Antisense oligonucleotides for the treatment of peroxisome biogenesis disorders in the zellweger spectrum (PBD-ZSD)

Abstract

The present invention relates to antisense oligonucleotides (AONs) that can mediate RNA editing by binding to a target RNA nucleic acid molecule, preferably an RNA transcript molecule, in a cell and recruiting an endogenous deaminating enzyme in the cell to deaminate one or more target adenosine nucleotides in the target RNA molecule to an inosine. The target RNA molecule is a transcript molecule from the PEX1 gene that encodes a variant PEX1 protein with an impaired functionality. The RNA editing of the target adenosine, preferably changing an aspartic acid residue to a glycine residue at position 843 in the protein (e.g., reversing a G843D mutation), results in a PEX1 protein with a wildtype functionality, thereby providing a potential treatment for a Peroxisome Biogenesis Disorder in the Zellweger Spectrum (PBD-ZSD).

Description

ANTISENSE OLIGONUCLEOTIDES FOR THE TREATMENT OF PEROXISOME BIOGENESIS DISORDERS IN THE ZELLWEGER SPECTRUM (PBD-ZSD)TECHNICAL FIELD

[0001] This disclosure relates to the field of medicine, and particularly to the field of peroxisome biogenesis disorders in the Zellweger spectrum (PBD-ZSD). The disclosure describes antisense oligonucleotides that mediate nucleotide-specific RNA editing in transcripts of the PEX1 gene encoding human peroxin 1 (also referred to as peroxisomal biogenesis factor 1), a 147 kDa member of the AAA protein family of ATPases to bring about amino acid changes of the encoded protein that influence its crucial activity in formation and function of peroxisomes.BACKGROUND

[0002] Peroxisomes are ubiquitous components of eukaryotic cells. These membrane-bound organelles are formed by replication through fission from pre-existing peroxisomes or can originate de novo by budding from the endoplasmic reticulum, and number up to several hundred per mammalian cell. They are indispensable for normal life and highly conserved throughout evolution amongst most eukaryotes. Peroxisomes were first identified by biochemist Christian DeDuve in the 1960s as cytoplasmic particles containing hydrogen peroxide-generating oxidases. Also, in the 1960’s, neurologist Hans Zellweger reported a familial condition that presented in children with craniofacial dysmorphism in addition to malformations in the brain, eye, liver, and kidney. These cases were initially described as “cerebro-hepato-renal-syndrome”, which was subsequently renamed as Zellweger syndrome (ZS). The absence of morphologically identifiable peroxisomes in hepatocytes and renal tubular cells of affected individuals (by using the peroxisomal matrix enzyme marker catalase) demonstrated a connection between ZS and peroxisome dysfunction. However, the underlying etiology was not recognized until the identification of multiple peroxisomal enzymes and their deficiencies in affected patients.

[0003] Contained within the peroxisome matrix of mammalian cells are over 70 distinct enzymes required for normal lipid metabolism and a host of other biochemical processes critical for normal health and development. Metabolic functions of peroxisomes are widespread and include hydrogen peroxide-based respiration, oxidation of Very Long Chain Fatty Acids (VLCFA), and biosynthesis of ether phospholipids, bile acids, and isoprene compounds. Several human diseases are caused by mutations in genes encoding peroxisomal metabolic enzymes, but the most dramatic loss of peroxisome function is observed in the Peroxisome Biogenesis Disorders (PBDs). For PBDs two broad clinical spectra can be distinguished, the Zellweger Spectrum Disorder (ZSD), accounting for about80% of all PBD patients, and the Rhizomelia Chondrodysplasia Punctata (RCDP) spectrum. PBDs are a heterogenous group of autosomal recessive diseases caused by mutations in any of the PEX genes. It is estimated that 1 in 50,000 births are affected by PBDs in North America, which estimates may increase with the introduction of newborn screening for peroxisomal disorders. Mutations in the PEX1 gene account for nearly 70% of all PBD-ZSD cases. Among the mutations identified in ZSD, the PEX7-p.Gly843Asp (p.G843D; c.2528G>A; rs61750420) founder allele occurs in high frequency. Although Zellweger spectrum disorders are all severe, the presence of at least one PEX7-p.G843D allele predicts a phenotype relatively milder than classic Zellweger syndrome. PEX1 (or peroxin 1) is an ATPase Associated with diverse cellular Activities (AAA) that oligomerizes with PEX6 and when complexed to PEX6, recycles PEX5 from the peroxisome membrane to the cytosol. The glycine residue at position 843 is in the D2 domain of PEX1 and the mutation to an aspartic acid residue is linked to a reduction of complex stability and is proposed to be involved in ATP binding. The interaction between PEX1-p.G843D and PEX6 is significantly reduced, and it has been reported that the residual activity of PEX1 -p.G843D mutant is down to levels of approximately 3-15% of wild type (Walter C et al. 2001. Am J Hum Genet 69:35- 48; Geisbrecht BV et al. 1998. Proc Natl Acad Sci USA 95:8630-8635).

[0004] Prior to the discovery of the shared peroxisomal basis, three different syndromes were historically described: i) Zellweger syndrome (also referred to as cerebrohepatorenal syndrome), ii) neonatal adreno-leukodystrophy (NALD), and iii) infantile Refsum disease. It is now generally accepted that these names ought to be replaced by “Peroxisome Biogenesis Disorders in the Zellweger Spectrum”, abbreviated to PBD-ZSD as indicated above, ranging from severe, intermediate, and mild phenotypes. This highlights the fact that the individual clinical pictures are along a spectrum of disease severity and often do not fit into the original assigned categories. Besides that, other variant phenotypes continue to be described.

[0005] As a result of impaired peroxisomal activities, individuals with PBD-ZSD can manifest a complex spectrum of clinical phenotypes that typically result in shortened life spans. The variability in disease manifestations ranging from onset of profound neurologic symptoms in newborns to progressive degenerative disease in adults presents practical challenges in disease diagnosis and medical management. There is no therapy for ZSD, and management is currently only supportive. One-half of the patients have a phenotype milder than classic Zellweger syndrome and exhibit a progressive disease course. In general, patients would benefit if therapies became available and were instituted early.

[0006] PBDs can be diagnosed by demonstrating abnormalities in several peroxisome biochemical functions that can be monitored in bodily fluids, with the primary step generallyinvolving the detection of elevated VLCFA in a fasting plasma sample. Elevations of C26:0 and C26:1 fatty acid and the ratios of C24:0 / C22:0 and C26:0 / C22:0 are consistent with a peroxisomal fatty acid p oxidation defect. On top of this, several other biomarkers exist. Next-generation sequencing panels for PEX genes are being used more frequently as confirmatory tests. Heterozygosity for PEX1-p.G843D typically predicts a milder phenotype, but also in this category there is a range of intellectual impairment and a shortened lifespan. Symptoms generally occur already in infancy and include hypotonia, retinitis pigmentosa, sensorineural hearing impairment, and hepatomegaly. Other manifestations may include leukodystrophy, moderate to severe psychomotor retardation, postnatal growth failure, and osteopenia. The onset of symptoms is typically short after birth with new complications developing over time. Life expectancy may be as short as less than twelve months for the most severe forms, to well into adulthood for the least severe forms.

[0007] Although there is some progress in the treatment of patients suffering from PBD-ZSD caused by the PEX1-p.G843D mutation, such as managing their symptoms and use of medical interventions and assistive devices, together with a dietary approach (avoiding food containing phytanic acid), these do not target the underlying cause of the disease. Most PBD-ZSD patients have progressive disorders in the intermediate or milder end of the spectrum and could benefit from therapeutics that could stop further deterioration. Hence, there is a high unmet need for improved and other mechanistic efforts to target PEX1- p.G843D and thereby restore the activity of the PEX1 protein.SUMMARY

[0008] Disclosed herein is an antisense oligonucleotide (AON) that is capable of recruiting an endogenous (= naturally present) ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target transcript molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target transcript molecule is a pre-mRNA or mRNA transcript molecule of a human PEX1 gene. In a preferred aspect, the human PEX1 gene encodes a loss-of-function variant of the peroxin 1 ATPase (PEX1) protein, or a PEX1 protein with an impaired functionality in comparison to a PEX1 protein based on NCBI Reference Sequence: NM_000466. Preferably, the cell is a mammalian cell with an impaired peroxisome activity, more preferably caused by a mutated PEX1 gene, even more preferably caused by the presence of two mutated PEX1 genes. In a preferred aspect, the loss-of-function variant is selected from the group consisting of p.Gly843Asp (G843D), p.Trp1250Ter (W1250*), p.Trp388Ter (W388*), p.Trp839Ter (W839*), and p.Trp2Ter (W2*). In a preferred aspect,the variant is G843D, which is caused by a c.2528G>A mutation in the PEX1 gene, wherein the target adenosine is the adenosine in the GAU codon for aspartic acid at position 843 of the encoded protein, and the deamination of the adenosine results in an inosine. The change to inosine makes that the codon encodes a glycine, which represents the wildtype version of the protein.

[0009] In a preferred aspect, the orphan nucleotide is a deoxynucleotide.

[0010] In a preferred aspect, the orphan nucleotide comprises a cytosine, a cytosine analog, an uracil, or an uracil analog.

[0011] In a preferred aspect, the cytosine analog is a 6-amino-5-nitro-3-yl-2(1 H)-pyridone nucleobase.

[0012] In another preferred aspect, the cytosine analog is according to formula (X):or any of its tautomeric forms, wherein: R1, R2, R4and / or R5is H; OH; SH; =0; NH2; a halogen; a linear or branched lower (C1-C10) alkyl; or a Ci-Ce cycloalkyl, wherein the alkyl and / or cycloalkyl is optionally interrupted by one or more heteroatoms; and R3is H; OH; SH; =0; NH2; or a halogen. More preferably, all of R1, R2, R3, R4and R5are H.

[0013] In another preferred aspect, the orphan nucleotide comprises an uracil anaolog, wherein the uracil analog is an iso-uracil nucleobase.

[0014] In one aspect, the nucleotide numbering is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end. Preferably, the first nucleotide 3’ from the orphan nucleotide (position -1) is a nucleotide that can induce a syn conformation of the guanosine positioned opposite in the target sequence. In a preferred aspect, the nucleotide at position -1 in the AON is 7-deaza-2’-deoxyadenosine (7-deaza Ad), 3-deaza-2’-deoxyadenosine (3- deaza Ad), or 2’-deoxyinosine (Id).

[0015] In a preferred aspect, the AON is 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length.

[0016] In one preferred aspect, the AON comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP),sulfonylphosphoramidate, (1 ,3-dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi), or a mesyl phosphoramidate (PNms) internucleotide linkage. In one aspect, the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein linkage position -2 is an MP linkage or a PNms linkage, preferably a PNms linkage. In a preferred aspect, the linkage between the most terminal two nucleotides on the 5’ and / or 3’ terminus of the AON is a PNdmi linkage or a PNms linkage.

[0017] In one aspect, the AON comprises one or more nucleotides comprising a mono- or disubstitution at the 2', 3' and / or 5' position of the ribose, each independently selected from the group consisting of: -OH; -F; substituted or unsubstituted, linear or branched lower (Ci- C ) 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.

[0018] In a preferred aspect, the oligonucleotide comprises or consists of the sequence of any one of SEQ ID NO: 434, 435, 427, 442, 401 , 431 , 430, 432, 433, 421 , 358, 359, 355, 400,399, 135, 429, 444, 439, 443, 420, 423, 331 , 335, 344, 226, 222, 208, 212, 213, 216, 217,218, 219, 220, 225, 227, 234, 235, 236, 237, 238, 240, 597, 608, 610, 605, 779, 607, 617,591 , 722, 549, and 241.

[0019] The present disclosure relates to a vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON as disclosed herein.

[0020] The present disclosure relates to a pharmaceutical composition comprising an AON as disclosed herein, or a vector as disclosed herein, and a pharmaceutically acceptable carrier.

[0021] The present disclosure also relates to an AON as disclosed herein, for use in the treatment of a Peroxisome Biogenesis Disorder in the Zellweger Spectrum (PBD-ZSD).

[0022] The present disclosure also relates to a use of an AON as disclosed herein, in the manufacture of a medicament for the treatment of a PBD-ZSD.

[0023] The present disclosure also relates to a method of editing a human PEX1 polynucleotide in a cell, wherein the human PEX1 polynucleotide is a pre-mRNA or mRNA molecule transcribed from a variant PEX1 gene that encodes a PEX1 protein with an impaired functionality, the method comprising contacting the PEX1 polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, preferably wherein the AON is as disclosed herein, thereby editing the PEX1 polynucleotide to encodea PEX1 protein with a wildtype functionality, preferably wherein the PEX1 protein with an impaired functionality is selected from the group consisting of p.Gly843Asp, p.Trp1250Ter, p.Trp388Ter, p.Trp839Ter, and p.Trp2Ter. Preferably the method comprises the step of deaminating the adenosine in the codon GAU, coding for aspartic acid at position 843 in the G843D-mutated human PEX1 ATPase enzyme, thereby rendering the change from GAU to GIU, which is read as GGU by the translation machinery, and thereby changing the codon for aspartic acid to a codon for glycine.

[0024] The present disclosure also relates to a method of treating, ameliorating, or slowing down the progression of a PBD-ZSD, the method comprising administering to said subject an AON as disclosed herein, thereby contacting a PEX1 polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination, thereby contacting a PEX1 polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination at position 2528 in the human PEX1 coding sequence, thereby editing the PEX1 polynucleotide to encode a PEX1 protein with a wildtype activity, thereby treating the subject.

[0025] The present disclosure also relates to an in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in an PEX1 target transcript molecule in a cell, the method comprising the steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the target transcript molecule; and (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target transcript molecule to an inosine. In a preferred aspect, the in vivo method as disclosed herein comprises further step (v) of using a functional read-out to identify the presence of the inosine in the target transcript molecule.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0027] FIG. 1 shows the 5’ to 3’ sequence of part of the human PEX1 mRNA transcript in which the GAU codon coding for aspartic acid (D) at position 843 in the PEX1 protein is in bold (SEQ ID NO:207). The underlined adenosine is the target for RNA editing as disclosed herein, resulting in a codon coding for a glycine (G) residue (GIU / GGU) at this position after editing. Below the target sequence a set of AONs is provided (from 5’ to 3’, with their respective aliases, SEQ ID NO’s, and RM numbers, if applicable), that were designed to target the target adenosine in human PEX1 transcripts as exemplified by SEQ ID NQ:207. The chemical modifications are as follows: Ae and Ge are 2’-MOE modified adenosine and guanosine, respectively; m5Ce is 2’-MOE modified 5-methyl-cytidine; m5Ue is 2’-MOE modified thymidine (nucleotide carrying a 5-methyluracil nucleobase (= thymine) with a 2’-MOE substitution in the ribose sugar moiety; elsewhere also referred to as ‘Te’); Cm, Am, Um, and Gm are 2’-0Me modified cytidine, adenosine, uridine, and guanosine, respectively; Gf, Cf, Af, and Uf are 2’-fluoro modified guanosine, cytidine, adenosine, and uridine, respectively; Zd (orphan nucleotide) is a deoxynucleotide (deoxycytidine analog) carrying a Benner’s base; Ad is deoxyadenosine; Cd is deoxycytidine; Id is a deoxynucleotide carrying a hypoxanthine nucleobase; 8Ud (at the orphan nucleotide position) is a deoxynucleotide carrying an iso-uracil nucleobase; m5UI is a locked nucleic acid (LNA) carrying a 5- methyluracil nucleobase; 7Ad refers to 7-deaza-2’-deoxyadenosine; “I” refers to a PNdmi linkage; “A” refers to a MP linkage; “*” refers to a PS linkage; “#” refers to a PNms linkage, and “e” refers to a phosphodiester (PO) linkage. “N” refers to a nucleotide at position -1 in the AON (which is positioned opposite the guanosine that is 5’ of the target adenosine in the target sequence), that is a nucleotide analog that can induce a syn conformation of the opposite guanosine. In a preferred aspect, N is a deoxynucleotide. In a more preferred aspect, N is a deoxynucleotide and N comprises a modified purine nucleobase. In an even more preferred aspect, this modified purine is 7-deaza-2’-deoxyadenosine (7-deaza Ad). Other preferred nucleotides that can be positioned at the N position are 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 (Ir); 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-1 / 7-imidazol-1-yl) riboside; beta-(4-amidino-1 / 7-imidazol-1-yl) 2’-deoxyriboside; beta-(4-amidino-1 / 7-imidazol-1-yl) 2’-ara-fluoro-2’-deoxyriboside; and beta-(4-amidino-1 / 7- imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside.

[0028] FIG. 2 shows general AON sequences, wherein:SEQ ID NO: 175 represents the basic 32 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-01 to 04;SEQ ID NO: 176 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-05 to 08;SEQ ID NO: 177 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-09 to 12;SEQ ID NO: 178 represents the basic 26 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-13 to 16;SEQ ID NO: 179 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-17 to 20;SEQ ID NO: 180 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-21 to 24;SEQ ID NO: 181 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-25 to 29;SEQ ID NO: 182 represents the basic 27 nucleotide sequence (without modifications, except for the orphan nucleotide) in AON PEX1-30;SEQ ID NO: 183 represents the basic 32 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-31 to 34;SEQ ID NO: 184 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-35 to 38;SEQ ID NO: 185 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-39 to 42;SEQ ID NO: 186 represents the basic 26 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-43 to 46;SEQ ID NO: 187 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-47 to 50;SEQ ID NO: 188 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-51 to 54;SEQ ID NO: 189 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide) in AONs PEX1-55 to 59;SEQ ID NO: 190 represents the basic 27 nucleotide sequence (without modifications, except for the orphan nucleotide) in AON PEX1-60;SEQ ID NO: 191 represents the basic 32 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-61 to 64;SEQ ID NO: 192 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-65 to 68;SEQ ID NO: 193 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-69 to 72;SEQ ID NO: 194 represents the basic 26 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-73 to 76;SEQ ID NO: 195 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-77 to 80;SEQ ID NO: 196 represents the basic 28 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-81 to 84;SEQ ID NO: 197 represents the basic 30 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-85 to 89;SEQ ID NO: 198 represents the basic 27 nucleotide sequence (without modifications, except for the orphan nucleotide and the nucleotide at position -1) in AONs PEX1-90;SEQ ID NO: 199 represents the basic 32 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-91 to 94;SEQ ID NO:200 represents the basic 30 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-95 to 98;SEQ ID NO:201 represents the basic 28 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-99 to 102;SEQ ID NO:202 represents the basic 26 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-103 to 106;SEQ ID NO:203 represents the basic 30 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-107 to 110; SEQ ID NO:204 represents the basic 28 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-111 to 114;SEQ ID NO:205 represents the basic 30 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-115 to 119; SEQ ID NO:206 represents the basic 27 nucleotide sequence (without modifications, except for the nucleotide at position -1) in AONs PEX1-120; and wherein “Z” (orphan nucleotide) refers to a nucleotide carrying a Benner’s base; wherein “8U” refers to a nucleotide carrying an iso-uracil nucleobase; and wherein “N” refers to a nucleotide at position -1 in the AON (which is positioned opposite the guanosine that is 5’ ofthe target adenosine in the target sequence), that is a nucleotide analog that can induce a syn conformation of the opposite guanosine, preferably selected from the group as outlined above.

