Impact of IgG2 high molecular weight species on neonatal Fc receptor binding assays

Analytical Biochemistry 489 (2015) 25e31

Contents lists available at ScienceDirect

Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio

Impact of IgG2 high molecular weight species on neonatal Fc receptor binding assays Yuling Zhang a, *, Abhishek Mathur b, Gwen Maher c, Thomas Arroll d, Robert Bailey a a

Analytical Sciences, Amgen, Seattle, WA 98119, USA Regeneron Pharmaceuticals, Rensselaer, NY 12144, USA c Functional Biocharacterization, Amgen, Thousand Oaks, CA 91320, USA d Seattle Genetics, Bothell, WA 98021, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 December 2014 Received in revised form 28 July 2015 Accepted 29 July 2015 Available online 7 August 2015

A cell-based assay and a solution neonatal Fc receptor (FcRn) binding assay were implemented for the characterization of an IgG2 antibody after observation that different product lots exhibited unexpected differences in FcRn binding in the cell-based format with membrane-bound FcRn. The experiments described here suggest that the apparent differences observed in the FcRn binding across different product lots in the cell-based format can be attributed to the different levels of the higher order high molecular weight species (HMWs) in them. A strong correlation between FcRn binding in the cell-based format and the percentage (%) higher order HMWs suggests that small amounts (~0.1%) of the latter could cause the enhanced apparent FcRn binding (% relative binding ranging from 50 to 100%) in the format. However, when the binding was assessed with recombinant FcRn in soluble form, avidity effects were minimal and the assay format exhibited less sensitivity toward the differences in higher order HMWs levels across product lots. In conclusion, a solution-based assay may be a more appropriate assay to assess FcRn binding of the dominant species of an Fc-fusion protein or monoclonal antibody if minor differences in product variants such as higher order HMWs are shown to affect the binding significantly. © 2015 Elsevier Inc. All rights reserved.

Keywords: HMWs Aggregates SEC Antibody FcRn binding

Fc-fusion proteins and monoclonal antibodies (mAbs) have been a major focus for biopharmaceutical industries and academics for vital drug development for serious diseases such as rheumatoid and psoriatic arthritis and different types of cancers [1e7]. All such proteins contain an Fc domain that is known to bind to the neonatal Fc receptor (FcRn) [8,9]. The binding of the Fc domain of IgG-based molecules to the FcRn has been reported to affect their in vivo pharmacokinetic profile by maintaining the protein in circulation for longer periods, thereby prolonging their serum half-life and potentially influencing their pharmacokinetics (PK) [10e19]. Structurally, FcRn is a heterodimer composed of a transmembrane a-chain homologous to major histocompatibility

Abbreviations: mAb, monoclonal antibody; FcRn, neonatal Fc receptor; PK, pharmacokinetics; AlphaScreen, amplified luminescent proximity homogeneous assay screen; SEC, size exclusion chromatography; HMWs, high molecular weight species; CV, coefficient of variation; rCEeSDS, reduced capillary electrophoresis with sodium dodecyl sulfate; nrCEeSDS, non-reduced CE with SDS; SVeAUC, sedimentation velocity analytical ultracentrifugation. * Corresponding author. E-mail address: [email protected] (Y. Zhang). http://dx.doi.org/10.1016/j.ab.2015.07.017 0003-2697/© 2015 Elsevier Inc. All rights reserved.

complex (MHC) class-I-like molecules and a soluble light chain, b2 microglobulin. FcRn is expressed on a variety of tissues and cell types, including vascular endothelial cells [11]. FcRn binds to the interface between CH2 and CH3 domains of IgG Fc heavy chains under mildly acidic conditions in the endosome (~pH 6.0e6.5) and releases it at neutral to mildly basic pH (~7.0e7.5) at the cell surfaceeplasma interface. By this highly pH-dependent interaction, FcRn mediates IgG homeostasis in human adults by maintaining serum IgG levels and is also known to transfer maternal gamma globulins from mother to fetus antenatally via the neonatal intestine [20,21]. The FcRn binding assay has become a regulatory agency expectation and current industry standard for analytical characterization during process development of Fc-fusion proteins and mAbs [8,22,23]. FcRn binding assays can be performed using different formats that may use FcRn in immobilized form (e.g., membrane-bound FcRn in a cell-based assay) [24e27] or soluble form (e.g., Biacore) [28,29]. A cell-based assay, where the Fc region of IgG binds to the cell surface-expressed FcRn, is viewed as more appropriate for measuring FcRn binding because the cell-bound form closely represents the way in which FcRn is presented

