Analysis of Protein Glycosylation by Mass Spectrometry

David J. Harvey

Glycobiology Institute, Department of Biochemistry, University of Oxford, Oxford, UK

Introduction

It is estimated that about half of all proteins are glycosylated with the glycan portion responsible for many of the biophysical properties of the molecules [1, 2]. An enormous amount of work has been devoted to their analysis and much of this is covered in recent reviews [3-22]. This chapter summarizes the structure of these compounds and describes the main mass spectrometric methods used in their analysis.

General Structures of Carbohydrates

Carbohydrates are the most abundant and structurally diverse compounds found in nature. They can exist as small monosaccharides or they can link together to give polymeric compounds such as cellulose (d-G1cP-1 ^ 4)- linear repeat) and chitin d-G1cNAcP(1 ^ 4)- linear) with molecular weights that can exceed 1 MDa. Unlike linear biopolymers such as proteins and nucleic acids, oligomeric (2-10 monosaccharide residues) and polymeric carbohydrates typically form branched structures because linkage of the constituent monosaccharides can occur at any of the hydroxyl groups of the adjacent residue. Consequently, very large numbers of isomers are possible. For example, it has been calculated that for a simple hexasaccharide, there are more than 10¹² possible isomeric structures [23], presenting what has been referred to as an “isomer barrier” to the analyst. It is doubtful, therefore, that any single analytical technique will ever provide all of the structural information necessary to characterize such a large number of oligosaccharides. Fortunately, however, this hypothetical situation does not occur in nature because the biosynthetic

Analysis of Protein Post-Translational Modifications by Mass Spectrometry,

First Edition. Edited by John R. Griffiths and Richard D. Unwin.

© 2017 John Wiley & Sons, Inc. Published 2017 by John Wiley & Sons, Inc.

pathways creating most complex carbohydrates are specific and limited by the available glycosyltransferases. Consequently, only a very few of the theoretically possible isomers are ever encountered, a property that can be utilized structurally if the source of the glycan is known.

The basic building blocks of oligo- and polysaccharides are monosaccharides with the general formula C_nH_2nO_n where n = 3 (trioses) to 9 (nonoses) with n = 6 (hexoses) being the most common. The monosaccharides contain n - 1 hydroxy groups and one aldehyde (reducing sugars) or keto group and can exist in a number of isomeric forms as the result of equilibration between linear and cyclic forms (typically 6- (pyranose) or 5- (furanose) membered rings). Rings can also exist in different conformations, for example, chair or boat forms. Cyclization produces an additional chiral center giving rise to a- and p-anomers. Formation of a bond to the anomeric carbon prevents ring opening and fixes the ring size and conformation of that monosaccharide. Many modified monosaccharides exist such as those with missing hydroxyl groups, primary hydroxyl groups oxidized to carboxylic acids, or hydroxy groups replaced by, for example, a primary amine or M-acylamino group. Hydroxy groups can also be methylated or acetylated (see Kennedy [24] for more information on carbohydrate structure).