[0029] FIG. 3 shows the editing percentage in an in vitro biochemical editing assay (BEA) using a set of AONs that differ predominantly in the nucleotide at the -1 position. Negative controls were a sample in which no AON was used (No EON), a sample without target RNA and a sample in which no purified ADAR was applied.

[0030] FIG. 4A shows the editing percentage after pyrosequencing after the BEA with RM 120721 using four different concentrations as indicated. A maximum of 55% is reached after 20 minutes. FIG. 4B shows the editing percentage after pyrosequencing after the BEA with RM 120770 using four different concentrations as indicated. A maximum of 70% is reached after 20 minutes.

[0031] FIG. 5 shows the editing percentage in human Zellweger patient fibroblasts (GM16514) that are heterozygous for the c.2528G>A (rs61750420) mutation in the PEX1 gene, after transfection with the indicated AONs (details shown in FIG. 1). Controls are a non-treated (NT) sample, a Mock transfection, and a transfection with a beta-ACTIN specific AON (ACTB). The wildtype signal observed in the three negative controls (approximately reaching 30% in the ddPCR assay) has been deducted from the totals for each AON.

[0032] FIG. 6 shows an additional set of AONs targeting the c.2528G>A (rs61750420) in the human PEX1 transcript for deamination of the adenosine into inosine. The chemical modifications are as described in FIG. 1 , wherein Gd is deoxyguanosine, 3Ad refers to a 3- deaza-2’-deoxyadenosine (3-deaza Ad), and Ed is a deoxynucleotide comprising an E-base.

[0033] FIG. 7 shows a part of the human PEX1 transcript (5’ to 3’; SEQ ID NO:462), the equivalent part of the monkey (macaque) PEX1 transcript (5’ to 3’; SEQ ID NO:463), and the equivalent part of the mouse (Mus musculus) mPex1 transcript (5’ to 3’; SEQ ID NO:464), with the target adenosine of the c.2528G>A; rs61750420 mutation (in human) and its equivalents in large bold font. Underlined are nucleotides that differ from the human sequence. Below the three target sequences two sequences and modifications are provided for AONs that are 100% complementary to the mouse target sequence (except for the orphan position versus the target adenosine). Chemical modifications are as provided in FIG. 1 and FIG. 6.

[0034] FIG. 8 shows an additional set of AONs targeting the c.2528G>A (rs61750420) in the human PEX1 transcript for deamination of the adenosine into inosine. The chemical modifications are as described in FIG. 1 and FIG. 6, wherein If is a 2’-fluoro modified nucleotide carrying a hypoxanthine nucleobase, le is a 2’-MOE modified nucleotide carryinga hypoxanthine nucleobase, and Im is a 2’-0Me modified nucleotide carrying a hypoxanthine nucleobase.DETAILED DESCRIPTION

[0035] The present disclosure relates to an approach of restoring the activity of PEX1 , namely by using antisense oligonucleotides (AONs) and the cell’s own nucleic acid post- transcriptional modification machinery to specifically target and amend a nucleotide in the PEX1 (pre-) mRNA transcript. In the human PEX1 protein (see NCBI Reference Sequence: NM_000466), the glycine at position 843 is encoded by the codon GGT (or GGU in the mRNA transcript), whereas the aspartic acid in the G843D mutation is encoded by GAT (or GAU in the mRNA transcript), which results from the G>A mutation at the middle nucleotide position. The present invention relates to changing the GAU codon in the (pre-)mRNA of the variant transcript to GGU, which then encodes the glycine at that position and providing a protein with wildtype activity / function. The technology that the present disclosure relates to is generally referred to as ‘RNA editing’.

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

[0037] ADAR is a multi-domain protein, comprising of a catalytic domain and two to three double-stranded RNA recognition domains, depending on the enzyme in question. Each recognition domain recognizes a specific double-stranded RNA (dsRNA) sequence and / or conformation. The catalytic domain does also play a role in recognizing and binding a part of the dsRNA helix, although the key function of the catalytic domain is to convert an A into an I in a nearby, predefined, position in the target RNA, by deamination of the nucleobase. Inosine is read as guanosine by the translational machinery of the cell, meaning that, if an edited adenosine is in a coding region of an mRNA or pre-mRNA, it can recode the protein sequence. A-to-l 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’ untranslated region (UTR) or other non-coding parts of the transcript, which may affect the processing and / or stability of the RNA. In addition, A-to-l conversions may take place in splice elements in introns or exons in pre-mRNAs, thereby altering the pattern of splicing. As a result, exons may be included or skipped. The enzymes catalysing adenosine deamination are within an enzyme family of ADARs, which include human deaminases hADARI and hADAR2, as well as hADAR3. However, for hADAR3 no deaminase activity has been demonstrated.

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

[0039] It is known that ADAR may act on any dsRNA. Through a process sometimes referred to as ‘promiscuous editing’, the enzyme will edit multiple A’s in the dsRNA. Hence, there was a need for methods and means that circumvent such promiscuous editing and only target specific adenosines in a target RNA molecule to become therapeutic applicable. Vogel et al. (2014) showed that such off-target editing can be suppressed by using 2’-O-methyl (2’- OMe) modified nucleosides in the oligonucleotide at positions opposite to adenosines that should not be edited and used a non-modified nucleoside directly opposite to the specifically targeted adenosine on the target RNA. However, the specific editing effect at the target nucleotide has not been shown to take place without the use of recombinant ADAR enzymes having covalent bonds with the AON. Several publications have now shown that the recruitment of endogenous ADAR (hence without the need for an exogenous and / or recombinant source) is feasible while maintaining a specificity in which a single adenosine within a target RNA molecule can be targeted and deaminated to an inosine. Int. Patent Application Publication No. WO2016 / 097212 discloses AONs for the targeted editing of RNA, wherein the AONs are characterized by a sequence that is complementary to a target RNA sequence (therein referred to as the ‘targeting portion’) and by the presence of a stemloop / hairpin structure (therein referred to as the ‘recruitment portion’), which is preferablynon-complementary to the target RNA. Such oligonucleotides are referred to as ‘self-looping AONs’. The recruitment portion acts in recruiting a natural ADAR enzyme present in the cell (endogenously present) to the dsRNA formed by hybridization of the target sequence with the targeting portion. Due to the recruitment portion, there is no need for conjugated entities or presence of modified recombinant ADAR enzymes. Int. Patent Application Publication No. WO2016 / 097212 describes the recruitment portion as being a stem-loop structure mimicking either a natural substrate (e.g., the GluB receptor) or a Z-DNA structure known to be recognized by the dsRNA binding domains, or Z-DNA binding domains, of ADAR enzymes. A stem-loop structure can be an intermolecular stem-loop structure, formed by two separate nucleic acid strands, or an intramolecular stem loop structure, formed within a single nucleic acid strand. The stem-loop structure of the recruitment portion as described is an intramolecular stem-loop structure, formed within the AON itself, and are thought to attract (endogenous) ADAR. Similar stem-loop structure-comprising systems for RNA editing have since then been described in Int. Patent Application Publication Nos. WO2017 / 050306, W02020 / 001793, WO2017 / 010556, US11 ,390,865, W02020 / 246560, andWO2022 / 078995.

[0040] Int. Patent Application Publication Nos. WO2017 / 220751 and WO2018 / 041973 describe a next generation type of AONs that do not comprise such a stem-loop structure but that are (almost fully) complementary to the targeted area, and that appeared still capable of attracting endogenous ADAR enzymes. In one embodiment, one or more mismatching nucleotides, wobbles, or bulges exist between the oligonucleotide and the target sequence. A sole mismatch may be at the site of the nucleoside opposite the target adenosine, but in other embodiments AONs (or “RNA editing oligonucleotides” - even though the deamination reaction is carried out by the ADAR enzyme - and often abbreviated to ‘EONs’) were described with multiple bulges and / or wobbles when attached to the target sequence area. It appeared possible to achieve in vitro, ex vivo and in vivo RNA editing with AONs lacking a stem-loop structure and with endogenous ADAR enzymes when the sequence of the AON was carefully selected such that it could attract / recruit ADAR. The ‘orphan nucleoside’, which is defined as the nucleoside in the AON that is positioned directly opposite the target adenosine in the target RNA molecule, was a nucleotide with an unmodified cytosine nucleobase and that did not carry a 2’-OMe modification. The orphan nucleoside can be a deoxyribonucleoside (DNA), wherein the remainder of the AON could still carry 2’-O-alkyl modifications at the sugar entity (such as 2’-OMe), or the nucleotides directly surrounding the orphan nucleoside contained chemical modifications (such as DNA in comparison to RNA) that further improved the RNA editing efficiency and / or increased the resistance against nucleases. Such effects could even be further improved by using sense oligonucleotides (SONs) that ‘protected’ the AONs against breakdown upon delivery to thecells (described in Int. Patent Application Publication No. W02018 / 134301 and US Patent No. US11 ,274, 300).

[0041] The use of chemical modifications and particular structures in oligonucleotides that could be used in ADAR-mediated editing of specific adenosines in a target RNA have been the subject of numerous disclosures in the field, such as Int. Patent Application Publication Nos. WO2019 / 111957, WO2019 / 158475, W02020 / 165077, W02020 / 201406,W02020 / 211780, WO2021 / 008447, WO2021 / 020550, WO2021 / 060527, WO2021 / 117729, WO2021 / 136408, WO2021 / 182474, WO2021 / 216853, WO2021 / 242778, WO2021 / 242870, WO2021 / 242889, W02022 / 007803, W02022 / 018207, WO2022 / 026928, andWO2022 / 124345. The use of specific sugar moieties has been disclosed in for instance Int. Patent Application Publication Nos. W02020 / 154342, W02020 / 154343, W02020 / 154344, WO2022 / 103839, and WO2022 / 103852, whereas the use of stereo-defined linker moieties(in general for oligonucleotides that for instance can be used for exon skipping, in gapmers, in siRNA, or specifically for RNA-editing oligonucleotides, related to a wide variety of target sequences) has been described in Int. Patent Application Publication Nos. WO2011 / 005761 , WO2014 / 010250, W02014 / 012081 , WO2015 / 107425, WO2017 / 015575 (HTT), WO2017 / 062862, WO2017 / 160741 , WO2017 / 192664, WO2017 / 192679 (DMD), WO2017 / 198775, WO2017 / 210647, WO2018 / 067973, WO2018 / 098264, WO2018 / 223073(APOC3), WO2018 / 237194, WO2019 / 032607 (C9orf72), WO2019 / 055951 , WO2019 / 075357 (SMA / ALS), W02019 / 200185 (DM1), WO2019 / 217784 (DM1), WO2019 / 219581 , W02020 / 118246 (DM1), W02020 / 160336 (HTT), WO2020 / 191252, W02020 / 196662, WO2020 / 219981 (USH2A), WO2020 / 219983 (RHO), WO2020 / 227691 (C9orf72), WO2021 / 071788 (C9orf72), WO2021 / 071858, WO2021 / 178237 (MAPT), WO2021 / 234459, WO2021 / 237223, WO2022 / 099159, WO2021 / 030778, WO2022 / 174053, and WO2023 / 278589. Next to these disclosures, an extensive number of publications relate to the targeting of specific RNA target molecules, or specific adenosines within such RNA target molecules, be it to repair a mutation that resulted in a premature stop codon, or other mutation causing disease. Examples of such disclosures in which adenosines are targeted within specified target RNA molecules are Int. Patent Application Publication Nos.W02020 / 157008 and WO2021 / 136404 (USH2A); WO2021 / 113270 (APP);WO2021 / 113390 (CMT1A); W02021 / 209010 (IDUA, for Hurler syndrome);WO2021 / 231673 and WO2021 / 242903 (LRRK2); WO2021 / 231675 (ASS1);WO2021 / 231679 (GJB2); WO2019 / 071274 and WO2021 / 231680 (MECP2, for RETT syndrome); WO2021 / 231685 and WO2021 / 231692 (OTOF, for autosomal recessive non- syndromic hearing loss); WO2021 / 231691 (XLRS); WO2021 / 231698 (argininosuccinate lyase deficiency); W02021 / 130313 and WO2021 / 231830 (ABCA4, for Stargardt disease); WO2023 / 152371 (PCSK9); and WO2021 / 243023 (SERPINA1 ; for Alpha-1 -Antitrypsindeficiency; see also I nt. Patent Application Publication Nos. WO2016 / 097212, WO2017 / 220751 , and WO2018 / 041973).

[0042] The AONs as disclosed herein can recruit deaminating enzymes, such as ADAR1 and / or ADAR2 that are endogenously present (= naturally present) in a cell. An AON as disclosed herein can mediate RNA editing of a target adenosine present in a target RNA molecule after it is bound to the target RNA molecule, since the deaminating enzymes are recruited to the double-stranded AON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine.

[0043] The present invention aims to provide one or more alternative, and / or improved, compounds or compositions for use in the treatment of PBD-ZSD caused by the c.2528G>A mutation in the human PEX1 gene. Mutations in the human PEX1 gene account for nearly 70% of all Peroxisome Biogenesis Disorders in the Zellweger Spectrum (PBD-ZSD), with the c.2528G>A mutation (resulting in G843D) as the founder allele occurring in a high frequency of patients in the western world. No treatment is currently available for these patients, except dietary restrictions and supportive treatment to relieve symptoms, which start in early childhood and progress until an often-early death of the patient. The inventors here realized that this mutation is eligible for RNA editing, and that RNA editing of the adenosine at position 2528 in the PEX1 transcripts of these patients to an inosine (read as a guanosine by the translation machinery) would result in a wildtype version of the encoded PEX1 protein, thereby elevating disease. Hence, RNA editing using editing antisense oligonucleotides would be a first-in-class treatment for these PBD-ZSD patients.

[0044] RNA editing is not gene therapy, because it is transient 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 PEX1 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. For instance, transcripts of PEX genes different from PEX1 are not edited through the compounds as disclosed herein.

[0045] The AONs as disclosed herein can recruit deaminating enzymes, such as ADAR1 and / or ADAR2 that are endogenously present in a cell. An AON as disclosed herein can mediate RNA editing of a target adenosine present in a target RNA molecule after it is bound to the target RNA molecule, since the deaminating enzymes are recruited to the doublestranded AON / target RNA molecule complex and subsequently deaminate the target adenosine into an inosine.

[0046] The oligonucleotides are herein abbreviated to “AONs”, but sometimes also referred to as ‘editing oligonucleotides’, or ‘EONs’, even though the RNA editing event is performed by the deamination enzyme and the action of the oligonucleotide only triggers the RNA editing to take place. There is a constant need for improving the pharmacokinetic properties of the AONs without negatively affecting the efficiency in which the target adenosine is edited in the target RNA, and / or without negatively affecting the stability of the AON itself, which is constantly prone to breakdown because of nucleases present in a natural cell. Many chemical modifications are available for the generation of AONs (and many have been applied in the art). However, many of these properties are not always compatible with the desire of achieving efficient RNA editing. In the search for better pharmacokinetic properties, it was found earlier that a 2’-O-methoxyethyl (or 2’-methoxyethoxy, or 2’-MOE) modification of the ribose of some, but not all, nucleotides surprisingly appeared compatible with efficient ADAR engagement and editing (Int. Patent Application Publication No. WO2019 / 158475). In a similar fashion, it was found earlier that a PS linkage at some, but not all, internucleoside linkages surprisingly appeared compatible with efficient ADAR engagement and editing (Int. Patent Application Publication 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 AON appeared compatible with efficient engagement of an enzyme with nucleotide deamination activity and with subsequent deamination (Int. Patent Application Publication No. W02020 / 165077). Whereas the properties of phosphonoacetate and UNA modifications were known as such, the compatibility thereof with engagement of enzymes with nucleotide deamination activity and with the deamination reaction was not known.

[0047] Disclosed herein are AONs that can provide (mediate, cause, or trigger) RNA editing of a target adenosine in a target transcript molecule, such as pre-mRNA and / or mRNA. The target transcript molecule is preferably mutated at position 2528 (c.2528G>A) and the editing results in a transcript that encodes a protein with wildtype function properties.

[0048] RNA editing is often applied to correct G>A mutations that cause a disease. This is also the case in the present disclosure. Non-limiting examples of transcript molecules (as disclosed in the art) that are targeted using RNA editing for a variety of treatments are SERPINA 1 (for the treatment of alphal -antitrypsin (A1AT) deficiency; see e.g., Int. Patent Application Publication 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 (forthe treatment of Parkinson’s disease; see e.g., Int. Patent Application Publication Nos. WO2016 / 097212, WO2017 / 220751 , WO2018 / 041973, WO2021 / 231673 andWO2021 / 242903), ABCA4 (for the treatment of Stargardt disease; see e.g., Int. Patent Application Publication Nos. W02021 / 130313 and WO2021 / 231830), USH2A (for the treatment of Usher syndrome; see e.g., Int. Patent Application Publication Nos. W02020 / 157008, W02020 / 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).

[0049] The present disclosure relates to AONs that mediate RNA editing, using endogenous (naturally present) ADAR enzymes in the host cell (preferably cells with diminished or low activity of peroxisomes due to a mutated peroxin 1 protein) of an adenosines present in the transcript of the PEX1 gene. An AON as disclosed herein aims to restore the activity of the PEX1 protein in its lipid-metabolism function in peroxisomes. Targeting the adenosine in the codon for aspartic acid at position 843 in the PEX1 protein and changing it to an inosine resulting in a glycine at that position is a preferred example of a mutant PEX1 variant that is restored to a transcript that encodes a protein with a wildtype functionality.

