Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Feb 1.
Published in final edited form as: J Pharm Sci. 2016 Jan 11;105(2):575–587. doi: 10.1016/j.xphs.2015.10.024

Comparative evaluation of the chemical stability of four well-defined IgG1 Fc glycoforms

Olivier Mozziconacci 1, Solomon Okbazghi 1, Apurva S More 1, David B Volkin 1, Thomas Tolbert 1, Christian Schöneich 1,*
PMCID: PMC5118937  NIHMSID: NIHMS737274  PMID: 26869420

Abstract

As part of a series of papers in this special issue evaluating model IgG1-Fc glycoforms for biosimilarity analysis, three well-defined IgG1-Fc glycoforms (High mannose-Fc, Man5-Fc, GlcNAc-Fc), and a non-glycosylated Fc protein (N297Q-Fc) were examined in this work to elucidate chemical degradation pathways. The four proteins underwent a combination of accelerated thermal stability studies, and four independent forced degradation studies (UV-light, metal-catalyzed oxidation, peroxyl radicals, and hydrogen peroxide) at pH 6.0. Our results highlight chemical degradations at Asn315, Met428, Trp277, and Trp313. A cross-comparison of the different Fc glycoforms, stress conditions, and the observed chemical reactions revealed that both the deamidation of Asn315 and the transformation of Trp277 into glycine hydroperoxide were glycan-dependent during incubation for three months at 40°C. Our data will show that not only different glycans affect chemical degradation differently, but also do lead to different impurity profiles, which can affect chemical degradation.

Keywords: antibody, biosimilar, glycoform, oxidation, stability, UV-light, H2O2, metal, peroxide

1. Introduction

Antibodies (i.e., immunoglobulins) are essential mediators of the immune system.1 Among these immunoglobulins, immunoglobulin G (IgG) is the most common protein involved in the immune response while additional isotypes have been identified in placental mammals such as IgA, IgD, IgE, and IgM.2 The various immune cells are activated through the binding of IgG/antigen complexes to membrane-bound Fcγ receptors (FcγRs) present on the immune cells (e.g., mast cells, macrophages, and neutrophils).3 It was initially thought that the immune response involved solely the recognition of antigenic proteins, driving the search for new epitopes toward the design of synthetic peptides with sequences similar to that of target proteins of interest.4 However, it is now well established that the nature of the glycans is important for the modulation of the immune response.5 Glycosylation of both IgG and FcγR modulates the interaction of these proteins.6,7 The glycosylation of Fc domains in human IgG occurs at asparagine 297 (Asn297) within the CH2 domain.8,9 Usually, the carbohydrate chain is a heptasaccharide consisting of N-acetylglucosamine (GlcNAc), and mannose (Man). The polysaccharide is often enriched with galactose, sialic acid, fucose and bisecting GlcNAc residues.

Although the glycan does not directly interact with the FcγRs,10 the absence or truncation of the polysaccharide chain changes the affinity of the IgG forms toward the FcγRs.11 Such modulation of the affinity between IgGs and FcγRs is explained by conformational changes in the CH2 domains. A direct or allosteric interaction between carbohydrates attached to both IgG and the Fcγ receptor might also be crucial for the modulation of affinity.12 For example, deglycosylated murine IgG1 shows a higher affinity towards the activating FcγRIII, whereas the affinity towards the inhibiting FcγRIIIb receptor is reduced,13 providing a rationale for the role of deglycosylated antibodies in an autoimmune response. In contrast, an elongation of the Fc-oligosaccharide chain (e.g., through sialylation) reduces the affinity of IgG1 to the FcγRs, resulting in an anti-inflammatory activity.14,15

The immune system is particularly sensitive to protein surface recognition.16 Therefore, any chemical change, which would modify the primary or higher-order structure of an IgG, could not only affect its stability, but also its potency and immune potential. Monoclonal antibodies have become a therapeutic option for various diseases.17,18 The success of immunoglobulins as therapeutics has resulted in the development of multiple novel monoclonal antibodies. Antibodies can be engineered to improve antigen binding affinity, pharmacokinetics, pharmaceutical properties and to address safety issues.19 Especially, the engineering of Fc domains to modulate antibody affinity to FcγRs has become a major objective over the last decade.20 As glycosylation evolved as an important parameter regulating the functions of human IgG,21 strategies to modify the glycosylation profile of human IgG1 in order to isolate the most efficient glycoforms have been explored.22-24

Protein stability represents an important concern for the development of protein therapeutics, where chemical and/or physical degradation may lead to loss of potency, aggregation, fragmentation and/or immunogenicity.25 Importantly, the chemical degradation of glycoproteins may not only involve covalent modifications of amino acid residues but also of the carbohydrate moiety. In addition, conformational properties of the protein influenced by the nature of the glycans may control chemical degradation reactions. However, little systematic research has addressed the role of glycan nature for the chemical stability of IgG. With the increasing interest in the development of biosimilars such questions will certainly become significant26 and may raise important concerns for immunogenicity.27-30

As part of a series of papers in this special issue of the Journal of Pharmaceutical Sciences, the use of four well-defined IgG1-Fc glycoforms as a model system for developing a mathematical approach for biosimilarity analysis is described. The purpose of this study is to evaluate and compare the chemical stability of these model Fc glycoforms. The other three papers in this series include (1) expressing and purifying IgG1-Fc glycoforms from yeast with high mannose glycosylation, followed by truncation by in vitro enzymatic synthesis, and characterizing their biological activities (see Okbazghi SZ et al),31 (2) determining the physical stability profiles of the four Fc glycoforms under a variety of solution conditions (see More AS et al),32 and (3) performing mathematical modeling of biosimilarity assessments using a machine learning approach with stability data generated with these model IgG1-Fc glycoforms (see Kim JH et al).33

In order to evaluate whether glycan nature has an effect on chemical stability of the model, well-defined IgG1-Fc glycoforms, we subjected them to accelerated degradation (storage at different temperatures) as well as forced degradation by UV-light, metal-catalyzed oxidation, peroxyl radicals, and hydrogen peroxide. The amino acid sequence of the IgG1-Fc is presented in Chart 1. Three separate glycoforms were prepared which contained either a high mannose carbohydrate structure (HM-Fc), containing 8-12 Man units, pentamannose (Man5-Fc), containing (Man)5(GlcNAc)2, and GlcNAc (GlcNAc-Fc). A deglycosylated Fc domain was prepared by expression of an Asn297Gln mutant (N297Q-Fc), which prevented glycosylation because of the absence of Asn297. The nomenclature and nature of the glycans are summarized in Table 1. All glycoforms were fully glycosylated, as confirmed by mass spectrometry.31 Chemical degradation was detected, and compared for all glycoforms, at Asn315 (deamidation), Met428, Trp277, and Trp313 (oxidation). We also evaluated the presence of 3,4-dihydroxyphenylalanine (DOPA). The chemical degradation products observed are discussed in terms of their various locations and the nature of the glycans.

Chart 1.

Chart 1

Open in a new tab

Amino acid sequence and tryptic cleavage sites of the IgG1-Fc. Tryptic peptides are labeled T1 to T16 as indicated in the figure. The T5* indicates the tryptic peptide glycosylated at Asn297. The amino acid sequence is same for the HM_Fc, Man5_Fc, and GlcNAc_Fc. For the mutant (N297Q_Fc) Asn297 (in T5*) was replaced by Gln297.

Table 1.

Labels and carbohydrate sequences at the N-linked glycosylation site (N297) of the four different IgG1-Fc glycoforms. The Fc glycoforms were prepared and characterized by intact protein mass spectrometry analyses as described in detail elsewhere (see Okbazghi SZ et al,31 a companion paper in this issue).

