WO2025149486A1

WO2025149486A1 - Assays for screening and validation of inhibitors of a-to-i rna editing enzymes

Info

Publication number: WO2025149486A1
Application number: PCT/EP2025/050258
Authority: WO
Inventors: Frank H. BÜTTNER; Natalia ZIETARA
Original assignee: Boehringer Ingelheim International GmbH
Current assignee: Boehringer Ingelheim International GmbH
Priority date: 2024-01-08
Filing date: 2025-01-07
Publication date: 2025-07-17
Anticipated expiration: 2026-07-08

Abstract

The present invention provides a high-throughput method for determining dsRNA editing activity of an A-to-l RNA editing enzyme in a competition assay and particularly this high- throughput method allows for screening of inhibitors of an A-to-l RNA editing enzyme, as well as a medium-throughput method for screening and validating of inhibitors of an A-to-l RNA editing enzyme at single A-to-l editing site. Additionally, a high-throughput method is provided for determining binding of a dsRNA binding protein, particularly an A-to-l RNA editing enzyme, to a dsRNA target, and particularly this high-throughput method can be used for screening or characterizing inhibitors of dsRNA binding proteins.

Description

Assays for screening and validation of inhibitors of A-to-l RNA editing enzymes

FIELD OF THE INVENTION

[0001] The present invention provides a high-throughput method for determining dsRNA editing activity of an A-to-l RNA editing enzyme in a competition assay and particularly this high- throughput method allows for screening of inhibitors of an A-to-l RNA editing enzyme, as well as a medium-throughput method for screening and validating of inhibitors of an A-to-l RNA editing enzyme at single A-to-l editing site. Additionally, a high-throughput method is provided for determining binding of a dsRNA binding protein, particularly an A-to-l RNA editing enzyme, to a dsRNA target, and particularly this high-throughput method can be used for screening or characterizing inhibitors of dsRNA binding proteins.

BACKGROUND OF THE INVENTION

[0002] The adenosine deaminase acting on RNA (ADAR) protein catalyzes adenosine-to-inosine (A-to-l) editing, a post-transcriptional modification of double-stranded (ds) RNA, which changes its sequence, coding potential and secondary structure. A-to-l RNA editing results in nucleotide substitution, because later I is recognized as G instead of A both by ribosomes and by RNA polymerases. Besides this change in sequence and resulting re-coding events, A-to-l substitutions can also cause dsRNA destabilization, as l:U mismatch base pairs are less stable than A:U base pairs. Three mammalian ADAR genes are known, of which two encode active deaminases (ADAR1 and ADAR2). A-to-l RNA editing occurs with both viral and cellular RNAs and affects a broad range of biological processes. The primary role of editing by ADAR1 in mammals is to suppress innate immune activation by unedited cell-derived endogenous dsRNA. Changes in RNA editing are associated with various pathological states, from severe autoimmune diseases (deficiency in editing caused by loss-of-function mutations in ADAR1) to cancer (caused by ADAR1 overexpression and silencing of the innate immune response) (Vesely C & Jantsch MF, Genes 2021 , 12(7): 1026, doi: 10.3390/genesl 2071026; Baker AR & Slack FJ, Trends in Genetics 2022, 38(8): P821-830).

[0003] Recent developments, including studies in preclinical cancer models (Ishizuka JJ et al, Nature 2019, 565(7737): 43-48) and the analysis of tumor transcriptome, identified ADAR1 as an attractive candidate for therapeutic inhibition in the context of cancer immunotherapy. Pharmacological inhibition of ADAR1 should result in activation and enhancing of cell intrinsic innate immune signaling in tumors (mainly via type I Interferons) and subsequent immune response towards an otherwise immunologically silent/cold tumors.

[0004] Also, ADAR2 has been implicated in various diseases, particularly neurological and neurodegenerative disease (Yang, Y., et al., 2021 , RNA Biology, 18(7): 999-1013).

[0005] An in vitro deamination assay for ADAR1 is disclosed in Mendoza et al., (Biochemistry, 2023, 62: 1376-1387), however, this assay is not suitable for analysis in medium- or high throughput format, due to the requirement of gel extraction and sequencing.

[0006] To date there are no FDA-approved inhibitors of ADAR1 or ADAR2. Furthermore, there are no publicly available assay protocols which would allow to a) assess in vitro ADAR1 enzymatic activity and/or RNA binding in high-throughput format, and b) to quantify in vitro ADAR1 editing in medium-throughput format, which would be subsequently suitable for independent and complementary inhibitors screening and validation.

SUMMARY OF THE INVENTION

[0007] The present invention provides a toolbox for ADAR (e.g. ADAR1 or ADAR2) inhibitor screening and validation in a medium- or high-throughput format. This toolbox comprises three assays that were developed to work together and are fully compatible in terms of reagents (dsRNA substrates and ADAR proteins). The assays allow, e.g., to assess in vitro ADAR1 enzymatic activity and/or RNA binding, both in high-throughput format, and to quantify in vitro ADAR1 editing in medium-throughput format, which are further suitable for inhibitors screening and validation. Further these assays can be adapted to screen and validate inhibitors of other RNA binding proteins, an emerging class of important targets in immuno-oncology.

[0008] Provided is a method for determining double-stranded (ds) RNA editing activity of an adenosine-to-inosine (A-to-l) RNA editing enzyme comprising: (a) reacting a dsRNA substrate with an A-to-l RNA editing enzyme in a reaction mixture comprising (i) contacting the dsRNA substrate comprising at least one adenosine with the A-to-l RNA editing enzyme, and (ii) incubating the reaction mixture under conditions that allow editing the at least one adenosine in the dsRNA substrate into an inosine thereby forming an edited RNA product comprising at least one inosine; (b) adding an RNA tracer comprising at least one inosine in an RNA, an anti-inosine binding molecule and a donor and acceptor pair to the reaction mixture and detecting binding of the anti-inosine binding molecule to the RNA tracer in an assay using the donor and acceptor pair, wherein (i) the RNA tracer is coupled to one partner of the donor and acceptor pair in the reaction mixture, (ii) the anti-inosine binding molecule is coupled to the other partner of the donor and acceptor pair in the reaction mixture; and wherein the donor and the acceptor are brought into proximity by the binding of the anti-inosine binding molecule to the at least one inosine in the RNA tracer, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal) and wherein the acceptor signal is detected; and (c) measuring competition of the edited RNA product with the RNA tracer for binding to the anti-inosine binding molecule, wherein the presence of edited RNA product reduces the acceptor signal upon excitation of the donor compared to the maximal acceptor signal of a control (positive control) without the A-to-l RNA editing enzyme. In certain embodiments, the method is a method for screening inhibitors of the adenosine-to-inosine (A-to-l) RNA editing enzyme, wherein the method further comprises adding a compound, a reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a). In said method, the compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme reduces or prevents formation of the edited RNA product, wherein the presence of less or no edited RNA product results in an increased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of the library of compounds or reagents. In a preferred embodiment the method for determining double-stranded (ds) RNA editing activity of an adenosine-to-inosine (A-to-l) RNA editing enzyme is a homogenous method.

[0009] In certain embodiments, the RNA tracer (a) is a double-stranded RNA or a single-stranded RNA, preferably a double-stranded RNA; (b) comprises a mammalian RNA; and/or (c) comprises a coupling-label and the donor or acceptor comprises a coupling-label binding proteins, preferably wherein the coupling-label is biotin or digoxigenin and the coupling-label binding protein is a biotin-binding protein or a digoxigenin binding protein.

[0010] In certain embodiments, the anti-inosine binding molecule (a) is an antibody or an aptamer; (b) is coupled directly or indirectly to the other partner of the donor and acceptor pair, and/or (c) is coupled indirectly using an antibody binding the anti-inosine binding molecule, preferably wherein the antibody is coupled via Protein A.

[0011] Also provided is a method for screening or validating inhibitors of an A-to-l RNA editing enzyme in an assay quantifying editing activity of an A-to-l RNA editing enzyme in vitro comprising (a) providing a compound, reagent or at least one member of the library of compounds or reagents; (b) incubating a double-stranded (ds) RNA substrate comprising at least one adenosine (A) and a purified A-to-l RNA editing enzyme in a reaction mixture under conditions allowing A- to-l editing of the dsRNA substrate to form the respective edited RNA product; wherein the compound, reagent or at least one member of the library of compounds or reagents is added prior to or during incubation to the reaction mixture; (c) quantifying edited RNA product and non-edited dsRNA substrate using quantitative real-time PCR (qPCR) comprising (i) purifying RNA from the reaction mixture of step (b) following incubation; (ii) reverse transcribing RNA into cDNA using reverse transcription-polymerase chain reaction (RT-PCR); and (iii) quantifying an amplicon of a sequence within the reverse-transcribed non-edited dsRNA substrate cDNA, and an amplicon of the respective sequence within the reverse-transcribed edited RNA product cDNA using quantitative real-time PCR (qPCR), wherein the sequence comprises the non-edited at least one A reverse-transcribed into deoxythymidine (T) and the edited at least one A reverse transcribed into deoxyguanosine (G), respectively; (d) determining whether the compound, reagent or at least one member of the library of compounds or reagents is an inhibitor of the A-to-l RNA editing enzyme, wherein a compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme results in a decrease ratio of edited RNA product cDNA to non-edited RNA substrate cDNA compared to a control not comprising said compound, reagent or at least one member of the library of compounds or reagents.

[0012] In certain embodiments of the methods of the invention, the A-to-l RNA editing enzyme is Adenosine Deaminase Acting on RNA 1 (ADAR1) or ADAR2, preferably ADAR1. According to the method of the invention the dsRNA substrate is (a) generated by annealing two complementary RNA strands; or (b) a single-stranded RNA that forms a secondary structure comprising double stands, preferably a single-stranded RNA comprising inverse complementary nucleotide sequence regions forming a hairpin structure.

[0013] Also provided is a method for determining binding of a double-stranded (ds) RNA binding protein to a double-stranded (ds) RNA target comprising detecting interaction of a dsRNA binding protein and a dsRNA target in an assay using a bead-based donor and acceptor pair, wherein the dsRNA binding protein is an A-to-l RNA editing enzyme, comprising (a) contacting the dsRNA target with the dsRNA binding protein in a reaction mixture, wherein (i) the dsRNA target is coupled to one partner of the donor and acceptor pair in the reaction mixture; and (ii) the dsRNA binding protein is a fusion protein further comprising an affinity tag, wherein the affinity tag is coupled to the other partner of the donor and acceptor pair in the reaction mixture; and (b) detecting binding of the dsRNA binding protein to the dsRNA target, wherein the donor and the acceptor are brought into proximity by the binding of the RNA binding protein to the dsRNA target, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal), wherein the acceptor signal is detected. Preferably, the assay using a bead-based donor and acceptor pair is a bead-based amplified luminescent proximity homogenous assay. In certain embodiments, the method for determining binding of a dsRNA binding protein to a dsRNA target is a method for screening inhibitors that prevent binding of the dsRNA binding protein to the dsRNA target, wherein the method further comprises adding a compound, reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a). In said method, the compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor that prevents binding of the dsRNA binding protein to the dsRNA target results in a decreased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of a library of compounds or reagents. In a preferred embodiment the method for determining binding of a dsRNA binding protein to a dsRNA target is a homogenous method.

[0014] In certain embodiments, the affinity tag is a polyhistidine-tag. The dsRNA binding protein is an A-to-l RNA editing enzyme, such as ADAR1 , ADAR2 or ADAR3. In certain embodiments, the dsRNA binding protein is selected from the group consisting of an ADAR1-p110 isoform, an ADAR1-p150 isoform, an ADAR1 deaminase domain and dsRNA-binding domain, ADAR2 and ADAR3. According to the method of the invention the dsRNA target is (a) generated by annealing two complementary RNA strands; or (b) a single-stranded RNA that forms a secondary structure comprising double stands, preferably a single-stranded RNA comprising inverse complementary nucleotide sequence regions forming a hairpin structure.

[0015] The compound, reagent or at least one member of a library of compounds or reagents to be screened in the methods of the invention may be (a) a reagent, wherein preferable the reagent is an antibody, an aptamer, a phage, an RNA molecule, an oligonucleotide or a peptide, preferably an oligonucleotide or a peptide; or (b) a compound, wherein the compound is a small molecule compound.

[0016] In certain embodiments, the method for determining double-stranded (ds) RNA editing activity of an adenosine-to-inosine (A-to-l) RNA editing enzyme is a homogenous method. Preferably wherein the signal transmission according to said method is an electron transfer or a singlet oxygen transfer. In certain embodiments the assay using a donor and acceptor pair is a homogenous time-resolved fluorescent assay (HTRF) or a bead-based amplified luminescent proximity homogenous assay.

DESCRIPTION OF THE FIGURES

[0017] Figure 1 : Schematic illustration of the ADAR1 Enzymatic Assay Principle. In a first step illustrated in the box, the enzyme ADAR1 edits adenosine containing RNA substrate in the inosine containing RNA product. For inhibitor screening this reaction is performed in the presence of the test compound, wherein an inhibitor prevents formation of the inosine containing RNA product. In a second step a biotinylated inosine containing RNA tracer (Bio-RNA-I) is added followed by an anti-inosine antibody and a streptavidin donor bead binding the Bio-RNA-I and a Protein A acceptor bead binding a secondary antibody that binds to the anti-inosine antibody, wherein upon excitation the proximity of the donor bead and the acceptor bead results in a detectable acceptor signal emission. The inosine containing RNA product resulting from the ADAR1 mediated A-to-l RNA editing competes with the tracer (Bio-RNA-I) for the anti-inosine antibody, thereby reducing the detectable acceptor signal emission. In the presence of a functional ADAR1 inhibitor inosine containing RNA product formation is suppressed and the acceptor signal emission is high or close to maximal acceptor signal emission.

