Abstract
Adenosine deaminase acting on RNA (ADAR)-mediated RNA base editing offers a safer alternative to genome editing for specific clinical applications because of nonpermanent editing of targets. Current guide RNA (gRNA) designs feature a fully complementary specificity domain with an A–C mismatch at the targeted adenosine. However, perfectly matched dsRNA is not the most effective ADAR substrate. Here we introduce MIRROR (mimicking inverted repeats to recruit ADARs using engineered oligoribonucleotides), an approach that implements structural motifs derived from highly edited inverted Alu repeats in human tissues to enable rational gRNA design for ADAR recruitment. We demonstrated that MIRROR is applicable to both short chemically synthesized gRNAs with modifications and long biologically generated gRNAs and surpasses current state-of-the-art approaches in both gRNA forms. It enhances editing efficiency by up to 5.7-fold in multiple human cell types and primary hepatocytes from an alpha-1 antitrypsin deficiency mouse model. Our findings improve programmable RNA editing in vitro and in vivo by rational design through the screening of highly edited natural substrate mimics.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$259.00 per year
only $21.58 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The sequence data were uploaded to the National Center for Biotechnology Information under BioProject PRJNA1163502. TxDb.Hsapiens.UCSC.hg19.knownGene is available online (https://bioconductor.org/packages/TxDb.Hsapiens.UCSC.hg19.knownGene/). Source data are provided with this paper.
Code availability
Executable programs and source code of MIRROR (version 1.0) are publicly available on GitHub (https://github.com/Y2C99/MIRROR) for free, noncommercial use.
References
Gagnidze, K., Rayon-Estrada, V., Harroch, S., Bulloch, K. & Papavasiliou, F. N. A new chapter in genetic medicine: RNA editing and its role in disease pathogenesis. Trends Mol. Med. 24, 294–303 (2018).
Gold, A., Levanon, E. Y. & Eisenberg, E. The new RNA-editing era—ethical considerations. Trends Genet. 37, 685–687 (2021).
Casati, B., Pinamonti, V., Pecori, R., Lindner, J. M. & Papavasiliou, F. N. Neoepitope formation through the generation of RNA-derived ‘editopes’. Preprint at bioRxiv https://doi.org/10.1101/2023.03.16.532918 (2023).
Booth, B. J. et al. RNA editing: expanding the potential of RNA therapeutics. Mol. Ther. 31, 1533–1549 (2023).
Diaz Quiroz, J. F., Siskel, L. D. & Rosenthal, J. J. C. Site-directed A → I RNA editing as a therapeutic tool: moving beyond genetic mutations. RNA 29, 498–505 (2023).
Dadush, A. et al. DNA and RNA base editors can correct the majority of pathogenic single nucleotide variants. NPJ Genom. Med. 9, 16 (2024).
Chen, L.-L. et al. Voices: challenges and opportunities in RNA biology. Cell Chem. Biol. 31, 10–13 (2024).
Khosravi, H. M. & Jantsch, M. F. Site-directed RNA editing: recent advances and open challenges. RNA Biol. 18, 41–50 (2021).
Pfeiffer, L. S. & Stafforst, T. Precision RNA base editing with engineered and endogenous effectors. Nat. Biotechnol. 41, 1526–1542 (2023).
Stafforst, T. & Schneider, M. F. An RNA-deaminase conjugate selectively repairs point mutations. Angew. Chem. Int. Ed. 51, 11166–11169 (2012).
Montiel-Gonzalez, M. F., Vallecillo-Viejo, I., Yudowski, G. A. & Rosenthal, J. J. C. Correction of mutations within the cystic fibrosis transmembrane conductance regulator by site-directed RNA editing. Proc. Natl Acad. Sci. USA 110, 18285–18290 (2013).
Cox, D. B. T. et al. RNA editing with CRISPR–Cas13. Science 358, 1019–1027 (2017).
Sinnamon, J. R. et al. Site-directed RNA repair of endogenous Mecp2 RNA in neurons. Proc. Natl Acad. Sci. USA 114, E9395–E9402 (2017).
