Investigating the relationship within and between White Caucasian and Afro-Caribbean populations through study of their mitochondrial DNA sequences.

Several different techniques for the extraction, amplification and sequencing of mtDNA were investigated, in an attempt to find a rapid, robust and reliable method of acquiring sequence information from human mitochondrial genomes. Optimal results were obtained following two rounds of symmetric, nested PCR. The subsequent double-stranded, biotinylated, DNA products were immobilised on magnetic beads and sequenced using Sequenase T7 DNA polymerase, in conjunction with either dye-labelled primers or dye-labelled terminators and an automated sequencer (Applied Biosystems). In this manner, mtDNA from 10 unrelated British Caucasian and 114 unrelated Afro-Caribbean blood samples was extracted, amplified and directly sequenced. Sequences of approximately 400 nucleotides were obtained, in both orientations, from two hypervariable segments within the non-coding region of the mitochondrial genome. The sequences were then assembled, analysed and compared; to an external reference sequence (Anderson et al, 1981), to each other (within and between population) and to other published data sets. Comparison to a reference sequence (Anderson et al, 1981) showed that, for each population, both hypervariable regions sequenced were polymorphic to similar extents. The nucleotide sequence present at the time of population splitting was also inferred from further comparisons to the reference sequence (Anderson et al, 1981). A total of 94 different sequences were observed in the Caucasian population, with an average nucleotide diversity of 1.0%. In the Afro-Caribbean population, however, a total of 105 different sequences were observed, with an average nucleotide diversity of 1.75%. The higher nucleotide diversity observed in the Afro-Caribbean data, along with a greater proportion of significant race specific positions, suggested an earlier evolutionary origin for this population. Phylogenetic analysis of the populations was carried out using PAUP (Swofford, 1991) and PHYLIP (Felsenstein, 1995). However, due to the non-clustering of samples known to have the same sequence, it was concluded that any inferences formed from the analysis of resultant genealogical trees had to be viewed accordingly.