Base editor technology, which uses CRISPR–Cas9 to direct cytidine deaminase enzymatic activity to specific genomic loci, enables the highly efficient introduction of precise cytidine-to-thymidine DNA alterations1,2,3,4,5,6. However, existing base editors create unwanted C-to-T alterations when more than one C is present in the enzyme's five-base-pair editing window. Here we describe a strategy for reducing bystander mutations using an engineered human APOBEC3A (eA3A) domain, which preferentially deaminates cytidines in specific motifs according to a TCR>TCY>VCN hierarchy. In direct comparisons with the widely used base editor 3 (BE3) fusion in human cells, our eA3A-BE3 fusion exhibits similar activities on cytidines in TC motifs but greatly reduced editing on cytidines in other sequence contexts. eA3A-BE3 corrects a human β-thalassemia promoter mutation with much higher (>40-fold) precision than BE3. We also demonstrate that eA3A-BE3 shows reduced mutation frequencies on known off-target sites of BE3, even when targeting promiscuous homopolymeric sites.
Subscribe to Journal
Get full journal access for 1 year
only $20.83 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
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).
Nishida, K. et al. Targeted nucleotide editing using hybrid prokaryotic and vertebrate adaptive immune systems. Science 353, aaf8729 (2016).
Kim, K. et al. Highly efficient RNA-guided base editing in mouse embryos. Nat. Biotechnol. 35, 435–437 (2017).
Rees, H.A. et al. Improving the DNA specificity and applicability of base editing through protein engineering and protein delivery. Nat. Commun. 8, 15790 (2017).
Shimatani, Z. et al. Targeted base editing in rice and tomato using a CRISPR-Cas9 cytidine deaminase fusion. Nat. Biotechnol. 35, 441–443 (2017).
Hess, G.T. et al. Directed evolution using dCas9-targeted somatic hypermutation in mammalian cells. Nat. Methods 13, 1036–1042 (2016).
Doudna, J.A. & Charpentier, E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science 346, 1258096 (2014).
Cox, D.B.T., Platt, R.J. & Zhang, F. Therapeutic genome editing: prospects and challenges. Nat. Med. 21, 121–131 (2015).
Sander, J.D. & Joung, J.K. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat. Biotechnol. 32, 347–355 (2014).
Komor, A.C. et al. Improved base excision repair inhibition and bacteriophage Mu Gam protein yields C:G-to-T:A base editors with higher efficiency and product purity. Sci. Adv. 3, eaao4774 (2017).
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).
Logue, E.C. et al. A DNA sequence recognition loop on APOBEC3A controls substrate specificity. PLoS One 9, e97062 (2014).
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).
Kouno, T. et al. Crystal structure of APOBEC3A bound to single-stranded DNA reveals structural basis for cytidine deamination and specificity. Nat. Commun. 8, 15024 (2017).
Bohn, M.-F. et al. The ssDNA mutator APOBEC3A is regulated by cooperative dimerization. Structure 23, 903–911 (2015).
Wang, M., Yang, Z., Rada, C. & Neuberger, M.S. AID upmutants isolated using a high-throughput screen highlight the immunity/cancer balance limiting DNA deaminase activity. Nat. Struct. Mol. Biol. 16, 769–776 (2009).
Kim, D. et al. Genome-wide target specificities of CRISPR RNA-guided programmable deaminases. Nat. Biotechnol. 35, 475–480 (2017).
Tsai, S.Q. et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat. Biotechnol. 33, 187–197 (2015).
Bae, S., Park, J. & Kim, J.-S. Cas-OFFinder: a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics 30, 1473–1475 (2014).
Kleinstiver, B.P. et al. High-fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects. Nature 529, 490–495 (2016).
Chen, J.S. et al. Enhanced proofreading governs CRISPR-Cas9 targeting accuracy. Nature 550, 407–410 (2017).
Wang, L. et al. Enhanced base editing by co-expression of free uracil DNA glycosylase inhibitor. Cell Res. 27, 1289–1292 (2017).
Cao, A. & Galanello, R. Beta-thalassemia. Genet. Med. 12, 61–76 (2010).
Liang, P. et al. Correction of β-thalassemia mutant by base editor in human embryos. Protein Cell 8, 811–822 (2017).
Eng, B. et al. Three new beta-globin gene promoter mutations identified through newborn screening. Hemoglobin 31, 129–134 (2007).
