Understanding the Impact of Methionine Oxidation on the Biological

Aug 30, 2016 - Yingda Xu , Dongdong Wang , Bruce Mason , Tony Rossomando , Ning Li , Dingjiang Liu , Jason K Cheung , Wei Xu , Smita Raghava , Amit ...
0 downloads 0 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Understanding the Impact of Methionine Oxidation on the Biological Functions of IgG1 Antibodies Using Hydrogen/Deuterium Exchange Mass Spectrometry Jingjie Mo, Qingrong Yan, Chi Kwong So, Tam Soden, Michael J. Lewis, and Ping Hu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b01958 • Publication Date (Web): 30 Aug 2016 Downloaded from http://pubs.acs.org on September 7, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Understanding the Impact of Methionine Oxidation on the Biological Functions of IgG1 Antibodies Using Hydrogen/Deuterium Exchange Mass Spectrometry Jingjie Mo*, Qingrong Yan, Chi Kwong So, Tam Soden, Michael J. Lewis, Ping Hu* Large Molecule Analytical Development, Pharmaceutical Development and Manufacturing Science, Janssen Research & Development LLC, 200 Great Valley Parkway, Malvern, Pennsylvania 19355, United States. Supporting Information

ABSTRACT: Hydrogen/deuterium exchange mass spectrometry (HDX MS) was used in two case studies to evaluate the impact of methionine (Met) oxidation on the biological functions of IgG1 antibodies. In the first case study, linear correlations were observed between the oxidation of the conserved Fc methionine residues and the loss of neonatal Fc receptor (FcRn) binding and complement-dependent cytotoxicity (CDC) activity. Both heavy chain (HC) residues Met257 and Met433 were located near the FcRn binding interface as indicated by HDX MS and structural modeling; however, HC Met257 oxidation was further demonstrated to have a more significant impact on FcRn binding than HC Met433 oxidation. In addition, oxidation of HC Met257 and HC Met433 could disrupt protein conformation at the CH2-CH3 interface and prevent IgG oligomerization, which is needed for C1q binding and subsequent CDC activity. In the second case study, HDX MS demonstrated that oxidation of the two complementary determining region (CDR) methionine residues had little or no impact on antigen binding of the antibody. Together, these results suggested that HDX MS is a powerful tool for evaluating the impact of individual post translational modifications (PTMs) on the biological activities of antibodies, even when the PTM levels are relatively low. The high selectivity and sensitivity of this method makes it a valuable tool for assisting the critical quality attributes (CQAs) assessment of antibodies.

Monoclonal antibodies are a large and rapidly growing category of biological therapeutics.1-4 There are currently more than forty approved monoclonal antibody (mAb) drugs, most of which are immunoglobulin isotype G (IgG) molecules,5 and more are in review and ongoing clinical trials. IgG antibodies are comprised of an antigen binding fragment (Fab) that recognizes a specific antigen and a crystallizable fragment (Fc) that is important for targeted cell killing and the relatively long half-life of IgG immunoglobulins. Similar to other protein therapeutics, IgG antibodies are susceptible to a variety of chemical modifications.6 Maintaining the chemical and structural integrity of therapeutic antibodies during manufacturing, distribution and storage remains a major challenge for pharmaceutical development. Methionine (Met) oxidation is a very common post translational modification (PTM) that can impact the bioactivity of the antibody and potentially induce an immunogenic response.7 Most IgG1 antibodies have two conserved heavy chain (HC) methionine residues located at the interface of the CH2 and CH3 domains. It has been reported that oxidation of these two methionine residues decreased the thermal stability,8,9 protein A binding,10,11 FcRn binding,10-12 and circulation half-life of IgG1 antibodies.13 There are also antibodies with methionine residues in the complementarity determining regions (CDRs), and oxidation could potentially impact the antigen binding. Therefore, it is essential to characterize the effect of methionine oxidation on the structure, stability, and biological activity of protein drugs.

