An Algorithm for Building Multi-State Classifiers Based on Collision-Induced Unfolding Data

Abstract

Collision-induced unfolding (CIU) has emerged as a valuable method for distinguishing iso-cross-sectional protein ions through their distinct gas-phase unfolding trajectories. CIU shows promise as a high-throughput, structure-sensitive screening technique with potential applications in drug discovery and biotherapeutic characterization. We recently developed a CIU classification workflow to support screening applications that utilized CIU data acquired from a single protein charge state to distinguish immunoglobulin (IgG) subtypes and membrane protein lipid binding. However, distinguishing highly similar protein structures, such as those associated with biotherapeutics, can be challenging. Here, we present an expansion of this classification method that includes CIU data from multiple charge states, or indeed any perturbation to protein structure that differentially affects CIU, into a combined classifier. Using this improved method, we are able to improve the accuracy of existing, single-state classifiers for IgG subtypes and develop an activation-state-sensitive classifier for selected Src kinase inhibitors when data from a single charge state was insufficient to do so. Finally, we employ the combination of multiple charge states and stress conditions to distinguish a highly similar innovator/biosimilar biotherapeutic pair, demonstrating the potential of CIU as a rapid screening tool for drug discovery and biotherapeutic analysis.

Toward Automation of Collision-Induced Unfolding Experiments through Online Size Exclusion Chromatography Coupled to Native Mass Spectrometry

“…For mAb subclass categorization, the centroid of the ATDs was standardized at CV = 0 V to ensure comparison solely on unfolding patterns, irrespective of drift time variations because of different masses. Adalimumab (IgG1), panitumumab (IgG2), and reslizumab (IgG4) were chosen as reference mAbs to build classification methods at intact and middle levels using the “Classification” module of the CIUSuite 2 software. , Univariate feature selection (UFS) plots were used to select the most diagnostic voltages, that is, with the higher scores, to classify clusterized mAbs (ofatumumab, trastuzumab, and nivolumab).…”

“…Adalimumab (IgG1), panitumumab (IgG2), and reslizumab (IgG4) were chosen as reference mAbs to build classification methods at intact and middle levels using the "Classification" module of the CIUSuite 2 software. 33,36 Univariate feature selection (UFS) plots were used to select the most diagnostic voltages, that is, with the higher scores, to classify clusterized mAbs (ofatumumab, trastuzumab, and nivolumab).…”

Rapid structural discrimination of IgG antibodies by multicharge-state collision-induced unfolding

“…Inspired by the pioneering efforts of Ruotolo's group, these challenges motivated the development of native IM-MS-based collision-induced unfolding (CIU), which has gained rapidly increasing attention owing to its ability to rapidly interrogate subtle tertiary structure differences and chemical modications in the protein isoforms based on their unfolding patterns and gas-phase stabilities. 9,33,34 Upon stepwise collisional activation of protein assemblies, a range of partially unfolded intermediates can occur in the gas phase, which provides a multitude of comparative and diagnostic information of protein complexes detected by IM-MS. 35,36 Unfortunately, only protein ions with a single charge state can be quadrupole-selected prior to CIU analysis, resulting in iterative data collection process or a loss of complementary information acquired by multiple charge states. Furthermore, the CIU patterns of protein assemblies are strongly dependent on charge-state selection, but to what extent charge states affect their gas-phase structures is still in an early stage of exploration.…”

Sialylation Fine-Tunes Glycoprotein Structural Microheterogeneity Associated with Alzheimer’s Disease as Captured by Native Ion Mobility-Mass Spectrometry

“…In this way, we manage to perform all ion unfolding/fragmentation sequentially on all sialylated glycoprotein isoforms. The IM-MS strategy additionally relies on further modifications of data visualization and quantitative characterization based on the CIUSuite [15][16][17] and UniDec 18 , which allow for not only reliable tracking of rapid, efficient and in situ releasing of sialic acids and the simultaneous stepwise unfolding of native-like glycoproteins, but also unprecedentedly revealing the differential roles of Neu5Ac and Neu5Gc sialylations in fine-tuning of glycoprotein structures and topologies. While bottom-up proteomics was employed to accurately determine the specific sialylation patterns in various TF samples, native gel electrophoresis experiments further extended our gas-phase findings to solution phase.…”