Li, Z. et al. A novel promoter mutation (HBB: c.-75G>T) was identified as a cause of β(+)-thalassemia. Hemoglobin 39, 115–120 (2015).
Ran, F.A. et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186–191 (2015).
Hu, J.H. et al. Evolved Cas9 variants with broad PAM compatibility and high DNA specificity. Nature 556, 57–63 (2018).
Rathore, A. et al. The local dinucleotide preference of APOBEC3G can be altered from 5′-CC to 5′-TC by a single amino acid substitution. J. Mol. Biol. 425, 4442–4454 (2013).
Pinello, L. et al. Analyzing CRISPR genome-editing experiments with CRISPResso. Nat. Biotechnol. 34, 695–697 (2016).
Kulcsár, P.I. et al. Crossing enhanced and high fidelity SpCas9 nucleases to optimize specificity and cleavage. Genome Biol. 18, 190 (2017).
Kim, S., Bae, T., Hwang, J. & Kim, J.-S. Rescueof high-specificity Cas9 variants using sgRNAs with matched 5′ nucleotides. Genome Biol. 18, 218 (2017).
Ye, L. et al. Genome editing using CRISPR-Cas9 to create the HPFH genotype in HSPCs: An approach for treating sickle cell disease and β-thalassemia. Proc. Natl. Acad. Sci. USA 113, 10661–10665 (2016).
J.K.J. was supported by grants from the National Institutes of Health (R35 GM118158 and RM1 HG009490), by the Desmond and Ann Heathwood MGH Research Scholar Award, and by a St. Jude Children's Research Hospital Collaborative Research Consortium award. J.M.G. was supported by the National Science Foundation Graduate Research Fellowship Program. D.E.B. was supported by NHLBI (DP2OD022716, P01HL032262) and the St. Jude Children's Research Hospital Collaborative Research Consortium. We thank A. Sousa for advice on performing GUIDE-seq experiments, P. Cabeceiras for assistance producing lentivirus, and J. Angstman, V. Pattanayak, and A. Mattei for helpful discussions and comments.
J.M.G. is currently a full-time employee of and holds equity in Beam Therapeutics. J.K.J. has financial interests in Beam Therapeutics, Editas Medicine, Monitor Biotechnologies, Pairwise Plants, Poseida Therapeutics, and Transposagen Biopharmaceuticals. J.K.J. holds equity in Endcadia, Inc. J.K.J.'s interests were reviewed and are managed by Massachusetts General Hospital and Partners HealthCare in accordance with their conflict of interest policies. J.M.G. and J.K.J. are inventors on a patent application that has been filed for engineered sequence-specific deaminase domains in base editor architectures.
Integrated supplementary information
Supplementary Figure 1 Base editing activities of engineered A3A-BE3 variants with mutations designed to disrupt non-specific interactions with substrate ssDNA.
Graphs illustrating the frequencies of C to T editing by a series of A3A-BE3 variants containing various pairs of mutations in A3A on bystander and cognate cytidines at four endogenous human gene target sites in single replicate. The reference sequence of each target site is shown at the top of each graph.
Supplementary Figure 2 Comparison of product purities and indel mutation frequencies for BE3 and eA3A-BE3 on endogenous human gene target sites.
(a) Graph showing normalized frequencies of cognate Cs edited to A, T, or G for twelve endogenous human gene target sites from Fig. 1d when targeting with BE3 or A3A (N57G)-BE3. (b) Graph showing indel mutation frequencies for BE3, the YE BE3s, and different engineered A3A-BE3 variants at the same 12 sites shown in (a). All data shown represent the mean of three biological replicates with error bars representing SEMs.
Supplementary Figure 3 eA3A-BE3-mediated editing of bystander cytidines typically occurs on the same allele as editing of cognate cytidines.
Allele frequency tables for four sites on which eA3A-BE3 exhibits cognate-to-bystander editing ratios less than five. Arrows with red outlines indicate bystander cytidines in the editing window, while black arrows indicate the target cytidine. The predicted nicking site by nSpCas9 is indicated by black dashed line.
Supplementary Figure 4 Mutations designed to decrease the catalytic rate of eA3A can increase the cognate-to-bystander editing ratio of eA3A-BE3.
