Age of the common ancestor of human mitochondrial DNA.

Abstract

Studying the evolutionary change of the control region of human (H) and chimpanzee (C) mitochondrial DNAs (mtDNAs), Vigilant et al. ( 199 1) concluded that the root of the phylogenetic tree for H mtDNAs is located at a sequence from African populations and that the age of the common H ancestral mtDNA is 166,000-249,000 years. Hedges et al. ( 1992) and Templeton ( 1992) challenged the first of these conclusions, showing that valid statistical tests do not support it. In this letter, I would like to show that Vigilant et al.? second conclusion is also questionable and that, if we consider statistical errors associated with the rate of nucleotide substitution, the range of the estimate of the age becomes much wider. In the control region, transitional nucleotide substitution is known to occur at a much higher rate than does transversional substitution, and the transitional differences between the H and C mtDNAs seem to have reached a saturation level at hypervariable sites of this region (Kocher and Wilson 199 1) . Vigilant et al. therefore used information on the transversional differences between H and C and the transition/ transversion ratio obtained from H data, to calibrate the rate of nucleotide substitution. Let n be the number of transversional differences between H and C, and let R be the ratio of transitional (s) and transversional (v) substitutions observed among H sequences. [ Hasegawa and Horai ( 199 1) suggested that R may be different between H and C, but K. Tamura’s (personal communication ) detailed analysis indicates that it is essentially the same for the two species. Here I consider only H data, following Vigilant et al.] The expected total number of nucleotide substitutions between H and C can then be estimated by n +nR. So, if the total number of nucleotides examined is m, then the total number of nucleotide substitutions, per site, between H and C is given by