Specific elimination of mutant mitochondrial genomes in patient-derived cells by mitoTALENs

Abstract

Mitochondrial diseases are commonly caused by mutated mitochondrial DNA (mtDNA), which in most cases coexists with wild-type mtDNA, resulting in mtDNA heteroplasmy. We have engineered transcription activator-like effector nucleases (TALENs) to localize to mitochondria and cleave different classes of pathogenic mtDNA mutations. Mitochondria-targeted TALEN (mitoTALEN) expression led to permanent reductions in deletion or point-mutant mtDNA in patient-derived cells, raising the possibility that these mitochondrial nucleases can be therapeutic for some mitochondrial diseases.

Figure 1. Using mitoTALEN to eliminate mtDNAs with the common deletion

(a) We designed mitoTALEN monomers that bind wild–type sequence adjacent to the breakpoint present in the mtDNA carrying the “common deletion”. In the wild–type mtDNA, these sites are far apart, blocking FokI dimerization. (b) Plasmids express one of the two monomers of the mitoTALEN, and a fluorescent marker (eGFP or mCherry) from the same transcriptional unit through the use of a T2A translation stuttering sequence. The plasmids also contained additional elements to optimize expression and mitochondrial targeting, such as 3′UTRs from nuclear encoded mitochondrial genes. (c) Anti–FLAG or anti–HA antibodies were used in immunocytochemistry, showing strict mitochondrial localization of Δ5–mitoTALEN monomers in 24 hrs–transfected COS7 cells. (d) Cells were sorted for eGFP and mCherry 48 hours after transfection with two individual plasmids coding for the Δ5–mitoTALEN monomers. DNA isolated from the cell fractions that were not expressing fluorescent markers (“Black”), only one marker (“Green”) and both eGFP and mCherry (“Yellow”) were subjected to the 3–primer PCR analysis to quantify the relative levels of deleted mtDNA (e). PCR was performed with the primers depicted by colored arrows in Fig. 1a. Panel f shows the quantitation of 3 independent experiments (t–Test unpaired between “black” and “yellow” values; n=3; ***P = <0.001). Error bars correspond to SD of the mean.

Figure 2. Using mitoTALEN to reduce the levels of the pathogenic m.14459A mtDNA point mutation

(a) Approach for the elimination of a point mutant (m.14459G>A). In this case, one of the mitoTALEN monomers binds to wild–type sequence (left monomer) and the other to sequence containing the point mutation within the MT–ND6 gene (right monomer). (b) Cybrids harboring the Leber’s–dystonia m.14,459A mtDNA mutation (Clones 1 and 3) were transiently transfected with 14459A–mitoTALEN expressing plasmids (Exp.1 and 2). Clone 1 harbors approximately 90% m.14459A whereas clone 3 harbors approximately 55%. The panel shows the RFLP analysis of the mutation load in the “black” and “yellow” cells 48 hours after transfection FACS sorted as shown in Fig. 1d. Quantification of n=3 independent experiments is shown (c). Untr=untransfected cells. (d) Quantitation of the total mtDNA levels by qPCR showed a decrease in clone 1 two days after transfection, but not at 14 d. Likewise, wild–type control 143B cells did not show a decrease in mtDNA levels, even after 2 d. (e) Enzyme activity (complex I/citrate synthase ratio) in control, mutant and mutant cells transfected with the 14459A mitoTALEN (14 d after transfection). (t–Test unpaired between “black” and “yellow” values; n=3; * P = <0.005, ** P = <0.002, *** P = <0.001). Error bars correspond to SD of the mean.