-fold as well (~36% editing vs ~29% editing). EON3 and EON7 were identical to EON1 and EON5, respectively, except that the orphan nucleotide was a 2’-OMe modified uridine nucleotide (Um), which should suppress efficient editing. Very low editing yields were indeed observed after using these EONs. EON4 and EON8 were scrambled versions of EON1 and EON5, respectively, and served as negative controls. Non-treated samples (NT) also served as negative controls. Table 8 shows the editing percentages depicted in bar diagrams in FIG.17A and 17B.

[0247] Table 8. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the Q96 and H101 (KCC2a) sites. Treatments were in triplicate. NA indicates a sample from which no results were retrieved. NT are the non-treated samples.

[0248] These results show that incorporation of the E base as a nucleobase in the nucleotide at the orphan position, which is directly opposite the target adenosine improves the RNA editing efficiency of a target adenosine in the human SLC12A5 transcript, in comparison to the Z base at this position. Example 8. Editing of human SLC12A5 transcripts using further designed EONs.

[0249] The same gymnotic experiment as outlined in Example 7 was performed using the EONs shown in FIGS.18A-18C (for the adenosine in the CAG codon for Q96 in KCC2a) and in FIGS. 19A-19C (for the adenosine in the CAU codon for H101 in KCC2a). The EONs shown in FIGS. 18A-18C were designed based on the results obtained with B96- 111, B96-112, and B96-116, see Table 4 above. The EONs shown in FIGS.19A-19C were designed based on the results obtained with B101-68, B101-71, B101-77, B101-78, and B101-79, see Table 6 above. The new set of EONs comprise a variation in 2’-F modifications, positions of the PNdmi linkages, and the introduction of inosine wobble base pairing between EON and target sequence at different positions.

[0250] RNA isolation, cDNA synthesis, RT-PCR and ddPCR was performed as described above, whereas primers and probes for each of the specific sites were as shown in Table 2.

[0251] The ddPCR results for the targeting of the adenosine in the CAG codon for glutamine at position 96 are shown in Table 9. The best results were obtained with B96- 130 (SEQ ID NO:377), B96-132 (SEQ ID NO:379), B96-148 (SEQ ID NO:395), and B96- 153 (SEQ ID NO:400), in which particularly the results with B96-148 (SEQ ID NO:395) were striking, reaching an average of almost 70% editing in a cell system without the aid of delivery agents, or transfection. It is though that the most important difference between B96-148 and the other EONs tested in this experiment is the presence of a 2’-OMe modified nucleotide comprising a hypoxanthine nucleobase at nucleotide position -2 (Im), which forms a I-C wobble base pair with cytidine in the target sequence. Without wishing to be bound by theory, it is held that the four consecutive cytidines in the target sequence preceding the adenosine in the CAG codon for glutamine at position 96 (which in principle may be targeted with an EON comprising four consecutive guanosines) is unfavourable because the four consecutive guanosines in the EON may form different kinds of secondary structures within the EON itself or with other (target) sequences, thereby lowering the editing efficiency. If this is the case, that problem is best solved by introducing two inosine nucleotides at positions -1 and -2 in the EONs targeting the adenosine in the CAG codonfor glutamine at position 96 in KCC2a (Id at position -1 and Im at position -2), thereby leaving only two consecutive guanosine residues at positions -3 and -4. At least this shows that when the adenosine in the CAG codon for glutamine at position 96 in the KCC2a isoform is targeted, it is preferred to have an inosine nucleotide at position -2 in the EON. It is even more preferred to have, in addition, an inosine nucleotide at position -1 in the EON. Combined with the data observed in Example 7, these inosine nucleotides are combined with an orphan nucleotide which is preferable a deoxynucleotide comprising a E base as nucleobase.

[0252] Table 9. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the Q96 (KCC2a) site. Treatments were in triplicate. NT indicates the non-treated negative control.

[0253] The ddPCR results for the targeting of the adenosine in the CAU codon for histidine at position 101 are shown in Table 10. The best results were obtained with B101-127 (SEQ ID NO:418) reaching an average editing above 31%. Unlike what was observed with the EONs targeting the adenosine in the CAG codon for glutamine at position 96, the introduction of a nucleotide comprising a hypoxanthine nucleobase (at position -5, see B101-147; SEQ ID NO:438) did not result in a significant increase in RNA editing.

[0254] Table 10. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the H101 (KCC2a) site. Treatments were in triplicate. NT indicates the non-treated negative control.Example 9. Intrathecal delivery of an editing nucleotide.