Heat maps showing C-to-T editing efficiencies for eA3A-BE3 and three eA3A-BE3 variants bearing the indicated mutations at three endogenous human gene target sites, each bearing a cognate cytidine preceded by a 5′ T and one or more bystander cytidines within the editing window. Editing efficiencies shown represent the mean of three biological replicates. Graph showing cognate-to-bystander editing ratios for all three target sites with eA3A-BE3 and each of the three eA3A-BE3 variants. Data points shown represent ratios of the means of the three biological replicates performed for each base editor. Mean values are indicated by a line and error bars represent the SEM.
Supplementary Figure 5 SpCas9 nuclease off-targets discovered by GUIDE-seq for the CTNNB1 and HBB -28 (A>G) gRNAs.
GUIDE-seq plots depicting the target site for both gRNAs (top of each figure panel) and off-target sites discovered by GUIDE-seq shown below with base positions containing mismatches to the on-target site highlighted with a colored box and RNA bulges indicated with a dash. Sites with more than one potential alignment (i.e. where either an RNA bulge or additional mismatches are both plausible target sites) are shown with brackets. GUIDE-seq read counts for each site discovered are shown in the right column. The on-target site for the CTNNB1 gRNA and the single mismatched on-target site for the HBB -28 (A>G) gRNA are indicated with a small black square to the left.
Supplementary Figure 6 On-target C to T editing efficiencies for each C in the editing window for sites in Figure 2a-c.
Bar plots showing the C-to-T editing efficiencies for each C in the editing windows of the on-target sites for the three gRNAs depicted in Figure 2a-c. The target sequence for each gRNA is depicted above each plot. Editing efficiencies shown represent the mean of n = 3 biologically independent samples. Center values represent the mean of n = 3 biologically independent samples and error bars represent SEM.
Supplementary Figure 7 The N57G mutation is important for decreased off-target activity of eA3A-BE3.
Editing frequencies for EMX1 site 1 and FANCF site 1 gRNAs with BE3, eA3A-BE3, and untransfected cells (from Figs. 2a – 2b) are re-plotted with editing frequencies with these same gRNAs observed with A3A-BE3 from a separate experiment. Sequences of the on- and off-target sites are shown to the left of the bar plot below the target sequence with mismatches relative to the on-target site highlighted with colored boxes and bulges with a grey highlighted dash. The on-target sites for the EMX1 site 1 and FANCF site 2 gRNAs are indicated with a small black square to the left. Editing efficiencies shown represent the mean of n = 3 biologically independent samples. Center values represent the mean of n = 3 biologically independent samples and error bars represent SEM.
Supplementary Figure 8 Effects of adding a second UGI domain and HF1 or Hypa high-fidelity mutations eA3A-BE3 on base editing product purities and indel frequencies at four endogenous human gene target sites.
Graph in top panel shows normalized frequencies of cognate Cs edited to A, T, or G for the four endogenous human gene target sites of Figures 2a – 2d when targeting each site with BE3, eA3A-BE3, or eA3A-BE3 variants incorporating HF1 or Hypa mutations and a second UGI domain. Graph in bottom panel shows indel frequencies induced by the same base editors at the same sites. All data shown represent the mean of three biological replicates and error bars represent SEMs.
Supplementary Figure 9 Efficient correction of the HBB -28 (A>G) pathogenic SNP in patient-derived erythroid precursor cells.
(a) Coomassie-stained SDS-PAGE gel showing titrations of purified eA3A-BE3 and A3A (N57Q)-BE3 protein next to a protein ladder. Molecular weights are indicated to the left of the image. Purification was performed for n = 3 biologically independent samples. (b) C-to-T, -G or -A editing frequencies at the -25 and -28 positions of the HBB -28 (A>G) allele in patient-derived erythroid precursor cells. Sequencing reads were bioinformatically separated according to whether they contained the -CTTT deletion in exon 2 that is present only in the allele without the HBB -28 (A>G) mutation. Editing efficiencies shown represent the mean of n = 3 biologically independent samples. Center values represent the mean of n = 3 biologically independent samples and error bars represent SEM. (c) RT-qPCR data examining HBB and HBG1/2 globin mRNA expression normalized to expression of HBA1/2 from the edited populations shown in panel (b) following terminal erythroid differentiation. Center values represent the mean of n = 3 biologically independent samples and error bars represent SEM. (d) Off-target editing by A3A (N57Q)-BE3 or eA3A-BE3 RNPs from the edited populations shown in panel (b) at six off-target sites. Percentage edits represent the sum of all C-to-D (D = A, G or T) editing frequencies in the editing window and represent the mean of n = 3 biologically independent replicates with error bars representing SEMs. Intended target sequence is shown at the top. On-target site is marked with a black diamond to the left and mismatches or bulges in the various off-target sites are shown with colored boxes or a dash in gray boxes, respectively. Off-target sites that lose the cognate TC motif within the editing window and thus might be expected to show lower off-target editing by eA3A, are noted with empty circles to the left. Asterisks indicate statistically significant differences in editing efficiencies observed between the indicated samples at each site (* p < 0.05, ** p < 0.005, *** p < 0.0005). All statistical testing was performed using two-tailed Student's t-test according to the method of Benjamini, Krieger, and Yekutieli without assuming equal variances between samples.