26

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

physiologically. A few other formats using engineered cell lines expressing human FcRn on the cell surface have also been reported in the literature [27]. However, cell-based formats are prone to avidity effects due to the presence of surfaceimmobilized receptors [24e27]. “Avidity” is defined as apparent affinity (strength of interaction) resulting from multiple points of interactions such as in the case of oligomers as opposed to a monomer. It is usually higher than “affinity”, which represents the strength of a single point of interaction (such as with monomer). As an alternative, non-cell-based assays using surface plasmon resonance (SPR) [29,30] and AlphaScreen (amplified luminescent proximity homogeneous assay screen) technologies [30] have routinely been used for studying the binding of the Fc domain of IgG molecules to the soluble form of recombinant FcRn. However, these methods evaluate the nonphysiological presentation of FcRn in soluble form. Size exclusion chromatography (SEC) separates large molecules such as polymers and proteins, including mAbs, into a few peaks based on their hydrodynamic volumes [31e35]. The hydrodynamic volumes are determined by the physical sizes and the conformational shapes of these molecules that define the elution orders. Usually, the high molecular weight species (HMWs) elute first, the monomer second, and the clips last. The HMWs contain highly aggregated materials [34e36], trimers [37,38], and dimers [39,40] (listed in the order of elution). The amounts of these aggregates are as low as 1e3% but may have significant impact on biologics in terms of immunogenicity in previous reported literature [22,41e44]. Moreover, the HMWs may also cause hypersensitivity in potency assays [42,45] and FcRn binding assays [24,27] due to avidity effects. In this article, we report the impact of small amounts of highly aggregated material (higher order HMWs) on an in vitro cellbased FcRn binding assay and compare it with a solution-based FcRn binding assay. As a result of the avidity effect, the small amount of higher order HMWs results in hyperbinding to the FcRn for an IgG2 mAb in the cell-based assay. An in-depth understanding of the hypersensitivity of higher order HMWs of the IgG2 mAb and a comprehensive comparison between membranebound and solution-based FcRn binding assays are described in this article. Materials and methods Materials A TOSOH TSKgel G3000SWXL column (7.8
300 mm, 5 mm, cat. no. 08541) and a TOSOH TSKgel G3000SWXL guard column (6.0
40 mm, 7 mm, cat. no. 08543) were purchased from TosoHass. Sodium phosphate, monobasic and monohydrate (NaH2PO4$H2O, FW ¼ 137.99, cat. no. 3818-05), sodium chloride (NaCl, FW ¼ 58.44, cat. no. 3624-01), ethanol (cat. no. L216-07), and sodium hydroxide, 50% (NaOH, cat. no. 3727-01), were purchased from J. T. Baker. Materials related to the cell-based FcRn binding assay are described in previously published literature [25]. Analytical separation by SEC The HMW component was separated from the main component (monomer) of the IgG2 mAb by SEC using two TOSOH TSK G3000SWXL columns (7.8
300 mm) connected in series, preceded by a TSK SWXL guard column (6.0
40 mm). A 10-mg aliquot of the IgG2 mAb solution with an injection volume of 10 ml was loaded and separated in an SEC column. The temperature for the column during the separation was room temperature (~23  C). The IgG2 mAb was isocratically eluted with a pH 7.3 mobile phase

