Conformational and Colloidal Stabilities of Isolated Constant Domains

Article pubs.acs.org/molecularpharmaceutics

Conformational and Colloidal Stabilities of Isolated Constant Domains of Human Immunoglobulin G and Their Impact on Antibody Aggregation under Acidic Conditions Seiki Yageta,† Timothy M. Lauer,‡ Bernhardt L. Trout,‡ and Shinya Honda*,†,§ †

Department of Medical Genome Sciences, Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8562, Japan ‡ Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02319, United States § Biomedical Research Institute, National Institute of Advanced Industrial Science and Technology, Central 6, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan S Supporting Information *

ABSTRACT: Antibody therapeutics are now in widespread use and provide a new approach for treating serious diseases such as rheumatic diseases and cancer. Monoclonal antibodies used as therapeutic agents must be of high quality, and their safety must be guaranteed. Aggregated antibody is a degradation product that may be generated during the manufacturing process. To maintain the high quality and safety of antibody therapeutics, it is necessary to understand the mechanism of aggregation and to develop technologies to strictly control aggregate formation. Here, we extensively investigated the conformational and colloidal characteristics of isolated antibody constant domains, and provided insights into the molecular mechanism of antibody aggregation. Isolated domains (CH2, CH3, CL, and CH1-CL dimer) of human immunoglobulin G were synthesized, solubilized using 49 sets of solution conditions (pH 2−8 and 0−300 mM NaCl), and characterized using circular dichroism, intrinsic tryptophan fluorescence, and dynamic light scattering. Salt-induced conformational changes and oligomer formation were kinetically analyzed by NaCl-jump measurements (from 0 to 300 mM at pH 3). Phase diagrams revealed that the domains have different conformational and colloidal stabilities. The unfolded fractions of CH3 and CH2 at pH 3 were larger than that of CL and CH1-CL dimer. The secondary and tertiary structures and particle sizes of CH3 and CH2 showed that, in non-native states, these domains were sensitive to salt concentration. Kinetic analyses suggest that oligomer formation by CH3 and CH2 proceeds through partially refolded conformations. The colloidal stability of CH3 in non-native states is the lowest of the four domains under the conditions tested. We propose that the impact of IgG constant domains on aggregation follows the order CH3 > CH2 > CH1-CL dimer > CL; furthermore, we suggest that CH3 plays the most critical role in driving intact antibody aggregation under acidic conditions. KEYWORDS: antibody aggregation, antibody domain, acid denaturation, conformational and colloidal stabilities, empirical phase diagram

■

often aggregate,3,8 suggesting that conformational stability toward such stresses may decrease the propensity to aggregate. However, some proteins remain monomeric and monodisperse in the unfolded state, suggesting that the colloidal stability of the unfolded state would impact the propensity of a protein to aggregate. The conformational and colloidal stabilities of a protein likely depend both on the protein (amino acid composition, sequence, and structure) and environment (buffer, salts, and other solvent components).9−12 Consequently, assessing the propensity and exploring the mechanisms

INTRODUCTION Monoclonal antibodies are successful drugs because they exhibit significant therapeutic potential with few side effects. However, antibodies are less stable than low molecular weight chemical compounds and are prone to chemical and physical degradation.1 Aggregated antibody is a degraded product that may be generated during the manufacturing process,2,3 and can exhibit low efficacy and trigger immunogenic responses.4,5 The generation of higher quality antibody therapeutics requires understanding the mechanisms underlying aggregation and establishing production technologies that can strictly control aggregate formation. The tendency of a protein to aggregate likely depends on its conformational and colloidal stabilities.6,7 Proteins unfold upon exposure to stresses such as heat, pH, and agitation, and then © 2015 American Chemical Society

Received: Revised: Accepted: Published: 1443

November 15, 2014 March 17, 2015 April 14, 2015 April 14, 2015 DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Molecular Pharmaceutics