[0018] Figure 2: Dependency of different biotinylated tracer on the detection range for the RNA product, exemplified for a biotin tracer concentration of 0.5nM. For the enzymatic ADAR1 assay the AlphaScreen detection part was set up to obtain a stable and reproducible S/B ratio and with the detection area of the RNA product within the guasi-linear range of the competition curve. Different biotinylated tracers, (A) RNA-I V5 (std), (B-F) biotin-RNA-l V5_1-5 and (G) biotin RNA-I V5 Z (see Table 4) were employed at a concentration of 0.5 nM. For each of the different biotinylated tracers an RNA V5 Z product was titrated covering a broad concentration range to monitor the detection sensitivity for the product (RNA V5 Z [nM]: 50, 20, 10, 5, 2.5, 1 .3, 0.63, 0.31 . 0.08, 0.04, 0.02, 0.01). Illustrated in each graph is the S/B ratio and location on the curve of 2.5 nM and 10 nM of product.

[0019] Figure 3: Comparison of different concentrations of RNA V5 Z (RNA Edit V5_Z; Table A, SEQ ID NOs: 30 and31) using different ADAR1 constructs, employing the enzymatic activity assay. The results for the ADAR1 constructs DAD (SEQ ID NO: 40), DAD mutant (SEQ ID NO: 44), DAD RBD (SEQ ID NO: 45), ADAR1 p110 (SEQ ID NO: 39), DAD-His (SEQ ID NO: 46), ADAR1 p110-His (SEQ ID NO: 43) are shown from left to right. The left-hand bars (black) show the results using 100 nM substrate (RNA V5 Z) and the right-hand bars (white) show the results using 50 nM substrate (RNA V5 Z) with incubation times of 3 h at 37°C employing 150 nM of each indicated enzyme. S/B ratio indicate the necessity of RNA binding domain.

[0020] Figure 4: (A) Michaelis constant (Km) determination with 5nM ADAR1 p110 and 25 nM V5 Z (RNA Edit V5_Z, Table A, SEQ ID NOs: 30 and 31), employing the enzymatic activity assay, with an incubation time of 3 h, providing a determined Km of ~50 nM. (B) Linearity of product generation over 4 h at 37°C.

[0021] Figure 5: Three independent examples of dose response curves for compounds measured with the enzyme activity assay using 4nM ADAR1 p110 (SEQ ID NO: 39) and 25 nM V5 Z (RNA Edit V5_Z SEQ ID NOs: 30 and 31).

[0022] Figure 6: (A) Z' factors of a full screening campaign for ~1.000.000 different compounds. Mean value over all plates is a Z'factor of 0.9. (B) Gaussian distribution of a full screening campaign for ~1.000.000 different compounds. Mean value of 99.9% CTL and a standard deviation (Stddev) of 1.4% CTL (% of control). (C) Screening run of 45 plates (1536 well plates) indicating the separation window of the high and low controls on each assay plate, wherein the dark circles represent low controls (positive control with no A-to-l RNA editing enzyme), the light circles represent high controls (negative control, 1% DMSO instead of compound) and (x) represents invalid data excluded from analysis.

[0023] Figure 7: Schematic illustration of the ADAR-RNA Binding Assay Principle. In a first step the his-tagged ADAR1 enzyme is incubated with a potential inhibitor followed by addition of a biotinylated RNA substrate (ligand) and incubation under conditions that allow binding of the ADAR1 enzyme to the RNA substrate (ligand). In a second step a streptavidin donor bead binding the biotinylated RNA substrate and an acceptor bead coupled to an anti-his-tag antibody are added to the reaction for detection, wherein upon excitation the proximity of the donor bead and the acceptor bead results in a detectable acceptor signal emission. The presence of an inhibitor that interferes with binding of the ADAR1 enzyme with the RNA substrate reduces the detectable acceptor signal emission.

[0024] Figure 8: (A) dissociation constant (Kd) determination for His tagged ADAR1 p110 using biotinylated-RNA-A (substrate) V5 Z (Biotin EDIT V5_Z; SEQ ID NOs: 34 and 31) employing the ADAR-RNA binding assay. (B) Kd determination for biotinylated RNA-A (substrate) V5 Z using His tagged ADAR1 p110 employing the ADAR-RNA binding assay. (C) Competition curve with untagged ADAR1 p110, IC50 1.78 nM [0025] Figure 9: 3 examples of dose response curves for compounds measured with the RNA- ADAR1 p110 binding assay.

[0026] Figure 10: Principle of medium throughput gPCR assay for the validation of candidate/potential inhibitors of A-to-l RNA editing enzymes comprising: (1) Generation of unedited dsRNA, (2) RNA isolation and A-to-l RNA editing reaction, (3) reverse-transcription PCR (RT-PCR) for cDNA synthesis, (4) guantitative real-time PCR (gPCR) using target specific primers, (5) data analysis showing Ct determination (A) and use of primer pairs differentiating between cDNA derived from edited RNA and non-edited RNA (B).

[0027] Figure 11 : Establishment of assay window for compounds validation. Various ADAR1 and ADAR2 proteins were tested at 250 nM, 500 nM or 750 nM different concentrations for editing AZIN1 as substrate at three different reaction times. Shown is the ratio edited/non-edited (wt) RNA.

[0028] Figure 12: Validation of 9 inhibitory compounds identified using the high-throughput enzymatic activity assay described in Example 1. The inhibitory compounds were used at (A) 10 pM or (B) 100 pM. Shown is % editing normalized to DMSO control. The circles represent each individual editing reaction and the bar represents the mean of the individual measurements.

DETAILED DESCRIPTION OF THE INVENTION

[0029] The present invention provides three assays for studying A-to-l RNA editing enzymes, such as ADARs, and particularly their application for screening and identifying inhibitors of these therapeutically relevant enzymes that interfere with enzyme activity or enzyme binding on a gualitative and/or guantitative level.

[0030] The term “A-to-l RNA editing enzyme” as used herein refers to adenosine to an inosine editing enzymes that mediates the most abundant post-transcriptional RNA modification, such as the enzymes Adenosine Deaminase Acting on RNA (ADARs), that catalyze hydrolytic deamination the amino group at the C6 position of adenosine (A) to produce inosine (I) in doublestranded RNA (dsRNA) substrates. Editing by ADARs results in nucleotide substitution in RNA, because the purine I generated as the result of the deamination reaction is recognized as G instead of A, both by ribosomes during translational decoding of mRNA and by RNA-dependent polymerases during RNA replication. Since Inosine contains a similar chemical structure with guanosine (G), inosine base pairs with cytidine instead of uridine. The double-stranded RNA recognized by ADARs is typically formed by internal repetitive complementary seguences. Due to the inability of hypoxanthine to pair with uracil, the A-to-l RNA editing tends to disrupt dsRNA structures and thereby modulates RNA post-transcriptionally. Inosine is a nucleoside that is formed when hypoxanthine is attached to a ribose ring.

[0031] A-to-l RNA editing occurs in both viral and cellular RNAs and affects a broad range of biological processes. These include virus growth and persistence, apoptosis and embryogenesis, neurotransmitter receptor and ion channel function, pancreatic cell function and post transcriptional gene regulation by microRNAs (miRNAs). These editing sites are also found in various genes expressed in the central nervous system (CNS) and play an important role in neurological development and brain function. A-to-l editing sites have been identified in both coding and non-coding RNA transcripts.

[0032] Three mammalian ADAR genes are known, of which two encode enzymatically active deaminases (ADAR1 and ADAR2). A third gene, ADAR3, is known but has been shown not to encode a catalytically active deaminase domain. All the enzymes contain a conserved deaminase domain and at least two double-stranded RNA binding domains (dsRBDs) that determine the substrate specificity. Alternative promoters together with alternative splicing events give rise to two ADAR1 proteins of different size: an interferon-inducible ADAR1-p150 deaminase that binds dsRNA and Z-DNA and is mainly present in the cytoplasm, and a constitutively expressed ADAR1-p110 deaminase, which is mainly localized in the nucleus, but can re-shuffle into the cytoplasm too. ADAR1-p110 and p150 isoforms have three dsRBDs and further comprise a Z- DNA binding domain, with p150 isoform comprising a Za and a Zp-domain and p110 isoform comprising one Zp-DNA binding domain. ADAR1-p110 is constitutively expressed in most tissues, while ADAR1-p150 isoform is expressed upon induction with type I Interferons (I IFNs). ADAR2 has two dsRBDs and is, like ADAR-p110, constitutively expressed, mainly in brain and testis. ADAR3 has two dsRBDs and an arginine-rich single-stranded RNA binding domain. ADAR3 shows brain-specific expression. Although ADAR3 lacks editing activity, it has been reported that ADAR3 functions as a negative regulator for RNA editing by competing with ADAR1 or ADAR2 for binding to editing substrates (Wang, Y. et al., Nucleic Acids Res. 2019, 47(29): 10801-10814). [0033] An A-to-l RNA editing enzyme or an Adenosine Deaminase Acting on RNA as referred to herein includes mutants and truncated forms thereof, as long as the enzyme has A-to-l RNA editing activity.

[0034] The term “editing” in the context of an A-to-l RNA editing enzyme as used herein refers to catalyzing hydrolytic deamination of the amino group at the C6 position of adenosine (A) to produce inosine (I) in double-stranded RNA (dsRNA) substrates, thereby editing a adenosine containing dsRNA substrate into an inosine containing edited RNA product.

[0035] The term “double-stranded (ds) RNA editing activity” as used herein refers to an adenosine-to inosine (A-to-l) RNA editing activity catalyzing hydrolytic deamination the amino group at the C6 position of adenosine (A) to produce inosine (I) in double-stranded RNA (dsRNA) substrates.

[0036] The term “Z-RNA” as used herein refers to a left-handed alternative conformation for the RNA double helix. Just like for Z-DNA, Z-RNA is favored by a sequence composed of purine/pyrimidine repeats and especially CG repeats. Z-RNA can be detected, e.g., using NMR, circular dichroism and anti-Z-RNA antibodies. Also, the Za-domain of ADAR1 binds and recognizes high affinity Z-RNA. The N-terminus of p150 ADAR1 includes two copies of a Z-DNA- binding domain designated Za and Zp, which are both encoded by exon 2. The p110 protein comprises only Zp and ADAR2 does not comprise a Z-DNA-binding. Although mutations of the Z-DNA-binding domain decreases the efficiency of ADAR1 editing of specific short (15 bp) dsRNA substrates, for the majority of the substrates the Za and Zp domains are not essential for dsRNA binding and/or for dsRNA editing activity of ADARs (Mboukou A et a., doi: https://doi.org/10.1101/2023.12.05.570066).

Enzymatic activity assay (qualitative determination)

[0037] In one aspect the invention relates to a method for determining double-stranded (ds) RNA editing activity of an adenosine-to-inosine (A-to-l) RNA editing enzyme comprising:

(a) reacting a dsRNA substrate with an A-to-l RNA editing enzyme comprising (i) contacting the dsRNA substrate comprising at least one adenosine with the A-to-l RNA editing enzyme in a reaction mixture, and (ii) incubating the reaction mixture under conditions that allow editing the at least one adenosine in the dsRNA substrate into an inosine, thereby forming an edited RNA product comprising at least one inosine;

(b) adding an RNA tracer comprising at least one inosine in an RNA, an anti-inosine binding molecule and a donor and acceptor pair to the reaction mixture and detecting binding of the anti- inosine binding molecule to the RNA tracer in an assay using the donor and acceptor pair, wherein

(i) the RNA tracer is coupled to one partner of the donor and acceptor pair in the reaction mixture,

(ii) the anti-inosine binding molecule is coupled to the other partner of the donor and acceptor pair in the reaction mixture, and wherein the donor and the acceptor are brought into proximity by the binding of the anti-inosine binding molecule to the at least one inosine in the RNA tracer, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal) and wherein the acceptor signal is detected, and

(c) measuring competition of the edited RNA product with the RNA tracer for binding to the antiinosine binding molecule, wherein the presence of edited RNA product reduces the acceptor signal upon excitation of the donor compared to the maximal acceptor signal of a control (positive control) without the A-to-l RNA editing enzyme. In certain embodiments, the A-to-l RNA editing enzyme is an Adenosine Deaminase Acting on RNA (ADAR) enzyme, such as RNA 1 (ADAR1) or ADAR2, preferably ADAR1 . This method is a method for a qualitative determination of dsRNA editing activity of an A-to-l RNA editing enzyme. Moreover, this method follows the principles of a “signal decrease assay”, which means that an A-to-l RNA editing activity increases edited RNA product competing with the RNA tracer for binding to the anti-inosine binding molecule, which decreases or reduces the acceptor signal compared to the positive control without the A-to-l RNA editing enzymes. This assay principle increases sensitivity of the assay. ADARs have a low turnover and are substrate inhibited. Thus, increasing substrate concentration does not increase sensitivity. Further, part of the edited RNA product is maintained within the enzyme and/or is released with a certain delay. Overall, only a small portion of the dsRNA substrate is edited and/or can be detected in an assay. Signal increase assays failed to show the required sensitivity for effective detection of A-to-R RNA editing activity.

[0038] While the method may be used for detecting dsRNA editing activity of an A-to-l RNA editing enzyme, e.g., for comparing different dsRNA substrates or A-to-l RNA editing enzymes, including mutans or truncated forms thereof, the method has been designed for screening inhibitors for an A-to-l RNA editing enzyme. A-to-l RNA editing enzymes have become promising therapeutic targets in various diseases and hence inhibitors may serve as research tool in further studying A-to-l RNA editing, but more importantly inhibitors to A-to-l RNA editing enzymes are potentially useful as compounds for use in therapy. Thus, in certain embodiments of the method of the invention, the method is a method for screening inhibitors of the adenosine-to-inosine (A- to-l) RNA editing enzyme, and wherein the method further comprises adding a compound, a reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a). The compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme, when added to the reaction mixture in step (a) reduces or prevents formation of the edited RNA product, wherein the presence of less or no edited RNA product results in less or no competition of the dsRNA product with the RNA tracer for binding to the anti-inosine binding molecule and an increased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of the library of compounds or reagents. The compound, reagent or at least one member of the library of compounds or reagents may be added to the reaction mixture prior to, simultaneously or after contacting the dsRNA substrate with the A-to-l RNA editing substrate. Preferably the compound, reagent or at least one member of the library of compounds or reagents is added to the reaction mixture prior to contacting the dsRNA substrate with the A-to-l RNA editing enzyme, more preferably the compound, reagent or at least one member of the library of compounds or reagents is preincubated with the A-to-l editing enzyme. The term “reaction mixture” as used herein refers to an aqueous solution comprising a buffer, and the specified components. Preferably the buffer has an about neutral pH, such as pH 6.5 to pH 8, preferably pH 7.0 to 7.5. The reaction mixture typically further comprises salts and/or RNAse inhibitors.