[0255] This example describes a non-human primate (NHP) study delivering an editing guide oligonucleotide via intrathecal (IT) route in cynomolgus monkeys. The animals were administered with an editing guide oligonucleotide (“test article”) according to the study design shown in Table 11. In this example the test article is a guide oligonucleotide according to SEQ ID NO:469, targeting Beta-actin transcripts. The study shows editing efficiency and exposure after IT delivery. SEQ ID NO:469 represents: 5’-Gm!Am*Am*Am*Gm*Cf*Am*Af*Um*Gf*m5Ce*Zd*Ad^Um*Cf*Ae *Cf*Cm*Uf*Cm*Cf*Cm*Cm*Um!Gm-3’ wherein the modifications are as provided in FIG. 5, FIGS. 7A-7L, FIGS. 10A-10C, FIGS.11A-11C, FIG.12, FIG.16, FIGS.18A-18C and FIGS.19A-19C.Methods

[0256] 1. Animals. Twenty drug-naïve female cynomolgus monkeys had an IT catheter implanted in the lumbar region to facilitate test article administration.

[0257] 2. Reagents. Test article consisted of a single-stranded editing oligonucleotide according to SEQ ID NO:469.

[0258] 3. In-Life Study Design. Dose groups and terminal necropsy time points are depicted in Table 11. Test article was administered via IT catheter at total doses, volumes, and concentrations depicted in Table 11. Assessment of toxicity was based on mortality, clinical observations, body weights, qualitative food consumption, neurobehavioral observations, and clinical and anatomic pathology.

[0259] 4. Tissue Collection. Serial plasma and cerebrospinal fluid (CSF) collection was performed on dosing days for up to 360 hours post-dose. Terminal tissue collection was performed on the necropsy day indicated in Table 11. Tissues (when present) from each animal were preserved, as indicated in Table 12.

[0260] 4.1 Histology. All tissues denoted by “E” in Table 12 (Necropsy, Organ Weights, and Macroscopic Observations section) from each monkey were incubated in 10% neutral buffered formalin (NBF) for at least 24 hrs, but no more than 28 hrs, at RT. Upon removal from NBF, tissues were rinsed three times with 1x PBS and stored in PBS at 4ºC for up to 72 hrs until processed to paraffin block. Once processed to paraffin block, tissues were sectioned at a nominal 5 micron, and slides were prepared and stained with haematoxylin and eosin. For spinal cord (mid-thoracic and mid-lumbar), one transverse and one transverse oblique section (including the injection site) were taken. For spinal cord (mid cervical) one transverse and one parasagittal longitudinal section was taken.

[0261] 4.2 Microscopic Observations. All tissues denoted by “E” in Table 12 (Necropsy, Organ Weights, and Macroscopic Observations section) from all animals were examined microscopically. Following completion of the primary microscopic evaluation, an independent peer review evaluation was performed.

[0262] 4.3 Frozen Tissue Collection for Exploratory Analysis. After the collection of any of the following tissue samples for microscopic evaluation, the following samples were collected from each monkey:

[0263] Brain, left frontal cortex; Brain left, temporal cortex; Brain, left caudate; Brain, left putamen; Brain, left hippocampus; Brain, left thalamus; Brain, left medulla; Brain left,pons; Brain left, substantia nigra; Brain left, amygdala; Brain left, parietal cortex; Brain left, cerebellum; Dorsal Root Ganglion, lumbar (to target L5); Dorsal Root Ganglion, thoracic; Dorsal Root Ganglion, cervical; Heart, apex; Kidney, left cortex; Liver, left lateral lobe; Sciatic, left; Spinal cord, cervical; Spinal cord, thoracic; Spinal cord, lumbar; Spleen; Superior cervical ganglion, left.

[0264] For the left brain hemisphere for each brain region (frontal cortex, temporal cortex, caudate, putamen, hippocampus, thalamus, medulla, pons, substantia nigra, amygdala, parietal cortex, and cerebellum), up to four samples (25 to 100 mg each) were collected and placed in a pre-chilled, 2 mL, RNase-free tube; snap frozen in liquid nitrogen; and placed on dry ice until transferred to a freezer, set to maintain 60 to 80°C. All sample weights were recorded.

[0265] For the spinal cord region (cervical, thoracic, and lumbar), up to four samples (25 to 100 mg each) were collected and placed in a pre-chilled, 2 mL, RNase-free tube; snap frozen in liquid nitrogen; and placed on dry ice until transferred to a freezer, set to maintain -60 to 80°C. All sample weights were recorded.

[0266] For the dorsal root ganglions (DRG), four DRG (two bilateral pairs) from each region (cervical, thoracic, and lumbar) were collected, weighed, and had nerve roots trimmed out. Each bilateral pair was placed in a pre-chilled, 2 mL, RNase-free tube. The left superior cervical ganglion was sampled whole, weighed, and placed into a pre-chilled, 2 mL, RNase-free tube. Each sample was snap frozen in liquid nitrogen and placed on dry ice until transferred to a freezer, set to maintain -60 to 80°C. All sample weights were recorded.

