Fig. 5: Programmable ADP-ribosylation enables specific and differentiated editing in human cells lacking TARG1.
All experiments were conducted in HEK293T ΔTARG1 cells. a, Extent of base mutagenesis based on the relative location of ADPr-T according to e. Cumulative thymine base editing across 21 sgRNA targets, within 37 5′-TYTN-3′ motifs at positions 3–14 (with position 1 is at the PAM-distal end). Solid black lines represent the median and gray lines represent the quartiles. Each dot represents the mean of three independent transient transfections without selection or sorting for a given sgRNA. b, Relationship between the outcome of base mutagenesis and the DarT2 recognition sequence according to e. Distributions were calculated for base mutations occurring at 33 DarT2 recognition motifs across 21 sgRNAs. c, Frequency of indels for DarT2D–nSpCas9 compared to SpCas9 at the same target sites. d, Frequency of kilobase-scale deletions for DarT2D–nSpCas9 compared to SpCas9 at the same target sites. e, Guide-independent base substitutions at editable bases within orthogonal R-loops formed with dSaCas9 distant from (left) or close to (right) the target site. Editing was assessed through append editing (DarT2D–nSpCas9) or cytosine base editing (BE4) in HEK293T ΔTARG1 cells. Editing was significantly higher at position C11 of the R-loop in PTEN with the targeted BE4 (P = 0.016) or a nontargeted BE4 (P = 0.003) and at positions C2 (P = 0.003), C6 (P = 0.007) and C8 (P = 0.006) of the R-loop in VEGFA for the targeted BE4. f, Frequency of base substitutions using programmable ADP-ribosylation (DarT2D–nSpCas9) or glycosylation (DAF-TBE) of thymine in HEK293T ΔTARG1 cells. Bars and error bars in b, c, e and f represent the mean and s.e.m. of three independent transient transfections without selection or sorting. Bars in d represent the mean of the two independent transient transfections without selection or sorting.