Wettengel, J., Reautschnig, P., Geisler, S., Kahle, P. J. & Stafforst, T. Harnessing human ADAR2 for RNA repair—recoding a PINK1 mutation rescues mitophagy. Nucleic Acids Res. 45, 2797–2808 (2017).
Fukuda, M. et al. Construction of a guide-RNA for site-directed RNA mutagenesis utilising intracellular A-to-I RNA editing. Sci. Rep. 7, 41478 (2017).
Vogel, P. et al. Efficient and precise editing of endogenous transcripts with SNAP-tagged ADARs. Nat. Methods 15, 535–538 (2018).
Katrekar, D. et al. In vivo RNA editing of point mutations via RNA-guided adenosine deaminases. Nat. Methods 16, 239–242 (2019).
Merkle, T. et al. Precise RNA editing by recruiting endogenous ADARs with antisense oligonucleotides. Nat. Biotechnol. 37, 133–138 (2019).
Reautschnig, P. et al. CLUSTER guide RNAs enable precise and efficient RNA editing with endogenous ADAR enzymes in vivo. Nat. Biotechnol. 40, 759–768 (2022).
Katrekar, D. et al. Efficient in vitro and in vivo RNA editing via recruitment of endogenous ADARs using circular guide RNAs. Nat. Biotechnol. 40, 938–945 (2022).
Yi, Z. et al. Engineered circular ADAR-recruiting RNAs increase the efficiency and fidelity of RNA editing in vitro and in vivo. Nat. Biotechnol. 40, 946–955 (2022).
Monian, P. et al. Endogenous ADAR-mediated RNA editing in non-human primates using stereopure chemically modified oligonucleotides. Nat. Biotechnol. 40, 1093–1102 (2022).
Wong, S. K., Sato, S. & Lazinski, D. W. Substrate recognition by ADAR1 and ADAR2. RNA 7, 846–858 (2001).
Lehmann, K. A. & Bass, B. L. The importance of internal loops within RNA substrates of ADAR1. J. Mol. Biol. 291, 1–13 (1999).
Tian, N. et al. A structural determinant required for RNA editing. Nucleic Acids Res. 39, 5669–5681 (2011).
Eggington, J. M., Greene, T. & Bass, B. L. Predicting sites of ADAR editing in double-stranded RNA. Nat. Commun. 2, 319 (2011).
Ramaswami, G. et al. Genetic mapping uncovers cis-regulatory landscape of RNA editing. Nat. Commun. 6, 8194 (2015).
Song, Y. et al. irCLASH reveals RNA substrates recognized by human ADARs. Nat. Struct. Mol. Biol. 27, 351–362 (2020).
Zhang, H. et al. Human A-to-I RNA editing SNP loci are enriched in GWAS signals for autoimmune diseases and under balancing selection. Genome Biol. 21, 288 (2020).
Liu, X. et al. Learning cis-regulatory principles of ADAR-based RNA editing from CRISPR-mediated mutagenesis. Nat. Commun. 12, 2165 (2021).
Uzonyi, A. et al. Deciphering the principles of the RNA editing code via large-scale systematic probing. Mol. Cell 81, 2374–2387.e3 (2021).
Zambrano-Mila, M. S. et al. Dissecting the basis for differential substrate specificity of ADAR1 and ADAR2. Nat. Commun. 14, 8212 (2023).
Jacobsen, C. S. et al. Library screening reveals sequence motifs that enable ADAR2 editing at recalcitrant sites. ACS Chem. Biol. 18, 2188–2199 (2023).
Diaz Quiroz, J. F. et al. Development of a selection assay for small guide RNAs that drive efficient site-directed RNA editing. Nucleic Acids Res. 51, e41 (2023).
Levanon, E. Y. et al. Systematic identification of abundant A-to-I editing sites in the human transcriptome. Nat. Biotechnol. 22, 1001–1005 (2004).
Bazak, L. et al. A-to-I RNA editing occurs at over a hundred million genomic sites, located in a majority of human genes. Genome Res. 24, 365–376 (2014).
Tan, M. H. et al. Dynamic landscape and regulation of RNA editing in mammals. Nature 550, 249–254 (2017).
Kleinberger, Y. & Eisenberg, E. Large-scale analysis of structural, sequence and thermodynamic characteristics of A-to-I RNA editing sites in human Alu repeats. BMC Genomics 11, 453 (2010).