[0267] For the liver (left lateral lobe), spleen, heart (apex), left kidney (cortex), and left sciatic nerve sample, up to four samples (25 to 100 mg each) of each tissue were collected and placed in a pre-chilled, 2 mL, RNase-free tube; snap frozen in liquid nitrogen; and placed on dry ice until transferred to a freezer, set to maintain -60 to 80°C. All sample weights were recorded.

[0268] 1. RNA Isolation. RNA was extracted from exploratory tissue lysates for exploratory pharmacodynamic endpoints (e.g. RNA editing efficiency). Frozen tissue chunks were transferred to 1 ml of cold TRIZOLTM(Life Technologies) and immediately homogenized using a Qiagen TissueRuptor or OMNI Soft Tissue Disruptor on ice (high speed for 10-30 sec, depending on the tissue). Homogenates equivalent to up to 35 mg oftissue weight (25 mg for liver, kidney, and spleen) were diluted with TRIZOLTMto a volume of 1 ml at RT. RNA was extracted by adding 200 µl of chloroform and shaking vigorously. After centrifugation, 450-500 µl of the upper aqueous phase was mixed with an equal volume of 70% ethanol. The mixture was subjected to RNA isolation using the RNeasy Mini Kit (Qiagen) following the manufacturer’s instructions. DNase on-column treatment was applied. RNA was eluted with 30 µl of RNase-free water. RNA concentration and quality were determined using an Agilent Bioanalyzer 4200 with RNA ScreenTape (Agilent), and RNA samples were stored at -80°C.

[0269] 2. cDNA Synthesis. A 0.5 µg RNA sample was mixed with 0.5 µl of 100 mM Oligo(dT)18 (Thermo Fisher Scientific), 0.5 µl of 100 µM Random Hexamer (Thermo Fisher Scientific), 1.0 µl of dNTP mix (Thermo Fisher Scientific), and RNase- and DNase- free water to make a final volume of 15.5 µl. The solution was incubated at 65°C for 5 min and cooled at 4°C for 1 min. After mixing with 4 µl of 5x RT buffer and 0.5 µl of Maxima Reverse Transcriptase (Thermo Fisher Scientific), the reaction solution was placed in a thermocycler and run under the following conditions: 25°C for 10 min, 60°C for 30 min, 85°C for 5 min, and 4°C indefinitely. After the reaction, cDNA samples were diluted with RNase- and DNase-free water at a 1:80 ratio.

[0270] 3. Digital droplet PCR.1 µl of cDNA sample was used to perform digital droplet PCR (ddPCR) with ddPCR Multiplex Supermix (4x) (BioRad) in a reaction volume of 20 µl. ddPCR probes and primers were custom-made by IDT.

[0271] The sequences of probes are as follows: / 56-FAM / AG GTG A+T+G +GCA TTG CTT TCG T / 3IABkFQ / (SEQ ID NO:470) and / 5HEX / AG +GTG A+T+A +GCA TTG CTT TCG TGT / 3IABkFQ / (SEQ ID NO:471).

[0272] The sequences of primers are as follows: 5’-AGT CCT CTC CCG AGT CCA CA-3’ (SEQ ID NO:472) and 5’-GGG GCA TGA AGG CTC ATT ATT CAA-3’ (SEQ ID NO:473).

[0273] A final concentration of 250 nM (probes) was used. A final concentration of 500 nM (primers) was used. Droplets were generated on a BioRad Automated Droplet Generator. PCR was performed in a C1000 Touch™ Thermal Cycler (BioRad) using the following conditions: 95°C for 10 min for 1 cycle; 95°C for 30 sec and 64°C for 1 min for40 cycles; 98°C for 10 min for 1 cycle; and 4°C indefinitely. After the reaction was completed, droplets were analysed on a QX600TMDroplet Reader (BioRad). Data was acquired and analysed using BioRad QX manager Standard Edition v.2.20 software. FAM- positive, HEX-positive, and FAM / HEX-positive populations were gated in the 2D plot of channel 1 and channel 2 amplitude. Editing efficiency percentage was calculated using the equation: G(cp / µl) / (G(cp / µl) + A(cp / µl)) x 100%.

[0274] Table 11. NHP study design of different doses, timepoints and repeat dosing. Group 1 did not receive the test article, but was treated in line with the other animals, and served as negative control.

[0275] Table 12. Organ / tissue collection.Results

[0276] The editing efficiencies measured across various tissues are provided in Table 13.