[0039] The inhibitor for an A-to-l RNA editing enzyme may be any compound or reagent that inhibits enzyme activity of an A-to-l RNA editing enzyme and/or that binds to an A-to-l RNA editing enzyme, and/or inhibits dsRNA substrate binding to the A-to-l RNA editing enzyme, and/or modifies the conformation of the enzyme thereby inhibiting the activity of or dsRNA substrate binding to the A-to-l RNA editing enzyme and/or prevents conformational change upon dsRNA substrate binding, preferably that inhibits enzyme activity of an A-to-l RNA editing enzyme. The inhibition may be reversible or irreversible, preferably reversible. In certain embodiments, the compound, reagent or at least one member of a library of compounds or reagents to be screened is (a) a reagent, wherein preferable the reagent is an antibody, an aptamer, a phage, an RNA, a peptide, a morpholino or an oligonucleotide, preferably a peptide or an oligonucleotide; or (b) a compound, wherein the compound is a small molecule compound. The term "small molecule as used herein refers to a low molecular weight (< 1000 daltons) organic compound. This term is used in contrast to larger polymeric structures such as nucleic acids and proteins and larger polysaccharides, although their constituent monomers (ribo- or deoxyribonucleotides, amino acids and monosaccharides) are typically small molecules. An oligonucleotide is a short DNA or RNA molecule (oligomer), wherein short means about 30 nucleotides or less, whereas RNA molecules are typically understood to be larger than 30 nucleotides. Aptamers as used herein are short sequences of artificial DNA, RNA, XNA or peptide that bind a specific target molecule and are also referred to as chemical antibodies. Xeno nucleic acids (XNA) are synthetic nucleic acid analogues that have a different backbone than the ribose or deoxyribose found in RNA and DNA. Aptamers and antibodies can be used in similar applications. Preferably the aptamer or antibody is an anti-A-to-l RNA editing enzyme antibody, i.e., directed against the enzyme. The term antibody is used herein it its broad sense and includes antibody fragments, such as Fab or F(ab’)2 fragments or single chain variable fragments (scFv), diabodies, minibodies and the like. Moreover, it refers to monospecific as well as bispecific or multi-specific antibodies and antibody format. It also includes all isotypes, including IgG, IgM, IgA, IgD or IgE, preferably IgG, particularly lgG1 , lgG2a, lgG2b, lgG3 and lgG4, preferably lgG1. For therapeutic uses an identified A-to-l RNA editing enzyme inhibitor, said inhibitor needs to act intracellularly, therefore small reagents or small molecule compounds may be preferred compounds, reagents or members of a library of compounds or reagents to be screened.

[0040] The method is suitable for high-throughput screening and is preferably performed using multiwell plates, preferably 96-well plates, more preferably 384-well plates or 1538-well plates. Moreover, the method is a homogenous method. The term “homogeneous method” as used herein refers to a “mix and measure” assay comprising no washing and no separation steps, i.e., an assay in which separation steps and washing steps are not required and have been eliminated. Since separation steps and washing steps are time consuming and increase the hand-on time of a method, homogenous methods comprising no washing and no separation steps are particularly suitable for high-throughput screening. Further, the method may be measured overtime (including in real-time following initiation of step (b)) or using end-point measurement. Preferably the RNA tracer, the anti-inosine binding molecule and the donor and acceptor pair in the reaction mixture are incubated for at least 30 min, preferably for at least 1 hour, e.g., between 15 and 37°C, preferably at room temperature (21 °C) or about 24°C.

[0041] Step (a) describes an enzymatic reaction. Thus, incubating the reaction mixture under conditions that allow editing the at least one adenosine in the dsRNA substrate into an inosine is typically performed at about 15°C to 37°C (preferably about 21 °C to 37°C, more preferably about 37°C) and may be performed for, e.g., 20 min or more, 30 min or more, 1 hour or more, preferably 2 hours or more, more preferably 2.5 hours or more. The incubation step may further be followed by a heating step to release the edited RNA product more efficiently from the enzyme. While steps (a) and (b) may be performed simultaneously, preferably the dsRNA substrate and the A-to-l RNA editing enzyme in the reaction mixture are incubated for a certain amount of time (preferably at about 37°C for 30 min to 1 hour or more) before the RNA tracer, the anti-inosine binding molecule and/or the donor and acceptor pair is added, more preferably before the RNA tracer, the anti- inosine binding molecule and the donor and acceptor pair is added. [0042] The method according to the invention comprises adding a donor and acceptor pair and detecting binding of the anti-inosine binding molecule to the RNA tracer in an assay using the donor and the acceptor pair, wherein the donor and acceptor pair are brought into proximity by the binding of the anti-inosine binding molecule to the RNA tracer, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal), and wherein the donor signal is detected. The signal transmission may be an electron transfer or a singlet oxygen transfer. Preferably the signal transmission is a singlet oxygen. In certain embodiments, the assay using a donor and acceptor pair is a homogenous time-resolved fluorescent assay (HTRF) or a bead-based amplified luminescent proximity homogenous assay. The donor and acceptor pair may be a fluorescent donor and acceptor pair and the assay using the donor and acceptor pair is a homogenous time-resolved fluorescent assay (HTRF), or the acceptor pair may be a beadbased donor and acceptor pair and the assay using the donor and acceptor pair is a bead-based amplified luminescence proximity homogenous assay, preferably the acceptor pair is a beadbased donor and acceptor pair and the assay using the donor and acceptor pair is a bead-based amplified luminescence proximity homogenous assay.

[0043] An homogenous time-resolved fluorescence assay (HTRF) (commercially available as LANCE™ (Wallac Oy, Turku, Finland) or HTRF® assay (PerkinElmer Cisbio) is based on the principle of time-resolved fluorescence (Forster) resonance energy transfer (TR-FRET), which combines the benefits of time-resolved fluorescence (FRT) with those of Forster resonance energy transfer (FRET), first described in 1988 by Morrison L. E. (Anal. Biochem. 174: 101-120). FRET allows the detection of two molecules within several nanometers. The mechanism of FRET involves a donor fluorophore in an excited electronic state which may transfer its excitation energy to a nearby acceptor chromophore in a non-radiative fashion through long-range dipole-dipole interaction. The europium chelate complex possesses a longer fluorescence lifetime (ps) than traditional fluorophores (ns), allowing signal to be collected beyond the lifetime of background fluorescence. For example, when exited at a wavelength of 320 nm, the interaction of an europium chelate (donor)-labeled first binding partner with an allophycocyanine (APC) (acceptor)-labeled second binding partner results in transfer of energy to APC, leading to maximal emission of a wavelength of 650 nm. This energy transfer is time-gated in order to reduce short-lived fluorescent background. The detection of binding distance is limited to 9 nm. Thus, in certain embodiments the donor and acceptor pair is a fluorescent donor and acceptor pair, such as an europium chelate (donor) and allophycocyanine (acceptor). However, other fluorescent donor and acceptor pairs are known in the art, including without being limited thereto fluorescent protein FRET pairs, such as CFP-GFP, CFP-YFP, eGFP-mCherry, fluorescent protein-dye FRET pairs, such as EGFR- Alexa Fluor 555, 546, 594 or 568 and dye-dye-biofluorochrome FRET pairs, such as Cy2-Cy3, Cy3-Cy5, FITC-TRITC, Europium-APC, Europium-Cy5, Atto 488-Atto 647N or 590, Alexa 488- Alexa 430, 514, 532, 546 or 610.

[0044] More recently a bead-based, non-radioactive amplified luminescent proximity homogeneous assay (first described in 1994 by Ullman, E. F., et al., Proc. Natl. Acad. Sci USA, 91(12): 5426-5430) and based on the principle of luminescence oxygen channeling and compatible for high-throughput screening has become commercially available (AlphaScreen). In this assay a donor and acceptor pair (donor and acceptor beads of 250 nm diameter reagent- coated polystyrene microbeads) are brought into proximity by a biomolecular interaction of binding partners immobilized to the beads. When the biological interaction brings the donor and acceptor beads together, a cascade of chemical reactions acts to produce a greatly amplified signal. On laser excitation (at 680 nm), a photosensitizer in the “donor” bead converts ambient oxygen to a more excited singlet state. The singlet state oxygen molecules diffuse across (up to 200 nm) to react with a thioxene derivative in the acceptor bead generating chemiluminescence at 370 nm that further activates fluorophores contained in the same bead. The fluorophores subsequently emit light at ~520-620 nm using a classical acceptor bead or europium is activated in an AlphaLisa Acceptor bead and light is emitted at ~615 nm. In the absence of a specific biological interaction, the singlet state oxygen molecules produced by the donor go undetected without the close proximity of the acceptor. As a result, only a very low background signal is produced. The donor bead generates about 60,000 singlet oxygen molecules, resulting in an amplified signal. Singlet oxygen has a short lifetime in aqueous solution (~4 psec) which allows a diffusion over a distance up to ~200 nm. Because the signal is very long lived, with a half-life in the second range, the detection system can be time-gated, thus eliminating short-lived background (AlphaScreen signal is measured with a delay between illumination and detection of 20 ms). Furthermore, the detection wavelength is shorter than the excitation wavelength, thus further reducing potential for fluorescence interference. The sensitivity of the assay derives from the very low background fluorescence. AlphaScreen provides a highly versatile, sensitive, homogeneous and miniaturizable technology to efficiently perform assay development and HTS resulting in higher throughput at lower costs. A comparable technology is Enlight OMEGA (Oxygen Mediated/Gated Assay; BioSignal² Inc. CA), were the beads have an extension wavelength of 680 nm and an emission wavelength of 615 nm. Thus, in certain embodiments the donor and acceptor pair is a bead-based donor and acceptor pair.

[0045] For analysis the detected counts of the test sample are normalized based on positive and negative control. Thus, the detected counts of the positive control (no A-to-l RNA editing enzyme; highest signal) are subtracted from the counts of the test sample (with the potential inhibitor) as well as the counts of the negative control (no inhibitor; lowest signal) and expressed as a ratio or percent of counts test sample (subtracted): counts negative control (subtracted), according to the following formula (for %control):

%CTL = 100 * ((Counts test sample - counts positive control) I (counts negative control - counts positive control))

Although test compounds or reagents may interfere with the assay reagents, the particular set up of the method based on a “signal decrease assay” reduces detecting false positive compounds or reagents. Interfering of a test compound or reagent with assay reagents (e.g. capturing oxygen or quenching fluorescence or luminescence) similarly affects the positive control and hence normalization to the positive control minimizes detection of false positive results. The method according to the invention therefore provides a highly reliable assay for detecting inhibitors of an A-to-l RNA editing enzyme, wherein the chance of detecting false positive compounds is minimized or reduced. The term “test sample” as used herein refers to a reaction mixture comprising a compound, reagent or at least one member of a library of compounds or reagents, e.g., to be tested as inhibitor of the A-to-l RNA editing enzyme (in the context of the method for screening inhibitors of an A-to-l RNA editing enzyme), but may also relate, without being limited thereto, e.g., to another A-to-l RNA editing enzyme, a mutant, variant, truncated form or fusion protein of an A-to-l RNA editing enzyme mutant ora truncated form thereof to be tested compared to the negative control (comprising a reference A-to-l RNA editing enzyme), as well as to another dsRNA substrate to be tested compared to the negative control (comprising a reference substrate).

[0046] The dsRNA substrate comprising at least one adenosine may also be referred to as “dsRNA substrate” only and is to be understood to always comprising at least one adenosine that may be edited into an inosine by an A-to-l RNA editing enzyme. Thus, the at least one adenosine is at least one A-to-l editing site, preferably a natural occurring A-to-l editing site. Typically, a dsRNA substrate for an A-to-l RNA editing enzyme comprises more than one adenosine (or A-to- I editing site), such as two, three, four or more adenosines (or A-to-l editing sites). The double- stranded RNA may be formed by two annealed complementary RNA strands or a single-stranded RNA that forms a secondary structure comprising double strands, such as a hairpin structures. A hairpin structure (also referred to as stem-loop or hairpin loop) describes a secondary structure of RNA, more specifically an intramolecular base pairing of a single-stranded RNA that occurs then two regions of the same strand, usually inverse complementary in nucleotide sequence, base-pair to form a double helix that ends in an unpaired loop. Thus, in certain embodiments, the dsRNA substrate is generated by annealing two complementary RNA strands. In other embodiments the dsRNA substrate is a single-stranded RNA that forms a secondary structure comprising double stands, preferably comprising inverse complementary nucleotide sequence regions forming a double strand, such as a hairpin structure. For example, two inverted Alu- elements give rise to double-stranded RNA. The dsRNA substrate may also comprise other secondary structures, such as a bulge or an internal loop, which may occur in both, (a) two complementary RNA strands, or (b) ssRNA that forms a secondary structure comprising double strands. The dsRNA typically forms a double helix (tertiary structure), which may be in the A- conformation or the Z-conformation. The term “Z-RNA” as used herein refers to the high-energy, left-handed conformation for the RNA double helix, while A-RNA describes a right-handed conformation of the RNA double helix. Although the right-handed B- and A-conformations of helical DNA and RNA are structurally distinct from each other, the helical parameters of Z-DNA and Z-RNA are very similar. Unlike the phosphate backbones of A-RNA, which follow a smooth curve, the backbone of the Z-conformation zig-zags.

