Mitigation of age-dependent accumulation of defective mitochondrial genomes

Unknown processes promote the accumulation of mitochondrial DNA (mtDNA) mutations during aging. Accumulation of defective mitochondrial genomes is thought to promote the progression of heteroplasmic mitochondrial diseases and degenerative changes with natural aging. We used a heteroplasmic Drosophila model to test 1) whether purifying selection acts to limit the abundance of deleterious mutations during development and aging, 2) whether quality control pathways contribute to purifying selection, 3) whether activation of quality control can mitigate accumulation of deleterious mutations, and 4) whether improved quality control improves health span. We show that purifying selection operates during development and growth but is ineffective during aging. Genetic manipulations suggest that a quality control process known to enforce purifying selection during oogenesis also suppresses accumulation of a deleterious mutation during growth and development. Flies with nuclear genotypes that enhance purifying selection sustained higher genome quality, retained more vigorous climbing activity, and lost fewer dopaminergic neurons. A pharmacological agent thought to enhance quality control produced similar benefits. Importantly, similar pharmacological treatment of aged mice reversed age-associated accumulation of a deleterious mtDNA mutation. Our findings reveal dynamic maintenance of mitochondrial genome fitness and reduction in the effectiveness of purifying selection during life. Importantly, we describe interventions that mitigate and even reverse age-associated genome degeneration in flies and in mice. Furthermore, mitigation of genome degeneration improved well-being in a Drosophila model of heteroplasmic mitochondrial disease.

Unlike nuclear genotype, which is largely stable during one’s lifetime, when genetically distinct mitochondrial genomes co-reside (heteroplasmy), their relative proportions shift during growth, development, and aging. This shift is not random (1–7). Mutant mitochondrial DNA (mtDNA) variants accumulate during aging and in the progression of some mitochondrial diseases (1–5). Stereotyped changes in abundance of particular alleles in different tissues in human and mouse indicate that selective forces favor different mitochondrial genomes (3, 6, 7). However, despite the importance of mitochondrial function to health and well-being, we have limited understanding of the processes underlying the accumulation of mitochondrial mutations with age.The nuclear genome encodes mechanisms of quality control that survey the function of mitochondria and eliminate or compromise the proliferation of defective mitochondria (8, 9). Two described mechanisms use PINK1, the product of a gene discovered as one of the causes of early-onset familial Parkinson’s disease, as a sensor of mitochondrial function. PINK1 accumulates on the surface of mitochondria having a reduced membrane potential, and its kinase activity in this location signals several downstream events (10–15). In one pathway defined largely in a cell culture model, PINK1 activates PARKIN, the product of another Parkinson’s disease gene, which then triggers elimination of compromised mitochondria by mitophagy (16). In a second pathway, acting in the Drosophila female germ line (17, 18), PINK1 acts in a PARKIN-independent pathway that targets a protein called Larp to inhibit its role in promoting biogenesis of the mitochondria (18) (SI Appendix, Fig. S2).Since deleterious mtDNA mutations compromise electron transport of mitochondria, it seems that quality control would put genomes carrying such mutations at a disadvantage, creating a purifying selection that leads to their elimination. However, this outcome is far from certain. Various factors such as the sharing of gene products among mitochondria as a result of dynamic fission and fusion could mask the consequences of heteroplasmic mutations, shielding them from quality control. Indeed, a number of studies suggest a contrast between the germ line, which exhibits purifying selection, and adult somatic tissues, which often show accumulation of mutations. Studies in Drosophila show that purifying selection acting in the female germ line eliminates deleterious mutations in a few generations (18–21). Genetic dissection revealed that this purifying selection depends on the PINK1/LARP pathway of quality control (18). On the other hand, an elegant study that induced heteroplasmic deletions in the flight muscle of the adult fly detected no substantial indications of purifying selection unless additional stressors were introduced (22). Furthermore, a detailed study in Drosophila carrying a proofreading defective mitochondrial DNA polymerase (mutator line) showed that adult flies accumulate a spectrum of mutations biased toward deleterious mutations (23). Since deleterious mutations would potentially be removed by purifying selection, this finding argues either that it was not operating or that it was opposed by a stronger selection favoring deleterious mutations (23). Similarly, studies in mouse suggest a discordance between germ line and soma. When mutations in the mitochondrial genome that were introduced in a mutator line were passed through subsequent generations in a wild-type