Optimisation of the use of sliding window deconvolution for comprehensive characterisation of trastuzumab and adalimumab charge variants by native high resolution mass spectrometry

Abstract

The biopharmaceutical industry continues to develop mAb-based biotherapeutics in increasing numbers. Due to their complexity, there are several critical quality attributes (CQAs) that need to be measured and controlled to guarantee product safety and efficacy. Charge variant analysis is a widely used method to monitor changes in product quality during the manufacturing process of monoclonal antibodies (mAbs) and, together with a bottom-up peptide centred approach, acts as a key analytical platform to fulfil regulatory requirements. Native MS measures biomolecules under conditions that preserve most aspects of protein tertiary and quaternary structure, enabling direct characterization of large intact proteins such as mAbs. The resulting native mass spectrum of a mAb is characterized by a narrower charge-state envelope that simplifies the spectra and also condenses the ion signals into fewer peaks, increasing the signal-to-noise ratio. Algorithmic spectral deconvolution is needed for routine accurate and rapid molecular weight determination, and consequently, multiple deconvolution algorithms have evolved over the past decade. Here, we demonstrate the utility of the sliding window algorithm as a robust and powerful deconvolution tool for comprehensive characterisation of charge variant analysis data for mAbs. Optimum performance is evaluated by studying the impact of critical software parameters on detection, identification and relative quantitation of protein isoforms. By combining molecular mass and retention time information, it was possible to identify multiple modifications on adalimumab and trastuzumab, both IgG1 mAbs, including lysine truncation, deamidation and succinimide formation, along with the N-glycan distribution of each of the identified charge variants. Sliding window deconvolution also provides a key benefit of low abundant variant detection in a single analysis and the ability to detect co-eluting components with different relative abundances. The studied mAbs demonstrate the algoritms applicability for efficient data processing of both simple and complex mAbs analysed using pH gradient cation exchange chromatography coupled to native mass spectrometry.