containing 50 mM sodium phosphate, 100 mM sodium chloride, and 10% ethanol. The separation used a flow rate of 0.5 ml/min, and the total separation time was 60 min, including the cleanup and reequilibration of the column. Eluted peaks were detected by fluorescence, with excitation at 280 nm and emission at 340 nm, and integrated using HPLC Empower software (Waters, Milford, MA, USA). Semi-preparative fractionation by SEC The main peak monomer component was fractionated by loading 1.0 mg of material instead of the analytical scale of 10 mg. The eluted peaks were detected by UV (ultraviolet) 280 nm. The collected main peak was concentrated and buffer exchanged into pH 5.0 formulation buffer and stored at less than 20  C. Cell-based FcRn binding assay The in vitro cell-based FcRn binding assay was developed in a competition binding format to test the binding of the Fc moiety of mAb to FcRn. The assay used a variant of the HEK (human embryonic kidney) cell line, 293T (293 cells expressing SV40 large T antigen), which expressed FcRn on the cell surface. The mAb test sample and the mAb reference standard were incubated with FcRnexpressing cells. The cell-expressed IgG-Fc was labeled with Alexa 488 dye. The cell-based FcRn binding assay was performed at room temperature at pH 6.0. After the incubation, the assay plate was read on a flow cytometer for cell-bound fluorescence. Fluorescence data from each well were recorded and analyzed. A representative dose-response curve of the cell bioassay was reported in Fig. 5 of Ref. [25]. After assessing similarity between response curves of test sample and reference standard, the test sample binding relative to the reference standard was determined and the results were reported as percentage (%) relative binding. Details related to the cellbased FcRn binding assay were reported previously [25]. FcRn AlphaScreen solution-based binding assay The AlphaScreen binding assay is a bead-based amplified luminescent proximity homogeneous assay that detects bimolecular interactions. The in vitro binding assay contains two bead types: an acceptor bead and a donor bead. The acceptor beads are coated with a hydrogel that contains thioxene derivatives as well as nickel chelate, which binds to the histidine domain of histidinelabeled FcRn (FcRneHis). The donor beads are coated with a hydrogel that contains phthalocyanine, a photosensitizer, and streptavidin, which binds to biotinylated CHO (Chinese hamster ovary)-derived human Fc. When FcRneHis and the biotinylated human Fc bind together, they also bring the acceptor and donor beads into close proximity. When laser light is applied to this complex, ambient oxygen is converted to singlet oxygen by the excitation of the donor bead. If the beads are in close proximity, an energy transfer to the acceptor bead occurs, resulting in light production (luminescence), which is measured by an Envision plate reader (PerkinElmer, Waltham, MA, USA) equipped for AlphaScreen signal detection. When mAb is present at sufficient concentrations to inhibit the binding of FcRneHis, at a fixed concentration, to the biotinylated human Fc, a dose-dependent decrease in emission at 520e620 nm is observed. After assessing and passing parallelism using a four-parameter curve fit, mAb binding relative to a reference standard was determined and reported as % relative binding. During method qualification, the accuracy (% recovery) and precision (% coefficient of variation, CV) of AlphaScreen binding assay were observed to be 98 and 8%, respectively. The assay has

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

Results Cell-based FcRn binding assessment for multiple product lots of an IgG2 mAb The cell-based FcRn binding assay was employed to analyze multiple lots of the IgG2 mAb. The % relative FcRn binding results are shown in Table 1. There were 8 lots that exhibited apparent FcRn binding in the range of 43e51%, whereas the other 5 lots had binding in the range of 74e100%, with respect to the IgG2 mAb reference standard. The results for these samples were repeatable with acceptable precision of CV < 16%, demonstrating that the assay itself performed consistently [25]. FcRn binding of purified main peak from SEC A main peak fraction from the IgG2 mAb reference standard was purified and concentrated to 10 mg/ml and analyzed by SEC to ensure the purity of the main peak fraction (Fig. 1). The higher order HMWs are the highly aggregated materials eluting before the dimer and 1 and ½ molecule as a leading shoulder shown in Fig. 1. The total HMWs include not only the higher order HMWs but also the dimer and 1 and ½ molecule. The main peak monomer sample and its starting material were then analyzed by the cell-based FcRn binding assay. The data are shown in Table 2. Removal of higher order HMWs reduced the FcRn down to 22% relative binding compared with the reference standard, whereas the starting material remained high at 79e81% relative binding.

based FcRn binding. An overlay of SEC chromatograms of lot 3 with 2.14% HMWs and lot 12 with 0.93% HMWs is shown in Fig. 3. The results show subtle variations and differences in the profiles, notably a higher order HMWs present in lot 3 that is absent in lot 12. This same higher order HMWs is observed in lots 1 to 5, corresponding to higher relative FcRn binding, but not in lots 6 to 13, where the relative FcRn binding is lower. Fig. 4 shows 1 of the product lots with 0.1% higher order HMWs and 3 product lots without the leading shoulder. The first lot exhibited FcRn binding of 100%, whereas the other 3 lots showed 45% FcRn binding. To

1600

1200 1000 800 600 400 200 0 0.0

10.0

20.0

30.0

40.0

50.0

60.0

Time (min) 24.0

B

22.0 20.0

Fluorescence

Correlation of cell-based FcRn binding versus total HMWs Based on the FcRn binding results of the purified main peak (Table 2), the elevated FcRn binding in samples 1e5 was believed to be due to the presence of total HMWs. Thus, to evaluate a correlation between FcRn binding and the amount of total HMWs present in each sample, a linear regression analysis was performed by plotting the % relative FcRn binding and the total % HMWs as shown in Fig. 2. An observed coefficient of determination of 0.63 suggested a poor correlation between the total % HMWs and the % relative FcRn binding.

A

1400

Fluorescence

appropriate system suitability criteria in place to yield acceptable accuracy and precision around relative binding values.

