Killer cell immunoglobulin-like receptor (KIR) genes and their HLA-C ligands in a Ugandan population

Abstract

Killer cell immunoglobulin-like receptor (KIR) genes are expressed by natural killer cells and encoded by a family of genes exhibiting considerable haplotypic and allelic variation. HLA-C molecules, the dominant ligands for KIR, are present in all individuals and are discriminated by two KIR epitopes, C1 and C2. We studied the frequencies of KIR genes and HLA-C1 and C2 groups in a large cohort (n = 492) from Kampala, Uganda, East Africa and compared our findings with published data from other populations in sub-Saharan Africa (SSA) and several European populations. We find considerably more KIR diversity and weaker linkage disequilibrium in SSA compared to the European populations and describe several novel KIR genotypes. C1 and C2 frequencies were similar to other SSA populations with a higher frequency of the C2 epitope (54.9 %) compared to Europe (average 39.7 %). Analysis of this large cohort from Uganda in the context of other African populations reveals variations in KIR and HLA-C1 and C2 that are consistent with migrations within Africa and potential selection pressures on these genes. Our results will help understand how KIR/HLA-C interactions contribute to resistance to pathogens and reproductive success.

Fig. 1

Principal component analysis based on seven KIR gene carrier frequencies for the 23 African populations described in Table 2. Arrows represent the relative contribution of each KIR gene frequency to the variability along the first two axes (PC1 and PC2)

Fig. 2

Worldwide HLA-C2 allele frequencies for 208 populations from the Allele Frequency Net database and the Ugandan population. Only non-migrant populations are included

Fig. 3

KIR genotype profiles and frequencies observed in the Ugandan population. Reference numbers are according to the Allele Frequency Net database. New genotypes identified in this study are described as “novel.” Genes are presented in the order observed on sequenced KIR haplotypes except for KIR2DS3S5 and KIR2DL5. Inhibitory genes are in red, activating genes are in blue, and the pseudo gene KIR2DP1 is in gray

Fig. 4

Inferred KIR haplotypes for 15 individuals representing 13 KIR genotypes. Genes are presented in the order observed on published sequenced KIR haplotypes. Inhibitory genes are in red, activating genes are in blue, and pseudo genes are in gray. Haplotype names are according to Pyo et al. (2013)

Fig. 5

Genotypes observed for the Ugandan population (n = 492), six other African populations from the Allele Frequency Net database (Gabon, Ghana, Senegal, South Africa San, and South Africa Xhosa, n = 354), and the UK population (n = 584), showing proportions of common and unique genotypes for each group

Fig. 6

Comparison of the linkage disequilibrium observed in the KIR region between the Ugandan population and six African and seven European populations. The LD values calculated for 11 KIR gene loci for Ugandan population are compared with the values obtained using the same method by Single et al. (2008). Red represents strong positive LD, blue represents strong negative LD