Sensitive direct detection of cancer antigens via user-defined peptide spectral libraries

ABSTRACT: Mass spectrometry (MS) enables precise identification of HLA-bound peptides, but challenges remain in achieving sensitive, accurate, and reproducible detection of rare yet clinically relevant antigens, including predicted tumor neoantigens. Reliable reporting of such rare peptides often requires confirmation with synthetic peptides, which are costly and time-intensive to produce or through prediction of chemical moieties, a strategy that has only recently been expanded to non-tryptic HLA peptides. To generate user-defined high-quality peptide libraries, containing up to 20,000 unique sequences, we introduce Pepyrus, a robust peptide expression system in E. coli. We demonstrate that Pepyrus-generated peptides exhibited close to identical fragmentation patterns, retention times and ion mobility distributions to conventional chemically synthesized peptides. With a HLA-specific data independent acquisition strategy, we recovered >75% of the expected sequences per single injection even at >10,000 peptide Pepyrus library. Pepyrus-assisted identification of rare neoantigens from 0.1 fmol in a complex background was feasible. Personalized Pepyrus libraries facilitated the identification of clinically relevant peptides from melanoma patients, several of which were previously undetected using MS. The customization of Pepyrus allows rapid generation of patient- or disease-specific peptide libraries that enable the confident identification of rare HLA peptides from immunopeptidomics data as well as the generation of large training datasets to improve spectrum, retention time, and ion mobility prediction tools.

INSTRUMENT(S): timsTOF Ultra

ORGANISM(S): Homo Sapiens (ncbitaxon:9606)

SUBMITTER: Steven A. Carr  

PROVIDER: MSV000096555 | MassIVE | Thu Nov 28 11:57:00 GMT 2024

SECONDARY ACCESSION(S): PXD058389

REPOSITORIES: MassIVE