[0047] The dsRNA substrate can be perfectly matched dsRNA or can have a more complex secondary structure, such as structures common in cellular RNA, including hairpin structures, bulges and internal loops. The dsRNA substrate may comprise adenosines within more complex secondary structures, such as hairpin structures, bulges and internal loops, and/or adenosines found in A-U pairs or in A-C mismatches, preferably in A-C mismatches. An A-U pair describes base paring between adenosine and uridine, while A-C mismatches describes adenosines found opposite cytidines, which are unable to undergo base paring. For example, ADARs have a preference for adenosines within a certain local sequence context, e.g., 5’-UAG-3’, and can bind to perfectly matched double-stranded RNA but can also be highly selective for specific adenosines found within more complex secondary structures common in cellular RNA. In addition, most efficient editing occurs at adenosines found opposite C, while A-U pairs are also edited. A- A and A-G mismatches are poorly tolerated by ADARs. Duplex RNA imperfections, such as mismatches, bulges, internal loops and hairpins in many naturally occurring RNA substrates play important roles in determining which adenosines are efficiently edited. Examples of ADAR dsRNA substrates (particularly ADAR1 or ADAR2 dsRNA substrates) are, e.g., the RNAs NEIL1 , TTYH2 and AJUBA and derivatives thereof, as described by Liu, X., et al., (Nat. Commun, 2021 , 12(1): 2165, doi: 10.1038/s41467-021-22489-2), or RNA-A V5-Z (“substrate”): RNA sequence (SEQ ID NO: 30): CGCGCGCGCGCGGGACAAAUUAACCAAGGAAAAUAACAAGGACAGGGACC

RNA sequence reverse (SEQ ID NO: 31):

GGUCCCUGUCCUUGUUAUUUUCCUUGGUUAAUUUGUCCCGCGCGCGCGCG

[0048] In certain embodiments the dsRNA substrate comprises a Z-RNA formation, preferably comprising a (CG)₆-repeat at its 5’ end, more preferably comprising the sequence of SEQ ID NO: 1 and/or SEQ ID NO: 4. Preferably the dsRNA substrate comprises at least 49 bp, (ii) is a noncoding RNA, (iii) has an alu-element, and/or (iv) is a naturally occurring mammalian RNA.

[0049] The edited RNA product comprising at least one inosine may also be referred to as “edited RNA product” only and is to be understood to always comprising at least one inosine that may be edited from an adenosine by an A-to-l RNA editing enzyme. Since the edited RNA product is the reaction product of the A-to-l RNA editing enzyme, it is the reaction product of the dsRNA substrate as described herein with the at least one adenosine edited into an inosine. Since the editing from an adenosine to an inosine may disrupt double-stranded RNA, the edited RNA product may be single-stranded or double-stranded or a mixture thereof.

[0050] The RNA tracer comprising at least one inosine in an RNA may also be referred to as “RNA tracer” only and is to be understood to always comprising at least one inosine in an RNA. Any RNA sequence comprising at least one inosine is suitable as RNA tracer, provided that the anti-inosine binding molecule binds to the at least one inosine in the RNA tracer comprising at least one inosine. The RNA tracer may be single-stranded or double-stranded. Preferably the RNA tracer is a double-stranded RNA. In certain embodiments, the RNA tracer comprises a mammalian RNA, such as a natural occurring mammalian RNA comprising at least one inosine. In a preferred embodiment the RNA tracer is the edited RNA product comprising at least one inosine. Thus, the RNA tracer may comprise the A-to-l edited sequence of the dsRNA substrate. A suitable RNA tracer is, e.g., RNA-I V5_1 :

RNA sequence (SEQ ID NO: 15):

CGGCCGGCCGGGGCCCCCUUIICCIIGGCCCCUCCCCCGGACAGGGACC RNA sequence reverse (SEQ ID NO: 16):

GGUCCCUGUCCGGGGGAGGGGCCUUGGUUAAGGGGGCCCCGGCCGGCCGG

More preferably the RNA tracer is biotinylated, e.g., Bio-RNA-I V5_1 : Biotinylated RNA sequence (SEQ ID NO: 25): Biotin-CGGCCGGCCGGGGCCCCCUUIICCIIGGCCCCUCCCCCGGACAGGGACC RNA sequence reverse (SEQ ID NO: 16): GGUCCCUGUCCGGGGGAGGGGCCUUGGUUAAGGGGGCCCCGGCCGGCCGG [0051] According to the method of the invention the RNA tracer is coupled to at least one partner of the donor and acceptor pair, preferably to the donor. The coupling to one partner of the donor or acceptor pair includes covalent binding and non-covalent binding. For example, the RNA tracer comprises a coupling label that binds non-covalently to a coupling-label binding protein on the donor or acceptor, preferably the donor. In certain embodiments the RNA tracer comprises a coupling label and the donor or acceptor comprises a coupling-label binding protein. Preferably the coupling-label is biotin or digoxigenin and the coupling-label binding protein is a biotin-binding protein or a digoxigenin binding protein. Wherein the biotin is covalently linked to the RNA tracer (biotinylated RNA tracer), preferably to the 5’- or 3’-end of the RNA tracer, more preferably the d’end of the RNA tracer. In the case of a double-stranded RNA tracer labeling one strand is sufficient. Methods for biotinylating RNA molecules, including end-terminal labeling are known in the art. Biotin-binding proteins are further known in the art and include without being limited thereto streptavidin and avidin. Suitable donor and acceptor pairs comprising streptavidin or avidin (preferably covalently linked) are known in the art and commercially available.

[0052] Digoxigenin (DIG) is a steroid small molecule with high antigenicity that is used as an immuno-tag in many molecular biology applications. Anti-digoxigenin antibodies (or derivatives thereof) with high affinities are commercially available and are used in a variety of biological immuno-assays (e.g., ELISA and in situ hybridization) or coupling of DIG-labelled molecules, particularly RNA or DNA. It may be conjugated to a single species of RNA nucleoside triphosphate (typically uridine), which is then incorporated into RNA (a riboprobe) as it is snythesized by the cellular machinery. Alternatively, it may be introduced chemically (conjugation) to the RNA tracer. Digoxigenin binding proteins may be antibodies specific for digoxygenin. Alternatively an artificially designed DIG-binding protein as described by Tinberg, C.E., et al., (Nature, 2013, 501 (7466): 212-216; doi:10.1038/nature12443. PMC 3898436) may be used. In certain embodiments the RNA tracer is biotinylated and the donor or acceptor comprises streptavidin or avidin or the RNA tracer is labelled with digoxigenin and the donor or acceptor comprises an antibody against digoxygenin (directly or indirectly bound via a secondary antibody).

[0053] According to the method of the invention the anti-inosine binding molecule binds to the RNA tracer, i.e., it binds to the at least one inosine in the RNA tracer comprising at least one inosine. Preferably, the anti-inosine binding molecule, is an anti-inosine binding aptamer or an anti-inosine binding antibody, preferably an anti-inosine binding antibody. The anti-inosine binding molecule is coupled (directly or indirectly) to the other partner of the donor and acceptor pair, preferably the acceptor. The coupling includes covalent binding and non-covalent binding. Preferably, the anti-inosine binding molecule is coupled indirectly using an antibody binding the anti-inosine binding molecule, preferably wherein the antibody is coupled via its Fc domain, such as Protein A. More preferably the anti-inosine binding molecule is an anti-inosine antibody coupled indirectly using a secondary antibody binding said anti-inosine antibody and wherein the secondary antibody is coupled via its Fc domain, such as via Protein A to the other partner of the donor or acceptor pair. Anti-inosine antibodies are known in the art and commercially available, such as a monoclonal mouse anti-inosine IgG antibody from Diagenode (Cat. No. C15200251).

[0054] The method according to the invention is for determining dsRNA editing activity of an adenosine-to inosine (A-to-l) RNA editing enzyme, including without being limited thereto Adenosine Deaminase Acting on RNA (ADAR) enzymes, particularly Adenosine Deaminase Acting on RNA 1 (ADAR1) or Adenosine Deaminase Acting on RNA 2 (ADAR2). This also includes mutant and truncated forms of ADAR1 or ADAR2. The person skilled in the art would understand that mutant and truncated forms of ADAR1 or ADAR2 with dsRNA editing activity are suitable for the method according to the present invention. This includes for example full-length (FL) ADAR1 , isoform p110, full-length (FL) ADAR1 , isoform p150, truncated ADAR1 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs), functional ADAR1 mutants, full length (FL) ADAR2, truncated ADAR2 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs) and functional ADAR2 mutants. In certain embodiments, the A-to-l RNA editing enzyme is an Adenosine Deaminase Acting on RNA (ADAR) enzyme, such as RNA 1 (ADAR1) or ADAR2, preferably ADAR1 . In certain embodiments, the ADAR enzyme comprises a sequence selected from the group consisting of SEQ ID NOs: 39, 40, 41 , 42 and 45. Quantitative medium-throughput enzymatic activity assay

[0055] In another aspect of the invention a method for screening and validating of (potential) inhibitors of an A-to-l RNA editing enzyme in an assay quantifying editing activity of an A-to-l RNA editing enzyme in vitro comprising

(a) providing a compound, reagent or at least one member of the library of compounds or reagents;

(b) incubating a double-stranded (ds) RNA substrate comprising at least one adenosine (A) and an A-to-l RNA editing enzyme (preferably a purified A-to-l RNA editing enzyme) in a reaction mixture under conditions allowing A-to-l editing of the dsRNA substrate to form the respective edited RNA product; wherein the compound, reagent or at least one member of the library of compounds or reagents is added prior to or during incubation to the reaction mixture;

(c) quantifying edited RNA product and non-edited dsRNA substrate using quantitative realtime PCR (qPCR) comprising (i) purifying RNA from the reaction mixture of step (b) following incubation, (ii) reverse transcribing RNA into cDNA using reverse transcription-polymerase chain reaction (RT-PCR) and (iii) quantifying an amplicon of a sequence within the reverse- transcribed non-edited dsRNA substrate cDNA, and an amplicon of the respective sequence within the reverse-transcribed edited RNA product cDNA using quantitative realtime PCR (qPCR), wherein the sequence comprises the non-edited at least one A reverse- transcribed into deoxythymidine (T) and the edited at least one A reverse transcribed into deoxyguanosine (G), respectively; and

(d) determining whether the compound, reagent or at least one member of the library of compounds or reagents is an inhibitor of the A-to-l RNA editing enzyme, wherein a compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme results in a decreased ratio of edited RNA product cDNA to non-edited RNA substrate cDNA compared to a control not comprising said compound, reagent or at least one member of the library of compounds or reagents. An overview of the assay principle is provided in Figure 10.

[0056] This method provides a medium-throughput assay to quantify A-to-l RNA editing and is particularly useful for inhibitor validation, particularly in combination with the method for determining dsRNA editing activity of an A-to-l RNA editing enzyme (qualitative determination) and the associated method for screening inhibitors of the A-to-l RNA editing enzyme. Thus, this method provides an independent validation of the enzymatic activity assay described herein. In certain embodiments of the method, the RNA is purified in multiwell plates from the reaction mixture of step (b) following incubation, such as 96-well plates or multiples thereof. Methods for RNA purification are known in the art, specifically medium-throughput methods are suitable in the context of the present invention such as silica-membrane spin columns or 96-well plates (e.g., miRNAeasy microKit; Quiagen). The skilled person will understand that the RNA is purified as total RNA. The method is performed using purified A-to-l RNA editing enzyme, preferably purified recombinant A-to-l RNA editing enzyme. For example, the recombinant A-to-l RNA editing enzyme may comprise an affinity tag for easier purification (e.g., a his-tag or another affinity tag as described herein). The term “purified A-to-l RNA editing enzyme” refers to the enzyme following at least one step of purification, such as precipitation or preferably column purification and does not encompass a cell lysate or cell extract.

[0057] The A-to-l RNA editing enzyme suitable for this method as is defined for the method for determining dsRNA editing activity of an A-to-l RNA editing enzyme according to the invention, including without being limited thereto Adenosine Deaminase Acting on RNA (ADAR) enzymes, particularly Adenosine Deaminase Acting on RNA 1 (ADAR1) or Adenosine Deaminase Acting on RNA 2 (ADAR2). This also includes mutant and truncated forms of ADAR1 or ADAR2. The person skilled in the art would understand that mutant and truncated forms of ADAR1 or ADAR2 with dsRNA editing activity are suitable for the method according to the present invention. This includes for example full-length (FL) ADAR1 , isoform p110, full-length (FL) ADAR1 , isoform p150, truncated ADAR1 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs), functional ADAR1 mutants, full length (FL) ADAR2, truncated ADAR2 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs) and functional ADAR2 mutants. In certain embodiments, the A-to-l RNA editing enzyme is an Adenosine Deaminase Acting on RNA (ADAR) enzyme, such as RNA 1 (ADAR1) or ADAR2, preferably ADAR1. In certain embodiments, the ADAR enzyme comprises a sequence selected from the group consisting of SEQ ID NOs: 39, 40, 41 , 42 and 45.

[0058] The double-stranded (ds) RNA substrate used in this method may be a synthetic dsRNA substrate as defined herein above for the method for determining dsRNA editing activity of an A- to-l RNA editing enzyme according to the invention. Preferably the dsRNA substrate is produced and isolated from mammalian cells transfected with a plasmid encoding and overexpressing said dsRNA substrate in the mammalian cells, such as Antizyme inhibitor 1 (AZIN1) RNA, Mouse Double Minute 2 (MDM2) RNA, or Cyclin Dependent Kinase 13 (CDK13), which are known RNA editing substrates for ADAR1 and also ADAR2 at certain conditions in vitro. More preferably the dsRNA substrate is produced and isolated from mammalian cell deficient in ADAR1 (p110 and p150) and/or ADAR2, such as HCT-116 ADAR1 knock-out cells generated from the human colon carcinoma cell line HCT-116. Producing and isolating dsRNA substrate from mammalian cells, provides RNA substrates in its native context and using ADAR1 and/or ADAR2 deficient cells, preferably ADAR1 deficient cells, increases the purity of the (non-edited) dsRNA substrate. Isolation of dsRNA substrate from mammalian cells means total RNA isolation and incubating total RNA comprising a double-stranded (ds) RNA substrate comprising at least one adenosine (A) and an A-to-l RNA editing enzyme (preferably a purified A-to-l RNA editing enzyme) in a reaction mixture in step (b).