Bazak, L., Levanon, E. Y. & Eisenberg, E. Genome-wide analysis of Alu editability. Nucleic Acids Res. 42, 6876–6884 (2014).
Yamamoto, R., Liu, Z., Choudhury, M. & Xiao, X. dsRID: in silico identification of dsRNA regions using long-read RNA-seq data. Bioinformatics 39, btad649 (2023).
Matthews, M. M. et al. Structures of human ADAR2 bound to dsRNA reveal base-flipping mechanism and basis for site selectivity. Nat. Struct. Mol. Biol. 23, 426–433 (2016).
Roberts, T. C., Langer, R. & Wood, M. J. A. Advances in oligonucleotide drug delivery. Nat. Rev. Drug Discov. 19, 673–694 (2020).
Strnad, P., McElvaney, N. G. & Lomas, D. A. Alpha1-Antitrypsin Deficiency. N. Engl. J. Med. 382, 1443–1455 (2020).
Doherty, E. E. et al. ADAR activation by inducing a syn conformation at guanosine adjacent to an editing site. Nucleic Acids Res. 50, 10857–10868 (2022).
Pecori, R. & Papavasiliou, N. F. It takes two (and some distance) to tango: how ADARs join to edit RNA. Nat. Struct. Mol. Biol. 27, 308–310 (2020).
Yu, G., Wang, L. G. & He, Q. Y. ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics 31, 2382–2383 (2015).
Ramaswami, G. & Li, J. B. RADAR: a rigorously annotated database of A-to-I RNA editing. Nucleic Acids Res. 42, D109–D113 (2014).
Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842 (2010).
Rehmsmeier, M., Steffen, P., Hochsmann, M. & Giegerich, R. Fast and effective prediction of microRNA/target duplexes. RNA 10, 1507–1517 (2004).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Litke, J. L. & Jaffrey, S. R. Highly efficient expression of circular RNA aptamers in cells using autocatalytic transcripts. Nat. Biotechnol. 37, 667–675 (2019).
Charni-Natan, M. & Goldstein, I. Protocol for primary mouse hepatocyte isolation. STAR Protoc. 1, 100086 (2020).
Acknowledgements
We thank Q. Lai, C. Li and Y. Niu for the technical assistance. We thank ChatGPT for assisting us in naming this method. This study was supported by grants from National Key R&D Program of China (2020YFA0509400 and 2024YFC3405901 to R.Z.), Guangzhou Basic and Applied Basic Research Funds (2024A04J3211 to R.Z.), Guangzhou Agricultural and Social Development Science and Technology Funds (2024B03J0004 to R.Z.), the Fundamental Research Funds for the Central Universities, Sun Yat-Sen University (24lgzy005 to R.Z.) and Guangdong Science and Technology Department (2021A1515012463 to W.B.Y.). The GTEx project was supported by the Common Fund of the Office of the Director of the National Institutes of Health (https://commonfund.nih.gov/genotype-tissue-expression-gtex).
Author information
Authors and Affiliations
Contributions
R.Z. conceptualized the project. Y.F.S., W.B.Y., Y.C., J.L., Y.H.H., W.H. and G.D.X. conducted the experiments. Y.C., Y.L.S. and Y.F.S. performed the bioinformatics analysis. R.Z. wrote the paper with input from all authors.
Corresponding authors
Ethics declarations
Competing interests
W.B.Y., R.Z. and J.L. are inventors of filed patents based on the work published here. R.Z. and W.B.Y. are cofounders and shareholders of RecoRNA Biotechnology. J.L., Y.H.H., G.D.X., W.H. and W.B.Y. are employees at RecoRNA Biotechnology. The other authors declare no competing interests.
Peer review
Peer review information
Nature Biotechnology thanks Michael Jantsch, Fotini Papavasiliou and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Characterization of Alu editing sites used for analysis.
a, Genic locations of Alu editing sites present in the GTEx tissues. b, The number of Alu editing sites for different triplet motifs. Sites that were uniquely mapped by ≥20 reads in ≥ 10 samples were used. c, Representative editing level distribution of sites with different triplet motifs.
