Animal mitochondrial DNA undergoes homologous recombination

Despite the evolutionary role of homologous recombination in protecting genome integrity, it has not been experimentally demonstrated in the animal mitochondrial DNA (mtDNA). In the study described below, by selecting against the two parental genomes in Drosophila, progeny with only recombinant mtDNA were recovered. Recombination can occur between highly diverged mitochondrial genomes and often involves long continuous stretches of exchange. In addition, some recombinants allowed us to map a novel mitochondrial trait to the non-coding region of the mtDNA. Thus, this work not only provides a definitive resolution to a long-standing controversy about the existence of recombination in animal mitochondria and its functional benefit, but also establishes recombination as a new genetic technique for the manipulation of animal mitochondrial genomes.

HFSP Long-Term Fellow Hansong Ma and colleagues
authored on Mon, 19 October 2015

Homologous recombination operates in organisms from bacteriophage to human. However, there is very little support for mitochondrial recombination in animals, exchange of homologous sequences among mitochondrial genomes has even been demonstrated under various conditions for plant and fungal species. If it were to occur, recombination among mitochondrial genomes is expected to generate favourable combination of alleles that can then be selected. It could also play a negative role by promoting deletions and insertions, thus contributing to the genetics of disease mutations in the germline and age associated accumulation of mtDNA mutations in the soma. Hence, a definitive characterization of recombination of mitochondrial genomes will impact our understanding of the evolution of this genome, the behavior of disease mutations, and the age associated degenerative changes of the mitochondrial genome.

In order to overcome the well-established difficulty of detecting rare recombination between similar sequences, we performed cytoplasmic transplantation (to overcome uniparental inheritance) to create heteroplasmic lines carrying two mitochondrial genomes with the desired characteristics in Drosophila. We then developed novel genetic tools to allow the application of two selections, each operating against one of the two starting ‘parental’ genomes, in order to select for recombinant genomes.

We isolated progeny with recombinant genomes in four scenarios. In three scenarios, we introduced double-stranded breaks (DSBs) by targeting restriction enzymes to germline mitochondria, cutting either only one parental genome, or both genomes at different loci. Linearizing both mitochondrial genomes isolated recombinant genomes with a high efficiency. We also used this system to force recombination between highly diverged sequences (mitochondrial genomes from two Drosophila species diverged about 10 MYA), and showed that DSB-induced recombination is not a simple local repair, but involves the exchange of long stretches of sequences. Linearizing only one genome reduced frequency of recombination, yet recombinants were still detected without too much effort.

In the last scenario, no DSBs were introduced. Instead, a defective mutant genome that is temperature-sensitive lethal was mixed with a functional genome to make a heteroplasmic line. Since the defective mutant has a strong drive for replication and transmission, it became more widespread and completely replaced the functional genome over a few generations. Flies with only the defective genome cannot propagate at the restrictive temperature, and thus the replacement led to the death of the population (a typical example of selfish behaviour of mtDNA). We followed 50 such heteroplasmic lines at a restrictive temperature and isolated three lines that survived such a selection. Using PacBio sequencing technology, we showed that each of these lines has a recombinant genome that lacked the temperature sensitive mutation but retained the high replicative drive. This study shows that even rare recombination can uncouple a positively selected drive from detrimental mutations and selection can then restore the health of the population. Thus, genetic exchange would prevent rogue genomes, which appear to arise repeatedly, from wiping out lineages. The isolated recombinants also allowed us to map mtDNA sequence that is responsible for the replicative drive.

In conclusion, this work shows that recombination can be harnessed to dissect function and evolution of the mitochondrial genome.  In addition, on the way to characterizing the recombination, we introduced genomes from other species and forced recombination with endogenous D. melanogaster genomes, suggesting that trans-species recombination can be used to identify the locus of diverged traits.


Selections that isolate recombinant mitochondrial genomes in animals. Hansong Ma and Patrick H O'Farrell eLife. 2015; 4: e07247. Published online 2015 Aug 3. doi: 10.7554/eLife.07247

Link to eLIFE article

Link to the F1000

Pubmed link