High throughput and high confidence sequence variant analysis in therapeutic antibodies using Evosep One liquid chromatography tandem mass spectrometry with synthetic heavy peptides.

Sequence variants are anomalous misincorporations of amino acids into the primary structure of therapeutic antibodies during DNA replication and protein biosynthesis. As these low abundance variants contribute to molecular heterogeneity and could negatively impact the safety and efficacy of a protein therapeutic, analytical methods like liquid chromatography tandem mass spectrometry (LC-MS2) are used to monitor them with the goal of establishing control strategies that limit their occurrence. Current LC-MS2 strategies depend on relatively long gradients that minimize coelution between abundant non-variant peptide peaks and trace-level variants to limit ion suppression that can potentially conceal the latter. However, lengthy LC gradients reduce the number of samples that can be analyzed per day, limiting the practicality of LC-MS2 when analyzing large sample sets. Furthermore, confident variant identification partly depends on capturing rich MS2 spectra that localize any amino acid misincorporations, which can be challenging due to the low abundance of this class of analyte. This work drastically reduces the cycle time to run each therapeutic antibody sample with roughly the same or even more variant identifications, compared to traditional LC-MS2 analysis, by integrating an Evosep One LC platform with an Orbitrap Fusion Lumos mass spectrometer. It also introduces a novel strategy using synthetic peptides that contain heavy isotopes placed near both termini to validate lower confidence variants in one targeted LC-MS2 run according to retention time, precursor mass signal, and MS2 fragment patterns shared with the heavy peptide variant. Taken together, this approach enables high-throughput sequence variant analysis at 30 samples per day as well as validation for lower confidence variants that can be integrated into therapeutic antibody process development and characterization.

Electron Transfer Dissociation Parameter Optimization Using Design of Experiments Increases Sequence Coverage of Monoclonal Antibodies.

Middle-down analysis of monoclonal antibodies (mAbs) by tandem mass spectrometry (MS2) can provide detailed insight into their primary structure with minimal sample preparation. The middle-down approach uses an enzyme to cleave mAbs into Fc/2, LC, and Fd subunits that are then analyzed by reversed phase liquid chromatography tandem mass spectrometry (RPLC-MS2). As maximum sequence coverage is desired to obtain meaningful structural information at the subunit level, a host of dissociation methods have been developed, and sometimes combined, to bolster fragmentation and increase the number of identified fragments. Here, we present a design of experiments (DOE) approach to optimize MS2 parameters, in particular those that may influence electron transfer dissociation (ETD) efficiency to increase the sequence coverage of antibody subunits. Applying this approach to the NIST monoclonal antibody standard (NISTmAb) using three RPLC-MS2 runs resulted in high sequence coverages of 67%, 67%, and 52% for Fc/2, LC, and Fd subunits, respectively. In addition, we apply this DOE strategy to model the parameters required to maximize the number of fragments produced in "low", "medium", and "high" mass ranges, which ultimately resulted in even higher sequence coverages of NISTmAb subunits (75%, 78%, and 64% for Fc/2, LC, and Fd subunits, respectively). The DOE approach provides high sequence coverage percentages utilizing only one fragmentation method, ETD, and could be extended to other state-of-the-art techniques that combine multiple fragmentation mechanisms to increase coverage.

Integration of liquid chromatography mass spectrometry with a heavy peptide response curve accurately measures unprocessed C-terminal lysine during peptide mapping analysis of therapeutic antibodies in a single run.

Therapeutic monoclonal and bispecific antibodies are susceptible to modification after protein biosynthesis. These post-translational modifications (PTMs) not only contribute to mass and charge heterogeneity, but they can also negatively impact the molecule's activity, half-life, and immunogenicity. Therefore, identification and quantification of PTMs are critical to ensure the safety and efficacy of an antibody therapeutic as well as demonstrate product consistency and process control. Unprocessed C-terminal lysine on the heavy chain (HC) is a prevalent modification that contributes to this charge heterogeneity in antibodies. Peptide mapping through liquid chromatography tandem mass spectrometry (LC-MS2) enjoys higher selectivity and sensitivity for measuring this PTM relative to global PTM methods, but differences in the ionization efficiencies of the unprocessed C-terminal K peptide and the truncated C-terminal K peptide result in its overestimation. Consequently, large discrepancies in this PTM's measured abundance may exist between different characterization assays used in regulatory filings, which can be further compounded by large variability when multiple mass spectrometers are used to quantify C-terminal K during a therapeutic's lifespan. In this study, we propose a simple new method to quantify unprocessed C-terminal K in antibodies in a single LC-MS2 run that incorporates heavy isotopic standards for both the unprocessed and truncated C-terminal K peptide to build a response curve and correct for the disparity in ionization efficiency between these two different peptide sequences. The approach was evaluated across two different Orbitrap-based mass spectrometers using multiple monoclonal and bispecific therapeutic antibodies, resulting in accurate (<10% error, as determined with peptide standards) and precise C-terminal K quantification during peptide mapping analysis.

Combination of FAIMS, Protein A Depletion, and Native Digest Conditions Enables Deep Proteomic Profiling of Host Cell Proteins in Monoclonal Antibodies.

Host cell proteins (HCPs) are residual impurities generated by the expression cell line during the production of biopharmaceuticals. Although the majority of these contaminants are removed during purification, HCPs can represent a considerable risk to the efficacy and safety of a therapeutic protein if not actively monitored. The enzyme-linked immunosorbent assay (ELISA) is commonly used throughout production to monitor HCP levels but has limited ability to identify novel HCPs or provide detailed quantification. Liquid chromatography tandem mass spectrometry (LC-MS2) methods are increasingly being used in conjunction with established ELISA techniques to provide rapid adaptability to increasingly complex samples as well as highly quantitative and informative results. However, MS-based methods are still hindered by the large dynamic range between high abundance biopharmaceutical proteins and low abundance HCPs. Here, we propose a multifactorial approach designed to optimize HCP detection in purified monoclonal antibody samples with LC-MS2. By first depleting the sample of antibody on a protein A column, then specifically digesting HCPs while precipitating remaining antibody, and finally reducing spectral complexity through compensation voltage (CV) switching using high-field asymmetric waveform ion mobility spectrometry (FAIMS), we identified multiple-fold more HCPs in the NIST monoclonal antibody standard than any single established mass spectrometry technique reported in the literature. Our analyses consistently identified over 600 high confidence mouse HCPs, a multifold increase over established methods, while maintaining high reproducibility.