Recently developed DNA base editing methods enable the direct generation of desired point mutations in genomic DNA without generating any double-strand breaks1,2,3, but the issue of off-target edits has limited the application of these methods. Although several previous studies have evaluated off-target mutations in genomic DNA4,5,6,7,8, it is now clear that the deaminases that are integral to commonly used DNA base editors often bind to RNA9,10,11,12,13. For example, the cytosine deaminase APOBEC1—which is used in cytosine base editors (CBEs)—targets both DNA and RNA12, and the adenine deaminase TadA—which is used in adenine base editors (ABEs)—induces site-specific inosine formation on RNA9,11. However, any potential RNA mutations caused by DNA base editors have not been evaluated. Adeno-associated viruses are the most common delivery system for gene therapies that involve DNA editing; these viruses can sustain long-term gene expression in vivo, so the extent of potential RNA mutations induced by DNA base editors is of great concern14,15,16. Here we quantitatively evaluated RNA single nucleotide variations (SNVs) that were induced by CBEs or ABEs. Both the cytosine base editor BE3 and the adenine base editor ABE7.10 generated tens of thousands of off-target RNA SNVs. Subsequently, by engineering deaminases, we found that three CBE variants and one ABE variant showed a reduction in off-target RNA SNVs to the baseline while maintaining efficient DNA on-target activity. This study reveals a previously overlooked aspect of off-target effects in DNA editing and also demonstrates that such effects can be eliminated by engineering deaminases.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Evolution of an adenine base editor into a small, efficient cytosine base editor with low off-target activity
Nature Biotechnology Open Access 10 November 2022
Cell Death Discovery Open Access 02 August 2022
Nature Communications Open Access 19 May 2022
Subscribe to Nature+
Get immediate online access to Nature and 55 other Nature journal
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
All the sequencing data have been deposited in the NCBI SRA under project accession numbers PRJNA528149 and PRJNA528561 or at http://www.biosino.org/node/project/detail/OEP000277. All materials are available upon reasonable request.
The authors declare that all code used in this study are available within the article and its Extended Data or from the corresponding author upon reasonable request.
Rees, H. A. & Liu, D. R. Base editing: precision chemistry on the genome and transcriptome of living cells. Nat. Rev. Genet. 19, 770–788 (2018).
Komor, A. C., Kim, Y. B., Packer, M. S., Zuris, J. A. & Liu, D. R. Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage. Nature 533, 420–424 (2016).
Gaudelli, N. M. et al. Programmable base editing of A.T to G.C in genomic DNA without DNA cleavage. Nature 551, 464–471 (2017).
Kim, D., Kim, D. E., Lee, G., Cho, S. I. & Kim, J. S. Genome-wide target specificity of CRISPR RNA-guided adenine base editors. Nat. Biotechnol. 37, 430–435 (2019).
Gehrke, J. M. et al. An APOBEC3A-Cas9 base editor with minimized bystander and off-target activities. Nat. Biotechnol. 36, 977–982 (2018).
Kim, D. et al. Genome-wide target specificities of CRISPR RNA-guided programmable deaminases. Nat. Biotechnol. 35, 475–480 (2017).
Zuo, E. et al. Cytosine base editor generates substantial off-target single-nucleotide variants in mouse embryos. Science 364, 289–292 (2019).
Jin, S. et al. Cytosine, but not adenine, base editors induce genome-wide off-target mutations in rice. Science 364, 292–295 (2019).
Poulsen, L. K., Larsen, N. W., Molin, S. & Andersson, P. Analysis of an Escherichia coli mutant strain resistant to the cell-killing function encoded by the gef gene family. Mol. Microbiol. 6, 895–905 (1992).
Sowden, M., Hamm, J. K. & Smith, H. C. Overexpression of APOBEC-1 results in mooring sequence-dependent promiscuous RNA editing. J. Biol. Chem. 271, 3011–3017 (1996).
Wolf, J., Gerber, A. P. & Keller, W. tadA, an essential tRNA-specific adenosine deaminase from Escherichia coli. EMBO J. 21, 3841–3851 (2002).
Conticello, S. G. The AID/APOBEC family of nucleic acid mutators. Genome Biol. 9, 229 (2008).