Label Carbohydrate sequence
HM From (Man)8 to (Man)12
Man5 (Man)5(GlcNAc)2
GlcNAc (GlcNAc)1
N297Q No glycosylation
Open in a new tab

2. Materials

2.1. Chemicals

Histidine, sucrose, sodium phosphate (NaH2PO4), iodoacetamide (IAA), ammonium bicarbonate (NH4HCO3), bis-(2-mercaptoethyl) sulfone (BMS), CuCl2, sodium ascorbate (Asc), sodium phosphate (NaH2PO4), potassium ferricyanide (K3Fe(CN)6), guanidine hydrochloride (GuHCl), 2,2′-Azobis(2-methylpropionamidine) dihydrochloride (AAPH), hydrogen peroxide (H2O2, 30%), and ethylenediaminetetraactic acid (EDTA) were supplied by Sigma-Aldrich (St Louis, MO) at the highest purity grade. Sequencing trypsin/LysC grade was supplied by Promega (Madison, WI). 4-(Aminomethyl) benzenesulfonic acid (ABS) was prepared according to a published protocol.34

2.2 Fc glycoforms

The Fc glycoforms were prepared as described elsewhere (see Okbazghi SZ et al, accompanying paper in this issue).31 Especially for purification of HM-Fc and Man5-Fc, the proteins were divided into two batches where dialysis was performed either in the absence or presence of 5 mM EDTA prior to chromatographic purification. The dialysis with 5 mM EDTA was included in order to evaluate the role of potentially contaminating redox-active transition metals in the chemical degradation. After purification, each of the four Fc glycoforms (0.2 mg/mL) were formulated in either a His/sucrose buffer (10 mM His, 85 mg/mL sucrose, pH 5.8) and stored at -80C. Upon thawing, samples were analyzed in the His/sucrose buffer or were dialyzed into a citrate-phosphate buffer (20 mM citrate-phosphate at pH 6.0 with the ionic strength adjusted to 0.15 using NaCl) (as described by More AS et al, in an accompanying paper in this issue).32

3. Methods

3.1 Oxidation of the Fc glycoforms

The four different Fc glycoforms (HM-Fc, Man5-Fc, GlcNAc-Fc, and N297Q-Fc) were exposed to chemical degradation according to five different protocols: i) UV-light, ii) metal-catalyzed oxidation, iii) peroxyl radicals generated via thermal decomposition of AAPH, iv) hydrogen peroxide, and v) longer-term incubation at 4°C and 40°C over several months.

3.1.1 Photo-irradiation

Fc glycoforms (0.16 mg/mL), formulated either in the presence of 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM of both citrate and phosphate buffer, pH 6.0, were placed in pyrex tubes for irradiation with λmax= 305 nm. For these experiments, a volume of 800 μL of the Fc protein solution (0.2 mg/mL) was mixed with either 200 μL of the His/sucrose solution (10 mM His and 85 mg/mL sucrose, pH 5.8) or 200 μL of the citrate-phosphate buffer (20 mM, pH 6.0). Photo-irradiation was performed by means of a UV-irradiator (Rayonet®, The Southern New England Ultraviolet Company, Branford, CN) equipped with four UV-lamps emitting at λmax= 305 nm (RMR-300Å) delivering a power of approximately 5 W/cm2. The lamps emitting at λmax= 305 nm have an emission spectrum between 280 nm and 330 nm. Thus, the samples were placed in pyrex tubes, which have a cut-off at ca. 290 nm, to avoid the absorption of photons with λ < 290 nm.

3.1.2 Metal-catalyzed oxidation

The Fc glycoforms (0.16 mg/mL) were oxidized in the presence of Cu(II) and Asc under air. Stock solutions of CuCl2 (1.25 mM) and sodium ascorbate (5 mM) were prepared in the presence of 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM citrate-phosphate buffer, pH 6.0. The reaction mixtures were prepared by the addition of 800 μL of the Fc glycoform stock solution (0.2 mg/mL in 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM citrate-phosphate buffer, pH 6.0) to 100 μL of CuCl2 (1.25 mM) and 100 μL of sodium ascorbate (5 mM). The final solution was incubated at 37°C for 4 hours. After four hours, the reaction mixtures were centrifuged (13,000 g for 20 min) using Amicon ultra-0.5 centrifugal unit devices (Millipore Corp. Bedford, MA, USA, cut-off 10kDa) to isolate the Fc glycoforms.

3.1.3 Oxidation by peroxyl radicals

The Fc glycoforms (0.16 mg/mL) were exposed to peroxyl radicals generated through the thermal decomposition of AAPH (1mM) in air-saturated aqueous solutions. A volume of 200 μL of AAPH stock solution (5 mM AAPH in 10 mM His and 85 mg/mL sucrose, pH 5.8, or 5 mM AAPH in 20 mM citrate-phosphate buffer, pH 6.0) was mixed with 800 μL of the Fc glycoform solution (0.2 mg/mL in 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM citrate-phosphate buffer, pH 6.0). The reaction mixtures were incubated at 37°C for 4 hours. The decomposition of one molecule of AAPH leads to the formation of two carbon-centered radicals (R•), which ultimately generate peroxyl radicals (ROO•) in the presence of oxygen. Peroxyl radicals are generated at a rate of d[ROO•]/dt = 1.36×10-6 [AAPH] (Ms-1) at 37°C.35 After 4 hours, the reaction mixtures were centrifuged (13,000 g for 20 min) using Amicon ultra-0.5 centrifugal unit devices (Millipore Corp. Bedford, MA, USA, cut-off 10kDa) to isolate the Fc glycoforms.

3.1.4 Oxidation by hydrogen peroxide

The Fc glycoforms (0.16 mg/mL) were exposed to 1 mM H2O2 in air-saturated aqueous solution. A volume of 200 μL H2O2 stock solution (5 mM H2O2 in either 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM citrate-phosphate buffer, pH 6.0) was mixed with 800 μL of the Fc protein solution (0.2 mg/mL in either 10 mM His and 85 mg/mL sucrose, pH 5.8, or in 20 mM citrate-phosphate buffer, pH 6.0). The reaction mixtures were incubated at 37°C for 4 hours. H2O2 was removed from the samples by centrifugation (13,000 g for 20 min) using Amicon ultra-0.5 centrifugal unit devices (Millipore Corp. Bedford, MA, USA, cut-off 10kDa) to isolate the protein.

3.1.4 Long-term thermal stability

The four IgG1-Fc glycoforms (0.2 mg/mL in 20 mM citrate-phosphate buffer, pH 6.0, containing NaCl to adjust ionic strength to 0.15) were incubated in stoppered glass vials (in the upright position) for 2, 4, and 12 weeks at 4°C and 40°C.

3.2 Analysis of the oxidation products

3.2.1 Digestion protocol

For peptide mapping of the Fc glycoforms, 100 μL of either the controls or the stressed samples were mixed with 250 μL of GuHCl solution (6 M, in sodium phosphate buffer, 110 mM, pH 8.2) and 50 μL of BMS solution (5 mM) for disulfide reduction. The mixtures were incubated at 50°C for one hour. The thiolate groups of the reduced cysteine residues were then alkylated by the addition of 50 μL of IAA (25 mM). After a second hour of incubation at 50°C, 50 μL of the reduced and alkylated samples were mixed with 300 μL of ammonium bicarbonate buffer (150 mM, pH 8.2). 1.8 μg of trypsin/LysC were added to each sample prior to incubation at 37 °C for two hours. Subsequently, a second aliquot of 1.8 μg of trypsin/LysC was added to each sample, and the incubation was continued for 6 hours at 37°C. After six hours, 30 μL of formic acid (10% v:v in H2O) were added to each sample to stop the digestion.

