Improved base excision repair inhibition and bacteriophage Mu Gam protein yields C:G-to-T:A base editors with higher efficiency and product purity

Abstract

We recently developed base editing, the programmable conversion of target C:G base pairs to T:A without inducing double-stranded DNA breaks (DSBs) or requiring homology-directed repair using engineered fusions of Cas9 variants and cytidine deaminases. Over the past year, the third-generation base editor (BE3) and related technologies have been successfully used by many researchers in a wide range of organisms. The product distribution of base editing-the frequency with which the target C:G is converted to mixtures of undesired by-products, along with the desired T:A product-varies in a target site-dependent manner. We characterize determinants of base editing outcomes in human cells and establish that the formation of undesired products is dependent on uracil N-glycosylase (UNG) and is more likely to occur at target sites containing only a single C within the base editing activity window. We engineered CDA1-BE3 and AID-BE3, which use cytidine deaminase homologs that increase base editing efficiency for some sequences. On the basis of these observations, we engineered fourth-generation base editors (BE4 and SaBE4) that increase the efficiency of C:G to T:A base editing by approximately 50%, while halving the frequency of undesired by-products compared to BE3. Fusing BE3, BE4, SaBE3, or SaBE4 to Gam, a bacteriophage Mu protein that binds DSBs greatly reduces indel formation during base editing, in most cases to below 1.5%, and further improves product purity. BE4, SaBE4, BE4-Gam, and SaBE4-Gam represent the state of the art in C:G-to-T:A base editing, and we recommend their use in future efforts.

Fig. 1. Effects of knocking out UNG on base editing product purity.

(A) Architecture of BE3. (B) Protospacers and PAM (blue) sequences of the genomic loci tested, with the target C’s analyzed in (A) shown in red. (C) HAP1 (UNG⁺) and HAP1 UNG⁻ cells were treated with BE3, as described in Materials and Methods. The product distribution among edited DNA sequencing reads (reads in which the target C is mutated) is shown. See fig. S1 for C-to-T editing efficiencies, which generally varied between 15 and 45%. (D) Frequency of indel formation following treatment with BE3 in HAP1 or HAP1 UNG⁻ cells. Values and error bars reflect the means and SD of three independent biological replicates performed on different days. ns (not significant), P ≥ 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, by two-tailed Student’s t test.

Fig. 2. Effects of multi-C base editing on product purity.

(A) Architectures of BE3, CDA1-BE3, and AID-BE3. (B) Representative high-throughput sequencing data of BE3-, CDA1-BE3–, and AID-BE3–treated human HEK293T cells. The sequence of the protospacer is shown at the top, with the PAM in blue and the target C’s in red, with subscripted numbers indicating their position within the protospacer. Underneath each sequence are the percentages of total sequencing reads with the corresponding base. The relative percentage of target C’s that are cleanly edited to T rather than to non-T bases are much higher for cells treated with AID-BE3, which edits three C’s at this locus, than for cells treated with BE3, which edits only one C. (C) HEK293T cells were treated with BE3, CDA1-BE3, and AID-BE3, as described in Materials and Methods. The product distribution among edited DNA sequencing reads (reads in which the target C is mutated) is shown. (D) Protospacers and PAM (blue) sequences of genomic loci studied, with the target C’s analyzed in (B) shown in red. (E) Frequency of indel formation (see Materials and Methods) following the treatment in (A). Values and error bars reflect the means and SD of three independent biological replicates performed on different days. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, by two-tailed Student’s t test.

Fig. 3. Effects of changing the architecture of BE3 on C-to-T editing efficiencies and product purities.

(A) Architectures of BE3, SSB-BE3, N-UGI-BE3, and BE3-2xUGI. (B) Protospacers and PAM (blue) sequences of genomic loci studied, with the target C’s in (C) shown in purple and red, and the target C’s in (B) shown in red. (C) HEK293T cells were treated with BE3, SSB-BE3, N-UGI-BE3, and BE3-2xUGI, as described in Materials and Methods. The product distribution among edited DNA sequencing reads (reads in which the target C is mutated) is shown for BE3, N-UGI-BE3, and BE3-2xUGI. (D) C-to-T base editing efficiencies. Values and error bars reflect the means and SD of three independent biological replicates performed on different days. ns, P ≥ 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, by two-tailed Student’s t test.

Fig. 4. Effects of linker length variation in BE3 on C-to-T editing efficiencies and product purities.

(A) Architecture of BE3, BE3C, BE3D, and BE3E. (B) Protospacers and PAM (blue) sequences of genomic loci studied, with the target C’s in (C) shown in purple and red, and target C’s in (D) shown in red. (C) HEK293T cells were treated with BE3, BE3C, BE3D, or BE3E, as described in Materials and Methods. C-to-T base editing efficiencies are shown. (D) The product distribution among edited DNA sequencing reads (reads in which the target C is mutated) is shown for BE3, BE3C, BE3D, and BE3E. Values and error bars reflect the means and SD of three independent biological replicates performed on different days. ns, P ≥ 0.05; *P < 0.05; **P < 0.01, by two-tailed Student’s t test.

Fig. 5. BE4 increases base editing efficiency and product purities compared to BE3.

(A) Architectures of BE3, BE4, and Target-AID. (B) Protospacers and PAM (blue) sequences of genomic loci studied, with the target C’s in (C) shown in purple and red, and the target C’s in (D) shown in red. (C) HEK293T cells were treated with BE3, BE4, or Target-AID, as described in Materials and Methods. C-to-T base editing efficiencies are shown. (D) The product distribution among edited DNA sequencing reads (reads in which the target C is mutated) is shown for BE3 and BE4. Values and error bars reflect the means and SD of three independent biological replicates performed on different days. ns, P ≥ 0.05; *P < 0.05; **P < 0.01, by two-tailed Student’s t test.

Fig. 6. Fusion with Gam from bacteriophage Mu reduces indel frequencies.

(A) Architectures of BE3-Gam and BE4-Gam. (B) HEK293T cells were treated with BE3, BE3-Gam, BE4, BE4-Gam, SaBE3, SaBE3-Gam, SaBE4, or SaBE4-Gam, as described in Materials and Methods. C-to-T base editing efficiencies are shown. (C) Frequency of indel formation (see Materials and Methods) following the treatment in (B). (D) Product distribution among edited DNA sequencing reads (reads in which the target C is mutated). (E) Protospacers and PAM (blue) sequences of genomic loci studied, with the target Cs in (B) shown in purple and red, and the target Cs in (D) shown in red. Values and error bars of BE3-Gam, SaBE3-Gam, BE4-Gam, and SaBE4-Gam reflect the means and SD of three independent biological replicates performed on different days. Values and error bars of BE3, SaBE3, BE4, and SaBE4 reflect the means and SD of six independent biological replicates performed on different days by two different researchers. ns, P ≥ 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, by two-tailed Student’s t test.