[0050] Examples of pathogenic PEX1 variants that can be modified by targeted RNA editing (including the p.Gly843Asp encoded variant described herein in more detail) and that form part of the present invention comprise:

[0051] The G843D variant is based on isoform 1 of the protein, whereas G786D is the same position in the human PEX1 protein but then related to isoform 2 (which lacks the 57 amino acids of exon 12). The G635D variant is the same position but then related to isoform 3 (which has an alternative start site in exon 5). In any case, when the variant herein is referred to as G843D, which is reversed by the AONs as disclosed herein (D843G), the equivalent position is meant in isoform 2 and 3.Definitions

[0052] 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’ or a ‘deoxyoligoribonucleotide’, the nucleotide may comprise the nucleobase adenine, guanine, uracil, cytosine, hypoxanthine, 6-amino-5-nitro-3-yl-2(1 H)-pyridone, iso-uracil, thymine (= 5- methyluracil), or any other nucleobase known in the art. The nucleotide may also lack a base (= a-basic). An AON as disclosed herein may comprise a mix of ribonucleotides and deoxyribonucleotides. When a deoxyribonucleotide is used, hence without a modification at the 2’ position of the sugar, the nucleotide is often abbreviated to dA (or Ad), dC (or Cd), dG (or Gd) or m5Ud (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.

[0053] The term ‘nucleoside’ refers to the nucleobase linked to the (deoxy)ribosyl sugar, without phosphate groups. A ‘nucleotide’ is composed of a nucleoside and one or more phosphate groups. The term ‘nucleotide’ thus refers to the respective nucleobase- (deoxy)ribosyl-phospholinker, as well as any chemical modifications of the ribose moiety or the phospho group. Thus, the term would include a nucleotide including a locked ribosyl moiety (comprising a 2’-4’ bridge, comprising a methylene group or any other group), an unlocked nucleic acid (UNA), a threose nucleic acid (TNA), a nucleotide including a linker comprising a phosphodiester, phosphonoacetate, phosphotriester, PS, phosphoro(di)thioate, MP (or MeP), methyl thiophosphonate, phosphoramidate linkages, PNdmi, and a linkage according to the structure of formula (IX) 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 aneighbouring nucleoside and the linkage between these nucleosides is modified. As stated herein, a nucleotide is a nucleoside plus one or more phosphate groups. The terms ‘ribonucleoside’ and ‘deoxyribonucleoside’, or ‘ribose’ and ‘deoxyribose’ are as used in the art.

[0054] 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 (sometimes referred to as m5U) and is an uracil (II) 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.

[0055] Whenever reference is made to nucleotides in the oligonucleotide, such as cytosine, 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, 5-acetylcytosine, 5- hydroxycytosine, and p-D-glucosyl-5-hydroxymethylcytosine are included. Whenever reference is made to adenine, N6-methyladenine, 8-oxo-adenine, 2,6-diaminopurine and 7-methyladenine are included. Whenever reference is made to uracil, 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 and the linkage structure according to formula (IX), further outlined in detail below.

[0056] The term ‘comprising’ encompasses ‘including’ as well as ‘consisting of’, e.g., a composition ‘comprising X’ may consist exclusively of X or may include something additional, e.g., X + Y. The term ‘about’ in relation to a numerical value x is optional and means, e.g., x+10%.

[0057] 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.

[0058] The term ‘conducive to’ or ‘mediate’ can be used interchangeably with ‘capable of facilitating’. When used in the context of an AON that is conducive to ADAR editing (or can mediate ADAR editing), this means that the AON, after entry into the cell, interacts with the target RNA sequence, thereby forming a double stranded structure which is recognized by the ADAR enzyme, which can then deaminate the target adenosine into an inosine. Hence, the AON itself does not have the enzymatic function (the ADAR enzyme has), but it can trigger, induce, cause, organize, mediate, provide, give, produce, facilitate, result in RNA editing after binding to the target RNA molecule.

[0059] The term ‘mismatch’ is used herein to refer to opposing nucleotides in a double stranded RNA complex which do not form perfect base pairs according to the Watson-Crick base pairing rules. In the historical sense, mismatched nucleotides are G-A, C-A, ll-C, A-A, G-G, C-C, Il-Il pairs. In some embodiments AONs as disclosed herein comprise fewer than four mismatches with the target sequence, for example 0, 1 or 2 mismatches. ‘Wobble’ base pairs are G-ll, l-ll, l-A, and l-C base pairs. When a II is placed opposite the target A, there is no mismatch, and the AON may be 100% complementary, although an iso-uridine (iso-ll) opposite the target adenosine qualifies as a mismatch, since it does not pair according to the Watson-Crick rules of base pairing. When a C is placed opposite the target A, there is at least 1 mismatch between the AON and the target sequence. Although a G:G pairing would be considered a mismatch, that does not necessarily mean that the interaction is unstable, which means that the term ‘mismatch’ may be somewhat outdated based on the current disclosure where a Hoogsteen base-pairing may be seen as a mismatch based on the origin of the nucleotide but still be relatively stable. An isolated G:G pairing in duplex RNA can for instance be quite stable but will still be defined as a mismatch. Analysis of natural targets of ADAR enzymes has indicated that these generally include mismatches between the two strands that form the RNA helix edited by ADAR1 or 2. It has been suggested that these mismatches enhance the specificity of the editing reaction (Stefl et al. 2006. Structure 14(2):345-355; Tian et al. 2011. Nucleic Acids Res 39(13):5669-5681). Characterization of optimal patterns of paired / mismatched nucleotides between the AONs and the target RNA also appears important to the development of efficient ADAR-based AON therapy.

[0060] The term ‘complementary’ as used herein refers to the fact that the AON hybridizes under physiological conditions to a second nucleic acid strand. Examples are (i) when the AON as a first nucleic acid strand (= guide oligonucleotide) forms a heteroduplex RNAediting oligonucleotide complex with second complementary nucleic acid strand (in vitro), or (ii) when it forms a double stranded complex with the target RNA molecule. The term does not necessarily mean that each nucleotide in a nucleic acid strand has a perfect pairing with its opposite nucleotide in the opposite sequence. In other words, while an AON may be complementary to a target sequence, there may be mismatches, wobbles and / or bulges between the AON and the target sequence, while under physiological conditions that AON still hybridizes to the target sequence such that the cellular RNA editing enzymes can deaminate the target adenosine to an inosine. The term ‘substantially complementary’ therefore also means that despite the presence of the mismatches, wobbles, and / or bulges, the AON has enough matching nucleotides with the target sequence that under physiological conditions the AON hybridizes to the target RNA molecule. As shown herein, an AON may be complementary, but may also comprise one or more mismatches, wobbles and / or bulges with the target sequence, if under physiological conditions the AON is able to hybridize to its target.

[0061] 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 O.M. et al. Biochemistry. 2000. 39(40): 12243-12251). Also, because ADARs require duplex RNA for activity, their reaction can be directed to specific adenosines in different transcripts 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). This is also the subject of the present disclosure. 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. Common. 2011. 2(319):DOI:10.1038 / ncomms1324). This preference is explained by structural studies of ADAR2 bound to transition state analogcontaining 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. 2016). Earlier work with fusion proteins bearing ADAR deaminase domains indicated that editing efficiency at 5’-GA sites could beimproved with a G-A or G-G pair at the nearest 5’ 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 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. Int. Patent Application Publication No. WO2024 / 013361 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.

[0062] The term ‘orphan nucleotide’ relates to the nucleotide in the AON that is directly opposite the target adenosine, which is the adenosine that is deaminated by the deaminating enzyme. The orphan nucleotide may be a natural cytidine, a deoxycytidine, a uridine, or a deoxyuridine. It may also be a chemically modified nucleotide, as further described in detail below, or a known or chemically modified analog of a natural (deoxy)cytidine, such as a nucleotide carrying a Benner’s base, or a known or chemically modified analog of a natural (deoxy)uridine, such as iso-uridine, as further outlined in detail below.

[0063] 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.

[0064] The term ‘downstream’ in relation to a nucleic acid sequence means further along the sequence in the 3' direction; the term ‘upstream’ means the converse. Thus, in any sequence encoding a polypeptide, the start codon is upstream of the stop codon in the sense strand but is downstream of the stop codon in the antisense strand. The same holds true for the AONs as disclosed herein. Nucleotides that are upstream of the orphan nucleotide in the antisense oligonucleotide are located towards the 5’ terminus, and nucleotides that are downstream of the orphan nucleotide are located towards the 3’ terminus.

[0065] The nucleotide ‘numbering’ in an AON as disclosed herein is such that the orphan nucleotide is number 0 and the nucleotide 5’ from the orphan nucleotide is number +1. Counting is further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, wherein the first nucleotide 3’ from the orphan nucleotideis number -1. The internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end.

[0066] 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.

[0067] The term ‘splice mutation’ relates to a mutation in a gene that encodes fora pre-mRNA, wherein the splicing machinery is dysfunctional in the sense that splicing of introns from exons is disturbed and due to the aberrant splicing, the subsequent translation is out of frame resulting in premature termination of the encoded protein. Often such shortened proteins degrade rapidly and do not have any functional activity.

[0068] Whenever a ‘naked’ form in relation to the AON as disclosed herein is referred to, it means that the AON is manufactured in a laboratory or manufacturing facility, through which it is generally chemically modified to prevent it from rapid degradation after it enters the mammalian body or a tissue, or cell, upon administration. The naked form of an AON is therefore different from a form in which the AON is encoded (and delivered) by a viral genome or within a plasmid vector. When such viral vectors or plasmid vectors are administered, the encoded AON is expressed from the viral vector genome or from the plasmid in the cell to which the viral vector or plasmid vector is delivered. Consequently, the AON is then not chemically modified and comprises solely naturally occurring RNA nucleotides.

[0069] The length of the AON as disclosed herein, and when delivered in a naked form is preferably 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. However, when the AON as disclosed herein is to be delivered through the expression of a viral vector, then the AON may be longer, such as 70, 80, 90, 100, 150, or 200 or more nucleotides in length.

[0070] The term ‘HEON’ refers to a heteroduplex double-stranded complex molecule wherein an AON as disclosed herein is hybridized to a partially or fully complementary, partially of fully overlapping sense oligonucleotide. Because the AON as disclosed herein often has specified chemical modifications that are different from the chemical modifications in the sense strand, the two strands form such a heteroduplex RNA editing oligonucleotide complex. The sense strand may be chemically modified almost in its entirety, similar ordifferent to what is performed in the AON as disclosed herein, for example by providing nucleotides with a ribose sugar moiety carrying a 2’-0Me substitution, a 2’-F substitution, or a 2’-M0E substitution. It is to be understood that the sense strand present in the HEON is a different entity in comparison to the target RNA molecule in the cell. The sense strand in an HEON is preferably 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. The HEON is often generated in vitro and used as a delivery tool to protect the AON from degradation when administered to the cell. In other words, the HEON is preferably formed before the AON is administered to the cell. Preferred aspects of HEONs that may be used for AONs as disclosed herein are discussed in Int. Patent Application Publication No. W02024 / 084048.Embodiments

[0071] Disclosed here is an AON that is capable of recruiting an endogenous ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target RNA nucleic acid molecule is a transcript molecule of the human PEX1 gene encoding the variant PEX1- p.G843D protein. In one aspect, the transcript molecule is a pre-mRNA or an mRNA molecule. In one aspect, the PEX1 gene is mutated in comparison to the ‘wildtype’ version referenced by NM_000466, and the target adenosine is the adenosine in the GAU codon coding for aspartic acid (D) at position 843 of the PEX1 protein, and wherein the deamination of the adenosine changes the amino acid to a glycine (G), and wherein the deamination of the target adenosine results in a PEX1 ATPase enzyme that has a wildtype function. The invention relates to reversing the G843D variant to a D843G version that has wildtype functionality. In one embodiment, disclosed herein is an AON according to the disclosure, wherein the PEX1 gene, preferably based on NCBI Reference Sequence: NM_000466, comprises a variant selected from the group consisting of c.2528G>A (representing a protein р.Gly843Asp variant), c.3750G>A (representing a protein p.Trp1250Ter variant), с.1163G>A (representing a p.Trp388Ter variant), c.2516G>A (representing a protein p.Trp839Ter variant), and c.5G>A (representing a protein p.Trp2Ter variant).

[0072] In one aspect, the oligonucleotide as disclosed herein comprises or consists of the sequence of any one of SEQ ID NO: 135, 173, 123, 124, 125, 138, 122, 121 , and 136. In one aspect, an AON as disclosed herein comprises or consists of any of the sequences provided in FIG. 2, optionally including the chemical modifications (as outlined herein and as shownas preferred embodiments in FIG. 1) to the nucleobase, sugar moiety and / or linkage, or any combination thereof.

[0073] In one aspect, the orphan nucleotide in an AON as disclosed herein does not comprise a natural cytosine nucleobase. In one aspect, the orphan nucleotide in an AON as disclosed herein does not comprise a 2’-OMe substituted ribose. In one aspect, the orphan nucleotide in an AON as disclosed herein is a deoxycytidine or a deoxyuridine. In one aspect, the orphan nucleotide in an AON as disclosed herein is a cytidine analog or a uridine analog. In one aspect, a cytidine analog is a deoxynucleotide comprising a 6-amino-5-nitro-3-yl-2(1 H)- pyridone nucleobase. In one aspect, a uridine analog is a deoxynucleotide comprising an iso-uracil nucleobase.

[0074] In one aspect, the nucleotide numbering in an AON as disclosed herein is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein the first nucleotide 3’ from the orphan nucleotide (-1) is a nucleotide analog that can induce a syn conformation of the guanosine that is located opposite in the target sequence. Preferred nucleotide analogs that can be used for the -1 position in the AON as disclosed herein are listed in Int. Patent Application Publication No. WO2024 / 013361 and as used herein (above and below).

[0075] In one aspect, at least one nucleotide or nucleotide analog in the AON comprises a substitution at the 2' position of the ribose, wherein the substitution is selected from the group consisting of: H (DNA); OH (RNA); F; ara-F; diF; 2’-C-methyl’-2’-F; substituted or unsubstituted, linear or branched lower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, that may be interrupted by one or more heteroatoms; O-, S-, or N-alkyl; O-, S-, or N- alkenyl; O-, S-, or N-alkynyl; O-, S-, or N-allyl; O-alkyl-O-alkyl; methoxy; aminopropoxy; meth oxy ethoxy; dimethylamino oxyethoxy; and dimethylaminoethoxyethoxy; with the proviso that the nucleotide that is opposite the target adenosine does not comprise a 2’-O- methyl or a 2’-(2-methoxy)ethyl ribose modification when the nucleobase is cytosine. In one aspect, at least one nucleotide or nucleotide analog in the AON is an arabinonucleic acid. In one aspect, the nucleotide analog comprises a modified nucleobase. In one aspect, the nucleotide analog at position -1 in the AON comprises a modified purine nucleobase. In one aspect, this modified purine nucleobase comprises a 7-deaza-purine modification, preferably a 7-deaza-adenine modification, or a 3-deaza-purine modification, preferably a 3- deaza-adenine modification, or a 3,7-dideaza modification, preferably a 3,7-dideaza- adenine modification, preferably wherein the nucleotide is a deoxyribonucleotide. In one aspect, the modified purine nucleobase comprises a hydrogen bond donor at N1 that has apKa that is higher than 3.7 and lower than 9.5. In one aspect, the nucleotide analog at position -1 in the AON is selected from the group consisting of:

[0076] 7-deaza-2’-deoxyadenosine (7-deaza Ad);

[0077] 7-deaza-2’-adenosine (7-deaza A);

[0078] 7-deaza-2’-deoxy-2’-fluoroadenosine (7-deaza Af);

[0079] 7-deaza-2’-deoxy-2’-ara-fluoroadenosine;

[0080] 7-deaza-2’-deoxy-2’,2’-difluoroadenosine;

[0081] 3-deaza-2’-deoxyadenosine (3-deaza Ad);

[0082] 3-deaza-2’-adenosine (3-deaza A);

[0083] 3-deaza-2’-deoxy-2’-fluoroadenosine (3-deaza Af);

[0084] 3-deaza-2’-deoxy-2’-ara-fluoroadenosine;

[0085] 3-deaza-2’-deoxy-2’,2’-difluoroadenosine;

[0086] 3,7-dideaza-2’-deoxyadenosine (3,7-dideaza Ad);

[0087] 3,7-dideaza-2’-adenosine (3,7-dideaza A);

[0088] 3,7-dideaza-2’-deoxy-2’-fluoroadenosine (3,7-dideaza Af);

[0089] 3,7-dideaza-2’-deoxy-2’-ara-fluoroadenosine;

[0090] 3,7-dideaza-2’-deoxy-2’,2’-difluoroadenosine;

[0091] 3-deaza-2’-[2-(methoxy)ethyl] adenosine;

[0092] 3-deaza-2’-O-[2-methylamino-2-oxoethyl] adenosine;

[0093] 2’-deoxy-2’-fluoroguanosine;

[0094] 2’-ara-fluoro guanosine (FANA G);

[0095] 2’,2’-difluoro guanosine;

[0096] 2’-deoxyinosine;

[0097] 2’-OH-inosine;

[0098] 2’-fluoroinosine;

[0099] 2‘-ara-fluoro inosine (FANA I);

[0100] 2’,2’-difluoro inosine;

[0101] 5-formylindole-2’-deoxyriboside;

[0102] 5-formyl-2’-fluoro-2’-deoxyriboside

[0103] 5-formylindole-2’-ara-fluoro-2’-deoxyriboside;

[0104] 5-formylindole-2’,2’-difluoro-2’-deoxyriboside;

[0105] 5-formylindole-2’-O-methylriboside;

[0106] 5-formylindole-2’-O-[2-(methoxy)ethyl]riboside;

[0107] 5-formylindole-2’-O-[2-methylamino-2-oxoethyl]riboside

[0108] beta-(4-amidino-1 / 7-imidazol-1-yl) riboside;

[0109] beta-(4-amidino-1 / 7-imidazol-1-yl) 2’-deoxyriboside;

[0110] beta-(4-amidino-1 / 7-imidazol-1-yl) 2’-ara-fluoro-2’-deoxyriboside; and

[0111] beta-(4-amidino-1 / 7-imidazol-1-yl) 2’,2’-difluoro-2’-deoxyriboside.

[0112] Preferably, the analog at the -1 position in the AON is 7-deaza-2’-deoxyadenosine (7- deaza Ad), 3-deaza-2’-deoxyadenosine (3-deaza Ad), or 2’-deoxyinosine (Id).

[0113] In one aspect, the nucleotide opposite the 5’-G in the target sequence comprises a base moiety with the structure of formula (I), (II), (III), (IV), (V), or (VI):wherein R is:- H;an electron-donating moiety;- -C(=O)H- -OR1;-SR1; or- -NR1R2 wherein R1 is (Ci-Ce)-alkoxy, methoxy, ethoxy, isopropoxy, cyclopropoxy, (Ci-Ce)- alkyl, methyl, ethyl, isopropyl, or cyclopropyl, and R2 is H, (Ci-Ce)-alkoxy, methoxy, ethoxy, isopropoxy, cyclopropoxy, (Ci-Ce)-alkyl, methyl, ethyl, isopropyl, or cyclopropyl.