[0059] In step (b) the method comprises incubating the dsRNA substrate comprising at least one adenosine and a purified A-to-l RNA editing enzyme in a reaction mixture. Thus, the dsRNA substrate comprises at least one A-to-l editing site as described herein. Step (b) describes an enzymatic reaction. Thus, incubating the reaction mixture under conditions that allow editing the at least one adenosine in the dsRNA substrate into an inosine is typically, without being limited thereto, performed at about 37°C for 30-60 min or more, preferably 45-60 min. Following incubation (i.e. , following step (b) and priorto RNA purification in step (c)(i)), the A-to-l RNA editing enzyme is preferably heat-inactivated, such as for about 5 min at about 70°C.

[0060] The edited RNA product and the non-edited dsRNA substrate are quantified in step (c) using quantitative real-time PCR (qPCR), more specifically qPCR following reverse transcription of the RNA into cDNA. Step (c) comprises (i) purifying RNA from the reaction mixture of step (b) following incubation, followed by (ii) reverse transcribing RNA into cDNA using reverse transcription-polymerase chain reaction (RT-PCR). The person skilled in the art would understand that RNA purification is typically total RNA purification as described above, preferably using multiwell plates, such as 96-well plates or multiples thereof. Methods for reverse transcribing RNA into cDNA using reverse transcription-polymerase chain reaction (RT-PCR), i.e., first-strand cDNA synthesis, are known in the art. Typically primers targeting mRNA in a target-unspecific manner, such as oligo(dT)-primer are used for reverse transcription, although theoretically also target specific primers may be used. Reverse-transcriptases are enzymes able to polymerize a strand of DNA (cDNA) that is complementary to an original RNA template and suitable reversetranscriptases, such as Super Script (II) (Invitrogen) are known in the art. RT-PCR further uses the cDNA for PCR amplification. Reverse-transcription-qPCR (RT-qPCR) combines RT-PCR amplification with detection and quantification (qPCR). Thus, step (c) further comprises amplifying cDNA. Amplification of cDNA provides an amplicon. The amplicon is typically generated using target specific primer pairs. [0061] The method further comprises (iii) quantifying an amplicon of a sequence within the reverse-transcribed non-edited dsRNA substrate cDNA, and an amplicon of the respective sequence within the reverse-transcribed edited RNA product cDNA using quantitative real-time PCR (qPCR), wherein the sequence comprises the non-edited at least one A reverse-transcribed into deoxythymidine (T) and the edited at least one A reverse transcribed into deoxyguanosine (G), respectively. The term “amplicon” as used herein is a piece of DNA that is the product of an amplification reaction, particularly wherein the amplification reaction is quantitative real-time PCR (qPCR). The terms “quantitative PCR”, “real-time PCR” or “quantitative real-time PCR” are used synonymously herein and abbreviated as qPCR and describes monitoring the amplification of a target DNA molecule during the PCR (i.e. , in real-time), which allows quantification. Methods for qPCR and differentiation and quantification of amplicons are known in the art. Two common methods for the detection of PCR products (amplicons) in quantitative real-time PCR (qPCR) are (1) non-specific fluorescent dyes that intercalate with any double-stranded DNA (e.g., SYBR green) and (2) sequence-specific DNA probes consisting of oligonucleotides that are labelled with a fluorescent reporter, which permits detection only after hybridization of the probe with its complementary sequence. While both qPCR methods may be applied in step (c) of this method of the present invention, qPCR method (1) using non-specific fluorescent dyes that intercalate with any double-stranded DNA (e.g., SYBR green) is preferred. For example, primer pairs binding differentially to reverse-transcribed non-edited dsRNA substrate cDNA and reverse transcribed edited RNA product cDNA may be used for amplification (in separate vials or wells) and amplification detected using a fluorescent dye that intercalate with any double-stranded DNA (e.g., SYBR green). During editing A is edited into I, which is transcribed into guanosine (or deoxyguanosine in a DNA sequence). Thus, primers can be designed to bind to and amplify reverse-transcribed edited or non-edited A-to-l RNA editing sites. Editing may be quantified by providing the ratio of edited vs. non-edited (wild type, WT) RNA or the percentage of edited RNA. [0062] In step (d) the method comprises determining whether the compound, reagent or at least one member of the library of compounds or reagents is an inhibitor of the A-to-l RNA editing enzyme, wherein a compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme results in a decreased ratio of edited RNA product cDNA to non-edited RNA substrate cDNA compared to a control not comprising said compound, reagent or at least one member of the library of compounds or reagents. The control is typically treated identical to a sample comprising the compound, reagent or at least one member of the library of compounds or reagents in the reaction mixture with the solvent used for diluting the compound or reagent at the same amount as present when adding the compound, reagent or at least one member of the library of compounds or reagents.

[0063] In a preferred embodiment multiple compounds, reagents or members of a library of compounds or reagents is tested in parallel using the method for screening or validating of (potential) inhibitors of an A-to-l RNA editing enzyme in an assay quantifying editing activity of an A-to-l RNA editing enzyme in vitro, such as 2 or more, 3 or more, 4 or more, 5 or more, 10 or more, 25 or more or 100 or more compounds, reagents or members of a library of compounds or reagents. Preferably the method is performed using one or more 96-well plates or multiples thereof, such as 384-well plates, more preferably the RT-qPCR reaction (such as steps (c)(ii) and (iii)) is performed using 384 well plates and the ADAR1 reaction (such as step b) and RNA isolation using 96 well plates (such as step c(i) and optionally generation of the dsRNA substrate). dsRNA Binding Assay

[0064] In another aspect, the invention relates to a method for determining binding of a doublestranded (ds) RNA binding protein to a double-stranded (ds) RNA target comprising detecting interaction of a dsRNA binding protein and a dsRNA target in an assay using a bead-based donor and acceptor pair, wherein the dsRNA binding protein is an A-to-l RNA editing enzyme, comprising

(a) contacting the dsRNA target with the dsRNA binding protein in a reaction mixture, wherein (i) the dsRNA target is coupled to one partner of the donor and acceptor pair in the reaction mixture; and (ii) the dsRNA binding protein is a fusion protein further comprising an affinity tag, wherein the affinity tag is coupled to the other partner of the donor and acceptor pair in the reaction mixture, and

(b) detecting binding of the dsRNA binding protein to the dsRNA target, wherein the donor and the acceptor are brought into proximity by the binding of the RNA binding protein to the dsRNA target, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal), wherein the acceptor signal is detected. According to the invention, the dsRNA binding protein is an A-to-l RNA editing enzyme, such as an Adenosine Deaminase Acting on RNA (ADAR) enzyme, e.g., RNA 1 (ADAR1), ADAR2 or ADAR3.

[0065] While the method may be used for detecting dsRNA binding of an dsRNA binding protein, e.g., for comparing different dsRNA targets or dsRNA binding protein variants, including mutans or truncated forms thereof, the method has been designed for screening (binding) inhibitors for dsRNA binding proteins, specifically A-to-l RNA editing enzymes. Particularly this method may be combined with the enzymatic activity assay described above to further characterize inhibitors identified in the method for determining dsRNA editing activity of an A-to-l RNA editing enzyme and/or the method for screening inhibitors of the A-to-l RNA editing enzyme according to the invention. The method for determining binding of a dsRNA binding protein to a dsRNA target provides the additional information whether the inhibitor interferes with binding of the dsRNA binding protein to the dsRNA target. Interfering with binding includes, without being limited thereto interfering with the binding event itself, i.e., blocking the interaction of the dsRNA binding protein to the dsRNA target, but also modifying the dsRNA binding protein in a way that it cannot bind to the dsRNA anymore (e.g., by inducing a conformational change resulting in loss of dsRNA binding affinity) or binding to the dsRNA and/or modifying the dsRNA, such as modifying the secondary structure of the dsRNA, particularly separating the double-strand of the RNA, preventing binding of the dsRNA binding protein and/or recognition as a substrate of the dsRNA binding protein (such as the A-to-l RNA editing enzyme). Thus, in certain embodiments the dsRNA binding protein is an A-to-l RNA editing enzyme as described herein and the dsRNA target may be a dsRNA substrate comprising at least one adenosine or the edited RNA product comprising at least one inosine.

[0066] In preferred embodiments the method is a method for screening inhibitors that prevent binding of the dsRNA binding protein (i.e., the A-to-l RNA editing enzyme) to the dsRNA target, wherein the method further comprises adding a compound, reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a). The compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor that prevents binding of the dsRNA binding protein (i.e., the A-to-l RNA editing enzyme) to the dsRNA target results in a decreased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of a library of compounds or reagents.

[0067] The compound, reagent or at least one member of the library of compounds or reagents may be added to the reaction mixture prior to, simultaneously or after contacting the dsRNA target with the dsRNA binding protein. More specifically, the compound, reagent or at least one member of the library of compounds or reagents may be (i) added to the dsRNA target and/or the dsRNA binding protein, preferably to the dsRNA binding protein, prior to contacting in the reaction mixture in step (a), and/or (ii) added to the dsRNA target and the dsRNA binding protein simultaneously with the contacting in the reaction mixture in step (a) and/or (iii) added following contacting the dsRNA target with the RNA binding enzyme in the reaction mixture in step (a) and prior to detecting binding of the RNA binding enzyme to the substrate. The reaction mixture is an aqueous solution comprising a buffer in addition to the specified components. Preferably the buffer has an about neutral pH, such as pH 6.5 to pH 8, preferably pH 7.0 to 7.5. The reaction mixture typically further comprises salts and/or RNAse inhibitors.

[0068] The inhibitor that prevents binding of the dsRNA binding protein (i.e., the A-to-l RNA editing enzyme) to the dsRNA target may be any compound or reagent that inhibits or interferes with dsRNA target binding to the dsRNA binding protein, and/or that binds to the dsRNA binding protein, and/or modifies the conformation of the dsRNA binding protein thereby dsRNA target binding to the dsRNA binding protein and/or prevents conformational change upon dsRNA substrate binding, preferably that inhibits or interferes with dsRNA target binding to the dsRNA binding protein, e.g., by blocking the interaction via binding to the binding site of the dsRNA binding protein and/or by sterically hindering dsRNA target binding to the dsRNA binding site. The inhibition may be reversible or irreversible, preferably reversible. In certain embodiments, the compound, reagent or at least one member of a library of compounds or reagents to be screened is (a) a reagent, such as an antibody, an aptamer, a phage, an RNA, a peptide, a morpholino or an oligonucleotide, wherein preferable the reagent is an antibody, an aptamer, a phage, a peptide, a morpholino or an oligonucleotide; or (b) a compound, wherein the compound is a small molecule compound. In a preferred embodiment the reagent or at least one member of a library of compounds or reagents to be screened is a small molecule compound.

[0069] The method is suitable for high-throughput screening and is preferably performed using multiwell plates, preferably 96-well or more plates, more preferably 384-well or more plates or even more preferably 1538-well or more plates. Moreover, the method is a homogenous method or assay, i.e., a mix and measure assay comprising no washing and no separation steps.

[0070] The method may be measured over time (including in real-time following initiation of step (b)) or using end-point measurement. Preferably the dsRNA target and the dsRNA binding protein are incubated following contacting in the presence of the donor and acceptor pair in the reaction mixture or prior to addition of the donor and acceptor pair into the reaction mixture in which case a further incubation step may be required. The incubation may be for about 10 min or more, about 15 min or more, about 30 min or more, preferably for about 1 hour or more, more preferably for about 2 hours or more, e.g., between 15 and 37°C, preferably at room temperature (21°C) or about 24°C.

[0071] The method for determining binding of a dsRNA binding protein (i.e., the A-to-l RNA editing enzyme) to a dsRNA target according to the invention comprises adding a bead-based donor and acceptor pair and detecting interaction of the dsRNA binding protein to the dsRNA target in an assay using the bead-based donor and the acceptor pair, wherein the donor and acceptor pair is brought into proximity by the binding of the dsRNA binding protein to the dsRNA target, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal), and wherein the acceptor signal is detected. Preferably the signal transmission is a singlet oxygen. More preferably, the assay using a donor and acceptor pair is a bead-based amplified luminescence proximity homogenous assay. Thus, the donor and acceptor pair is a bead-based donor and acceptor pair and the assay using the donor and acceptor pair is a bead-based amplified luminescence proximity homogenous assay. Exemplary assays suitable as bead-based donor and acceptor pair, without being limited thereto, are as disclosed herein above in the context of the method for determining dsRNA editing activity of an A-to-l RNA editing enzyme according to the invention.

[0072] The dsRNA binding protein, wherein the dsRNA binding protein is an A-to-l RNA editing enzyme, is a fusion protein further comprising an affinity tag. Thus, the fusion protein comprises the dsRNA binding protein (i.e., the A-to-l RNA editing enzyme) fused to an affinity tag, and wherein the affinity tag is a polypeptide-tag. Typically, the affinity tag is N-terminally or C- terminally fused to the dsRNA binding protein. The term “affinity tag” as used herein refers to a polypeptide that can be coupled to one of the partners of the donor and the acceptor pair (bead) comprising an affinity tag binding reagent, i.e., if dsRNA is coupled to one partner, the affinity tag is coupled to the other partner. Affinity tags are known in the art and include without being limited thereto, e.g., chitin binding protein (CBP), maltose binding protein (MBP), Strep-tag, glutathione- S transferase (GST), polyhistidine-tag (preferably hexahistidine tag (6xHis-tag)), biotin and immunoglobulin Fc domain (Fc domain). Preferably the affinity tag is a polyhistidine-tag or an Fc domain. The person skilled in the art would know that an affinity tag can also be an immune tag or epitope tag, i.e., peptide sequences against which a high affinity antibody can be reliably produced, including without being limited thereto e.g., ALFA-tag, V5-tag, Myc-tag, HA-tag, Spottag, T7-tag and NE-tag. Optionally the dsRNA binding protein and the affinity tag are separated by a specific enzymatic cleavage site, such as a TEV cleavage site.

