Eurasian and African mitochondrial DNA influences in the Saudi Arabian population

Abstract

Background: Genetic studies of the Arabian Peninsula are scarce even though the region was the center of ancient trade routes and empires and may have been the southern corridor for the earliest human migration from Africa to Asia. A total of 120 mtDNA Saudi Arab lineages were analyzed for HVSI/II sequences and for haplogroup confirmatory coding diagnostic positions. A phylogeny of the most abundant haplogroup (preHV)1 (R0a) was constructed based on 13 whole mtDNA genomes.

Results: The Saudi Arabian group showed greatest similarity to other Arabian Peninsula populations (Bedouin from the Negev desert and Yemeni) and to Levantine populations. Nearly all the main western Asia haplogroups were detected in the Saudi sample, including the rare U9 clade. Saudi Arabs had only a minority sub-Saharan Africa component (7%), similar to the specific North-African contribution (5%). In addition, a small Indian influence (3%) was also detected.

Conclusion: The majority of the Saudi-Arab mitochondrial DNA lineages (85%) have a western Asia provenance. Although the still large confidence intervals, the coalescence and phylogeography of (preHV)1 haplogroup (accounting for 18 % of Saudi Arabian lineages) matches a Neolithic expansion in Saudi Arabia.

Figure 1

Phylogenetic position of the haplogroup M Arab 201 sequence. All mutation differences are listed with respect to the revised Cambridge Reference Sequence (rCRS) [66]. This sequence has accession number DQ904234 in GenBank.

Figure 2

Haplogroup (preHV)1 phylogeny based on thirteen complete or nearly complete sequences. The Iberian Peninsula (IP969) and the seven Saudi Arab (Ar) sequences are from this study. Five additional sequences were taken from the literature as detailed in Methods. Numbers along links refer to nucleotide positions with i indicating insertions, d indicating deletions, underlining indicating recurrent mutations in the (preHV)1 haplogroup. From the star all individuals present the following mutations with respect to rCRS: 263, 750, 1438, 2706, 4769, 7028, 8860, 14766, 15326. All mutation differences are detailed with respect to the revised Cambridge Reference sequence (rCRS) [66]. Our eight sequences were given GenBank accession numbers [GenBank: DQ904235], [GenBank: DQ904236], [GenBank: DQ904237], [GenBank: DQ904238], [GenBank: DQ904239], [GenBank: DQ904240], [GenBank: DQ904241] and [GenBank: DQ904242] for sequences # Ar20, Ar440, Ar505, Ar439, Ar448, Ar194, Ar222 and IP969 respectively.

Figure 3

Reduced median network relating (preHV)1 HVSI sequences. The central motif (star) differs from rCRS at positions 16126 and 16362 in HVI control region. Numbers along links refer to nucleotide positions minus 16000. Positions not used in diversity estimations are in italics. The broken lines are less probable links and/or recurrent mutations. Size of boxes is proportional to the number of individuals included. Codes are: ARA, Arab; BAL, Balkanian; BED, Bedouin; CAS, Caspian; CAU, Caucasus; DRZ, Druze; EGY, Egyptian; ETH, Ethiopian; EUR, European; IND, Indian; IP, Iberian Peninsula; IRK, Iraki; IRN, Iranian; JBA, Baltic Jew; JET, Ethiopian Jew; JEU, European Jew; JIP, Iberian Jew; JIR, Iraki Jew; JOR, Jordanian; JRN, Iranian Jew; JYE, Yemeni Jew; KEN, Kenian; KUR, Kurd; MAU, Mauritanian; MBE, Moroccan Berber; MOR, Moroccan; NUB, Nubian; PAK, Pakistani; PAL, Palestinian; SAH, Saharan; SEN, Senegalese; SOM, Somalian; SYR, Syrian; TUK, Turkish; TUN, Tunisian; and YEM, Yemeni.

Figure 4

Graphical relationships among the studied populations. Codes are as in Table 1. (A) MDS plot based on F_ST haplotypic distances. Stress value is 0.086. Dimension 1 axe has been shortened to include the Druze sample. (B) PC analysis based on haplogroup frequencies. The two components represent 37% of the total variance.