There are many strong associations of the chicken MHC with infectious diseases, but the strong associations for the human MHC are with autoimmune diseases. Over the last 20 years, we have provided a molecular explanation for this functional difference, which is rooted in the evolutionary history of the MHC, particularly changes in genomic organisation that affect the interaction between proteins of the antigen presentation pathway. The result is that humans and most placental mammals express a multi-gene family of MHC molecules at high levels, so that each MHC haplotype confers more-or-less resistance to most pathogens, leading to weak genetic associations. In contrast, each chicken MHC haplotype expresses only one class I molecule at a high level, and the properties of this “dominantly-expressed” MHC molecule in large part determines the immune response, leading to strong genetic associations. For a couple of viral infections and vaccinations, we have shown that the peptide-binding specificity of the particular MHC class I molecule can explain life or death of particular chickens.

We have found that the dominantly-expressed class I molecules from some MHC haplotypes (like B4/B13, B12, B15 and B19) have peptide-binding specificities that are really “fastidious”, requiring particular amino acids as “anchor residues” in at least three places within an eight amino acid peptide. However, for other haplotypes (like B2, B14 and B21), the dominantly-expressed class I molecules have peptide-binding specificities that are really “promiscuous”, binding an astonishing variety of peptides. The structures of these promiscuous class I molecules explain this breadth of peptide binding. These two kinds of haplotypes differ in a suite of properties, including cell surface expression level, peptide-transporter (“TAP”) specificity and resistance to the oncogenic herpesvirus that causes Marek’s disease.

Our current hypothesis is that the promiscuous class I molecules allow recognition by a much wider variety of T cells, resulting in a greater and more effective immune response, more like class II molecules than human class I molecules. Such promiscuous class I molecules might be considered as “generalists” that confer resistance to many pathogens, with the more fastidious “specialists” being strongly selected for particular new or deadly pathogens.

However, such wider recognition of promiscuous class I molecules would lead to a great deletion of T cells during negative selection in the thymus, which would not be beneficial. Our current hypothesis is that the peptide-binding repertoire is inversely correlated with cell surface expression exactly to balance deletion of T cells in the thymus in order to ensure an optimal T cell response in the periphery.
Such properties are more difficult to discern in mammals, which generally have a multigene family of well-expressed class I molecules. However, we have found that certain class I alleles in humans also show an inverse correlation between the peptide-binding repertoire and cell surface expression. In particular, HLA-B*57:01 and 27:05 are fastidious high expressing class I molecules conferring long-term non-progression from HIV infection to AIDS (a new and dangerous pathogen for humans), whereas HLA-B*07:02 and 35:01 are more promiscuous and confer rapid progression. Thus, the simple MHC of chickens has allowed us to discover fundamental properties of MHC class I molecules that eluded the biomedical community for decades.