Glycosylation of cell surface, secreted, and circulating proteins is one of the most common types of post‐translational modification. These modifications occur most commonly as one of three major classes: N‐linked glycosylation on asparagine residues, O‐linked glycosylation on serine or threonine residues, or as glycosaminoglycan oligosaccharide polymers on serine. Specifically, for N‐linked glycans, an endoglycosidase enzyme, peptide N‐glycosidase F (PNGase F), cleaves the attached oligosaccharides between the asparagine and first sugar. A method to analyze released N‐glycans and map them to specific locations within a tissue is presented here. The PNGase F is applied by solvent sprayer as a molecular layer on frozen or formalin‐fixed tissues and all released N‐glycans in a given region of tissue are detected using matrix‐assisted laser desorption/ionization (MALDI) imaging mass spectrometry (MALDI‐IMS). Using the described MALDI‐IMS protocol, at least 40 or more individual N‐glycans can be mapped to tissue histopathology and extracted for further structural analysis approaches. © 2018 by John Wiley & Sons, Inc.

The early detection of pancreatic ductal adenocarcinoma (PDAC) is a complex clinical obstacle yet is key to improving the overall likelihood of patient survival. Current and prospective carbohydrate biomarkers carbohydrate antigen 19-9 (CA19-9) and sialylated tumor-related antigen (sTRA) are sufficient for surveilling disease progression yet are not approved for delineating PDAC from other abdominal cancers and noncancerous pancreatic pathologies. To further understand these glycan epitopes, an imaging mass spectrometry (IMS) approach was used to assess the N-glycome of the human pancreas and pancreatic cancer in a cohort of patients with PDAC represented by tissue microarrays and whole-tissue sections. Orthogonally, these same tissues were characterized by multiround immunofluorescence that defined expression of CA19-9 and sTRA as well as other lectins toward carbohydrate epitopes with the potential to improve PDAC diagnosis. These analyses revealed distinct differences not only in N-glycan spatial localization across both healthy and diseased tissues but importantly between different biomarker-categorized tissue samples. Unique sulfated biantennary N-glycans were detected specifically in normal pancreatic islets. N-glycans from CA19-9-expressing tissues tended to be biantennary, triantennary, and tetra-antennary structures with both core and terminal fucose residues and bisecting GlcNAc. These N-glycans were detected in less abundance in sTRA-expressing tumor tissues, which favored triantennary and tetra-antennary structures with polylactosamine extensions. Increased sialylation of N-glycans was detected in all tumor tissues. A candidate new biomarker derived from IMS was further explored by fluorescence staining with selected lectins on the same tissues. The lectins confirmed the expression of the epitopes in cancer cells and revealed different tumor-associated staining patterns between glycans with bisecting GlcNAc and those with terminal GlcNAc. Thus, the combination of lectin-immunohistochemistry and lectin-IMS techniques produces more complete information for tumor classification than the individual analyses alone. These findings potentiate the development of early assessment technologies to rapidly and specifically identify PDAC in the clinic that may directly impact patient outcomes.