[0073] The fusion protein comprises the dsRNA binding protein and the affinity tag. The affinity tag is coupled (directly or indirectly) to the other partner of the donor and acceptor pair, in the reaction mixture, preferably to the acceptor. The affinity tag is captured (i.e., bound) by an affinity tag binding reagent, such as an anti-affinity tag antibody, which is directly or indirectly coupled to the donor or acceptor, e.g., via its Fc domain. In a preferred embodiment the affinity tag is a polyhistidine tag and the affinity tag binding reagent is an anti-polyhistidine antibody. The person known in the art would know reagents available for capturing fusion proteins comprising an affinity tag. For example, fluorescently labeled anti-polyhistidine antibodies and anti-6xHis acceptor beads are commercially available. Other suitable pairs of affinity tag and affinity tag binding reagent are Fc domain and Protein A or G (preferably A), biotin and streptavidin, GST and glutathione or any affinity tag and an affinity tag specific antibody.

[0074] The dsRNA binding protein may be any A-to-l RNA editing enzyme and/or ADAR protein, such as ADAR1 , ADAR2 or ADAR3. Double-stranded RNA binding protein (DRBP) are a family of eukaryotic, prokaryotic and viral-encoded products that share a common evolutionarily conserved motif facilitating interaction with dsRNA. These proteins comprise at least one dsRNA binding domain (DRBD), which binds dsRNA (e.g., about 11 pb or more of dsRNA). Typically, dsRNA binding proteins comprise more than one dsRNA binding domain, such as 2, 3, 4 or 5 dsRNA binding domains. Examples of dsRNA binding proteins are without being limited thereto dsRNA-dependent protein kinase (PKR), Staufen, ADAR, RNA helicase A (RHA), Dicer, and spermatid perinulear RNA binding protein (SPNR). The dsRNA binding protein according to the present invention is an A-to-l RNA editing enzyme and may be a full-length protein or a truncated or mutated form thereof comprising a catalytical domain and/or at least one dsRNA binding domain, preferably at least one dsRNA binding domain and optionally a catalytical domain. Preferably the dsRNA binding protein comprises more than one dsRNA binding domain, such as 2, 3, or 5 dsRNA binding domains.

[0075] In a specific embodiment dsRNA binding protein is an Adenosine Deaminase Acting on RNA (ADAR) protein, particularly Adenosine Deaminase Acting on RNA 1 (ADAR1), Adenosine Deaminase Acting on RNA 2 (ADAR2) or Adenosine Deaminase Acting on RNA 3 (ADAR3). This also includes mutant and truncated forms of ADAR1 , ADAR2 or ADAR3. The person skilled in the art would understand that mutant and truncated forms of ADAR1 , ADAR2 or ADAR3 with dsRNA binding activity are suitable for the method for determining binding of a dsRNA binding protein to a dsRNA target according to the present invention. This includes for example full-length (FL) ADAR1 , isoform p110, full-length (FL) ADAR1 , isoform p150, truncated ADAR1 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs), dsRNA binding ADAR1 mutants, full length (FL) ADAR2, truncated ADAR2 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs), dsRNA binding ADAR2 mutants, full length (FL) ADAR3, truncated ADAR3 comprising deaminase domain (DAD) and dsRNA binding domains (dsRBDs) and dsRNA binding ADAR3 mutants. In certain embodiments, the dsRNA binding protein comprises a sequence selected from the group consisting of SEQ ID NOs: 39, 40, 41 , 42 and 45. While the ADARs require at least one dsRNA binding domain preferably 2 or 3, ADAR1 may further comprise a Za and/or Zp domain. In certain embodiments, the dsRNA binding protein ADAR1 or ADAR2, preferably ADAR1. For example, the RNA binding protein may be ADAR1 fused to a polyhistidine tag, preferably the fusion protein comprises the amino acid sequence of SEQ ID NO: 43, or ADAR2 fused to a polyhistidine tag, preferably the fusion protein comprising the amino acid sequence of SEQ ID NO: 38.

[0076] According to the method of the invention the dsRNA target is coupled to at least one partner of the donor and acceptor pair, preferably to the donor. The dsRNA target may be (a) generated by annealing two complementary RNA strands; or (b) a single-stranded RNA that forms a secondary structure comprising double stands, such as hairpin structures, preferably a singlestranded RNA comprising inverse complementary nucleotide sequence regions forming a double strand such as a hairpin structure. Thus, in certain embodiments, the dsRNA substrate is generated by annealing two complementary RNA strands. In other embodiments the dsRNA substrate is a single-stranded RNA that forms a secondary structure comprising double stands, preferably comprising inverse complementary nucleotide sequence regions forming a double strand, such as a hairpin structure. For example, two inverted Alu-elements give rise to doublestranded RNA. The dsRNA substrate may also comprise other secondary structures, such as a bulge or an internal loop, which may occur in both, (a) two complementary RNA strands, or (b) ssRNA that forms a secondary structure comprising double strands. The dsRNA typically forms a double helix (tertiary structure), which may be in the A-conformation or the Z-conformation. The dsRNA target may further comprise at least one adenosine, particularly a dsRNA target that comprises at least one adenosine which is a dsRNA substrate for an A-to-l RNA editing enzyme. Thus, the dsRNA target may be identical to the dsRNA substrate comprising at least one adenosine as described herein. Particularly when the two methods according to the invention, the method for determining dsRNA editing activity and the method for determining binding of a dsRNA binding protein to a dsRNA target are used in combination to characterize the same A-to-l RNA editing enzyme or to screen an inhibitor for said enzyme, it may be advantageous if the dsRNA target and the dsRNA substrate are identical, and optionally that further the RNA tracer is identical to the edited RNA product of said dsRNA substrate. Examples of dsRNA targets are therefore the same as specified above for dsRNA substrates. [0077] According to the method of the invention the dsRNA target is coupled to at least one partner of the donor and acceptor pair, preferably to the donor. The coupling to one partner of the donor or acceptor pair includes covalent binding and non-covalent binding. For example, the dsRNA target comprises a coupling label that binds non-covalently to a coupling-label binding protein on the donor or acceptor, preferably the donor. Preferably the coupling label and the affinity tag are different, particularly, if the coupling label is biotin, the affinity tag is not biotin. In certain embodiments the dsRNA target comprising a coupling label and the donor or acceptor comprises a coupling-label binding protein. Preferably the coupling-label is biotin or digoxigenin and the coupling-label binding protein is a biotin-binding protein or a digoxigenin binding protein. Wherein the biotin is covalently linked to the dsRNA target (biotinylated dsRNA target), preferably to the 5’- or 3’-end of the dsRNA target, more preferably the 5’-end of the dsRNA target. In case the double-stranded RNA target is generated by annealing two complementary RNA strands labeling one strand is sufficient. Methods for biotinylating RNA molecules, including end-terminal labeling are known in the art. Biotin-binding proteins are further known in the art and include without being limited thereto streptavidin, streptactin and avidin. Suitable donor and acceptor pairs comprising streptavidin, streptactin or avidin (preferably covalently-linked) are known in the art and commercially available. In certain embodiments the dsRNA target is biotinylated, and the donor or acceptor comprises streptavidin or avidin or the dsRNA target is labelled with digoxigenin and the donor or acceptor comprises an antibody against digoxygenin (directly or indirectly bound via a secondary antibody).

[0078] An exemplary assay using a bead-based donor and acceptor pair is a bead-based, nonradioactive amplified luminescent proximity homogeneous assay (first described in 1994 by Ullman, E. F., et al., Proc. Natl. Acad. Sci USA, 91 (12): 5426-5430) and based on the principle of luminescence oxygen channeling and compatible for high-throughput screening has become commercially available (AlphaScreen). In this assay a donor and acceptor pair (donor and acceptor beads of 250 nm diameter reagent-coated polystyrene microbeads) are brought into proximity by a biomolecular interaction of binding partners immobilized to the beads. When the biological interaction brings the donor and acceptor beads together, a cascade of chemical reactions acts to produce a greatly amplified signal. On laser excitation (at 680 nm), a photosensitizer in the “donor” bead converts ambient oxygen to a more excited singlet state. The singlet state oxygen molecules diffuse across (up to 200 nm) to react with a thioxene derivative in the acceptor bead generating chemiluminescence at 370 nm that further activates fluorophores contained in the same bead. The fluorophores subsequently emit light at ~520-620 nm using a classical acceptor bead or europium is activated in an AlphaLisa Acceptor bead and light is emitted at ~615 nm. In the absence of a specific biological interaction, the singlet state oxygen molecules produced by the donor go undetected without the close proximity of the acceptor. As a result, only a very low background signal is produced. The donor bead generates about 60,000 singlet oxygen molecules, resulting in an amplified signal. Singlet oxygen has a short lifetime in aqueous solution (~4 psec) which allows a diffusion over a distance up to ~200 nm. Because the signal is very long lived, with a half-life in the second range, the detection system can be timegated, thus eliminating short-lived background (AlphaScreen signal is measured with a delay between illumination and detection of 20 ms). Furthermore, the detection wavelength is shorter than the excitation wavelength, thus further reducing potential for fluorescence interference. The sensitivity of the assay derives from the very low background fluorescence. AlphaScreen provides a highly versatile, sensitive, homogeneous and miniaturizable technology to efficiently perform assay development and HTS resulting in higher throughput at lower costs. A comparable technology is Enlight OMEGA (Oxygen Mediated/Gated Assay; BioSignal² Inc. CA), were the beads have an extension wavelength of 680 nm and an emission wavelength of 615 nm.

EXAMPLES

Example 1: Enzymatic activity assay for ADAR1 (Figures 1-6)

Reagents / material:

[0079] The following reagents and materials have been used: 1536 plates (Greiner Bio-One GmbH), EDTA, Hepes, ZnCI₂ and IP6 (Sigma Aldrich), NaCI and Tween 20 (Roth), BSA (Serva), SUPERase-ln RNAse inhibitor and DEPC-treated water (Thermo Fisher; Cat. No. AM2696), antiinosine antibody (Diagenode, Cat. No. C152000251), rabbit anti-mouse IgG (Merck, clone RM 104, Car. No. 04-1643), AlphaScreen IgG protein A detection kit (PerkinElmer, Cat. No. 6760617), ADAR enzymes (as described herein), RNA substrates, products and biotinylated tracers (custom made, Eurofins Genomics):

Exemplary dsRNA substrate and biotinylated RNA tracer:

RNA-A V5-Z (“substrate”)

RNA sequence (SEQ ID NO: 30):

CGCGCGCGCGCGGGACAAAUUAACCAAGGAAAAUAACAAGGACAGGGACC

RNA sequence reverse (SEQ ID NO: 31): GGUCCCUGUCCUUGUUAUUUUCCUUGGUUAAUUUGUCCCGCGCGCGCGCG Bio-RNA-I V5_1 (“tracer”)

RNA sequence (SEQ ID NO: 25):

BIOTEG](CGGCCGGCCGGGGCCCCCUUIICCIIGGCCCCUCCCCCGGACAGGGACC) RNA sequence reverse (SEQ ID NO: 16): GGUCCCUGUCCGGGGGAGGGGCCUUGGUUAAGGGGGCCCCGGCCGGCCGG

To generate double-stranded RNA the complementary strands were annealed at a 1 :1 ratio in RNAse-free water (2 min, 95°C, cool down to 25°C (ramp -4°C/min), cool down to 4°C).

[0080] Single dose testing of potential inhibitors (test compound) was performed at compound concentrations of 5 pg/ml to 50 pg/ml. For dose response testing compounds were diluted in 1 :5 or 1 :3.16 dilution steps or other useful variations, with a compound stock concentration of 5 mg/ml. Assay buffer (20 mM Hepes pH 7.5, 100 mM NaCI, 0.02% Tween20, 0.1% BSA, 4.3 pM ZnCI₂, 1 pM IP6, 0.03 U/pl SUPERase-ln RNAse-lnhibitor) was used throughout for dilutions and in the reaction mix.

Assay procedure

[0081] The assay was performed as a competition assay using the AlphaScreen IgG protein A detection kit comprising a Protein A-acceptor bead and a streptavidin donor bead in a 1536-well plate. For the competing enzyme reaction 50 nl test compound (from proprietary Screening Pool collections (all chemically based collections)) and 1 pl EDTA (final cone, of 1 mM) were added per well of the 1536-well plate and incubated for 10 min in a humidified incubator at 24°C, before 2 pl of ADAR protein (final concentration of 4 nM), 2 pl of assay buffer (positive control), or 2 pl of product standard (2.5 nM - 0.02 nM) were added followed again by incubation for 10 min in a humidified incubator at 24°C. 2 pl of RNA-substrate was added and the plate was incubated for at least 2.5 h in a humidified incubator at 37°C to allow A-to-l RNA editing. Following completion of the enzymatic reaction, the reagents for detection were added. First, 2 pl of biotinylated inosine- containing-RNA (Bio-RNA-I) was added, followed by the addition of 2 pl of beads-antibodies-mix (containing a final concentrations of 10 pg/ml beads each, 2 nM of anti-inosine mouse IgG antibody and 3 nM of anti-mouse-IgG antibody) and incubated for 1 h in a humidified incubator at 24°C. The AlphaScreen signal was detected with an Envision Reader with Ex 680nm and Em 520-620nm. Analyzing Data:

[0082] Each assay plate contains wells as negative control with no inhibitor (1 % DMSO instead of compound) as reference (100% CTL, low signal), and wells as positive control with no A-to-l RNA editing enzyme (1% DMSO and without enzyme (0% CTL, high signal). The abbreviation CTL means % of control. The raw data are analysed by direct normalization of the raw data (see variation 1) or by transformation into concentrations via a standard curve following normalization (see variation 2):

Variation 1: Normalization of raw data based on positive and negative control

%CTL = 100 * (((Counts (sample)-Counts (positive control)) I (Counts(negative control)- Counts(positive)))

Variation 2: Transfer of raw data into product concentration using a 4-parameter sigmoidal standard curve Log (Cone) vs. Signal

Y=Bottom + (Top-Bottom)/(1 + 10^A((LoglC50-X))*Slope)

The transferred data (product concentrations) are further normalized using positive and negative controls

%CTL = 100 * (((Transformed Data(sample)- Transformed Data (positive))/(Transformed Data (negative)- Transformed Data (positive)))

IC₅o data were calculated with normalized data with 4-parameter sigmoidal dose response formula.