3.2.2 UPLC-nano-LC and nano-electrospray ionization time-of-flight MS and MS/MS analysis

LC-MS analyses of the peptide maps were performed on a NanoAcquity ultra performance liquid chromatography (nanoAcquity-UPLC) system (Waters Corporation, Milford, MA). 1 μL of each sample was injected onto a Symmetry C18 Waters trap column (2G-V/MTrap, 180 μm × 20 mm, 5 μm) connected to an analytical nanoAcquity ultra performance Waters column (HSS T3, 75 μm × 150 mm, 1.8 μm). Mobile phases consisted of water/acetonitrile/formic acid at a ratio of 99%, 1%, 0.08% (v:v:v) for solvent A and a ratio of 1%, 99%, 0.06% (v:v:v) for solvent B. The injected sample was first loaded onto a trapping column at 8 μL/min for 4 min with 97% of solvent A and 3% of solvent B. After 4 min, the peptides were directed towards the analytical column, and were separated with a linear gradient (3-35% of solvent B within 130 min) delivered at a rate of 0.3 μL/min. Mass spectrometry was performed on a Waters Xevo-G2 (Waters Corp., Milford, MA) operating in the positive mode. The desolvation gas flow and the desolvation temperature were set to 1000 L/h and 100 °C, respectively. A nanoflow gas pressure was set to 0.2 bar, with a cone gas flow set to 4 L/h and a source temperature of 95 °C. The capillary voltage and cone voltage were set to 2800 V and 35 V, respectively. The Xevo-G2 acquisition rate was set to 0.5 s with a 0.0 s inter-scan delay. Argon was employed as the E collision gas. The instrument was operated in the MS mode. The instrument was operated with the first resolving quadrupole in a wide pass mode with the collision cell operating with different alternating energies. To acquire the non-fragmented MS1 spectrum the collision cell was operated at 5 eV. The fragmented MS1 ion spectra were acquired by ramping the collision cell energies from 18 eV to 55 eV. The data were collected into separate data channels. All analyses were acquired using the lockspray to ensure accuracy and reproducibility; [Glu]1-fibrinopeptide B was used as the lock mass (m/z 785.8426, doubly charged) at a concentration of 2 pmol/μL and a flow rate of 0.15 μL/min. Data were collected in t h e centroid mode, the lockspray frequency was set to 5 s, and data were averaged over 10 scans. The data were analyzed with the softwares ProteinLynx Global Server and BiopharmaLynx (Waters Corp., Milford, MA). During the acquisition of LC-MSE data, multiple precursor ions are fragmented simultaneously. In order to assign the correct fragment ions to their parent ions, the properties that are used by the algorithm to parse MSE data include retention time, precursor and product ion intensities, charge states, and, crucially, the accurate masses of both the precursor and product ions from the LC-MSE data. This strategy has been shown to be effective for the identification of proteins in both simple and complex samples. Tentative peptide and protein identifications were ranked and scored by their relative correlation to a number of well-established models of known and empirically derived physicochemical attributes of proteins and peptides. The algorithm utilizes reverse or random decoy databases for automatically determining the false positive identification rate. Our data were obtained using a default acceptable false positive rate of 4%.

3.2.3 Fluorogenic tagging of intact oxidized Fc glycoforms

The ABS molecule permits an easy, efficient, and specific fluorogenic tagging of DOPA as detailed in Scheme 1.34,36-38 This chemical labeling method allows for a rapid screening of the level of oxidation of undigested proteins. After oxidation, the intact oxidized and non-oxidized Fc glycoforms were dialyzed into 100 mM sodium phosphate buffer (pH 9.0) using Amicon ultra-0.5 centrifugal filter devices (Millipore Corp. Bedford, MA, USA) for ABS fluorogenic tagging. ABS tagging of oxidized Fc glycoforms was performed by incubation of the oxidized proteins with the ABS derivatization reagents in the dark at room temperature under the following conditions: i) the molar ratio of K3Fe(CN)6:protein was 5:1, ii) the ABS concentration was 10 mM, and iii) the reaction time was 120 min. These parameters were optimized to obtain the maximal transformation yield of DOPA into benzoxazole. The formation of the resulting fluorescent benzoxazole derivatives (Scheme 1) was monitored by a steady-state fluorescence spectrometer (SpectraMax Gemini, Molecular Devices, Sunnyvale, CA). Similar derivatizing conditions were applied to the controls.

Scheme 1.

Scheme 1

Open in a new tab

Fluorogenic tagging of 3,4-dihydroxyphenylalanine (DOPA) by ABS.

4. Results

The tryptic digests of the Fc glycoforms generated 16 peptides and covered 94% of the amino acid sequence as shown in Chart 1. An example of the peptide map obtained after LC-MS analysis of one of the glycoforms (Man5-Fc) is presented in Fig. 1. The chemical stability results are presented in three sections. The first and second section focus on the forced oxidative degradation studies of the Fc glycoforms in the presence and absence of sucrose. Sucrose is used to protect immunoglobulins during long-term storage,39 and has been previously used to store related IgG-Fc glycoproteins frozen at -80°C.40 Since sucrose can degrade at elevated temperatures and acidic pH, it can be desirable to remove it from the formulations during accelerated stability studies. We have, therefore, stressed the Fc glycoforms in the presence and absence of sucrose. In these sections, we will compare for the four different glycoforms the nature of the oxidation products generated after exposure of the proteins to UV-light, metal-catalyzed oxidation, peroxyl radicals and hydrogen peroxide. In the third section, we will focus on the accelerated stability studies of the four glycoforms. We will compare the nature of the deamidation and oxidation products generated after incubation at 4°C and 40°C for 2, 4, and 12 weeks.

Figure 1.

Figure 1

Open in a new tab

LC-MS analysis of the tryptic digest of Man5-Fc (right). The fragmentation (left) of the tryptic peptide T5* indicates the presence of the glycan (Man)5-(GlcNAc)2.

4.1 Forced degradation study in the presence of sucrose

4.1.1 Formation of methionine sulfoxide

Met428, present in the tryptic peptide T15 (Chart 1), was oxidized to methionine sulfoxide (MetSO). The presence of MetSO was demonstrated by the MS/MS spectrum of the oxidized tryptic peptide T15 (Fig. S1), where the fragment ions y11, y14, and b13 confirm the oxidation of Met428 to MetSO. The transformation of Met428 to MetSO accounts for 1.4%, 1.0%, and 0.8% of the total Fc glycoforms after H2O2 exposure of HM-Fc, GlcNAc-Fc, and N297Q-Fc, respectively (Fig. 2, A. In stark contrast, the exposure of Man5-Fc to H2O2 yields less than 0.1% MetSO relative to total protein (Fig. 2, A. This result indicates that the nature of the glycan can have a significant effect on protein oxidation (see also below). We also observed significant losses of the native tryptic peptides T3, T4, and T9 after metal-catalyzed oxidation of the Fc glycoforms. These results will be discussed below. The losses of the tryptic peptides T3, T4, and T9 were taken into account for the relative quantification of product formation.

Figure 2.

Figure 2

Open in a new tab

A) Relative amount of MetSO detected for Met428. B) Relative amount of deamidated Asn315. C) Relative amount of the translocation of the side-chain of Trp313 to Lys317. The relative amounts of oxidation products were calculated as follows: 100 × [ion counts of the oxidized tryptic peptide] / [total ion counts]. The losses of tryptic peptides T3, T4, and T9 observed during metal-catalyzed oxidation were taken into account to calculate these relative amounts of oxidized product (See Table 3). The glycoforms of the Fc proteins are displayed by colors. Man5 (green), HM (blue), GlcNac (purple), and N297Q mutant (orange). See Table 1 for description of the Fc glycoforms.

4.1.2 Deamidation of Asn315

Asn315, present in the tryptic peptide T7 (Chart 1), underwent deamidation, as indicated through the presence of the y2, y3, and b14 ions in the MS/MS spectra of the deamidated tryptic peptide T7 (Fig. S2).

