MitoTALEN: A General Approach to Reduce Mutant mtDNA Loads and Restore Oxidative Phosphorylation Function in Mitochondrial Diseases

Abstract

We have designed mitochondrially targeted transcription activator-like effector nucleases or mitoTALENs to cleave specific sequences in the mitochondrial DNA (mtDNA) with the goal of eliminating mtDNA carrying pathogenic point mutations. To test the generality of the approach, we designed mitoTALENs to target two relatively common pathogenic mtDNA point mutations associated with mitochondrial diseases: the m.8344A>G tRNA(Lys) gene mutation associated with myoclonic epilepsy with ragged red fibers (MERRF) and the m.13513G>A ND5 mutation associated with MELAS/Leigh syndrome. Transmitochondrial cybrid cells harbouring the respective heteroplasmic mtDNA mutations were transfected with the respective mitoTALEN and analyzed after different time periods. MitoTALENs efficiently reduced the levels of the targeted pathogenic mtDNAs in the respective cell lines. Functional assays showed that cells with heteroplasmic mutant mtDNA were able to recover respiratory capacity and oxidative phosphorylation enzymes activity after transfection with the mitoTALEN. To improve the design in the context of the low complexity of mtDNA, we designed shorter versions of the mitoTALEN specific for the MERRF m.8344A>G mutation. These shorter mitoTALENs also eliminated the mutant mtDNA. These reductions in size will improve our ability to package these large sequences into viral vectors, bringing the use of these genetic tools closer to clinical trials.

Figure 1

Development of mitoTALEN for two pathogenic mtDNA mutations. (a) The structure of the mitoTALEN monomers is illustrated. They contain a mitochondrial localization signal (MLS), an immunological tag (FLAG or HA) and the TALE DNA-binding domain. The latter is fused to a FokI that works as a heterodimer (EL only dimerizes with KK). (b) Using the established TALE code, we designed TALE binding domains against mtDNA regions harboring two pathogenic mutations. The first region was at position m.8344A>G in the tRNA^Lys gene, which is associated with the MERRF syndrome. The second region, surrounding position m.13513G>A in the ND5 gene is associated with MELAS/Leigh syndrome. We have designed two pairs for each region. The figure shows only one for each, but the second pair is described in Supplementary Figure S1. The base pair at position T0 is shown in green color and the mutated position in red (when they coincide they are labeled red). Arrows indicate the mutated base pair.

Figure 2

mitoTALEN monomers are expressed and localized in mitochondria. (a) Taking advantage of the immune tags, we transfected COS7 cells with the respective pair of plasmids coding for the m.8344A>G mitoTALEN monomers. After 48 hours, we immunostained the cells with antibodies against FLAG or HA depicted in green. Cells were incubated with mitotracker red before fixation. DAPI was used for nuclear staining. Western blots (right panels) showed the presence of specific proteins. Double bands may indicate incomplete removal of the MLS. (b) The same experiment described above was performed for the mitoTALEN monomers against the m.13513G>A mtDNA mutation. Molecular weight of the different monomers were: m.8344A>G-15.5 RVDs=102.5 KDa; m.8344A>G-9.5 RVDs = 92.5 KDa; m.13513G>A-15.5 RVDs = 102.5KDa; m.13513G>A-9.5 RVDs = 92.5KDa. Normalization was done with monoclonal actin and tubulin antibodies.

Figure 3

mitoTALEN against single nucleotide changes can effectively reduce mutant mtDNA loads. (a) Transmitochondrial cybrids harboring heteroplasmic levels of the m.8344A>G mtDNA mutation were simultaneously transfected with two plasmids, coding for the two mitoTALEN monomers. After 48 hours, cells were sorted for the fluorescent markers and the sorted population expressing both monomers (mCherry and eGFP) were referred to as the “Yellow” population (both fluorescent markers). The mtDNA heteroplasmy in these cells was compared to the one in the “Black” population (sorted cells with no detectable fluorescence, Supplementary Figure S4). Sorted cell populations had their DNA extracted were analyzed by PCR/RFLP as described after digestion with BglI to differentiate the mutant from the WT load. (b) Comparison of change in heteroplasmy at 2, 15, or 30 days. (c) The same experiment was performed for transmitochondrial cybrids harboring the m.13513G>A mutation with two different pairs of mitoTALENs, using MboI digestion. The gel illustrates the results with one pair, whereas the quantification shows the results for both (mean ± SD; n = 3). The binding region for the different monomers is described in Figure 1 and Supplementary Figure S1. Unt, Untransfected.

Figure 4

Functional recovery of mitoTALEN-treated MERRF m.8344A>G tRNA lysine transmitochondrial cybrids. (a) Transmitochondrial cybrids harboring heteroplasmic levels of the m.8344A>G mtDNA mutation were treated with mitoTALEN and the “Yellow” population expanded for 14 days. Cells were assayed for oxygen consumption and compared with the parental line. Oligomycin was used to inhibit respiration. To determine the maximum respiration, the uncoupler CCCP was added in a stepwise manner. Quantification of the respiratory rates (endogenous and maximal) is depicted in the lower part of panel a. (b) The same cells were analyzed for the steady-state levels of OXPHOS proteins by western blot. To exacerbate the defect, cells were treated with different concentrations of doxycycline for 72 hours, a mitochondrial protein synthesis inhibitor. (c) Enzymatic activity of complex IV (CIV) and citrate synthase (CS) were measured spectrophotometrically in mitochondria from cultured cells treated (Yell) and untreated (Unt) with mitoTALENs and in 143B osteosarcoma control line. Values of specific activity of CIV/CS in the ND5 mutant clone were compared with the values in mutant clone after TALEN treatment. Panel a: mean±SD; n = 3; t-test. Panel c: mean ± SEM; n = 3; ANOVA, Tukey post hoc.

Figure 5

Functional recovery of mitoTALEN-treated MELAS/Leigh Syndrome m.13513G>A ND5 transmitochondrial cybrids. (a) Transmitochondrial cybrids harboring heteroplasmic levels of the m.13513G>A mtDNA mutation were treated with mitoTALENs and the “Yellow” (Yell) population expanded for 14 days. Cells were assayed for oxygen consumption and compared with the untreated parental line (Unt.). (b) The activity of Complex I and cytochrome oxidase (panel c) was measured in these cells, plus a 143B osteosarcoma control line, using spectrophotometric assays. Values were normalized to citrate synthase activity. Error bars correspond to mean ± SEM; n = 3; ANOVA, with Tukey post hoc.

Figure 6

Shorter mitoTALENs can effectively reduce mutant mtDNA loads. We compared the original functional m.8344A>G mitoTALEN monomers, having 15.5 or 9.5 RVDs, with shorter monomers of 10.5 and 7.5 RVDs, respectively. The specific binding sites are described in Supplementary Figure S5 and Supplementary Text. All different combinations between sense and antisense monomers promoted reduction in mutant loads, as shown by comparing heteroplasmy in “Black” and “Yellow” sorted cell populations. The bar graph represents three independent experiments (mean ± SEM; n = 3).