- Email: [email protected]
Effect of the conserved oligosaccharides of recombinant monoclonal antibodies on the separation by protein A and protein G chromatography
Journal of Chromatography A, 1216 (2009) 2382–2387
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Effect of the conserved oligosaccharides of recombinant monoclonal antibodies on the separation by protein A and protein G chromatography Georgeen Gaza-Bulseco, Keith Hickman, Sara Sinicropi-Yao, Karen Hurkmans, Chris Chumsae, Hongcheng Liu ∗ Process Sciences Department, Abbott Bioresearch Center, 100 Research Drive, Worcester, MA 01605, USA
a r t i c l e
i n f o
Article history: Received 10 September 2008 Received in revised form 30 December 2008 Accepted 6 January 2009 Available online 15 January 2009 Keywords: Recombinant monoclonal antibodies Protein A Protein G Oligosaccharides Mass spectrometry
a b s t r a c t Glycosylation of the conserved asparagine residue in CH2 domains of IgG molecules is an important post-translational modification. The presence of oligosaccharides is critical for structure, stability and biological function of IgG antibodies. Effect of the glycosylation states of recombinant monoclonal antibodies on protein A and protein G chromatography was evaluated. Antibodies lacking oligosaccharides eluted later from protein A and earlier from protein G columns than antibodies with oligosaccharides using a gradient of decreasing pH. Interestingly, different types of oligosaccharides also affected the elution of the antibodies. Antibodies with high mannose type oligosaccharides were enriched in later eluting fractions from protein A and earlier eluting fractions from protein G. While antibodies with more mature oligosaccharides, such as core fucosylated biantennary complex oligosaccharides with zero (Gal 0), one (Gal 1) or two (Gal 2) terminal galactoses, were enriched in earlier eluting fractions from protein A and in the later eluting fractions from protein G. However, analysis by enzyme-linked immunosorbent assay (ELISA) revealed that antibody binding affinity to protein A and protein G was not affected by the absence or presence of oligosaccharides. It was thus concluded that the elution difference of antibodies with or without oligosaccharides and antibodies with different types of oligosaccharides were due to differential structural changes around the CH2–CH3 domain interface under the low pH conditions used for protein A and protein G chromatography. © 2009 Elsevier B.V. All rights reserved.
1. Introduction IgG molecules contain N-linked oligosaccharides on the conserved asparagine (Asn) residue in the CH2 domains. The absence of oligosaccharides as well as different types of oligosaccharides at this site does not affect antigen binding affinity [1–9], due to the structural separation of the Fab and Fc region by a highly flexible hinge. However, oligosaccharides are critical for the structural integrity, stability and the biological functions of the Fc region. It has been demonstrated that the absence of oligosaccharides altered the structure of the CH2 domains of antibodies and resulted in decreased stability [10–15]. Conformational changes caused by the absence of oligosaccharides are subtle and local to the regions around the glycosylation sites [1,14–16]. However, the consequences were significant on effector functions, since the Fc␥ receptors and the first component of the complement (C1q) binding sites are located in the hinge proximal region of the CH2 domains [17–19], where the structure was significantly affected. Absence of
∗ Corresponding author. Tel.: +1 508 849 2591; fax: +1 508 793 4885. E-mail address: [email protected] (H. Liu). 0021-9673/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2009.01.014
oligosaccharides resulted in antibodies with decreased to complete loss of binding to Fc␥ receptors [4,5,12–15,20–25], which resulted in decreased to complete abolishment of biological functions such as antibody-dependent-cellular-cytotoxicity (ADCC) [4,5,8,14,26]. Absence of oligosaccharides also caused a decrease or loss of binding to C1q [15,17,23,24], which resulted in decreased or complete loss of complement activation and complement dependent cytotoxicity (CDC) [4,5,7,8,13,15,17,23,24]. Antibodies containing different types of oligosaccharides also showed significant differences in structure, stability and biological functions [1,2,8,12,13,22,26–29]. The presence of different types of oligosaccharides also caused a subtle conformational change around the glycosylation sites [1,16]. Antibodies with high mannose type oligosaccharides were deficient in C1q binding, complement consumption and CDC [1,2]. Antibodies that contained oligosaccharides without the core fucose showed a dramatic increase in the binding affinity to FcRIIIa and enhancement of ADCC [30–36]. Protein A and protein G share similar as well as distinct binding sites around the CH2–CH3 domain interface [37–39]. Although, Matsuda et al. [15] reported that human IgG-Fc expressed in E. coli, presumably lacking oligosaccharides, required a slightly lower pH for elution from a protein A column compared to Fc with oligosac-
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
charides, it is generally agreed that binding of IgG antibodies to protein A was not affected by the absence or presence of different types of oligosaccharides [3,4,8,15,23,40]. Protein G can bind more IgG subtypes and with stronger binding affinities [41,42]. The effect of oligosaccharides on the binding of IgG to protein G was unclear. Protein A and protein G are two important ligands of IgGFc and are commonly used for affinity chromatography to purify antibodies. To better understand the role of oligosaccharides, experiments were carried out in this study to investigate the effect of the absence of oligosaccharides and the presence of different types of oligosaccharides on the binding of antibodies to protein A and protein G. Two recombinant monoclonal antibodies, one fully human (MAb-A) and one humanized (MAb-B) were used in this study. MAb-A contains core fucosylated biantennary complex structures with zero (Gal 0), one (Gal 1) or two (Gal 2) terminal galactose residues. MAb-A also contains low levels of high mannose structures with five (M5), six (M6) or seven (M7) mannose residues. MAb-B contains approximately 30% non-glycosylated heavy chains, in addition to the same oligosaccharide structures as MAb-A.
