Framework for analyzing MAE-derived immunopeptidomes from cell lines with shared HLA haplotypes

Abstract

Background: The goal in vaccinology is to identify candidate antigens for clinical trials that will elicit an immune response for a significant portion of the target population. Unfortunately, promising data generated at the preclinical level often cannot be replicated in larger sample sizes. The goal of this project was to develop a methodology for processing MAE-generated data to identify MHC epitopes, minimize non-specific contaminants, find binding motifs, and utilize genetic connections among donors to determine which peptides were presented by specific MHC alleles.

Results: Our approach demonstrated that mild acid elution of peptides from seven consanguineous B-lymphocyte lines accurately reflects the HLA genotypes within family members, highlighting the specificity of MAE. Additionally, the data successfully reproduced known MHC binding motifs and partially deconvoluted the originating HLA alleles of the epitopes.

Conclusions: These findings suggest that our approach could be applied to numerous cell lines globally to evaluate a wide array of HLA haplotypes. This may help to reveal candidate vaccine antigens that induce immune protection for a wider population.

Fig 1. Using MAE to isolate MHC I and MHC II ligands.

(A) Sequence logo (top) generated from epitopes that were found in the WT cell but not in the HLA-A*02 knockdown closely matches the HLA-A*02:01 motif from naturally presented ligands (bottom, NetMHCpan). (B) A volcano plot of the log2 fold change of proteins in a whole cell lysate of WT (LBL-721) and mutant (LBL-721-174) cell lines. Proteins associated with the MHC I (red) and MHC II (blue) antigen presentation machinery are highlighted by colored dots. Venn diagram of epitopes in the (C) 9-11 residues long MHC I and (D) 12-20 residues long category found in the WT and mutant cells.

Fig 2. Benchmarking epitope sequences derived from peptide data.

(A) A flowchart describing how MAE-derived peptide data are processed, resulting in sequences that are either rejected or retained for further analysis. (B) These sequences are then benchmarked by using NetMHCpan (for epitopes of 9-11 amino acids in length) or NetMHCIIpan (for epitopes of 12 + amino acids in length) that tag entries either as a specific binder (weak or strong) or a non-binder against any of the queried MHC alleles, thereby allowing each MAE-derived sequence to be categorized as a true positive (TP), false negative (FN), false positive (FP), or true negative (TN).

Fig 3. HLA genetic relatedness of B-lymphocytes from a seven-member donor family.

(A) HLA alleles of each family member are shown. (B) Highly similar clustergrams were generated from E_i values for both epitopes in the 9-11 mer group (top, MHC I) and 12-20 mer group (bottom, MHC II) that mirror the HLA genetic relatedness of the family members. Non-numbered individuals refer to the parents and numbered individuals are the children based on birth order.

Fig 4. Strategy for analyzing consanguineous cell lines.

(A) Analyzing the E_i profiles of epitopes presented by theoretical cell lines X, Y, and Z that have partial overlap of HLA-A, HLA-B, and HLA-C genotypes can be used to trace the origin of a given epitope back to its originating allele. (B) A flow chart of how MAE-derived peptides from related cell lines can be processed together through a series of steps (condensing peptides into epitopes, hierarchical clustering, and multiple sequence alignment (MSA)) to predict MHC binding motifs.

Fig 5. Deriving HLA allele-specific consensus sequences from MAE-derived peptide data.

Theoretical Ei profiles for (A) MHC I and (B) MHC II genes for all seven members of the donor family, with non-numbered individuals referring to the parents and numbered individuals are the children based on birth order. Grey cells indicate E_i > 0 while white cells refer to E_i = 0. Positions 1-9 are predicted to be the region that fits into the MHC binding grove, with typical anchor positions highlighted (shaded circles). Shown are examples of (C) MHC I and (D) MHC II nodes that contain aligned epitopes, visualized as sequence logos, with similar E_i profile family members. Epitopes in these nodes were aligned and visualized as sequence logos. Corresponding known HLA motifs from NetMHCpan and NetMHCIIpan derived from a database of naturally bound ligands [29] are shown as evidence of this technique’s ability to reproduce them. Positions 1-9 are predicted to be the region that fits into the MHC binding grove, with typical anchor positions highlighted (shaded circles).

Fig 6. Different ways to display a sequence logo.

One example node with epitopes (I) displayed as a sequence logo, as a single line output with (II) relative Amino Acid Positional Score (S) in grayscale (white = 0%, black = 100% or if S > 1) for amino acids that exceed threshold R (bold), or amino acids that pass threshold only when “grouped” by their physiochemical properties (not bold). (III) Shows only the amino acids with the top 4 highest S and also pass threshold, and (IV) similar to (III) but also displays non-significant positions contained within the consensus are represented by dashed lines.

Fig 7. Effects of adding more cell lines to the epitope clustering approach.

This graph demonstrates that for each cell line (available at the International Histocompatibility Working Group) added to the epitope clustering approach, it increases the possibility to trace a ligand back to its origin MHC by approximately one more allele.