■

underlying aggregation require the systematic investigation of both the conformational and colloidal stability of a protein in a wide range of solution conditions. Immunoglobulin G (IgG) is a multidomain protein exhibiting complex molecular behavior, but recent studies have reported a relationship between the aggregation reaction of the antibody and the conformational and colloidal stabilities of its domains. Calorimetry experiments showed that the CH2, CH3, and Fab regions unfold at different temperatures.13,14 Enk et al. reported that thermally unfolded aglycosylated CH2 region led to aggregation of Fc, and that the presence of anions destabilized the CH2 region and accelerated the aggregation reaction.13 Kim et al. reported that aggregation rates for intact antibody were strongly influenced by the conformational stability of the Fab region.14 Furthermore, Buchner and coworkers showed that murine IgG1 domains (i.e., whole-IgG1, Fab, CH3, VH, VL, CH1, and CL) form molten-globule-like intermediate structures under specific acidic conditions (pH 2 and 100 mM NaCl; CL: pH 2, 175 mM NaCl).15−18 These intermediate structures, also called the alternatively folded state (AFS), exhibit molecular properties unique from both native and random coil conformations. Moreover, antibody domains in the AFS generally oligomerize, suggesting that the AFS is involved in antibody aggregation mechanisms. Taken together, the evidence to date suggests that each antibody domain has the potential to induce antibody aggregation, and that aggregation occurs through complicated interactions between multiple antibody domains, each of which may have different conformational and colloidal stabilities. Since understanding antibody aggregation mechanisms is clearly challenging, we propose that an extensive investigation of the conformational and colloidal stabilities of individual antibody domains is a useful approach toward understanding antibody aggregation mechanisms. Here, we synthesized four recombinant proteins corresponding to the constant domains of human IgG1 (CH2, CH3, CL, and CH1-CL dimer), characterized their conformational and colloidal stabilities under a wide range of solvent conditions (pH 2−8, 0−300 mM NaCl), and deduced the impact of their stabilities on the aggregation of the intact antibody. We focused on unfolding under acid conditions at room temperature because antibody therapeutics are exposed to acidic conditions during the manufacturing process (e.g., protein A affinity chromatography and virus inactivation). These acidic conditions are known to increase the risk of antibody aggregation.3,19−22 The elution buffer with a pH between 2.5 and 4.1 is used for elution of bound antibodies from protein A column.3,23−25 Effective virus inactivation is performed pH CH3 > CH1-CL dimer > CL. Determination of the Order of Colloidal Instabilities among Antibody Domains. NaCl induced oligomerization of CH3 and CH2 (Figures 2 and 5): oligomerization of CH3 was observed at 50 mM NaCl at pH 2 and at 100 mM NaCl at pH 3 (Figure 2a). Although CH3 showed a unique conformation corresponding to the PPNN state at 100 mM NaCl at pH 2 and pH 3 (Figure 1a,b), the particle size continued to increase as the NaCl concentration increased up to 300 mM, suggesting that CH3 in the PPNN state is prone to aggregation, and increasing NaCl concentration increasingly

Figure 3. Oligomeric state analyses using blue native gel polyacrylamide electrophoresis (BN−PAGE) and size exclusion chromatography (SEC). (a) BN−PAGE of CH2 (lanes 1 and 5), CH3 (lanes 2 and 6), CL (lanes 3 and 7), and CH1-CL dimer (lanes 4 and 8) at pH 6.8. BN−PAGE was electrophoresed on 4−20% polyacrylamide gradient gels. The disulfide bonds of samples applied to lanes 1 to 4 were reduced by treating with 5% beta-mercaptoethanol (β-ME) prior to analysis, whereas the cysteines in the samples applied to lanes 5 to 8 were oxidized. SEC of the domains at pH 7, 150 mM NaCl (b), pH 2, 0 mM NaCl (c), and pH 3, 300 mM NaCl (d).

concentrations of 175 mM and higher, but CL remained monomeric. The physicochemical characteristics of murine and human CL would also be very similar. Determination of the Order of Conformational Instabilities among Antibody Domains. The conformational transition from N state to non-native state(s) was observed at a different acidic pH for each domain (Figures 1 and 5). The EPDs show boundaries between pH 3 and pH 4 1449

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Article

Molecular Pharmaceutics

Figure 5. Generalized phase diagrams for the antibody constant domains. The N state, MMNN state, PPNN state, and a partially distorted state in CH3 (a), CH2 (b), CL (c), and CH1-CL dimer (d) are shown, as applicable.