[0277] Table 13. Editing efficiency measured across tissues.Exposure analyses. Tissue Materials, Methods, and Data from HPLC / HRMS Analysis.

[0278] To analyse test article delivery to tissues and understand exposure response relationships, tissue samples were collected at timepoints corresponding to the editing measurements, and the exposure of the total guide oligonucleotide in tissue was quantified by HP-LC / MS. Briefly, tissue standards were prepared in control tissue homogenate. To control assay variability, an internal standard was added to all standards and samples.

[0279] Tissue samples were homogenized in cell lysis buffer. For total guide oligonucleotide measurements, tissue standards and samples were digested with proteinase K prior to being loaded onto an Oasis Wax micro-elution solid phase extraction (SPE) plate (Waters Inc, Milford, MA) for isolation. The SPE plate was washed with wash buffers and then analytes were eluted with elution buffer. Eluants from the SPE plates were dried, reconstituted, and injected onto an LC / MS system.

[0280] The total guide oligonucleotide concentrations were measured using a Orbitrap Exploris 240 (Thermo Fisher Scientific) mass spectrometer using the guide oligonucleotide peak for quantification. The mass spectrometer was operated in negative ion detection mode. All data were processed using XCALIBURTMversion 4.4 (Thermo Fisher Scientific).

[0281] Table 14 shows that animals dosed intrathecally with test article had measurable exposure in multiple central nervous system regions. The variability in brain exposure in some groups is likely indicative of incomplete IT dosing. Peripheral tissues (liver and kidney) were also exposed to test article, indicating a significant amount of IT dosed compound reaching the systemic circulation. Kidney concentrations were consistently higher than liver.

[0282] Table 14. Summary of exposure data.

[0283] The IT study demonstrates that direct administration into the brain is a feasible way of getting an RNA editing ASO at the cell of interest, especially brain cells. Example 10: In vitro stability of EONs in mouse liver homogenates.

[0284] EON stability was tested in an in vitro stability assay using mouse liver homogenates. In this assay, 2 μL of 1 mg / mL EON was added to a deep well plate containing 265 μL of mouse liver homogenate solution containing 100 mM Tris-HCL buffer (pH 6) with 1 mM MgCl2 and premium mouse liver homogenate diluted in PBS in a 1:3 (w:v) ratio to a final protein concentration of 0.25 mg / mL. At baseline and 12 hrs (T0 and T12, respectively), 20 μL of the liver homogenate solution was added to 140 μL clarity lysis buffer. From the homogenate in clarity lysis buffer, percent of EON remaining was measured following solid phase extraction using LC / MS. Intrinsic clearance was derived from the concentrations at T0 and T12. The results for in vitro stability are presented in Table 15.

[0285] Table 15. Stability of EONs in mouse liver homogenate using a 12-hour timepoint.Example 11: In vitro KCC2 activity using fluorescent imaging plate reader (FLIPR)

[0286] Changes in functional activity due to editing of the KCC2 mRNA were measured using a FLIPRTMPotassium Assay Kit (Molecular Devices). KCC2 is capable of transporting thallium (Tl+), thereby regulating its intracellular concentrations, in a manner analogous to potassium. Using a highly sensitive Tl+dye, the activity of KCC2 was determined by the flux of Tl+from cells expressing KCC2.

[0287] HEK293 cells that stably overexpress human KCC2 were seeded on a poly-D-lysine coated clear bottom plate and cultured overnight. Cells were transfected with EON (final concentration 200nM) using FUGENE®SI (Promega) and tested for functional activity 72 hrs later. Cells were treated using the FLIPRTMPotassium Assay kit with the addition of 2.5 mM probenecid, 10 μM of bumetanide, and 20 μM of oubain in the dye loading buffer. Flux rate was measured over the first minute following the addition of Tl+, compared to baseline. KCC2 activity, measured in triplicate, was normalized against KCC2 expression. KCC2 expression was measured via fluorescent intensity of cells stained for KCC2 protein using a KCC2 primary antibody (clone N1 / 12 IgG2b, K) (Biolegend) and secondary antibody Alexa Fluor 555 conjugated goat anti mouse IgG(H+L) (Thermo Fisher Scientific). Percent of editing and fold-increase in KCC2 activity is presented in Table 16. B96-6 (SEQ ID NO:58) and B96-83 (SEQ ID NO:238) scored particularly good, with an almost 2-fold increase in KCC2 activity.

[0288] Table 16. In vitro KCC2 editing efficiency and increased activity.Example 12. Editing of human SLC12A5 transcripts using further designed EONs.

