A limit to the divergent allele advantage model supported by variable pathogen recognition across HLA‐DRB1 allele lineages

Abstract

Genetic diversity in human leukocyte antigen (HLA) molecules is thought to have arisen from the co‐evolution between host and pathogen and maintained by balancing selection. Heterozygote advantage is a common proposed scenario for maintaining high levels of diversity in HLA genes, and extending from this, the divergent allele advantage (DAA) model suggests that individuals with more divergent HLA alleles bind and recognize a wider array of antigens. While the DAA model seems biologically suitable for driving HLA diversity, there is likely an upper threshold to the amount of sequence divergence. We used peptide‐binding and pathogen‐recognition capacity of DRB1 alleles as a model to further explore the DAA model; within the DRB1 locus, we examined binding predictions based on two distinct phylogenetic groups (denoted group A and B) previously identified based on non‐peptide‐binding region (PBR) nucleotide sequences. Predictions in this study support that group A allele and group B allele lineages have contrasting binding/recognition capacity, with only the latter supporting the DAA model. Furthermore, computer simulations revealed an inconsistency in the DAA model alone with observed extent of polymorphisms, supporting that the DAA model could only work effectively in combination with other mechanisms. Overall, we support that the mechanisms driving HLA diversity are non‐exclusive. By investigating the relationships among HLA alleles, and pathogens recognized, we can provide further insights into the mechanisms on how humans have adapted to infectious diseases over time.

HLA class I molecular variation and peptide-binding properties suggest a model of joint divergent asymmetric selection

“…3 . This latest observation parallels a suggestion by Lau et al of an upper limit to sequence divergence generated by the DAA model at HLA class II locus, HLA-DRB1 (Lau et al 2015 ). One reason invoked for explaining this limit is that a too high sequence divergence between HLA molecules would reduce the T cell repertoire during thymus maturation (Lau et al 2015 ; Lenz 2011 ), in the same way to what has been proposed for a too high number of MHC molecules (Nowak et al 1992 ; Woelfing et al 2009 ).…”

“…This latest observation parallels a suggestion by Lau et al of an upper limit to sequence divergence generated by the DAA model at HLA class II locus, HLA-DRB1 (Lau et al 2015 ). One reason invoked for explaining this limit is that a too high sequence divergence between HLA molecules would reduce the T cell repertoire during thymus maturation (Lau et al 2015 ; Lenz 2011 ), in the same way to what has been proposed for a too high number of MHC molecules (Nowak et al 1992 ; Woelfing et al 2009 ). Some empirical evidence for this hypothesis comes from studies of mate choice in fish, where intermediate rather than maximum MHC sequence dissimilarity would be preferred (Forsberg et al 2007 ; Lenz et al 2009 ; Nowak et al 1992 ; Woelfing et al 2009 ).…”

Divergent Allele Advantage at Human MHC Genes: Signatures of Past and Ongoing Selection

“…The investigation of the DRB1 locus has been further extended by considering two distinct phylogenetic groups of alleles, denoted as group A and B ( Yasukochi and Satta 2014 ). The same pattern of increased pathogen recognition capacity was observed only for those alleles that in the phylogenetic tree cluster together with primate alleles forming a polyphyletic group (group B) ( Lau et al. 2015 ).…”

“…Phylogenetic analysis of the HLA-DRB1 gene has revealed two subgroups of allelic lineages: a human-specific monophyletic group (Group A), and a polyphyletic group with primates (Group B) ( Yasukochi and Satta 2014 ). It has been proposed that group A and B allele lineages have evolved with contrasting binding capacity, and only the alleles from the polyphyletic group B showed increased presentation of pathogen peptides with increasing sequence divergence ( Lau et al. 2015 ).…”

“…Such an association is predicted by the divergent allele advantage hypothesis ( Wakeland et al. 1990 ) and had previously been investigated only for the HLA-DRB1 gene in humans ( Lenz 2011 ; Lau et al. 2015 ).…”