Extended Data Fig. 2 Characterization and verification of inverted Alu pairs.
a, An example of the secondary structure of an inverted Alu pair predicted by RNAhybrid. b, Cumulative distribution showing the number of editing sites in each Alu element. c, The proportion of edited and unedited ECSs. d, Editing levels decreased with increasing distances between the two arms of edited inverted Alu pairs. The mean editing level is defined as the average editing level across all Alu editing sites within an inverted Alu pair. Box plots show the median and the 25th and 75th percentiles, with whiskers extending 1.5 times the interquartile range. Sample sizes are provided in the source data. e, The proportion of single Alus and inverted Alus that overlapped with dsRID-identified dsRNA regions.
Extended Data Fig. 3 Editing levels of top editing sites with different triplet complementary motifs.
Box plots show the median and the 25th and 75th percentiles, with whiskers extending 1.5 times the interquartile range. Sample sizes are provided in the Source data.
Extended Data Fig. 4 The development of the high-throughput editing efficiency measurement system.
a, Details of targeted sequencing for editing reporters. The regions containing the barcode and part of the target, including the editing sites, were amplified. The library size was approximately 300 bp, and sequencing was performed using a PE150 strategy. We ensured that the editing sites were sequenced in Read 1, while the barcode was sequenced in Read 2. Sequencing depth was set to ensure that the number of reads was at least 5,000 times larger than the number of gRNAs in the library. b, The pipeline used to analyze targeted RNA-seq data to determine the editing efficiency of each gRNA. c, Schematic illustrating the design of 41 and 71 nt gRNAs to target a UAG site in the GAPDH gene. d, Scatter plot showing the correlation of editing efficiencies between replicates. Pearson correlation coefficients are indicated. e, The relationship between original editing levels in GTEx data and MIRROR gRNA-mediated editing efficiencies in the screening system. gRNAs were grouped based on the original editing levels of their mirrored substrates. Sample sizes are provided in the source data. f, The relationship between the numbers of unpaired bases (wobble pairs, loops, and bulges) and MIRROR gRNA-mediated editing efficiencies in the screening system. e,f, Box plots show the median and the 25th and 75th percentiles, with whiskers extending 1.5 times the interquartile range. Sample sizes are provided in the source data.
Extended Data Fig. 5 The reproducibility and reliability of the high-throughput screening data of short MIRROR gRNAs.
a, Scatter plot showing the correlation of editing efficiencies between replicates. The fully complementary gRNAs with an A-C mismatch at the targeted adenosine position are shown in orange. b, Expression levels of ADAR1 and ADAR2 across various cell lines. Data are from https://www.proteinatlas.org/. c, Heatmap showing the Pearson correlation of gRNA editing efficiencies across various cell lines. The mean editing efficiency of each gRNA, averaged across biological replicates, was used for the analysis in each cell line. d, Comparison between editing efficiencies measured from the pooled screening data and those obtained by cloning individual gRNAs for editing efficiency measurement. e, Scatter plot showing the correlation of editing efficiencies between replicates for the 9 additional sites.
Extended Data Fig. 6 Application of two structural rules to additional sites.
We chose two MIRROR gRNAs that significantly improved editing efficiency at the UAC and UAU sites in the GAPDH gene, and applied their structural features to four other sites. Values are presented as mean ± s.e.m. n = 3. P values, unpaired one-sided Student’s t-test.
Extended Data Fig. 7 gRNA concentrations in liver tissues.
a, gRNA concentrations in livers for mice treated with MIRROR gRNAs with different doses. The mice underwent a single intravenous tail vein injection, and their livers were collected two days later for gRNA quantification. Values are presented as mean ± s.e.m. n = 4. b, gRNA concentrations in livers for indicated MIRROR gRNAs over time. Mice were injected at a single dose of 3 mg kg−1 and livers were collected at different time points across two weeks. Values are presented as mean ± s.e.m. n = 4.