2. Materials and methods 2.1. Materials MAb-A, a recombinant fully human monoclonal IgG1 antibody was produced by a transfected Chinese hamster ovary (CHO) cell line and purified by multiple chromatography steps (Abbott Bioresearch Center, Worcester, MA, USA). MAb-B, a recombinant humanized monoclonal IgG1 antibody was expressed by transient transfection of 293 cells and purified by protein A chromatography (Abbott Bioresearch Center). Dithiothreitol (DTT) was purchased from Sigma (St. Louis, MO, USA). Formic acid (FA) was purchased from EMD (Gibbbstown, NJ, USA). Lys-C, phenylmethylsulfonyl fluoride (PMSF) and N-octylglucoside were purchased from Roche (Indianapolis, IN, USA). Acetonitrile was purchased from J.T. Baker (Phillipsburg, NJ, USA) and peptide-N-glycosidase F (PNGaseF) was purchased from Prozyme (San Leandro, CA, USA).
2383
2.2. Deglycosylation of the recombinant monoclonal antibody MAb-A at approximately 70 mg/mL in formulation buffer (5.57 mM sodium phosphate monobasic, 8.69 mM sodium phosphate dibasic, 106.69 mM sodium chloride, 1.07 mM sodium citrate, 6.45 mM citric acid, 66.68 mM mannitol and 0.1% Tween) at pH, 5.2 was diluted to 10 mg/mL with water deionized by a Milli-Q system (Millipore, Bedford, MA, USA). N-Octylglucoside at a final concentration of 1% was included in the sample preparation to facilitate deglycosylation. PNGaseF (2.5 mU/L) was added to the diluted sample at the ratio of 1 L enzyme to 500 g antibody. A sample prepared the same as the sample for PNGaseF digestion but without the addition of PNGaseF was used as a control. The samples were incubated at 37 ◦ C for 18 h. 2.3. Protein A chromatography A Shimadzu HPLC and a Poros A column (50 mm × 4.6 mm, Applied Biosystems, Framingham, MA, USA) were used for sample analysis and fraction collection. One hundred L of each sample (10 mg/mL in PBS) was loaded at 100% mobile phase A [phosphate buffered saline (PBS), pH 7.4]. The column was washed using 100% mobile phase A for 10 min, and then eluted with an increase of mobile phase B (0.1 M acetic acid and 0.15 M sodium chloride, pH 2.9) from 0% to 60% in 30 min. The column was washed using 100% mobile phase B for 10 min and then equilibrated using 100% mobile phase A for another 10 min. The flow-rate was set at 2 mL/min. For fraction collection, five mg of either MAb-A or MAb-B was injected and analyzed by following the same procedure. Fractions were collected across the entire elution peak. Four mL of each fraction was collected in a 2-min elution time window. 2.4. Protein G chromatography A Shimadzu HPLC and a Poros G column (50 mm × 4.6 mm, Applied Biosystems) were used for sample analysis and fraction collection. One hundred L of each sample (10 mg/mL in PBS) was loaded at 80% mobile phase A and 20% mobile phase B with the pH adjusted to 2.5. The column was washed using 80% mobile phase A and 20% mobile phase B for 10 min, and then eluted with an increase
Fig. 1. Mass spectra of the reduced Fc of antibody MAb-A. The data was acquired from MAb-A with oligosaccharides (A) and from MAb-A after removal of the oligosaccharides using PNGaseF (B). Peak identities are labeled in the spectra.
