I'm interested in expressing a synthetic gene from the mitochondrial genome. It is possible to knock a transgene into the mitochondrial genome such that it's passed on to subsequent generations?
The other way to do it might be to transfect entire mitochodrial genomes. Look up the TFAM system.
In either case, you'll have the problem that the mitochodrial genome pool in the cell is highly dynamic, and can probably even be replenished from other cells. To get germline transmission, or even any stability at all, you'd have to somehow create a selective advantage for your edited copy to outcompete the non-edited cellular pool. This is not well established. I saw a poster once where they expressed a restriction enzyme from the transgenic mtDNA to chop up wild-type, but not transgenic mtDNA. I never saw what came of that though.
The Riken group in Japan had at least some transient expression in plant mitochondria using a cell penetrating peptide complexed with both DNA and a mitochondrial targeting peptide.