Introduction
The approval of biopharmaceuticals requires in-depth characterization at the molecular level involving confirmation of the amino-acid sequence as well as comprehensive assessment of post-translational modifications (PTMs). Consistent glycosylation constitutes an important quality attribute as it may impact the efficacy and safety of the therapeutic product. Current analytical methods comprise the characterization of released glycans or of glycopeptides, providing detailed information on the average molecular composition. Alternatively, the existence and relative abundance of specific proteoforms can be revealed by the determination of intact protein masses4, as has been accomplished for several therapeutic monoclonal antibodies (mAbs). Conventional mass determination under denaturing conditions is restricted to samples of limited complexity due to the overlap of broad signal clusters observed for the respective charge states. In contrast, under the conditions applied in native mass spectrometry (MS), protein folding is preserved resulting in lower charge states situated at higher m/z in the corresponding mass spectrum. The inherently higher spatial resolution at high m/z allows separation and thus deconvolution of these broad signal clusters. This technique is well established for the characterization of non-covalent protein complexes, e.g., large protein assemblies or antibody–drug conjugates. Native MS has also been applied for qualitative and semi-quantitative analysis of composite mixtures of mAbs as well as for characterization of micro-heterogeneity in therapeutic proteins arising from, for example, N-glycosylation variants. In this context, the glycan heterogeneity of human erythropoietin (26–30 kDa) and human plasma properdin (54 kDa) was successfully revealed upon integration of intact protein mass determination by native MS, middle–down analysis of proteolytic glycopeptides, and enzymatic deglycosylation, which facilitated the assignment of PTM compositions to the detected intact protein masses.
Etanercept, the active pharmaceutical ingredient of Enbrel, is a highly glycosylated therapeutic Fc-fusion protein. This biopharmaceutical acts as an inhibitor for tumor necrosis factor (TNF), an important mediator protein of inflammatory cell responses, and has been licensed for the treatment of autoimmune disorders such as rheumatoid and psoriatic arthritis. Etanercept consists of a TNF-α receptor (TNFR) domain fused to the Fc portion of human IgG1 and forms dimers stabilized by three intermolecular disulfide bonds resulting in a theoretical protein mass of 102.4 kDa. The resultant protein comprises multiple glycosylation sites: four N- and 26 O-glycosylation sites in the dimeric TNFR domain, as well as two N-glycan sites in the Fc domain, as previously characterized via released N- and O-glycan analysis by hydrophilic interaction liquid chromatography with fluorescence detection. In the same study, O-glycopeptides were analyzed by HPLC-MS with collision-induced- and electron transfer dissociation. The O-glycans were found to be predominantly of the core 1 subtype (Galβ1-3GalNAc-) substituted by up to two sialic acid residues (N-acetylneuraminic acid, Neu5Ac).
