Why Glycan Analysis is an Essential Part of Therapeutic Protein Characterization
A recent report on the growth of the glycan analysis market¹ highlighted the importance of assessing the glycosylation of biotherapeutics. The carbohydrate structures attached to the crystallizable fragment (Fc) of monoclonal antibodies can influence efficacy and safety and so they are often classed as critical quality attributes. For this reason the analytical characterization of glycosylation forms an important part of protein structure analysis throughout the development process.
This article is posted on our Sartorius Blog.
Glycosylation significantly impacts effector functions of the immune system, including complement-dependent cytotoxicity (CDC), antibody-dependent cell-mediated cytotoxicity (ADCC ) and antibody-dependent cellular phagocytosis (ADCP). It is of particular importance when one or more of these effector functions is the primary mechanism of action, but should be considered nonetheless for any therapeutic protein containing an Fc region.
Combining Protein Structure Analysis with Binding and Functional Assays
A dramatic example of the possible effect of glycosylation on effector function is shown below (Figure 1). Enzymatic removal of the Fc glycans attached to a therapeutic monoclonal antibody (mAb) completely impairs C1q binding (and hence CDC activity).
Figure: A dilution series of the mAb (glycosylated or deglycosylated) was bound to an ELISA plate prior to the addition of HRP-labelled C1q. A chromogenic substrate was added and the resulting color change was measured. yellow – reference standard, black – glycosylated mAb, dark gray – deglycosylated mAb, light gray – blank.
There can be a wide heterogeneity of glycan structures in biotherapeutics and the structural variation between samples of different origin are, of course, more subtle than the scenario described above. We have explored the impact of glycan variation on antibody function by comparison of a candidate biosimilar to the innovator molecule². Orthogonal protein characterization methods, including intact mass and released N-glycan analysis by LC-MS, were used to assess glycosylation. By combining these data with those from binding and functional assays, we were able to measure the effect of structural differences on CDC and ADCC activity (Table 1).
LC-MS analysis revealed elevated levels of fucosylation and galactosylation in the candidate biosimilar, relative to the innovator. The increase in terminal galactose was reflected in higher relative C1q binding and increased CDC potency. Meanwhile, a reduction in relative CD16a binding and resultant decrease in ADCC activity was attributed to the increase in core fucose, which is known to hinder formation of the Fc-receptor complex.
Table 1: Comparison of an innovator and candidate biosimilar mAb in terms of glycan profile, binding and effector functions.
Analytical Characterization of Glycosylation
The potential for variation in monosaccharide composition, linkage and branching pattern leads to a vast array of possible glycan structures, many with similar physicochemical properties. As such, it is useful to bring together protein characterization methods that provide orthogonal selectivities. A great example of this is the use of LC-MS for released glycan analysis. For example, the isomers (a) and (b) in Figure 2 have identical mass but can be separated by hydrophilic interaction liquid chromatography (HILIC). In contrast, (c) and (d) have similar HILIC retention times but have different masses and are easily resolved by high resolution mass spectrometry. Thus the combination of these two techniques provides deeper analytical characterization of glycosylation for protein structure analysis with increased confidence.
Figure 2: Symbolic representations of glycan structures using CFG nomenclature. Glycans commonly known as: (a) G1F, (b) G1’F, (c) Man5, (d) G0FB.