Structural Changes and Aggregation Mechanisms of Two Different

Subscriber access provided by EKU Libraries

Article

Structural Changes and Aggregation Mechanisms of Two Different Dimers of an IgG2 Monoclonal Antibody Jun Zhang, Christopher Woods, Feng He, Mei Han, Michael J. Treuheit, and David B. Volkin Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b00575 • Publication Date (Web): 24 Aug 2018 Downloaded from http://pubs.acs.org on August 27, 2018

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 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 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.

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 41 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

Biochemistry

Structural Changes and Aggregation Mechanisms of Two Different Dimers of an IgG2 Monoclonal Antibody Jun Zhang1,*, Christopher Woods1,3,†, Feng He1,‡, Mei Han2, Michael J. Treuheit1, David B. Volkin3 1

Process Development, Amgen Inc., Thousand Oaks, CA 91320, USA Pharmacokinetics & Drug Metabolism, Amgen Inc., South San Francisco, CA 94080, USA 3 Department of Pharmaceutical Chemistry; Macromolecule and Vaccine Stabilization Center, University of Kansas; Lawrence, KS 66049, USA 2

*

Address correspondence to

Jun Zhang Process Development, Amgen Inc. One Amgen Center Drive, Thousand Oaks, CA 91320, USA Phone: 805-313-3359 E-mail: [email protected] †

Current address: Omeros Corporation, Seattle, WA 98119, USA

‡

Current address: Shanghai Escugen Biotechnology Co., Ltd. Shanghai, 201210, P.R. China

For Submission to Biochemistry

Key words: monoclonal antibody, immunoglobulin G, dimer, aggregation, protein structure, protein dynamics, hydrogen deuterium exchange, mass spectrometry

ACS Paragon Plus Environment

Biochemistry 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

Page 2 of 41

ABSTRACT

Protein therapeutics, monoclonal antibodies (mAbs) in particular, are large, structurally complex molecules that are prone to numerous modes of degradation during their production and longterm storage. Physical degradation via protein aggregation is a major concern when developing protein therapeutic candidates for clinical use. A dimer is perhaps the simplest element of protein aggregation, and thus, a better understanding of protein dimers in terms of their structures, intermolecular interactions, and chemical nature will help in the development of rational strategies for reducing aggregation propensity. In this study, two different mAb dimers were generated from an IgG2 monoclonal antibody solution, i.e., a native dimer generated under the long-term storage and a thermal dimer from a thermal stress condition. Both IgG2 dimers were characterized for their chemical/physical properties, bioactivity, and conformational dynamics. The native IgG2 dimer was formed mainly through non-covalent association. It displayed minimal differences in biophysical properties and higher order structure compared to the monomer, yet showed compromised in vitro potency, likely due to steric hindrances. In contrast, the thermal IgG2 dimer was mainly disulfide-linked, but even so, no new non-native disulfide bonds were detected by peptide mapping. Two regions within Fc-CH2 domain of the thermal IgG2 dimer exhibited significantly increased flexibility as measured by hydrogen deuterium exchange mass spectrometry, and notably these regions are connected by an intra-chain disulfide bond under natively folded conditions. These findings provide a greater understanding of dimer formation under long-term storage and thermal stress conditions for this IgG2 mAb, and possible aggregation

mechanisms


are

discussed.

Page 3 of 41 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

Biochemistry

INTRODUCTION

Monoclonal antibodies (mAbs), specifically those of the immunoglobulin G (IgG) subclass, have become important protein therapeutic options for treatment of a variety of diseases. As of 2017, more than ~70 mAb drug products have been approved by the regulatory agencies in US and Europe for a wide number of therapeutic uses 1. Compared to small molecule drugs, the complexity of mAb-based drug products are orders of magnitude greater, due to their much higher molecular weight, various post-translational modifications, and their delicate threedimensional structure which dictates their biological activities.