Although many studies have established correlations between IgG biological functions and methionine oxidation, the connections to specific methionine residues were rarely evaluated due to the difficulty of generating materials with oxidation on a single methionine residue.14,15 In recent years, hydrogen/deuterium exchange mass spectrometry (HDX MS) has been used extensively to study the impact of PTMs on protein conformation,9,12,16-18 and characterize protein/protein interactions.19-22 A few studies have used this technology to evaluate the impact of PTMs on protein/protein interactions.9,23 In this study, HDX MS was used to link the changes in primary structure caused by methionine oxidation with changes in Fc receptor and antigen binding. With the aid of structural modeling, we demonstrated that HDX MS has the resolution to differentiate the impact of oxidation at individual methionine residues, which provides highly valuable information to assist critical quality attributes (CQAs) assessment of antibody therapeutics. HDX MS is also a sensitive technique that can detect the local conformational changes caused by a low percentage of methionine oxidation. This is because oxidation results in a 16 Da mass shift and so the modified and unmodified peptides could be readily resolved by MS and studied separately, suggesting there was no need to generate stress samples with high levels of oxidation for HDX MS analysis.

MATERIALS AND METHODS

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Materials. Two IgG1 mAbs designated as mAb1 and mAb2 were expressed in CHO cells24 and purified from the clarified cell culture media by Protein A affinity and ion exchange chromatography.25 Deuterium oxide (D2O) was purchased from Sigma-Aldrich (99.9% deuterium). All other reagents were HPLC purity grade and purchased from Sigma-Aldrich unless otherwise specified. Sample Preparation. mAb1 was incubated at 30°C for 2 hours in its formulation buffer (25 mM sodium acetate pH 5.5, 60 mM NaCl, 140 mM mannitol, 0.04% (w/v) Polysorbate-20) with additional 0 ~ 750 µM peracetic acid. mAb2 was incubated at 25°C for 7 days in its formulation buffer (10 mM Acetate, 5.0% (w/v) Sorbitol, 0.04% (w/v) Polysorbate-20, pH 5.0) with additional 0 ~ 1500 µM peracetic acid. The oxidized proteins were then exchanged back into their corresponding formulation buffer to remove peracetic acid. Based on a KD of 0.6 µM for the IgG/FcRn interaction,23 the mAb1/FcRn complex was prepared by mixing mAb1 (control or oxidized) and FcRn at a 1:4 molar ratio and then diluting the mixture with buffer E1 (25mM K2HPO4, 25 mM in KH2PO4 and 50mM NaCl in H2O, pH6.2) to a final concentration of 22 µM of mAb1. The mAb2/antigen complex was prepared by mixing mAb2 (control or oxidized) and IL-6 at a 1:2 molar ratio and then diluting the mixture with buffer E2 (25 mM K2HPO4 and 25 mM in KH2PO4 in H2O, pH7.0) to a final concentration of 33 µM of mAb2. Peptide Mapping. Each sample was diluted to a final concentration of 5 mg/mL with 6.0 M guanadine-HCl, 100 mM Tris-HCl, 2.5 mM EDTA (pH8.0), reduced with 25 mM dithiothreitol (DTT) and incubated at 37 °C for 1 hour and then alkylated with 60 mM sodium iodoacetate and incubated at room temperature for 1 hour in dark. Another 35 mM DTT was added to quench the alkylation and the mixture was buffer exchanged with the digestion buffer (50 mM Tris, pH8.0) using NAP-5 column (GE Healthcare). Lys-C was added to the each sample at a 1:50 (enzyme:protein) ratio and incubated at 37 °C for 4 hours. The digestion was quenched by adding 5% (v/v) TFA. The digested samples were then analyzed by LC/MS/MS using an Agilent 1100 HPLC coupled to an ion trap mass spectrometer (Thermo LTQ). Peptide mapping data were analyzed using the Thermo Xcalibur software. Differential Scanning Calorimetry (DSC). Samples were dialyzed against phosphate-buffered saline (PBS) buffer, diluted to 0.5 mg/ml and analyzed with a MicroCal VPCapillary DSC instrument. Scans were collected from 10 to 95 °C at 60 °C/hour and analyzed using the Origin 7.0 MicroCal LLC DSC analysis software. The unfolding transitions of each sample were fit using the non-two state model with 3 (mAb1) or 4 (mAb2) transition midpoints. Hydrogen/Deuterium Exchange Mass Spectrometry (HDX MS). HDX MS analysis was performed using a Waters UPLC HDX system coupled with a Synapt G2-S mass spectrometer. Samples were diluted 10~15 fold into labeling buffer to initiate the HDX reactions (10 fold for the FcRn study and 15 fold for other studies). The labeling buffer used for comparing mAb1 and its FcRn complex was 25 mM K2HPO4, 25 mM KH2PO4 and 50 mM NaCl in D2O (pD6.2). pD was adjusted with DCl or NaOD and the value was calculated by add 0.4 unit to the pH meter reading.26 The labeling buffer used for all other comparison was 25 mM