[0289] Based on the results with the position 96 and position 101 targeting EONs, provided in example 7 and 8, two new sets of EONs were designed, depicted in FIGS. 20A-20C. B96-155 to B96-167 represent EONs targeting the adenosine in the CAG codon for glutamine at position 96. B96-155 is identical to RM123434 (EON2, SEQ ID NO:335, see above). B101-148 to B101-153 represent EONs targeting the adenosine in the CAU codon for histidine at position 101. B101-148 is identical to RM123435 (EON6; SEQ ID NO:338). Editing percentages in human iPSC-derived (hiPSC) forebrain neurons were investigated as outlined above, using gymnotic uptake and 14 days washout protocol. For the Q96 site, a first experiment was performed with B96-155 to B96-167, whereas a second separate experiment (but identical in setup) was performed with B96-168 to B96-180). Percentages of editing are presented in Table 17 (Q96) and Table 18 (H101).

[0290] Table 17. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the Q96 (KCC2a) site. Treatments were in triplicate, with averages (avg) given in the last column.

[0291] Table 18. Editing percentages in iPSC-derived forebrain neurons after treatment with 5 μM EON targeting the H101 (KCC2a) site. Treatments were in triplicate, with averages (avg) given in the last column.

[0292] As can be seen in the obtained editing percentages on the Q96 site in Table 17, it appears that an editing plateau (maximum) was reached using these EONs under these gymnotic uptake circumstances in the human iPSC derived forebrain neurons, with editing levels of approximately 60-71%, with B96-179 (RM129114, SEQ ID NO:462, formula X) reaching more than 71% editing on average. Most EONs performed equally good. Even though no distinction could be made based on certain chemical modifications or specifics, the level if editing, without using transfection aids, is remarkably high in these in vitro cell assays, applying an endogenous KCC2 target transcript and endogenous ADAR enzymes.

[0293] Editing levels in respect of the H101 site were not as high as observed with the Q96 site, as can be seen from the data presented in Table 18. Nevertheless, also here editing levels, using gymnotic uptake, were significantly high, reaching levels of almost 36% in the case of B101-148, which carries an E base at the orphan position. The six EONs performed equally good. Example 13. In vivo editing of mouse Slc12a5 transcripts in a pharmacodynamic study.

[0294] Further to the in vivo editing experiments shown in Example 6 above, it was investigated whether editing could be increased in vivo based on the acquired knowledge with the newly designed EONs that performed particularly well as shown in the in vitro experiments of Examples 7, 8, and 12. For this a set of mouse-specific EONs were designed, that are shown in FIGS. 21A-21B. The twelve mB96 EONs differ from the human SLC12A5 targeting B96 EONs at positions -5, +10, and +22 due to a different target sequence in mice. The six mB101 EONs differ from the human SLC12A5 targeting B101 EONs at position +7 for the same reason. A mouse App targeting EON (provided in SEQ ID NO:496) was used as a control in these experiments.

[0295] Nineteen groups of C57BL / 6 mice (with 3-4 mice per group) received a single dose of 200 μg EON through intraventricular (ICV) administration, with 12 groups receiving the EONs targeting the Q96 site (mB96-79 to -148, respectively as shown in FIGS.21A-21B), 6 groups receiving the EONs targeting the H101 site (mB101-18 to -127, respectively as shown in FIGS.21A-21B), and 1 group receiving the App control EON. Necropsy was at day 14 after EON administration. Mouse Slc12a5 editing levels were determined in the cortex, midbrain, hippocampus, and cerebellum brain regions, as well as in the cervical, thoracic, and lumbar spinal cord, using the methods and means as described in Example 6. Percentages of editing are presented in Table 19 (Q96) and Table 20 (H101).

[0296] Table 19. Editing percentages in mouse brain and spinal cord tissues, 14 days after ICV administration of 200 μg EON targeting the Q96 (Kcc2a) site, as indicated. mB96-118 was not well tolerated and did not provide trustworthy results in all mice (shown is data from one mouse only). Average calculations (Avg) are given in the rows with the bold values. CT = cortex, HC = hippocampus, MB = midbrain, CB = cerebellum, SC Cer = spinal cord cervical, SC Tho = spinal cord thoracic, and SC Lum = spinal cord lumbar.

[0297] Table 20. Editing percentages in mouse brain and spinal cord tissues, 14 days after ICV administration of 200 μg EON targeting the H101 (Kcc2a) site, as indicated. mB101- 68 and mB101-78 were not well tolerated by all mice and did not provide trustworthy results in some mice. Average calculations (Avg) are given in the rows with the bold values. CT = cortex, HC = hippocampus, MB = midbrain, CB = cerebellum, SC Cer = spinal cord cervical, SC Tho = spinal cord thoracic, and SC Lum = spinal cord lumbar.