27

16.0 12.0 8.0 4.0 0.0 25.0

30.0

35.0

40.0

45.0

50.0

Time (min)

Correlation of cell-based FcRn binding versus higher order HMWs Further investigation was conducted to determine whether different HMWs were responsible for differences in relative cell-

Fig.1. Overlay of SEC chromatograms of IgG2 mAb reference standard (black trace) and the purified main peak (red trace): (A) full scale; (B) zoomed scale. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Table 1 Cell-based membrane-bound % relative FcRn binding versus various product lots of IgG2 mAb. Various lots of IgG2

% Total HMWs from SEC

Average % relative FcRn binding

Standard deviation of FcRn binding assay

CV of FcRn binding assay (%)

1 2 3 4 5 6 7 8 9 10 11 12 13

1.75 1.33 2.14 1.33 1.54 1.01 1.03 1.00 0.96 0.89 1.09 0.93 0.94

100 81 74 74 92 50 51 43 46 48 48 46 46

0.00 0.04 0.03 0.03 0.07 0.04 0.04 0.06 0.03 0.03 0.05 0.05 0.08

0 4 5 4 8 8 8 13 7 6 11 12 16

28

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

Table 2 Cell-based FcRn binding of purified main peak fraction of IgG2 mAb reference standard compared with original reference standard. Sample names

% Relative cell-based FcRn binding to IgG2 mAb reference standard

IgG2 mAb reference standard (lot 1) IgG2 mAb containing HMWs (lot 2) Purified main peak fraction from reference standard (lot 1)

Replicate 1

Replicate 2

100 81 13

100 79 30

evaluate whether the higher order HMWs observed in lots 1 to 5 correlate with FcRn binding, the cell-based FcRn % relative binding results were plotted against the % higher order HMWs, as shown in Fig. 5. The R2 was observed to be 0.93, indicating a strong correlation between % higher order HMWs and FcRn % relative binding. Solution-based FcRn binding comparison with cell-based FcRn binding assay To evaluate whether the binding assay format is affected by the presence of the higher order HMWs, a solution-based FcRn binding assay was developed using the AlphaScreen technology and the performance of this FcRn binding assay format was compared with that of the cell-based binding assay (Table 3). The results of the comparison between the two assay formats showed that the solution-based FcRn binding is less sensitive to the presence of HMWs as compared with membrane-bound FcRn binding assay. The results of the solution-based FcRn binding assay showed similar % relative FcRn binding for all product lots (within the variability of the assay), in contrast to the membrane-bound cellbased assay, which showed 2-fold higher binding for some of the lots. Discussion Extensive investigations related to the product quality and the FcRn assay formats were conducted to understand the differences in FcRn binding observed for different product lots. First, the product quality of these lots was evaluated and numerous analytical techniques were employed, including potency assays, pH, appearance, color, osmolality, polysorbate 20, cation exchange chromatography, reduced and non-reduced capillary electrophoresis with sodium dodecyl sulfate (rCEeSDS and nrCEeSDS, respectively), SEC, and sedimentation velocity analytical ultracentrifugation (SVeAUC). The N-linked glycosylation, charge variants,

Average % relative cell-based FcRn binding

e 80 22

clips, and deamidation profiles of these lots did not show any differences that could correlate with the differences observed in FcRn binding. HMWs (aggregates) measured in rCEeSDS and nrCEeSDS, equivalent to SDSePAGE (polyacrylamide gel electrophoresis) or SVeAUC, did not suggest a significant amount of aggregates (HMWs) (data not shown). Only the SEC results showed the possibility of a potential correlation between the HMWs and the FcRn binding because the samples with higher FcRn binding contained 0.6e1.5% more HMWs than the lots with lower FcRn binding (Table 1). To evaluate the hypothesis that the presence of HMWs led to an apparent increase in % relative FcRn binding, fractionation was performed by SEC to remove the HMWs from the sample and cellbased FcRn binding of the newly purified samples was performed. Removal of higher order HMWs reduced the FcRn down to 13e30% relative binding, supporting the hypothesis that HMWs were the primary cause of the hypersensitivity in cell-based FcRn binding assay (Table 2). However, when the total HMWs was plotted against the cell-based FcRn relative binding, the correlation was poor, suggesting that total HMWs alone was not solely responsible for the hypersensitivity observed. Because the total HMWs did not show a good correlation to FcRn binding differences, further investigation of the HMWs by SEC was conducted to determine whether specific HMWs might be responsible for the differences in relative FcRn binding. SEC chromatograms in Fig. 3 show subtle variations and differences in the profiles; however, they seem to be insignificant. The % dimers and partially aggregated materials were plotted against % relative FcRn binding for different lots. No correlation between the amount of the dimer or partially aggregated materials and the % relative FcRn binding was observed (data not shown), which suggested that the dimer or the partially aggregated materials were not the root cause

110.0 y = 39.543x + 14.075 R² = 0.63

% Relative FcRn binding

100.0 90.0 80.0 70.0 60.0 50.0 40.0 0.5

1.0

1.5

2.0

2.5

% Total HMWs

Fig.2. Correlation of % FcRn binding from cell-based assay versus % HMWs observed from multiple product lots of IgG2 mAb.

