AbstractThe all-protein cytosine base editor DdCBE uses TALE proteins and a double-stranded DNA-specific cytidine deaminase (DddA) to mediate targeted C•G-to-T•A editing. To improve editing efficiency and overcome the strict TC sequence-context constraint of DddA, we used phage-assisted non-continuous and continuous evolution to evolve DddA variants with improved activity and expanded targeting scope. Compared to canonical DdCBEs, base editors with evolved DddA6 improved mitochondrial DNA (mtDNA) editing efficiencies at TC by 3.3-fold on average. DdCBEs containing evolved DddA11 offered a broadened HC (H = A, C or T) sequence compatibility for both mitochondrial and nuclear base editing, increasing average editing efficiencies at AC and CC targets from less than 10% for canonical DdCBE to 15–30% and up to 50% in cell populations sorted to express both halves of DdCBE. We used these evolved DdCBEs to efficiently install disease-associated mtDNA mutations in human cells at non-TC target sites. DddA6 and DddA11 substantially increase the effectiveness and applicability of all-protein base editing.
MainEach human cell can contain several hundred copies of circular mtDNA that encodes RNAs and proteins that mediate ATP production1,2,3. Owing to the essential role of the mitochondria in energy homeostasis, single-nucleotide mutations in the mtDNA can contribute to developmental disorders, neuromuscular disease, cancer progression and a growing number of other human diseases4,5,6,7. Technologies that enable the precise installation of point mutations within mtDNA could reveal the role of these mutations in pathogenesis and provide ways to correct them for potential therapeutic applications.
Programmable nucleases can make targeted double-strand breaks within mtDNA copies that contain specific mutations, resulting in the elimination of those copies8,9,10,11,12. Nucleases, however, cannot introduce specified sequence changes. Genome editing agents, including base editors13,14 and prime editors15, directly install precise changes in a target DNA sequence but typically rely on a guide RNA sequence to direct CRISPR–Cas proteins for binding to its target DNA. Owing to the challenge of importing guide RNAs into the mitochondria, CRISPR-based systems have, thus far, not been used reliably for mtDNA engineering16,17.
To begin to address this challenge, we recently developed DdCBE to enable targeted C•G-to-T•A conversions within mtDNA18. DdCBE uses two TALE proteins to specify the double-stranded DNA (dsDNA) region for editing. Each TALE is fused to a non-toxic half of DddA and one copy of uracil glycosylase inhibitor (UGI) protein. Binding of two TALE fused split-DddA–UGI fusions to adjacent sites promotes reassembly of functional DddA for deamination of target cytosines within the dsDNA spacing region. DdCBEs have been applied for mitochondrial base editing in human embryo, mice, zebrafish and plants19,20,21,22,23,24.
In our initial studies, we observed a range of mtDNA editing efficiencies (4.6–49%) depending on the position of the target C within the spacing region between the DNA-bound DdCBE halves18. We hypothesized that enhancing the activity of split DddA could increase mtDNA editing efficiencies at putative 5′-TC contexts by improving the compatibility of DddA with different TALE designs and deaminase orientations.
Given the strict sequence preference of DddA, our initial DdCBE is limited predominantly to TC targets. In this study, we sought to increase DdCBE activity at both TC and non-TC targets by applying rapid phage-assisted continuous evolution (PACE) and related phage-assisted non-continuous evolution (PANCE) methods25,26. Development of a selection circuit for DdCBE activity followed by PANCE and PACE resulted in several DddA variants with conserved mutations enriched during evolution. Evolved variants DddA6 and DddA11 mediated ~4.3-fold average improvement in mtDNA base editing efficiency at TC targets compared to wild-type DddA. Notably, DddA11 increased average bulk editing levels at AC and CC targets in the mtDNA and nucleus from less than 10% with canonical DdCBE to ~15–30%. These variants collectively enable the installation or correction of C•G-to-T•A point mutations at both TC and non-TC targets, substantially expanding the overall utility of all-protein DdCBEs.
ResultsAdapting BE-PACE to evolve TALE-based DdCBEsPACE uses an M13 phage that is modified to contain an evolving gene in place of gene III (gIII)27. gIII encodes a capsid protein pIII that is essential for producing infectious phage progeny. To establish a selection circuit, gIII is encoded in an accessory plasmid (AP) within the Escherichia coli host cell such that gIII expression is dependent on the evolving activity. We previously reported a BE-PACE system to evolve CRISPR cytosine base editors25. In this system, the AP encodes gIII under the control of a T7 promoter. A complementary plasmid (CP) encodes T7 RNA polymerase (T7 RNAP) fused to a degron through a 2-amino-acid linker (Fig. 1a). In the absence of C•G-to-T•A editing of the linker sequence, the degron triggers constitutive proteolysis of T7 RNAP, preventing gIII expression (Fig. 1b). The target cytosines for DdCBE-mediated editing in this selection are C6 and C7, where the subscripted numbers refer to their positions in the spacing region, counting the DNA nucleotide immediately after the binding site of the left-side TALE (TALE3) as position 1 (Fig. 1c). Successful C•G-to-T•A editing of either or both C6 and C7 targets introduces a stop codon within the linker to prevent translation of the degron tag. Active T7 RNAP then initiates gIII expression (Fig. 1b). The nucleotide at position 8 may be modified to A, T, C or G to enable selection against TC and non-TC contexts (Fig. 1b).