[0114] In an embodiment, the nucleotide opposite the 5’-G in the target sequence comprises a base structure according to formula (VII) or (VIII):wherein:X = CH, CR6or N;Ri = H, OH, halogen, SH, (Ci-C3)-alkoxy, or NH2;R2 = H, OH, NH2, methyl, ethyl, or cyclopropyl;R3 = H, OH, NH2, methyl, ethyl, cyclopropyl, or (Ci-C3)-alkoxy;R4 = H, OH, NH2, methyl, ethyl, cyclopropyl, or (Ci-C3)-alkoxy;Rs = H, OH, NH2, methyl, ethyl, cyclopropyl, or (Ci-C3)-alkoxy; andRe = an electron-donating moiety, methyl, ethyl cyclopropyl, or (Ci-C3)-alkoxy.

[0115] In one aspect, the AON as disclosed herein is 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length. In one aspect, an AON as disclosed herein is in a naked form. In one aspect, an AON as disclosed herein is in a circular format.

[0116] In one aspect, an AON as disclosed herein is not in a naked form but is expressed from the genome of a viral vector.

[0117] In one aspect, an AON as disclosed herein is not in a naked form but is expressed from an expression vector such as a plasmid.

[0118] In one aspect, when the AON as disclosed herein is not in a naked form, the AON is 15 to 60 nucleotides in length as indicated above, or in another embodiment, from 61 to 300 nucleotides in length.

[0119] In one aspect, an AON as disclosed herein comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP), sulfonylphosphoramidate, PNdmi internucleotide linkage, or a linkage moiety with the structure according to formula (IX):wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group.

[0120] Disclosed herein is an AON, wherein the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein linkage position -2 is an MP linkage, or a PNms linkage. In one aspect, the linkage between the most terminal two nucleotides on the 5’ and / or 3’ terminus of the AON as disclosed herein is a PNdmi linkage, a MP linkage, or a PNms linkage, or a combination thereof in respect of the 5’ and 3’ terminus. Preferred are PNms linkages.

[0121] Disclosed herein is an AON, wherein the AON comprises one or more nucleotides comprising a mono- or di-substitution at the 2', 3' and / or 5' position of the ribose, each independently selected from the group consisting of: -OH; -F; substituted or unsubstituted, linear or branched lower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, that may be interrupted by one or more heteroatoms; -O-, S-, or N-alkyl; -O-, S-, or N-alkenyl; -O-, S-, orN-alkynyl; -O-, S-, or N-allyl; -O-alkyl-O-alkyl; -methoxy; -aminopropoxy; -meth oxy ethoxy; - dimethylamino oxyethoxy; and -dimethylaminoethoxyethoxy.

[0122] In one preferred aspect, the AON as disclosed herein is selected from the group consisting of SEQ ID NO: 434, 435, 427, 442, 401 , 431 , 430, 432, 433, 421 , 358, 359, 355, 400, 399, 135, 429, 444, 439, 443, 420, 423, 331 , 335, 344, 226, 222, 208, 212, 213, 216, 217, 218, 219, 220, 225, 227, 234, 235, 236, 237, 238, 240, 597, 608, 610, 605, 779, 607, 617, 591 , 722, 549, and 241.

[0123] Disclosed herein is a vector, preferably a viral vector, more preferably an adeno- associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON as disclosed herein.

[0124] Disclosed herein is a pharmaceutical composition comprising an AON as disclosed herein, or a vector as disclosed herein, and a pharmaceutically acceptable carrier.

[0125] Disclosed herein is an AON for use in the treatment of PBD-ZSD caused by PEX1- P.G843D.

[0126] Disclosed herein is a use of an AON as disclosed herein, for use in the manufacture of a medicament for the treatment of PBD-ZSD caused by PEX1-p.G843D.

[0127] Disclosed herein is a method of editing a human PEX1-c.2528G>A polynucleotide in a cell, wherein the human PEX1-c.2528G>A polynucleotide is a pre-mRNA or mRNA molecule, the method comprising contacting the PEX1-c.2528G>A polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, thereby editing the PEX1-c.2528G>A polynucleotide to encode a PEX1 protein with a wildtype functionality.

[0128] Disclosed is a method of treating, ameliorating, or slowing down PBD-ZSD caused by PEX1-p.G843D in a human subject in need thereof, the method comprising administering to said subject an AON as disclosed herein, or a vector as disclosed herein, or a pharmaceutical composition as disclosed herein, thereby contacting a PEX1 polynucleotide, preferably a pre-mRNA or an mRNA transcript molecule, such as transcribed from a mutant PEX7-c.2528G>A gene, in a cell of the subject with an AON capable of effecting an ADAR- mediated adenosine to inosine deamination, thereby editing the PEX1 polynucleotide to encode a PEX1 protein with a wildtype functionality.

[0129] Disclosed herein is an in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in a human PEX1 pre-mRNA or mRNA molecule in a cell, the method comprising the steps of: (i) providing the cell with an AON as disclosed herein; (ii) allowing uptake by the cell of the AON; (iii) allowing annealing of the AON to the PEX1 transcribed pre-mRNA or mRNA target molecule; and (iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target RNA molecule to an inosine. In one aspect, the in vivo methodas disclosed herein comprises step (v) of using a functional read-out to identify the presence of the inosine in the target RNA molecule. A preferred read-out is measuring VLCFA levels in plasma samples from a treated subject, like what is used in diagnosis of the disease. Other bodily fluid available biomarkers known to the person skilled in the art, especially in the art of diagnosing and treating PBD-ZSD, can be used. Examples are pipecolic acid and BA intermediates in urine / blood, reduced plasmalogen levels in RBC membranes and phytanic and / or pristanic acid in blood.Chemical modifications

[0130] Various chemistries and modifications are known in the field of oligonucleotides that can be readily used in accordance with the invention. All chemical modifications listed herein that may be used in the AON as disclosed herein may also be used for a sense strand that is complementary to the AON, when the AON and the complementary strand form a HEON complex, such as described in I nt. Patent Application Publication No. W02024 / 084048, except that the opposite sense strand does not have an orphan nucleotide. Hence, the modification related to the orphan nucleotide relate only to the AON as disclosed herein, but all other modifications relate to the AON as disclosed herein and any (protecting) sense oligonucleotide that may be used together with the AON in a pharmaceutical product. This includes the use of hydrophobic moieties (such as tocopherol and cholesterol) and cellspecific ligands, that have also been described herein, and in detail in Int. Patent Application Publication No. W02024 / 084048, which may either be bound to the AON or its opposite strand, or both.

[0131] The skilled person knows that an oligonucleotide, such as an AON as outlined herein, generally consists of repeating monomers. Such a monomer is most often a nucleotide or a chemically modified nucleotide. The most common naturally occurring nucleotides in RNA are adenosine monophosphate (A), cytidine monophosphate (C), guanosine monophosphate (G), and uridine monophosphate (II). These consist of a pentose sugar, a ribose, a 5’-linked phosphate group which is linked via a phosphate ester, and a T-linked base. The sugar connects the base and the phosphate and is therefore often referred to as the “scaffold” of the nucleotide. A modification in the pentose sugar is therefore often referred to as a ‘scaffold modification’. The original pentose sugar may be replaced in its entirety by another moiety that similarly connects the base and the phosphate. It is therefore understood that while a pentose sugar is often a scaffold, a scaffold is not necessarily a pentose sugar. Examples of scaffold modifications that may be applied in the monomers of the AON as disclosed herein are disclosed in Int. Patent Application Publication Nos. W02020 / 154342, W02020 / 154343, and WO2020 / 154344.

[0132] A nucleoside in the AON as disclosed herein may be a natural nucleoside (deoxyribonucleoside or ribonucleoside) or a non-natural nucleoside. It is noted that for RNA editing, in which double-stranded RNA is generally the substrate for enzymes with deamination activity (such as ADARs), ribonucleosides are considered ‘natural’, while deoxyribonucleosides may then be, for the sake of argument, considered as non-natural, or considered as modified, simply because DNA is not present in the RNA-RNA double stranded (natural) substrate configurations. The skilled person appreciates that when the nucleotide has a natural ribose moiety, it may still be non-naturally modified in the base and / or the linkage.

[0133] It is recognized in the art that common limiting factors in oligonucleotide-based therapies are the oligonucleotide’s ability to be taken up by the cell (when delivered per se, or ‘naked’ without applying a delivery vehicle such as a viral vector or plasmid), the biodistribution and the resistance to nuclease-mediated breakdown. The skilled person is aware, and it has been described in detail in the art, that a variety of chemical modifications can assist in overcoming such limitations. Examples of such now commonly used chemical modifications are the 2’-OMe, 2’-F and 2’-MOE modifications of the sugar and the use of PS linkages between nucleosides, as described herein.Scaffold modifications (ribose)

[0134] The ribose 2’ groups in all nucleotides of the AON as disclosed herein, except for the ribose sugar moiety of the orphan nucleotide that has certain limitations in respect of compatibility with RNA editing, can be independently selected from 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-OMe, 2’-MOE, 2’-F, or 2’-4’-linked (for instance a locked nucleic acid (LNA)), or other ribosyl T-substitutions, 2’ substitutions, 3’ substitutions, 4’ substitutions or 5’ substitutions. The orphan nucleotide in the AON that comprises no other chemical modifications to the ribose sugar, the base, or the linkage preferably does not carry a 2’- OMe or 2’-MOE substitution when the nucleobase is a naturally occurring cytosine, but may carry a 2’-F, a 2’,2’-difluoro (diF), or 2’-ara-F (FANA) substitution or may be DNA. Int. Patent Application Publication No. W02024 / 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).

[0135] An AON as disclosed herein may comprise one or more nucleotides carrying a 2’-MOE ribose modification. Also, an AON as disclosed herein may comprise one or more nucleotides not carrying a 2’-MOE ribose modification, or wherein the 2’-MOE ribosemodifications are at positions that do not prevent the enzyme with adenosine deaminase activity from deaminating the target adenosine. An AON as disclosed herein may comprise a 2’-0Me ribose modification at a position that does not comprise a 2’-MOE ribose modification. An AON as disclosed herein may comprise deoxynucleotides at positions that do not comprise a 2’-MOE or a 2’-0Me ribose modification, or other 2’ ribose substitution.

[0136] An AON as disclosed herein may comprise one or more nucleotides comprising a 2’ substitution comprising a 2’-MOE, 2’-OMe, 2’-OH, 2’-deoxy, TNA, 2’-fluoro (2’-F), 2’, 2’- difluoro (diF) modification, 2’-fluoro-2’-C-methyl modification, or a 2’-4’-linkage (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA or examples mentioned in e.g., Int. Patent Application Publication No. WO2018 / 007475)). Other nucleic acid monomers that may be used in an AON as disclosed herein are arabinonucleic acids and 2’-deoxy-2’- fluoroarabinonucleic acid (FANA), for instance for improved affinity purposes. The 2’-4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker. A wide variety of 2’ modifications that may present in an AON as disclosed herein are known in the art, including but not limited to the modifications outlined in detail in Int. Patent Application Publication Nos. WO2016 / 097212, WO2017 / 220751 , WO201 8 / 041973, WO2018 / 134301 , WO2019 / 219581 , WO2019 / 158475, andWO2022 / 099159. In all cases, the modifications should be compatible with RNA editing such that the AON fulfils its role as an oligonucleotide that can form a double stranded complex with the target RNA and by generating this double-stranded nucleic acid complex, recruit a deaminating enzyme, which can subsequently deaminate the target adenosine. Where a monomer in an AON as disclosed herein comprises an unlocked nucleic acid (UNA) ribose modification, that monomer can have a 2’ position comprising the same modifications discussed above, such as a 2’-MOE, a 2’-OMe, a 2’-OH, a 2’-deoxy, a 2’-F, a 2’,2’-diF, a 2’- fluoro-2’-C-methyl, an arabinonucleic acid, a FANA, or a 2’-4’-linkage (i.e., a bridged nucleic acids such as a locked nucleic acid (LNA)).

[0137] In an aspect, the AON as disclosed herein comprises at least one nucleotide comprising a threose nucleic acid (TNA) ribose modification. In one aspect, the AON as disclosed herein comprises at least one nucleotide with a sugar moiety that comprises a 2’- fluoro (2’-F) modification. A preferred position for the nucleotide that carries a 2’-F modification is position -3 in the AON, which may be present together with an identical 2’ modification in the orphan nucleotide as discussed above.Base modifications

[0138] 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 bondto another nucleobase through H-bonds, polarized bonds (such as through CF moieties) or aromatic electronic interactions. Cytosine, thymine, and uracil are pyrimidine bases, and are generally linked to the scaffold through their 1 -nitrogen. Adenine and guanine are purine bases and are generally linked to the scaffold through their 9-nitrogen. The terms ‘adenine’, ‘guanine’, ‘cytosine’, ‘thymine’, ‘uracil’ and ‘hypoxanthine’ as used herein refer to the nucleobases as such. The terms ‘adenosine’, ‘guanosine’, ‘cytidine’, ‘thymidine’, ‘uridine’ and ‘inosine’ refer to the nucleobases linked to the (deoxy)ribosyl sugar. The nucleobases in an AON as disclosed herein can be adenine, cytosine, guanine, thymine, or uracil or any other moiety able to interact with another nucleobase through H-bonds, polarized bonds (such as CF) or aromatic electronic interactions. The nucleobases at any position in the AON as disclosed herein can be a modified form of adenine, cytosine, guanine, or uracil, such as hypoxanthine (the nucleobase in inosine), pseudouracil, pseudocytosine, isouracil, N3- glycosylated uracil, 1 -methylpseudouracil, orotic acid, agmatidine, lysidine, 2-thiouracil, 2- thiothymine, 5-substituted pyrimidine (e.g., 5-halouracil, 5-halomethyluracil, 5- trifluoromethyluracil, 5-propynyluracil, 5-propynylcytosine, 5-aminomethyluracil, 5- hydroxymethyluracil, 5-formyl uracil, 5-aminomethylcytosine, 5-formylcytosine), 5- hydroxymethylcytosine, 7-deazaguanine, 7-deazaadenine, 7-deaza-2,6-diaminopurine, 8- aza-7-deazaguanine, 8-aza-7-deazaadenine, 8-aza-7-deaza-2,6-diaminopurine, 8-oxo- adenine, 3-deazapurine (such as a 3-deaza-adenosine), pseudoisocytosine, N4- ethylcytosine, N2-cyclopentylguanine, N2-cyclopentyl-2-aminopurine, N2-propyl-2- aminopurine, 2,6-diaminopurine, 2-aminopurine, G-clamp and its derivatives, Super A, Super T, Super G, amino-modified nucleobases or derivatives thereof; and degenerate or universal bases, like 2,6-difluorotoluene, or absent like abasic sites (e.g. 1 -deoxyribose, 1 ,2- dideoxyribose, 1-deoxy-2-O-methylribose, azaribose). Modified bases comprise synthetic and natural bases such as inosine, xanthine, hypoxanthine and other -aza, deaza, -hydroxy, -halo, -thio, thiol, -alkyl, -alkenyl, -alkynyl, thioalkyl derivatives of pyrimidine and purine bases that are or will be known in the art. Purine nucleobases and / or pyrimidine nucleobases may be modified to alter their properties, for example by amination or deamination of the heterocyclic rings. The exact chemistry and formats may vary from oligonucleotide construct to oligonucleotide construct and from application to application, and may be worked out in accordance with the wishes and preferences of those of skill in the art.

[0139] 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’-0-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’-0- alkoxycarbonyl such as 2’-O-[2-(methoxycarbonyl)ethyl] (MOCE), 2’-O-[2-N- methylcarbamoyl)ethyl] (MCE), 2’-O-[2-( / V, / \ / -dimethylcarbamoyl)ethyl] (DCME); 2’-halo e.g. 2’-F, FANA; 2'-O-[2-(methylamino)-2-oxoethyl] (NMA); a bicyclic or bridged nucleic acid (BNA) scaffold modification such as a conformationally restricted nucleotide (CRN) monomer, a locked nucleic acid (LNA) monomer, a xy / o-LNA monomer, an a-LNA monomer, an a-l-LNA monomer, a p-d-LNA monomer, a 2’-amino-LNA monomer, a 2’-(alkylamino)- LNA monomer, a 2’-(acylamino)-LNA monomer, a 2’- / V-substituted 2’-amino-LNA monomer, a 2’-thio-LNA monomer, a (2’-O,4’-C) constrained ethyl (cEt) BNA monomer, a (2’-O,4’-C) constrained methoxyethyl (cMOE) BNA monomer, a 2’,4’-BNANC(NH) monomer, a 2’,4’- BNANC(NMe) monomer, a 2’,4’-BNANC(NBn) monomer, an ethylene-bridged nucleic acid (ENA) monomer, a carba-LNA (cLNA) monomer, a 3,4-dihydro-2 / 7-pyran nucleic acid (DpNA) monomer, a 2’-C-bridged bicyclic nucleotide (CBBN) monomer, an oxo-CBBN monomer, a heterocyclic-bridged BNA monomer (such as triazolyl or tetrazolyl-linked), an amido-bridged BNA monomer (such as AmNA), an urea-bridged BNA monomer, a sulfonamide-bridged BNA monomer, a bicyclic carbocyclic nucleotide monomer, a TriNA monomer, an a-l-TriNA monomer, a bicyclo DNA (bcDNA) monomer, an F-bcDNA monomer, a tricyclo DNA (tcDNA) monomer, an F-tcDNA monomer, an alpha anomeric bicyclo DNA (abcDNA) monomer, an oxetane nucleotide monomer, a locked PMO monomer derived from 2’-amino LNA, a guanidine-bridged nucleic acid (GuNA) monomer, a spirocyclopropylene-bridged nucleic acid (scpBNA) monomer, and derivatives thereof; cyclohexenyl nucleic acid (CeNA) monomer, altriol nucleic acid (ANA) monomer, hexitol nucleic acid (HNA) monomer, fluorinated HNA (F-HNA) monomer, pyranosyl-RNA (p-RNA) monomer, 3’-deoxypyranosyl DNA (p-DNA), unlocked nucleic acid UNA); an inverted version of any of the monomers above. All these modifications are known to the person skilled in the art.The orphan nucleotide

[0140] 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’ asthe case may be), as indicated above. The crystal structure of ADAR2 E488Q bound to double stranded RNA (dsRNA) revealed that the glutamine (Gin; Q) side chain at position 488 can donate an H-bond to the N3 position of the orphan cytidine, which leads to the increased catalytic rate of ADAR2 E488Q. In the wild-type enzyme, wherein a glutamate (or glutamic acid; Glu; E) is present at position 488 instead of a glutamine (Gin) the amide group of the glutamine is absent and is instead a carboxylic acid. To obtain the same contact of the orphan cytidine with the E488Q mutant would then, for the wild-type situation, require protonation for this contact to occur. To make use of endogenously expressed ADAR2 to correct disease relevant mutations, it is essential to maximize the editing efficiency of the wild type ADAR2 enzyme present in the cell. Int. Patent Application Publication No. WO2020 / 252376 discloses the use of AONs with modified RNA bases, especially at the position of the orphan cytidine to mimic the hydrogen-bonding pattern observed by the E488Q ADAR2 mutant. By replacing the nucleotide opposite the target adenosine in the AON with cytidine analogs that serve as H-bond donors at N3, it was envisioned that it would be possible to stabilize the same contact that is believed to provide the increase in catalytic rate for the mutant enzyme. Two cytidine analogs were of particular interest: pseudoisocytidine (also referred to as ‘piC’; Lu et al. J Org Chem 2009. 74(21):8021-8030; Burchenal et al. (1976) Cancer Res 36:1520-1523) and Benner’s base Z (also referred to as ‘dZ’ or ‘Zd’; Yang et al. Nucl Acid Res 2006. 34(21):6095-6101) that were initially selected because they offer hydrogen-bond donation at N3 with minimal perturbation to the shape of the nucleobase. Benner’s base is also referred to as 6-amino-5-nitro-3-yl-2(1 H)-pyridone.

