Enhanced mitochondrial DNA editing in mice using nuclear-exported TALE-linked deaminases and nucleases

Abstract

We present two methods for enhancing the efficiency of mitochondrial DNA (mtDNA) editing in mice with DddA-derived cytosine base editors (DdCBEs). First, we fused DdCBEs to a nuclear export signal (DdCBE-NES) to avoid off-target C-to-T conversions in the nuclear genome and improve editing efficiency in mtDNA. Second, mtDNA-targeted TALENs (mitoTALENs) are co-injected into mouse embryos to cleave unedited mtDNA. We generated a mouse model with the m.G12918A mutation in the MT-ND5 gene, associated with mitochondrial genetic disorders in humans. The mutant mice show hunched appearances, damaged mitochondria in kidney and brown adipose tissues, and hippocampal atrophy, resulting in premature death.

Keywords: DdCBE; Mitochondrial DNA editing; NES; mitoTALEN; mtDNA.

Fig. 1

Improved cytosine base editing efficiency in mouse blastocysts. a Plasmid construct expressing DdCBE-NES. The amino acid sequence of the NES from the MVM NS2 protein is shown. b Base editing targets in the mitochondrial and nuclear genomes. C(G)-to-T(A) conversion targets are shown in red, and mismatches between mtDNA and nuclear DNA are shown in lowercase blue font. c–e Target C(G)-to-T(A) editing efficiencies in mouse blastocysts. The sequencing data were obtained from cultured blastocysts that developed after one-cell stage embryos were microinjected with mRNA encoding the DdCBE or DdCBE-NES. mtDNA and nuclear DNA amplicons were respectively analyzed for the MT-ND5 gene and a potential off-target site on chromosome 4 (c), the MT-TrnA gene and an identical target site on chromosome 5, and MT-Rnr2 and an identical target site on chromosome 6. For mitochondrial-specific amplicons, data points from DdCBE-injected blastocysts are shown as magenta and those from DdCBE-NES-injected blastocysts as green. For nuclear DNA-specific amplicons, purple indicates DdCBE and light blue DdCBE-NES. All data sets represented in the graphs were obtained from at least three biologically independent samples. The exact p-values are *0.0189 and **0.0045 for c, *0.0327 for d, and *0.0881 for e. (N ≥ 3; n.s., not significant, *p < 0.05, **p < 0.01, and ***p < 0.001 using Student’s two-tailed t-test)

Fig. 2

Improved editing efficiency of DdCBE-NES in the presence of mitoTALEN in mice. a Base editing target for generating the m.G12918A mutation. The TALE binding sequences for the DdCBE are highlighted in green and for the mitoTALEN in orange. The mitoTALEN was designed to recognize a site with a single mismatch with the wild-type mtDNA sequence, as denoted with a lowercase letter. b Effect of mitoTALEN, used to eliminate wild-type mtDNA in mouse blastocysts, on the m.G12918A base editing efficiency. Targeted deep sequencing data were obtained from blastocysts that developed after microinjection of zygotes with mRNAs encoding the indicated constructs. Exact p-values are **0.0006 for DdCBE compared with DdCBE + mitoTALEN, **0.0022 for DdCBE-NES compared with DdCBE-NES + mitoTALEN, and *0.0467 for DdCBE compared with DdCBE-NES; n.s. is 0.3519. (N ≥ 3; n.s., not significant, *p < 0.05, **p < 0.01, and ***p < 0.001 using Student’s two-tailed t-test). c Images of a mouse harboring the m.G12918A point mutation (ptpup-18) with a wild-type C57BL/6 mouse. d Transmission electron micrographs of the mitochondria in the kidney and brown adipose tissue from wild-type and ptpup-18 mice. e Immunohistochemistry images of the brain sections from wild-type and m.G12918A mutant mice. Brain sections, stained with anti-NeuN antibody visualized with DAB (3,3′-diaminobenzidine), from the forebrain and midbrain regions of brains from a wild-type mouse, pt119 (a littermate of the pt121 mouse with normal behavior), and the pt121 mouse with a hunchback phenotype. The arrow on the pt121 forebrain section indicates one of the enlarged lateral ventricles, and the arrow on the pt121 midbrain section shows the asymmetric hippocampus in this mouse harboring the m.G12918A mutation