In chickens and humans, classical class I molecules of the major histocompatibility complex (MHC) can have a hierarchy of correlated properties, including cell surface expression and peptide repertoire. Chicken BF2 alleles that are less well-expressed on the cell surface and bind a very wide range of peptides are expressed by MHC haplotypes that confer protection from a variety of economically-important infectious diseases, while certain human HLA-B alleles that are well-expressed and bind a narrow range of peptides lead to slow progression from HIV infection to AIDS. Understanding the impact of these promiscuous generalists and fastidious specialists is of considerable interest. The promiscuous BF2 molecule from the chicken B21 haplotype, BF2*21:01, binds a wide range of peptides by remodelling the peptide-binding site, allowing co-variation of the anchor residues at peptide positions P 2 and P c-2, and binding of an anchor residue at Pc. By using in vitro refolding assays with peptides and peptide libraries, determining thermostability and crystal structures, and analysing a chicken B21 cell line by immunopeptidomics, we found that BF2*21:01 will accommodate many possible combinations at P 2 and P c-2, as well as several hydrophobic amino acids at P c. However, marked preferences for particular peptide lengths, particular amino acids at the three anchor residues, combinations of amino acids at P 2 and P c-2, and amino acids at P 3 and P c-3 affecting stability lead to high frequencies of major peptides while still allowing the possibility of presenting a wide peptide repertoire. These simplifying principles may eventually allow predictions of pathogen peptides with stable binding for this iconic promiscuous class I molecule, as well as providing the data for more sophisticated peptide prediction methods.
The major histocompatibility complex (MHC) is a genetic region defined by the presence of a few highly polymorphic genes encoding classical class I and class II molecules, originally identified as transplantation antigens but now known to play crucial roles in the immune response to pathogens and tumours. The human MHC is an enormous genomic region with much recombination among hundreds of genes with a vast array of functions, with multigene families of classical class I and class II genes, strongly associated with autoimmunity and allergies but not so strongly with infectious pathogens. In contrast, the chicken MHC is much smaller and simpler, with single dominantly-expressed class I and class II genes, and strong associations of MHC haplotypes with resistance and susceptibility to certain economically-important pathogens.
In trying to understand the strong associations of the chicken MHC haplotypes with infectious disease, we discovered a hierarchy of alleles of the dominantly-expressed class I molecule, BF2. The original insight came from flow cytometry: erythrocytes from the B21 haplotype conferring resistance to the oncogenic herpesvirus that causes Marek’s disease have ten-fold less class I on the cell surface than from the susceptible B4, B15, B12 and B19 haplotypes. Since then, a suite of properties was found to correlate with this hierarchy, the most important being the peptide repertoire, defined as the range of different peptide sequences bound. An inverse correlation between the level of cell surface expression and the breadth of peptide binding was found, with less well-expressed alleles that have wide peptide repertoires correlating with resistance to Marek’s disease and other economically-important pathogens.
A similar inverse correlation between predicted peptide repertoire and cell surface expression was found for some human class I molecules, but a disease association that was the opposite: well-expressed “elite controller” alleles with narrow peptide repertoires correlated with slow progression from HIV infection to AIDS. This finding led to the hypothesis of generalist and specialist class I alleles, with low-expressing promiscuous alleles generally protecting from many pathogens while high-expressing fastidious alleles bound special peptides to protect against particular pathogens. A hierarchy of tapasin-dependence and ease of in vitro refolding for certain human class I alleles was found to correlate with cell surface expression and predicted peptide repertoire and eventually shown to associate with slow progression once elite controllers were removed from analysis, thus supporting the generalist-specialist hypothesis. In addition, a hierarchy of human class II alleles based on predicted peptide repertoire was reported.
There are presumably various biochemical mechanisms leading to this hierarchy of class I alleles, including specificity of peptide transport by TAPs in chickens, dependence of class I alleles on tapasin (and perhaps TAPBPR) in humans, and specificity of peptide binding by the class I alleles themselves. Several modes of promiscuous binding have been reported for dominantly-expressed chicken class I molecules.
Despite requiring three anchor residues, BF2*21:01 achieves a wide peptide repertoire by remodelling the peptide-binding site in a previously unprecedented manner that involves co-variation between the amino acids at positions P 2 and P c-2, as shown by structures bearing six peptides with very divergent sequences. Since this iconic chicken class I molecule is so frequent across the world and so important for resistance to economically-important viral diseases, we wished to better understand the rules by which peptides bind to BF2*21:01, starting with refolding the class I molecule with different peptides and peptide libraries, determining thermostabilities, determining structures and creating models to illustrate some of the results, and finishing with immunopeptidomics of class I molecules from a B21 cell line.
Gas phase sequencing of peptides eluted from B21 class I molecules and separated by HPLC gave a wide range of sizes and sequences with no obvious anchor residues, in stark contrast to the class I molecules from B4, B12, B15 and B19 which gave mostly octamers with clear peptide motifs. The heterogeneity in length of peptides from blood and spleen cells is likely to have been due to proteolysis, since the single preparation from a B21 chicken cell line gave only 10mers and 11mers, including TNPESKVFYL from which the octamer PESKVFYL from blood samples (that failed to refold in vitro) was apparently derived. The wide range of sequences were eventually understood to mean that BF2*21:01 molecules would accommodate a variety of anchor residues at three positions, with co-variation between the anchor residues at P 2 and P c-2, as found by refolding and crystallography.