[0083] For converting raw values into concentrations, a calibration curve was used. A calibration curve is a plot of known concentrations of a standard solution against their corresponding raw values (e.g., absorbance, fluorescence, or other measurable signals) and fitting a curve to the data points, which can be linear, logarithmic, or sigmoidal (e.g., 4-parameter logistic model) depending on the relationship between the raw values and concentrations. The curve fitting can be done using statistical software or spreadsheet programs. In the present case a 4-parameter sigmoidal standard curve log(conc.) vs. signal was used.

[0084] The calibration curve is then used to determine the concentration of unknown samples based on their raw values. Alternatively, the equation of the fitted curve (as shown above) may be used to calculate the concentration directly from the raw value. A 4 parameter curve fit for a sigmoidal curve using the equation above, wherein the fitting algorithm is generating the top and the bottom values of the curve is e.g., shown in Figure 2.

[0085] For ADAR1 no high throughput compatible enzymatic activity assay is described in literature to date. A high throughput compatible assay should be sensitive, robust, reproducible, miniaturizable to save reagents and costs, fully automatable and should allow the test of hundred thousand of chemical compounds, peptides, proteins, antibodies etc.

[0086] Therefore, an enzymatic activity assay for ADAR1 based on the AlphaScreen technology as one of the most sensitive technologies available was developed, of which the assay principle is described in Figure 1. A biotinylated double-strand RNA containing several inosines (ADAR1 product, Tracer) is immobilized to a streptavidin coated AlphaScreen donor bead. The inosines are detected via an anti-inosine specific antibody which is bound via an anti IgG specific antibody to an AlphaScreen acceptor bead. If these interactions occur a high AlphaScreen signal is generated after stimulation. If one runs a biochemical ADAR1 reaction employing an adenosine containing double-strand RNA as a substrate, ADAR1 will edit the adenosines to inosines, and one obtains an edited RNA product. Since this product is not biotinylated it will compete with its biotinylated product (Tracer) to the binding to the anti-inosine antibody. In the presence of an active enzyme, unbiotinylated RNA product is generated which leads to a signal reduction. A potential inhibitor blocks the enzymatic activity less unbiotinylated RNA product is generated resulting in a high AlphaScreen signal. This assay format has the advantage that any possible reagent interfering with the assay setup or technology itself provides AlphaScreen counts comparable to an active enzymatic reaction. Therefor this assay format reduces the identification of potential false positive inhibitors.

[0087] For the assay development mentioned in Figure 1 , it was of importance to identify an appropriate biotinylated RNA Tracer, an appropriate RNA substrate, a functional enzymatic construct as well as the establishment of the final assay conditions and reagent concentrations.

[0088] As illustrated in Figure 2, several different biotinylated double-stranded RNAs, containing different editing sites, as listed in Table B, were systematically tested by employing the AlphaScreen detection part illustrated in Figure 1. V5 Z is a better substrate compared to V5 std. (Table 1) and V5 1-5 did not show superiority against V5 Z. Table 1 : Comparison of RNA V5 standard (std) (EDIT_V5_std; SEQ ID NOs: 1 and 4) and RNA V5 Z (EDIT_V5_Z; SEQ ID NOs: 30 and 31) using different concentration of full length ADAR1 p110 employing enzymatic activity assay. Signal/Background (S/B) ratio indicate superiority of ^NA V5 Z as a substrate. Incubation times 3h at 37°C.

Table A: Variant 5 (HT) sequences used in enzymatic activity and binding assay [0089] As shown in Figure 2, appropriate RNA products were used to generate titration curves to monitor the detection range for the different biotinylated Tracers employed. For the final assay conditions, it is of importance to adjust the enzymatic reaction in that way, that the generated RNA product can be detected in the quasi-linear range of the detection curve. Due to the highest sensitivity generated and the best reproducibility, the biotinylated RNA-I V5_1 Tracer was chosen.

Table B: RNA combinations used to generate data presented in Figure 2

[0090] For the next set of experiments, the enzymatic constructs as well as the substrates were optimized. Deaminase domain (DAD) (SEQ ID NO: 40), a mutated DAD for higher activity (SEQ ID NO: 44), DAD containing RNA binding domains (RBD) (SEQ ID NO: 45), full length ADAR1 p110 (SEQ ID NO: 39), a His tagged DAD (SEQ ID NO: 46) and a full length ADAR1 His tagged (SEQ ID NO: 43) version was tested.

Table C: ADAR proteins used in all three assays

[0091] For the substrates we selected V5 std (SEQ ID NOs: 1 and 4) and V5 Z (SEQ ID NO: 30 and 31). Both containing comparable editing sites, the later possesses sequences causing a specific folding of the RNA. The results are summarized in Table 2 and Figure 3.

Table 2: Comparison of RNA V5 standard (std) (EDIT_V5_std; SEQ ID NOs: 1 and 4) and RNA V5 Z (EDIT_V5_Z, SEQ ID NOs: 30 and 31) using different ADAR1 constructs, employing the enzymatic activity assay. S/B ratio indicate superiority of RNA V5 Z as a substrate. Incubation times 3h at 37°C.

[0092] Each enzyme was tested with a concentration of 150 nM and two different substrate concentrations 100 nM and 50 nM as well as two different substrates V5 std and V5 Z for 3h at 37°C. There is a clear activity distribution within the different enzymes tested: highest activity has full length ADAR1 p110, followed by the DAD containing an RNA binding site, followed with a clear lower activity of the mutated DAD and DAD alone, shown by the different signal to background ratios (S/B) in Table 2 and Figure 3. A clear substrate dependency was shown, with higher S/B ratios with the higher substrate concentration. There was a clear superiority for the V5 Z substrate compared to V5 std, especially shown with DAD-RBD and full length ADAR1 p110. [0093] Using these findings, we tested different ADAR1 p110 concentrations (7.5, 6, 5 nM) with increasing concentrations of substrate V5 Z (12.5, 25, 35, 50 nM), incubating the enzymatic reaction for 2 h at 37°C. The results are summarized in Table 3, showing a decrease of the S/B ratios with lower enzyme concentrations and a clear favourable substrate concentration of 25 nM.

Table 3: Comparison of different concentrations of RNA V5 Z (EDIT_V5_Z; SEQ ID NOs: 30 and 31) using different ADAR1 concentrations, employing the enzymatic activity assay. S/B ratio indicate dependency of RNA and enzyme concentrations. Incubation times 2 h at 37°C.

[0094] The characterization of the enzyme and to use the generated information to adjust the assay conditions is an important step during assay development. Figure 4A summarizes the Km determination of full length ADAR1 p110 (5 nM) with increasing concentrations of substrate V5 Z. Km was determined after an incubation time of 3h at 37°C with 51.22 nM. The determination curve could give hints for a substrate inhibition and/or could be interpreted as a kind of saturation of the available enzymes over time without an immediate product release. This would be in line with data summarized in Table 3. Figure 4B shows the linearity of the enzymatic reaction of ADAR1 p110 (5 nM) and V5 Z (25 nM) substrate over 4h of incubation at 37°C, demonstrating that no substrate depletion occurs under the conditions tested. [0095] Using the conditions described in the section “assay procedure” and final enzyme concentrations for full length ADAR1 p110 with 4 nM and 25 nM of substrate V5 Z, compound were screened for inhibition of ADAR1 by generating e.g., dose response curves. Figure 5 shows 3 examples of IC50 curves determined for three different compounds. Aiming for a highly sensitive assay allowing to perform fully automated ultra-high throughput screening in a 1536 assay format, the assay clearly allows to work below Km of the substrate, which opens the chance to identify compounds with different mode of actions, including a low assay wall. Figure 6 summarizes the results of a fully automated high throughput screening (HTS) campaign testing ~ 1.000.000 of different compounds and underline the robustness and reproducible quality of this 1536 compatible enzymatic activity assay for ADAR1. Z' factor was 0.9 (Figure 6A), and the assay demonstrated a clear and reproducible separation window (S/B)(6C) and a small standard deviation of 1.4% CTL with a mean CTL value of a Gaussian Distribution of 99.9% (Figure 6B).

[0096] Using this enzymatic activity assay, a high throughput screening campaign was performed. In total ~1 .254.376 compounds were tested. Using a hit criterion of <75% CTL 22.699 primary hits were identified. After running a hit confirmation with fresh compound samples, ~15.728 hits could be confirmed, which translates into a confirmation rate of 69.3% and a confirmed hit rate of 1.25%. After performing a chemoinformatics analysis, containing e.g. compounds which are known as frequent hitters that regularly appear in screenings and employing structural clustering processes, ~5.092 compounds were selected for generating dose response curves (such as shown in Figure 5).

Example 2: ADAR / RNA binding assay (Figures 7-9):

Reagents / material:

[0097] The following reagents and materials have been used: 1536 plates (Greiner Bio-One GmbH, PS, hibase, white, med-binding), EDTA, Hepes, ZnCI₂ and IP6 (Sigma Aldrich), NaCI and Tween 20 (Roth), BSA (Serva), SUPERase-ln RNAse inhibitor and DEPC-treated water (Thermo Fisher, Cat. No. AM2696), AlphaScreen streptavidin donor beads and anti-6xhis Alphalisa acceptor beads (PerkinElmer, Cat. No. 6760002 and AL178R, respectively), ADAR enzymes (as described herein), and biotinylated RNA targets (custom made, Eurofins Genomics). Exemplary biotinylated RNA target: Bio-RNA-A V5-Z (“target”)

RNA sequence (SEQ ID NO: 34):

CGCGCGCGCGCGGGACAAAUUAACCAAGGAAAAUAACAAGGACAGGGACC

RNA sequence reverse (SEQ ID NO: 31): GGUCCCUGUCCUUGUUAUUUUCCUUGGUUAAUUUGUCCCGCGCGCGCGCG [0098] To generate double-stranded RNA the complementary strands were annealed at a 1 :1 ratio in RNAse-free water (2 min, 95°C, cool down to 25°C (ramp -4°C/min), cool down to 4°C). [0099] Single dose testing of potential inhibitors (test compound) was performed at compound concentrations of 5 pg/ml to 50 pg/ml. For dose response testing compounds were diluted in 1 :5 or 1 :3.16 dilution steps or other useful variations, with a compound stock concentration of 5 mg/ml. Assay buffer (20 mM Hepes pH 7.5, 100 mM NaCI, 0.02% Tween20, 0.1% BSA, 4.3 pM ZnCI₂, 1 pM IP6, 0.03 U/pl SUPERase-ln RNAse-lnhibitor) was used throughout for dilutions and in the reaction mix.

Assay procedure

[00100] The assay was performed as a competition assay using the AlphaScreen streptavidin donor beads and anti-6xhis Alphalisa acceptor beads in a 1536-well plate. 50 nl test compound and 1 pl EDTA (final cone, of 1 mM) were added per well of the 1536-well plate and incubated for 10 min in a humidified incubator at 24°C, before 2 pl of assay buffer (positive control) or 2 pl of the his-tagged enzyme (final concentration of 12.5 nM) was added followed again by incubation for 10 min in a humidified incubator at 24°C. 2 pl of biotinylated RNA-substrate (target) was added and the plate was incubated for 2 h in a humidified incubator at 24°C to binding of the enzyme to the RNA-substrate. The reagents for detection were added comprising adding 4 pl of bead-mix (containing a final concentrations of 20 pg/ml beads each) and incubated for 2 h in a humidified incubator at 24°C. The AlphaScreen signals was detected with an Envision Reader with Ex 680 nm and Em 615 nm.

Analyzing Data:

[00101] Each assay plate contains wells as negative control with no inhibitor (1% DMSO instead of compound) as reference (100% CTL, low signal), and wells as positive control with no his-tagged enzyme (1 % DMSO and without enzyme) added (0% CTL, high signal). The raw data were analysed by direct normalization of the raw data (see variation 1) based on the positive and negative controls: %CTL = 100 * (((Counts(sample)-Counts(positive))/(Counts(negative)- Counts(positive)))

IC₅o data were calculated from the normalized data with 4-parameter sigmoidal dose response formula.

[00102] A prerequisite for the editing activity of ADARs is the interaction with their substrates. Therefor we developed a so far not described ADAR - RNA binding assay, illustrated in Figure 7, which may be adapted to dsRNA binding proteins other than A-to-l RNA editing enzymes. A biotinylated RNA double-stranded RNA containing adenosines (substrate) interacts with an His tagged ADAR p110. After addition of a streptavidin coated donor bead and an anti- His tagged acceptor bead, a high AlphaScreen signal is generated due to the proximity of the reagents. The addition of a compound interfering with the binding causes a signal reduction. Interfering with the binding includes, without being limited thereto interfering with the binding event itself, i.e., blocking the interaction, but also modifying the ADAR protein in a way that it cannot bind to the dsRNA anymore (e.g., by a conformational change) or binding to the dsRNA and/or modifying the dsRNA, such as the secondary structure of the dsRNA in a way that it cannot serve as a substrate and/or bind to the ADAR protein any longer. Therefor this assay serves as a screening assay to identify compounds interfering with the interaction of the ADAR protein with a substrate (or a dsRNA binding protein to a dsRNA target). In addition, the assay can be used to characterize the mode of action of already existing compounds knowing to inhibit the enzymatic activity of ADARs. Similar to the enzymatic activity assay, this assay is suitable for and has been successfully use with 1536 well plates. Since the assay can be set up in a 1536 format and is automatable, it is also high throughput compatible.