Such complexities create

significant challenges for biopharmaceutical manufacturers to ensure the safety, efficacy, and quality of their therapeutic protein products during manufacturing, long-term storage, transportation, and administration. Of the various physicochemical degradation pathways associated with mAb therapeutics, aggregation has been one of the key concerns during biopharmaceutical development 2-7. Protein therapeutics are often exposed to common environmental stresses such as pH and temperature changes, light exposure, physical agitation and/or contact with various surfaces and materials during the manufacturing processes, long-term storage, and final administration of the drug product to patients. Such exposure tends to alter protein conformation, physical, or chemical properties, and leads to aggregation, particularly in high concentration formulations required for many mAb therapeutic products, especially those to be self-administered in small volumes by patients via subcutaneous injection 6, 8, 9. Because protein aggregates are impurities of the active component itself, their presence may limit protein efficacy and/or promote immunogenic reactions (e.g., formation of neutralizing anti-drug antibodies) 6, 10.


Biochemistry 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

Page 4 of 41

One of the predominant oligomeric species that are found in mAb-containing solutions are dimers, consisting of two mAb monomer molecules (i.e., monomers are defined as intact IgGs with a molar mass of ~ 150 kDa). Dimer formation is commonly thought to be the initial step in protein aggregation leading to formation of larger oligomers and particulates 11. To date, much effort has been devoted to investigating the mechanism of mAb dimer formation and consequently various pathways for mAb aggregation have been proposed in the literature 12-15. In many of these studies, however, spectroscopic techniques were used predominantly, but such analyses lack sufficient resolution to reveal the specific local regions or domains involved in intermolecular contacts in the dimer formation. Recently, hydrogen deuterium exchange coupled with mass spectrometry (HDX-MS) has offered opportunities to obtain higher-resolution information with high sensitivity to study higher order structure of protein therapeutics even in complex drug product dosage forms

16-19

. The bottom-up HDX-MS methodology involves

exposing a protein to deuterium buffer at various time intervals, followed by proteolysis and mass analysis under quenching conditions (low pH and low temperature to prevent deuterium back exchange)20, 21. This approach can achieve localized deuterium incorporation information in a protein down to the peptide level of 5 – 10 amino acid residues. The deuterium uptake information can be improved to single-residue level by using subtraction of deuterium uptake from overlapping peptides22 or the low energy dissociation methods23,

24

, e.g., ETD or ECD.

HDX-MS reports on the hydrogen bonding and solvent accessibility of the protein backbone amide hydrogens perturbations

26-28

25

and is sensitive to protein structural changes caused by different

. Most importantly, key insights into the protein – protein interactions can be

obtained by comparing HDX profiles to map the interfaces for the mAb reversible selfassociation

29, 30

, and pinpoint the regions with structural changes induced by aggregation31, 32.


Page 5 of 41 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

Biochemistry

Understanding protein aggregation at the molecular level could lead to rational design strategies for reducing mAb aggregation during product design, development, and manufacturing. In this study, our focus was on how dimers of an IgG2 mAb (mAb2) formed under two different conditions: (1) during long-term storage at the recommended storage condition, and (2) during short-term exposure to a high temperature thermal stress. To this end, we isolated two dimer species using the mAb2, one from 5°C storage after 2 years and the other from 50°C incubation after 3 days. Subsequently, we employed a comprehensive set of analytical characterization tools including biophysical measurements, potency assays, limited proteolysis, and HDX-MS to better understand the differences between these two mAb2 dimers as well as comparisons to the original monomer molecule. The results presented here provide insights into the role that structural changes in specific regions of this mAb2 play during the initial events when the monomers of this mAb2 begin to aggregate into dimers.