Fig. 1: Phage-assisted evolution of DddA-derived cytosine base editor for improved activity and expanded targeting scope.a, Selection to evolve DdCBE using PANCE and PACE. An AP (purple) contains gIII driven by the T7 promoter. The CP (orange) expresses a T7 RNAP–degron fusion. The evolving T7-DdCBE containing DddA split at G1397 is encoded in the SP (blue). Where relevant, the promoters are indicated. b, A 2-amino-acid linker connects T7 RNAP to the degron. The linker sequence contains cytidines C6 and C7 that are targets for DdCBE editing. The nucleotide at position 8 can be varied to T, A, C or G to form plasmids CP-TCC, CP-ACC, CP-CCC and CP-GCC, respectively. In the absence of target C-to-T editing, expression of degron (brown) results in proteolysis of T7 RNAP (orange) and inhibition of gIII expression. Active T7-DdCBE edits one or both target cytidines to install a stop codon (*) within the linker, thus restoring active T7 RNAP to mediate gIII expression. c, Architecture of T7-DdCBE and the 15-bp target spacing region. Nucleotides corresponding to DNA sequences within T7 RNAP, linker and degron genes are colored in orange, gray and brown, respectively.
To enhance phage propagation in the selection circuit, we hypothesized that a DdCBE architecture with maximal editing efficiency would provide a favorable starting point to evolve activity against TC and non-TC targets. We designed a DdCBE that consisted of a left-side TALE (TALE3) and a right-side TALE (TALE4) flanking a 15-base pair (bp) spacing region, with targets C6 and C7 within the transcription template strand (Fig. 1c). We fused one copy of UGI to the N-terminus of the TALE protein and split DddA at G1397 to maximize editing of cytosine targets in the transcription template strand18. The resulting UGI–TALE3–DddA-G1397-N and UGI–TALE4–DddA-G1397-C fusions, which we refer to hereafter as T7-DdCBE, were encoded in the selection phage (SP) to co-evolve both halves of DdCBE (Fig. 1a). The phage genome is continuously mutagenized by an arabinose-inducible mutagenesis plasmid (MP6)28 (Fig. 1a).
To modulate selection stringency, we generated host strains 1–4. Each host strain contained combinations of AP and CP with different ribosome-binding-site strengths, such that strain 1 resulted in the lowest selection stringency and strain 4 provided the highest stringency. All tested CPs encoded the TCC linker sequence (Extended Data Fig. 1a). We then tested overnight SP propagation in these host strains. At the highest stringency, we observed ~100-fold overnight phage propagation of an SP containing an active T7-DdCBE, consistent with DdCBE’s ability to edit 5′-TC targets. Notably, phage containing an inactivating E1347A DddA mutation (dead T7-DdCBE phage) did not propagate (Extended Data Fig. 1b). These results establish the dependence of phage propagation on DdCBE activity and that BE-PACE can be successfully adapted to select TALE-based DdCBEs.
Phage-assisted evolution of DdCBE toward higher editing efficiency at 5′-TC
We reasoned that beginning evolution with PANCE may be useful to increase activity and phage propagation before moving into PACE26. PANCE is less stringent because fresh host cells are manually infected with SP from a preceding passage, so no phage is lost to continuous dilution.
To evolve DdCBEs for higher activity at TC targets, we initiated PANCE of canonical T7-DdCBE by infecting SP into high-stringency strain 4 transformed with MP6 (Extended Data Fig. 1a). After seven passages, phage populations from all four replicates propagated approximately 10,000-fold overnight (Extended Data Fig. 1c). Isolated clonal phages from two or more independent replicates were enriched for the mutations T1372I, M1379I and T1380I within the DddA gene (Supplementary Table 1).
To validate the editing activity associated with these DddA genotypes, we incorporated each mutation into our previously published G1397 split DdCBEs that targeted human MT-ATP8, MT-ND4 and MT-ND5 (Supplementary Note 1)18. We plasmid-transfected HEK293T cells with canonical versions of ATP8-DdCBE, ND4-DdCBE or ND5.2-DdCBE and compared their editing efficiencies to those produced from the corresponding mutant DdCBEs. Although T1372I and M1379I impaired editing, T1380I increased C•G-to-T•A conversions by an average of 1.2-fold to 2.0-fold across the three mtDNA sites (Extended Data Fig. 1d). It is possible that the benefit of T1372I and M1379I may require additional mutations evolved during PANCE but not tested in mammalian cells. These results indicate that PANCE of canonical T7-DdCBE was able to yield a DddA variant that modestly improved TC editing. We refer to the DddA (T1380I) mutant as DddA1 (Fig. 2a).
