Analysis of serum
protein glycovariants has the potential to identify
new biomarkers of human disease. However, the inability to rapidly
quantify glycans in a site-specific fashion remains the major barrier
to applying such biomarkers clinically. Advancements in sample preparation
and glycopeptide quantification are thus needed to better bridge glycoscience
with biomarker discovery research. We present here the successful
utilization of several sample preparation techniques, including multienzyme
digestion and glycopeptide enrichment, to increase the repertoire
of glycopeptides that can be generated from serum glycoproteins. These
techniques combined with glycopeptide retention time prediction and
UHPLC-QqQ conditions optimization were then used to develop a dynamic
multiple-reaction monitoring (dMRM)-based strategy to simultaneously
monitor over 100 glycosylation sites across 50 serum glycoproteins.
In total, the abundances of over 600 glycopeptides were simultaneously
monitored, some of which were identified by utilizing theoretically
predicted ion products and presumed m/z values. The dMRM method was found to have good sensitivity. In the
targeted dMRM mode, the limit of quantitation (LOQ) of nine standard
glycoproteins reached femtomole levels with dynamic ranges spanning
3–4 orders of magnitude. The dMRM-based strategy also showed
high reproducibility with regards to both instrument and sample preparation
performance. The high coverage of the serum glycoproteins that can
be quantitated to the glycopeptide level makes this method especially
suitable for the biomarker discovery from large sample sets. We predict
that, in the near future, biomarkers, such as these, will be deployed
clinically, especially in the fields of cancer and autoimmunity.