Supplementary Figure 10 eA3A-BE3 bearing VRQR or xCas9 mutations that alter PAM recognition specificity can be used to edit sites bearing NGA or NGT PAMs for highly precise single nucleotide editing.
Heat maps showing C-to-T editing efficiencies for VRQR BE3, eA3A-VRQR BE3, xCas9 BE3, and eA3A-xCas9 BE3 at 14 endogenous human gene target sites. VRQR target sites all bear NGAN PAM sequences, while xCas9 target sites bear NGN PAM sequences. Each site bears a cognate cytidine preceded by a 5′ T and one or more bystander cytidines within the editing window. Editing efficiencies shown represent the mean of three biological replicates. Graph shows cognate-to-bystander C to T editing ratios for VRQR BE3, eA3A-VRQR BE3, xCas9 BE3 and eA3A-xCas9 BE3. Data points shown represent ratios calculated from the mean values shown in the heat maps. Median values are indicated by a red line and error bars represent interquartile ranges. Values shown represent the mean of n = 3 biologically independent replicates.
Off-target editing frequencies of two gRNAs targeted to PPP1R12C VRQR site 1 or 3 with BE3 or eA3A-BE3. Percentage edits represent the sum of all edited Cs in the editing window and represent the mean of three biological replicates with error bars representing SEMs. Intended target sequence is shown at the top of each graph. Mismatches or bulges in the various off-target sites are shown with colored boxes or a dash in gray boxes, respectively. Off-target sites that lose the cognate TC motif within the editing window and thus might be expected to show lower off-target editing by eA3A, are noted with empty circles to the left. Asterisks indicate statistically significant differences in editing efficiencies observed between the indicated samples at each site (* p < 0.05, ** p < 0.005, *** p < 0.0005). All statistical testing was performed using two-tailed Student's t-test according to the method of Benjamini, Krieger, and Yekutieli without assuming equal variances between samples.
Supplementary Figure 12 Incorporating a self-cleaving hammerhead ribozyme on the 5′ end of the gRNA preserves perfect matching of the spacer and target site and rescues the activities of the eA3A-BE3 variants bearing HF1 and Hypa high-fidelity mutations.
Heat maps showing C-to-T editing efficiencies for eA3A variants incorporating HF1 or Hypa mutations targeting HBB -28 (A>G) using a gRNA with a 5′ mismatched guanine (top), a self-cleaving hammerhead ribozyme with 6 nucleotides of self-complementarity (middle), or a self-cleaving hammerhead ribozyme with 8 nucleotides of self-complementarity (bottom). The 5′ mismatched guanine or the ribozyme sequence is shown in red, while the spacer sequence is shown in black. Editing efficiencies shown represent the mean of three biological replicates.
Supplementary Figures 1–12 (PDF 3250 kb)
All standard deviation values derived from high-throughput sequencing reads of the 12 endogenous sites depicted in Figure 1d. (XLSX 117 kb)
Supplementary Tables 2–4 (XLSX 29 kb)
About this article
Cite this article
Gehrke, J., Cervantes, O., Clement, M. et al. An APOBEC3A-Cas9 base editor with minimized bystander and off-target activities. Nat Biotechnol 36, 977–982 (2018) doi:10.1038/nbt.4199
Cell Regeneration (2019)
Materials and Methods (2019)
Scientific Reports (2019)
Trends in Plant Science (2019)
Human Genetics (2019)