Fig. 2: Evolved DddA variants improve mitochondrial base editing activity at 5′-TC.a, Mutations within the DddA gene of T7-DdCBE. Variants were isolated after evolution of canonical T7-DdCBE using PANCE and PACE in strain 4 transformed with MP6 (Extended Data Fig. 1a). DddA6 was rationally designed by incorporating the T1413I mutation into DddA5. b, Crystal structure of DddA (gray, PDB 6U08) complexed with DddI immunity protein (not shown). Positions of mutations enriched after PANCE and PACE are colored in orange. The catalytic residue E1347 is shown. DddA was split at G1397 (red) to generate T7-DdCBE. c, d, mtDNA editing efficiencies and indel frequencies of HEK293T cells treated with ND5.2-DdCBE (c) or ATP8-DdCBE (d). The genotypes of DddA variants correspond to a. For each base editor, the DNA spacing region, target cytosines and DddA split orientation are shown. e, Frequencies of MT-ND5 alleles produced by DddA6 in c. f, Frequencies of MT-ATP8 alleles produced by DddA6 in d. For e and f, tables are representative of n = 3 independent biological replicates. For c–f, values and errors reflect the mean ± s.d. of n = 3 independent biological replicates.
To further increase selection stringency, we conducted PACE using an SP encoding the DddA1 variant of T7-DdCBE (T7-DdCBE-DddA1). After 140 hours of continuous propagation at a flow rate of 1.5–3.0 lagoon volumes per hour, distinct mutations enriched across the four replicates, with the starting T1380I mutation maintained in all lagoons (Supplementary Table 2). We selected the most enriched genotype in each of the four replicates (DddA2, DddA3, DddA4 and DddA5) and tested their mtDNA editing efficiencies (Fig. 2a,b). DddA2, DddA3, DddA4 and DddA5 improved average TC editing efficiencies from 7.6 ± 2.4% with starting DddA to 14 ± 5.8%, 22 ± 6.1%, 21 ± 7.9% and 24 ± 4.4%, respectively, within MT-ND5 and MT-ATP8 (Fig. 2c,d).
The T1413I mutation in DddA4, which is in the C-terminal half of split DddA, improved base editing efficiency of DddA4 by an average of 1.6-fold compared to DddA1. Given that T1413I is positioned along the interface between the two split DddA halves (Fig. 2b), we hypothesized that this mutation might promote the reconstitution of split DddA halves. Incorporating T1413I into DddA5 to form DddA6 (Q1310R + S1330I + T1380I + T1413I) resulted in a modest editing efficiency improvement to 26 ± 3.7%, a 3.4-fold average improvement in TC editing activity compared to wild-type DddA (Fig. 2c d). Close to 90% of the edited alleles produced from DddA6 contained a TCC-to-TTT conversion, suggesting that consecutive cytosines are likely targets for processive base editing (Fig. 2e,f). These results establish DddA6 as a dsDNA cytidine deaminase variant with enhanced editing activity at TC sequences.
We evolved DddA6 from DddA split at G1397. To check if DddA6 is compatible with the G1333 split, we tested DddA6 at three mtDNA sites using DdCBEs split at G1333 and observed a 1.3-fold to 3.6-fold improvement in editing efficiencies compared to wild-type DddA (Extended Data Fig. 2a–c). These data indicate that mutations in DddA6 can enhance mtDNA editing efficiencies of the G1333 split variant, but the extent of improvement is lower than with the G1397 split. We noted that editing improvements mediated by DddA6 were modest at sites that exhibit efficient editing even with wild-type DddA, such as MT-ND1 and MT-ND4 (Extended Data Fig. 2a,d). For sites already efficiently edited with canonical DdCBEs, other deaminase-independent factors, such as mtDNA repair, could limit editing efficiency more than deaminase activity.
Evolving DddA variants with expanded sequence context compatibilityTo assess if the enhanced activity of DddA6 would enable base editing at target cytosines not in the native TC sequence context, bacteria expressing the evolved T7-DdCBE were transformed with a plasmid library encoding NC7N targets, where N = A, T, C or G. After overnight incubation, the plasmid library was isolated and subjected to high-throughput sequencing to measure the C•G-to-T•A conversion at each of the 16 NC7N targets (Fig. 3a).
Fig. 3: Evolved DddA variants show enhanced editing at TC and non-TC target sequences in mtDNA.a, Bacterial plasmid assay to profile sequence preferences of evolved DddA variants. T7-DdCBE edits the NC7N sequence of the target plasmid library. b, Heat map showing C•G-to-T•A editing efficiencies of NC7N sequence in each target plasmid, including the second cytosine in NCC6 sequences. Genotypes of listed variants correspond to Figs. 2a and 3c. Mock-treated cells did not express T7-DdCBE and contained only the library of target plasmids. Shading levels reflect the mean of n = 3 independent biological replicates. c, Genotypes of DddA variants after evolving T7-DdCBE-DddA1 using context-specific PANCE and PACE. Mutations enriched for activity on a CCC linker or GCC linker are highlighted in red and blue, respectively. d, e, Mitochondrial C•G-to-T•A editing efficiencies of HEK293T cells treated with canonical and evolved variants of ND5.2-DdCBE (d) or ATP8-DdCBE (e). Target spacing regions and split DddA orientations are shown for each base editor. Cytosines highlighted in light purple and dark purple are in non-TC contexts. f, Mitochondrial base editing efficiencies of reversion mutants from ATP8-DdCBE-DddA11 (labeled as 11) in HEK293T cells. Reversion mutants are designated 11a–11h. Amino acids that differ from those in canonical ATP8-DdCBE are indicated, so the absence of an amino acid indicates a reversion to the corresponding canonical amino acid in the first column. g, Average percentage of genome-wide C•G-to-T•A off-target editing in mtDNA for indicated DdCBE and controls in HEK293T cells. For d–g, values and error bars reflect the mean ± s.d. of n = 3 independent biological replicates.