[0141] The presence of the cytidine analog in the AON may exist in addition to modifications to the ribose 2’ group. The ribose 2’ groups in the orphan nucleotide can be independently selected from 2’-H (i.e., DNA), 2’-OH (i.e., RNA), 2’-OMe, 2’-MOE, 2’-F, or 2’-4’-linked (i.e., a bridged nucleic acid such as a locked nucleic acid (LNA)), or other 2’ substitutions. The 2’- 4’ linkage can be selected from linkers known in the art, such as a methylene linker or constrained ethyl linker.

[0142] Another preferred cytidine analog is a deoxyribonucleoside comprising a cytosine analog nucleobase according to formula (X):or any of its tautomeric forms, wherein: R1, R2, R4and / or R5is H; OH; SH; =0; NH2; a halogen; a linear or branched lower (C1-C10) alkyl; or a Ci-Ce cycloalkyl, wherein the alkyl and / or cycloalkyl is optionally interrupted by one or more heteroatoms; and R3is H; OH; SH; =0; NH2; or a halogen. When all of 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-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 enolamino and enol-imino tautomers. Examples of keto tautomers of the E base are ketoamino and keto-imino tautomers.

[0143] The field of nucleoside analogs has long been instrumental in developing therapeutics targeting viral infections and epigenetic modifications. Among these, 5-aza-5,6-di hydro cytosine represents a structurally novel cytosine (= nucleobase) analog with distinct chemical and biological properties. Unlike as observed in conventional nucleosides, this base features a saturated 5,6 bond, significantly altering its tautomeric equilibrium and basepairing 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 basepairing promiscuity, which underlies its mutagenic effects (Li et al. 2014).

[0144] 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; Matousova 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 cytidinehas 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 oral formulations, which would overcome the limitations of current hypomethylating agents requiring parenteral delivery (Matousova et al. 2011).

[0145] The orphan nucleotide in the AON as disclosed herein is preferably a cytidine or analog thereof (such as a nucleotide carrying a Benner’s base or an E-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 an embodiment at least one and in another embodiment both the neighbouring (directly adjacent) nucleotides flanking the orphan nucleotide do not comprise a 2’-OMe modification. Complete modification wherein all nucleotides of the oligonucleotide hold a 2’-OMe modification (including the orphan nucleotide), with natural bases, results in a non-functional oligonucleotide as far as RNA editing goes (known in the art), presumably because it hinders the ADAR activity at the targeted position. In general, an adenosine in a target RNA can be protected from editing by providing an opposing nucleotide with a 2'-OMe group (at least when there are no other chemical substitutions or modifications within the nucleotide), or by providing a guanine or adenine as opposing base, as these two nucleobases are also able to reduce editing of the opposing adenosine.Linkage modifications

[0146] A nucleoside is generally connected to neighboring nucleosides through condensation of its 5’-phosphate moiety to the 3’-hydroxyl moiety of the neighboring nucleotide monomer. Similarly, its 3’-hydroxyl moiety is generally connected to the 5’-phosphate of a neighboring nucleotide monomer. This forms phosphodiester bonds. The phosphodiesters and the scaffold form an alternating copolymer. The bases are grafted on this copolymer, namely to the scaffold moieties. Because of this characteristic, the alternating copolymer formed by linked scaffolds of an oligonucleotide is often called the ‘backbone’ of the oligonucleotide. Because phosphodiester bonds connect neighboring monomers together, they are often referred to as ‘backbone linkages’. It is understood that when a phosphate group is modified so that it is instead an analogous moiety such as a phosphorothioate, such a moiety is still referred to as the backbone linkage of the monomer. This is referred to as a ‘backbonelinkage modification’. In general terms, the backbone of an oligonucleotide comprises alternating scaffolds and backbone linkages.

[0147] As outlined in detail herein, naked AONs as disclosed herein comprise at least one, preferably multiple linkage modifications. It is generally preferred that the AON as disclosed herein comprises linkage modifications at most, and potentially all positions if the AON is capable of mediating RNA editing through the deamination enzyme when the AON is bound to the target RNA nucleic acid molecule. A linkage modification can be, but is not limited to, a modified version of the phosphodiester present in RNA, such as phosphorothioate (PS), chirally pure PS, (R)-PS, (S)-PS, methyl phosphonate (MP or MeP), chirally pure MP, (R)- MP, (S)-MP, phosphoryl guanidine (such as PNdmi), chirally pure phosphoryl guanidine, (R)-phosphoryl guanidine, (S)-phosphoryl guanidine, phosphorodithioate (PS2), phosphonacetate (PACE), phosphonoacetamide (PACA), thiophosphonoacetate, thiophosphonoacetamide, methyl phosphorohioate, methyl thiophosphonate, PS prodrug, alkylated PS, H-phosphonate, ethyl phosphate, ethyl PS, boranophosphate, borano PS, metyl boranophosphate, methyl borano PS, methyl boranophosphonate, methyl boranophosphothioate, phosphate, phosphotriester, aminoalkylphosphotriester, and their derivatives. Another modification includes phosphoramidite, phosphoramidate, N3’^P5’ phosphoramidate, phosphorodiamidate, phosphorothiodiamidate, sulfamate, diethylenesulfoxide, amide, sulfonate, siloxane, sulfide, sulfone, formacetyl, alkenyl, methylenehydrazino, sulfonamide, triazole, oxalyl, carbamate, methyleneimino (MM I), and thioacetamide nucleic acid (TANA); and their derivatives. Various salts, mixed salts, deprotonated, protonated, tautomeric, and free acid forms are also included, as well as 3’->3’ and 2’->5’ linkages. An AON as disclosed herein may also comprise one or more linkage modifications according to the structure of formula (IX), wherein: X = O or S; and R = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl and the linkage modification is referred to as mesyl phosphoramidate, MsPA or PNms.

[0148] In a preferred aspect, the AON as disclosed herein comprises an internucleoside linkage of the structure of formula (IX), wherein X = O and R = CH3, which linkage is generally referred to herein as a PNms linkage (mesyl phosphoramidate). In other preferred aspects, R equals one of the following structures (a), (b), (c), (d), (e), (f), (g), (h), or (i):

[0149] The one or more PN linkages as depicted in formula (IX), present in an AON as disclosed herein, can be independently of each other of R or SP chirality, or stereo random.

[0150] The one or more PN linkages as depicted in formula (IX), in an AON as disclosed herein, can be in tautomeric and / or pH-dependent (de)protonated form, including but not limited to the structures (A), (B), (C), (D), and (E):wherein X and R are as indicated above for formula (IX).

[0151] Disclosed herein is also an AON that is able to mediate adenosine deamination by recruitment of a deaminating enzyme in a cell after the AON has formed a double-stranded complex with a region of a target RNA nucleic acid molecule in a cell, wherein the region comprises a target adenosine, wherein the deaminating enzyme can deaminate the target adenosine into an inosine, and wherein the AON comprises a moiety at one and / or both termini with a structure according to formula (XI):wherein: X = O or S;Y = O' or S'; andR = an aryl, a substituted aryl, a heterocycle, a substituted heterocycle, an aromatic heterocycle, a substituted aromatic heterocycle, a Ci-Ce alkoxy, a substituted Ci-Ce alkoxy, a C1-C20 alkyl, a substituted C1-C20 alkyl, a Ci-Ce alkenyl, a Ci-Ce substituted alkenyl, a Ci-Ce alkynyl, a substituted Ci-Ce alkynyl, or a conjugate group. In a preferred embodiment, X = O and R = methyl.

[0152] An AON as disclosed herein may comprise a substitution of one of the non-bridging oxygens in the phosphodiester linkage. This modification slightly destabilizes base pairing but adds significant resistance to nuclease degradation. A preferred nucleotide analogue or equivalent comprises PS, phosphonoacetate, phosphorodithioate, phosphotriester, aminoalkylphosphotriester, H-phosphonate, methyl and other alkyl phosphonate including 3'-alkylene phosphonate, 5'-alkylene phosphonate and chiral phosphonate, phosphinate, phosphoramidate including 3'-amino phosphoramidate and aminoalkylphosphoramidate, thionophosphoramidate, thionoalkylphosphonate, thionoalkylphosphotriester, selenophosphate or boranophosphate. Particularly preferred are internucleoside linkages that are modified to contain a PS. Particularly preferred are internucleoside linkages that are modified to contain a PNms. Particularly preferred are internucleoside linkages that are modified to contain a PNdmi. The regular internucleosidic linkages between the nucleotides may be altered by mono- or di-thioation of the phosphodiester bonds to yield PS esters or phosphorodithioate esters, respectively. Other modifications of the internucleosidic linkages are possible, including amidation and peptide linkers. The skilled person can determine for what target RNA nucleic acid molecule the AON comprises a certain linkage modification ateach linkage position of the AON as disclosed herein to generate the most effective and stable oligonucleotide compound.

[0153] Many of the non-naturally occurring modifications of the linkage, such as PS, are chiral. This means that there are Rp and Sp configurations, known to the person skilled in the art. In one embodiment, the chirality of the PS linkages is controlled, which means that each of the linkages is either in the Rp or in the Sp configuration, whichever is preferred. The choice of an Rp or Sp configuration at a specified linkage position may depend on the target sequence and the efficiency of binding and induction of causing RNA editing of the target adenosine. However, if such is not specifically desired, a composition may comprise AONs as active compounds with both Rp and Sp configurations at a certain specified linkage position. Mixtures of such AONs are also feasible, wherein certain positions preferably have either one of the configurations, while for other positions such does not matter. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PS linkages. In one aspect, the AON as disclosed herein comprises one of more (chirally pure or chirally mixed) phosphoramidate (PN) linkages. In one aspect, the AON as disclosed herein comprises one or more (chirally pure or chirally mixed) PNms linkages. In one aspect, a PN linkage connects the terminal two nucleotides on each end of the AON. AONs as disclosed herein may also comprise linkage modifications at all positions that are not chirally controlled. The AON as disclosed herein may also comprise one or more naturally occurring internucleoside linkages. The choice and number of modified linkages may depend on the specific target, the sequence, the length, and the stability of the AON observed in a particular cell type of interest, which can be assessed by methods known to the person skilled in the art. In one aspect, at least one, at least two, at least three, or at least four internucleoside linkages between the 5’ and / or the 3’ terminal two, three, four, or five nucleosides respectively of the AON as disclosed herein are modified internucleoside linkages. In one aspect, the AON as disclosed herein comprises at least one MP internucleoside linkage according to the structure of formula (XII):

[0154] As was noted in the art, a preferred position for an MP linkage in an AON is linkage position -2, thereby connecting the nucleoside at position -1 with the nucleoside at position-2. Int. Patent Application Publication No. W02024 / 200278 describes that PNms at this position renders the AON more stable than when a MP linkage is present. Hence, in a preferred embodiment, this -2 position, in an AON as disclosed herein, comprises a linkage modification according to the structure of formula (XI), but more preferably a linkage modification according to the structure of formula (IX), such as a PNms linkage, instead of an MP linkage. Int. Patent Application Publication No. W02020 / 201406 discloses the use of MP linkage modifications at certain positions surrounding the orphan nucleotide in the first nucleic acid strand. Although the presence of MP linkages is compatible with RNA editing by human ADAR enzymes, introducing MP linkages during the manufacturing of oligonucleotides is also challenging in view of additional manufacturing (purification) steps in the coupling and decoupling process. Hence, in one aspect, the AON does not comprise an MP linkage, and the linkage position at -2 is a PNms linkage instead.

[0155] In one aspect, the AON as disclosed herein comprises at least one PNdmi linkage, preferably linking the most terminal two nucleosides at the 5’ and / or 3’ end of the AON. A PNdmi linkage as preferably used in the AONs as disclosed herein has the structure of formula (XIII):PNdmi linkage (XIII)

[0156] Other internucleoside linkages that may be used in the AONs as disclosed herein are those that are disclosed in Int. Patent Application Publication No. WO2023 / 278589. In one aspect, the AON as disclosed herein comprises at least one phosphonoacetate and / or at least one phosphonoacetamide internucleoside linkage.Conjugate chemistries

[0157] In one aspect, the AON as disclosed herein, or the sense strand to which it may be annealed before entering a target cell (in an HEON as disclosed herein), is bound to a hydrophobic moiety, such as palmityl or an analog thereof, cholesterol or analog thereof, or tocopherol or analog thereof. It is preferably bound to the 5’ terminus. In case a hydrophobic moiety is bound to the 5’ terminus as well as to the 3’ terminus, such hydrophobic moieties may the same or different. The hydrophobic moiety bound to the oligonucleotide may bebound directly or indirectly mediated by another substance. When the hydrophobic moiety is bound directly, it is sufficient if the moiety is bound via a covalent bond, an ionic bond, a hydrogen bond, or the like. When the hydrophobic moiety is bound indirectly, it may be bound via a linking group (a linker). The linker may be a cleavable or an uncleavable linker. A cleavable linker refers to a linker that can be cleaved under physiological conditions, for example, in a cell or an animal body (e.g., a human body). A cleavable linker is selectively cleaved by an endogenous enzyme such as a nuclease, or by physiological circumstances specific to parts of the body or cell, such as pH or reducing environment (such as glutathione concentrations). Examples of a cleavable linker comprise, but is not limited to, an amide, an ester, one or both esters of a phosphodiester, a phosphoester, a carbamate, and a disulfide bond, as well as a natural DNA linker. Cleavable linkers also include self-immolative linkers. An uncleavable linker refers to a linker that is not cleaved under physiological conditions, or very slowly compared to a cleavable linker, for example, in a PS linkage, modified or unmodified deoxyribonucleosides linked by a PS linkage, a spacer connected through a PS bond and a linker consisting of modified or unmodified ribonucleosides. There is no restriction on the chain length, when a linker is a nucleic acid such as DNA, or an oligonucleotide. However, it may be usually from 2 to 20 bases in length, from 3 to 10 bases in length, or from 4 to 6 bases in length. There is no restriction on the length or composition of a spacer that is connects the ligand and the oligonucleotide, and may include for example ethylene glycol, triethylene glycol (TEG), HEG, alkyl chains, propyl, 6-aminohexyl, or dodecyl. One or more other types of molecules may be bound to the AON through one or more linkers, including peptides, sugars, vitamins, polymers, aptamers, (fragments of) antibodies, small molecules, and the like.General

[0158] In addition to the specific preferred chemical modifications at certain positions in compounds as disclosed herein, AONs as disclosed herein may comprise one or more (additional) modifications to the nucleobase, scaffold and / or backbone linkage, which may or may not be present in the same monomer, for instance at the 3’ and / or 5’ position. In one aspect, the AON as disclosed herein comprises at least one internucleoside linkage according to the structure of formula (IX), and / or the AON further comprises at least one nucleotide with a sugar moiety that comprises a 2’-OMe modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’-MOE modification, and / or the AON comprises at least one nucleotide with a sugar moiety that comprises a 2’- F modification, and / or the AON comprises an orphan nucleotide that carries a 2’-H in the sugar moiety and is therefore referred to as a DNA nucleotide, even though additional modifications may exist in its base and / or linkage to its neighbouring nucleosides. In oneaspect, the orphan nucleotide carries a 2’-F in the sugar moiety. In one aspect, the orphan nucleotide carries a di F substitution in the sugar moiety. In one aspect, the orphan nucleotide carries a 2’-F and a 2’-C-methyl in the sugar moiety. In one aspect, the orphan nucleotide comprises a 2’-F in the arabinose configuration (FANA) in the sugar moiety.

[0159] In one aspect, the AON is an antisense oligonucleotide that can form a double stranded nucleic acid complex with a target RNA molecule, wherein the double stranded nucleic acid complex can recruit an adenosine deaminating enzyme for deamination of a target adenosine in the target RNA molecule, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, and wherein the orphan nucleotide has the structure of formula (XIV):wherein: X is O, NH, OCH2, CH2, Se, or S; B is a nitrogenous base selected from the group consisting of: cytosine, uracil, isouracil, N3-glycosylated uracil, pseudoisocytosine, 8-oxo- adenine, and 6-amino-5-nitro-3-yl-2(1 H)-pyridone; R1 and R2 are both selected, independently, from H, OH, F or CH3; R3 is the part of the AON that is 5’ of the orphan nucleotide, consisting of 7 to 30 nucleotides; and R4 is the part of the AON that is 3’ of the orphan nucleotide, consisting of 4 to 25 nucleotides.

[0160] The nucleotide 3’ and / or 5’ from the orphan nucleotide may be DNA, more preferably the nucleotide at the 3’ (position -1).

[0161] Other chemical modifications of the AON as disclosed herein include the substitution of one or more than one of any of the hydrogen atoms with deuterium or tritium, examples of which can be found in e.g., Int. Patent Application Publication Nos. WO2014 / 022566 or WO201 5 / 011694. Again, in all cases, the modifications should be compatible with editing such that the AON fulfils its role as an oligonucleotide that can, after binding to its target sequence, recruit an adenosine deaminase enzyme because of the double-stranded nucleic acid entity that arises. In all aspects of the disclosure, the enzyme with adenosine deaminase activity is preferably ADAR1 , ADAR2, or ADAT.