Extended Data Fig. 8 A framework for high-throughput measurements of long MIRROR gRNA editing efficiencies.
a, Schematic overview of the high-throughput editing efficiency measurement system for long gRNAs. The reporter expresses both substrates (reporter) and individual gRNAs in the 3′ UTR region of the GFP gene, with a linker employed to connect these two components. Each gRNA is uniquely associated with a barcode, ensuring accurate gRNA assignment. In contrast to the short MIRROR gRNA screening system, where reporter and gRNAs can be sequenced in the same reactions, the lengths of reporters with gRNAs surpassed the sequencing length limitations of NGS. Consequently, we sequenced the reporter and barcode, as well as the barcode and gRNAs, respectively. Subsequently, we integrated both sets of data to determine the corresponding editing efficiencies for each gRNA. b, The pipeline used to analyze targeted RNA-seq to assign editing efficiency to each gRNA. We analyzed the targeted RNA-seq data and the plasmid DNA-seq data separately and then integrated both sets of data to determine the editing efficiencies corresponding to each gRNA.
Extended Data Fig. 9 The diagram of CLUSTER gRNA design.
a, Illustration of the binding sites and target sequences of the three CLUSTER gRNAs targeting TPT1, SRSF1, and RAB7A genes, generated using the recruitment-cluster-finder. b, Secondary structure prediction of the circular CLUSTER gRNAs, generated using the ViennaRNA Package 2.0.
Extended Data Fig. 10 Low-throughput screening of MIRROR gRNAs and feature analysis of high-throughput MIRROR gRNAs.
a, The program for low-throughput screening was applied to one randomly selected site in the ACTB and GAPDH transcripts, respectively. For each site, the top 6 chemically modified MIRROR gRNAs were synthesized, and their editing efficiencies were compared to fully complementary gRNAs with an A-C mismatch in HeLa cells. gRNA concentration, 40nM. Values are presented as mean ± s.e.m. n = 3. P values, unpaired one-sided Student’s t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. b, Framework for XGBoost model training and SHAP value interpretation. Editing levels obtained from high-throughput screening were used as labels for model training. Structural features were extracted from RNAhybrid predictions, with each feature assigned a specific value. These structural features were encoded based on their distance from the editing site. c, SHAP values from the XGBoost model prediction on the test dataset. The top 12 features are shown, with the color bar indicating the feature value in the input dataset. If a specific structural type was present, its feature value was set to 1; otherwise, it was set to 0. Each dot represents a gRNA, the Y-axis shows the feature categories, and the X-axis indicates the impact of each feature on the prediction of each gRNA. Features downstream of the editing site were denoted by ‘+,’ while those upstream were denoted by ‘−’. d, Intermolecular structures of validated modified gRNAs with improved editing efficiency. The matched features for each MIRROR gRNA are highlighted in (c).
Supplementary information
Supplementary Information
Supplementary Fig. 1 and Notes 1 and 2.
Supplementary Tables 1–7
Supplementary Table 1: Editing sites and inverted Alu pair annotations. Supplementary Table 2: Long biologically generated gRNA sequences for triplet motif analysis. Supplementary Table 3: Reporter sequences used for RNA base editing. Supplementary Table 4: Statistics for the oligo pools and targeted RNA-seq and DNA-seq libraries. Supplementary Table 5: Chemically modified gRNA sequences. Supplementary Table 6: Long biologically generated gRNA sequences. Supplementary Table 7: Primers used for targeted RNA-seq and Sanger sequencing.
Supplementary Software 1
Code for MIRROR gRNA design.
Source data
Source Data Fig. 1
Statistical source data.
Source Data Fig. 2
Statistical source data.
Source Data Fig. 3
Statistical source data.
Source Data Fig. 4
Statistical source data.
Source Data Fig. 5
Statistical source data.
Source Data Extended Data Fig. 1
Statistical source data.
Source Data Extended Data Fig. 2
Statistical source data.
Source Data Extended Data Fig. 3
Statistical source data.
Source Data Extended Data Fig. 4
Statistical source data.
Source Data Extended Data Fig. 6
Statistical source data.
Source Data Extended Data Fig. 7
Statistical source data.
Source Data Extended Data Fig. 10
Statistical source data.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Sun, Y., Cao, Y., Song, Y. et al. Improved RNA base editing with guide RNAs mimicking highly edited endogenous ADAR substrates. Nat Biotechnol (2025). https://doi.org/10.1038/s41587-025-02628-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41587-025-02628-6