Consistent with earlier human mtDNA editing results, DddA6 improved the average editing efficiencies of bacterial plasmids containing TC7N substrates by approximately 1.3-fold. DddA6-mediated editing at non-TC sequences, however, remained negligible ( A mutation in a CCA context. Both mutations were previously implicated in renal oncocytoma37 (Fig. 6a).
Fig. 6: Application of DddA11 variant to install pathogenic mutations at non-TC targets in HEK293T cells.a, Use of DdCBEs to install disease-associated target mutations in human mtDNA. (V, valine; I, isoleucine; A, alanine; T, threonine; Q, glutamine; ∗, stop). b–d, Mitochondrial base editing efficiencies of HEK293T cells treated with canonical or evolved ND4.3-DdCBE (b), ND4.2-DdCBE (c) and ND5.4-DdCBE (d). On-target cytosines are colored green, blue or red, respectively. Cells expressing the DddA11 variant of DdCBE were isolated by FACS for high-throughput sequencing. The split orientation, target spacing region and corresponding encoded amino acids are shown. Nucleotide sequences boxed in dotted lines are part of the TALE binding site. e, f, Oxygen consumption rate (OCR) (e) and relative values of respiratory parameters (f) in sorted HEK293T cells treated with the DddA11 variant of ND4.2-DdCBE or ND5.4-DdCBE. For b–f, values and error bars reflect the mean ± s.d. of n = 3 independent biological replicates, except that ND4.2-DdCBE in e and f reflects n = 2 independent biological replicates. *P 30 at a given position were taken into account when calling SNPs at that particular position. For Extended Data Fig. 6a–f, all SNPs that were called in untreated samples were excluded from the analyses of treated samples. Each SNP was called in treated samples if it appeared in at least one biological replicate, and the average frequency was calculated by taking the average of all replicate(s) in which the SNP was present.
Calculation of average off-target C•G-to-T•A editing frequencyAnalysis was performed as previously described18,43. To calculate the mitochondrial genome-wide average off-target editing frequency for each DdCBE in Fig. 3g, REDItools was used (version 1.2.1)44. All nucleobases except cytosines and guanines were removed, and the number of reads covering each C•G base pair with a PHRED quality score greater than 30 (Q > 30) was calculated. The on-target C•G base pairs (depending on the DdCBE used in each treatment) were excluded to consider only off-target effects. C•G-to-T•A SNVs present at high frequencies (>50%) in both treated and untreated samples (that, therefore, did not arise from DdCBE treatment) were also excluded. The average off-target editing frequency was then calculated independently for each biological replicate of each treatment condition as: (number of reads in which a given C•G base pair was called as a T•A base pair, summed over all non-target C•G base pairs) / (total number of reads that covered all non-target C•G base pair).
Oxygen consumption rate analyses by Seahorse XF analyzerA Seahorse plate was coated with 0.01% (w/v) poly-ʟ-lysine (Sigma-Aldrich). Next, 1.6 × 104 cells were seeded on the coated Seahorse plate 16 hours before the analysis in the Seahorse XFe96 Analyzer (Agilent). Analysis was performed in the Seahorse XF DMEM Medium pH 7.4 (Agilent) supplemented with 10 mM glucose (Agilent), 2 mM L-glutamine (Gibco) and 1 mM sodium pyruvate (Gibco). Mito stress protocol was applied with the use of 1.5 µM oligomycin, 1 µM FCCP and 1 µM piericidin + 1 µM antimycin.
Reporting SummaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article.
Data availabilityHigh-throughput sequencing and whole mitochondria sequencing data have been deposited in the National Center of Biotechnology Information Sequence Read Archive under accession code PRJNA753136. SNP identification for whole mitochondrial genome sequencing used the NC_012920 genome as a reference. Plasmids are available from Addgene. Amino acid sequences of all base editors in this study are provided in the Supplementary Information as Supplementary Sequences 1–3. TALE sequences for SIRT6-DdCBE and JAK2-DdCBE are from Addgene plasmids TAL2406, TAL2407, TAL2454 and TAL2455.
ReferencesBruser, C., Keller-Findeisen, J. & Jakobs, S. The TFAM-to-mtDNA ratio defines inner-cellular nucleoid populations with distinct activity levels. Cell Rep. 37, 110000 (2021).
PubMed
Google Scholar
Kukat, C. et al. Super-resolution microscopy reveals that mammalian mitochondrial nucleoids have a uniform size and frequently contain a single copy of mtDNA. Proc. Natl Acad. Sci. USA 108, 13534–13539 (2011).
CAS
PubMed
PubMed Central
Google Scholar
Robin, E. D. & Wong, R. Mitochondrial DNA molecules and virtual number of mitochondria per cell in mammalian cells. J. Cell. Physiol. 136, 507–513 (1988).
CAS
PubMed
Google Scholar
Yuan, Y. et al. Comprehensive molecular characterization of mitochondrial genomes in human cancers. Nat. Genet. 52, 342–352 (2020).