Biochemistry 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 AND METHODS

Sample Preparation A monoclonal antibody IgG2, mAb2 (Amgen, Inc.), was buffer-exchanged into 20 mM acetic acid, pH 5.0, using Slide-A-Lyzer™ Dialysis Cassettes (Life Technologies, Grand Island, NY) and then diluted to 20 mg/mL using the same buffer. The mAb2 solution was then stored at 5°C for two years for the recommended storage condition or at 50°C for 3 days for the thermal stress condition. Analytical size exclusion chromatography (SEC) The relative size distribution of mAb2 monomers and dimers were determined by SEC using a Waters H-Class UHPLC System (Waters Corporation, Milford, Massachusetts) equipped with an ultraviolet (UV) diode array detector. For each sample, 60 µg of protein was injected onto a gel filtration column (ACQUITY UPLC PrST SEC Column, 200Å, 1.7 µm, 4.6 mm X 300 mm; Waters Corporation, Milford, Massachusetts) equilibrated in a running buffer mobile phase comprised of 100 mM sodium phosphate, 250 mM NaCl, pH 6.8 under isocratic conditions with a flow rate of 0.4 mL/min. Chromeleon® 7.2 Chromatography Data System (Dionex, Sunnyvale, California) was used to analyze results and peak integration was calculated using the UV absorbance of 280 nm. A reference standard mAb sample was injected in triplicate at the beginning and end of each sequence to determine the SEC assay variability. Protein fractionation using SEC The monomer and dimer species of mAb2 were fractionated using an ÄKTAexplorer 100 FPLC instrument equipped with a fraction collector (GE Healthcare Bio-Sciences, Pittsburgh, PA) and a HPLC gel filtration column (TSKgel G3000SW, 250Å, 13 µm, 21 mm X 60 cm, TOSOH Bioscience LLC, King of Prussia, PA). The instrument was programmed to begin collecting 1


Page 6 of 41

Page 7 of 41 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

Biochemistry

mL fractions into a 96 deep well plate when the absorbance at 280 nm exceeded 0.20 mAU. Multiple injections were made to generate sufficient amounts of the monomer and dimer fractions. Upon completion of all injections, individual wells corresponding to the apexes of either monomer or dimer peaks were pooled together and concentrated to 3 mg/mL protein using Amicon Ultra-15 Centrifugal Filter Units (EMD Millipore, Billerica, MA). Monomer and dimer fractions of mAb2 were buffer exchanged to an acetate buffer containing sucrose at pH 5.0 for future analyses. SEC Coupled with Online Multi-Angle Light Scattering Detection (SEC-MALS) SEC-MALS analysis was performed to determine the molar mass of monomer and dimer species using an Agilent 1100 HPLC system with a TSK-GEL G3000SWxl column (5 µm particle size, 7.8 mm ID x 300 mm length; Tosoh Biosep, 08541) with a Metasaver 0.5 µm pre-column filter (Varian, A6005). The three detectors used included a Wyatt HELEOS MALS detector (light scattering), a Wyatt Optilab rEX RI detector (refractive index), and an Agilent UV detector with wavelength set at 280 nm. The SEC-MALS runs were performed at room temperature, with 100 mM sodium phosphate, 250 mM sodium chloride, pH 6.8 buffer used as the mobile phase, and the flow rate was set to 0.5 mL/min. Sample volumes were adjusted as necessary to ensure that 300 µg of each sample was injected, without dilution, into the SEC-LS system. For molecular weight (MW) calculation, instrument software was used with an extinction coefficient value for the mAb species of 1.5 mL/(mg*cm) at 280 nm. Potency Assay The potency assay for mAb2 is a reporter gene bioassay utilizing a human erythroleukemic cell line transfected with a reporter gene construct. These cells express mAb2 antigen receptor and are activated in the presence of the antigen. Upon antigen binding to the receptor, transcription


Biochemistry 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

factors are activated which lead to the transcription of luciferase reporter gene. Steady-Glo® luciferase assay substrate reagent was added to the plates and then they were read in a luminometer (EnVision). The amount of luciferase activity is directly proportional to the amount of antigen and inversely proportional to the concentration of mAb2. The biological activity of mAb2 is determined by normalizing test samples to a reference standard. The potency assay was performed in triplicate for each sample. FcRn Binding Activity The cell-based FcRn binding assay was developed in a competitive binding format to test the binding of the Fc moiety of monoclonal antibodies to FcRn. The assay utilized a variant of human embryonic kidney cell line, 293T (293 cells expressing SV40 large T antigen), developed internally by Amgen, which expressed FcRn on the cell surface. Varying concentrations of the mAb2 test samples and the reference standard were incubated with FcRn-expressing cells and a fixed concentration of Alexa488®-labeled IgG-Fc at room temperature and at pH 6. After the incubation, the assay plate was read on an acumen® Cellista laser scanning imaging cytometer (TPP Labtech Inc, Cambridge, MA) for cell bound fluorescence. Fluorescence data from each well were recorded and analyzed using Softmax Pro version 5.4.1 (Molecular Devices, LLC, Sunnyvale, CA). After assessing similarity between response curves of test samples and a reference standard, the test sample binding relative to the reference standard was determined and the results were reported as percent relative binding (% relative binding). Each sample test was performed in triplicate and the averaged % relative binding and standard deviation were reported. Sedimentation Velocity measured by Analytical Ultracentrifugation (SV-AUC) Sedimentation velocity analytical ultracentrifugation was used as an orthogonal approach to SEC for size distribution analysis using a Proteomelab XL-I analytical ultracentrifuge instrument


