Amino acid signatures in the HLA class II peptide-binding region associated with protection/susceptibility to the severe West Nile Virus disease

Abstract

The MHC class II region in humans is highly polymorphic. Each MHC molecule is formed by an α and a β chain, produced by different genes, creating an antigen-binding groove. In the groove there are several pockets into which antigens anchor and fit. The affinity of this fitting determines the recognition specificity of a given peptide. Here, based on our previous results about the association of MHC class II with the WNV disease, we examined the role of the binding pockets of HLA-DPA1, -DQA1 and-DRB1 in the severe form of the disease. In HLA-DQA1, variants in all pockets 1, 6 and 9 were found to be associated with either protection and/or susceptibility to neuroinvasion caused by WNV. Similarly, pockets 7, 9 and 10 in HLA-DRB1 were associated with severe disease. Protein modeling of these molecules revealed structural and functional differences among alleles with opposite roles concerning the development of the disease. Different amino acids in positions α52 and α66 (HLA-DQA1) significantly influenced the peptide binding while DYWLR/EFA combination (HLA-DRB1) was associated with neuronal damage. Further studies could help us understand the selectivity of pocket variants in order to create suitable peptides for an effective response.

Fig 1. MHC class II molecules are heterodimers consisting of α (green) and β (orange) chains.

The antigen binding groove is formed by two alpha-helices above a beta-sheet floor. Antigens (red) fit in the cleft and stabilized by electrostatic interactions and/or hydrogen bonds to anchor peptides of the groove (regions with light shades). The structure being illustrated is based on the structure with PDB ID: 2NNA [12].

Fig 2. The amino acid sequence of exon 2 (partial) of the alleles that were identified in HLA-DPA1, -DQA1 and DRB1 in our WNV cases.

The residues that form pockets P1, P4, P6, P7, P9 and P10 have been noted accordingly. Positions that have the same amino acid with the first sequence in other alleles are shown with dots.

Fig 3

The distribution of alleles and pocket variants in WNND (blue) and WNF (light red). Next to each group are listed the alleles that were unique for the group. In the boxes below, the alleles/pocket variants that were associated positively or negatively with neuroinvasion are shown.

Fig 4. Graphical representation of DQA1/DQB1 molecules.

α chain is shown in red and β chain in green. In the center of the groove, the peptide (chain C from PDB ID: 2NNA) is shown in cyan sticks. The residues of the heterodimer that are in the neighbor of the peptide are shown in CPK colored sticks (DQA1*01:02 in yellow and DQA1*03:01 in magenta). Amino acids from chain β that form H-bonds are shown in salmon. Hydrogen bonds are shown in yellow dotted lines. A. Heterodimer DQA1*01:02-DQB1*03:01. B. Heterodimer DQA1*03:01-DQB1*03:01. All neighbor residues to the peptide are shown (in radius 4 Å). Only the amino acids that form hydrogen bonds are labeled. C. Selected residues that have different orientation and/or show different interaction with the peptide between the two structures (residues 52, 55, 63, 69, 70 and 73). Variable residues of the alleles that were found in close proximity to the peptide are also shown (positions 9, 53, 54 and 59).

Fig 5. Graphical representation of DRA1/DRB1 molecules.

α chain is shown in red and β chain in green. In the center of the groove, the peptide (chain D from PDB ID: 2WBJ) is shown in cyan sticks. The residues of the heterodimer that are in the neighbor of the peptide and belong in P7, P9 and P10 are shown in CPK colored sticks (DRB1*11:04 in yellow and DRB1*14:04 in magenta). Amino acids from chain α that form H-bonds are shown in salmon. Hydrogen bonds are shown in yellow dotted lines. A. Heterodimer DRA1*01– DRB1*11:04. B. Heterodimer DRA1*01– DRB1*14:04. All neighbor residues to the peptide are shown (in radius 4 Å). Only the amino acids that form hydrogen bonds are labeled. C. Selected residues that have different orientation and/or show different interaction with the peptide between the two structures (residues 28, 56, 60, 61 and 82). Variable residues of the alleles that were found in close proximity to the peptide are also shown (positions 57, 67 and 71).