Mass Spectrometry of Human Leukocyte Antigen Class I Peptidomes Reveals Strong Effects of Protein Abundance and Turnover on Antigen Presentation

HLA class I molecules reflect the health state of cells to cytotoxic T cells by presenting a repertoire of endogenously derived peptides. However, the extent to which the proteome shapes the peptidome is still largely unknown. Here we present a high-throughput mass-spectrometry-based workflow that allows stringent and accurate identification of thousands of such peptides and direct determination of binding motifs. Applying the workflow to seven cancer cell lines and primary cells, yielded more than 22,000 unique HLA peptides across different allelic binding specificities. By computing a score representing the HLA-I sampling density, we show a strong link between protein abundance and HLA-presentation (p < 0.0001). When analyzing overpresented proteins – those with at least fivefold higher density score than expected for their abundance – we noticed that they are degraded almost 3 h faster than similar but nonpresented proteins (top 20% abundance class; median half-life 20.8h versus 23.6h, p < 0.0001). This validates protein degradation as an important factor for HLA presentation. Ribosomal, mitochondrial respiratory chain, and nucleosomal proteins are particularly well presented. Taking a set of proteins associated with cancer, we compared the predicted immunogenicity of previously validated T-cell epitopes with other peptides from these proteins in our data set. The validated epitopes indeed tend to have higher immunogenic scores than the other detected HLA peptides. Remarkably, we identified five mutated peptides from a human colon cancer cell line, which have very recently been predicted to be HLA-I binders. Altogether, we demonstrate the usefulness of combining MS-analysis with immunogenesis prediction for identifying, ranking, and selecting peptides for therapeutic use.The highly polymorphic Human Leukocyte Antigen class I (HLA-I)¹ genes are encoded by three loci (HLA-A, B, and C) in a gene-rich region on chromosome 6. They produce up to six unique cell surface receptors that bind and present the so-called HLA class I peptidome, which consists of peptides derived from proteolysis of intracellular proteins. Their function is to reflect the health state of the body''s cells to CD8+ cytotoxic T cells. During thymic maturation T cells that react to self-peptides are eliminated (1), leaving T cells with the capability to recognize peptides from viruses and bacteria. This recognition is interpreted as a danger signal, leading to removal of infected cells. Transformed, preneoplastic and cancer cells also tend to display atypical self-peptides from mutated or excessively expressed self-proteins, known as tumor associated antigens (TAAs). Although HLA-I molecules are indispensable in prevention of disease, they also pose a substantial health problem by causing allergies (2), life-threatening autoimmune diseases (3), and the often fatal rejection of donor organs because of recognition of both major and minor histocompatibility antigens (4).Finding the rules for peptide generation and selection is regarded as the most important open issue in the field of HLA-I biology by leading experts (5). Although the antigen presentation pathway is well characterized, it is still unclear how basic properties such as protein abundance, turnover, and subcellular localization influence and shape the HLA-I presented peptidome (6–10). One expectation is that protein abundance should correlate with presentation (11), but previous studies have reported conflicting and contradicting results that mostly argue against a strong link (6, 7, 10, 12, 13). It is also not fully understood why only some HLA-sampled self-peptides from cancer antigens spontaneously activate T cells, whereas others do not.The majority of HLA-I peptides are derived from proteasomal degradation (5). Although the proteasome generates an excess of peptides, only some have the required sequence motifs for HLA binding, resulting in a selective sampling of available peptides (14). The presented peptides are typically nine amino acids long, but the length can range from eight to 15. The high degree of genetic variance of HLA-I receptors translates into allele-specific peptide-binding motifs defined by anchor positions, which are usually the second and the last positions in a peptide (15). Each cell has around 200,000 cell-surface-expressed HLA complexes, which bind about 10,000 unique peptide sequences (16). The affinity of a peptide toward the presenting HLA molecule does not correlate strongly with its immunogenicity, and neither does the number of presented HLA complexes (17). Instead, the most robust predictor of peptide immunogenicity appears to be the number of potential reactive T-cell clones (17–19).The longer the source protein, the higher the chances it will contain sequences that fit to a certain HLA motif, which would inflate the representation of longer proteins regardless of biological role. Furthermore, some HLA-I peptide sequences can be mapped to multiple proteins, potentially causing a problem in determining the number of observed HLA peptides per protein (13). This illustrates that careful accounting of the potentially and actually presented HLA peptides is important in properly delineating trends in propensity of peptide presentation.In cancer immunotherapy, T cells can be directed against tumors, based on the pattern of cancer associated HLA peptides. Therefore, there is great interest in determining the identity of these immunogenic peptides. Bioinformatic methods that attempt to predict HLA peptides of cancer proteins of interest are easily accessible and most commonly used. They typically score sequences with respect to proteasomal degradation, transport into the ER via the transporter associate with antigen processing (TAP) and binding to different HLA-I alleles (20). However, their precision success is modest (21, 22). The second approach is to directly capture the naturally presented peptides using mass spectrometry; however, this requires the relevant biological sample and sophisticated instruments and workflows, which have become accessible only recently for large-scale work (23–28). Although identification of cancer associated HLA peptides by MS, if performed stringently, establish the in vivo existence of the peptide, it still does not guarantee that it will elicit a potent T-cell response, which is required for further development into therapeutics (29). Therefore, like in the case of in silico predicted peptides, the immunogenicity of the peptides must in any case be tested empirically.We here present a rich and high confidence HLA-I peptidome, established by applying state-of-the-art mass-spectrometric techniques on a collection of seven cell lines. We investigate how abundance affects the propensity of proteins to be presented as measurable HLA peptides and whether or not there are specific protein classes that are overrepresented even independent of abundance. Likewise, we explore how to use in silico immunogenicity tools on the set of identified HLA peptides from cancer-associated proteins, with a view to select vaccine candidates.