weakened intermolecular repulsion. In contrast, higher NaCl concentrations were required for CH2 to adopt the PPNN state at pH 2 and pH 3 (Figures 2b and 5b), suggesting that the nonnative states of CH2 are more resistant to salt-induced oligomerization. CH2 oligomerized at pH 4 and 200 mM NaCl (Figure 2b); the conformation of CH2 under these conditions likely represents the lowest conformational stability of any of the IgG domains. The multimodal particle distribution of CH2 at pH 4 (Figure S4i,j in the Supporting Information) indicates the presence of a small fraction of unfolded CH2, which would oligomerize. Therefore, under conditions of mild acidic pH (∼pH 4) and high NaCl concentration (∼200 mM), the low conformational stability of CH2 would be the primary trigger for antibody aggregation. At pH 2 and pH 3, both CL and CH1-CL dimer did not oligomerize (Figures 2c,d, and 5c,d), suggesting that CL and CH1-CL dimer have the highest colloidal stabilities among the antibody constant domains. Taking the particle size of each domain in the PPNN state at pH 3, 300 mM NaCl (Table 1), as a quantitative indicator of colloidal instability, the order of the colloidal instabilities of the antibody constant domains is CH3 > CH2 > CH1-CL dimer ≈ CL. Impact of the Different Stabilities of the Constant Domains on Whole Antibody Aggregation. Our equilibrium and kinetic analyses revealed that each constant domain exhibited different conformational and colloidal stabilities, with CH2 being the least stable under acidic conditions (Figures 1 and 6). Studies of heat-induced antibody unfolding showed that conformational instability followed the order CH2 > Fab region > CH3; the presence or absence of the oligosaccharide of CH2 had no effect.13,14,47−50 Therefore, the order of the acidinduced conformational instability of antibody domains, presented here, is similar (but not identical) to the order observed for heat-induced instability. The colloidal stabilities of CH3 and CH2 are lower than those of CL and CH1-CL dimer (Figures 2 and 5; Table 1). Several reports suggest that the Fc region aggregates under heat and acid stress prior to the whole antibody aggregating,13,50,51

in agreement with our experimental results with isolated antibody domains. Therefore, although isolated domains and the same domains covalently integrated into the intact antibody may exhibit somewhat different molecular behavior, aggregation of the intact antibody is largely dependent on the molecular behavior of the individual antibody domain, each with its unique conformational and colloidal stability. Although Kim et al. reported that aggregation rates are strongly influenced by the conformational instability of the Fab region,14 we observed that CL and CH1-CL dimer did not oligomerize (Figure 2c,d), suggesting that the propensity of the Fab region to aggregate could be largely due to the instability of the VH and VL domains. Indeed, Galber and Demarest reported that Fab regions from different antibodies showed different stabilities and aggregation propensities.47 Taken together with the stabilities of antibody constant domains presented here and those reported recently, we propose that the order of the impact of the antibody constant domains on the aggregation of the intact antibody is CH3 > CH2 > CH1-CL dimer > CL. Aggregation Propensities of Antibody Domains Estimated from Several Protein Properties. What characteristics contribute to the different aggregation propensities? Although the structures of the four constant domains are almost identical (RMSD of backbone atoms CL > CH1-CL dimer (Figure 8e); this order is very similar to the order of colloidal instability determined in the present study (CH3 > CH2 > CH1-CL dimer ≈ CL).

We next searched the aggregation prone regions (APRs) of the domains using several sequence-based aggregation prediction algorithms: Aggrescan,33 PASTA,34 Zyggregator,35−38 and Tango.39 All algorithms predicted several APRs at approximately similar positions (Figure S6 in the Supporting Information). However, the order of the output scores of three of the algorithms did not coincide with the experimentally observed order of colloidal instability (Figure 7): only Zyggregator provided an output consistent with our experimental data. Zyggregator estimates the aggregation propensity score based on several properties of the polypeptide chain; the hydrophobicity, α-helical or β-sheet propensities, charge, hydrophobic patterns, and gatekeeper residues. The weights of properties were obtained by fitting on a database of protein aggregation rates determined experimentally. Although each independent amino acid composition did not suggest any correlation as mentioned above, appropriate combinations of 1451

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Article

Molecular Pharmaceutics

Figure 7. Sequence-based estimation of aggregation propensity. Na4vSS (Aggrescan) (a), best energy (PASTA) (b), Zagg (Zyggregator) (c), and the Agg parameter (Tango) (d) were plotted against the experimental order of colloidal instability of each domain.

Figure 8. Surface-based estimation of aggregation propensity. SAP values were mapped on the structures of CH1 (a), CH2 (b), CH3 (c), and CL (d). interface side and outer side are illustrated using the native structure of IgG (PDB ID: 1N8Z and 3D6G). Green regions represent hydrophobic regions, and white regions are hydrophilic regions. (e) The calculation results of the SAP score, net charge, and DI of the domains.