Contrary to the formation of MetSO at Met428, our results do not show any significant effect of the glycan nature on the deamidation of Asn315 (Fig. 2, B. Interestingly, the controls of HM-Fc and GlcNAc-Fc reveal a higher amount of deamidation as compared to the stressed samples. The latter might be rationalized in several ways such as, e.g., the formation of unknown oxidation products, which would not deamidate, or the potential presence of a carbohydrate at Asn315, which would hydrolyze faster during sample preparation than the carbohydrate at Asn297, resulting in the formation of a deamidated Asn315. A comparison of the different glycoforms, and the four different applied stresses, reveals that no more than 2% of the total Fc glycoforms underwent deamidation (Fig. 2, B.

4.1.3 Transformation of Trp313 into Gly313 and addition of 3-methylene-indolenine to Lys317

During forced degradation, the most abundant modification of the Fc-glycoforms occurred at Trp313, which was transformed into Gly313. Parallel to this transformation, 3-methylene-indolenine, obtained through fragmentation of the original Trp313 added to Lys317. Such transformations have recently been documented by us for the UV exposure of octreotide41 and IgG1.42 The structure of the product is displayed in Table 2. The new product is fully characterized by its MS/MS spectrum (Fig. S3, fragment ion b14) and the increase of mass by 129 Da of the original residue Lys317 (Fig. S3, fragment ions y4 and b16).

Table 2.

A summary of the amino acid residues of IgGl-Fc glycoforms modified during the accelerated and forced degradation studies.

Amino acid Products detected by HPLC-MS/MS analysis
Accelerated degradation Forced degradation
Asn315 Asn315 deamidation Asn315 deamidation
Met428 - Met sulfoxide
Trp277 graphic file with name nihms737274t1.jpg Not observed
Trp313 and Lys317 Not observed graphic file with name nihms737274t2.jpg
Open in a new tab

We observed that approximately 1.0-1.5% of Man5-Fc and HM-Fc underwent transformation of Trp313 to Gly313 (Fig. 2, C, green and blue bars) during UV-exposure, but detected no significant transformation of GlcNAc-Fc and N297Q-Fc relative to controls. However, we noted up to 3.5% transformation of Trp313 into Gly313 for HM-Fc and GlcNAc-Fc in the presence of H2O2 (Fig. 2, C, blue and purple bars). This was rather unexpected and suggested transition metal catalysis, as the Trp side chain fragmentation requires one-electron oxidation (see below).42 Importantly, 3% of the N297Q, but none of the other glycoforms, underwent transformation of Trp313 to Gly313 during oxidation by peroxyl radicals generated from AAPH (Fig. 2, C, orange).

4.1.4 Loss of the tryptic peptides T3, T4, and T9 after metal-catalyzed oxidation

We observed that specifically after metal-catalyzed oxidation of the Fc glycoforms, the recovery of the tryptic peptides T3, T4, and T9 varied significantly in comparison to that for the non-oxidized proteins (Table 3). Depending on the glycoform, after four hours of metal-catalyzed oxidation, the losses of T3, T4, and T9 ranged between 20%-98%. Our results indicate that for T3 and T4 the presence of HM and Man5 prevent the degradation of these peptides. In the case of the trytpic peptide T9, all the glycoforms revealed a loss of >97% (Table 3). These observations suggest potential covalent cross-linking. The nature of cross-links will be addressed in the future.

Table 3.

Losses of the tryptic peptides T3, T4 and T9 after metal-catalyzed oxidation.

In the presence of His/sucrosea In the absence of His/sucrose
Peptide T3 % loss % loss
Man5 40.6 96.8
HM 54.4 97.4
GclNac 79.5 96.0
N297Q 73.1 19.1
Peptide T4 % %
Man5 11.5 1.5
HM 23.2 59.1
GclNac 24.8 27.1
N297Q 49.2 60.1
Peptide T9 % %
Man5 97.0 53.2
HM 98.4 59.1
GclNac 98.5 65.8
N297Q 96.5 74.5
Open in a new tab
a

The His/sucrose buffer consisted of 10 mM His and 85 mg/mL sucrose, pH 5.8. The percentages were averaged from triplicates.

4.1.5 Formation of DOPA

The formation of DOPA was monitored after derivatization with ABS (Scheme 1). All controls showed presence of DOPA without exposure to any stress (Fig. 3, green bars). This result would be consistent with the presence of redox-active transition metals in the glycoform samples, inducing DOPA formation during protein preparation and purification, indicated also by the conversion of Trp313 to Gly313 through hydrogen peroxide (see above). A slight increase of the fluorescence (Fig. 3, blue bars) is observed after exposure of the glycoforms to UV light, metal-catalyzed oxidation, peroxyl radicals and H2O2. Here, the mannose-containing glycoforms Man5-Fc and HM-Fc show a higher increase of fluorescence for metal-catalyzed oxidation and the exposure to H2O2. Because of the presence of DOPA in the controls and the increase of DOPA upon the addition of H2O2, we suspected the presence of trace amounts of redox-active transition metals in the samples. Therefore, specifically HM-Fc and Man5-FC were dialyzed against a solution containing 5 mM EDTA prior to chromatographic purification. Treatment with 5 mM EDTA reduced the yields of fluorescence significantly (Fig. 4, left A and B), and the addition of H2O2 did not result in an additional increase of fluorescence (Fig. 4, left, B and C). It is also important to note that the levels of DOPA for HM-Fc and Man5-Fc controls (Fig. 4) were 3-fold lower in this batch than in the batch of Fc glycoforms used for the forced degradation study (Fig. 3), which consisted of Fc glycoforms purified without the use of EDTA. This result indicates that transition metal contamination can vary significantly in different batches.

Figure 3.

Figure 3

Open in a new tab

Fluorescence detection of DOPA observed after forced degradation reactions of Fc proteins. The glycoforms are noted: Man5 (5 mannoses), HM (high-mannoses), GlcNAc (N-acetyl-glucosamine), and N297Q (aglycosylated; Asn297 is replaced by Gln297). Fluorescence of the benzoxazole group formed after derivatization of DOPA with ABS is measured at λex= 360 nm, and λem= 490 nm.

Figure 4.

Figure 4

Open in a new tab

Fluorescence detection of DOPA observed after forced degradation reactions of HM and Man5 Fc proteins. “No chelator” indicates that the Fc proteins were purified without using EDTA. “EDTA” indicates that the Fc proteins were dialyzed against 5 mM EDTA prior to chromatographic purification. A) The controls were incubated at 4°C without H2O2 for 4 hours prior to ABS tagging. B) The Fc glycoforms were incubated at 40°C for 4 hours in the absence of H2O2. C) The Fc glycoforms were incubated at 40°C in the presence of H2O2 (1 mM). The fluorescence of the benzoxazole product formed after derivatization of DOPA with ABS is measured at λex= 360 nm and λem= 490 nm. See Table 1 for description of the Fc glycoforms.

4.2 Forced degradation in the absence of sucrose

4.2.1 Conversion of Trp277 to Gly hydroperoxide

The Fc glycoforms were prepared in citrate-phosphate buffer (20 mM, pH 6.0). In citrate-phosphate buffer, the formation of MetSO followed the same trend as observed when the Fc glycoforms were oxidized in 10 mM His/85 mg/mL sucrose, pH 5.8 (Fig. 2, A, and B). The major difference between both formulations involved Trp313 and Trp277. In the presence of His/sucrose, the oxidation of the Fc glycoforms led to the transformation of Trp313 to Gly313. In the presence of citrate-phosphate buffer, Trp313 was not oxidized. Instead, Trp277 in the tryptic peptide T4 (Chart 1) was converted to glycine hydroperoxide. However, among the four glycoforms, only Man5-Fc, HM-Fc and N297Q-Fc showed conversion of Trp277 to glycine hydroperoxide after exposure to hydrogen peroxide (Fig. 5) with yields between 1.5% and 2.5%. The other stresses, i.e. UV-exposure, peroxyl radicals and metal-catalyzed oxidation, did not reveal any oxidation at Trp277. The Gly hydroperoxide at the original Trp277 was characterized by its MS/MS spectrum (Fig. S4), especially through the presence of the b2, b3, y11, and y13 fragment ions.