Blanc, V. & Davidson, N. O. APOBEC-1-mediated RNA editing. Wiley Interdiscip. Rev. Syst. Biol. Med. 2, 594–602 (2010).
Villiger, L. et al. Treatment of a metabolic liver disease by in vivo genome base editing in adult mice. Nat. Med. 24, 1519–1525 (2018).
Maeder, M. L. et al. Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10. Nat. Med. 25, 229–233 (2019).
Rossidis, A. C. et al. In utero CRISPR-mediated therapeutic editing of metabolic genes. Nat. Med. 24, 1513–1518 (2018).
Green, P., Ewing, B., Miller, W., Thomas, P. J. & Green, E. D. Transcription-associated mutational asymmetry in mammalian evolution. Nat. Genet. 33, 514–517 (2003).
Mitchell, A. & Graur, D. Inferring the pattern of spontaneous mutation from the pattern of substitution in unitary pseudogenes of Mycobacterium leprae and a comparison of mutation patterns among distantly related organisms. J. Mol. Evol. 61, 795–803 (2005).
Duret, L. Mutation patterns in the human genome: more variable than expected. Plos Biol. 7, 217–219 (2009).
Chen, K. M. et al. Structure of the DNA deaminase domain of the HIV-1 restriction factor APOBEC3G. Nature 452, 116–119 (2008).
Holden, L. G. et al. Crystal structure of the anti-viral APOBEC3G catalytic domain and functional implications. Nature 456, 121–124 (2008).
Kim, Y. B. et al. Increasing the genome-targeting scope and precision of base editing with engineered Cas9-cytidine deaminase fusions. Nat. Biotechnol. 35, 371–376 (2017).
Wang, X. et al. Efficient base editing in methylated regions with a human APOBEC3A-Cas9 fusion. Nat. Biotechnol. 36, 946–949 (2018).
Stauch, B. et al. Model structure of APOBEC3C reveals a binding pocket modulating ribonucleic acid interaction required for encapsidation. Proc. Natl Acad. Sci. USA 106, 12079–12084 (2009).
Shi, K. et al. Structural basis for targeted DNA cytosine deamination and mutagenesis by APOBEC3A and APOBEC3B. Nat. Struct. Mol. Biol. 24, 131–139 (2017).
Xiang, S., Short, S. A., Wolfenden, R. & Carter, C. W. Jr The structure of the cytidine deaminase-product complex provides evidence for efficient proton transfer and ground-state destabilization. Biochemistry 36, 4768–4774 (1997).
Kim, J. et al. Structural and kinetic characterization of Escherichia coli TadA, the wobble-specific tRNA deaminase. Biochemistry 45, 6407–6416 (2006).
Grünewald, J. et al. Transcriptome-wide off-target RNA editing induced by CRISPR-guided DNA base editors. Nature 569, 433–437 (2019).
Rees, H. A., Wilson, C., Doman, J. L. & Liu, D. R. Analysis and minimization of cellular RNA editing by DNA adenine base editors. Sci. Adv. 5, eaax5717 (2019).
Grünewald, J. et al. CRISPR adenine and cytosine base editors with reduced RNA off-target activities. Preprint at https://doi.org/10.1101/631721 (2019).
Koblan, L. W. et al. Improving cytidine and adenine base editors by expression optimization and ancestral reconstruction. Nat. Biotechnol. 36, 843–846 (2018).
Zafra, M. P. et al. Optimized base editors enable efficient editing in cells, organoids and mice. Nat. Biotechnol. 36, 888–893 (2018).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
McKenna, A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303 (2010).
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-seq data with or without a reference genome. BMC Bioinformatics 12, 323 (2011).
Crooks, G. E. et al. WebLogo: a sequence logo generator. Genome Res. 14, 1188–1190 (2004).
Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. Protocols 9, 171–181 (2014).
Gu, C., Liu, S., Wu, Q., Zhang, L. & Guo, F. Integrative single-cell analysis of transcriptome, DNA methylome and chromatin accessibility in mouse oocytes. Cell Res. 29, 110–123 (2019).
Anders, S., Pyl, P. T. & Huber, W. HTSeq—a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015).