Page 2 of 9

K2HPO4 and 25 mM KH2PO4 in D2O (pD7.0). The exchange reactions were maintained at 20 °C for various lengths of time between 1 and 300 min. The reactions were then stopped by 1:1 (v:v) dilution with quenching buffer (100 mM K2HPO4, 100 mM KH2PO4, 4 M Guanidine HCl, and 0.7 M TCEP, pH 2.5)27,28 at 1°C. The quenched samples were immediately injected into the Waters HDX system, where they were first digested by flowing through an online immobilized pepsin column (AB applied science, 2.1 × 30 mm), then desalted on a Waters VanGuard precolumn (2.1 × 5 mm), and finally separated on a reversed phase UPLC column (Waters Acquity BEH300 column, 1.0 × 100 mm) holding at 1 °C. The eluent was directly injected into the mass spectrometer and samples were analyzed in the ESI positive mode. The peptides produced by online pepsin digestion were identified in a separate experiment, in which exact mass and MSE data were collected without deuterium labeling and then analyzed using the Waters ProteinLynx Global Server (PLGS) software. The extent of deuterium uptake for each peptide at each time point was calculated using Waters DynamX software. HDX difference charts were plotted to compare any two protein states and statistics analysis was used to set up the criteria based on intermediate precision, which was calculated as the weighted 3SD from 3 replicates of a standard mAb that were run on 3 different days.12,27,29 For this HDX MS system, differences of peptides from 2 samples that met the following two criteria simultaneously were considered statistically significant with a 99.7% confidence interval: (1) the measured individual HDX difference was greater than 0.3 Da for at least one labeling time point and (2) the total HDX difference summed from all time points was greater than 0.8 Da. These criteria served as a general screening tool for locating the differences between two protein states. However, individual comparisons were performed for all peptides of interest using the student t-test because experimental variability differed slightly between different peptides. Structural Modeling. A structural model of the mAb1 Fc/human FcRn complex was created by the Accelrys Discovery Studio 4.0 software using a crystal structure of the rat FcRn/Fc complex (PDB: 1FRT) as a template. The mAb1 hexamer model was built with the crystal structure of the human anti-HIV gp120 IgG1-b12 (PDB: 1HZH) using Pymol software. All structure model figures were generated by Pymol.

RESULTS AND DISCUSSION Impact of Fc Methionine Oxidation on FcRn Binding and CDC Activity. mAb1 contains three methionine residues in the Fc domain (HC Met257, HC Met363 and HC Met433), two methionine residues in the Fab domain (HC Met34 and HC Met83), and no methionine residues on the light chain (LC). Peracetic acid is a reactive oxidant that can break down to hydrogen peroxide.30 When treated with peroxide, antibodies were oxidized dominantly on methionine residues and converted methionine to methionine sulfoxide and similar results were obtained with peracetic acid.31,32 A summary of the peptide mapping results for mAb1 samples treated with peracetic acid are presented in Table S1. Only methionine residues were oxidized and methionine sulfoxide was the only product. As shown in Figure 1A, there were linear increases in oxidation levels for all five methionine residues measured by peptide mapping with increasing concentrations of peracetic

2 ACS Paragon Plus Environment

Page 3 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

acid. In the most oxidized sample, oxidation levels of all three Fc methionine residues were >80%, while the oxidation levels of the two Fab methionine residues were