The aggregation propensities predicted by SAP/DI calculations on the static crystal structure agreed with our

experimental results, despite the aggregation-prone conformation of the PPNN state being clearly different from the native 1452

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Article

Molecular Pharmaceutics

only for the understanding of the antibody aggregation mechanism but also for development of the manufacturing process, drug formulation, and antibody engineering based on the molecular behavior of antibody domains.

structure. The PPNN state was also different from the random coil structure. This led to speculation that domains in the PPNN state adopt partially structured conformations whose surface properties, such as hydrophobicity and electrostatic potential, are somewhat similar to those of the native structure in the N state. The relationship between SAP/DI and Zyggregator with our experimental result also suggests that relatively small hydrophobic surface patches and/or short hydrophobic sequence segments primarily participate in aggregate formation. DI calculations suggested that the CH3 homodimer is less prone to aggregation than the CH3 monomer (Figure 8e). This correlates with our experimental results that CH3 remained a stable, dispersed dimer in the absence of acid stress, and that the CH3 dimer dissociated into monomers and oligomerized at pH 3 and below (Figures 2a, 3b−d, and 5a). Therefore, the dissociation of CH3 dimer by acidic conditions is strongly associated with the higher aggregation propensity of CH3. This viewpoint raises the possibility that CH1 aggregates when the CH1-CL dimer is dissociated by acid conditions. Feige et al. reported that murine CH1 formed oligomers at pH 2, 100 mM NaCl.18 However, CH1-CL dimer did not form oligomers in the present experiments. The higher colloidal stability of CH1CL dimer may arise from the close proximity of CH1 and CL, since the two domains are covalently connected through their C terminal disulfide bond, allowing CL to function as a “solubility tag” and improve the solubility of CH1.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional materials and methods, additional results, additional discussion, Tables S1 and S2, and Figures S1−S6. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: +81 29 862 6737. Fax: +81 29 861 6194. E-mail: s. [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Hideki Watanabe, Dr. Hisashi Takahashi, Dr. Hiroshi Imamura, Ayako Ooishi, and Dr. Feng Yan Wen in our research group for their technical support, discussions, and advice. This study was financially supported in part by a grant for the project focused on developing key technology of discovering and manufacturing drug for next-generation treatment and diagnosis from the Ministry of Economy, Trade and Industry, Japan.

■

■

CONCLUSIONS We conducted a systematic investigation of the conformational and colloidal stabilities of the isolated constant domains (CH3, CH2, CL, and CH1-CL dimer) of human IgG under a wide range of acidic pH and salt conditions (pH 2−8 and 0−300 mM NaCl). The data indicate that the 49 substates can be classified into three major states: the N state, MMNN state, and PPNN state. CD and fluorescence measurements showed that the order of conformational instability is CH2 > CH3 > CH1CL dimer > CL; DLS measurements showed that the order of colloidal instability is CH3 > CH2 > CH1-CL dimer ≈ CL. Taken together, the data suggest that the order of the impact of the antibody constant domains on the aggregation of the intact antibody is CH3 > CH2 > CH1-CL dimer > CL. High NaCl concentrations at pH 2 and pH 3 induced CH3 and CH2 oligomerization; furthermore, CH3 and CH2 adopted conformations distinct from native and random coil structures, corresponding to the PPNN state. Kinetic analyses showed that CH3 and CH2 oligomerization proceeded through partially refolded conformations characteristic of the PPNN state. The colloidal stability of CH3 in the PPNN state was the lowest of the antibody constant domains; consequently, CH3 oligomerized the fastest and formed the largest aggregates. Our experiments also suggest that the low conformational stability of CH2 may trigger the aggregation of intact antibody at mild acidic pH (∼pH 4) and high salt concentrations (∼200 mM NaCl). Taken together with recent reports, current evidence suggests that the tendency of an intact antibody to aggregate will largely depend on the conformational and colloidal stabilities of the domains. The colloidal stabilities estimated by SAP/DI calculations agreed with our experimentally determined order, suggesting that acid-induced antibody aggregation involves partially structured conformations with somewhat native-like surface properties and not random coil structures. Our investigation will provide useful information not

