New Intact Mass Workflow Resolves Over 300 Glycoforms in a Complex Fusion Protein
A recent study published in the Journal of Pharmaceutical and Biomedical Analysis presents a new workflow for characterizing highly heterogeneous glycoproteins at the intact protein level, resolving more than 300 distinct glycoforms of a densely glycosylated Fc-fusion protein. To confirm the method could reliably detect real glycan changes and not just produce a well-resolved spectrum, the researchers turned to our SialEXO™ sialidase mix to enzymatically remove sialic acid and generate a controlled, well-defined shift in glycan composition.
The study focused on CD24-Fc, a homodimeric fusion protein combining human CD24 with an IgG1 Fc fragment. With eight clustered O-glycosylation sites and two N-glycosylation sites on each CD24 chain, plus an additional N-glycosylation site per Fc domain, the intact dimer carries 16 O-glycosylation sites and 6 N-glycosylation sites, making it a particularly demanding test case for intact mass analysis.
To resolve this complexity, Zhang et al. combined hydrophilic interaction chromatography (HILIC) with a scan-by-scan deconvolution strategy, processing each individual MS scan independently rather than summing the full chromatographic window into one congested spectrum. The approach improved signal-to-noise from approximately 1.5 in summed spectra to over 35 in single-scan spectra, ultimately identifying more than 300 glycoforms across a 32–38 kDa mass range.
With only one large-scale batch of CD24-Fc available, the researchers needed a controlled way to simulate batch variability and test whether their method could actually detect it. Using SialEXO™, they enzymatically removed sialic acid from CD24-Fc, then compared the treated and untreated material with the same workflow. The result was unambiguous: a peak at 34,802.0 Da shifted to 34,219.0 Da after treatment, a loss of roughly 583 Da consistent with the removal of two NeuAc residues, alongside a recurring 365 Da mass shift indicating an accompanying change in O-glycan occupancy.
This confirmed that the HILIC-MS and scan-by-scan deconvolution workflow can sensitively and specifically detect compositional glycan changes at the intact protein level, a capability directly relevant to glycoform monitoring, batch comparability, and biosimilarity assessment for the growing class of Fc-fusion, cytokine-fusion, and enzyme-based biologics that carry substantially more glycosylation complexity than a typical monoclonal antibody.
