SNP profile within the human major histocompatibility complex reveals an extreme and interrupted level of nucleotide diversity

Abstract

The human major histocompatibility complex (MHC) is characterized by polymorphic multicopy gene families, such as HLA and MIC (PERB11); duplications; insertions and deletions (indels); and uneven rates of recombination. Polymorphisms at the antigen recognition sites of the HLA class I and II genes and at associated neutral sites have been attributed to balancing selection and a hitchhiking effect, respectively. We, and others, have previously shown that nucleotide diversity between MHC haplotypes at non-HLA sites is unusually high (>10%) and up to several times greater than elsewhere in the genome (0.08%-0.2%). We report here the most extensive analysis of nucleotide diversity within a continuous sequence in the genome. We constructed a single nucleotide polymorphism (SNP) profile that reveals a pattern of extreme but interrupted levels of nucleotide diversity by comparing a continuous sequence within haplotypes in three genomic subregions of the MHC. A comparison of several haplotypes within one of the genomic subregions containing the HLA-B and -C loci suggests that positive selection is operating over the whole subgenomic region, including HLA and non-HLA genes. [The sequence data for the multiple haplotype comparisons within the class I region have been submitted to DDBJ/EMBL/GenBank under accession nos. AF029061, AF029062, and AB031005-AB031010. Additional sequence data have been submitted to the DDBJ data library under accession nos. AB031005-AB03101 and AF029061-AF029062.]

Figure 1

(A) SNP profiles within the human MHC and another region on 6p23 containing the polymorphic SCA1 locus. The graphs depict the level of nucleotide diversity between two human haplotypes in three continuous sequences within the class II and class I (α and β blocks) regions of the MHC and the SCA1 region on 6p23. The different haplotype comparisons within the class II region and the β block of the class I region are shown by horizontal bars and designated as a–c and d–e, respectively. The details of the comparisons are listed in Table 1. The position and transcription orientation of the HLA and MIC (PERB11) loci are shown as black and open boxes, respectively. Nucleotide diversity for each graph was calculated from a 100-nucleotide window. The position of the duplicated segments containing MIC (PERB11) genes and HLA class I genes, HLA-B and -C, are shown by black and shaded horizontal bars, respectively (Gaudieri et al. 1997a). Indels within the duplicated segments are shown by shaded circles. The 10-kb region between the HLA-B and -C segments duplicated in the telomeric region near HLA-E is indicated by a double black line. The five regions compared between different MHC haplotypes in Table 2 are indicated by a thin black line and labeled as i–v. The multisegment duplications within the α block are shown as horizontal black lines and labeled I–III (Kulski et al. 1999b). (B) The G+C% and CpG% within 100-nucleotide windows are shown below the SNP profile. (C) The retroelement sequences from the LTR/HERV, Alu, and LINE1 groups are depicted below the graph. The large boxes within the HERV sequences are indicated by name below the line. The sharp increase within the α block SNP profile occurs within a cosmid and, therefore, is unlikely to be a result of a chimeric haplotype comparison (shown as a black horizontal bar below the α block (class I) SNP profile.