Research Article: Determination of true ratios of different N-glycan structures in electrospray ionization mass spectrometry

Date Published: March 7, 2017

Publisher: Springer Berlin Heidelberg

Author(s): Clemens Grünwald-Gruber, Andreas Thader, Daniel Maresch, Thomas Dalik, Friedrich Altmann.


An ideal method for the analysis of N-glycans would both identify the isomeric structure and deliver a true picture of the relative, if not absolute, amounts of the various structures in one sample. Porous graphitic carbon chromatography coupled with electrospray ionization mass spectrometry (ESI-MS) detection has emerged as a method with a particularly high potential of resolving isomeric oligosaccharides, but little attention has so far been paid to quantitation of the results obtained. In this work, we isolated a range of structures from Man5 to complex type N-glycans with zero to four sialic acids and blended them into an equimolar “glyco tune mix”. When subjected to liquid chromatography–ESI-MS in positive and negative modes, the glyco tune mix clearly demonstrated the futility of quantitation of N-glycans of different overall composition, different number of sialic acids, and strongly differing size without compensation for their very different molar responses. Relative quantitation of human plasma N-glycans was performed with correction factors deduced from this external glyco tune mix. Addition of just one isotope-coded internal standard with enzymatically added 13C-galactose led to absolute quantification in the same experiment.

Partial Text

Structural assignment of N-glycans and O-glycans has been a major issue in the last few decades, with the clearest advances having been achieved with mass spectrometry (MS) [1] and even more with liquid chromatography (LC)–MS [2, 3]. In terms of power to define a particular structure, MS surpasses chromatographic techniques such as hydrophilic interaction LC (HILIC)–high-performance LC (HPLC) or capillary zone electrophoresis of fluorescently labeled glycans [1]. These techniques, however, offer the advantage of an inherently identical molar response of all N-glycan species in a sample because of the invariable stoichiometry of the fluorophore [4, 5]. Correct relative quantitation of the components within a sample therefore requires well-separated peaks as facilitated by the latest ultraperformance HILIC columns [6]. Even then, with samples being more complex than antibody or plasma N-glycans, peak overlap will increasingly become a problem and requires the higher definition power of MS.

External calibration using a set of N-glycans of known—preferably equal—concentrations appears to be the minimum requirement for quantitative evaluation of glycomics data obtained by ESI-MS of underivatized glycans. Without this correction, the data are hardly more than incidental peak height ratios that will in addition vary from day to day and instrument to instrument. A recent attempt at external calibration showed impressive linearity over several orders of magnitude of permethylated glycans for a Thermo Fisher LTQ Orbitrap instrument [35]. Two nongalactosylated, neutral glycans (GnGnbi and Gn[GnGn]bi in ProGlycAn nomenclature) and one disialoglycan (Na6-4Na6-4) were the subjects of this study. A larger panel of structures was used in a study that introduced the application of stable-isotope-labeled glycans for internal calibration [30]. This work focused on IgG glycans and thus on diantennary structures with zero to two galactoses and fucose and up to two sialic acids. However, as IgG glycans were analyzed by MALDI-TOF MS, no further attention was paid to sialylated species.