Applications of Mass Spectrometric Assays for Analysis of Serum Protein N-glycosylation Associated with Pancreatic Cancer.

Abstract

Pancreatic cancer has the worst prognosis among all cancers, mainly due to lack of effective diagnostics capable of early-stage detection. There have been many studies on pancreatic cancer biomarkers, among which serum protein markers are of particular interest as a minimally-invasive technique. In this dissertation, multiple mass spectrometric assays have been utilized to characterize protein N-glycosylation at the glycan, glycopeptide, and peptide levels both qualitatively and quantitatively, for identification of serum-based pancreatic cancer biomarkers. The assays incorporated optimization of sample pre-processing, liquid chromatography separation, and analysis by mass spectrometry. State-of-the-art mass spectrometric fragmentation methods including collision-induced dissociation (CID), electron transfer dissociation (ETD) and higher-energy collisional dissociation (HCD) have been utilized to study N-glycosylation of both individual proteins isolated from human serum and depleted human serum. Chapter 2 describes a MALDI-MS/MS based method for analysis of N-glycans which integrates N-glycan extraction, desialylation, permethylation, structure elucidation, and fucosylation degree measurement. The fucosylation degree of human serum haptoglobin is elevated in pancreatic cancer relative to non-cancer conditions, including normal controls, chronic pancreatitis and type II diabetes. Chapter 3 presents a CID/ETD-MS/MS method for detailing the site-specific glycosylation patterns of human serum alpha-2-macroglobulin (A2MG). For targeted analysis of A2MG site-specific core-fucosylation, an endoglycosidase-assisted strategy was utilized. All eight potential N-glycosylations sites were identified with six of them found to be core-fucosylated. Chapter 4 extends the work in Chapter 3 to a quantitative level and investigates the alterations of A2MG core-fucosylation at specific sites in pancreatic cancer using a label-free method. Core-fucosylation ratios at sites N396, N410 and N1424 were found to decrease in pancreatic cancer relative to normal controls. Chapter 5 describes total core-fucosylation profiling of human serum, where the core-fucosylated glycopeptide enrichment and mass spectrometric methods were optimized, leading to identification of 135 core-fucosylation sites in serum. The quantitative aspect of this assay used an isobaric labeling strategy, which may prove potentially useful for future high-throughput serum pancreatic cancer biomarker screening. The results of this thesis provide insight into the potential of protein fucosylation alterations as pancreatic cancer biomarkers, and relevant strategies will be useful in other clinical applications.