[0162] AONs as disclosed herein preferably do not include a 5’-terminal O6-benzylguanosine or a 5’-terminal amino modification and preferably are not covalently linked to a SNAP-tag domain (an engineered O6-alkylguanosine-DNA-alkyl transferase). An AON as disclosed herein preferably does not comprise a boxB RNA hairpin sequence. In one aspect, an AONas disclosed herein comprises 0, 1 , 2 or 3 wobble base pairs with the target sequence, and / or 0, 1 , 2, 3, 4, 5, 6, 7, or 8 mismatching base pairs with the target RNA sequence. No mismatch exists when the orphan nucleotide is uridine, which may be defined differently when the orphan nucleotide is a uridine analog or derivative. One alternative for uridine is positioning an iso-uridine opposite the target adenosine, which likely does not pair like G pairs with II. Preferably, the target adenosine in the target sequence forms a mismatch base pair with the nucleoside in the AON that is directly opposite the target adenosine.

[0163] As outlined above, an AON as disclosed herein makes use of specific nucleotide modifications at predefined spots to ensure stability as well as proper ADAR binding and activity. These changes may vary and may include modifications in the backbone of the AON, in the sugar moiety of the nucleotides as well as in the nucleobases or the phosphodiester linkages, as outlined in detail herein. They may also be variably distributed throughout the sequence of the AON. Specific modifications may be needed to support interactions of different amino acid residues within the RNA-binding domains of ADAR enzymes, as well as those in the deaminase domain. For example, PS linkages between nucleotides or 2’-OMe or 2’-MOE modifications may be tolerated in some parts of the AON, while in other parts they should be avoided so as not to disrupt crucial interactions of the enzyme with the phosphate and 2’-OH groups. Specific nucleotide modifications may also be necessary to enhance the editing activity on substrate RNAs where the target sequence is not optimal for ADAR editing. Previous work has established that certain sequence contexts are more amenable to editing. For example, a target sequence 5’-UAG-3’ (with the target A in the middle) contains the preferred nearest-neighbor nucleotides for ADAR2, whereas a 5’-CAA-3’ target sequence is disfavored (Schneider et al. 2014. Nucleic Acids Res 42(10):e87). The structural analysis of ADAR2 deaminase domain hints at the possibility of enhancing editing by careful selection of the nucleotides that are opposite to the target trinucleotide. For example, the 5’-CAA-3’ target sequence, paired to a 3’-GCU-5’ sequence on the opposing strand (with the A-C mismatch formed in the middle), is disfavored because the guanosine base sterically clashes with an amino acid side chain of ADAR2. The guanosine opposite the C in such circumstances is preferably replaced by an inosine (hence, at the -1 position within the AON), preferably a deoxynucleotide carrying a hypoxanthine nucleobase.

[0164] The AON as disclosed herein, in contrast to what has been described for siRNA, or gapmers and their relation towards RNase breakdown and the use of such gapmers in double-stranded complexes (see for instance EP 3954395 A1), does not comprise a stretch of DNA nucleotides which would make a target sequence (or a sense nucleic acid strand) a target for RNase-mediated breakdown. It is not desired that the target transcript molecule is degraded through the binding of the AON to the transcript molecule. In one embodiment,the AON does not comprise four or more consecutive DNA nucleotides anywhere within its sequence. In an embodiment, the AON is composed of as much (chemically) modified nucleotides as possible to enhance the resistance towards RNase-mediated breakdown, while at the same time being as efficient as possible in producing an RNA editing effect. This means that the orphan nucleotide and several other nucleotides within the AON may be DNA, but also that there is no stretch of four or more consecutive DNA nucleotides within the AON. Hence, the AON as disclosed herein is not a gapmer. A gapmer reduces the expression of a target transcript but does not produce RNA editing of a specified adenosine within the target transcript. A gapmer is in principle a single-stranded nucleic acid consisting of a central region (DNA gap region with at least four consecutive deoxyribonucleotides) and wing regions positioned directly at the 5’ end (5’ wing region) and the 3’ end (3’ wing region) thereof. In contrast, the AON 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.

[0165] The AONs as disclosed herein may also be administered in the context of aids that will increase the entry of the AON into the target cell and / or its endosomal escape as soon as it is in the cell. Moieties that can be applied for such applications are for example a set of chemical compounds (generally purified from nature) referred to as saponins or triterpene glycosides, as outlined infra. A 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.

[0166] Disclosed herein is also a pharmaceutical composition comprising the AON as disclosed herein, and further comprising a pharmaceutically acceptable carrier, solvent, diluent, and / or other additive (such as a saponin or triterpene glycoside like AG1856 (as discussed above), which in fact may also be administered separately from the AON) and may be dissolved in a pharmaceutically acceptable organic solvent, or the like. Dosage forms in which the AON or the pharmaceutical composition are administered may depend on the disorder to be treated and the tissue that needs to be targeted and can be selected according to common procedures in the art. The pharmaceutical compositions may be administered by a single-dose administration or by multiple dose administration. It may be administered daily or at appropriate time intervals, which may be determined using common general knowledge in the field and may be adjusted based on the disorder, the severity thereof, and the efficacy of the active ingredient.

[0167] Although in a preferred embodiment, the AON as disclosed herein is a single-stranded oligonucleotide comprising an orphan nucleotide opposite the target adenosine, wherein the orphan nucleotide is chemically modified as disclosed herein, and wherein the remainder of the oligonucleotide is chemically modified to prevent it from nuclease breakdown also as disclosed herein, in another embodiment, disclosed is any kind of oligonucleotide or heteroduplex oligonucleotide complex, that may or may not be bound to hairpin structures (internally or at the terminal end(s)), that may be bound to ADAR or catalytic domains thereof, or wherein the oligonucleotide is in a circular format. In a preferred aspect, the AON as disclosed herein is a ‘naked’ oligonucleotide, comprising a variety of chemical modifications in the ribose sugar and / or the base of one or more of the nucleotides within the sequence, that preferably comprises at least one linkage according to the structure of formula (IX) as disclosed herein, that can hybridize to the target transcript or a part thereof that includes the target adenosine, and can recruit endogenous (naturally present) ADAR in the target cell for the deamination of the target adenosine. In another aspect, the AON as disclosed herein, that is delivered in a ‘naked’ form, does not comprise a stem-loop structure for recruitment of the deaminating enzyme, which allows for a shorter AON and improved cellular delivery and trafficking.

[0168] It is known in the art that RNA editing entities (such as human ADAR enzymes) edit dsRNA structures with varying specificity, depending on several factors. One important factor is the degree of complementarity of the two strands making up the dsRNA sequence. Perfect complementarity of the two strands usually causes the catalytic domain of human ADAR to deaminate adenosines in a non-discriminative manner, reacting with any adenosine it encounters. The specificity of hADARI and 2 can be increased by introducing chemical modifications and / or ensuring several mismatches in the dsRNA, which presumably helps to position the dsRNA binding domains in a way that has not been clearly defined yet. Additionally, the deamination reaction itself can be enhanced by providing an oligonucleotide that comprises a mismatch opposite the adenosine to be edited. Following the instructions in the present application, those of skill in the art will be capable of designing the complementary portion of the oligonucleotide according to their needs.

[0169] It will be understood by a person having ordinary skill in the art that the extent to which the editing enzymes inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing enzyme. The exact modification may be determined through some trial and error and / or through computational methods based on structural interactions between the AON and the recognition domain of the editing enzyme. In addition, or alternatively, the degree of recruiting and redirecting the editing enzyme resident in the cell may be regulated by thedosing and the dosing regimen of the AON. This is something to be determined by the experimenter {in vitro) or the clinician, usually in phase I and / or II clinical trials.

[0170] Disclosed herein is the site-specific editing of target adenosines in RNA sequences in eukaryotic, preferably metazoan, more preferably mammalian, more preferably human cells, more preferably human neuronal cells, and most preferably human cells of the CNS. The target cell can be located in vitro, ex vivo or in vivo. One advantage of the AON as disclosed herein is that it can be used with cells in situ in a living organism, but it can also be used with cells in culture. In some embodiments cells are treated ex vivo and are then introduced into a living organism (e.g., re-introduced into an organism from whom they were originally derived). The AON as disclosed herein can also be used to edit target RNA sequences in cells from a transplant or within a so-called organoid, e.g., a brain tissue organoid. Organoids can be thought of as three-dimensional in v / tro-derived tissues but are driven using specific conditions to generate individual, isolated tissues.

[0171] Without wishing to be bound by theory, the RNA editing through human ADAR2 for example is thought to take place on primary transcripts in the nucleus, during transcription or splicing, or in the cytoplasm, where e.g., mature mRNA, miRNA or ncRNA can be edited. Generally spoken, RNA editing may be used to create RNA sequences with different properties. Such properties may be coding properties (creating proteins with different sequences or length, leading to altered protein properties or functions), or binding properties (causing inhibition or over-expression of the RNA itself or a target or binding partner; entire expression pathways may be altered by recoding miRNAs or their cognate sequences on target RNAs). Protein function or localization may be changed at will, by functional domains or recognition motifs, including but not limited to signal sequences, targeting or localization signals, recognition sites for proteolytic cleavage or co- or post-translational modification, catalytic sites of enzymes, binding sites for binding partners, signals for degradation or activation and so on. These and other forms of RNA and protein “engineering”, whether to prevent, delay or treat disease or for any other purpose, in medicine or biotechnology, as diagnostic, prophylactic, therapeutic, research tool or otherwise, are encompassed by the present disclosure.

[0172] The amount of AON to be administered, the dosage and the dosing regimen can vary from cell type to cell type, the disease to be treated, the target population, the mode of administration e.g., systemic versus local), the severity of disease and the acceptable level of side activity, but these can and should be assessed by trial and error during in vitro research, in pre-clinical and clinical trials. The trials are particularly straightforward when the modified sequence leads to an easily detected phenotypic change, or a change in (the level of, or activity of) a specified biomarker. It is possible that higher doses of AONs couldcompete for binding to an ADAR enzyme within a cell, thereby depleting the amount of the enzyme, which is free to take part in RNA editing, but routine dosing trials will reveal any such effects for a given AON and a given target.

[0173] One suitable trial technique involves delivering the AON to cell lines, or a test organism and then taking biopsy samples at various time points thereafter. The sequence of the target RNA can be assessed in the biopsy sample and the proportion of cells having the modification can easily be followed. A potential biomarker that can be used for determining the increase in PEX1 activity is the concentration of Very Long Chain Fatty Acids (VLCFA) in fasting blood samples. 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.

[0174] After RNA editing has occurred in a cell, the modified RNA can become diluted over time, for example due to cell division, limited half-life of the edited RNAs, etc. Thus, in practical therapeutic terms a method as disclosed herein may involve repeated delivery of an AON until enough target RNAs have been modified to provide a tangible benefit to the patient and / or to maintain the benefits over time.

[0175] AONs as disclosed herein are particularly suitable for therapeutic use, and so disclosed is also a pharmaceutical composition comprising an AON as disclosed herein and a pharmaceutically acceptable carrier, solvent, or diluent. In some embodiments the pharmaceutically acceptable carrier can simply be a saline solution. This can usefully be isotonic or hypotonic, particularly for pulmonary delivery. The AON as disclosed herein is suitably administrated in aqueous solution, e.g. saline, or in suspension, optionally comprising additives, excipients and other ingredients, compatible with pharmaceutical use, at concentrations ranging from 1 ng / ml to 1 g / ml, preferably from 10 ng / ml to 500 mg / ml, more preferably from 100 ng / ml to 100 mg / ml. Dosage may suitably range from between about 1 pg / kg to about 100 mg / kg, preferably from about 10 pg / kg to about 10 mg / kg, more preferably from about 100 pg / kg to about 1 mg / kg. Administration may be by injection or infusion, intracranially, intrathecally, intranasally, orally, intravenously, subcutaneously, intradermally, intramuscularly, intra-tracheally, intra-peritoneally, intrarectally, intra-cisterna magna, parenterally, and the like. Administration may be in solid form, in the form of apowder, a pill, a gel, a solution, a slow-release formulation, or in any other form compatible with pharmaceutical use in humans.

[0176] In one embodiment, depending on the ultimate deamination effect of A to I conversion, the identification step of whether the editing has taken place, comprises the following steps: sequencing the target RNA; assessing the presence or absence of a non-, or less-functional protein; assessing whether splicing of the pre-mRNA was altered by the deamination; or using a functional read-out, because the target RNA after the deamination should encode a PEX1 protein with a wildtype function, which means restoration (at least in part) of the peroxisome functioning. The identification of the deamination into inosine may be a functional read-out using a suitable biomarker, such as levels of VLCFAs. A functional assessment will generally be according to methods known to the skilled person. A suitable manner to identify the presence of an inosine after deamination of the target adenosine is of course dPCR or even sequencing, using methods that are well-known to the person skilled in the art. However, the person skilled in the art of metabolic disease will preferably apply tests to monitor certain biomarkers related to metabolic function(s).

[0177] In one embodiment, a method as disclosed herein comprises the steps of administering to the subject an AON or pharmaceutical composition as disclosed herein, allowing the formation of a double stranded nucleic acid complex of the AON with its specific complementary target nucleic acid molecule in a cell in the subject; allowing the engagement of an endogenous present adenosine deaminating enzyme, such as ADAR 1 or ADAR2; and allowing the enzyme to deaminate the target adenosine in the target nucleic target molecule to an inosine, thereby alleviating, treating, ameliorating, or slowing down progression of the peroxisome-related disease, or more in particular the PBD-ZSD.

[0178] RNA editing molecules present in the cell will usually be proteinaceous in nature, such as the ADAR enzymes found in metazoans, including mammals. The ones of most interest are the human ADARs, hADARI and hADAR2, including any isoforms thereof. RNA editing enzymes known in the art, for which oligonucleotide constructs as disclosed herein may conveniently be designed, include the adenosine deaminases acting on RNA (ADARs), such as hADARI and hADAR2 in humans or human cells and cytidine deaminases. It is known that hADARI exists in two isoforms; a long 150 kDa interferon inducible version and a shorter, 110 kDa version, that is produced through alternative splicing from a common pre- mRNA. Consequently, the level of the 150 kDa isoform available in the cell may be influenced by interferon, particularly interferon-gamma (IFN-y). hADARI is also inducible by TNF-a. This provides an opportunity to develop combination therapy, whereby IFN-y orTNF- a and AONs as disclosed herein are administered to a patient either as a combination product, or as separate products, either simultaneously or subsequently, in any order.Certain disease conditions may already coincide with increased IFN-y or TNF-a levels in certain tissues of a patient, creating further opportunities to make editing more specific for diseased tissues. It will be understood by a person having ordinary skill in the art that the extent to which the editing entities inside the cell are redirected to other target sites may be regulated by varying the affinity of the first nucleic acid strand for the recognition domain of the editing molecule.

[0179] An AON as disclosed herein can utilise endogenous cellular pathways and naturally available ADAR enzymes to specifically edit a target adenosine in the target RNA sequence. An AON as disclosed herein is capable of recruiting ADAR and complex with it and then facilitates the deamination of a (single) specific target adenosine nucleotide in a target RNA sequence to which it is bound. Ideally, only one adenosine is deaminated. An AON as disclosed herein, when complexed to ADAR, preferably brings about the deamination of a single target adenosine.

[0180] An AON as disclosed herein, especially when it is in a naked form, is normally longer than 10 nucleotides, preferably more than 11 , 12, 13, 14, 15, 16, still more preferably more than 17 nucleotides. In one aspect the AON as disclosed herein is longer than 20 nucleotides. The AON as disclosed herein is preferably shorter than 100 nucleotides, still more preferably shorter than 60 nucleotides, still more preferably shorter than 50 nucleotides. In a preferred aspect, the AON as disclosed herein comprises 18 to 70 nucleotides, more preferably comprises 18 to 60 nucleotides, and even more preferably comprises 18 to 50 nucleotides. Hence, in a particularly preferred aspect, the AON as disclosed herein comprises 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides. In one embodiment, the AON is 27, 28, 29, or 30 nucleotides in length.EXAMPLES

[0181] Example 1. Use ofRNA editing guide oligonucleotides forA-to-l editing of human PEX1 target RNA in an in vitro biochemical editing assay.

[0182] It was initially investigated whether AONs can be applied for specific editing of human PEX1 mRNA. An initial set of AONs was designed to target the adenosine in the GAU codon encoding aspartic acid at position 843 in the mutated human PEX1 protein. The nucleotide sequences, designs and chemical modifications of these AONs are provided in FIG. 1. The editing efficacy of RM 118062 (SEQ ID NO:265), RM 118071 , RM 118073, RM 118074,RM120721 (SEQ ID NO:121), RM120724 (SEQ ID NO:124), RM120725 (SEQ ID NO:125),RM120726 (SEQ ID NO:126), RM120727 (SEQ ID NO:127), RM120761 (SEQ ID NO:163),RM120770 (SEQ ID NQ:170), RM120771 (SEQ ID NO:171), RM120772 (SEQ ID NO:172),RM120773 (SEQ ID NO:173), and RM120774 (SEQ ID NO:174) was measured and compared in an in vitro biochemical editing assay (BEA).

