High-throughput glycomics for discovery of cancer biomarkers.

9 Background: While the importance of glycosylation in many cancers is well established, the use of glycomics in biomarker research has lagged behind genomics and proteomics. This is due, in part, to the lack of practical platforms capable of analyzing clinically relevant sample numbers. To address these challenges, we have developed a novel glycomics technology (the GlycanMap platform) that combines a high-throughput assay with custom bioinformatics and rapidly provides both biomarker candidates and information on the underlying biology. Methods: N-glycans were enzymatically released from their parent glycoproteins and captured on chemoselective beads. After washing to remove non-glycan components, purified glycans were derivatized to stabilize labile sialic acids and released from the beads. The steps described above were automated on a 96-well format robotics system to maximize throughput and reduce variability and can be performed in less than 24 hours. Released glycans were analyzed by MALDI-TOF MS using internal standards to facilitate quantitation. In addition to comparing individual glycans between groups, glycan changes were also analyzed with respect to known glycan biosynthetic pathways. Results: The automated assay was compatible with multiple biological sample types, including serum/plasma, tissue, and cell lysates. Human serum was used to assess assay performance and yielded 50-60 glycans with CVs of 10-15% and good linearity. The lower limit of detection was approximately 100 nM. The assay was applied to drug-treated colon cancer cells (HCT116) and revealed significant (> 2-fold) changes in 17 glycans. Projection of these glycan changes on the known N-glycan pathway showed that the most significant changes occurred in the medial-Golgi. Conclusions: We have developed and optimized a high-throughput glycomics platform to facilitate large-scale biomarker studies and assured its practical performance in terms of sensitivity, repeatability, and linearity. Application of this assay to drug-treated colon cancer cells demonstrated that projection of individual glycan changes against known glycan pathways provided additional information about biological mechanism and relevance.