Fig.3. Overlay SEC chromatograms of IgG2 mAb lot 12 with low levels of HMWs (blue trace) and lot 3 with high levels of HMWs (black trace). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

29

Fig.4. SEC chromatograms of 4 IgG2 mAb lots: (A) lot 1; (B) lot 11; (C) lot 12; (D) lot 13.

% Relatvie FcRn binding

of the observed nearly 2-fold increase in FcRn binding of product lots 1 to 5 in Table 1. On closer investigation of the SEC chromatograms, a subtle variation was observed in peaks between 23 and 29 min, as shown in Figs. 3 and 4. A leading shoulder was apparent in the lots with higher FcRn binding, whereas it was absent in the other lots with approximately 50% FcRn binding. Further evaluation of SEC peak percentages of the leading shoulder that correspond to higher

y = 429.16x + 45.436 R² = 0.93

110 100 90 80 70 60 50 40 30 0.00

0.02

0.04

0.06

0.08

0.10

0.12

Higher order HMWs Fig.5. Correlation of % FcRn binding from cell-based assay versus % higher order HMWs observed from multiple product lots of IgG2 mAb.

order HMWs showed that product lots 1 to 5 contained approximately 0.05e0.1% of higher order HMWs, whereas product lots 6 to 13 did not have them. These results suggest that the higher order HMWs are responsible for the differences in relative FcRn binding. To further confirm this hypothesis, the cell-based FcRn % relative binding results were plotted against the % higher order HMWs, as shown in Fig. 5. The R2 was observed to be 0.93, indicating a strong correlation between % higher order HMWs and FcRn % relative binding. The observations that relative FcRn binding was higher in lots containing higher order HMWs, that the correlation between relative FcRn binding was good, and that subsequent removal of the higher order species through purification eliminates FcRn binding differences all strongly suggest that the higher order HMWs are responsible for FcRn binding differences in the cell-based assay. A mechanistic explanation for this observation was that the apparent affinity of higher order HMWs is expected to be greater than the monomeric species due to avidity effects [28,29]. Avidity describes the combined strength of multiple bonds, as compared with affinity, which describes the strength of one bond. For the IgG2 mAb, the higher order HMWs are likely forming multiple bonds with the Fc receptor, making them more difficult to dislodge than the monomer (main peak). The experiments described here suggested that the apparent differences in FcRn % relative binding could be attributed to the presence of higher order HMWs. The reason for the apparent increased % relative binding of the higher order HMWs was likely due to higher binding avidity compared

30

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

Table 3 Comparison of cell-based and solution-based AlphaScreen FcRn binding assays of four different product lots of IgG2 mAb. Lot ID

1 7 8 9

Cell-based FcRn binding assay

Solution-based AlphaScreen FcRn binding assay

Average % relative FcRn binding

CV (%)

Number of assay replicates

Average % relative FcRn binding

CV (%)

Number of assay replicates

100 51 43 46

8 8 13 7

10 3 3 3

93 87 85 92

11 14 12 12

10 3 3 3

with the monomeric and other HMWs. A strong correlation of the cell-based FcRn relative binding to the % higher order HMWs suggested that small amounts (~0.1%) of higher order HMWs could cause the increase observed in apparent FcRn % relative binding (from ~50 to 100%). A solution-based FcRn binding assay was developed using the AlphaScreen technology to determine whether a different assay format would be less affected by avidity effects. The results in Table 3 show that solution-based FcRn binding is less sensitive to the presence of HMWs as compared with the membrane-bound cell-based method that uses 293T cells that express FcRn on the cell surface. With membrane-bound FcRn, avidity effects come into play and result in hyper-binding to higher order HMWs. As a result, different % relative binding was observed for samples that have higher order HMWs levels different from the reference standard. It was confirmed that the sensitivity of the cell-based assay toward higher order HMWs could not be modulated much by altering cell densities or critical reagent (labeled competitor) concentrations used in the respective systems (data not shown). Thus, the differences in the binding values across two FcRn binding methods (~50% vs. ~100%) could be attributed to different levels of sensitivity toward higher order HMWs due to differences in the way FcRn is presented in each format. When the binding was assessed with FcRn in soluble form, avidity effects were minimal and the assay format became less sensitive to the differences in higher order HMWs levels. The AlphaScreen format used a soluble form of FcRn. In this format, the FcRn binding occurred in the solution phase during the first incubation. AlphaScreen beads were added only after the solution-phase FcRn binding reaction with the IgG2 mAb neared completion. Reaction with beads during the second incubation was minimal to prevent the possible rearrangements that might introduce the avidity effects, resulting in differential binding as observed with methods using immobilized FcRn. Because higher order HMWs resulted in the avidity effects in the cell-based assay, the assay format was unable to properly distinguish affinity of the predominant monomeric IgG2 mAb across different product lots. Due to these avidity effects and the inability to measure the predominant IgG2 mAb form, the membrane-bound cell-based assay was deemed to be inappropriate for use to measure the FcRn binding affinity of the IgG2 mAb. To ascertain that avidity effects were not an artifact of the cell-based system only and were actually a result of FcRn presentation in the immobilized form, another format using immobilized FcRn on AlphaScreen beads was developed and different product lots were compared in it. It also exhibited sensitivity to the higher order HMWs due to avidity effects similar to the cell-based assay. We demonstrated a strong correlation between the FcRn binding with the assay using immobilized receptor and the levels of higher order HMWs of an IgG2 antibody, presumably due to avidity effects. Similar avidity effects were also observed in the AlphaScreen assay using beads pre-immobilized with FcRn. Thus, even though the presentation of FcRn is closer to the physiological presentation in the assay formats using immobilized FcRn, they