Figure 5.

Figure 5

Open in a new tab

Relative amount of the oxidized tryptic peptide T4 at Trp277 from different IgG1-Fc glycoforms. Trp is oxidized to glycine hydroperoxide. See Table 1 for description of the Fc glycoforms. Glycine hydroperoxide at former Trp277 was not observed in GlcNAc-Fc glycoform.

4.2.2 Loss of the tryptic peptides T3, T4, and T9 after metal-catalyzed oxidation

In the absence of His and sucrose, we observed that after metal-catalyzed oxidation of the four glycoforms, the recovery from the peptide mapping of the tryptic peptides T3, T4 and T9 varied significantly (Table 3). A cross comparison of the losses of these tryptic peptides for the four glycoforms in the presence and in the absence of His/sucrose, revealed that i) for the tryptic peptides T3, the presence or absence of His/sucrose had minor effects on the recovery of the tryptic peptide after metal-catalyzed oxidation, ii) for the tryptic peptide T4, major losses were observed after metal-catalyzed oxidation of Man5-Fc, HM-Fc and GlcNAc-Fc in the absence of His/sucrose, iii) for the tryptic peptide T9, the absence of sucrose permitted the recovery of between 25% to 50% of the peptide when the glycoforms were exposed to metal-catalyzed oxidation in the absence of sucrose.

4.3 Accelerated stability study

The four IgG1-Fc glycoforms were prepared in citrate-phosphate buffer (20 mM, pH 6.0 with NaCl added to ionic strength of 0.15). After 2, 4, and 12 weeks of incubation of the Fc glycoforms at 4°C and 40°C, we observed deamidation of Asn315 and the conversion of Trp277 into glycine hydroperoxide.

4.3.1 Deamidation of Asn315

At 4°C, the deamidation of Asn315 slightly increased over four weeks from ca. 1.6-1.7% at t = 0 to 1.7%, 2.0%, 2.4%, and 2.2% for HM-Fc, Man5-Fc, GlcNac-Fc, and N297Q-Fc, respectively (Fig. 6, Insert). Over a 12 weeks period, additional deamidation of Asn315 was only observed for Man5-Fc and HM-Fc. At 40°C, HM-Fc and Man5-Fc displayed a steady increase of Asn315 deamidation over 12 weeks (Fig. 6), accounting for ca. 0.25% deamidation per month relative to total protein. In contrast, GlcNAc-Fc and N297Q-Fc showed a rapid increase of Asn315 deamidation over four weeks at 40°C (Fig. 6), accounting for ca. 3.6% and 3.9%, respectively, relative to total protein. Thereafter, deamidation continued over 12 weeks with a significantly slower rate.

Figure 6.

Figure 6

Open in a new tab

Deamidation of Asn315 in the IgG1-Fc glycoforms during incubation for 2, 4, and 12 weeks at 40°C at pH 6.0. Insert: Deamidation of Asn315 during incubation for 2, 4, and 12 weeks at 4°C. See Table 1 for description of the Fc glycoforms.

4.3.2 Conversion of Trp277 into Gly hydroperoxide

All four glycoforms displayed a base level of Trp277 conversion into Gly hydroperoxide between 0.85% and 1.3% relative to total protein at t = 0 and 4°C (Fig.7, Insert). Over 2 weeks at 4°C, only a slight increase of this modification could be observed for GlcNAc-Fc, while over 4 weeks a slight increase was observed for N297Q-Fc. After 4 weeks of incubation at 4°C, the yields of Trp277 conversion steadily decreased. At 40°C, maximal yields of Trp277 conversion were observed for all glycoforms at 4 weeks of incubation (3.5%, 5.5%, 5.1%, and 3.9% relative to total protein for GlcNAc-Fc, HM-Fc, Man5-Fc, and N297Q-Fc, respectively (Fig. 7)), after which the yields steadily decreased. It is likely that especially at 40°C the Gly hydroperoxide is unstable over time, and may have suffered cleavage according to the mechanism displayed in Scheme 2. In fact, such cleavage products were detected for Gly hydroperoxide formed in IgG1.42 Such cleavage leads to the release of a C-terminal amidated protein, and the formation of a glyoxal derivative.41 The latter is a reactive intermediate in the Maillard reaction.43 That is, the protein fragment with the glyoxal would possibly form crosslinks with nucleophiles.44 During digestion, the C-terminal amidated protein fragment would release a dipeptide (FN), which could not be retained on the column for MS/MS analysis.

Figure 7.

Figure 7

Open in a new tab

Transformation of Trp277 into glycine hydroperoxide in the IgG1-Fc glycoforms during incubation for 2, 4, and 12 weeks at 40°C. Insert: Transformation of Trp277 into glycine hydroperoxide during incubation for 2, 4, and 12 weeks at 4oC. See Table 1 for description of the Fc glycoforms.

Scheme 2.

Scheme 2

Open in a new tab

Proposed scheme to rationalize the degradation of glycine hydroperoxide at former Trp277.

The MS/MS data for the oxidized tryptic peptide T4, where Trp277 was transformed into Gly hydroperoxide were similar to the data recorded during the forced degradation studies (Fig. S4).

5. Discussion

The chemical degradation studies of four different well-defined IgG1-Fc glycoforms highlight significant differences for specific degradation mechanisms, affected by the nature of the glycans. During the accelerated stability study, we observed that over four weeks of incubation at 4°C, the rate of conversion of Trp277 to Gly hydroperoxide did not show any glycan-dependent trend (Fig. 7, insert). However, during four weeks of incubation at 40°C we noted similar behavior for HM-Fc and Man5-Fc, i.e. a lag time of ca. two weeks, followed by a significant increase of Gly hydroperoxide formation, corresponding to ca. 5-5.5% relative to total protein (Fig. 7). Previously, for a Fc glycoform containing mannose, i.e. (Fuc-GlcNAc-GlcNAc-Man-(Man)2-(GlcNAc)2-(Gal)2), Trp277 was identified as an oxidation site during acid-induced unfolding and aggregation of IgG1 and IgG2.45 Both GlcNAc-Fc and N297Q-Fc also showed maximal yields of Trp277 conversion to Gly hydroperoxide after four weeks incubation at 40°C, albeit with yields 1.5-times lower as compared to HM-Fc and Man5-Fc. However, while N297Q-Fc showed a lag-time comparable to HM-Fc and Man5-Fc, GlcNAc-Fc did not show a lag time. When our Fc glycoform was N-linked to a single carbohydrate (GlcNAc), or non glycosylated, the amount of Gly hydroperoxide formed after four weeks of incubation was 1.5 fold less than the amount measured for the HM-Fc and Man5-Fc (Fig. 7). At this point, we cannot rationalize in detail how the glycan structure affects Trp277 conversion to Gly hydroperoxide. However, in a first approximation, it appears that a decrease in the size of the carbohydrate, N-linked to the Asn297 site in the CH2 domain of the Fc, decreases the formation rate of Gly hydroperoxide from Trp277.

