It has long been known that defects in the structure of the mitochondrial genome can cause various neuromuscular and neurodegenerative diseases. Nevertheless, at present there is no effective method for treating mitochondrial diseases. The major problem with the treatment of such diseases is associated with mitochondrial DNA (mtDNA) heteroplasmy. It means that due to a high copy number of the mitochondrial genome, mutant copies of mtDNA coexist with wild-type molecules in the same organelle. The clinical symptoms of mitochondrial diseases and the degree of their manifestation directly depend on the number of mutant mtDNA molecules in the cell. The possible way to reduce adverse effects of the mutation is by shifting the level of heteroplasmy towards the wild-type mtDNA molecules. Using this idea, several gene therapeutic approaches based on TALE and ZF nucleases have been developed for this purpose. However, the construction of protein domains of such systems is rather long and laborious process. Meanwhile, the CRISPR/Cas9 system is fundamentally different from protein systems in that it is easy to use, highly efficiency and has a different mechanism of action. All the characteristics and capabilities of the CRISPR/Cas9 system make it a promising tool in mitochondrial genetic engineering. In this article, we demonstrate for the first time that the modification of gRNA by integration of specific mitochondrial import determinants in the gRNA scaffold does not affect the activity of the gRNA/Cas9 complex in vitro.