[0183] For this, transcribed target RNA used for the BEA was obtained from Integrated DNA Technologies and comprised the following PEX7-specific 64 nucleotides (SEQ ID NO:316) in which the target adenosine is underlined:5'-CUAGAGACCUGGGUUGGGACAAGAUUGAUGGGUUACAUGAAGUUAGGCAGAUACUCAUGGAUAC-3'

[0184] AONs were annealed to the PEX1 target RNA, which was added to a buffer (5 mM Tris-CI pH 7.4, 0.5 mM EDTA and 10 mM NaCI) at a 1 :3 ratio of target RNA and oligonucleotide (200 nM target RNA and 600 nM AON). The samples were heated at 95°C for 3 min and then slowly cooled down to 4°C and diluted 10x. Next, the editing reaction was carried out. The annealed oligonucleotide I target RNA was mixed with protease inhibitor (complete™, Mini, EDTA-free Protease I, Sigma-Aldrich), RNase inhibitor (RNasin, Promega), poly A (Qiagen), tRNA (Invitrogen) and editing reaction buffer (15 mM Tris-CI pH 7.4, 1.5 mM EDTA, 3% glycerol, 60 mM KOI, 0.003% NP-40, 3 mM MgCI2and 0.5 mM DTT) such that their final concentration was 6 nM AON and 2 nM target RNA. The reaction was started by adding purified ADAR2 (GenScript) to a final concentration of 6 nM into the mix and incubated for predetermined time points (either 0 sec and 1 min; or 0 sec, 30 sec,1 min, 2 min, 5 min, 10 min, 15 min, 30 min, 45 min and 60 min) at 37 °C. Each reaction was stopped by removing 5 pl of reaction mixture and adding it to 95 pl of 3 mM EDTA followed by incubation at 95 °C for 5 min. A 6 pl aliquot of the stopped reaction mixture was then used as template for cDNA synthesis using Maxima reverse transcriptase kit (Thermo Fisher) with reverse transcriptase (RT) binding primer (IDT). Initial denaturation of tRNA / EON duplex was performed in the presence of the primer and dNTPs at 95 °C for 5 min, followed by slowcooling to 10 °C, after which first strand synthesis was carried out according to the manufacturer’s instructions in a total volume of 20 pl, using an extension temperature of 60 °C. Products were amplified for pyrosequencing analysis by PCR, using the Amplitaq gold 360 DNA Polymerase kit (Applied Biosystems) according to the manufacturer’s instructions, with 2 pl of the cDNA as template. The following primers were used at a concentration of 10 pM: FW BEA standard, 5’-ATCGGGTCAGAGCCTGTAACT-3’ (SEQ ID NO:317), and RV biotinylated BEA standard, 5’- / 5Biosg / CCCAAGGAGCTGGAAAATC-3’ (SEQ ID NO:318). PCR was performed using the following thermal cycling protocol: Initial denaturation at 95°C for 5 min, followed by 45 cycles of 95°C for 15 sec, 60°C for 30 sec and 72°C for 30 sec, and a final extension of 72°C for 5 min. Because inosines base-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 pyrosequencing. Pyrosequencing of the PCR products and data analysis were performed by the PyroMark Q48 Autoprep instrument (QIAGEN) following the manufacturer’s instructions with 10 pl input of the PCR product and 4 pM of the following sequencing primer: Sequencing primer BEA standard: 5’- GTAACTCCAAGGGTTCA-3’ (SEQ ID NO:319). The settings specifically defined for this target RNA strand included two sets of sequence information. The first of these defines the sequence for the instrument to analyse, in which the potential for a particular position to contain either an adenosine or a guanosine is indicated by a5’- CTAGAGACCTGGGTTGGGACAAGATTGA / GTGGGTTACA-3’ (SEQ ID NQ:320). The dispensation order was defined for this analysis as follows: 5’-ACTAGAGACT GGTGGACAGATAGAGTGTAT-3’ (SEQ ID NO:321). The analysis performed by the instrument provides the results for the selected nucleotide as a percentage of adenosine and guanosine detected in that position, and the extent of A-to-l editing at a chosen position will therefore be measured by the percentage of guanosine in that position.

[0185] The results are shown in FIG. 3. Clearly, all tested AONs provided editing of the target adenosine, and no editing was observed with the negative controls. Interestingly, AONs RM120770, RM120771 , RM120772, RM120773, and RM120774, comprising a 7-deaza-2’- deoxyadenosine (7Ad) at the -1 position in the AON, which is opposite the guanosine in the GAU codon, performed best in line with what has been disclosed in Inti. Patent Application Publication Number WQ2024 / 013361. In contrast, the AONs comprising a cytidine at that position (RM 118062, RM 118071 , RM 118073, and RM 118074) provided significantly lower editing in the BEA. The AONs with an inosine at position -1 (RM120721 , RM120724, RM120725, RM120726, RM120727, and RM120761) provided proper intermediate editing results. The only difference between RM120724 and RM120761 is the presence of a PNmslinkage at linkage position -2 in RM120761 instead of an MP linkage in RM120724, which replacement does not appear to hamper editing in the BEA.

[0186] The pyrosequencing results after use of 3, 6, 9, and 12 nM of RM120721 (32-mer) and RM 120770 (25-mer) are provided in FIG. 4A and FIG. 4B, respectively. In line with the results provided in FIG. 3, RM 120770 reached a higher percentage editing (approximately 70%) than RM120721 (approximately 55%) with a clear dose-response effect. Together, this shows that it was possible to obtain specific A-to-l editing of the target adenosine in the GUA codon encoding aspartic acid in the mutated human PEX1 protein.

[0187] Example 2. RNA editing of the PEX1 transcript in Zellweger patient fibroblasts.

[0188] To test whether editing could also be achieved in cells, applying endogenous ADAR enzymes, on an endogenously present target transcript, it was first investigated whether commercially available Zellweger patient fibroblasts carried the G843D mutation. For this, cell lines GM00228, GM08040, GM16514, and GM06094 (Coriell) were screened for the mutation. It turned out that GM00228 appeared wildtype for PEX1. GM08040 was wildtype at the mutation site but lacked exon 9 and 14. GM16514 and GM06094 both contained the c.2528G>A (p.Gly843Asp; rs6175420) mutation, but GM06094 appeared to be a very slow growing cell line in culture. It was concluded that it was best that Zellweger patient fibroblasts GM16514 were going to be used for further studies. The next investigation showed that these cells were heterozygous for the c.2528G>A mutation, which complicates things in the sense that wildtype background signal is present in an editing assay. However, this background signal can be deducted.

[0189] In the first setup, the 54 AONs with SEQ ID NO: 121 to 174 (see FIG. 1) were tested.

[0190] On day 0, GM 16514 fibroblasts (5.0x104cells / well for 96 wells plate) were seeded. On day 1 , cells were transfected with 100 nM AONs, in triplicate, using Lipofectamine® RNAiMAX Reagent, following the protocol of the manufacturer. The plates containing cells, medium and AON were held at 37°C, 5% CO2, and the medium was refreshed 24 hrs after transfection / plating. On day 2 (48 hrs post transfection / plating) the supernatants were discarded, and subsequent analysis was performed as follows. Cells were collected and used for RNA isolation using a ReliaPrep RNA Cell Miniprep kit (Promega-Z6012) according to the manufacturer’s instructions. Isolated RNA was then reverse-transcribed using the Maxima Reverse Transcriptase (Thermo-EP0742) kit with oligo-dT primer, and / or random Hexamer Primer, and dNTP Mix (10 mM each). A quantitative PCR was then performed with the QI Acuity Digital PCR System (QIAGEN, QIAcuity Four)) in 12 pl aliquots of reaction mixtures containing cDNA, appropriate pairs of primers and probes, as well as the dPCR 4x Mastermix of the QIAcuity Probe PCR kit(QIAGEN- 250102). The primers given in Table 1 were used with a PCR program that was as follows: 2 min at 95°C; 40 cycles for 15 sec at 95°C and 30 sec at 61-63°C.

[0191] The plate was placed in the system where partitioning PCR and imaging is performed automatically. Thereafter the copy / pL values were retrieved from the instrument. The editing percentage was calculated per replicate according to the formula: score = G / (A+G) * 100

[0192] Table 1. Primers and probes with their respective sequences for quantitative PCR assays. The SEQ ID NO of each primer or probe is given between brackets following the respective sequence in the middle column. The + indicates a Locked Nucleic Acid (LNA) on the 3’ side of the symbol.

[0193] For each AON, the average editing score was calculated from the individual replicates.

[0194] First it was analysed whether a skip of exon 15 occurred after incubation with the respective AONs. However, very low levels of skipping were observed (data not shown), indicating that the AONs apparently did not interfere with splicing events on the target PEX1 pre-mRNA.

[0195] The editing results of the 1sttransfection screen in human cells are shown in FIG. 5. It should be noted that the background editing level, that scored up to around 30% on average in the three negative controls: Non-Treated (NT), Mock, and ACTB control AON, was deducted from the signal. Hence, the editing percentage obtained for each AON (averaged) minus the background (wt) level, is depicted. Whereas some AONs, such as RM120742 and RM 120761 did not show any editing above background in this assay, some AONs outperformed. The best performing AONs with a range of approximately 5% to 14% editing above background were RM 120735 (SEQ ID NO: 135), RM 120736 (SEQ ID NO: 136), RM120721 (SEQ ID NO:121 , which goes by the names of PEX1_G843D_78 and PEX1_G843D_135, see FIG. 1), RM120722 (SEQ ID NO:122), RM120738 (SEQ ID NO:138), RM120723 (SEQ ID NO:123), RM120724 (SEQ ID NO:124), RM120725 (SEQ ID NO:125), and RM120773 (SEQ ID NO:173).

[0196] In a 2ndscreen, RM123054 to RM123109 (SEQ ID NO’s, sequences and modifications shown in FIG. 1) were compared to RM120721 and RM120735 (from the 1stscreen above) and tested in an identical experimental setup (GM16514 cells, transfection, 100 nM AON,48 hr incubation) as described above. The editing percentages of the three replicates and their averages are provided in Table 2.

[0197] Table 2. Results of the 2ndAON screen using the AONs with the RM numbers indicated in human GM 16514 cells after transfection of 100 nM AON and 48 hr incubation. The column with the Avg shows the average editing percentage of the three replicates in the N=1 experiment.

[0198] These results show that whereas RM 120735 (SEQ ID NO: 135) still gave a high percentage level of 17.26% on average (above background), other AONs, such as RM123068 (SEQ ID NO:220), RM123067 (SEQ ID NO:219), RM123061 (SEQ ID NO:213),RM123066 (SEQ ID NO:218), RM123072 (SEQ ID NO:227), RM123060 (SEQ ID NO:212),RM123056 (SEQ ID NO:226), RM123079 (SEQ ID NO:234), RM123065 (SEQ ID NO:217),RM123071 (SEQ ID NO:225), RM123085 (SEQ ID NQ:240), RM123057 (SEQ ID NQ:208),RM123080 (SEQ ID NO:235), and RM123081 (SEQ ID NO:236) gave editing percentages above 16%, with RM123068 (SEQ ID NQ:220) showing a result of 19.32%.

[0199] Based on the results from the 1stand 2ndscreen above a further set of AONs was designed, shown as RM125212 (SEQ ID NO:322), and as RM125486 to RM125517 (SEQ ID NO:323 to 354, respectively) in FIG. 6.

[0200] In a 3rdscreen, AONs RM125212, and RM125486 to RM125517, shown in FIG. 6, were compared to RM 123054, RM 123055, and RM 123056, shown in FIG. 1 and used in the 2ndscreen, and RM 120735 which performed best in the 1stscreen and properly in the 2ndscreen, in an identical experimental setup as performed in the 1stand 2ndscreen. Now, two experiments were performed (N=1 and N=2), each with three technical replicates. Theediting results of the two experiments with each an average of its three replicates as well as an average of the two averaged experiments are provided in Table 3.

[0201] Table 3. Results of the 3rdAON screen using the AONs with the RM numbers indicated, in human GM16514 cells after transfection of 100 nM AON and 48 hr incubation. The column with the Final avg shows the average editing percentage of the average in the N=1 experiment and the average in the N=2 experiment.

[0202] These results shows that whereas RM 120735 (SEQ ID NO: 135) still gave a high percentage level of 14.63% (above background) in this experiment. RM 125494 (SEQ ID NO:331) outperformed with 16.93% editing (above background), whereas RM125507 (SEQ ID NO:344) and RM125498 (SEQ ID NO:335) also showed a good activity.

[0203] From these three transfection experiments it was concluded that the best performing AONs in these assays were at least those as provided as SEQ ID NO: 331 , 135, 344, 335, 220, 219, 213, 218, 227, 212, 226, 234, 217, 225, 240, 208, 235, 236, 136, 121 , 122, 138, 123, 124, 125, and 173, which therefore represent preferred embodiments of the present disclosure.

[0204] Example 3. RNA editing of the PEX1 transcript in Zellweger patient fibroblasts using AONs with a variety of chemical modifications.

[0205] Further to the experiments performed in example 2 and based on the designs of the AONs and their editing efficiencies, a variety of AONs was designed to study: i) the need for and the positions of 2’-F modifications of the nucleotides within the AONs; ii) modifications of the orphan nucleotide; iii) modifications of the -1 position in the AON, which nucleotide is opposite the 5’ guanosine next to the target adenosine in the c.2528G>A (rs61750420) PEX1 mRNA; and iv) the presence or absence of PNms linkage modifications. The AONs used here are shown in FIG. 6.

[0206] All AONs used in these experiments were tested for editing efficiency using GM 16514 fibroblasts and Lipofectamine RNAiMAX transfection using 100 nM AON and 48 hr incubation, with two experiments (N=1 and N=2) each with three technical replicates, as outlined in example 2. RNA isolation and editing efficiencies were also determined as described above.

[0207] AON RM 120735 was used in six separate transfection experiments, each with three technical replicates. These six experiments include the two experiments (N=1 and N=2) shown in Table 3, with an average score of 14.63%. The total average editing percentage that was found (17.66% editing) from the six experiments is provided in T able 4. RM 125489, RM125507, RM125508, RM125509, and RM125512 were all transfected four times, each with three technical replicates, including the N=1 and N=2 experiments provided in Table 3. RM 125498 was transfected three times, each with three technical replicates. The average editing percentages of these six AONs are also provided in Table 4. The other AONs provided in Table 4 were transfected twice (N=1 and N=2), except for RM126218 and RM 126219 for which no results were obtained in the second transfection experiment.

[0208] Table 4. Results of an AON screen using the AONs with the RM numbers indicated, in human GM 16514 cells after transfection of 100 nM AON and 48 hr incubation. The column with the Total avg shows the average editing percentage of the average in the N=1 experiment and the average in the N=2 experiment, except for RM 120735, RM 125489, RM 125498, RM 125507, RM 125508, RM 125509, and RM 125512, that were transfected more than twice, as indicated. The Total avg is the average of all average calculations from each of the N experiments, per AON.

[0209] Within these results, RM 120735, RM 125497, RM 125486, RM 125487, RM 125488, RM125489, RM125490, RM125495, and RM125496 were compared in relation to the 2’-F modification patterns in the AONs. The results that are based on the percentages provided in Table 4 show that RM 120735 (SEQ ID NO: 135) outperformed these other AONs. RM 125487, which has a consecutive stretch of six 2’-F modified nucleotides from position +2 to +7, in fact performed less well, indicating the preference of an alternating 2’-F modification pattern at this region of the AON, as observed in RM120735. RM125488 (SEQ ID NO:325) also performed relatively good.

[0210] Then, RM120735, RM126151 , RM125512, RM126152, RM125489, and RM126153 were compared to see whether the introduction of the E-base (see above: formula (X) and the beneficial properties of this cytidine analog) at the orphan position could influence the editing efficiency if the target A in the human PEX1 target transcript. RM 120735 (comprising a Benner’s base at the orphan nucleotide) was compared to RM126151 , which is identical to RM 120735 except that it carries the E-base (5-aza-5,6-dihydro cytosine nucleobase) at the orphan nucleotide. RM125512 (comprising a Benner’s base at the orphan nucleotide) was compared to RM 126152 (identical but comprising an E-base at the orphan position), and RM 125489 (comprising a Benner’s base at the orphan nucleotide) was compared to RM 126153 (identical but comprising an E-base at the orphan position). The results based on what is provided in Table 4 show that in this experiment the change of the Benner’s base to an E-base may increase editing at this level. RM 126151 (SEQ ID NO:355) showed a 19.62% average editing and performed better than RM120735. Also, RM126152 (SEQ ID NO:356) and RM126153 (SEQ ID NO:357) showed an increase in editing when compared to their counterparts that comprise a Benner’s base at the orphan position (RM125512 and RM 125489, respectively). These results indicate the beneficial properties of using an E-baseat the orphan position in the AON that is applied for deamination of the adenosine in the c.2528G>A (rs61750420) mutated PEX1 transcript.

[0211] Then, because at the 5’ side of the target adenosine in the human PEX1 c.2528G>A mutation a guanosine (also referred to as a 5’-G) is positioned (see FIG. 1 and SEQ ID NO:207), and in view of what is disclosed in Int. Patent Application Publication No. W02024 / 013361 , it was investigated whether the introduction of a 7-deaza-2’- deoxyadenosine (7-deaza Ad) or a 3-deaza-2’-deoxyadenosine (3-deaza Ad) nucleotide at the -1 position in the AON could beneficially influence editing. For this, RM 120735 (SEQ ID NO: 135) was compared to RM 123055 (SEQ ID NO:222) that is identical to RM 120735 except for the presence of a 7-deaza Ad at position -1 instead of a deoxyinosine (Id). RM125512 (SEQ ID NO:349) comprising an Id at position -1 was compared to RM125494 (SEQ ID NO:331) comprising a 7-deaza Ad at position -1 and with RM125154 (SEQ ID NO:358) comprising a 3-deaza Ad at position -1 and further being identical. RM 125489 (SEQ ID NO:326) comprising an Id at position -1 was compared to RM 125493 (SEQ ID NQ:330) comprising a 7-deaza Ad at position -1 and with RM 126155 (SEQ ID NO:359) comprising a 3-deaza Ad at position -1 and further being identical. The results show that using Id, 7-deaza Ad or 3-deaza Ad at position -1 are all functional and differences likely depend on the further chemical modifications of the AON. In respect of 7-deaza Ad, RM125494 (7-deaza Ad; 16.93% editing) clearly outperforms RM125512 (Id; 10.35% editing) and RM126154 (3- deaza Ad; 8.88% editing). In contrast, RM126155 (3-deaza Ad; 11.77% editing) outperforms RM125489 (Id; 10.08% editing) and RM125493 (7-deaza Ad; 9.72% editing). In respect of Id, RM120735 (Id; 17.66% editing) outperforms RM123055 (7-deaza Ad; 12.93% editing). The overall conclusion regarding the -1 position in the AON is that, apparently in respect of the PEX1 target transcript, targeting the c.2528G>A mutation (where a 5’-G is present next to the target adenosine) one can select Id, 7-deaza Ad and 3-deaza Ad for the -1 position in the AON, depending on the further modifications of the AON, since all may positively contribute to editing levels.

[0212] It was also studied whether adding PNms linkages instead of PO or PS linkages at a variety of positions within the AON would influence RNA editing efficiency. RM120735 was compared to RM125512, RM125498, RM125507, RM125508, RM125509, RM125489, RM125498, and RM126134 to RM126149, using the percentages provided in Table 4. This showed that the addition of PNms linkages to the AONs does not negatively influence editing efficiency. For example, RM126144 comprises 11 PNms linkages at a total of 32 linkages and gave 13.80% editing (with RM120735 giving 17.66% editing). In general, it can be stated that decreasing the number of PNms linkages also does not negatively influence editing efficiency.

