Understanding differences between phylogenetic and pedigree-derived mtDNA mutation rate: a model using families from the Azores Islands (Portugal)

Abstract

We analyzed the control region of the mitochondrial DNA (mtDNA) from maternally related individuals originating from the Azores Islands (Portugal) in order to estimate the mutation rate of mtDNA and to gain insights into the process by which a new mutation arises and segregates into heteroplasmy. Length and/or point heteroplasmies were found at least in one individual of 72% of the studied families. Eleven new point substitutions were found, all of them in heteroplasmy, from which five appear to be somatic mutations and six can be considered germinal, evidencing the high frequency of somatic mutations in mtDNA in healthy young individuals. Different values of the mutation rate according to different assumptions were estimated. When considering all the germinal mutations, the value of the mutation rate obtained is one of the highest reported so far in family studies. However, when corrected for gender (assuming that the mutations present in men have the same evolutionary weight of somatic mutations because they will inevitably be lost) and for the probability of intraindividual fixation, the value for the mutation rate obtained for HVRI and HVRII (0.2415 mutations/site/Myr) was in the upper end of the values provided by phylogenetic estimations. These results indicate that the discrepancy, that has been reported previously, between the human mtDNA mutation rates observed along evolutionary timescales and the estimations obtained using family pedigrees can be minimized when corrections for gender proportions in newborn individuals and for the probability of intraindividual fixation are introduced. The analyses performed support the hypothesis that (1) in a constant, tight bottleneck genetic drift alone can explain different patterns of heteroplasmy segregation and (2) in neutral conditions, the destiny of a new mutation is strictly related to the initial proportion of the new variant. Another important point arising from the data obtained is that, even in the absence of a paternal contribution of mtDNA, recombination may occur between mtDNA molecules present in an individual, which is only observable if it occurs between mtDNA types that differ at two or more positions.