Pathogen selection drives nonoverlapping associations between HLA loci

Abstract

Pathogen-mediated selection is commonly invoked as an explanation for the exceptional polymorphism of the HLA gene cluster, but its role in generating and maintaining linkage disequilibrium between HLA loci is unclear. Here we show that pathogenmediated selection can promote nonrandom associations between HLA loci. These associations may be distinguished from linkage disequilibrium generated by other population genetic processes by virtue of being nonoverlapping as well as nonrandom. Within our framework, immune selection forces the pathogen population to exist as a set of antigenically discrete strains; this then drives nonoverlapping associations between the HLA loci through which recognition of these antigens is mediated. We demonstrate that this signature of pathogen-driven selection can be observed in existing data, and propose that analyses of HLA population structure can be combined with laboratory studies to help us uncover the functional relationships between HLA alleles. In a wider coevolutionary context, our framework also shows that the inclusion of memory immunity can lead to robust cyclical dynamics across a range of host-pathogen systems.infectious disease | major histocompatibility complex | mathematical model | human evolution | population genetics H LAs, found on the surface of all nucleated cells, present pathogen peptides to T lymphocytes and are thus a keystone of adaptive immunity. Demonstrable associations of particular HLA alleles with resistance or susceptibility to severe disease (1, 2) underscore the importance of their role in protection against death from infection. The genes encoding HLAs are found in the 3.6-Mb-long MHC on chromosome 6 and are distinguished by their exceptional polymorphism (3), which is likely the result of selection from pathogens (4-6) Despite this enormous diversity, most human populations are dominated by a relatively small number of combinations of the alleles present at the class I HLA (A, B, C) and the principal class II HLA (DP, DQ, and DR) loci (7-12). Here we present a coevolutionary model demonstrating that pathogen selection can drive such long-term, long-range associations between HLAs. We show that this mechanistic process can be distinguished from other evolutionary effects by virtue of generating a higher degree of nonoverlap between HLA repertoires than might be expected under founder effects or hitchhiking.A Multilocus Model for Host-Pathogen Coevolution with Allele-Specific Adaptive Immunity We first explored the properties of a deterministic epidemiological model (Methods) in which (i) the pathogen population was represented by four potential strains defined by two antigenic loci containing alleles (a, b) and (x, y), respectively, and (ii) we defined within a diploid host, alleles (A,B) and (X,Y) at two linked "recognition loci" (i.e., HLA loci), each only capable of responding to the corresponding parasite allele (or epitope) given in lowercase above. We assumed that immunity developed in an allele-specific manner conferring complete protectio...

Excess of Deleterious Mutations around HLA Genes Reveals Evolutionary Cost of Balancing Selection

“…For instance, the well-documented COX haplotype covers the entire MHC region and has a population frequency of about 10% in Northern Europe (Stewart et al 2004). Epistasis among different HLA genes, for instance due to advantageous allele combinations may select against recombination and maintain such long-range haplotypes (Penman et al 2013).…”

Diversity and the maintenance of sex by parasites

“…Theoreticians often work with limited diversity in their models as this reduces the complexity of the system, but our results show that the number of loci involved in specificity is crucial for the maintenance of sex. The genetic underpinnings of host-parasite interactions are increasingly being uncovered in natural systems, many of which appear to be governed by multiple loci (Jones & Dangl, 2006;Scanlan et al, 2011;Penman et al, 2013;Barribeau et al, 2014). When discussing multiple loci and the evolution of sex, it is important to distinguish between loci that are under selection due to interspecific interactions (RQH) and loci that may negatively affect fitness due to the presence of deleterious mutations.…”

Parasitic Castration Promotes Coevolutionary Cycling but Also Imposes a Cost on Sex

“…Persistent fluctuations in host and parasite gene frequencies, often referred to as “coevolutionary cycling,” are of special interest to evolutionary biologists as they may help to resolve prominent questions such as why certain genes are highly polymorphic (e.g., those involved in the major histocompatibility complex; Penman et al. ) and why many organisms reproduce sexually. The evolution of sex has received considerable attention from both theoreticians and empiricists in the context of the Red Queen hypothesis (RQH), which posits that costs associated with sex (e.g., the twofold cost of males, sexual conflict, etc.)…”