appear to be influenced by avidity of minor species such as higher order HMWs. In contrast, when the binding was assessed with FcRn in soluble form, avidity effects were minimal and the assay format became less sensitive to the differences in higher order HMWs levels. Based on the results discussed here, we conclude that a solution-based FcRn binding assay would be more appropriate to assess FcRn binding of the dominant species of an Fc-fusion protein or monoclonal antibody, when the conventional formats with membrane-bound (immobilized) FcRn are shown to exhibit undesirable sensitivity to the minor differences in higher order HMWs levels. Acknowledgments We thank Randal Bass, Richard Rogers, Yilong Zhang, and Dick Ill for critical feedback and constant support. We also acknowledge Joanne Ho for providing some of the FcRn binding data during the investigation. References [1] E. Koren, L.A. Zuckerman, A.R. Mire-Sluis, Immune responses to therapeutic proteins in humans: clinical significance, assessment, and prediction, Curr. Pharm. Biotechnol. 3 (2002) 349e360. [2] E.M. Lewiecki, Monoclonal antibodies for the treatment of osteoporosis, Expert Opin. Biol. Ther. 13 (2013) 183e196. [3] N.A.P.S. Buss, S.J. Henderson, M. McFarlane, J.M. Shenton, L.D. Haan, Monoclonal antibody therapeutics: history and future, Curr. Opin. Pharmacol. 12 (2012) 615e622. [4] W.R. Strohl, Optimization of Fc-mediated effector functions of monoclonal antibodies, Curr. Opin. Biotechnol. 20 (2009) 685e691. [5] X. Jiang, A. Song, S. Bergelson, T. Arroll, B. Parekh, K. May, S. Chung, R. Strouse, A. Mire-Sluis, M. Schenerman, Advances in the assessment and control of the effector functions of therapeutic antibodies, Nat. Rev. Drug Dis. 10 (2011) 101e111. [6] D.M. Czajkowsky, J. Hu, Z. Shao, R.J. Pleass, Fc-fusion proteins: new developments and future perspectives, EMBO Mol. Med. 4 (2012) 1015e1028. [7] S.J. Kim, Y. Park, H. Hong, Antibody engineering for the development of therapeutic antibodies, J. Mol. Cells 20 (2005) 17e29. [8] T.T. Kuo, K. Baker, M. Yoshida, S.W. Qiao, V.G. Aveson, W.I. Lencer, R.S. Blumberg, Neonatal Fc receptor: from immunity to therapeutics, J. Clin. Immunol. 30 (2010) 777e789. rani, [9] A. Beck, E. Wagner-Rousset, D. Ayoub, A.V. Dorsselaer, S. Sanglier-Cianfe Characterization of therapeutic antibodies and related products, Anal. Chem. 85 (2013) 715e736. [10] M. Raghavan, P.J. Bjorkman, Fc receptors and their interactions with immunoglobulins, Annu. Rev. Cell Dev. Biol. 12 (1996) 181e220. [11] D.C. Roopenian, S. Akilesh, FcRn: the neonatal Fc receptor comes of age, Nat. Rev. Immunol. 7 (2007) 715e725. [12] L.G. Presta, Molecular engineering and design of therapeutic antibodies, Curr. Opin. Immunol. 20 (2008) 460e470. [13] W.F. Dall'Acqua, P.A. Kiener, H. Wu, Properties of human IgG1s engineered for enhanced binding to the neonatal Fc receptor (FcRn), J. Biol. Chem. 281 (2006) 23514e23524. [14] J. Zalevsky, A.K. Chamberlain, H.M. Horton, S. Karki, I.W. Leung, T.J. Sproule, G.A. Lazar, D.C. Roopenian, J.R. Desjarlais, Enhanced antibody half-life improves in vivo activity, Nat. Biotechnol. 28 (2010) 157e159. [15] T. Suzuki, A. Ishii-Watabe, M. Tada, T. Kobayashi, T. Kanayasu-Toyoda, T. Kawanishi, T. Yamaguchi, Importance of neonatal FcR in regulating the serum half-life of therapeutic proteins containing the Fc domain of human IgG1: a comparative study of the affinity of monoclonal antibodies and Fcfusion proteins to human neonatal FcR, J. Immunol. 184 (2010) 1968e1976. [16] P.R. Hinton, M.G. Johlfs, J.M. Xiong, K. Hanestad, K.C. Ong, C. Bullock, S. Keller, squez, N. Tsurushita, Engineered human IgG M.T. Tang, J.Y. Tso, M. Va

