OR40 Integrative analysis of HLA NGS genotyping data and single antigen bead assays for mapping serologic epitopes

Abstract

Aim Defining HLA serologic epitopes contributing to antibody-mediated rejection (AMR) is a key to untangle the molecular mechanisms of epitope recognition. We proposed a reverse engineering strategy to define new epitopes from the protein sequences of transplant recipients sharing the same donor specific antibody, and developed a set of tools implementing the idea. Methods The hypothesis is that amino acid (AA) polymorphisms shared between the recipient and the donor will not be immunogenic and therefore cannot lead to the generation of antibodies. Without other assumptions, we developed a program which: 1. retrieves all AA triplets of the recipient by sliding a window of size three with a step of size one starting from the first AA on the full AA sequences of all the alleles, 2. reads the AA sequences of the donor with a sliding window in the same manner and outputs the allele, the position, and the triplet itself whenever an AA triplet is not in the triplet set of the recipient. We call the results unbiased donor specific AA triplets because we do not restrict the comparison within the same class nor to just the mature protein. We also developed web-based functions in the 17th IHIWS database to collect the unbiased triplets so that we could group the transplant pairs with the same DSAs and find shared unbiased triplets within the groups as potential epitopes. Please see the figure for an illustration. Results To verify the strategy, we collected the NGS HLA typing combined with the donor specific antibodies identified by Luminex® IgG and Bio-C1q SAB for 206 renal and 133 cardiac transplant recipients with AMR from nine labs. We ran the program and imported the results into the 17th IHIWS database, further mapping the epitopes onto the 3D protein structure for reference. Conclusions We proposed and verified an unbiased reverse engineering strategy to define new serologic epitopes.