Page 8 of 41

Page 9 of 41 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

Biochemistry

(Beckman Coulter, Fullerton, CA). 12-mm charcoal-filled Epon 2-channel centerpieces (Beckman Coulter) and 4-hole An60 Ti analytical rotor pre-equilibrated to 20.0 °C were used. Experiments were conducted at a temperature of 20.0 °C and absorbance was recorded at 280 nm with radial scan increment of 0.003 cm. The rotor angular velocity was 45,000 rpm. The data were analyzed by the continuous c(s) distribution model in Sedfit (version 9.4) 33 using fitting parameters described previously for monoclonal antibodies 34. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) SDS-PAGE was performed using 4-20% Tris-Glycine Gel from Bio-Rad Laboratories, Inc. (Hercules, CA). Samples were prepared in non-reduced and reduced forms. For each sample, 10 µg of protein was mixed with 2x Tris-Glycine SDS sample buffer at 1:1 ratio in the presence of 15 mM iodoacetimide for non-reduced sample preparation and 25 mM dithiothreitol (DTT) for reduced sample preparation. All sample mixtures were then heated at 70oC for 10 minutes. A constant voltage of 150V was applied for 40 minutes. The gel was then stained with Simply Blue Safe Stain from Life Technologies (Grand Island, NY) and destained with water. Limited Proteolysis by FabRICATOR® Enzyme Approximately 60 µg of each sample was digested in 30 µL of 10 mM acetate, pH 5.2 containing 60 units of FabRICATOR enzyme (IdeS). The reaction was incubated at 37°C overnight. For reduced samples, the digested material was treated with a buffer containing 4 M Guanidine-HCl, 50 mM Tris, pH 8.3, with 50 mM DTT. Reduced and non-reduced digests were directly separated by reversed phase chromatography with a BEH Phenyl 2.1 X 150 mm column (Waters Corporation, Milford, Massachusetts) and the mass measurements were obtained using a Waters Premier Q-Tof mass spectrometer with the resolving quadruple of 4k Da (Waters Corporation, Milford, Massachusetts).


Biochemistry 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

Disulfide peptide mapping with mass spectrometry The mAb2 samples were denatured by diluting 33 µL of 3 mg/mL proteins in 90 µL of 8 M Guanidine-HCl, 10 mM N-Ethylmaleimide with 100 mM sodium acetate, pH 5.2, and incubated at 37 C for 3 hours. The mixture was then diluted with 325 µL of 4 M urea, 20 mM hydroxylamine, 100 mM Tris, pH 7.0, and Lys-C was added to the mixture to achieve an enzyme to substrate ratio of 1:20. To prepare reduced digested samples, 200 µL of each digests were added with 4 µL of 0.5 M TCEP solution and incubated at room temperature for 30 min. The reduced and non-reduced Lys-C digests were analyzed by reversed phase HPLC with mass spectrometry using a Waters Acquity UPLC with an Orbitrap mass spectrometer (Elite, ThermoFisher Scientific, San Jose, CA). Approximately, 30 µg of protein digest was injected onto a Waters BEH300 C4 2.1 X 150 mm column. Mobile phase A was 0.1 % TFA in water and mobile phase B was 0.1% TFA in 90% acetonitrile. The column was equilibrated with 2% B using a flow rate of 0.2 mL/min. After sample injection, the column was washed with 2% mobile phase B for 5 minutes. A linear gradient to 20% mobile phase B over 35 minutes was applied, followed by another linear gradient of 20 to 40% B over next 80 minutes. Hydrogen Deuterium Exchange Reactions and Mass Spectrometric Analysis (HDX MS) HDX MS experiments were performed with a Twin HTS PAL liquid handling robot (LEAP Technologies, Carrboro. NC) interfaced with an Orbitrap mass spectrometer (Elite, ThermoFisher Scientific, San Jose, CA), as previously described 35, 36. The protein concentration was adjusted with 10 mM acetate (pH 5.2) to 3 mg/mL. The H/D exchange reaction was initiated by 5-fold dilution of 3 mg/mL protein samples with 10 mM acetate in D2O (pD 5.2) as indicated for a predetermined time (10, 30 s, 1, 10 min, 1, and 4 h) at 25 °C. The exchange reaction was quenched by mixing 1:1 with ice-cold 200 mM sodium phosphate, 4 M guanidine HCl, 0.5 M