CAS
PubMed
PubMed Central
Google Scholar
Schapira, A. H. Mitochondrial diseases. Lancet 379, 1825–1834 (2012).
CAS
PubMed
Google Scholar
Vafai, S. B. & Mootha, V. K. Mitochondrial disorders as windows into an ancient organelle. Nature 491, 374–383 (2012).
CAS
PubMed
Google Scholar
Stewart, J. B. & Chinnery, P. F. Extreme heterogeneity of human mitochondrial DNA from organelles to populations. Nat. Rev. Genet. 22, 106–118 (2021).
CAS
PubMed
Google Scholar
Bacman, S. R., Williams, S. L., Pinto, M., Peralta, S. & Moraes, C. T. Specific elimination of mutant mitochondrial genomes in patient-derived cells by mitoTALENs. Nat. Med. 19, 1111–1113 (2013).
CAS
PubMed
PubMed Central
Google Scholar
Gammage, P. A., Rorbach, J., Vincent, A. I., Rebar, E. J. & Minczuk, M. Mitochondrially targeted ZFNs for selective degradation of pathogenic mitochondrial genomes bearing large-scale deletions or point mutations. EMBO Mol. Med. 6, 458–466 (2014).
CAS
PubMed
PubMed Central
Google Scholar
Peeva, V. et al. Linear mitochondrial DNA is rapidly degraded by components of the replication machinery. Nat. Commun. 9, 1727 (2018).
PubMed
PubMed Central
Google Scholar
Nissanka, N., Bacman, S. R., Plastini, M. J. & Moraes, C. T. The mitochondrial DNA polymerase gamma degrades linear DNA fragments precluding the formation of deletions. Nat. Commun. 9, 2491 (2018).
PubMed
PubMed Central
Google Scholar
Jackson, C. B., Turnbull, D. M., Minczuk, M. & Gammage, P. A. Therapeutic manipulation of mtDNA heteroplasmy: a shifting perspective. Trends Mol. Med. 26, 698–709 (2020).
CAS
PubMed
Google Scholar
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).
PubMed Central
Google Scholar
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).
PubMed
PubMed Central
Google Scholar
Anzalone, A. V. et al. Search-and-replace genome editing without double-strand breaks or donor DNA. Nature 576, 149–157 (2019).
CAS
PubMed
PubMed Central
Google Scholar
Gammage, P. A., Moraes, C. T. & Minczuk, M. Mitochondrial genome engineering: the revolution may not be CRISPR-ized. Trends Genet. 34, 101–110 (2018).
CAS
PubMed
PubMed Central
Google Scholar
Silva-Pinheiro, P. & Minczuk, M. The potential of mitochondrial genome engineering. Nat. Rev. Genet. https://doi.org/10.1038/s41576-021-00432-x (2021).
Mok, B. Y. et al. A bacterial cytidine deaminase toxin enables CRISPR-free mitochondrial base editing. Nature 583, 631–637 (2020).
CAS
PubMed
PubMed Central
Google Scholar
Lee, H. et al. Mitochondrial DNA editing in mice with DddA-TALE fusion deaminases. Nat. Commun. 12, 1190 (2021).
CAS
PubMed
PubMed Central
Google Scholar
Kang, B. C. et al. Chloroplast and mitochondrial DNA editing in plants. Nat Plants 7, 899–905 (2021).
CAS
PubMed
PubMed Central
Google Scholar
Guo, J. et al. Precision modeling of mitochondrial diseases in zebrafish via DdCBE-mediated mtDNA base editing. Cell Discov. 7, 78 (2021).
PubMed
PubMed Central
Google Scholar
Guo, J. et al. DdCBE mediates efficient and inheritable modifications in mouse mitochondrial genome. Mol. Ther. Nucleic Acids 27, 73–80 (2022).
CAS
PubMed
Google Scholar
Wei, Y. et al. Human cleaving embryos enable efficient mitochondrial base-editing with DdCBE. Cell Discov. 8, 7 (2022).
CAS
PubMed
PubMed Central
Google Scholar
Sabharwal, A. et al. The FusX TALE Base Editor (FusXTBE) for rapid mitochondrial DNA programming of human cells in vitro and zebrafish disease models in vivo. CRISPR J. 4, 799–821 (2021).
CAS
PubMed
PubMed Central
Google Scholar
Thuronyi, B. W. et al. Continuous evolution of base editors with expanded target compatibility and improved activity. Nat. Biotechnol. 37, 1070–1079 (2019).
CAS
PubMed
PubMed Central
Google Scholar
Roth, T. B., Woolston, B. M., Stephanopoulos, G. & Liu, D. R. Phage-assisted evolution of Bacillus methanolicus methanol dehydrogenase 2. ACS Synth. Biol. 8, 796–806 (2019).
CAS
PubMed
PubMed Central
Google Scholar
Esvelt, K. M., Carlson, J. C. & Liu, D. R. A system for the continuous directed evolution of biomolecules. Nature 472, 499–503 (2011).
CAS
PubMed
PubMed Central
Google Scholar
Badran, A. H. & Liu, D. R. Development of potent in vivo mutagenesis plasmids with broad mutational spectra. Nat. Commun. 6, 8425 (2015).
