Mitochondrial DNA (mtDNA) is a maternally inherited 16.6 kbp circular genome that codes for 13 subunits of the mitochondrial respiratory chain, 2 rRNAs and 22 tRNAs. The mutation rate in mtDNA is high and therefore, mutations have accumulated sequentially to lineages that have diverged tens of thousands of years ago. The neutral theory predicts that a proportion of these variations may be slightly deleterious, associated with diseases and selected against, but the issue is still controversial.
This study reports an analysis of selection against mutations in mtDNA. First, the population prevalence of one of the most pathogenic mtDNA mutations, the common MELAS mutation (3243A>G), was determined in a population-based screening setting in Northern Ostrobothnia, and the reproductive capacity, or genetic fitness, of women with the mutation was estimated in order to measure for the first time the degree of host-level selection against this highly pathogenic mutation. The frequency of 3243A>G was high, as the minimum estimate for the prevalence was 10.2/100,000, and this together with the geographical distribution of maternal ancestors of the mutation carriers suggested that nuclear genes may be involved in the population history of the mutation. Surprisingly, the genetic fitness of mutation carriers was not reduced, suggesting that the average host-level selection against carriers is not strong. Second, all available complete human mtDNA sequences worldwide (N=847) were collected into a database and analysed for evidence to support the hypothesis concerning slightly deleterious mutations and selective constraints imposed by lineage-specific interactions. 465 distinct missense and 6 nonsense mutations were identified. 48% of the amino acid replacements changed the polarity, 44% hydropathy, 32% aliphaticity, 26% size, 13% aromaticity, and 8% charge. Nonconservative amino acid replacements were found to be more common among the evolutionarily recent mutations than among the older ones, and mutations that have arisen more than once during human evolution showed different properties from the remaining ones. The major continent-specific mtDNA lineages were analysed in terms of nucleotide diversity indices, neutrality tests and nonsynonymous/synonymous rate ratios, and patterns suggesting selective constraints possibly due to lineage-specific interactions were identified. Moreover, a general correlation between nucleotide position and nucleotide polymorphism was identified in the mtDNA.
The results are compatible with the assumption that selection has a marked role in human mtDNA evolution and that selective constraints may vary between populations, so that the pathogenic potential of a given mutation may depend markedly on the presence of other, interacting mutations.
Another study indicating that natural selection works on mtDNA It is a functional thing, not ‘junk’ so there was no reason to reason to suppose it didn’t.