Y. Zhang et al. / Analytical Biochemistry 489 (2015) 25e31

[17]

[18]

[19]

[20]

[21]

[22] [23] [24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

antibodies with longer serum half-lives in primates, J. Biol. Chem. 279 (2004) 6213e6216. P.R. Hinton, J.M. Xiong, M.G. Johlfs, M.T. Tang, S. Keller, N. Tsurushita, An engineered human IgG1 antibody with longer serum half-life, J. Immunol. 176 (2006) 346e356. W. Wang, J. Vlasak, Y. Li, P. Pristatsky, Y. Fang, T. Pittman, J. Roman, Y. Wang, T. Prueksaritanont, R. Ionescu, Impact of methionine oxidation in human IgG1 Fc on serum half-life of monoclonal antibodies, Mol. Immunol. 48 (2011) 860e866. P. Wang, Y. Lu, L. Fang, T. Hamuro, B. Pittman, J. Carr, T. Hochman, Monoclonal antibodies with identical Fc sequences can bind to FcRn differentially with pharmacokinetic consequences, Drug Metab. Dispos. 39 (2011) 1469e1477. J.L. Leach, D.D. Sedmak, J.M. Osborne, B. Rahill, M.D. Lairmore, C.L. Anderson, Isolation from human placenta of the IgG transporter, FcRn, and localization to the syncytiotrophoblast: implications for maternalefetal antibody transport, J. Immunol. 157 (1996) 3317e3322. N.E. Simister, C.M. Story, H.L. Chen, J.S. Hunt, An IgG transporting Fc receptor expressed in the syncytiotrophoblast of human placenta, Eur. J. Immunol. 26 (1996) 1527e1531. A.S. Rosenberg, Effects of protein aggregates: an immunologic perspective, AAPS J. 8 (2006) 501e507. A.C. Chan, P.J. Carter, Therapeutic antibodies for autoimmunity and inflammation, Nat. Rev. Immunol. 10 (2010) 301e316. Y. Lu, J.M. Vernes, N. Chiang, Q. Ou, J. Ding, C. Adams, K. Hong, B.T. Truong, D. Ng, A. Shen, G. Nakamura, Q. Gong, L.G. Presta, M. Beresini, B. Kelley, H. Lowman, W.L. Wong, Y.G. Meng, Identification of IgG1 variants with increased affinity to FcgRIIIa and unaltered affinity to FcgRI and FcRn: comparison of soluble receptor-based and cell-based binding assays, J. Immunol. Methods 365 (2011) 132e141. A. Mathur, T. Arora, L. Liu, J. Crouse-Zeineddini, V. Mukku, Qualification of a homogeneous cell-based neonatal Fc receptor (FcRn) binding assay and its application to studies on Fc functionality of IgG-based therapeutics, J. Immunol. Methods 390 (2013) 81e91. R.J. Ober, C.G. Radu, V. Ghetie, E.S. Ward, Differences in promiscuity for antibodyeFcRn interactions across species: implications for therapeutic antibodies, Int. Immunol. 13 (2001) 1551e1559. A. Ratcliffe, D.R. Stanworth, The localization of the binding site(s) on human IgG1 for the Fc receptors on homologous monocytes and heterologous mouse macrophages, Immunology 50 (1983) 93e100. H. Pan, K. Chen, L. Chu, F. Kinderman, I. Apostol, G. Huang, Methionine oxidation in human IgG2 Fc decreases binding affinities to protein A and FcRn, Protein Sci. 18 (2009) 424e433. A. Bertolotti-Ciarlet, W. Wang, R. Lownes, P. Pristatsky, Y. Fang, T. McKelvey, Y. Li, J. Drummond, T. Prueksaritanont, J. Vlasak, Impact of methionine oxidation on the binding of human IgG1 to FcRn and Fc gamma receptors, Mol. Immunol. 46 (2009) 1878e1882. A. K. Chamberlain, B. I. Dahiyat, J. R. Desjarlais, S. B. Karki, Fc variants with altered binding to FcRn, U.S. patent application 20090041770 (2007). H.G. Barth, B.E. Boyes, C. Jackson, Size exclusion chromatography and related separation techniques, Anal. Chem. 70 (1998) 251Re278R.