Page 10 of 41

Page 11 of 41 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

Biochemistry

Tris(2-carboxyethyl)phosphine (TCEP), pH 2.4. The quenched protein mixture was passed over a custom-packed 2 mm × 2 cm pepsin (Fisher Scientific, Pittsburgh, PA) column (Agilent Technologies, Santa Clara, CA) at a flow rate of 200 µL/min. Digested peptides were captured on a 2 mm × 1 cm C18 trap column (Waters Corporation, Milford, MA) and desalted for 3 minutes at a flow rate of 0.2 mL/min. Peptides were then separated by using a 2.1 mm × 5 cm C18 column (1.9 µm Hypersil Gold; Thermo Fisher Scientific, Waltham, MA) with a 9.5 minute linear gradient of 5–40 % acetonitrile in 0.1% formic acid at a flow rate of 0.2 mL/min. Protein digestion and peptide separation were carried out in thermal chamber maintained at 1°C to reduce back exchange. LC-MS data were acquired with a mass resolving power of 60,000 for ions of m/z 400. Each experiment was performed in duplicate to provide an estimate of variation. Tandem mass spectrometry (MS/MS) experiments were performed under the same conditions as described above. Product-ion spectra were acquired in a data-dependent mode, and the 10 most abundant ions were selected for product-ion analysis. All data were processed with the software MassAnalyzer37 for the peptide identification and the deuterium level calculation. Approximately 500 peptides were analyzed with sequence coverage of at least 97% for all mAb2 polypeptide chains. All HDX-MS data were normalized to 100% deuterium concentration and the percent deuterium incorporation was plotted against labeling time in log scale with Prism v 6.02 (Graphpad Software, La Jolla, CA).


Biochemistry 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

RESULTS Size Analyses of mAb2 HMW Species Samples of the mAb2 solution stored under two different conditions were analyzed by SEC. The apparent size distribution by SEC (Figure 1A) indicates the predominant population remains monomer. For mAb2 solution stored at 5°C after 2 years (black trace in Figure 1A), there is a small amount of higher molecular weight (HMW) species that elute before the monomer. As a comparison, thermal incubation of the same mAb2 solution at 50°C for 3 days resulted in a significant increase in the HMW peak (red trace in Figure 1A). Moreover, there was a small decrease in total peak area (6.1%) for the thermally stressed sample compared to the sample stored at 5°C for 2 years, although the absolute amount of protein injected was the same, which suggests that insoluble HMW aggregates might have formed which are too large to enter the column for SEC analysis. The relative abundances of the HMW species for mAb2 under these two conditions are 1% and 5.5%, respectively (Figure 1C). To better understand the composition of the HMW species, we isolated the HMW species using SEC fractionation. After fraction collection, sample fractions were analyzed by SEC, as shown in Figure 1B. HMW fractions, determined to be dimers as shown below, regardless of the conditions by which they were generated, retained a certain portion of monomer (Figure 1C), and thus were enriched dimers but not completely purified. It is unclear whether the monomer present in the dimer fractions is a result of co-purification or due to disassociation of the HMW species upon purification. Noticeably, fractionated samples showed wider SEC peaks compared to the bulk materials prior to fractionation. The increase peak width is likely a result of the larger injection volume 38. It should be noted that the bulk material is at 20 mg/mL and the fractionated