CAS
PubMed
Google Scholar
Carlson, J. C., Badran, A. H., Guggiana-Nilo, D. A. & Liu, D. R. Negative selection and stringency modulation in phage-assisted continuous evolution. Nat. Chem. Biol. 10, 216–222 (2014).
CAS
PubMed
PubMed Central
Google Scholar
Szymczak, A. L. et al. Correction of multi-gene deficiency in vivo using a single ‘self-cleaving’ 2A peptide–based retroviral vector. Nat. Biotechnol. 22, 589–594 (2004).
CAS
PubMed
Google Scholar
Cuculis, L., Abil, Z., Zhao, H. & Schroeder, C. M. Direct observation of TALE protein dynamics reveals a two-state search mechanism. Nat. Commun. 6, 7277 (2015).
CAS
PubMed
Google Scholar
Streubel, J., Blücher, C., Landgraf, A. & Boch, J. TAL effector RVD specificities and efficiencies. Nat. Biotechnol. 30, 593–595 (2012).
CAS
PubMed
Google Scholar
Jain, S. et al. TALEN outperforms Cas9 in editing heterochromatin target sites. Nat. Commun. 12, 606 (2021).
CAS
PubMed
PubMed Central
Google Scholar
Reyon, D. et al. FLASH assembly of TALENs for high-throughput genome editing. Nat. Biotechnol. 30, 460–465 (2012).
CAS
PubMed
PubMed Central
Google Scholar
Fine, E. J., Cradick, T. J., Zhao, C. L., Lin, Y. & Bao, G. An online bioinformatics tool predicts zinc finger and TALE nuclease off-target cleavage. Nucleic Acids Res. 42, e42 (2013).
PubMed
PubMed Central
Google Scholar
Liao, Z. et al. The ND4 G11696A mutation may influence the phenotypic manifestation of the deafness-associated 12S rRNA A1555G mutation in a four-generation Chinese family. Biochem. Biophys. Res. Commun. 362, 670–676 (2007).
CAS
PubMed
PubMed Central
Google Scholar
Gopal, R. K. et al. Early loss of mitochondrial complex I and rewiring of glutathione metabolism in renal oncocytoma. Proc. Natl Acad. Sci. USA 115, E6283–E6290 (2018).
PubMed
PubMed Central
Google Scholar
Gehrke, J. M. et al. An APOBEC3A-Cas9 base editor with minimized bystander and off-target activities. Nat. Biotechnol. 36, 977–982 (2018).
CAS
PubMed
PubMed Central
Google Scholar
Hubbard, B. P. et al. Continuous directed evolution of DNA-binding proteins to improve TALEN specificity. Nat. Methods 12, 939–942 (2015).
CAS
PubMed
PubMed Central
Google Scholar
Badran, A. H. et al. Continuous evolution of Bacillus thuringiensis toxins overcomes insect resistance. Nature 533, 58–63 (2016).
CAS
PubMed
PubMed Central
Google Scholar
Clement, K. et al. CRISPResso2 provides accurate and rapid genome editing sequence analysis. Nat. Biotechnol. 37, 224–226 (2019).
CAS
PubMed
PubMed Central
Google Scholar
Shen, W., Le, S., Li, Y. & Hu, F. SeqKit: a cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PLoS ONE 11, e0163962 (2016).
PubMed
PubMed Central
Google Scholar
Doman, J. L., Raguram, A., Newby, G. A. & Liu, D. R. Evaluation and minimization of Cas9-independent off-target DNA editing by cytosine base editors. Nat. Biotechnol. 38, 620–628 (2020).
CAS
PubMed
PubMed Central
Google Scholar
Diroma, M. A., Ciaccia, L., Pesole, G. & Picardi, E. Elucidating the editome: bioinformatics approaches for RNA editing detection. Brief. Bioinform. 20, 436–447 (2019).
CAS
PubMed
Google Scholar
Download references
AcknowledgementsThis work was supported by the Merkin Institute of Transformative Technologies in Healthcare; US National Institutes of Health (NIH) grants RM1HG009490 (D.R.L), R01EB027793 (D.R.L), R01EB031172 (D.R.L), R35GM118062 (D.R.L), R35GM122455 (V.K.M) and U01AI142756 (D.R.L); and the Howard Hughes Medical Institute. We thank K. Zhao, J. Doman, M. Negeubauer, X. D. Gao, P. Chen, M. de Moraes and J. Mougous for materials, discussions and technical advice. B.Y.M. was supported by the Singapore A*STAR NSS fellowship. A.V.K. is supported by the Jane Coffin Childs Memorial Fund for Medical Research fellowship. A.R. is supported by NIH T32 GM095450 and a National Science Foundation graduate research fellowship. V.K.M. and D.R.L. are supported by the Howard Hughes Medical Institute.