Importantly, while the conversion of Trp277 to Gly hydroperoxide is enhanced when the size of the carbohydrate chain increases, the rate of deamidation appears to slow down when the size of the carbohydrate chain increases. Indeed, the amount of deamidated Asn315 increases with a decrease of the size of the carbohydrate chain (HM < Man5 < GlcNAc < N297Q) (Fig. 6). Interestingly, we did not observe the formation of MetSO at either Met252 or Met428 during our thermal stability study. The oxidation of Met252 and Met428 is known to have an effect on the thermal stability of the CH2 domain of IgG1,46,47 and has been observed during the formulation of IgG1.48,49 The lack of MetSO formation during accelerated stability studies of the Fc glycoforms may indicate a role for the Fab domain on MetSO formation in intact antibodies.

As we observed the losses of tryptic peptides T3, T4, and T9 after four hours of metal-catalyzed oxidation, which could potentially indicate the formation of covalent crosslinks, we also observed that the total ion counts (TIC) (measured by LC-MS of the digested samples) decreased over three months of incubation at 40°C. For HM-Fc, Man5-Fc, GlcNAc-Fc, and N297Q- Fc, we observed linear decreases of the TIC at rates of 1075 ions/h, 1080 ions/h, 1585 ions/h, and 2800 ions/h respectively (Fig. 8). These observations indicate that the degradation rates of the glycoforms, represented by the losses of T3, T4 and T9, increase with a decrease of the size of the carbohydrate chain.

Figure 8.

Figure 8

Open in a new tab

Total ion counts of the tryptic digests measured after LC-MS analyses of HM (●), Man5 (●), GlcNAc (●), and N297Q (●) Fc proteins after incubation for 2, 4, and 12 weeks at 40oC. See Table 1 for description of the Fc glycoforms.

In our opinion, the loss of T3, T4 and T9 suggests the formation of covalent cross-links though we did not investigate the nature of the potential crosslinks. However, we know that the loss of these peptides was not related to a faulty digest. In fact, the LC-MS analysis covered 94% of the sequence of all Fc glycoforms (Fig. 1). Also, the tryptic peptides T3, T4, and T9 were recovered after stressing the proteins with UV-light, peroxyl radicals, and hydrogen peroxide. The percentages of loss of the tryptic peptides T3, T4 and T9 are documented in Table 3. Without knowing the exact nature of the products in which these peptides could be involved, the interpretation of Table 3 remains speculative. However, a comparison of the samples which were oxidized with Cu(II)/Asc, in the presence and absence of His/sucrose, highlights that the nature of the formulation affected the percentage of recovery of a given tryptic peptide. For example, for the tryptic peptides T3 and T9, it appears that the modification of the formulation inverted the amounts of peptides recovered for each of the glycoforms.

In the presence of His/sucrose, three different degradation pathways, i.e. MetSO formation at Met428, deamidation of Asn315, and the transformation of Trp313 into Gly313, were identified by LC-MS analyses during the forced degradation studies of the four different glycoforms. However, only the yield of MetSO and that of the transformation of Trp313 into Gly313 varied between the four different glycoforms.

The formation of MetSO at Met428 appears as a major oxidation product for HM-Fc, GlcNAc-Fc, and N297Q-Fc, when these proteins were exposed to H2O2 (Fig. 2, A. Non-negligible amounts of MetSO were also observed after exposure of N297Q-Fc to UV-light and AAPH (Fig. 2, A, orange bars).

The transformation of Trp313 into Gly313 was observed for Man5-Fc, HM-Fc, and GlcNAc-Fc during the exposure to H2O2 or UV-light (Fig. 2, C. Such transformation of Trp has been documented earlier during the exposure of IgG1 to UV light, and involves intermediary radical cation formation of Trp.41 However, the transformation of Trp313 into Gly313 through mere reaction of Fc glycoforms with H2O2 is difficult to rationalize. We, therefore, believe that traces of redox-active metals in the samples may catalyze the transformation of Trp313 into Gly313 by H2O2. Indeed, traces of metals would permit a Fenton-like reaction.50-53 Interestingly, peroxyl radicals promoted the transformation of Trp313 into Gly3131 only for N297Q-Fc (Fig. 2, C, orange bars). As a result, not only the sequence variability of the protein has to be taken into account for biosimilar comparison, but also the impurities present in the formulations are important factors to consider. As shown above, the variability of the amount of redox-active metals present in different batches affect differently the chemical degradation of the proteins. Therefore, biosimilar comparison studies have to take into consideration the variability of the impurity profiles of the different batches.54,55

In the absence of His/sucrose, the amount of MetSO at Met428 and deamidated Asn315 did not significantly change in comparison to the oxidation of the Fc proteins in the presence of sucrose. However, no oxidation of Trp313 was observed. Instead, we observed the oxidation of Trp277 to Gly hydroperoxide (Fig. 5). In absence of His/sucrose, the exposure of Man5-Fc, HM-Fc and N297Q-Fc to H2O2 converted Trp277 into Gly hydroperoxide, with comparable yields for all three glycoforms (Fig. 5). Other studies had reported the formation of kynurenin and N-formylkynurenin during UV-irradiation and metal-catalyzed oxidation of IgG1. 56-58 In our case, mass spectrometry analysis of the Fc glycoforms revealed neither the formation of kynurenin nor of N-formylkynurenin.

6. Conclusion

Our data indicate that, depending on the nature of the stress (i.e., accelerated stability vs. forced degradation studies), Trp277 and Trp313 residues in the IgG1-Fc glycoforms showed differential sensitivity towards oxidation. During the accelerated stability study, we noticed that HM-Fc and Man5-Fc showed transformation of Trp277 into Gly hydroperoxide. The losses of the tryptic peptides T3, T4, and T9 during the metal-catalyzed oxidation as well as the formation of potential cross-links generated during the accelerated thermal stability study, indicate that a systematic follow-up comparative study on the stability of glycoforms will have to focus on the formation of covalent cross-links.

