Mitochondria are the energy-making machines in the cell. Also, since evolutionarily they are thought to evolve from symbiotic bacteria, they have their own little genomes, which are mainly evolved through the maternal side, and also maybe because of their important function, mutations in genes affecting them are very damaging, and even lethal. However, mitochondrial DNA was not available to editing tools like CRISPR, so changes there were simply not possible, and any chance of recovering from such a mutation could only be either by approaching the symptoms, of from the womb via a three-parent baby with a mitochondrial donor. Now, however, mitochondrial gene editing is possible.
Thanks to a bacterium. Specifically, to a bacterial toxin. As a work published recently in Nature explained 1, the issue with most gene editing techniques is that they need single stranded DNA to operate their changes, and to open the double helix they need an RNA primer molecule that serves as a model. It is this model that could not access the mitochondria. However, it seems that the toxin secreted by the bacteria Burkholderia cenocepacia offers a way around this issue: it can modify double-stranded DNA. This capacity, which is originally used by the bacteria to destroy their victims, could then be used to perform mitochondrial gene editing.
But not in its original form, obviously, for it would have been as lethal and as unspecific as the bacterial toxin. To make it useful as a tool, and restrain the cutting to where intended, the researches split the protein into two non-toxic parts that only upon union would be capable of cutting, and they linked them to TALEN proteins, which can be designed to target specific DNA sections.
As a first effort, I consider their results on cell culture, where they achieved up to approximately 50% efficient C>T (cytosine to thymine) changes in the mitochondrial DNA quite a success.
Obviously, much work needs to be done before it can even be thought of as a possibility for mitochondrial gene editing in humans. It would still need to improve efficiency, expand the type of base modifications it can achieve, and see if it also works in vivo. But I certainly like the idea that there is a –future– therapeutic option for those suffering from one of the over 150 mitochondrial syndromes known today.