[0298] Average editing percentages for the mouse App target were 0.8% (cortex), 2.1% (hippocampus), 2.2% (midbrain), 2.3% (cerebellum), 3.4% (spinal cord cervical), 3.9% (spinal cord thoracic), and 2.6% (spinal cord lumbar).

[0299] These results clearly show an improved results with significant increase in editing percentages in comparison to the results shown in FIG. 13E, FIG. 13F, FIG. 13G, and FIG.14D.

[0300] In respect of the Q96R edit, > 50% editing was observed in the hippocampus of a mouse treated with mB96-112 (SEQ ID NO:485). The highest editing was observed with EONs having a length of 31 nucleotides (mB79-79, -81, and -83) with an average editing of ~25% in the spinal cord regions. All EONs showed the highest editing percentages in the spinal cord regions anyhow. mB96-148 is not the best performer, but it is one of the better performing 29-nucleotide EONs that were tested.

[0301] In respect of the H101R edit, up to 20% editing was observed in the spinal cord lumbar tissue of a mouse treated with mB101-71 (SEQ ID NO:492), even though two animals were likely mis-dosed in the same group. Editing efficiency with H101 targeting EONs in the spinal cord tissues was like what was observed with the Q96 targeting EONs.

[0302] Together, these results show that potentially clinically relevant editing percentages of editing can be achieved in vivo using EONs that cause a change from glutamine at position 96 (in the KCC2a isoform) to arginine, thereby providing the KCC2 protein with a gain-of-function. The same holds true for EONs that cause a change from histidine at position 101 (in the KCC2a isoform) to arginine, also thereby providing the KCC2 protein with a gain-of-function.

[0303] Such gain-of-function effects are studied in a mouse model for epilepsy, in which epileptic attacks are simulated by induction, and EON treatment targeting the mouse Slc12a5 transcript as outlined in detail above, prevents the occurrence or the frequency of such epileptic events. Furthermore, KCC2 functionality is further investigated before and after EON induced editing as outlined above, by measuring the K+ influx and / or the membrane potential, investigating and emphasizing the importance of the interaction of the N-terminal peptide with its respective binding partners elsewhere in the protein as discussed in detail above.

Claims

WHAT IS CLAIMED IS:

1. An RNA editing oligonucleotide (EON) forming a double-stranded complex with a human SLC12A5 RNA molecule in a cell, said SLC12A5 RNA molecule comprising a target adenosine, wherein the target adenosine is: i) the A in the UAC codon encoding tyrosine (Y) at position 68; ii) the A in the ACC codon encoding threonine (T) at position 69; iii) the first A in the AAC codon encoding asparagine (N) at position 70; iv) the second A in the AAC codon encoding asparagine (N) at position 70; v) the A in the CAG codon encoding glutamine (Q) at position 73; vi) the A in the CAU codon encoding histidine (H) at position 78; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82; ix) the A in the UAC codon encoding tyrosine (Y) at position 446; x) the A in the AGU codon encoding serine (S) at position 932; and / or xi) the A in the ACG codon encoding threonine (T) at position 906; wherein each amino acid position is according to the amino acid sequence of SEQ ID NO:477, wherein a nucleotide in the EON that is directly opposite the target adenosine is the orphan nucleotide, wherein counting of the nucleotides in the EON is such that the orphan nucleotide is number 0 and the nucleotides 5’ from the orphan nucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, optionally wherein the double-stranded complex can recruit an endogenous ADAR enzyme in the cell to deaminate the target adenosine into an inosine, thereby editing the SLC12A5 RNA molecule.

2. The EON according to claim 1, wherein the EON is 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49 or 50 nucleotides in length.

3. The EON according to claim 1 or 2, wherein the deamination of the target adenosine results in a SLC12A5-encoded KCC2 protein with an increased activity, optionally resulting in anincreased K+ flux, optionally wherein the increased activity results in a higher GABAergic inhibition.

4. The EON according to any one of claims 1 to 3, wherein the cell is a neuron, preferably a brain cell, and wherein the RNA molecule is a pre-mRNA or an mRNA molecule.

5. The EON according to any one of claims 1 to 4, comprising at least one non-naturally occurring chemical modification, and / or comprising one or more additional non-naturally occurring chemical modifications in the ribose, linkage, or base moiety.

6. The EON according to claim 5, wherein the one or more additional modifications in the linkage moiety is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methylphosphonate (MP), sulfonylphosphoramidate, mesyl phosphoramidate (PNms), and a (1,3- dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi) internucleotide linkage.

7. The EON according to claim 5 or 6, wherein the one or more additional modifications in the ribose moiety is 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.

8. The EON according to any one of claims 1 to 7, wherein the orphan nucleotide is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase (Z).