We thank M. Poo for discussions and comments on this manuscript; D. Li for discussions; and FACS facility H. Wu and L. Quan in the Institute of Neuroscience (ION) and M. Zhang in the Institute Pasteur of Shanghai (IPS), L. Yuan in Big Data Platform, Shanghai Institutes for Biological Sciences (SIBS) and Chinese Academy of Sciences (CAS). This work was supported by R&D Program of China (2017YFC1001302, 2018YFC2000100, 2018YFA0107701, and 2018YFC1003401), CAS Strategic Priority Research Program (XDB32060000), National Natural Science Foundation of China (31871502, 31522037, 31822035, 31822035, 31771590), Shanghai Municipal Science and Technology Major Project (2018SHZDZX05), Shanghai City Committee of Science and Technology Project (18411953700, 18JC1410100), National Science and Technology Major Project (2015ZX10004801-005) and National Key Research and Development Program of China (2017YFA0505500, 2016YFC0901704).
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
The schematics show the plasmids used in this study.
a, Representative distributions of off-target RNA SNVs on human chromosomes for APOBEC1, BE3–site 3, BE3–RNF2, TadA–TadA*, ABE7.10–site 1 and ABE7.10–site 2. b, Schematics of BE3(FNLS) and ABEmax. Note that BE3(FNLS)31 and ABEmax32 have previously been reported to greatly increase the expression of base editors. c, Expression of APOBEC1 in cells transfected with BE3–site 3 for 36 or 72 h, or with BE3(FNLS)–site 3 for 72 h. d, Expression level of in cells transfected with ABE7.10–site 1 for 36 and 72 h, or with ABEmax–site 1 for 72 h. e, The number of off-target RNA SNVs in cells transfected with BE3–site 3 for 36 or 72 h, or with BE3(FNLS)–site 3 for 72 h. f, The number of off-target RNA SNVs in cells transfected with ABE7.10–site 1 for 36 or 72 h, or with ABEmax–site 1 for 72 h. Transfections: GFP for 36 h; BE3–site 3 for 36 h; GFP for 72 h; BE3–site 3 for 72 h; BE3(FNLS)–site 3 for 72 h; ABE7.10–site 1 for 36 h; ABE7.10–site 1 for 72 h; ABEmax–site 1 for 72 h. RSEM, RNA-seq by expectation maximization. All values are presented as mean ± s.e.m. Number above the bar indicates the number of biologically independent samples. *P < 0.05, **P < 0.01, ***P < 0.001, two-sided unpaired t-test. Exact P values are provided in Supplementary Table 15.
a, Distribution of mutation types in each repeat of all groups. The number in each cell indicates the percentage of a certain type of mutation among all mutations. b, Expression of genes containing overlapping off-target RNA SNVs and random simulated genes in all groups transfected with BE3 (n = 8 biologically independent samples) or ABE7.10 (n = 9 biologically independent samples). Two-sided unpaired t-test. Box-and-whisker plots: centre line indicates median, box represents first and third quantiles, and whiskers indicate maximum and minimum values.
a, Similarity between adjacent sequences of off-target RNA SNVs and on-target sequences. n = 2 biologically independent samples for BE3–site 2, n = 3 for BE3–RNF2, n = 3 for ABE7.10–site 1, n = 3 for ABE7.10–site 2. b, Sanger sequencing chromatograms show that C to U mutation was observed only in RNA but not DNA for two BE3 off-target sites. OF, off-target. Gene names, amino acid mutations and single nucleotide conversions are indicated by blue, red and green, respectively. c, Sanger sequencing chromatograms show that U to C mutation was observed only in the RNA of three ABE7.10 off-target sites.
a, Percentages of different locations of SNVs for GFP, BE3 (BE3, BE3–site 3 and BE3–RNF2) and ABE7.10 (ABE7.10, ABE7.10–site 1 and ABE7.10–site 2) groups. All values are presented as mean ± s.e.m. n denotes biologically independent samples. *P < 0.05, **P < 0.01, ***P < 0.001, two-sided unpaired t-test. Exact P values are provided in Supplementary Table 16. b, Editing rate of BE3-induced non-synonymous mutations located on oncogenes and tumour suppressor genes. c, Editing rate of ABE7.10-induced non-synonymous mutations located on oncogenes and tumour suppressor genes. Gene names, amino acid mutations and single nucleotide conversions are indicated by blue, red and green, respectively.
