Glycoform Analysis of Glycoproteins

 Glycoform is defined as a subunit of molecules with identical polypeptide sequences but with different glycans present at the sites of glycosylation. It is very important to analyze the glycoform, because different glycoforms of the same glycoprotein have different biological properties (1). It has been estimated, for example, that recombinant tissue plasminogen activator (rtPA) may contain as many as 11,500 different glycoforms (2).

Ion-exchange chromatography, reversed phase HPLC, and a variety of electrophoretic methods are used to separate the glycoforms. In these cases, the separation, which is negatively charged at neutral pH values, is attributed to the hydrophobicity of the samples, and the different degrees of sialic acids. In general, it is quite difficult to analyze the the glycoform populations of the glycoprotein directly, even when careful selection of the conditions can be made. Therefore, it is common practice to first release carbohydrates by chemical or enzymatic methods before analyzing them. The characterization methods of carbohydrates released from glycoproteins used so far are as follows.

1) Two-dimensional sugar mapping of pyridylamino (PA)-oligosaccharides, which analyzes the released oligosaccharides with both reversed and normal phase HPLC (3).
2) The empirical characterization, focused on "structural reporter group" with NMR (4).
3) The analysis of released oligosaccharides with soft ionization methods, such as fast atom bombardment (FAB) (5), electrospray (ESI) (6), matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOFMS) (6). In addition, the hyphenated method via either LC or CE coupled with these mass spectrometric techniques (LC-MS or CE-MS) has also been used (7), because the analysis can be conducted without isolating each component of the samples.
Fig. Flow chart of the preparation of oligosaccharides sample from glycoprotein.
Since the first two methods are based on the empirical approach using a kind of database, their applications to unknown samples are generally considered to be difficult. The latter methods can be applied to unknowns. It is not, however, suitable for the analysis of monosaccharide composition, glycosidic linkage, or anomeric configuration, because they are based on molecular weight information. Therefore, the key to characterization of the glycoform lies in the use of an appropriate combination of different analytical methodologies.

Recently, an attempt at direct analysis of the glycoprotein by combination of CE and off-line MALDI-TOFMS has been reported (2). However, the separation was not satisfactory. At present this application may be limited for the demonstration of lot-to- lot consistency of different batches of recombinant biopharmaceuticals.
Masahiko OKAMOTO (Environmental Health Science Laboratory,
Sumitomo Chemical Co., Ltd)
References (1) TW, Rademacher, RB, Parekh, RA, Dwek Annu. Rev. Biochem. 58, 785, 1988
(2) JA, Chakel, E, Pungor Jr, WS, Hancock, SA, Swedberg J. Chromatogr. B. 689, 215, 1997
(3) N, Tomiya, J, Awaya, N, Kurono, S, Endo, Y, Arata, N, Takahashi Anal. Biochem. 171, 73, 1988
(4) JFG, Vliegenthart, L, Dorland, H van Halbeek Adv. Carbohydr. Chem. Biochem, 41, 209, 1983
(5) A, Dell Adv. Carbohydr. Chem. Biochem, 45, 19, 1987
(6) J, Charlwood, H, Birrell, P, Camilleri, J. Chromatogr. B. 734, 169, 1999
(7) J, Suzuki-Sawada, Y, Umeda, A, Kondo, I, Kato Anal. Biochem. 207, 203, 1992
Jun. 15, 2000

GlycoscienceNow INDEX Return to Top Page