[00103] For validation of the assay a dissociation constant (Kd) determination was performed providing for the biotinylated RNA-V5Z (Biotin EDIT V5_Z; SEQ ID NOs: 34 and 31) as dsRNA target (substrate) a Kd of 1.77 nM using a constant concentration of 12 nM His tagged ADAR1 p110 as the dsRNA binding protein (Figure 8A) and providing a Kd of 27.29 nM for the His tagged ADAR1 p110 using a constant concentration of 12 nM biotinylated RNA-V5Z (Biotin EDIT V5_Z; SEQ ID NOs: 34 and 31) as the dsRNA target (Figure 8B). A competition experiment with ADAR1 p110 without a his tag showed a dose response curve with an IC₅o of 1.78 nmol/l underlining the sensitivity of this assay format (Figure 8C).

[00104] Figure 9 shows examples of IC50 determinations of different chemical compounds tested with this assay set up. The assay was successfully applied to test the same ~5092 compounds selected using the enzymatic activity assay for dose response testing to analyse if they are interfering with the interaction of the enzyme and the substrate, giving a first hint for their potential mode of action.

Example 3: Medium-throughput (MT) compatible quantification of RNA editing by ADAR1 (Figures 10-12)

[00105] In order to quantify RNA editing an assay was established that measures the ratio of edited versus non-edited (WT) RNA. The assay principle is illustrated in Figure 10. Human AZIN1 (SEQ ID NO: 36) and MDM2 (SEQ ID NO: 37) cDNAs encoding dsRNA substrate were overexpressed (standard transient transfection of plasmid DNA in lipofectamine) from commercially available plasmids (pCMV6-Myc-DKK (Origene) and pCMV-Sport 6, (Addgene), respectively) in human colon carcinoma cell line HCT-116, genetically engineered (CRISPR/Cas9) to lack both isoforms of ADAR1 (p110 and p150 knockout cell line, referred to as HCT-116 ADAR1 KO).

[00106] Total RNA was isolated using commercially available miRNAeasy kit from Qiagen. Concentration of RNA was measured, and 1 pg of RNA was used for subsequent reactions with ADAR proteins.

[00107] Initially various ADAR proteins were tested at different time points to optimize conditions. ADAR proteins (ADAR1 p110 FL (SEQ ID NO: 39), ADAR1 DAD (SEQ ID NO: 40) and ADAR2 FL (SEQ ID NO: 41)) were pre-dilute to the desired final concentrations (250 nM, 500 nM and 750 nM) in ADAR reaction buffer (20 mM HEPES, 100 mM NaCI, 0.02% Tween20, 0.1 % BSA, 4.3 pM ZnCI2, 1 pM IP6, 0.03 U/ml SUPERASE-RNAse Inhibitor, pH 7.5) and incubated with isolated RNA for 15, 45 or 60 min (Figure 11). The reaction was stopped after indicated timepoints by incubating the mixture at 70°C for 5 min and subsequently cooled down to 4°C.

[00108] For testing potential inhibitors, the assay was essentially performed as described above using the following optimized conditions. Compounds identified as potential inhibitors of ADAR1 in the assay described in Example 1 (high throughput enzymatic assay) with known IC50 were diluted in DMSO to a final concentration 10 pM and 100 pM and used pre-diluted in assay buffer. 500 nM ADAR p110 FL protein was pre-incubate with compounds 01 to 09 with increasing IC₅o as shown in Table 4 below or DMSO (control) before adding RNA for 10 min at 24°C. Next, 1 pg of purified RNA was added and incubated for 45 min at 37°C. The reaction was stopped after 45 min by incubating the mixture at 70°C for 5 min and subsequently cooled down to 4°C. Table 4: Inhibitory compounds with respective IC₅o

[00109] In the next step RNA was purified using ethanol precipitation on 96-well plate with commercially available PureLink Pro 96 RNA Purification Kit (Thermofisher) and standard protocol provided by the supplier.

[00110] Next, cDNA was synthesized using reverse-transcriptase Super Script II (Invitrogen) and Oligo(dT)50 primers following standard protocol provided by supplier.

[00111] Quantitative RT-PCR reaction was performed using SYBR Green to quantify the number of unedited transcripts (WT) and edited transcripts (EDIT) using the following primer pairs (each at a final concentration of 10 pM) and conditions:

Table 5: Primers

Table 6: PCR cycle Data analysis and interpretation:

[00112] Ratio edit/wt RNA depicted on Figure 11 has been calculated using following formula:

2-(Ct mean (wt)-Ct mean (edit))

Ct values above 35 (assay sensitivity detection threshold) have been omitted. Each Ct (copy per reaction) value represents individual reaction. As shown in Figure 11 editing of the same dsRNA substrate (AZIN1) was different depending on the ADAR protein used. Similar results were obtained using MDM2 as substrate. Most efficient editing was detected with ADAR1 p110 FL, while almost no editing was detected for ADAR2 FL. This confirms that AZIN1 is not a substrate for ADAR2, as expected. Moreover, ADAR1 DAD shows editing activity only at higher concentration. Similar results have been observed in the high-throughput enzymatic activity assay (Example 1). Based on these results ADAR1 p110 FL at a concentration of 500 nM and a reaction time of 45 min was chosen to establish the assay with inhibitory compounds.

[00113] Figure 12 shows the results of an editing reaction performed with dsRNA (MDM2; SEQ ID NO: 49) as substrate and ADAR1 p110 FL protein, which was pre-incubated with nine compounds (01-09) previously identified as potential ADAR1 inhibitory compounds using the high- throughput enzymatic activity assay described in Example 1. The compounds with an IC50 ascending from compound 01 to 09 (Table 4) have been tested at 10 pM (Figure 12A) and 100 pM (Figure 12B). Particularly at 100 pM Figure 12 shows inhibition with most of the compounds and increasing IC₅o values correlated with decreasing inhibitory effects.

[00114] Overall, the results demonstrate the robustness and flexibility of the assay (applicable for two different dsRNA substrates and different ADAR proteins) as well it’s compatibility with the high-throughput enzymatic activity assay (Example 1). Not all tested compounds, identified previously in the high-throughput assay as inhibitory compounds has been confirmed in this medium-throughput assay, which demonstrate the necessity and value of validation with an independent assay and technology.

Claims

1 . A method for determining double-stranded (ds) RNA editing activity of an adenosine-to- inosine (A-to-l) RNA editing enzyme comprising:

(a) reacting a dsRNA substrate with an A-to-l RNA editing enzyme in a reaction mixture comprising

(i) contacting the dsRNA substrate comprising at least one adenosine with the A-to-l RNA editing enzyme, and

(ii) incubating the reaction mixture under conditions that allow editing the at least one adenosine in the dsRNA substrate into an inosine thereby forming an edited RNA product comprising at least one inosine;

(b) adding an RNA tracer comprising at least one inosine in an RNA, an anti-inosine binding molecule and a donor and acceptor pair to the reaction mixture and detecting binding of the anti-inosine binding molecule to the RNA tracer in an assay using the donor and acceptor pair, wherein

(c) measuring competition of the edited RNA product with the RNA tracer for binding to the anti-inosine binding molecule, wherein the presence of edited RNA product reduces the acceptor signal upon excitation of the donor compared to the maximal acceptor signal of a control (positive control) without the A-to-l RNA editing enzyme.

2. The method of claim 1 , wherein the method is a method for screening inhibitors of the adenosine-to-inosine (A-to-l) RNA editing enzyme, and wherein the method further comprises adding a compound, a reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a).

3. The method of claim 2, wherein the compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme reduces or prevents formation of the edited RNA product, wherein the presence of less or no edited RNA product results in an increased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of the library of compounds or reagents.

4. The method of any one of the preceding claims, wherein the RNA tracer

(a) is a double-stranded RNA or a single-stranded RNA, preferably a double-stranded RNA;

(b) comprises a mammalian RNA; and/or

(c) comprises a coupling-label and the donor or acceptor comprises a coupling-label binding proteins, preferably wherein the coupling-label is biotin or digoxigenin and the coupling-label binding protein is a biotin-binding protein or a digoxigenin binding protein.

5. The method of any one of the preceding claims, wherein the anti-inosine binding molecule

(a) is an antibody or an aptamer;

(b) is coupled directly or indirectly to the other partner of the donor and acceptor pair, and/or

(c) is coupled indirectly using an antibody binding the anti-inosine binding molecule, preferably wherein the antibody is coupled via Protein A.

6. A method for screening or validating inhibitors of an A-to-l RNA editing enzyme in an assay quantifying editing activity of an A-to-l RNA editing enzyme in vitro comprising

(b) incubating a double-stranded (ds) RNA substrate comprising at least one adenosine (A) and a purified A-to-l RNA editing enzyme in a reaction mixture under conditions allowing A- to-l editing of the dsRNA substrate to form the respective edited RNA product; wherein the compound, reagent or at least one member of the library of compounds or reagents is added prior to or during incubation to the reaction mixture;

(c) quantifying edited RNA product and non-edited dsRNA substrate using quantitative realtime PCR (qPCR) comprising

(i) purifying RNA from the reaction mixture of step (b) following incubation; (ii) reverse transcribing RNA into cDNA using reverse transcription-polymerase chain reaction (RT-PCR); and

(iii) quantifying an amplicon of a sequence within the reverse-transcribed nonedited dsRNA substrate cDNA, and an amplicon of the respective sequence within the reverse-transcribed edited RNA product cDNA using quantitative real-time PCR (qPCR), wherein the sequence comprises the non-edited at least one A reverse-transcribed into deoxythymidine (T) and the edited at least one A reverse transcribed into deoxyguanosine (G), respectively;

(d) determining whether the compound, reagent or at least one member of the library of compounds or reagents is an inhibitor of the A-to-l RNA editing enzyme, wherein a compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor for the A-to-l RNA editing enzyme results in a decrease ratio of edited RNA product cDNA to non-edited RNA substrate cDNA compared to a control not comprising said compound, reagent or at least one member of the library of compounds or reagents.

7. The method of any one of the preceding claims, wherein the A-to-l RNA editing enzyme is Adenosine Deaminase Acting on RNA 1 (ADAR1) or ADAR2, preferably ADAR1 .

8. The method of any one of the preceding claims, wherein the dsRNA substrate is

(a) generated by annealing two complementary RNA strands; or

(b) a single-stranded RNA that forms a secondary structure comprising double stands, preferably a single-stranded RNA comprising inverse complementary nucleotide sequence regions forming a hairpin structure.

9. A method for determining binding of a double-stranded (ds) RNA binding protein to a double-stranded (ds) RNA target comprising: detecting interaction of a dsRNA binding protein and a dsRNA target in an assay using a bead-based donor and acceptor pair, wherein the dsRNA binding protein is an A-to-l RNA editing enzyme, comprising

(a) contacting the dsRNA target with the dsRNA binding protein in a reaction mixture, wherein

(i) the dsRNA target is coupled to one partner of the donor and acceptor pair in the reaction mixture; and (ii) the dsRNA binding protein is a fusion protein further comprising an affinity tag, wherein the affinity tag is coupled to the other partner of the donor and acceptor pair in the reaction mixture, and

(b) detecting binding of the dsRNA binding protein to the dsRNA target, wherein the donor and the acceptor are brought into proximity by the binding of the RNA binding protein to the dsRNA target, allowing upon excitation of the donor a signal transmission from the donor to the acceptor (donor signal) and emission of a fluorescent or luminescent signal from the acceptor (acceptor signal), wherein the acceptor signal is detected.

10. The method of claim 9, wherein the method is a method for screening inhibitors that prevent binding of the dsRNA binding protein to the dsRNA target, wherein the method further comprises adding a compound, reagent or at least one member of a library of compounds or reagents to the reaction mixture in step (a).

11. The method of claim 10, wherein the compound, reagent or at least one member of the library of compounds or reagents that acts as an inhibitor that prevents binding of the dsRNA binding protein to the dsRNA target results in a decreased acceptor signal compared to a control (negative control) not comprising said compound, reagent or at least one member of a library of compounds or reagents.

12. The method of claims 2, 3, 6, 10 or 11 , wherein the compound, reagent or at least one member of a library of compounds or reagents to be screened is

(a) a reagent, wherein preferable the reagent is an antibody, an aptamer, a phage, an RNA molecule, an oligonucleotide or a peptide, preferably an oligonucleotide or a peptide; or

(b) a compound, wherein the compound is a small molecule compound.

13. The method of any one of claims 1-5, wherein the method is a homogenous method.

14. The method of claim 13, wherein the signal transmission is an electron transfer or a singlet oxygen transfer.

15. The method of claim 13 or 14, wherein the assay using a donor and acceptor pair is a homogenous time-resolved fluorescent assay (HTRF) or a bead-based amplified luminescent proximity homogenous assay.

16. The method of any one of claims 9-11 , wherein

(a) the affinity tag is a polyhistidine-tag;

(b) the signal transmission is a singlet oxygen transfer; and/or

(c) wherein the assay using the bead-based donor and acceptor pair is a bead-based amplified luminescent proximity homogenous assay.

17. The method of any one of claims 9-11 , wherein the dsRNA binding protein is an A-to-l RNA editing enzyme selected from the group consisting of ADAR1 , ADAR2 and ADAR3.

18. The method of claim 17, wherein the dsRNA binding protein is selected from the group consisting of an ADAR1-p110 isoform, an ADAR1-p150 isoform, an ADAR1 deaminase domain and dsRNA-binding domain, ADAR2 and ADAR3.

19. The method of any one of claims 9-11 , wherein the dsRNA target is

(a) generated by annealing two complementary RNA strands; or