Mitochondrial adenine base editing by TALED in cultured mouse cells.

a Schematic overview of the TALED adenine base editor architectures tested. The TadA8e enzyme is responsible for A:T to G:C editing, whilst DddAtox and Nt.BspD61(C) make the DNA accessible for enzyme activity. Black boxes indicate MTS and DNA binding elements. For full schematic see Supplementary Fig. 1. b On-target editing percentage values calculated by next generation sequencing of the six adenine base editor combinations targeted to the Mt-Cytb, Mt-CoII and Mt-Atp6 genes using dTALED, sTALED and mTALED. Sequence and mouse mtDNA position is indicated. n = 2. c Average mtDNA-wide off-target editing frequency values calculated by next generation sequencing of the adenine base editor combinations targeted to the Mt-Cytb, Mt-CoII and Mt-Atp6 genes from b. n = 2.

Mitochondrial adenine base editing by mitoBE in cultured mouse cells.

a Schematic overview of the mitoBE adenine base editor architectures tested. The TadA8e enzyme is responsible for A:T to G:C editing, while Nt.BspD61(C) make the DNA accessible for editing activity. Black boxes indicate MTS and DNA binding elements. For full schematic see Supplementary Fig. 1. b On-target editing percentage values calculated by next generation sequencing of base editors targeted to Mt-Atp6 gene using mitoBE, containing the Nt.BspD61(C) nickase. n = 2. c Average mtDNA-wide off-target editing frequency values from d. Multiple t test comparison performed.

In vivo mouse mtDNA editing using adenine base editors.

a Scheme of in vivo experiments with neonatal mice. TadA8e(V28R), DddAtox and BspD61(C) were encoded in separate AAV genomes, encapsidated in AAV9 then each pair (TadA8e(V28R) + DddAtox or TadA8e(V28R) + BspD61(C)) was imultaneously administered by temporal vein injection at 5 × 1011 vg/mouse of each monomer. Animals were sacrificed 4-weeks post-injection and their heart, skeletal muscle and liver tissues were examined for mtDNA editing. b On-target editing percentage values calculated by next generation sequencing of mouse tissues analysed following adenine base editor AAV delivery to the Mt-Atp6 gene. Sequence and mouse mtDNA position is indicated. c Average mtDNA-wide off-target editing frequency values calculated by next generation sequencing from mice in b. Bars represent the mean and error bars represent ± SEM (n = 3). d Real-Time qPCR relative quantification of mtDNA copy number (CN) in 4-week-old mouse tissues following adenine base editor AAV delivery. Bars represent the mean and error bars represent ± SEM (n = 3). Multiple t test comparison performed.

Strategies for adenine base editing.

Schematics showing the four main methods used for adenine base editing in mitochondria. Dimeric TALED (dTALED) uses a TALE-binding mtDNA sequence specific protein attached to the catalytically dead DddAtox (for DNA unwinding) on one monomer, and the TadA8e adenine deaminase to the monomer binding the opposite mtDNA strand. Split DddA TALED (sTALED) which uses the G1397 split of the DddAtox attached to two TALED monomers, alongside the TadA8e. Monomeric TALED (mTALED) which contains only one sequence specific TALE protein attached to both the DddAtox and the TadA8e. Dimeric mitoBEs use nickase enzyme BspD61(C) to allow for single stranded activity of TadA8e.

Adenine base editing of Mt-Cytb.

Sanger sequencing traces of the six adenine base editor combinations targeted to the Mt-Cytb gene. Only the on-target editing window is shown.

Adenine base editing of Mt-CoII.

Sanger sequencing traces of the six adenine base editor combinations targeted to the Mt-CoII gene. Only the on-target editing window is shown.

Adenine base editing of Mt-Atp6.

Sanger sequencing traces of the six adenine base editor combinations targeted to the Mt-Atp6 gene. Only the on-target editing window is shown.

Adenine base editing of Mt-Atp6 by mitoBEs.

Sanger sequencing traces for the two mitoBE combinations targeting to the Mt-Atp6 gene, compared to dTALED editing directed to the same gene. Only the on-target editing window is shown.

Cell survival following transfection with AAV plasmids encoding TadA8e variants.

Cell survival percentage calculated using Countess Automated Cell Counter following 4 days post transfection of AAV plasmids spiked with empty GFP plasmid and sorting for enrichment. n = 4 biological replicates. Error bars indicate ±SEM Statistical significance was calculated from a one-tailed Student’s t test **** P-value < 0.0001.

Delivery of adenine base editor AAVs to mouse skeletal muscle.

Western blot analysis of the levels of FLAG-tagged AAV DddA or BspD61C (L) and HA-tagged AAV TadA8e (H) in skeletal muscle (quadriceps) of mice at 4 weeks post-injection. Coomassie Brilliant Blue (CBB) was used as loading control. * indicates a non-specific band detected by the anti-FLAG antibody.

Delivery of adenine base editor AAVs to mouse heart.

Western blot analysis of the levels of FLAG-tagged AAV DddA or BspD61C (L) and HA-tagged AAV TadA8e (H) in heart of mice at 4 weeks post-injection. Coomassie Brilliant Blue (CBB) was used as loading control. * indicates a non-specific band detected by the anti-FLAG antibody.

Assessment of OXPHOS complex levels in skeletal muscle following adenine base editor AAV delivery.

a Western blot analysis of steady-state levels of components of OXPHOS in skeletal muscle of 4-week old control mice, mice injected with AAVs for TadA8e + DddA or TadA8e + BspD61(C). Coomassie Brilliant Blue (CBB) was used as loading control. b The western blots from a quantified using ImageJ. Error bars indicate ±SEM. Multiple t test comparison performed. *p<0.05.

Assessment of OXPHOS complex levels in heart following adenine base editor AAV delivery.

A Western blot analysis of steady-state levels of components of OXPHOS in heart of 4-week old control mice, mice injected with AAVs for TadA8e + DddA or TadA8e + BspD61(C). Coomassie Brilliant Blue (CBB) was used as loading control. b The western blots from a quantified using ImageJ. Error bars indicate ± SEM. Multiple t test comparison performed.