How HLA diversity is apportioned: influence of selection and relevance to transplantation

Abstract

In his 1972 paper ‘The apportionment of human diversity’, Lewontin showed that, when averaged over loci, genetic diversity is predominantly attributable to differences among individuals within populations. However, selection can alter the apportionment of diversity of specific genes or genomic regions. We examine genetic diversity at the human leucocyte antigen (HLA) loci, located within the major histocompatibility complex (MHC) region. HLA genes code for proteins that are critical to adaptive immunity and are well-documented targets of balancing selection. The single-nucleotide polymorphisms (SNPs) within HLA genes show strong signatures of balancing selection on large timescales and are broadly shared among populations, displaying low F ST values. However, when we analyse haplotypes defined by these SNPs (which define ‘HLA alleles’), we find marked differences in frequencies between geographic regions. These differences are not reflected in the F ST values because of the extreme polymorphism at HLA loci, illustrating challenges in interpreting F ST . Differences in the frequency of HLA alleles among geographic regions are relevant to bone-marrow transplantation, which requires genetic identity at HLA loci between patient and donor. We discuss the case of Brazil's bone marrow registry, where a deficit of enrolled volunteers with African ancestry reduces the chance of finding donors for individuals with an MHC region of African ancestry. This article is part of the theme issue ‘Celebrating 50 years since Lewontin's apportionment of human diversity’.

The background and legacy of Lewontin's apportionment of human genetic diversity

“…Over time, the 85% number would be replicated with some deviations (e.g. 80-95%) across a number of studies using related approaches on similar data, including by Nei and Roychoudhury that same year and by others shortly after using similar data [51,[56][57][58][59][60], and with novel forms of data such as alleles at HLA loci ( [56,61]; also see [62]).…”

Mathematical constraints onFST: multiallelic markers in arbitrarily many populations

“…In particular, we caution users to examine upper bounds on F ST to assess how mathematical constraints influence observations. As the constraints are strongest for K = 2, this step is valuable in pairwise comparisons; it is also useful when the frequency M of the most frequent allele can be small in relation to the number of populations K , such as for high-diversity forensic [ 28 ] and immunological [ 29 ] loci in human populations. Visual inspection of the values of M and F ST within their bounds can suggest that constraints have an effect.…”

Celebrating 50 years since Lewontin's apportionment of human diversity

“…The genetic differences that do exist can be relevant in biomedical settings, in part because they can affect population-level differences in Mendelian disease allele frequencies, genetic risk factors for complex disease, transplantation matching probabilities and treatment responses (e.g. [28,29]). As discussed by Novembre [2], Lewontin's paper [1] has been referenced beyond the field of human evolutionary genetics as supporting an oversimplified view of human variation, in which population structure is completely ignored even in situations in which it might be relevant, such as matching problems in forensics or transplantation, where population variation in match probabilities affects the societal use of population-genetic computations [27,29].…”

“…Maróstica et al [29] examine human population structure in the setting of the human leucocyte antigen (HLA) and bone-marrow transplantation matching. In transplantation matching, the goal is to find donors with a sufficiently close genetic match to patients at HLA loci in order to prevent rejection of the transplanted tissue by the patient.…”

“…[28,29]). As discussed by Novembre [2], Lewontin's paper [1] has been referenced beyond the field of human evolutionary genetics as supporting an oversimplified view of human variation, in which population structure is completely ignored even in situations in which it might be relevant, such as matching problems in forensics or transplantation, where population variation in match probabilities affects the societal use of population-genetic computations [27,29].…”