ABBREVIATIONS USED AFS, alternatively folded state; APR, aggregation prone region; BN−PAGE, blue native polyacrylamide gel electrophoresis; CD, circular dichroism; DI, developability index; DLS, dynamic light scattering; EPD, empirical phase diagram; Fab, fragment antigen binding; Fc, fragment crystallizable; IgG, immunoglobulin G; N state, native state; MMNN state, monomeric monodispersed non-native state; PCR, polymerase chain reaction; PPNN state, polymeric polydispersed non-native state; PSD, particle size diagram; RMSD, root-mean-square deviation; SAP, spatial aggregation propensity; SDS−PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SEC, size-exclusion chromatography; SVD, singular value decomposition

■

REFERENCES

(1) Wang, W.; Singh, S.; Zeng, D. L.; King, K.; Nema, S. Antibody structure, instability, and formulation. J. Pharm. Sci. 2007, 96 (1), 1− 26. (2) ICH Guideline Q6B: Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products. 10 March 1999. (3) Vázquez-Rey, M.; Lang, D. A. Aggregates in monoclonal antibody manufacturing processes. Biotechnol. Bioeng. 2011, 108 (7), 1494−508. (4) Rosenberg, A. S. Effects of protein aggregates: an immunologic perspective. AAPS J. 2006, 8 (3), E501−7. (5) Wang, W.; Singh, S. K.; Li, N.; Toler, M. R.; King, K. R.; Nema, S. Immunogenicity of protein aggregates–concerns and realities. Int. J. Pharm. 2012, 431 (1−2), 1−11. (6) Chi, E. Y.; Krishnan, S.; Randolph, T. W.; Carpenter, J. F. Physical stability of proteins in aqueous solution: mechanism and driving forces in nonnative protein aggregation. Pharm. Res. 2003, 20 (9), 1325−36. (7) Saito, S.; Hasegawa, J.; Kobayashi, N.; Tomitsuka, T.; Uchiyama, S.; Fukui, K. Effects of ionic strength and sugars on the aggregation propensity of monoclonal antibodies: influence of colloidal and conformational stabilities. Pharm. Res. 2013, 30 (5), 1263−80. 1453