9. The EON according to any one of claims 1 to 7, wherein the orphan nucleotide is a deoxynucleotide comprising a cytosine analog nucleobase according to formula (I):or any of its tautomeric forms, wherein: - R1, R2, R4and / or R5is H; OH; SH; =O; NH2; a halogen; a linear or branched lower (C1- C10) alkyl; or a C1-C6 cycloalkyl, wherein the alkyl and / or cycloalkyl is optionally interrupted by one or more heteroatoms; and - R3is H; OH; SH; =O; NH2; or a halogen.

10. The EON according to claim 9, wherein R1, R2, R3, R4and R5are H.

11. The EON according to any one of claims 1 to 10, wherein the target adenosine is: - the first A in the GAA codon encoding glutamic acid (E) at position 79; - the first A in the GAA codon encoding glutamic acid (E) at position 82; - the A in the AGU codon encoding serine (S) at position 932; and / or - the A in the ACG codon encoding threonine (T) at position 906, wherein the nucleotide on the -1 position in the EON can induce a syn conformation of the guanosine that is located directly 5’ from the target adenosine, preferably wherein the nucleotide on the -1 position is a 7-deaza-2’-deoxyadenosine.

12. The EON according to any one of claims 1 to 10, wherein the target adenosine is: - the A in the ACC codon encoding threonine (T) at position 69; - the first A in the AAC codon encoding asparagine (N) at position 70; - the A in the CAG codon encoding glutamine (Q) at position 73; and / or - the A in the CAU codon encoding histidine (H) at position 78; wherein the nucleotide on the -1 position in the EON is a deoxyinosine.

13. The EON according to claim 12, wherein the target adenosine is the A in the CAG codon encoding glutamine (Q) at position 73, wherein the nucleotide on the -2 position in theEON is a nucleotide comprising a hypoxanthine nucleobase, and preferably wherein said nucleotide at position -2 is 2’-OMe modified.

14. The EON according to claim 12, wherein the target adenosine is the A in the CAG codon encoding glutamine (Q) at position 73, wherein the EON comprises the sequence: 5’-…N5*N4N3N2*N1*O*M1^M2M3…-3’ wherein: - N5 is a 2’-F modified nucleotide comprising an uracil nucleobase (Uf); - N4 is Uf, or a 2’-OMe modified nucleotide comprising an uracil nucleobase (Um); - N3 is a 2’-OMe modified cytidine nucleotide (Cm), or a 2’-F modified cytidine nucleotide Cf; - N2 is Cf; - N1 is a deoxycytidine (Cd); - O is the orphan nucleotide that is directly opposite the target adenosine, wherein O is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1H)-pyridone nucleobase (Zd), or a deoxynucleotide comprising a 5-aza-5,6-dihydro cytosine nucleobase (Ed); - M1 is deoxynucleotide comprising a hypoxanthine nucleobase (Id); - M2 is a 2’-OMe modified nucleotide comprising a guanine nucleobase (Gm), or a hypoxanthine nucleobase (Im); - M3 is a 2’-F modified nucleotide comprising a guanine nucleobase (Gf), or a hypoxanthine nucleobase (If); - * is a phosphorothioate linkage; - ^ is a methylphosphonate linkage; - the linkage between N4 and N3 is a phosphorothioate linkage, or a (1,3- dimethylimidazolidin-2-ylidene) phosphoramidate linkage; - the linkage between N3 and N2 is a phosphodiester linkage, or a phosphorothioate linkage; and - the linkage between M2 and M3 is a phosphodiester linkage, or a phosphorothioate linkage.

15. The EON according to claim 12, wherein the target adenosine is the A in the CAG codon encoding glutamine (Q) at position 73, wherein the EON is selected from the groupconsisting of SEQ ID NO:53-122, 226-275, 335, 368-401, and 439-463, preferably wherein the EON is selected from the group consisting of SEQ ID NO: 238, 266, 394, 462, 335, 395, 377, 379, 400, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 463, 58, 73, 84, 98, 111, 112, 113, 114, 115, 120, 234, 236, 248, 267, and 271.

16. The EON according to claim 12, wherein the target adenosine is the A in the CAU codon encoding histidine (H) at position 78, wherein the EON is selected from the group consisting of SEQ ID NO: 123-182, 276-325, 338, 402-438, and 464-468, preferably wherein the EON is selected from the group consisting of SEQ ID NO:338, 418, 283, 404, 405, 411, 413, 420, 427, 428, 436, 464, 465, 466, 467, 468, 126, 125, 127, 130, 131, 137, 138, 140, 152, 179, 286, 292, 293, and 294.

