Selfish drive trumps function when animal mitochondrial genomes compete

In addition to favouring traits that enhance organismal fitness, evolution favours selfish gains in replication or transmission. By generating various Drosophila lines with two mitochondrial genotypes, we reveal the impact of selfish selection on how co-residing mitochondrial genomes compete for transmission.

HFSP Long-Term Fellow Hansong Ma and colleagues
authored on Mon, 18 July 2016

For most animal species, mitochondria are the principal generators of cellular ATP with their own genome (known as mtDNA). The genes encoded by mtDNA contribute directly or indirectly to electron transport, which is of paramount importance to the host. Preservation of fitness requires selection of mitochondrial genomes that serve the interest of the host. Indeed, previous work from our group and others had revealed an expected purifying selection that filters out mutations detrimental to electron transport function when closely related genomes are mixed for competition (Fan et al., 2008; Hill et al., 2014; Ma et al., 2014; Stewart et al., 2008). However, with few constraints on replication and segregation, the multi-copied mtDNA might exhibit freewheeling competition readily influenced by selfish behaviours.

In our paper, we show that mitochondrial genomes from diverged strains or different Drosophila species function well in collaboration with the D. melanogaster nuclear genome, but the diverged genomes do not compete based on electron transport function. Some genomes, that we call “bully” genomes, displace “wimpy” genomes even when the bully genome is functionally deficient.  We refer to the advantage carried by such bully genomes as “selfish drive”. In one particular example, a “bully” genome with a temperature sensitive defect in oxidative phosphorylation was paired with a functional but “wimpy” genome: Initially, the stock was healthy, but then the flies died after 5-6 generations as the temperature sensitive genome displaced the functional genome. In another example, selection for one genome based on selfish drive is counterbalanced by purifying selection for another genome, leading to a long-term stable co-transmission of two genomes. By isolating recombinant mitochondrial genomes, we map the selfish drive trait to the non-coding region of the mitochondrial genome. The functions of the non-coding region are poorly understood but replication starts within these sequences.

We then show that genomes from other species differ in their ability to displace the endogenous D. melanogaster mitochondrial genome. The divergence in the ability to compete is not related to the competence of these alien genomes to provide function, as they substituted robustly when the endogenous genome was eliminated. Since the non-coding sequences diverge in evolution faster than the coding sequences, this led us to something that we believe is a generally important evolutionary feature of the mitochondrial genome:  sequences providing electron transport function (the coding region) change little because they are preserved by a purifying selection that conserves function, while sequences governing replication of the genome (the non-coding region) diverge rapidly under a constant selfish selection to outdo each other.  Consequently, when two diverged genomes compete with each other, selfish drive, not function, is the dominant determinant of which mitochondrial genome takes over.

We think that the phenomena we describe are important because the action of selfish selection appears to be a major determinant of the evolutionary trajectory of mitochondrial genomes.  Moreover, the realization that mixing foreign genomes gives rise to such dramatic results has taken on new significance with a new approach in fertility clinics where a mitochondrial donor contributes to the production of “Three-parent embryos”—an approach that has been approved by the UK government and is currently under consideration by FDA (Vogel, 2014). Since not all cytoplasm of the original mother will be eliminated in the procedure, some three-parent babies will be heteroplasmic initially. Our data suggest that successful replacement and likelihood of maintenance of the mutant genome in a stable heteroplasmy will depend on the relative selfish drive of donor and recipient genomes.  Thus, our findings suggest that diverged human mitochondrial genomes (different haplotypes) ought to be assessed for competitive strength and that donors be limited to those with strong selfish drive.


Selfish drive can trump function when animal mitochondrial genomes compete. Hansong Ma, Patrick H. O’Farrell. Journal citation: Nature Genetics 48, 798–802 (2016), doi:10.1038/ng.3587.

Other References

[1]  A mouse model of mitochondrial disease reveals germline selection against severe mtDNA mutations. Fan, W., Waymire, K.G., Narula, N., Li, P., Rocher, C., Coskun, P.E., Vannan, M.A., Narula, J., Macgregor, G.R., Wallace, D.C., 2008.  Science 319, 958–962. doi:10.1126/science.1147786

[2]   Selective propagation of functional mitochondrial DNA during oogenesis restricts the transmission of a deleterious mitochondrial variant. Hill, J.H., Chen, Z., Xu, H., 2014. Nat Genet 46, 389–392. doi:10.1038/ng.2920

[3]   Transmission of mitochondrial mutations and action of purifying selection in Drosophila melanogaster. Ma, H., Xu, H., O'Farrell, P.H., 2014.  Nat Genet 46, 393–397. doi:10.1038/ng.2919

[4]   Strong purifying selection in transmission of mammalian mitochondrial DNA. Stewart, J.B., Freyer, C., Elson, J.L., Wredenberg, A., Cansu, Z., Trifunovic, A., Larsson, N.-G., 2008.  PLoS Biol 6, e10–71. doi:10.1371/journal.pbio.0060010

[5]   Assisted reproduction. FDA considers trials of 'three-parent embryos'. Vogel, G., 2014.   Science 343, 827–828. doi:10.1126/science.343.6173.827

Link to paper

Pubmed link