Extended Data Fig. 6 Expression of transfected vectors and mutation types of off-target RNA SNVs in single cells.
a, Expression of GFP, APOBEC1 and TadA–TadA* was quantified in all sequenced single cells. Thresholds are indicated by blue dashed lines. Thresholds of log2 (FPKM + 1) for GFP, BE3 and ABE7.10 are 0.3, 1 and 0.3, respectively. Cells with expression levels higher than the threshold were included for further analysis. b, c, Cells with high expression of TadA–TadA* or APOBEC1 showed greater numbers of RNA SNVs than those with low expressions in the ABE7.10 (n = 9 cells) or BE3 group (n = 4 cells), respectively. Box-and-whisker plots: centre line indicates median value, box represents first and third quantile, whisker indicates maximum and minimum values. d, Distribution of mutation types for GFP-transfected single cells (n = 16 cells). e, Distribution of mutation types for BE3–site 3-transfected single cells (n = 31 cells). Cells with expression of APOBEC1 higher than the threshold are included in the red square. f, Distribution of mutation types for ABE7.10–site 1-transfected single cells (n = 28 cells). Cells with expression of TadA–TadA* higher than the threshold are included in the red square. The number indicates the percentage of a certain type of mutation among all mutations. SC, single cell.
Extended Data Fig. 7 Distribution of off-target RNA SNVs on human chromosomes for all single cells with expression above thresholds.
a, Distribution of off-target RNA SNVs on human chromosomes for GFP-transfected single cells (n = 15 cells). b, Distribution of off-target RNA SNVs on human chromosomes for BE3–site 3-transfected single cells (n = 4 cells). c, Distribution of off-target RNA SNVs on human chromosomes for ABE7.10–site 1-transfected single cells (n = 9 cells).
a, b, The ratio of shared SNVs between any two samples in the same group or with predicted off-target sites by Cas-OFFinder. The proportion in each cell is calculated by the number of overlapping SNVs between two samples divided by the sample in the row. c, Editing rate of BE3-induced non-synonymous mutations on oncogenes and tumour suppressor genes in single cells. d, Editing rate of ABE7.10-induced non-synonymous mutations on oncogenes and tumour suppressor genes in single cells. Gene names, amino acid mutations and single nucleotide conversions are indicated by blue, red and green, respectively.
Extended Data Fig. 9 Characteristics of off-target RNA SNVs for engineered BE3 and ABE7.10 variants.
a, Schematic of BE3 and ABE7.10 variants. Point mutations are indicated by red lines. b, Representative distributions of off-target RNA SNVs on human chromosomes. c, Distribution of mutation types for each sample of the engineered variants of BE3 and ABE7.10. d, Ratio of shared RNA SNVs between any two samples in the engineered variants of BE3 and ABE7.10 or with off-target sites predicted by Cas-OFFinder. The proportion in each cell was calculated by the number of overlapping RNA SNVs between two samples divided by the number of RNA SNVs in the row.
a, Comparison of the width of editing windows between ABE7.10 and ABE7.10F148A. n = 3 biologically independent samples for each group. b, DNA on-target efficiency on site 1 of TadA–TadA*–Cas9n (wild-type TadA–evolved TadA heterodimer), TadAF148A–TadA*–Cas9n and TadAF148A–TadA*F148A–Cas9n. n = 3 biologically independent samples for each group. c, The number of RNA SNVs in the GFP, TadA–TadA*–Cas9n, TadAF148A–TadA*–Cas9n and TadAF148A–TadA*F148A–Cas9n groups. n = 3 biologically independent samples for each group. All values are presented as mean ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001. Two-sided unpaired t-test. Exact P values are provided in Supplementary Table 17.
This file contains Supplementary Tables 1-3 and Supplementary Tables 8-17.
SNVs located on oncogenes or tumour suppressor genes of BE3-transfected samples.
SNVs located on oncogenes or tumour suppressor genes of ABE7.10-transfected samples.
SNVs located on oncogenes or tumour suppressor genes of BE3-transfected single cells.
SNVs located on oncogenes or tumour suppressor genes of ABE7.10-transfected single cells.
About this article
Cite this article
Zhou, C., Sun, Y., Yan, R. et al. Off-target RNA mutation induced by DNA base editing and its elimination by mutagenesis. Nature 571, 275–278 (2019). https://doi.org/10.1038/s41586-019-1314-0