A World Wide Survey on Human Specific Alu Insertion Polymorphisms

Alu insertion polymorphisms are widely distributed in the human genome. They are readily analysed and provide an interesting set of markers for human population genetic and forensic studies. The aim of this article is to present preliminary results from our ongoing project on a survey of Alu insertion polymorphisms in various human populations. INTRODUCTION Repetitive DNA sequences comprise a significant portion of the human genome (Fig 1). Alu repeats are classified as short interspersed elements (SINEs) and make up 5-10 % of the human genome (1,2). A typical Alu is a heterodimer of two subunits, derived from the 7SL RNA gene. This retroposon invaded the pre-simian and early simian genomes in great copy numbers. Thus, they can serve as milestones in the anthropological and evolutionary studies (Fig. 2). The insertion of a Alu element at a particular locus can be regarded as a unique event. Once inserted, most Alu elements do not appear to be subject to loss or rearrangement, therefore being stable genetic markers. Alu deletions are rare and even then the deletion leaves behind a footprint (3). The possibility of a deletion, where the two cut points would be exactly at the both ends of the insertion, is negligible. Human specific Alu insertions Alu elements that are mostly, but not exclusively restricted to the human genome have been termed humanspecific (HS) (4,5) (Fig. 2, Fig 3). The HS Alu insertions are further classified into subfamilies Ya5, Ya8 and Yb8 according to base substitutions along their sequence (5). Since the bi-allelic genotyping of Alu insertions is feasible using PCR, agarose gel electrophoresis and ethidium bromide staining, they provide an interesting approach for studies of human history. Indeed, the usefulness of HS Alu polymorphisms as a tool in human population studies has been proven in previous studies (3, 6, 7). MATERIALS AND METHODS We have analysed eight Alu insertion polymorphisms (APO, PV92, TPA25, FXIIIB, D1, ACE, A25, and B65) in 57 populations consisting of a total of 2308 unrelated individuals that were sampled from 16 African populations, from 5 Middle Eastern populations, from 12 European populations, from 17 Asian populations and from 3 and 4 populations from Sahlu and New World, respectively. The samples included data from Bengs et al. (unpublished) and Stoneking et al. (7). Genomic DNA was prepared from peripheral blood lymphocytes using standard protocols. Eight human specific polymorphic Alu loci were amplified using polymerase chain reaction (PCR) and locus specific primers. PCR reactions were carried out in a 25 ml volume containing 20100 ng of template DNA, 5 nmol of each dNTP, 25 pmol of each primer and 1 unit of Taq DNA polymerase (Promega) in 50 mMTris-HCl, pH 8.8, 15 mM (NH4)2SO2 4,1,5 mM MgCl2, 0,1 % Triton X-100, 0,01 % gelatin. 30 cycles of 95 °C for 1 min, 54 °C for 1 min, and 72 °C for 1 min were used in a MJ Research PTL-225 thermal cycler for loci PV92, TPA25, APO, and FXIIIB. 30 cycles of 95 °C for 1 min, 56 °C for 1 min, and 72 °C for 1 min were used in a MJ Research PTL-225 thermal cycler for loci ACE, A25 and B65. Touchdown PCR with annealing temperatures from 61 °C to 58 °C in a total of 30 cycles was used for the D1 locus. PCR products were visualized in UV-light after separation in a 2 % agarose gel and ethidium bromide staining. Allele frequencies, heterozygosity and GST values were calculated using GENEPOP (ver. 3) (8) computer package. In addition, calculation of Nei's standard genetic distances (D) between populations, DA distances between populations and construction of phylogenetic trees were performed using the DISPAN computer program (9). Analysis of molecular variance (AMOVA) was calculated using ARLEQUIN computer package (10). A World Wide Survey on Human Specific Alu Insertion Polymorphisms 2 RESULTS AND DISCUSSION Hardy-Weinberg equilibrium All the populations were tested for the Hardy-Weinberg equilibrium (HWE). No violation of the assumption of the HWE could be demonstrated when using the exact test or the global test across all loci in all populations. Frequency of the Alu insertion and heterozygosity values in various populations Our results indicate a difference in the frequencies of the Alu insertions between various human population groups. For example, at locus PV92 the frequency of the Alu insertion varied from 0.08 to 0.91 and at locus TPA25 the frequency varied from 0.19 to 0.56 in the populations studied. Furthermore, at locus APO the frequency of the Alu insertion varied from 0.17 to 1.00. It is noteworthy that in Eurasian and Middle Eastern population groups the frequency varied from 0.94 to 1.00, whereas in SubSaharan and Saharan population groups the frequency varied from 0.17 to 0.76. Also at locus FXIIIB the frequency of the Alu insertion was lower (0.11-0.30) in SubSaharan and Saharan Africans compared to that in the Arabs and the Jews of Middle East (0.47-0.56) and to that in the Eurasians (0.37-0.96). Interestingly, at locus D1 the frequency of the Alu insertion did not exceed 0.50 in any of the population groups studied. The heterozygosity values across all loci varied from 0.28 to 0.39 in Eurasian populations, from 0.32 to 0.41 in Arabs and Middle Eastern Jews, and from 0.35 to 0.54 in Sub-Saharan and Saharan African populations. Genetic distances and construction of phylogenetic trees The genetic distances between populations showed the largest values between the African and non-African populations. This is also shown in phylogenetic trees constructed by using the neighbor-joining (NJ) method (11) from matrices of either D or DA distances (data not shown). In all the phylogenetic trees constructed, the African populations are closest to the outgroup. As an outgroup we used the genotypes of ancestral state of the Alu insertions (non-insertion). Thus, our data supports the findings in phylogenetic trees contructed using other gentic markers, such as mitochondrial DNA, autosomal microsatellites and serological markers.