Author informationAffiliationsMerkin Institute of Transformative Technologies in Healthcare, Broad Institute of MIT and Harvard, Cambridge, MA, USA
Beverly Y. Mok, Aditya Raguram, Tony P. Huang & David R. Liu
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA, USA
Beverly Y. Mok, Aditya Raguram, Tony P. Huang & David R. Liu
Howard Hughes Medical Institute, Harvard University, Cambridge, MA, USA
Beverly Y. Mok, Aditya Raguram, Tony P. Huang & David R. Liu
Howard Hughes Medical Institute and Department of Molecular Biology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
Anna V. Kotrys & Vamsi K. Mootha
Broad Institute of MIT and Harvard, Cambridge, MA, USA
Anna V. Kotrys & Vamsi K. Mootha
ContributionsB.Y.M. designed, performed and analyzed evolution experiments; characterized DddA variants in bacteria and mammalian cells; and wrote the manuscript. A.V.K. designed, performed and analyzed mitochondrial biology experiments, with supervision from V.K.M., and wrote the manuscript. A.R. designed the selection circuit, assisted with flow cytometry and performed off-target analyses. T.P.H. designed the sequence context profiling assay in bacteria. D.R.L. supervised the research and wrote the manuscript.
Corresponding authorCorrespondence to
David R. Liu.
Ethics declarations
Competing interests
The authors declare competing financial interests: B.Y.M., A.R. and D.R.L have filed patent applications on this work. D.R.L. is a consultant for Prime Medicine, Beam Therapeutics, Pairwise Plants, Chroma Medicine, and Resonance Medicine, which are companies that use genome editing, genome engineering, or PACE, and owns equity in these companies. V.K.M. is a consultant to 5am Ventures and Janssen Pharmaceuticals. Direct correspondence to drliu@fas.harvard.edu.
Peer review
Peer review information
Nature Biotechnology thanks the anonymous reviewers for their contribution to the peer review of this work.
Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended dataExtended Data Fig. 1 Evolution of canonical T7-DdCBE for improved TC activity using PANCE.a, Strains for screening selection stringency. Strains were generated by transformation with a variant of an AP and a variant of a CP. All CPs encode a TCC linker. Relative RBS strengths of SD8, sd8, sd2 and sd4U are 1.0, 0.20, 0.010 and 0.00040, respectively. b, Overnight phage propagation of indicated SPs in host strains with increasing stringencies. Dead T7-DdCBE phage contained the catalytically inactivating E1347A mutation in DddA. The fold phage propagation is the output phage titer divided by the input titer. c, Phage passage schedule for canonical T7-DdCBE evolution in PANCE using strain 4 transformed with MP6. Table indicates the dilution factor for the input phage population. Output phage titers for each replicate (A, B, C and D) are shown for each passage. Average fold propagation was obtained by averaging the fold propagation obtained from the four replicates A-D. d, Mitochondrial base editing efficiencies of HEK293T cells treated with canonical DdCBE or with DdCBEs containing the indicated mutations within DddA. For each base editor, the DddA split orientation and target cytosine (purple) within the spacing region is indicated. For b and d, values and error bars reflect the mean±s.d of n = 3 independent biological replicates.
Extended Data Fig. 2 DddA6 is compatible with split-G1333 and split-G1397 DdCBE orientations.a-d, Mitochondrial base editing efficiencies of HEK293T cells treated with (a) ND1.1-DdCBE, (b) ND1.2-DdCBE, (c) ND2-DdCBE and (d) ND4-DdCBE. Target spacing regions and split DddA orientations are shown above each plot. Values and error bars reflect the mean±s.d of n = 3 independent biological replicates.
Extended Data Fig. 3 Evolution of DddA1-containing T7-DdCBE for expanded targeting scope using PANCE.a, Strains for overnight phage propagation assays on non-TC linker substrates. b, Overnight fold propagation of indicated SP in host strains encoding TC or non-TC linkers. Strains correspond to Extended Data Fig. 3a. T7-DdCBE-DddA1 phage harbors a T1380I mutation in DddA. Dead T7-DdCBE-DddA1 phage contains an additional catalytically inactivating E1347A mutation in DddA. Values and error bars reflect the mean±s.d of n = 3 independent biological replicates. c-e, Phage passage schedule for T7-DdCBE-DddA1 evolution in PANCE using (c) strain 5 transformed with MP6, (d) strain 6 transformed with MP6 or (e) strain 7 transformed with MP6. Tables indicate the dilution factor for the input phage population (see Methods for drifting procedure) For a given linker target, the output phage titers for each replicate (A, B, C and D) are shown for each passage. Average fold propagations above the dotted line in each graph represent propagation >1-fold. Average fold propagation was obtained by averaging the fold propagation obtained from the four replicates.
Extended Data Fig. 4 Allele compositions from mitochondrial and nuclear editing by DddA11-containing DdCBEs.a, Frequencies of mitochondrial MT-ND5 alleles produced by DddA11 variant of ND5.2-DdCBE. b, Frequencies of mitochondrial MT-ATP8 alleles produced by DddA11 variant of ATP8-DdCBE. c, Frequencies of nuclear SIRT6 alleles produced by DddA11 variant of SIRT6-DdCBE. d, Frequencies of nuclear JAK2 alleles produced by DddA11 variant of JAK2-DdCBE. Each table is representative of n = 3 independent biological replicates. Values and errors reflect the mean±s.d of n = 3 independent biological replicates.
