Fig. 1: Generation of a heteroplasmic MT-ND5 gene knockout mutant mouse model and genotyping.

a Schematic illustration of DdCBE in the mouse MT-ND5 gene. The base editing window is annotated in bold. Possible editing loci are shown in red (m.C12336T) and light blue (m.G12341A). Transcription activator-like effector (TALE)-binding arrays are depicted in different colors for each of the four nucleotides. b Diagram of designed base editing in MT-ND5. Base editing targets are indicated in red (m.C12336T) and light blue (m.G12341A). The nonsense mutation is introduced at the 199th amino acid. c Editing efficiency of m.C12336T in DdCBE-treated mice. Targeted deep sequencing data were obtained from the genomic DNA of the pups’ toes. The exact p value was 2.4E-11 (*p < 0.05, **p < 0.01, and ***p < 0.001 via two-tailed Student’s t test). d Efficiency of editing various tissues from F0, F1, and F2 mice (from the left). Genomic DNA was obtained from tissues at least 6 weeks after birth. The dark and light bars indicate the frequencies of m.C12336T and m.G12341A, respectively. The error bars indicate the standard error of the mean (s.e.m.) for replicates (n = 3). e Timeline of genomic DNA collection for targeted deep sequencing data and enumerations of mutant sequences from F0 and F1 newborn pups. Genomic DNA was isolated one week after birth from feces at four-week intervals and after 30 weeks. Edited bases are highlighted in red. f Oxygen consumption rates of mouse embryonic fibroblasts. Each respiratory parameter was normalized to the value of the WT. For MEF-1 and MEF-2, the m.C12336T-editing efficiencies were 19% and 21%, respectively. The error bars indicate the SEMs for biologically independent replicates (n = 5).