Patterns of haplotype diversity within the serpin gene cluster at 14q32.1: insights into the natural history of the alpha1-antitrypsin polymorphism.

The levels of haplotype diversity associated with different alpha1-antitrypsin (PI) alleles were assessed by the analysis of three microsatellites located within or close to corticosteroid-binding globulin (CBG), alpha1-antitrypsin [PI-(TG)n] and protein C inhibitor [PCI-(TG)n] loci in three populations with different historic backgrounds: Portugal, the Basque Country and São Tomé Príncipe (Gulf of Guinea). Unlike the more distant PCI-(TG)n repeat, allelic variation at PI-(TG)n reflected distinct phases of mutational recovery of microsatellite diversity around different founder alleles and showed a considerable differentiation between alpha1-antitrypsin protein variants. In accordance with population history, the Basque sample presented overall reduced levels of microsatellite variation. The African sample, although presenting the highest PCI-(TG)n diversity, showed a lineage-specific reduction in PI-(TG)n heterozygosity within the oldest M1Ala213 variant that could have been caused by (1) selection at a closely linked locus or (2) biases in the microsatellite mutation process leading to a stable equilibrium distribution. Age estimates of alpha1-antitrypsin variants based on microsatellite variation suggest that the Z deficiency allele appeared 107-135 generations ago and could have been spread in Neolithic times. The S mutation has an older 279- to 470-generation age, indicating that its high frequencies in Iberia did not result from a recent bottleneck and that PI*S could have originated in this region. M2 and M3 types had lower age estimates than would be expected from their wide geographical distributions, suggesting that their dispersion in Europe might have been preceded by important bottlenecks.