Supplementary Material

1
2

Acknowledgments

This work was supported by grant NIPTE U01-KS-2014. Funding for this manuscript was also made possible, in part, by the Food and Drug Administration through grant 1U01FD005285-01, views expressed in this publication do not necessarily reflect the official policies of the Department of Health and Human Services nor does any mention of trade names, commercial practices, or organization imply endorsement by the United States Government. The authors wish to acknowledge additional financial support of T.J.T. via NIH grant NIGMS R01 GM090080 and S.Z.O. via NIH biotechnology training grant 5-T32-GM008359.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Janeway CA, Travers P, Walport M, Shlomchik M. Immunobiology. 5. Garland; 2001. [Google Scholar]
  • 2.Salfeld JG. Isotype selection in antibody engineering. Nat Biotechnol. 2007;25(12):1369–1372. doi: 10.1038/nbt1207-1369. [DOI] [PubMed] [Google Scholar]
  • 3.Nimmerjahn F, Ravetch JV. Fcgamma receptors: old friends and new family members. Immunity. 2006;24(1):19–28. doi: 10.1016/j.immuni.2005.11.010. [DOI] [PubMed] [Google Scholar]
  • 4.Mehra V, Sweetser D, Young RA. Efficient mapping of protein antigenic determinants. Proc Natl Acad Sci USA. 1986;83(18):7013–7017. doi: 10.1073/pnas.83.18.7013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Jefferis R. Recombinant antibody therapeutics: the impact of glycosylation on mechanisms of action. Trends Pharmacol Sci. 2009;30(7):356–362. doi: 10.1016/j.tips.2009.04.007. [DOI] [PubMed] [Google Scholar]
  • 6.Wright A, Tao MH, Kabat EA, Morrison SL. Antibody Variable Region Glycosylation - Position Effects on Antigen-Binding and Carbohydrate Structure. EMBO J. 1991;10(10):2717–2723. doi: 10.1002/j.1460-2075.1991.tb07819.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Wallick SC, Kabat EA, Morrison SL. Glycosylation of a VH residue of a monoclonal antibody against alpha (1----6) dextran increases its affinity for antigen. J Exp Med. 1988;168(3):1099–1109. doi: 10.1084/jem.168.3.1099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Huhn C, Selman MHJ, Ruhaak LR, Deelder AM, Wuhrer M. IgG glycosylation analysis. Proteomics. 2009;9(4):882–913. doi: 10.1002/pmic.200800715. [DOI] [PubMed] [Google Scholar]
  • 9.Takahashi M, Kuroki Y, Ohtsubo K, Taniguchi N. Core fucose and bisecting GlcNAc, the direct modifiers of the N-glycan core: their functions and target proteins. Carbohydr Res. 2009;344(12):1387–1390. doi: 10.1016/j.carres.2009.04.031. [DOI] [PubMed] [Google Scholar]
  • 10.Sondermann P, Huber R, Oosthuizen V, Jacob U. The 3.2-A crystal structure of the human IgG1 Fc fragment-Fc gammaRIII complex. Nature. 2000;406(6793):267–273. doi: 10.1038/35018508. [DOI] [PubMed] [Google Scholar]
  • 11.Krapp S, Mimura Y, Jefferis R, Huber R, Sondermann P. Structural analysis of human IgG-Fc glycoforms reveals a correlation between glycosylation and structural integrity. J Mol Biol. 2003;325(5):979–989. doi: 10.1016/s0022-2836(02)01250-0. [DOI] [PubMed] [Google Scholar]
  • 12.Ferrara C, Grau S, Jäger C, et al. Unique carbohydrate-carbohydrate interactions are required for high affinity binding between FcgammaRIII and antibodies lacking core fucose. Proc Natl Acad Sci USA. 2011;108(31):12669–12674. doi: 10.1073/pnas.1108455108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Nimmerjahn F, Anthony RM, Ravetch JV. Agalactosylated IgG antibodies depend on cellular Fc receptors for in vivo activity. 2007;104(20):8433–8437. doi: 10.1073/pnas.0702936104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kaneko Y, Nimmerjahn F, Ravetch JV. Anti-inflammatory activity of immunoglobulin G resulting from Fc sialylation. Science. 2006;313(5787):670–673. doi: 10.1126/science.1129594. [DOI] [PubMed] [Google Scholar]
  • 15.Anthony RM, Wermeling F, Karlsson MCI, Ravetch JV. Identification of a receptor required for the anti-inflammatory activity of IVIG. Proc Natl Acad Sci USA. 2008;105(50):19571–19578. doi: 10.1073/pnas.0810163105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Janeway C, Travers P, Walport M, Shlomchik M. Immunobiology: the Immune System in Health and Disease. New York: Garland Publishing Inc.; 1997. [Google Scholar]
  • 17.Chan AC, Carter PJ. Therapeutic antibodies for autoimmunity and inflammation. Nat Rev Immunol. 2010;10(5):301–316. doi: 10.1038/nri2761. [DOI] [PubMed] [Google Scholar]
  • 18.Weiner LM, Surana R, Wang S. Monoclonal antibodies: versatile platforms for cancer immunotherapy. Nat Rev Immunol. 2010;10(5):317–327. doi: 10.1038/nri2744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bhagwat A, Campi EM, Potdar MK, Jackson WR, Hearn MTW. Efficient nucleophilic substitution of halopyridines using ethanol as solvent with microwave heating: synthesis of 2-aminoethylsulfanylpyridines. Green Chem Lett Rev. 2012;5(4):595–601. [Google Scholar]
  • 20.Jefferis R. Glycosylation as a strategy to improve antibody-based therapeutics. Nat Rev Drug Discov. 2009;8(3):226–234. doi: 10.1038/nrd2804. [DOI] [PubMed] [Google Scholar]
  • 21.Arnold JN, Wormald MR, Sim RB, Rudd PM, Dwek RA. The impact of glycosylation on the biological function and structure of human immunoglobulins. Annu Rev Immunol. 2007;25:21–50. doi: 10.1146/annurev.immunol.25.022106.141702. [DOI] [PubMed] [Google Scholar]
  • 22.Siberil S, de Romeuf C, Bihoreau N, et al. Selection of a human anti-RhD monoclonal antibody for therapeutic use: Impact of IgG glycosylation on activating and inhibitory Fc gamma R functions. Clin Immunol. 2006;118(2-3):170–179. doi: 10.1016/j.clim.2005.10.008. [DOI] [PubMed] [Google Scholar]
  • 23.Yamane-Ohnuki N, Kinoshita S, Inoue-Urakubo M, et al. Establishment of FUT8 knockout Chinese hamster ovary cells: An ideal host cell line for producing completely defucosylated antibodies with enhanced antibody-dependent cellular cytotoxicity. Biotechnol Bioeng. 2004;87(5):614–622. doi: 10.1002/bit.20151. [DOI] [PubMed] [Google Scholar]
  • 24.Davies J, Jiang L, Pan LZ, LaBarre MJ, Anderson D, Reff M. Expression of GnTIII in a recombinant anti-CD20 CHO production cell line: Expression of antibodies with altered glycoforms leads to an increase in ADCC through higher affinity for FCγRIII. Biotechnol Bioeng. 2001;74(4):288–294. [PubMed] [Google Scholar]
  • 25.Wang W, Singh SK, Li N, Toler MR, King KR, Nema S. Immunogenicity of protein aggregates--concerns and realities. Int J Pharm. 2012;431(1-2):1–11. doi: 10.1016/j.ijpharm.2012.04.040. [DOI] [PubMed] [Google Scholar]
  • 26.Berkowitz SA, Engen JR, Mazzeo JR, Jones GB. Analytical tools for characterizing biopharmaceuticals and the implications for biosimilars. Nat Rev Drug Discov. 2012;11(7):527–540. doi: 10.1038/nrd3746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Buettel IC, Chamberlain P, Chowers Y, et al. Taking immunogenicity assessment of therapeutic proteins to the next level. Biologicals. 2011;39(2):100–109. doi: 10.1016/j.biologicals.2011.01.006. [DOI] [PubMed] [Google Scholar]
  • 28.Baker MP, Reynolds HM, Lumicisi B, Bryson CJ. Immunogenicity of protein therapeutics: The key causes, consequences and challenges. Self Nonself. 2010;1(4):314–322. doi: 10.4161/self.1.4.13904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Singh SK, Cousens LP, Alvarez D, Mahajan PB. Determinants of immunogenic response to protein therapeutics. Biologicals. 2012;40(5):364–368. doi: 10.1016/j.biologicals.2012.06.001. [DOI] [PubMed] [Google Scholar]
