mtDNA Variation in the Altai-Kizhi Population of Southern Siberia: A Synthesis of Genetic Variation

Human Biology, Aug 2006 by Phillips-Krawczak, Christine, Devor, Eric, Zlojutro, Mark, Moffat-Wilson, Kristin, Crawford, Michael H

The sample consists of 98 individuals from the village of Mendur-Sokkon. Informed consent was obtained by the Russian members of the research team when the samples were collected in 1994. The village population numbers approximately 1,000 individuals subdivided into three major patrilineal clans (Irkit, Kipchak, and Todosh) plus a number of smaller clans.

The three Altaian samples considered here are referred to as Altai 1 (present study of Mendur-Sokkon), Altai 2 (Shields et al. 1993), and Altai 3 (Derenko et al. 2003). The surrounding populations used for comparison with the Altai samples consist of Kirghiz, Kazakh, Uighur, Khakassians, Tuva, Sojots, Buryats, Mongol, Han Chinese, and Russians (Figure 2).

mtDNA Analysis. DNA was extracted from the blood samples using a standard phenol-chloroform technique (Birren et al. 1997). The DNA samples were analyzed for mtDNA haplogroups corresponding to Asian- and Europeanspecific lineages. Using standard PCR methods, we analyzed amplified mtDNA fragments for diagnostic variant sites using restriction fragment length polymorphisms (RFLPs). Table 1 identifies the polymorphic sites and the restriction enzymes used for the mtDNA haplogroup designations.

Restricted fragments were separated and typed by electrophoresis on 3% NuSieve agarose gels. Gels were stained with ethidium bromide, and DNA fragments were visualized using an ultraviolet transilluminator. Sequencing of the mtDNA HVS-I from nucleotide position 16151 to 16383 was performed at Integrated DNA Technologies on an ABI 310 capillary sequencer to further delineate haplogroup membership into individual haplotypes. Region 16151-16383 was the longest consecutive sequence that encompassed the largest number of successful runs. Failed runs were also attempted using the reverse primer in addition to the forward primer, without success. The failed runs were likely caused by sample degradation, which prevented successful sequencing; thus 61 samples were used in the HVS-I analyses.

Analytical Methods

Median-Joining Network Analysis. Median-joining network analysis was carried out, and phylogenetic networks were constructed by hand with the aid of Network, version 4.0 (Fluxus Engineering; available at http://www.fluxusengineering.com) for 61 sequences from Mendur-Sokkon village. The medianjoining algorithm used generates the most parsimonious tree composed of proportionately sized circles that represent observed haplotype frequencies (Bandelt et al. 1995, 1999). It resolves parallelisms and reversals with a compatibility argument justified by character-state frequency criteria.

Mismatch Distributions. Mismatch distributions were constructed using the program IWAVE (Sherry et al. 1994). These distributions were used to identify characteristics in the genetic structure of natural populations that can be attributed to the action of demographic processes (Kingman 1982). Two common signatures of population growth are starlike genealogies (Slatkin and Hudson 1991) and unimodal distributions of pairwise differences (mismatch distributions) (Harpending et al. 1993). Mismatch distributions for the three Altai population samples, as well as the distributions for the comparative population data sets, were constructed.