Glycosylation Patterns and Glycopeptide Biomarkers
Proteomics is one of the most dynamically improving research field recently focusing on post-translational modifications (eg. phosphorylation, glycosylation). Glycosylation is immensely important; two typical examples are biopharmaceuticals and FDA–accepted biomarkers, most of which are glycoproteins. There is a lack of adequate techniques for determining glycosylation patterns, suitable high-throughput analytical methodologies are just starting to appear. In this presentation we will show a novel high-throughput approach to determine site-specific glycosylation patterns.
We have developed a method which is capable for studying not only a purified glycoprotein sample, but is capable of dealing with glycoprotein mixtures, like blood plasma. This is an essential step for a high throughput approach. The first step of the analytical strategy is glycoprotein enrichment of blood sample. Then glycoproteins are enzimatically digested by trypsin resulting in a mixture of peptides and glycopeptides; and these are identified using nanoUPLC-MS/MS. Tryptic glycopeptides retain information of both the protein and the glycosylation pattern, so their analysis will retain all relevant information. (This is in contrast to the more common approach of glycan analysis following PNGase digestion, as this latter technique looses the information which connects protein and glycan chains.) After the glycopeptides are identified, the next step is their quantification. As suitable internal standards are not available, only relative quantification is feasible (i.e. determ ining the abundance ratio of e.g. triantennary, single fucosylated vs. triantennary, nonfucosylated), performed using nano-HPLC-MS. Even this is plagued with problems, as glycopeptides easily fragment under conventional conditions in a mass spectrometer, which does result in significant bias. Beside the complications, there is the practical problem of the very large number of glycosylated structures involved, so manual analysis is almost impossible. Computer software was also developed to automatically interpret and process this huge amount of data.
A workflow and a practical solution to these problems are presented, which allows simple determination of the glycosylation pattern of various serum glycoproteins from biological matrices.