Page 12 of 41

Page 13 of 41 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

Biochemistry

sample is at 3 mg/mL. Thus, the injection volumes are 3 µL and 20 µL, respectively, given the constant injection amount of 60 µg for SEC analyses. A

B

C Relative Peak Area Percentage (%) Before fractionation

After fractionation 2y at 5°C

Species 2y at 5°C

3d at 50°C

Dimer

1.0

Monomer

98.9

3d at 50°C

5.5

Monomer fraction 0.1

Dimer fraction 91.1

Monomer fraction 0.2

Dimer fraction 82.3

94.0

99.9

8.9

99.8

17.7

Figure 1. Size analyses of mAb2 species as measured by SEC. (A) Chromatogram overlay of mAb2 samples at 5°C after 2 years, in black, and at 50°C after three days, in red. (B) Chromatogram overlay of mAb2 samples after fractionation. Black traces represent the enriched fractions from the native sample after two-year storage at 5°C and red traces represent the enriched factions from the thermally stressed sample after three-day storage at 50°C. The monomer fractions are in solid traces and the dimer fractions are in dashed traces. (C) Relative percentages of peak area for dimer and monomer species in mAb2 solution, as well as in monomer and dimer fractions after purification. The error of the SEC method calculated using multiple injections of a reference standard was ± 0.3%.


Biochemistry 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

Page 14 of 41

SEC-MALS was used to assess the molar mass of the fractionated HMW species. It is interesting that the molar mass for each of the HMW species, formed under two different conditions, is ~ 300 g/mol (Table 1), twice the monomeric IgG mAb molar mass of ~150 g/mol, suggesting that the HMW species collected from each storage condition for mAb2 is dimer. Sedimentation Velocity measurements by Analytical Ultracentrifugation (SV-AUC) was also performed as an orthogonal approach to SEC. SV-AUC was able to detect two species, e.g., the monomer and dimer, and no other species, nor larger aggregates or protein fragments, were detected (Supplemental Figure S1). Both SEC-MALS and SV-AUC were effective at distinguishing monomer from dimer; however, no meaningful observations can be made with regard to potential size difference between the two mAb2 dimers generated under different conditions (Table 1). Table 1. Summary of key size and potency values of mAb2 monomer, native dimer, and thermal dimer. The average values and the standard deviations were reported as mentioned in the Methods. Average Relative Average Relative Potency (%) FcRn Binding (%)

Molar Mass (g/mol)

Sedimentation Velocity (S) *

Native Dimer

288 ± 1.7

9.3

49 ± 3

128 ± 8

Thermal Dimer

293 ± 1.6

9.3

28 ± 1

44 ± 8

Monomer

150 ± 0.8

6.4

114 ± 13

99 ± 14

Samples

* Only single replicate was performed.

For simplicity in nomenclature, the dimer fractions collected from mAb2 bulk stored at 5°C for 2 years and incubated at 50°C for 3 days were denoted as “native dimer” and “thermal dimer”, respectively. The native dimer and thermal dimer of this mAb2 were characterized and compared to the original mAb2 monomer for all subsequent studies. Biophysical Characterization of mAb2 Dimer Species In order to differentiate two dimer species, we performed biophysical characterizations on fractionated samples. Firstly, DSC analyses were performed to assess thermal stability. There