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

18. A pharmaceutical composition comprising an EON according to any one of claims 1 to 16, or a vector according to claim 17, and a pharmaceutically acceptable carrier.

19. An EON according to any one of claims 1 to 16, or a vector according to claim 17, or the pharmaceutical composition of claim 18 for use as a medicament or for use in therapy.

20. An EON according to any one of claims 1 to 16, or a vector according to claim 17, or the pharmaceutical composition of claim 18 for use in the treatment of a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity.

21. An EON according to any one of claims 1 to 16, or a vector according to claim 17, or the pharmaceutical composition according to claim 18 for use in the treatment of a disorder, wherein the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, and / or epilepsy.

22. Use of an EON according to any one of claims 1 to 16, or a vector according to claim 17, in the manufacture of a medicament for the treatment of a disorder caused by a diminished GABAergic inhibition, preferably caused by a diminished KCC2 activity.

23. Use according to claim 22, wherein the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, and / or epilepsy.

24. A method of editing a human SLC12A5 polynucleotide, the method comprising contacting the SLC12A5 polynucleotide with an EON according to any one of claims 1 to 16 to effect an ADAR-mediated A to I deamination of a target adenosine in a codon encoding an amino acid, thereby editing the SLC12A5 polynucleotide, optionally wherein the amino acid is associated with: i) an autoinhibitory folding of the N-terminal peptide of SEQ ID NO:28 in the SLC12A5-encoded protein KCC2 protein; ii) K+ binding and / or transport; iii) phosphorylation of the SLC12A5-encoded protein KCC2; or a combination thereof.

25. The method according to claim 24, wherein the target adenosine is: i) the A in the UAC codon encoding tyrosine (Y) at position 68; ii) the A in the ACC codon encoding threonine (T) at position 69; iii) the first A in the AAC codon encoding asparagine (N) at position 70; iv) the second A in the AAC codon encoding asparagine (N) at position 70; v) the A in the CAG codon encoding glutamine (Q) at position 73; vi) the A in the CAU codon encoding histidine (H) at position 78; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82; ix) the A in the UAC codon encoding tyrosine (Y) at position 446; x) the A in the AGU codon encoding serine (S) at position 932; and / or xi) the A in the ACG codon encoding threonine (T) at position 906; wherein the amino acid position is according to the amino acid sequence of SEQ ID NO:

477.

26. A method of treating a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorder, epilepsy, or a disorder caused by a diminished GABAergic inhibition,preferably caused by a diminished KCC2 activity, in a subject in need thereof, the method comprising administering to the subject a therapeutically effective amount of an EON according to any one of claims 1 to 16, the vector according to claim 17, or the pharmaceutical composition according to claim 18.

27. The method according to claim 26, wherein administering is to the central nervous system, optionally via an intrathecal delivery.

28. The method according to claim 26 or 27, wherein the disorder is a neurodevelopment disorder, a neuropsychiatric disorder, a chronic pain disorders, and / or epilepsy.

29. A method of deaminating a target adenosine in an SLC12A5 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: i) providing the cell with an EON according to any one of claims 1 to 16; ii) allowing uptake by the cell of the EON; iii) allowing annealing of the EON to the SLC12A5 pre-mRNA or mRNA molecule; iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine; and optionally v) identifying the presence of the inosine in the target RNA molecule.

30. The method according to claim 29, wherein step v) comprises: a) determining the sequence of the SLC12A5 pre-mRNA or mRNA molecule; b) assessing the presence of an SLC12A5- encoded KCC2 protein with a lower phosphorylation rate or a gain-of-function in K+ binding and / or transport; or c) using a functional read-out, preferably assessing the level of GABAergic inhibition in the cell.

31. A nucleic acid molecule for editing a target adenosine in a human SLC12A5 pre-mRNA or mRNA molecule, wherein the target adenosine is: i) the A in the UAC codon encoding tyrosine (Y) at position 68; ii) the A in the ACC codon encoding threonine (T) at position 69; iii) the first A in the AAC codon encoding asparagine (N) at position 70; iv) the second A in the AAC codon encoding asparagine (N) at position 70;v) the A in the CAG codon encoding glutamine (Q) at position 73; vi) the A in the CAU codon encoding histidine (H) at position 78; vii) the first A in the GAA codon encoding glutamic acid (E) at position 79; viii) the first A in the GAA codon encoding glutamic acid (E) at position 82; ix) the A in the UAC codon encoding tyrosine (Y) at position 446; x) the A in the AGU codon encoding serine (S) at position 932; and / or xi) the A in the ACG codon encoding threonine (T) at position 906; wherein the amino acid position is according to the amino acid sequence of SEQ ID NO:477.

32. The nucleic acid molecule of claim 31 comprising an EON of claim 1.