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Article

Molecular Pharmaceutics

(28) Kim, J. H.; Iyer, V.; Joshi, S. B.; Volkin, D. B.; Middaugh, C. R. Improved data visualization techniques for analyzing macromolecule structural changes. Protein Sci. 2012, 21 (10), 1540−53. (29) Hendler, R. W.; Shrager, R. I. Deconvolutions based on singular value decomposition and the pseudoinverse: a guide for beginners. J. Biochem. Biophys. Methods 1994, 28 (1), 1−33. (30) Wittig, I.; Braun, H. P.; Schägger, H. Blue native PAGE. Nat. Protoc. 2006, 1 (1), 418−28. (31) Goh, C. S.; Lan, N.; Douglas, S. M.; Wu, B.; Echols, N.; Smith, A.; Milburn, D.; Montelione, G. T.; Zhao, H.; Gerstein, M. Mining the structural genomics pipeline: identification of protein properties that affect high-throughput experimental analysis. J. Mol. Biol. 2004, 336 (1), 115−30. (32) Idicula-Thomas, S.; Balaji, P. V. Understanding the relationship between the primary structure of proteins and its propensity to be soluble on overexpression in Escherichia coli. Protein Sci. 2005, 14 (3), 582−92. (33) Conchillo-Solé, O.; de Groot, N. S.; Avilés, F. X.; Vendrell, J.; Daura, X.; Ventura, S. AGGRESCAN: a server for the prediction and evaluation of “hot spots” of aggregation in polypeptides. BMC Bioinf. 2007, 8, 65. (34) Trovato, A.; Chiti, F.; Maritan, A.; Seno, F. Insight into the structure of amyloid fibrils from the analysis of globular proteins. PLoS Comput. Biol. 2006, 2 (12), 1608−18. (35) DuBay, K. F.; Pawar, A. P.; Chiti, F.; Zurdo, J.; Dobson, C. M.; Vendruscolo, M. Prediction of the absolute aggregation rates of amyloidogenic polypeptide chains. J. Mol. Biol. 2004, 341 (5), 1317− 26. (36) Pawar, A. P.; Dubay, K. F.; Zurdo, J.; Chiti, F.; Vendruscolo, M.; Dobson, C. M. Prediction of “aggregation-prone” and “aggregationsusceptible” regions in proteins associated with neurodegenerative diseases. J. Mol. Biol. 2005, 350 (2), 379−92. (37) Tartaglia, G. G.; Pawar, A. P.; Campioni, S.; Dobson, C. M.; Chiti, F.; Vendruscolo, M. Prediction of aggregation-prone regions in structured proteins. J. Mol. Biol. 2008, 380 (2), 425−36. (38) Tartaglia, G. G.; Vendruscolo, M. The Zyggregator method for predicting protein aggregation propensities. Chem. Soc. Rev. 2008, 37 (7), 1395−401. (39) Fernandez-Escamilla, A. M.; Rousseau, F.; Schymkowitz, J.; Serrano, L. Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins. Nat. Biotechnol. 2004, 22 (10), 1302−6. (40) Chennamsetty, N.; Voynov, V.; Kayser, V.; Helk, B.; Trout, B. Design of therapeutic proteins with enhanced stability. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (29), 11937−42. (41) Chennamsetty, N.; Helk, B.; Voynov, V.; Kayser, V.; Trout, B. Aggregation-Prone Motifs in Human Immunoglobulin G. J. Mol. Biol. 2009, 391 (2), 404−13. (42) Chennamsetty, N.; Voynov, V.; Kayser, V.; Helk, B.; Trout, B. Prediction of Aggregation Prone Regions of Therapeutic Proteins. J. Phys. Chem. B 2010, 114 (19), 6614−24. (43) Lauer, T. M.; Agrawal, N. J.; Chennamsetty, N.; Egodage, K.; Helk, B.; Trout, B. L. Developability index: a rapid in silico tool for the screening of antibody aggregation propensity. J. Pharm. Sci. 2012, 101 (1), 102−15. (44) Cho, H. S.; Mason, K.; Ramyar, K. X.; Stanley, A. M.; Gabelli, S. B.; Denney, D. W.; Leahy, D. J. Structure of the extracellular region of HER2 alone and in complex with the Herceptin Fab. Nature 2003, 421 (6924), 756−60. (45) Moiani, D.; Salvalaglio, M.; Cavallotti, C.; Bujacz, A.; Redzynia, I.; Bujacz, G.; Dinon, F.; Pengo, P.; Fassina, G. Structural characterization of a Protein A mimetic peptide dendrimer bound to human IgG. J. Phys. Chem. B 2009, 113 (50), 16268−75. (46) Olsson, M.; Sondergaard, C.; Rostkowski, M.; Jensen, J. PROPKA3: Consistent Treatment of Internal and Surface Residues in Empirical pK(a) Predictions. J. Chem. Theory Comput. 2011, 7 (2), 525−37.

