Intact glycopeptide characterization is an imperative yet challenging component of glycoprotein analysis. Elucidation of both glycan and peptide requires specific sample preparation workflows that, in combination with multiple tandem mass spectrometry (MS/MS) approaches, enable identification of single glycopeptide species (1). Here, we systematically optimized and evaluated the strengths and weaknesses of the glycoproteomics workflows using synthetic glycopeptides from sample preparation to its implication on data analysis.
A library of >100 synthetic N-glycopeptides representing human serum glycoproteins was synthesized. These synthetic glycopeptides fostered the systematic investigation to explore the advantages and limitations of the glycoproteomics workflow and comment on their suitability for high-throughput glycoproteomic studies. All experiments were performed on a quadrupole-Orbitrap-linear ion trap Tribrid mass spectrometer.
A simplified approach to purify and produce a panel of glycosylated amino acids carrying N-linked glycans with various structures was developed (2). These building blocks were used to synthesise a library containing >100 glycopeptides and their unglycosylated counterparts. Recently developed "Drop-HILIC" (3)(hydrophilic interaction chromatography) enrichment was further improved to allow for efficient enrichment of both N- and O-glycopeptides by using DMSO in solubilization buffer. Next, glycopeptide fragmentation characteristics were evaluated using different fragmentation techniques (HCD, SCE-HCD, ETciD, and EthcD) to assess the merit of each method in terms of (a) peptide backbone sequence coverage, (b) glycan composition, (c) proportion of signal in different fragment ion types (e.g., oxonium ions, Y-type ions, and peptide backbone fragment ions, and (d) unambiguous identification of the glycosylation site. We also evaluated and compared various software tools (Byonic, SugarQb) in their ability to identify these glycopeptides reliably. In the end, using the optimized workflow, we are in the process of establishing Human GlycoAtlas containing protein-specific, site-specific N- and O-glycosylation profile of plasma proteins.
To our knowledge, this constitutes the first broad and systematic analysis of the LC-MS/MS properties of glycopeptides using synthetic glycopeptides, allowing optimized bottom-up glycoproteomics experiments. The availability of these glycopeptides and spectra will facilitate the development and improvement of further experimental and computational strategies.