2384
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
of mobile phase B from 20% to 100% in 40 min. The column was washed using 100% mobile phase B for 10 min and then equilibrated using 80% mobile phase A and 20% mobile phase B for an additional 10 min. The flow-rate was set at 2 mL/min. For fraction collection, five mg of MAb-A or MAb-B was injected and analyzed following the same procedure as described previously. Fractions were collected across the entire elution peak. Four mL of each fraction was collected in a 2-min elution time window. 2.5. Lys-C digestion and LC–MS analysis MAb-A, MAb-B and Protein A and G chromatography fractions collected from both antibodies were diluted to 0.5 mg/mL with PBS to a final volume of 50 L. Lys-C was added to each fraction at a final ratio of 1:200 (w/w) Lys-C: antibody, and digestion was carried out at 37 ◦ C for 30 min. PMSF dissolved in methanol was added to each sample at a final concentration of 2 mM to stop the Lys-C digestion. The samples were then reduced with 20 mM DTT at 37 ◦ C for 5 min. An Agilent HPLC (Santa Clara, CA) equipped with a protein C4 column (Vydac, 150 mm × 1 mm I.D., 5 m particle size, 300 A pore size) and a Qstar mass spectrometer (Applied Biosystems) were used to analyze the Lys-C digested samples. Ten L of each sample was loaded separately at 95% mobile phase A (0.08% FA in water deionized by a Milli-Q system (Millipore)) and 5% mobile phase B (0.08% FA in acetonitrile). After 5 min at 5% mobile phase B, proteins were eluted off the column by increasing mobile phase B to 65% over a period of 35 min. The column was washed with 95% mobile phase B for 5 min and then re-equilibrated at 5% mobile phase B for 10 min before the next injection. The flow-rate was set at 50 L/min. The column oven was set at 60 ◦ C. The mass spectrometer was operated at positive mode with a scan range of m/z 800–2500. IonSpray voltage was set at 4500 V, and the source temperature was set at 350 ◦ C. 2.6. ELISA The binding affinity of the antibody with and without oligosaccharides to protein A and protein G was compared by ELISA. The native and the deglycosylated antibody (20 g/mL) in 50 mM sodium bicarbonate buffer, pH 9.4, were diluted serially with a 2 fold dilution of each step. One hundred L of the serially diluted samples were immobilized onto three 96-well plates by incubating the plates overnight at 4 ◦ C with shaking. After removal of the unbound antibodies, the plates were blocked with SuperBlock blocking buffer (Pierce, Rockford, IL, USA) (300 L/well) for 1 h at 37 ◦ C with shaking. The plates were then washed 5 times with PBS with 0.1% Triton X-100 using a plate washer (Tecan, Zurich, Switzerland). The plates were incubated for 1 h at 37 ◦ C with 100 L/well
Fig. 2. Protein A and protein G chromatograms of MAb-A. MAb-A with oligosaccharides (Native) and after the removal of oligosaccharides (deglycosylated) were analyzed by protein A (A) and protein G (B) chromatography. Arrows indicate the points at which fractions were collected (mass spectra in Figs 4 and 5).
protein A-horse radish peroxidase (HRP) conjugate (Millipore, Billerica, MA), protein G-HRP or goat-anti-human IgG-HRP (Jackson ImmunoResearch, West Grove, PA, USA) that was diluted 50 000 times in SuperBlock. After one wash with PBS, 100 L/well of tetramethylbenzidine (TMB) substrate, K Blue (Neogen, Lansing, MI, USA), was added then incubated at room temperature for 5 min. The reaction was stopped by the addition of 100 L/well of 1 N phosphoric acid. The plates were then read at 450 nm using a Spectramax Plus plate reader (Molecular Devices, Sunnyvale, CA, USA). The data was fitted using Soft-max Pro Version 4.8 (Molecular Devices). 3. Results 3.1. Deglycosylation of MAb-A MAb-A contains two major post-translational modifications, glycosylation of the conserved Asn residue in the CH2 domains and incomplete processing of the C-terminal lysine (Lys) residues. To determine the effect of oligosaccharides on the binding of this antibody to protein A and protein G, oligosaccharides attached to MAb-A were removed by digestion with PNGaseF. The Lys-C digested and reduced MAb before and after PNGaseF digestion
Fig. 3. Mass spectrum of the reduced Fc of antibody MAb-B. Peak identities are labeled in the spectrum.