background, there was selective elimination of deleterious mutations, a strong signal of purifying selection (7). In contrast, zygotically accumulated mutations in the mutator mouse exhibited high levels of deleterious mutations, suggesting a lack of purifying selection in the soma. Furthermore, Parkin mutant mice did not show a significant increase in mutations in adult wild-type or mutator mice, suggesting that Parkin-dependent quality control does not contribute to purifying selection (16). While these studies suggest major changes in the efficiency of purifying selection, other studies have suggested continued purifying selection in some circumstances. For example, adult human T cells of mitochondrial disease patients show exceptionally low heteroplasmy levels, suggesting cell type-specific action of purifying selection (24). We sought to measure purifying selection and to understand the nature of quality control in the soma during growth, development, and aging using an experimental model developed in Drosophila.A previously described heteroplasmic line of Drosophila melanogaster carries a wild-type mitochondrial genome (Yak-mt) from another species, Drosophila yakuba, and a D. melanogaster genome (Mel-mt^ts) crippled by a temperature-sensitive mutation in cytochrome oxidase subunit 1 (mt:CoI^T300I) [Fig. 1A and (25)]. The Mel-mt^ts genome has an intrinsic advantage in replication. At a permissive temperature, it gradually displaces the Yak-mt genome, resulting in the loss of the D. yakuba genome in a few generations (25). At 29 °C, a temperature at which the CoI^T3000I mutant cannot support viability (26), purifying selection counters the replicative advantage of the mutant Mel-mt^ts genome, preventing it from taking over (25). qPCR gives a measure of the ratio Yak-mt/total-mt. The difference in this ratio between permissive and restrictive temperatures allows us to assess the impact of a functional disparity on competition between the two genomes and provides a measure of purifying selection.Open in a separate windowFig. 1.Quality control modulates the ratio of heteroplasmic mitochondrial genomes during development. (A) Heteroplasmy for schematized Yak-mt and Mel-mt^ts genomes was established by transferring cytoplasm of D. yakuba embryos into D. melanogaster embryos carrying the doubly mutant genome mt:ND2^del + mt:Col^T300I (Mel-mt^ts). (B) The proportion of Yak-mt (Yak-mt/total-mt) following development from egg to adult shows action of quality control. Eggs (2-h collection at 29 °C) were assayed (gray bar) or allowed to develop to 5 d after eclosion at 22 °C (blue) or 29 °C (amber). Adult females have a large contribution from oocyte mtDNA. (C) Quality control operates in multiple tissues with different effectiveness. The blue and amber bars (22 °C or 29 °C, respectively) show the proportion of Yak-mt in different tissues of late third instar larvae. (D) The impact of purifying selection declined with age in the CNS. (E) The decline in relative abundance of wild-type, indicated by the dashed regression lines (orange for 29°C and blue for 22°C), Yak-mt following eclosion was fast. (F) The signature of quality control is absent during maturation of adults. Yak-mt/total-mt ratios from gut and brain taken from 5-d (gray bars) or 20-d adults aged at 22 °C or 29 °C (blue and amber bars, respectively). Here and below, *P < 0.05; **P < 0.01; and ***P < 0.001 by one-way ANOVA/Tukey’s multiple comparison test. Data represent eight independent biological repeats, with each repeat being an average of ratios assessed in three samples of eggs or adults. For tissues, data represent tissues dissected from eight individuals. Error bars represent standard error. In E, slopes differ (*P < 0.05) by linear regression. Cyto, cytoplasm; AED, after egg deposition; ns, not significant.Past work using this heteroplasmic line focused on changes in the relative abundance of the two genomes from one generation to the next and uncovered the action of purifying selection during oogenesis (18–21, 25, 27). It is notable that this selection depends on quality control, occurs by competition between mitochondria within the oocyte, and does not involve selection for organismal fitness (18, 21). Changes in the ratio of Yak-mt/total-mt during the lifetime of the fly suggested that maintenance of this ratio is dynamic (25). Here, we examined changes in the relative abundance of the two genomes in the soma. Shifts in the ratio of Yak-mt/total-mt provide a measure of the effectiveness of purifying selection during the life of the fly and a means of assessing whether purifying selection can be genetically or pharmacologically modified. We found evidence that purifying selection is active in the soma during growth and development. The influence of mutations in quality control genes suggests that somatic mechanisms of quality control overlap those operating in the germ line. However, the effectiveness of purifying selection declines with age and has no obvious impact in tested tissues beyond 5 d after eclosion of adult flies. Importantly, we found that quality control can be stimulated in the adult by genetic alterations or by feeding kinetin and that such measures forestall mutational accumulation and aging phenotypes in Drosophila and can reverse mutational accumulation in aged mice.