Rapid Evaluation of the Extent of Haptoglobin Glycosylation Using Orthogonal Intact-Mass MS Approaches and Multivariate Analysis

Probing the Architecture of Multisubunit Protein Complexes with In-line Disulfide Reduction and Native MS Analysis

Native mass spectrometry (MS) continues to enjoy growing popularity as a means of providing a wealth of information on noncovalent biopolymer assemblies ranging from composition and binding stoichiometry to characterization of the topology of these assemblies. The latter frequently relies on supplementing MS measurements with limited fragmentation of the noncovalent complexes in the gas phase to identify the pairs of neighboring subunits. While this approach has met with much success in the past two decades, its implementation remains difficult (and the success record relatively modest) within one class of noncovalent assemblies: protein complexes in which at least one binding partner has multiple subunits cross-linked by disulfide bonds. We approach this problem by inducing chemical reduction of disulfide bonds under nondenaturing conditions in solution followed by native MS analysis with online buffer exchange to remove unconsumed reagents that are incompatible with the electrospray ionization process. While this approach works well with systems comprised of thiol-linked subunits that remain stable upon reduction of the disulfide bridges (such as immunoglobulins), chemical reduction frequently gives rise to species that are unstable (prone to aggregation). This problem is circumvented by taking advantage of the recently introduced cross-path reactive chromatography platform (XPRC), which allows the disulfide reduction to be carried out in-line, thereby minimizing the loss of metastable protein subunits and their noncovalent complexes with the binding partners prior to MS analysis. The feasibility of this approach is demonstrated using hemoglobin complexes with haptoglobin 1-1, a glycoprotein consisting of four polypeptide chains cross-linked by disulfide bonds.

Mass spectrometry-based methods to characterize highly heterogeneous biopharmaceuticals, vaccines, and nonbiological complex drugs at the intact-mass level

The intact-mass MS measurements are becoming increasingly popular in characterization of a range of biopolymers, especially those of interest to biopharmaceutical industry. However, as the complexity of protein therapeutics and other macromolecular medicines increases, the new challenges arise, one of which is the high levels of structural heterogeneity that are frequently exhibited by such products. The very notion of the molecular mass measurement loses its clear and intuitive meaning when applied to an extremely heterogenous system that cannot be characterized by a unique mass, but instead requires that a mass distribution be considered. Furthermore, convoluted mass distributions frequently give rise to unresolved ionic signal in mass spectra, from which little-to-none meaningful information can be extracted using standard approaches that work well for homogeneous systems. However, a range of technological advances made in the last decade, such as the hyphenation of intact-mass MS measurements with front-end separations, better integration of ion mobility in MS workflows, development of an impressive arsenal of gas-phase ion chemistry tools to supplement MS methods, as well as the revival of the charge detection MS and its triumphant entry into the field of bioanalysis already made impressive contributions towards addressing the structural heterogeneity challenge. An overview of these techniques is accompanied by critical analysis of the strengths and weaknesses of different approaches, and a brief overview of their applications to specific classes of biopharmaceutical products, vaccines, and nonbiological complex drugs.

First bioelectronic immunoplatform for quantitative secretomic analysis of total and metastasis-driven glycosylated haptoglobin

The glycosylation status of proteins is increasingly used as biomarker to improve the reliability in the diagnosis and prognosis of diseases as relevant as cancer. This feeds the need for tools that allow its simple and reliable analysis and are compatible with applicability in the clinic. With this objective in mind, this work reports the first bioelectronic immunoplatforms described to date for the determination of glycosylated haptoglobin (Hp) and the simultaneous determination of total and glycosylated Hp. The bioelectronic immunoplatform is based on the implementation of non-competitive bioassays using two different antibodies or an antibody and a lectin on the surface of commercial magnetic microcarriers. The resulting bioconjugates are labeled with the horseradish peroxidase (HRP) enzyme, and after their magnetic capture on disposable electroplatforms, the amperometric transduction using the H2O2/hydroquinone (HQ) system allows the single or multiple detection. The developed immunoplatform achieves limits of detection (LODs) of 0.07 and 0.46 ng mL−1 for total and glycosylated Hp in buffer solution, respectively. The immunoplatform allows accurate determination using simple and relatively short protocols (approx. 75 min) of total and glycosylated Hp in the secretomes of in vitro–cultured colorectal cancer (CRC) cells with different metastatic potentials, which is not feasible, due to lack of sensitivity, by means of some commercial ELISA kits and Western blot methodology.

Graphical abstract

Haptoglobin polymorphism affects its N-glycosylation pattern in serum

•

Polymorphism affects glycosylation pattern of haptoglobin in healthy population.

•

Sample phenotype classification was done based on the number and type of α-chains.

•

Glycoproteomic analyses of haptoglobin were done using enzyme-assisted LC-MS/MS.

•

Significant differences were obseerved in branching, sialylation and fucosylation.

Introduction

Haptoglobin (Hp) is an abundant acute-phase protein secreted mainly by the liver into the bloodstream. There are three Hp protein phenotypes (Hp type 1–1, 2–1, and 2–2), which differ in the number of α- and β-chains, type of α-chain (the β-chain type remains the same in all the Hp phenotypes), and the polymers that they form via disulfide bonds. Hp has four N-glycosylation sites on the β-chain. Glycosylation of Hp has been reported frequently as a potential glycobiomarker for many diseases; however, whether Hp polymorphism affects its glycosylation has not yet been addressed extensively or in depth.

Objectives

This study investigated the differences between the glycosylation patterns of Hp phenotypes using serum from 12 healthy individuals (four for each Hp phenotype).

Method

An efficient method for isolating Hp from serum was established and subsequently the Hp phenotype of each sample was characterized by immunoblotting. Then, LC-MS/MS analysis of isolated Hp after treatment with three exoglycosidases (sialidase, α2-3 neuraminidase, Endo F3) was performed to characterize the glycosylation pattern of Hp for each individual sample.

Results

The data reveal significant differences among the branching, sialylation, and fucosylation of Hp types, documenting the effect of Hp polymorphism on its glycosylation.

Conclusion

Overall, the study suggests that Hp phenotype characterization should be considered during the investigation of Hp glycosylation.