Data-Independent Immunopeptidomics Discovery of Low-Abundant Bacterial Epitopes.

Mass spectrometry-based immunopeptidomics is a powerful approach to uncover peptides presented by human leukocyte antigen (HLA) molecules that can guide vaccine design and immunotherapies. While data-dependent acquisition (DDA) has been the standard for navigating through the complexity associated with nonenzymatic immunopeptide database searches, data-independent acquisition (DIA) is increasingly adopted in immunopeptidomics research. In this work, we compare diaPASEF to conventional ddaPASEF in terms of global immunopeptidome profiling and bacterial epitope discovery of the model intracellular pathogen Listeria monocytogenes. We show that DIA spectrum-centric workflows that search pseudo-MS/MS spectra complement DDA analysis by uncovering additional human and bacterial immunopeptides. Furthermore, we leveraged DIA-NN for generating and searching proteome-wide predicted HLA class I peptide spectral libraries, scoring approximately 150 million immunopeptide peptide precursors. This approach outperformed other spectrum-based methods in the identification of MHC class I peptides and recovered low-abundant peptide precursors missed by other methods. Taken together, our results demonstrate how both DIA spectrum- and peptide-centric immunopeptidomics analysis are promising strategies to identify low-abundant immunopeptides.