Extended Data Fig. 5 Evolved DddA variants mediate mitochondrial base editing in multiple human cell lines.a-c, Mitochondrial DNA editing efficiencies of canonical and evolved ND5.2-DdCBE in (a) HeLa cells, (b) K562 cells, and (c) U2OS cells. Editing efficiencies were measured for unsorted cells (bulk) and isolated DdCBE-expressing cells (sorted). Target spacing region and split DddA orientation are shown. Cytosines highlighted in light purple and dark purple are in non-TC contexts. Values and error bars reflect the mean±s.d of n = 3 independent biological replicates.
Extended Data Fig. 6 Mitochondrial genome-wide off-target C•G-to-T•A mutations.Average frequency and mitochondrial genome position of each unique C•G-to-T•A single nucleotide variant (SNV) is shown for HEK293T cells treated with (a) canonical ATP8-DdCBE, (b) ATP8-DdCBE containing DddA6, (c) ATP8-DdCBE containing DddA11, (d) canonical ND5.2-DdCBE, (e) ND5.2-DdCBE containing DddA6 and (f) ND5.2-DdCBE containing DddA11. g, Ratio of average on-target:off-target editing for the indicated canonical and evolved DdCBE. The ratio was calculated for each treatment condition as: (average frequency of all on-target C•G base pairs)÷(average frequency of non-target C•G base pairs present in the mitochondrial genome).
Extended Data Fig. 7 Evolution of T7-DdCBE-DddA11 using PANCE for improved GC activity.a, The sequence encoding the T7 RNAP–degron linker was modified to contain GCA or GCG in an effort to evolve for higher activity on GC targets. T7-DdCBE must convert GC8 to GT8 to install a stop codon in the linker sequence and restore T7 RNAP activity. b, Strains for overnight phage propagation assays on GCA or GCG linkers. c, Overnight fold propagation of indicated SP in host strains encoding GCA or GCG linkers. Strains correspond to Extended Data Fig. 7b. T7-DdCBE-DddA11 phage contains the mutations S1330I, A1341V, N1342S, E1370K, T1380I and T1413I in DddA. Dead T7-DdCBE-DddA11 phage contains an additional inactivating E1347A mutation in DddA. Values and error bars reflect the mean±s.d of n = 3 independent biological replicates. d, Phage passage schedule for T7-DdCBE-DddA11 evolution in PANCE using strain 9 transformed with MP6 (red) or strain 10 transformed with MP6 (blue). The table indicates the dilution factor for the input phage population (see Methods for drifting procedure). Output phage titer and fold propagation are shown for a single replicate. Fold propagations above the dotted line in each graph represent propagation >1-fold.
Extended Data Fig. 8 Mitochondrial editing efficiencies of DdCBE variants evolved from GC-specific PANCE.a, Enriched mutations within the DddA gene of T7-DdCBE after PANCE against a GCA or GCG linker. T7-DdCBE-DddA11 was used as the input SP for PANCE. DddA mutations in the input SP are shown in beige. Mutations enriched after 9 or 12 PANCE passages are shown in blue. b-e, Heat maps of mitochondrial base editing efficiencies of HEK293T cells treated with canonical and evolved variants of (b) ND4.3-DdCBE, (c) ND5.4-DdCBE (d) ND5.2-DdCBE and (e) ATP8-DdCBE. Target spacing regions and split DddA orientations are shown for each base editor. For b-e, colors reflect the mean of n = 3 independent biological replicates.
Extended Data Fig. 9 Allele compositions at disease-relevant mtDNA sites in HEK293T cells following base editing by DddA11-containing DdCBE variants.a-c, Allele frequency table of HEK293T cells treated with DddA11-containing (a) ND4.3-DdCBE, (b) ND4.2-DdCBE and (c) ND5.4-DdCBE to install the non-TC mutations m.11642 G > A, m.11696 > A and m.13297 G > A, respectively. On-target cytosines are boxed. Each table is representative of n = 3 independent biological replicates. Values and errors reflect the mean±s.d of n = 3 independent biological replicates.
Extended Data Fig. 10 Structural alignment of DddA with ssDNA-bound APOBEC3G.a, Crystal structure of DddA (grey, PDB 6U08) complexed with DddI immunity protein (not shown). Positions of mutations common to the CCC- and GCC-specific evolutions are colored in purple. Additional mutations are colored according to Fig. 3c. DddA was split at G1397 (red) to generate T7-DdCBE for PANCE and PACE. b, DddA (PDB 6UO8, grey) was aligned to the catalytic domain of APOBEC3G (PDB 2KBO, red) complexed to its ssDNA 5’-CCA substrate (orange) using Pymol. The target C undergoing deamination by APOBEC3G is indicated as C0. Reversion analysis on the DddA11 mutant indicated that A1341V, N1342S and E1370K are critical for expanding the targeting scope of DddA (see Fig. 3f). D317 (red) confers 5’-CC specificity in APOBEC3G and loop 3 controls the catalytic activity of the APOBEC3G.
Supplementary informationRights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and Permissions
About this article
Cite this articleMok, B.Y., Kotrys, A.V., Raguram, A. et al. CRISPR-free base editors with enhanced activity and expanded targeting scope in mitochondrial and nuclear DNA.
Nat Biotechnol (2022). https://doi.org/10.1038/s41587-022-01256-8
Download citation
Received: 19 September 2021
Accepted: 08 February 2022
Published: 04 April 2022
DOI: https://doi.org/10.1038/s41587-022-01256-8
