Glycobiology Advance Access originally published online on October 11, 2008
Glycobiology 2009 19(3):201-211; doi:10.1093/glycob/cwn099
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press 2008.
Identification and quantification of N-linked oligosaccharides released from glycoproteins: An inter-laboratory study
2 Analytical Science, National Physical Laboratory, Teddington, Middlesex, TW11 0LW, UK
3 National Institute for Biological Standards and Control, Blanche Lane, South Mimms, Hertfordshire, EN6 3QG, UK
1 To whom correspondence should be addressed: Tel: +44-01707-641211; Fax: +44-01707-641050; e-mail: cjones{at}nibsc.ac.uk
Received on March 7, 2008; revised on September 5, 2008; accepted on September 28, 2008
As characterization of glycosylation is required for the licensing of recombinant glycoprotein therapeutics, technique comparability must be assessed. Eleven UK laboratories (seven industrial, two regulatory or government, two academic) participated in an inter-laboratory study to analyze N-glycans present in four mixtures prepared by PNGase F cleavage of commercial glycoproteins: human 
1-acid glycoprotein (H1), bovine 1-acid glycoprotein (B1), bovine pancreatic ribonuclease B (RNaseB), and human serum immunoglobulin G (hIgG). Participants applied their routine glycan mapping methodology using predominantly chromatography and mass spectrometry to identify and quantify components. Data interpretation focused on the relative amounts of different glycan structures present, the degree of sialylation, antennary and the galactosylation profiles, fucosylation and bisecting GlcNAc content, and the number of glycan components identified. All laboratories found high levels of sialylation for H1 and B1 (Z-numbers 271 ± 24 and 224 ± 18, respectively), but varying ratios of di-, tri-, and tetra-antennary chains. The Z-score for hIgG glycans had high variability as values obtained from mass spectrometric and chromatographic methods clustered separately. The proportion of the major penta-mannosyl chain from RNaseB was between 29 and 62%. Proportions of fucosylated and bisected GlcNAc chains from hIgG were between 58 and 96% and 9 and 23%, respectively. Mass spectrometric approaches consistently identified more glycan species, especially when both N-glycolylneuraminic acid (Neu5Gc) and N-acetylneuraminic acid (Neu5Ac) were present. These data highlight the need for well-characterized reference standards to support method validation and regulatory guidance on selection of approaches. Pharmacopoeial specifications must acknowledge method variability.
Key words: glycan analysis / glycoproteins / IgG / N-glycans / sialylation