Lost in Post-Translation, Found by Mass Spectrometry: Identification of Novel Biomarkers and Biologic Insights in Lymphoma Pathogenesis

Malignant lymphomas are neoplastic clonal expansions of B, T, or NK cells and are the seventh most common cause of cancer deaths. The majority (70-80%) of lymphomas that occur in the United States and Europe are non-Hodgkin lymphomas (NHL) while approximately 20 to 30 percent are classical Hodgkin lymphomas (CHL). Multi-parameter evaluation of protein expression is an important component of laboratory-based diagnosis and sub-classification of malignant lymphoma. The protein expression repertoires of lymphoma subtypes have not been well characterized and compendia of proteins expressed by lymphoma cells would constitute a valuable resource for tumor biomarkers for diagnosis and therapy. Liquid chromatography (LC) combined with high mass resolution tandem mass spectrometry (MS/MS) has emerged as a pivotal approach for large-scale analysis of proteins from complex mixtures. Importantly, the detection of proteins with low-stoichiometry, post-translational modifications requires enrichment strategies to isolate the proteins of interest. Strategies that exploit the unique biochemical characteristics of critical post-translational modifications such as phosphorylation and glycosylation can be used to specifically enrich for proteins containing those modifications. Tandem mass spectrometry is then used in an unbiased manner to identify and quantify the modified proteins as well as localize the specific amino acid residues bearing the modifications. Protein glycosylation is the most common post-translational modification that typically occurs by covalent modification of serine, threonine (O-glycosylation), or asparagine residues (N-glycosylation). In particular, N-glycosylation is prevalent in proteins that are destined for localization to membrane or intended for secretion into the extracellular environment. Thus these N-glycosylated proteins constitute an untapped reservoir of potential cellular biomarkers in extracellular fluids such as blood and urine. Indeed, many diagnostic protein biomarkers currently in clinical use are N-glycoproteins. These include many cluster of differentiation (CD) proteins that define hematopoietic sub-populations as well as hematologic neoplasms. Furthermore, many of these proteins represent important targets for therapeutic antibodies and immunotoxins. Enrichment techniques based on glycoprotein affinity for lectins, or linkage by specific chemistry of the sugar moiety to solid surfaces followed by their release using peptide N-glycosidase F (PNGase F) have been successfully employed for subsequent LC-MS/MS identification of N-glycosylated proteins. Using N-glycosylation enrichment strategies followed by LC-MS/MS, we catalogued the N-linked glycoproteins expressed in several human malignant lymphoma categories. We identified many known and several previously unidentified proteins with restricted expression in different lymphoma subtypes. Unsupervised hierarchical clustering based on N-glycoproteomic data classified primary tumors and cell lines into their appropriate disease classes. Orthogonal immunophenotypic studies of selected proteins in primary lymphoma tissue biopsies and lymphoma-derived cell lines corroborated the MS-based glycoproteomic data. Functional studies validated therapeutic targeting of a biomarker using a toxin-conjugated ligand. Overall, our studies indicate that mass spectrometry-driven proteomics provides a general strategy for discovery of cancer biomarkers and targets for precision therapeutics.