(8) Wang, W. Protein aggregation and its inhibition in biopharmaceutics. Int. J. Pharm. 2005, 289 (1−2), 1−30. (9) Kameoka, D.; Masuzaki, E.; Ueda, T.; Imoto, T. Effect of buffer species on the unfolding and the aggregation of humanized IgG. J. Biochem 2007, 142 (3), 383−91. (10) Feng, Y. W.; Ooishi, A.; Honda, S. Aggregation factor analysis for protein formulation by a systematic approach using FTIR, SEC and design of experiments techniques. J. Pharm. Biomed. Anal. 2012, 57, 143−52. (11) Bhambhani, A.; Kissmann, J. M.; Joshi, S. B.; Volkin, D. B.; Kashi, R. S.; Middaugh, C. R. Formulation design and high-throughput excipient selection based on structural integrity and conformational stability of dilute and highly concentrated IgG1 monoclonal antibody solutions. J. Pharm. Sci. 2012, 101 (3), 1120−35. (12) Mueller, M.; Loh, M. Q.; Tee, D. H.; Yang, Y.; Jungbauer, A. Liquid formulations for long-term storage of monoclonal IgGs. Appl. Biochem. Biotechnol. 2013, 169 (4), 1431−48. (13) Zhang-van Enk, J.; Mason, B. D.; Yu, L.; Zhang, L.; Hamouda, W.; Huang, G.; Liu, D.; Remmele, R. L.; Zhang, J. Perturbation of thermal unfolding and aggregation of human IgG1 Fc fragment by Hofmeister anions. Mol. Pharmaceutics 2013, 10 (2), 619−30. (14) Kim, N.; Remmele, R. L.; Liu, D.; Razinkov, V. I.; Fernandez, E. J.; Roberts, C. J. Aggregation of anti-streptavidin immunoglobulin gamma-1 involves Fab unfolding and competing growth pathways mediated by pH and salt concentration. Biophys. Chem. 2013, 172, 26− 36. (15) Buchner, J.; Renner, M.; Lilie, H.; Hinz, H. J.; Jaenicke, R.; Kiefhabel, T.; Rudolph, R. Alternatively folded states of an immunoglobulin. Biochemistry 1991, 30 (28), 6922−9. (16) Lilie, H.; Buchner, J. Domain interactions stabilize the alternatively folded state of an antibody Fab fragment. FEBS Lett. 1995, 362 (1), 43−6. (17) Thies, M. J.; Kammermeier, R.; Richter, K.; Buchner, J. The alternatively folded state of the antibody C(H)3 domain. J. Mol. Biol. 2001, 309 (5), 1077−85. (18) Feige, M. J.; Simpson, E. R.; Herold, E. M.; Bepperling, A.; Heger, K.; Buchner, J. Dissecting the alternatively folded state of the antibody Fab fragment. J. Mol. Biol. 2010, 399 (5), 719−30. (19) Phillips, J.; Drumm, A.; Harrison, P.; Bird, P.; Bhamra, K.; Berrie, E.; Hale, G. Manufacture and quality control of CAMPATH-1 antibodies for clinical trials. Cytotherapy 2001, 3 (3), 233−42. (20) Cromwell, M. E.; Hilario, E.; Jacobson, F. Protein aggregation and bioprocessing. AAPS J. 2006, 8 (3), E572−9. (21) Ejima, D.; Tsumoto, K.; Fukada, H.; Yumioka, R.; Nagase, K.; Arakawa, T.; Philo, J. S. Effects of acid exposure on the conformation, stability, and aggregation of monoclonal antibodies. Proteins 2007, 66 (4), 954−62. (22) Gagnon, P. Technology trends in antibody purification. J. Chromatogr. A 2012, 1221, 57−70. (23) Fahrner, R. L.; Knudsen, H. L.; Basey, C. D.; Galan, W.; Feuerhelm, D.; Vanderlaan, M.; Blank, G. S. Industrial purification of pharmaceutical antibodies: development, operation, and validation of chromatography processes. Biotechnol. Genet. Eng. Rev. 2001, 18, 301− 27. (24) Shukla, A. A.; Hubbard, B.; Tressel, T.; Guhan, S.; Low, D. Downstream processing of monoclonal antibodies–application of platform approaches. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2007, 848 (1), 28−39. (25) Ayyar, B. V.; Arora, S.; Murphy, C.; O’Kennedy, R. Affinity chromatography as a tool for antibody purification. Methods 2012, 56 (2), 116−29. (26) Brorson, K.; Krejci, S.; Lee, K.; Hamilton, E.; Stein, K.; Xu, Y. Bracketed generic inactivation of rodent retroviruses by low pH treatment for monoclonal antibodies and recombinant proteins. Biotechnol. Bioeng. 2003, 82 (3), 321−9. (27) Ramsey, J. D.; Gill, M. L.; Kamerzell, T. J.; Price, E. S.; Joshi, S. B.; Bishop, S. M.; Oliver, C. N.; Middaugh, C. R. Using empirical phase diagrams to understand the role of intramolecular dynamics in immunoglobulin G stability. J. Pharm. Sci. 2009, 98 (7), 2432−47. 1454

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455

Article

Molecular Pharmaceutics (47) Garber, E.; Demarest, S. J. A broad range of Fab stabilities within a host of therapeutic IgGs. Biochem. Biophys. Res. Commun. 2007, 355 (3), 751−7. (48) Fesinmeyer, R. M.; Hogan, S.; Saluja, A.; Brych, S. R.; Kras, E.; Narhi, L. O.; Brems, D. N.; Gokarn, Y. R. Effect of ions on agitationand temperature-induced aggregation reactions of antibodies. Pharm. Res. 2009, 26 (4), 903−13. (49) Hari, S. B.; Lau, H.; Razinkov, V. I.; Chen, S.; Latypov, R. F. Acid-induced aggregation of human monoclonal IgG1 and IgG2: molecular mechanism and the effect of solution composition. Biochemistry 2010, 49 (43), 9328−38. (50) Zheng, K.; Bantog, C.; Bayer, R. The impact of glycosylation on monoclonal antibody conformation and stability. MAbs 2011, 3 (6), 568−76. (51) Latypov, R. F.; 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), 1381−96.

1455

DOI: 10.1021/mp500759p Mol. Pharmaceutics 2015, 12, 1443−1455