Differential host response to a herpesvirus: Marek’s disease virus peptides on chicken MHC class II molecules are derived from only a few genes and illustrate a new mode of peptide binding

Viral diseases pose major threats to humans and other animals, including the billions of chickens that are an important food source as well as a public health concern due to zoonotic pathogens. Unlike humans and other typical mammals, the major histocompatibility complex (MHC) of chickens can confer decisive resistance or susceptibility to many viral diseases. An iconic example is Marek’s disease, caused by an oncogenic herpesvirus with over 100 genes. Classical MHC class I and class II molecules present antigenic peptides to T lymphocytes, and it has been hard to understand how such MHC molecules could be involved in susceptibility to Marek’s disease, given the potential number of peptides from over 100 genes. We used a new in vitro infection system and immunopeptidomics to determine peptide motifs for the two class II molecules expressed by the MHC haplotype B2, which is known to confer resistance to Marek’s disease. Surprisingly, we found that the vast majority of viral peptide epitopes presented by chicken class II molecules arise from only four viral genes, nearly all having the peptide motif for BL2, the dominantly-expressed class II molecule in chickens. We expressed BL2 linked to several MDV peptides, and determined one X-ray crystal structure, showing how a single small amino acid in the binding site causes a crinkle in the peptide, leading to core binding peptide of ten amino acids, compared to the nine amino acids in all other reported class II molecules. The limited number of potential T cell epitopes from such a complex virus can explain the differential MHC-determined resistance to MDV, but raises questions of mechanism and opportunities for vaccine targets in this important food species, as well as providing a basis for understanding class II molecules in other species including humans.


Introduction
The ongoing global pandemic of a coronavirus among humans highlights the enormous 55 challenge of viral disease, and the importance of the appropriate immune responses [1-3]. 56 The classical class I and class II molecules of the major histocompatibility complex (MHC) 57 play crucial roles in resistance to infection and response to vaccines, binding peptides for 58 presentation to thymus-derived (T) lymphocytes of the adaptive immune system as well as 59 natural killer (NK) cells of the innate immune system [4,5]. The importance of the classical 60 MHC molecules is underscored by their high level of allelic polymorphism, which is 61 mostly driven by molecular arms races with infectious pathogens [5]. However, the MHC 62 of humans and typical mammals is an enormous and complex genetic region encoding a 63 wide variety of molecules, with multigene families encoding class I and class II molecules 64 leading to strong genetic associations with autoimmune disease but relatively weak 65 associations with infectious diseases [4]. 66 In contrast to humans and other typical mammals, the level of resistance to many infectious 67 pathogens in chickens can be strongly determined by the MHC (that is, the BF-BL region 68 of the B locus), at least in part because the chicken MHC is much simpler, with single 69 dominantly-expressed class I and class II loci [6,7]. As a result of this simplicity, the 70 phenotypes are much clearer: either the dominantly-expressed MHC allele finds a 71 protective peptide to confer resistance or not, leading to strong genetic associations that are 72 easier to discover and dissect [7]. Moreover, the scale of viral challenges in the poultry 73 industry has been clear for decades, with many tens of billions of chickens each year beset 74 by a wide variety of poultry viruses [6], including the first coronavirus ever described as 75 such [8]. On top of the economic importance, zoonotic pathogens (including avian 76 influenza) have been a continuing concern for public health [9, 10]. Despite enormous 77 efforts in biosecurity, vaccination and genetic breeding, condemnation and slaughter of 78 huge numbers of infected chickens are relatively frequent [11]. 79 Among the economically-important diseases for which the chicken MHC is known to 80 determine resistance and susceptibility, Marek's disease (MD) caused by an oncogenic 81 herpesvirus was the first reported, is the best studied and remains an enormous burden for 82 the poultry industry, with continuing outbreaks despite routine vaccination [12][13][14][15][16]. Indeed, 83 current vaccines control disease but not transmission, leading to selection of more virulent 84 strains, which in turn have required more efficacious vaccines [17,18]. The virus 85 responsible for MD (MDV), in common with other herpesviruses, has a relatively large 86 genome with over 100 genes and a complex life cycle, so it is possible that many genes 87 contribute to resistance at different stages of infection, tumor growth and transmission [14, 88 16]. Several polymorphic genes located in the MHC have been proposed as candidates to 89 determine MD resistance, including the dominantly-expressed classical class I gene (BF2), 90 an NK receptor gene (B-NK), a gene with some similarities to mammalian butyrophilins 91 (BG1) and the classical class II B genes (BLB1 and BLB2) [19][20][21][22][23][24][25]. In comparison to the 92 MHC class I system, very little attention has been focused on chicken class II genes and 93 molecules [7,26]. 94 Mammalian class II molecules have been intensively studied, so many structural and 95 functional features are known in detail [27][28][29][30]. The heterodimer is composed of an α and a 96 β chain, each with two extracellular domains: membrane proximal immunoglobulin C-like 97 (Ig C) domains (α2 and β2 domains) and membrane distal domains composed of four β-4 98 strands forming a β-sheet surmounted by predominantly α-helical stretches (α1 and β1 99 domains). The domain organization, including the features of secondary structure and 100 location of intra-domain disulfide bonds, is similar to class I molecules. The membrane 101 distal domains together form a peptide-binding super-domain, with a core sequence from 102 the peptide binding across a groove, the T cell receptor recognizing the top, and the 103 dedicated chaperone DM interacting with key residues on the side for loading of 104 appropriate peptides. A glycosylation site near the end of the α1 domain is likely to be 105 involved in quality control during biosynthesis and peptide loading (as in class I, [31]), 106 while a loop in the β2 domain and part of the α2 domain bind the CD4 co-receptor that 107 contributes to T cell signaling (in the same place as CD8 interacts with class I, [30,32]).