  • 30.Singh SK. Impact of Product-Related Factors on Immunogenicity of Biotherapeutics. J Pharm Sci. 2011;100(2):354–387. doi: 10.1002/jps.22276. [DOI] [PubMed] [Google Scholar]
  • 31.Okbazghi SZ, et al. Production, characterization, and biological evaluations of well-defined IgG1-Fc glycoforms as model system for biosimilarity analysis. J Pharm Sci. 2015 doi: 10.1016/j.xphs.2015.11.003. Submitted. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.More AS, et al. Correlating the impact of different well-defined oligosaccharide structures on physical stability profiles of well-defined IgG1-Fc glycoforms. J Pharm Sci. 2015 doi: 10.1016/j.xphs.2015.10.014. Submitted. [DOI] [PubMed] [Google Scholar]
  • 33.Kim JH, et al. Biosimilarity assessments of model IgG1-Fc glycoforms using a machine learning approach. J Pharm Sci. 2015 doi: 10.1016/j.xphs.2015.10.013. Submitted. [DOI] [PubMed] [Google Scholar]
  • 34.Sharov VS, Dremina ES, Pennington J, et al. Selective Fluorogenic Derivatization of 3-Nitrotyrosine and 3,4-Dihydroxyphenylalanine in Peptides: A Method Designed for Quantitative Proteomic Analysis. Meth Enzymol. 2008;441:19–32. doi: 10.1016/S0076-6879(08)01202-0. [DOI] [PubMed] [Google Scholar]
  • 35.Niki E. Free radical initiators as source of water- or lipid-soluble peroxyl radicals. Meth Enzymol. 1990;186:100–108. doi: 10.1016/0076-6879(90)86095-d. [DOI] [PubMed] [Google Scholar]
  • 36.Schöneich C, Sharov VS. Mass spectrometry of protein modifications by reactive oxygen and nitrogen species. Free Radic Biol Med. 2006;41(10):1507–1520. doi: 10.1016/j.freeradbiomed.2006.08.013. [DOI] [PubMed] [Google Scholar]
  • 37.Sharov VS, Pal R, Dremina ES, Michaelis EK, Schöneich C. Fluorogenic tagging of protein 3-nitrotyrosine with 4-(aminomethyl)benzene sulfonate in tissues: A useful alternative to Immunohistochemistry for fluorescence microscopy imaging of protein nitration. Free Radic Biol Med. 2012;53(10):1877–1885. doi: 10.1016/j.freeradbiomed.2012.08.582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Sharov VS, Dremina ES, Galeva NA, et al. Fluorogenic Tagging of Peptide and Protein 3-Nitrotyrosine with 4-(Aminomethyl)benzenesulfonic Acid for Quantitative Analysis of Protein Tyrosine Nitration. Chromatographia. 2010;71(1-2):37–53. doi: 10.1365/s10337-009-1409-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Mueller M, Loh MQT, Tee DHY, Yang Y, Jungbauer A. Liquid formulations for long-term storage of monoclonal IgGs. Appl Biochem Biotechnol. 2013;169(4):1431–1448. doi: 10.1007/s12010-012-0084-z. [DOI] [PubMed] [Google Scholar]
  • 40.Alsenaidy MA, Okbazghi SZ, Kim JH, et al. Physical Stability Comparisons of IgG1-Fc Variants: Effects of N-Glycosylation Site Occupancy and Asp/Gln Residues at Site Asn 297. J Pharm Sci. 2014;103(6):1613–1627. doi: 10.1002/jps.23975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mozziconacci O, Schöneich C. Effect of conformation on the photodegradation of Trp- and cystine-containing cyclic peptides: Octreotide and somatostatin. Mol Pharm. 2014;11(10):3537–3546. doi: 10.1021/mp5003174. [DOI] [PubMed] [Google Scholar]
  • 42.Haywood J, Mozziconacci O, Allegre KM, Kerwin BA, Schöneich C. Light-Induced Conversion of Trp to Gly and Gly Hydroperoxide in IgG1. Mol Pharm. 2013;10(3):1146–1150. doi: 10.1021/mp300680c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Glomb MA, Monnier VM. Mechanism of protein modification by glyoxal and glycolaldehyde, reactive intermediates of the Maillard reaction. J Biol Chem. 1995;270(17):10017–10026. doi: 10.1074/jbc.270.17.10017. [DOI] [PubMed] [Google Scholar]
  • 44.Abordo EA, Minhas HS, Thornalley PJ. Accumulation of α-oxoaldehydes during oxidative stress: a role in cytotoxicity. Biochem Pharmacol. 1999;58(4):641–648. doi: 10.1016/s0006-2952(99)00132-x. [DOI] [PubMed] [Google Scholar]
  • 45.Latypov RF, Hogan S, Lau H, Gadgil H, Liu D. Elucidation of Acid-induced Unfolding and Aggregation of Human Immunoglobulin IgG1 and IgG2 Fc. J Biol Chem. 2012;287(2):1381–1396. doi: 10.1074/jbc.M111.297697. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Wang W, Vlasak J, Li Y, et al. Impact of methionine oxidation in human IgG1 Fc on serum half-life of monoclonal antibodies. Mol Immunol. 2011;48(6-7):860–866. doi: 10.1016/j.molimm.2010.12.009. [DOI] [PubMed] [Google Scholar]
  • 47.Liu D, Ren D, Huang H, et al. Structure and stability changes of human IgG1 Fc as a consequence of methionine oxidation. Biochemistry. 2008;47(18):5088–5100. doi: 10.1021/bi702238b. [DOI] [PubMed] [Google Scholar]
  • 48.Gao X, Ji JA, Veeravalli K, et al. Effect of individual Fc methionine oxidation on FcRn binding: Met252 oxidation impairs FcRn binding more profoundly than Met428 oxidation. In: Carpenter J, Hershenson S, Ho R, Uchiyama S, Volkin D, Winter G, editors. J Pharm Sci. 2. Vol. 104. 2015. pp. 368–377. [DOI] [PubMed] [Google Scholar]
  • 49.Zhang A, Hu P, MacGregor P, et al. Understanding the conformational impact of chemical modifications on monoclonal antibodies with diverse sequence variation using hydrogen/deuterium exchange mass spectrometry and structural modeling. Anal Chem. 2014;86(7):3468–3475. doi: 10.1021/ac404130a. [DOI] [PubMed] [Google Scholar]
  • 50.Rác M, Křupka M, Binder S, et al. Oxidative damage of U937 human leukemic cells caused by hydroxyl radical results in singlet oxygen formation. PLoS One. 2015;10(3):e0116958. doi: 10.1371/journal.pone.0116958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Kozlowski H, Kolkowska P, Watly J, Krzywoszynska K, Potocki S. General aspects of metal toxicity. Curr Med Chem. 2014;21(33):3721–3740. doi: 10.2174/0929867321666140716093838. [DOI] [PubMed] [Google Scholar]
  • 52.Hlavaty JJ, Nowak T. Affinity cleavage at the metal-binding site of phosphoenolpyruvate carboxykinase. Biochemistry. 1997;36(49):15514–15525. doi: 10.1021/bi970574p. [DOI] [PubMed] [Google Scholar]
  • 53.Kerwin BA. Polysorbates 20 and 80 used in the formulation of protein biotherapeutics: structure and degradation pathways. J Pharm Sci. 2008;97(8):2924–2935. doi: 10.1002/jps.21190. [DOI] [PubMed] [Google Scholar]
  • 54.Wang W, Ignatius AA, Thakkar SV. Impact of residual impurities and contaminants on protein stability. J Pharm Sci. 2014;103(5):1315–1330. doi: 10.1002/jps.23931. [DOI] [PubMed] [Google Scholar]
  • 55.Federici M, Lubiniecki A, Manikwar P, Volkin DB. Analytical lessons learned from selected therapeutic protein drug comparability studies. Biologicals. 2013;41(3):131–147. doi: 10.1016/j.biologicals.2012.10.001. [DOI] [PubMed] [Google Scholar]
  • 56.Pattison DI, Rahmanto AS, Davies MJ. Photo-oxidation of proteins. Photochem Photobiol Sci. 2011;11(1):38–53. doi: 10.1039/c1pp05164d. [DOI] [PubMed] [Google Scholar]
  • 57.Vanhooren A, Devreese B, Vanhee K, Van Beeumen J, Hanssens I. Photoexcitation of tryptophan groups induces reduction of two disulfide bonds in goat alpha-lactalbumin. Biochemistry. 2002;41(36):11035–11043. doi: 10.1021/bi0258851. [DOI] [PubMed] [Google Scholar]
  • 58.Griffiths HR, Lunec J, Blake DR. Oxygen radical induced fluorescence in proteins; identification of the fluorescent tryptophan metabolite, N-formyl kynurenine, as a biological index of radical damage. Amino Acids. 1992;3(2):183–194. doi: 10.1007/BF00806783. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS737274-supplement-1.docx (1.5MB, docx)
2
NIHMS737274-supplement-2.dtd (41.1KB, dtd)

RESOURCES