31

[32] F. He, D.H. Phan, S. Hogan, R. Bailey, G.W. Becker, L.O. Narhi, V.I. Razinkov, Detection of IgG aggregation by a high throughput method, based on extrinsic fluorescence, J. Pharm. Sci. 99 (2010) 2598e2608. [33] W. Wang, C.J. Roberts (Eds.), Aggregation of Therapeutic Proteins, John Wiley, Hoboken, NJ, 2007. [34] S. Krishnan, M.M. Pallitto, M.S. Ricci, Development of formulations for therapeutic monoclonal antibodies and Fc fusion proteins, in: F. Jameel, S. Hershenson (Eds.), Formulation and Process Development Strategies for Manufacturing Biopharmaceuticals, John Wiley, Hoboken, NJ, 2010, pp. 383e427. [35] B. Kukrer, V. Filipe, E. van Duijn, P.T. Kasper, R.J. Vreeken, A. Heck, W. Jiskoot, Mass spectrometric analysis of intact human monoclonal antibody aggregates fractionated by size-exclusion chromatography, Pharm. Res. 27 (2010) 2197e2204. [36] M.D. Bond, M.E. Panek, Z. Zhang, D. Wang, P. Mehndiratta, H. Zhao, K. Gunton, A. Ni, M.L. Nedved, S. Burman, D.B. Volkin, Evaluation of a dual-wavelength size exclusion HPLC method with improved sensitivity to detect protein aggregates and its use to better characterize degradation pathways of an IgG1 monoclonal antibody, J. Pharm. Sci. 99 (2010) 2582e2597. [37] S. Ito, K. Wakabayashi, O. Ubukata, S. Hayashi, F. Okada, T. Hata, Crystal structure of the extracellular domain of mouse RANK ligand at 2.2-Å resolution, J. Biol. Chem. 277 (2002) 6631e6636. [38] K.K. Arthur, J.P. Gabrielson, N. Hawkins, D. Anafi, J. Wypych, A. Nagi, J.K. Sullivan, P.V. Bondarenko, In vitro stoichiometry of complexes between the soluble RANK ligand and the monoclonal antibody denosumab, Biochemistry 51 (2012) 795e806. [39] X. Chen, G.C. Flynn, A high throughput dimer screening assay for monoclonal antibodies using chemical cross-linking and microchip electrophoresis, J. Chromatogr. B 877 (2009) 3012e3018. [40] J.L. Teeling, T. Jansen-Hendriks, T.W. Kuijpers, M. de Haas, J.G. van de Winkel, C.E. Hack, W.K. Bleeker, Therapeutic efficacy of intravenous immunoglobulin preparations depends on the immunoglobulin G dimers: studies in experimental immune thrombocytopenia, Blood 98 (2001) 1095e1099. [41] T. Fifis, A. Gamvrellis, B. Crimeen-Irwin, G.A. Pietersz, J. Li, P.L. Mottram, I.F. McKenzie, M. Plebanski, Size-dependent immunogenicity: therapeutic and protective properties of nano-vaccines against tumors, J. Immunol. 173 (2004) 3148e3154. [42] S. Hermeling, H. Schellekens, C. Maas, M.F. Gebbink, D.J. Crommelin, W. Jiskoot, Antibody response to aggregated human interferon a2b in wildtype and transgenic immune tolerant mice depends on type and level of aggregation, J. Pharm. Sci. 95 (2006) 1084e1096. [43] V.S. Purohit, C.R. Middaugh, S.V. Balasubramanian, Influence of aggregation on immunogenicity of recombinant human factor VIII in hemophilia A mice, J. Pharm. Sci. 95 (2006) 358e371. [44] A.H. Fradkin, J.F. Carpenter, T.W. Randolph, Immunogenicity of aggregates of recombinant human growth hormone in mouse models, J. Pharm. Sci. 98 (2009) 3247e3264. [45] C. Dolman, S.J. Thorpe, R. Thorpe, Enhanced efficacy of anti-D immunoglobulin for treating ITP is not explained by higher immunoglobulin polymer content, Biologicals 29 (2001) 75e79.