[0213] From the transfection experiments with the AONs shown in Table 2, 3, and 4, it can be concluded that the best performing AONs in these assays were RM 127512 (SEQ ID NO:434), RM127513 (SEQ ID NO:435), RM127505 (SEQ ID NO:427), RM127520 (SEQ ID NO:442), RM127479 (SEQ ID NQ:401), RM127509 (SEQ ID NO:431), RM127508 (SEQ ID NQ:430), RM127510 (SEQ ID NO:432), RM127511 (SEQ ID NO:433), RM127499 (SEQ ID NO:421), RM126151 (SEQ ID NO:355), RM127478 (SEQ ID NQ:400), RM127477 (SEQ ID NO:399), RM120735 (SEQ ID NO:135), RM127507 (SEQ ID NO:429), RM127522 (SEQ ID NO:444), RM127517 (SEQ ID NO:439), RM127521 (SEQ ID NO:443), RM127498 (SEQ ID NQ:420), RM127501 (SEQ ID NO:423), RM125494 (SEQ ID NO:331), RM125498 (SEQ ID NO:335), RM125507 (SEQ ID NO:344), RM123056 (SEQ ID NO:226), RM123055 (SEQ ID NO:222), RM123057 (SEQ ID NQ:208), RM123060 (SEQ ID NO:212), RM123061 (SEQ ID NO:213), RM123064 (SEQ ID NO:216), RM123065 (SEQ ID NO:217), RM123066 (SEQ ID NO:218), RM123067 (SEQ ID NO:219), RM123068 (SEQ ID NQ:220), RM123071 (SEQ ID NO:225), RM123072 (SEQ ID NO:227), RM123079 (SEQ ID NO:234), RM123080 (SEQ ID NO:235), RM123081 (SEQ ID NO:236), RM123082 (SEQ ID NO:237), RM123083 (SEQ ID NO:238), RM123085 (SEQ ID NQ:240), and RM123086 (SEQ ID NO:241).

[0214] In a stability assay, AONs RM125512, RM126152, RM125494, RM125489, RM126150, RM125498, RM125507, RM125508, RM125509, and RM125491 were subjected to a nuclease treatment for 24 hr to determine their stability under such harsh conditions. All AONs were still >93% intact after 24 hr, indicating the high stability of all these AONs (data not shown). The introduction of the E-base, the 7Ad modification at position -1 , and the introduction of the iso-uracil at the orphan position did not influence stability (data not shown).

[0215] Example 4. RNA editing of the PEX1 transcript in homozygous Zellweger patient fibroblasts.[0216J GM16514 fibroblasts are heterozygous for the c.2528G>A (p.Gly843Asp; rs6175420) mutation, causing a significant high background wildtype signal in editing assays. It was subsequently investigated what levels of editing could be achieved in patient fibroblasts that were homozygous for the c.2528G>A mutation. Two cell lines were available, referred to as cell line 7 and cell line 12. Both cell lines were seeded to -50,000 cells per well and transfections were performed with 100 nM AON using Lipofectamine RNAiMAX transfection reagent. Incubation was performed for 48 hr. RNA purification, cDNA synthesis, and editing measurements were subsequently performed as described above. RM120735, RM125512, and RM 125494 were transfected twice in each cell line (N=1 and N=2), with two technical replicates. The other AONs were transfected only once (N=1) in each cell line, with twotechnical replicates. The results are provided in Table 5 (cell line 7) and Table 6 (cell line 12).

[0217] Table 5. Results of an AON screen using the AONs with the RM numbers indicated, in human patient-derived fibroblasts that are homozygous for the PEX1 c.2528G>A; rs61750420 mutation (cell line 7), after transfection of 100 nM AON and 48 hr incubation. The column with the average shows the average editing percentage of the average in the N=1 experiment and the average in the N=2 experiment (where applicable). Background levels (~0%) are not shown.

[0218] Table 6. Results of an AON screen using the AONs with the RM numbers indicated, in human patient-derived fibroblasts that are homozygous for the PEX1 c.2528G>A; rs61750420 mutation (cell line 12), after transfection of 100 nM AON and 48 hr incubation. The column with the average shows the average editing percentage of the average in the N=1 experiment and the average in the N=2 experiment (where applicable). Background levels (~0%) are not shown.

[0219] Interestingly, transfection of one of the best performers in GM16514 cells, RM120735 did not result in high levels of editing (only -0.5% in both cell lines), whereas two other AONs, RM126154 (SEQ ID NO:358) and RM126155 (SEQ ID NO:359) provided levels of 11.38% and 8.78% editing (RM 126154) and 19.93% and 10.24% (RM 126155), in this case without wildtype background levels. Cell line 7 showed the highest editing percentages. Strikingly, RM 126154 and RM 126155 are both AONs carrying a 3Ad modification at position -1 , indicating the importance of this modification opposite the 5’-G in the human PEX1 transcript.

[0220] Example 5. RNA editing of the mPex1 transcript in homozygous mutated mouse embryonic fibroblasts (MEFs).

[0221] FIG. 7 shows the respective sequences surrounding the c.2528G>A mutation in the human sequence, and the equivalent sequences in monkey (macaque) and mouse (Mus musculus). There are only a few differences between the human versus the monkey sequence, and between the human versus the mouse sequence (underlined in FIG. 7). Around the mutation (in large bold font in FIG. 7) no differences are present between the human and the monkey sequence. Depending on the position of the AON targeting the adenosine of interest, 1 or 2 differences exist between the human and mouse sequence. To determine whether mouse embryonic fibroblasts (MEFs) carrying the equivalent mPex1 mutation (resulting in a G844D change) could be used to study RNA editing of the target adenosine, two available cell lines, cell line 28 and cell line 32 that are both homozygous for the mPex1 c.2531G>A (G844D) mutation, were transfected with two AONs that were used in the studies with the human cell lines: RM120735 and RM125494, and two AONs that - outside the orphan position vs the target adenosine - were 100% complementary to the mouse mPex1 sequence: RM123170 (SEQ ID NO:472) and RM123179 (SEQ ID NO:473).The sequences and respective chemical modifications of these two AONs are also shown in FIG. 7.

[0222] The MEFS were seeded to -50,000 cells per well and transfection with 100 nM AON was performed with Lipofectamine RNAiMAX, and incubated for 48 hrs, generally as outlined above. The primers and probes used in the dPCR are given in Table 7.

[0223] Table 7. mPex1 -specific primers and probes with their respective sequences for quantitative PCR assays. The SEQ ID NO of each primer or probe is given between brackets following the respective sequence. The + indicates a Locked Nucleic Acid (LNA) on the 3’ side of the symbol.

[0224] The editing results are provided in Table 8.

[0225] Table 8. Results of an AON screen using the AONs with the RM numbers indicated, in MEFs that are homozygous for the mPex1 c.2531G>A mutation, after transfection of 100 nM AON and 48 hr incubation. The column with the average shows the averages in editing percentage for both cell lines, except for RM127479 and RM127534 that were tested in cell line 31 only. Background levels (-0%) are not shown.

[0226] Even though RM120735 performed well in the heterozygous human cell lines (see above), it did not provide a high level of editing in the homozygous mutated MEFs (-1.1%). Its equivalent, RM 123170, that is 100% complementary to the mouse sequence (outside the orphan position) did not show a great increase in editing in either cell line (2 to 5%). Surprisingly, RM125494 (SEQ ID NO:331) outperformed its equivalent RM123179, that was 100% complementary to the mouse sequence (outside the orphan position), with >10% editing in cell line 28 and >12% editing in cell line 31. RM127534 (SEQ ID NO:455) that was also 100% complementary to the mouse sequence, except for the orphan position, was compared to its ‘human’ equivalent RM127479 (SEQ ID NQ:401) in cell line 31 only, in which both performed relatively well.

[0227] Example 6. Screening AONs for RNA editing of the PEX1 transcript in heterozygous Zellweger patient fibroblasts.

[0228] In a further screen, 380 AONs were designed and tested for human PEX1 editing in human iPSC-derived hepatocytes (hvs523) that are (like GM16514 fibroblasts) heterozygous for the c.2528G>A (p.Gly843Asp; rs6175420) mutation.

[0229] For cell differentiation, the compound heterozygous G843D PEX1 (c.2097dupT (p.lle700TyrfsX42) / c.2528G>A (p.G843D)) cell line hvs-523a, that is reprogrammed from patient fibroblasts and wildtype PEX1 (WT04) hiPSC were differentiated to hepatocyte-like cells using the STEMDIFF Hepatocyte Kit (STEMCell). Human iPSC cells were cultured at 37°C, 5% CO2 in mTeSR Plus Basal Medium (STEMCell) supplemented with mTeSR Plus 5x Supplement (STEMCell) and 1 % Pen / Strep on Cell Adhere laminin-521 coated flasks in single cells and used for differentiation when >80% confluency was reached. Briefly, -30.000 cells were seeded in 96-wells plates, with a full-medium change on the following day with STEMdiff endoderm Basal Medium supplemented with STEMdiff Definitive Endoderm Supplement MR and CJ (STEMCell), and a full-medium change with STEMdiff endoderm Basal Medium supplemented with STEMdiff Definitive Endoderm Supplement CJ on day 2, 3 and 4. On day 5, 6, 7 and 9 full-medium change was performed with STEMDiff TM Hepatic Progenitor Medium (STEMCell). On day 10 full-medium change was performed with STEMdiff Hepatocyte Medium (STEMCell). Cells were treated with oligonucleotides on day 10, or day 11 . Non-treated hvs523a cells were taken as negative control and non-treated WT04 cells as positive control.

[0230] For testing the oligonucleotides, 5 pM AON was incubated with the cells for gymnotic uptake for 72 hr. Editing analysis was performed as outlined above, using human specificprimers and probes. It should be noted that also the hvs523a cells have a basic background signal due to the wild type allele that is also present. The 380 AONs are shown in FIG. 8 with their respective SEQ ID NO’s indicated. The editing results are provided in Table 9.

[0231] Table 9. Results of an AON screen for editing of the PEX1 c.2528G>A mutation after gymnotic uptake into human iPSC-derived fibroblasts that are heterozygous for the mutation. One treatment per AON. The rows with % show the editing percentages of the AON mentioned left of the value, n.d. = not determined.

[0232] These results show that some AONs performed relatively well, such as RM 131318 (SEQ ID NO:597), RM131329 (SEQ ID NO:608), RM131331 (SEQ ID NO:610), RM131326 (SEQ ID NO:605), RM131500 (SEQ ID NO:779), RM131328 (SEQ ID NQ:607), RM131338 (SEQ ID NO:617), RM131312 (SEQ ID NO:591), RM131443 (SEQ ID NO:722), and RM131270 (SEQ ID NO:549), highlighted in grey in Table 9.

[0233] Example 7. Functional effect of editing the c.2528G>A (p.Gly843Asp; rs6175420) mutation in hvs523a cells.

[0234] The Acyl-CoA oxidase 1 (ACOX1) protein is synthesized in the cytosol as a large precursor protein and only processed into a smaller mature form after import into peroxisomes. In the heterozygous cell line hvs-523a the processing of ACOX1 is absent and therefore a sensitive read-out for functional peroxisomal matrix protein import in cells. The wild-type level in wild type cells was taken as 100%, and the level of ACOX1 protein processing was found to be 1% in hvs523a cells.

[0235] ACOX1 immunodetection was analysed by the Jess Automated Western Blot System (ProteinSimple, Bio-Techne). Cells were harvested in lysis buffer (RIPA, with protease inhibitor cocktail). Protein quantification was performed by BCA (Pierce) according to the manufacturer’s protocol and 1000 pg / mL total protein was loaded on pre-filled plates (Bio- Techne) on 12-230 kDa Fluorescence Separation Module with 25 capillary cartridge (Bio- Techne). The assay settings were according to the manufacturer’s standard settings with slight adjustments. The stacking matrix load time was increased to 20 sec and sample separation time was increased to 32 min. As primary antibody, a rabbit polyclonal againstAC0X1 (Proteintech; 1:50 dilution) was used. As a loading control, a mouse monoclonal antibody against vinculin (Sigma; 1 :100 dilution) was used. For visualization, an anti-rabbit secondary HRP antibody (Bio-Techne) and anti-mouse secondary NIR antibody (Bio- Techne; 1:20 dilution) was used. Analysis was performed with the area of the peak normalized to vinculin as loading control. For ACOX1 processing, the ratio of 20 kDa 170kDa area of the corresponding peak as percentage of wildtype control sample was measured.

[0236] After incubation with: a) RM127512 + AG1856 saponin; b) RM131328 + AG1856 saponin; c) RM127479 alone; and d) RM127508 alone, processing of ACOX1 was scored as 8%, 5%, 2%, and 4%, respectively, indicating that the RNA editing on the c.2528G>A PEX1 mutation results in a downstream functional effect on ACOX1 processing in these human cells.

Claims

CLAIMS1 . An antisense oligonucleotide (AON) that is capable of recruiting an endogenous ADAR enzyme in a human cell after the AON has formed a double-stranded complex with a region of a target transcript molecule in a cell, wherein the region comprises a target adenosine, wherein the nucleotide in the AON that is opposite the target adenosine is the orphan nucleotide, wherein the ADAR enzyme can deaminate the target adenosine into an inosine after binding to the double-stranded complex, and wherein the target transcript molecule is a pre-mRNA or mRNA transcript molecule of a human PEX1 gene encoding a loss-of-function variant of the PEX1 protein.

2. An AON according to claim 1, wherein the variant is selected from the group consisting of р.Gly843Asp, p.Trp1250Ter, p.Trp388Ter, p.Trp839Ter, and p.Trp2Ter.

3. An AON according to claim 2, wherein the variant is p.Gly843Asp that is caused by a с.2528G>A mutation in the PEX1 gene (rs6175420), and wherein the target adenosine is the adenosine in the GAU codon for aspartic acid at position 843 of the encoded protein.

4. An AON according to any one of claims 1 to 3, wherein the orphan nucleotide is a deoxynucleotide.

5. An AON according to any of claims 1 to 4, wherein the orphan nucleotide comprises a cytosine, a cytosine analog, an uracil, or an uracil analog.

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

7. An AON according to claim 5, wherein the cytosine analog is according to formula (X):or any of its tautomeric forms, wherein:76R1, R2, R4and / or R5is H; OH; SH; =0; NH2; a halogen; a linear or branched lower (C1-C10) alkyl; or a Ci-Ce cycloalkyl, wherein the alkyl and / or cycloalkyl is optionally interrupted by one or more heteroatoms; andR3is H; OH; SH; =0; NH2; or a halogen.

8. An EON according to claim 7, wherein all of R1, R2, R3, R4and R5are H.

9. An AON according to any one of claims 1 to 8, wherein the nucleotide numbering is such that the orphan nucleotide is number 0 and nucleotides are further positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein the first nucleotide 3’ from the orphan nucleotide (position -1) is 7-deaza-2’-deoxyadenosine (7-deaza Ad), 3-deaza-2’-deoxyadenosine (3-deaza Ad), or 2’-deoxyinosine (Id).

10. An AON according to any one of claims 1 to 9, wherein the AON is 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 , 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 , 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, or 60 nucleotides in length.11 . An AON according to any one of claims 1 to 10, wherein the AON comprises one or more modifications in the linkage moiety, which is each independently selected from a phosphorothioate (PS), phosphonoacetate, phosphorodithioate, methyl phosphonate (MP), sulfonylphosphoramidate, (1 ,3-dimethylimidazolidin-2-ylidene) phosphoramidate (PNdmi), or a mesyl phosphoramidate (PNms) internucleotide linkage.

12. An AON according to any one of claims 1 to 11 , wherein the internucleoside linkage numbering in the AON is such that linkage number 0 is the linkage 5’ from the orphan nucleotide, and the linkage positions in the oligonucleotide are positively (+) incremented towards the 5’ end and negatively (-) incremented towards the 3’ end, and wherein linkage position -2 is an MP linkage or a PNms linkage.

13. An AON according to any one of claims 1 to 12, wherein the AON comprises one or more nucleotides comprising a mono- or di-substitution at the 2', 3' and / or 5' position of the ribose, each independently selected from the group consisting of: -OH; -F; substituted or unsubstituted, linear or branched lower (C1-C10) alkyl, alkenyl, alkynyl, alkaryl, allyl, or aralkyl, that may be interrupted by one or more heteroatoms; -O-, S-, or N-alkyl; -O-, S-, or N-alkenyl; -O-, S-, or N-alkynyl; -O-, S-, or N-allyl; -O-alkyl-O-alkyl; -methoxy; -aminopropoxy; - methoxyethoxy; -dimethylamino oxyethoxy; and -dimethylaminoethoxyethoxy.

14. An AON according to any one of claims 1 to 13, wherein the oligonucleotide comprises or consists of the sequence of any one of SEQ ID NO:434, 435, 427, 442, 401 , 431 , 430, 432,77433, 421 , 358, 359, 355, 400, 399, 135, 429, 444, 439, 443, 420, 423, 331 , 335, 344, 226, 222, 208, 212, 213, 216, 217, 218, 219, 220, 225, 227, 234, 235, 236, 237, 238, 240, 597, 608, 610, 605, 779, 607, 617, 591 , 722, 549, and 241.

15. A vector, preferably a viral vector, more preferably an adeno-associated virus (AAV) vector, comprising a nucleic acid molecule encoding an AON according to any one of claims 1 to 3.

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

17. An AON according to any one of claims 1 to 14, for use in the treatment of a Peroxisome Biogenesis Disorder in the Zellweger Spectrum (PBD-ZSD).

18. Use of an AON according to any one of claims 1 to 14 in the manufacture of a medicament for the treatment of a PBD-ZSD.

19. A method of editing a human PEX1 polynucleotide in a cell, wherein the human PEX1 polynucleotide is a pre-mRNA or mRNA molecule transcribed from a variant PEX1 gene that encodes a PEX1 protein with an impaired functionality, the method comprising contacting the PEX1 polynucleotide with an AON capable of triggering an ADAR-mediated adenosine to inosine deamination, thereby editing the PEX1 polynucleotide to encode a PEX1 protein with a wildtype functionality.

20. A method according to claim 19, wherein the PEX1 protein with an impaired functionality is selected from the group consisting of p.Gly843Asp, p.Trp1250Ter, p.Trp388Ter, p.Trp839Ter, and p.Trp2Ter.

21. A method of treating, ameliorating, or slowing down the progression of a PBD-ZSD, the method comprising administering to said subject an AON according to any one of claims 1 to 14, thereby contacting a PEX1 polynucleotide in a cell of the subject with an AON capable of effecting an ADAR-mediated adenosine to inosine deamination, thereby editing the PEX1 polynucleotide to encode a PEX1 protein with a wildtype functionality, thereby treating the subject.

22. An in vitro, ex vivo, or in vivo method for the deamination of a target adenosine in an PEX1 target transcript molecule in a cell, the method comprising the steps of:(i) providing the cell with an AON according to any one of claims 1 to 14;(ii) allowing uptake by the cell of the AON;78(iii) allowing annealing of the AON to the target transcript molecule; and(iv) allowing an endogenous ADAR enzyme to deaminate the target adenosine in the target transcript molecule to an inosine.

23. An in vivo method according to claim 22, comprising step (v) of using a functional readout to identify the presence of the inosine in the target transcript molecule.

Images