Page 15 of 41 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

Biochemistry

was no differences in Tonset, which marks where the first structural transition initiates. The determined Tonset was ~ 61°C, which is much higher than the temperature (50°C) used for the thermal stress. Therefore, the thermal stress did not think to unfold the mAb2. In addition, there was only one transition midpoint (Tm) of thermal unfolding observed, and no differences were detected in thermal melting behavior between monomer and dimer species. The average Tm value based on duplicate runs was 75.2°C and representative thermograms are shown in Supplemental Figure S2. In addition, several spectroscopic based techniques were utilized to detect possible structural differences between monomer and dimer species (Supplemental Figure S3). The secondary structure was examined by far-UV CD and FT-IR spectroscopy and the results are shown in Supplemental Figure S3A and Supplemental Figure S3C, respectively. The spectra for both monomer and dimers are highly similar by visual comparison, suggesting that the overall secondary structure content was not notably altered. Likewise, nearUV CD (Supplemental Figure S4B) was not able to detect any differences in the overall tertiary structure between monomer and dimer species. These results suggest that the global structural differences between the two mAb2 dimers are rather small, and likely any notable structural differences are local in nature. Interestingly, an increase in fluorescent signal intensity was observed for the thermal dimer compared to monomer and native dimer using SYPRO® Orange dye, as shown in Supplemental Figure S4D. Since SYPRO® Orange dye undergoes a significant increase in quantum yield upon binding hydrophobic environments

39-42

, the increase

in fluorescent signal intensity indicated that the thermal dimer displayed increase exposure of hydrophobic surfaces compared to either the original monomer or the native dimer of the mAb2. Activity Comparison of mAb2 monomer and dimers


Biochemistry 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

Potency and neonatal Fc receptor (FcRn) binding assays were performed for the mAb2 monomer and the two dimers to evaluate the impacts of dimer formation on in vitro binding activities in the Fab and Fc regions, respectively. The results are shown in Table 1. Both the native and thermal dimer species were found with significantly reduced potency (49% and 28%, respectively) compared to monomer, suggesting that the target antigen binding in the Fab regions has been compromised in both dimers as compared to the monomer. On the other hand, the native dimer of mAb2 exhibited comparable FcRn binding relative to the monomer (128% vs. 99%), while the mAb2 thermal dimer showed a significant reduced ability to bind FcRn receptors (44%). FcRn binding results indicated that dimerization induced thermally might have resulted in altered local structure of the Fc region responsible for the Fc receptor binding, and this effect was not observed for the native dimer generated by long-term storage. Characterization of mAb2 Dimers by SDS-PAGE To understand the chemical nature of interactions between subunits for two dimers, we performed SDS-PAGE under both reducing and non-reducing conditions, as shown in Figure 2. When the samples were not reduced, the thermal dimer showed two bands, a major one at ~300 kDa and a relatively weaker one at ~150 kDa (lane 4), corresponding to dimer and monomer species, respectively. The majority of the thermal dimer could not be dissociated in SDS solution, indicating that the thermal dimer is mainly covalently bonded. As a comparison, the native dimer showed a strong band of monomer and a weak band of dimer (lane 6), suggesting that native dimer is mainly non-covalently bonded. These samples were also analyzed under reducing condition. All reduced samples only showed two main bands with lower molecular weight (lanes 9-12), corresponding to the heavy chain (HC) and light chain (LC). The covalently


Page 16 of 41

Page 17 of 41 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

Biochemistry

associated dimer species appear sensitive to reducing reagents, therefore the dimer is apparently disulfide-linked.

Figure 2. SDS-PAGE of mAb2 monomer and dimer fractions under non-reducing and reducing conditions. SeeBlue Standard Plus2 molecular weight ladder is shown in lane 1. Lanes 3-6 are from non-reduced samples and Lanes 912 are from reduced samples. From left to right, samples include mAb2 monomer, thermal dimer, monomer and native dimer, respectively. Lanes 2, 7 and 8 are blank controls.

Disulfide Mapping of mAb2 Monomer and Dimer Samples To determine whether the disulfide bond scrambling took place during dimer formation and where it might occur, we conducted peptide mapping for mAb2 monomer and the two different dimers. The base peak chromatograms of reduced (Figure 3A) and non-reduced (Figure 3B) mAb2 monomer and dimers were almost identical for both native and thermal samples, with some small differences in intensity observed (likely attributable to method variability). All cysteine-containing peptides were identified in the reduced Lys-C peptide maps and the expected disulfides were also detected in the non-reduced Lys-C peptide maps. In addition, there were no new peptides detected in the thermal dimer compared to the monomer, indicating no evidence of free Cys residues or newly formed disulfide bonds. Moreover, peptide mapping only detected low levels of increase in chemical modifications induced by dimerization (

Structural Changes and Aggregation Mechanisms of Two Different

Recommend Documents