108
As in human HLA-DR molecules, the chicken α-chain (encoded by the BLA gene) is nearly 109 monomorphic, with most residues on the top of the α1 domain identical with DRA 110 (although there is dimorphic position in the α-helix pointing up towards the T cells, and a 111 four amino acid insertion in one loop of the β-sheet) [33]. Virtually all of the variation 112 responsible for allelic polymorphism and thus for different peptide-binding specificities in 113 different chicken class II molecules is located in the β1 domain of the β chain, encoded by 114 either the BLB1 or BLB2 gene [34,35]. not. However, a major concern is how classical MHC molecules could confer susceptibility 118 to a virus such as MDV, since it is hard to imagine how a peptide conferring protection 119 would not be found among 100 viral molecules [19]. This conundrum is particularly an 120 issue for class II molecules for which the peptide motifs are relatively promiscuous 121 compared to class I molecules [28]. A significant barrier to examining this question has 122 been the low frequency of cells infected with MDV within chickens before tumors arise.

123
In this report, we use a novel culture system for bursal B cells [36], followed by 124 determination of the peptides bound to class II molecules by mass spectrometry (so-called 125 immunopeptidomics or MHC ligandomics, [37,38]). We identify the peptides bound to 126 class II molecules from the well-characterized MHC haplotype B2 known to confer 127 resistance to MD [13,39], with and without infection by the very virulent MDV strain RB-128 1B (also known as RB1B, 40) and the live attenuated virus strain CVI988 (widely used as 129 the Rispens vaccine, [41]). We then express the chicken class II molecule BL2*02 with clustering identified two nonamer (9mer) core motifs in all samples, but subsequent 152 structural evidence (detailed below) led to a re-analysis that showed one of these motifs to 153 be a decamer (10mer) (Fig 1A). The two motifs turn out to represent the two classical class  Table 1, Table S1). However, multiple species of a single epitope and/or multiple 174 peptide epitopes were found in several infected samples for four MDV proteins: 175 glycoprotein H (gH, encoded by the MDV034 gene), UL43 tegument protein (MDV056), 176 gI (MDV095) and gE (MDV096). Five peptide epitopes were found for gH, two of which 177 had multiple species, of which one was found in all five samples. The UL43 tegument 178 protein had one peptide epitope with 16 species, some of which were found in two to four 179 samples. For gI, one peptide epitope was found with three species, one of which was found 180 in four samples. For gE, two peptides were found, one of which had 18 species, four of 181 which were found in all five samples.

182
Most of these peptide species had the decamer motif, including those from three of the four 183 MDV proteins with abundant peptides. From the analysis by Gibbs clustering, eight peptide 184 species from five epitopes fit the nonamer BLB1 motif, 45 species from ten epitopes fit the 185 decamer BLB2 motif, nine species from six epitopes fit both motifs (and thus might 186 actually contain two epitopes), and two species each from one epitope didn't fit either motif 187 (Table 1). 188 Soluble BL2 molecules from the B2 haplotype were expressed in insect cells using  The overall structure of a chicken class II molecule is similar to those of mammals 213 Of the four soluble BL2 molecules from the B2 haplotype linked to MDV peptides that 214 were expressed, the one with a peptide from gE formed crystals that diffracted to 215 sufficiently high resolution to determine a structure (Fig 3A-C haplotype known to confer susceptibility to MD [13,44].

219
Overall, the structure of BL2 from B2 (BL2*02) is extremely similar to BL2 from B19 220 (BL2*19) and to the iconic mammalian class II molecule HLA-DR1 (Fig 3D, E), with root 221 mean square deviations for Cα atoms of 0.684 and 0.878 Å, respectively (Table S3). Most  BL2 from B2 illustrates a new mode of peptide binding for MHC class II molecules 246 The way in which peptides bind to mammalian class II molecules has been studied in detail, 247 showing that a nonamer core binds as a polyproline II helix, with main chain atoms forming structures that can be recognized by T cells [27,[45][46][47]. In fact, the P10 side chain of a 254 peptide with a nonamer core has been reported [48] in one structure to bind onto a 255 polymorphic shelf outside of the groove (Fig S5).

256
In stark contrast to all reported mammalian class II structures and to BL2*19, the BL2*02  tissues [26], has an unprecedented decamer core motif, with large aliphatic residues at P1 320 and P4, small and even acidic residues at P5, and large hydrophobic residues at P10.  (Table S1), but these genes are expressed early after infection of 329 B cells [49]. There are many other proteins present in the virion, and many other proteins  The overall structure of the chicken class II molecule is similar to that of mammals, which 341 is perhaps not such a surprise given the high level of sequence identity, particularly in the 342 BLA α chain [33]. Almost all the key features and residues involved in interactions with  explain the strong genetic associations with infectious pathogens [19][20][21][22].

366
No matter what the contribution of chicken class II molecules to differential resistance to   specificity, and oxidation of methionine residues was allowed as dynamic modification.

449
Precursor mass tolerance was set to 5 ppm, and fragment mass tolerance to 0.02 Da. False 450 discovery rate (FDR) was estimated using the Percolator node (67) and was limited to 5%.

505
Diffraction data was collected remotely on the I04-1 beamline (Diamond Light Source, 506 Oxford UK) at a wavelength of 0.978 Å. Data reduction and scaling were performed using 507 XDS (71) and SCALA (72). The crystal of the BL2*02 belongs to the C 1 2 1 space group, 508 and the structure was solved by basic molecular replacement deploying Phaser from the 509 CCP4i package (73) using HLA-DR1 (4X5W) as the search model (74). Further rounds of 510 manual model building and refinement were done using Coot (75) and Phenix (76). Further 511 details about collection and refinement are shown in