Recent male-mediated gene flow over a linguistic barrier in Iberia, suggested by analysis of a Y-chromosomal DNA polymorphism

Abstract

We have examined the worldwide distribution of a Y-chromosomal base-substitution polymorphism, the T/C transition at SRY-2627, where the T allele defines haplogroup 22; sequencing of primate homologues shows that the ancestral state cannot be determined unambiguously but is probably the C allele. Of 1,191 human Y chromosomes analyzed, 33 belong to haplogroup 22. Twenty-nine come from Iberia, and the highest frequencies are in Basques (11%; n=117) and Catalans (22%; n=32). Microsatellite and minisatellite (MSY1) diversity analysis shows that non-Iberian haplogroup-22 chromosomes are not significantly different from Iberian ones. The simplest interpretation of these data is that haplogroup 22 arose in Iberia and that non-Iberian cases reflect Iberian emigrants. Several different methods were used to date the origin of the polymorphism: microsatellite data gave ages of 1,650, 2,700, 3,100, or 3,450 years, and MSY1 gave ages of 1,000, 2,300, or 2,650 years, although 95% confidence intervals on all of these figures are wide. The age of the split between Basque and Catalan haplogroup-22 chromosomes was calculated as only 20% of the age of the lineage as a whole. This study thus provides evidence for direct or indirect gene flow over the substantial linguistic barrier between the Indo-European and non-Indo-European-speaking populations of the Catalans and the Basques, during the past few thousand years.

Figure 1

SRY-2627 polymorphism. A, Detection by PCR-RFLP analysis, and restriction map of PCR fragment. Digestion of the 1,242-bp product by BsiHKAI in the T allele (haplogroup 22) chromosomes gives fragments of 945 and 297 bp (digests shown are partial). Products of 946, 210, and 86 bp are obtained by digestion of the C allele (here, haplogroup 1) chromosomes with BanI. The fragment generated from chimp DNA contains a BanI site at the polymorphic position and therefore resembles the human C allele. hg = Haplogroup; Φ = HaeIII-digested ΦX174 phage DNA. Short arrows indicate PCR primers. B, Sequences of the SRY-2627 C and T alleles and of the homologous regions in DNAs of great apes. Arrows under the haplogroup 1 sequence indicate complete and partial copies of 12- and 15-bp direct repeats. Because of these, several alignments are possible, and the ancestral state cannot be unambiguously deduced; however, the polymorphism is at the third base of one of the 15-bp repeats in human DNA, and all other copies of the repeat, in both humans and apes, share a C at this position, suggesting that C is ancestral. In the alignment shown, the BanI site in chimp DNA is not homologous to that in DNA of the human C allele.

Figure 2

Geographical distribution of haplogroup-22 chromosomes. A, Worldwide distribution (see table 1). Data for South America are from Bianchi et al. (1997), as follows: pooled indigenous South Americans, 5/93 SRY-2627/T allele chromosomes; La Plata nonindigenous groups, 3/26 SRY-2627/T allele chromosomes. B, Distribution within Iberia and the Béarnais of southern France. Data on Madrid and Castilla la Mancha have been pooled.

Figure 3

Median joining microsatellite haplotype networks for haplogroup-22 chromosomes (A) and haplogroup-1 chromosomes (B) and (C) MSY1 network for haplogroup 22 chromosomes. A microsatellite or MSY1 haplotype is represented by a circle, with its area proportional to the number of individuals having that haplotype; haplotype number (see tables 2and 3) is given within each circle. A single arrow between circles represents a single mutational step in a haplotype, and its direction indicates an increase in allele length; the mutating microsatellite (e.g., “390,” representing DYS390) or MSY1 repeat block (e.g., “4,” representing the block of type 4 repeats) is indicated on the arrow. Arrows do not imply evolutionary pathways. A small blackened circle on a line indicates an intermediate, unobserved haplotype used as a node (i.e., median vector). Thick edges to circles highlight haplotypes shared by both haplogroups (haplogroup 1 only).