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
2385
Fig. 4. Mass spectra of representative protein A (left) and protein G (right) fractions of MAb-B from the early (A), middle (B) and later (C) sections of the chromatogram. Peak identities are labeled in the spectra.
was analyzed by LC–MS to confirm the glycosylation state of the samples. Lys-C cleaves the peptide bond between lysine and threonine in the hinge region to generate Fab and Fc. Three peaks were observed in the mass spectrum of the reduced Fc with molecular weights of 26617 Da, 26746 Da and 26780 Da, which corresponded to Fc with Gal 0 and without a C-terminal Lys, Gal 0 with a C-terminal Lys, and Gal 1 without a C-terminal Lys respectively (Fig. 1A). Only two peaks were observed after deglycosylation, with molecular weights of 25296 Da and 25171 Da, which corre-
sponded to the Fc with and without a C-terminal Lys respectively (Fig. 1B). 3.2. Protein A and protein G chromatography of MAb-A MAb-A with and without oligosaccharides was analyzed by protein A and protein G chromatography using a gradient of decreasing pH for elution. MAb-A with oligosaccharides eluted off the protein A column slightly earlier than the antibody without oligosaccharides
Fig. 5. Mass spectra of representative protein A (left) and protein G (right) fractions of MAb-A from the early (A), middle (B) and later (C) sections of the chromatogram. Peak identities are labeled in the spectra.
2386
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
(Fig. 2A). Interestingly, when the same samples were analyzed by protein G chromatography, MAb-A without oligosaccharides eluted off the column earlier than the antibody with oligosaccharides (Fig. 2B).
3.3. Protein A and protein G chromatography and mass spectrometry analysis of MAb-B Removal of oligosaccharides from MAb-A by PNGaseF digestion converted the original conserved Asn residue, to which the oligosaccharides were attached, to an aspartate (Asp) residue. To exclude the potential effect of this amino acid change on binding to protein A and protein G, a monoclonal antibody, MAb-B, that contained the original conserved Asn due to approximately 30% of the antibody being aglycosylated, was analyzed by protein A and protein G chromatography following the same procedure used for MAb-A. MAb-B also contained 70% glycosylated heavy chains with mainly Gal 0, Gal 1 and trace levels of Gal 2 and high mannose. As shown in Fig. 3, three peaks were observed in the mass spectrum of the reduced Fc of MAb-B with molecular weights of 25 119 Da, 26 564 Da and 26 727 Da, which corresponded to the reduced Fc without oligosaccharides, with Gal 0, and Gal 1, respectively. Heavy chain with Gal 2 was not detected. Protein A and protein G chromatography fractions of MAbB were collected across the entire elution peak. Mass spectra from representative fractions of the early, middle and late eluting sections of the elution peak from protein A and protein G chromatograms are shown in Fig. 4. There was a higher percentage of aglycosylated Fc that eluted in the late fractions from protein A and in the early fractions from protein G columns. This observation was in agreement with the results obtained with MAb-A. It was also interesting to note that there were slightly higher levels of Gal 0 than Gal 1 in the middle fraction and the appearance of Gal 2 in the late fraction from protein G.
3.4. LC–MS analysis of protein A and protein G fractions from MAb-A To investigate whether different types of oligosaccharides affected the elution from protein A and protein G, fractions from the elution peak of MAb-A from protein A and protein G chromatography were analyzed by LC–MS. Mass spectra of Fc fragments of representative fractions from the early, middle and late sections of the elution peaks from protein A and protein G chromatography are shown in Fig. 5. MAb-A with oligomannose eluted in the late fractions from protein A and in the early fractions from protein G columns.
3.5. ELISA An ELISA assay was used to compare the binding affinity of native and deglycosylated MAb-A to protein A and protein G. To assure that similar levels of the native and deglycosylated MAb-A were immobilized on the ELISA plates, the antibodies were probed with a goat-anti-human IgG-HRP. The dose response curves of the native and the deglycosylated MAb-A were similar, which confirmed that similar levels of the native and deglycosylated MAb-A were immobilized on the plates (Fig. 6C). The dose response curves of the native antibody and the deglycosylated antibody to either protein A-HRP or protein G-HRP were also similar, which indicated that antibody with and without oligosaccharides bound to protein A or protein G with similar affinity (Fig. 6A and B).
Fig. 6. Binding ELISA of the antibody to protein A and protein G. Native (diamond) and deglycosylated antibody (square) were immobilized and then detected with either protein A-HRP (A), protein G-HRP (B) or anti-human IgG-HRP (C).
4. Discussion The N-linked oligosaccharides of IgG are attached to the conserved Asn residue located in the hinge proximal region of the CH2 domain. The oligosaccharides at this site are mainly of the complex biantennary type, with low levels of oligomannose. The major binding sites of protein A and protein G are located in the CH2–CH3 interface and there are no known direct interactions with the Nlinked oligosaccharides [37,39]. The effect of oligosaccharides on the binding to protein A and protein G and elution of IgG from protein A as well as protein G resins were thus further investigated in the current study. Antibodies without oligosaccharides eluted later from protein A and earlier from protein G columns than native glycosylated antibodies. The elution differences were observed using antibodies without oligosaccharides containing either Asp or Asn. In contrast, when the binding of MAb-A with and without oligosaccharides to protein A and protein G was compared by ELISA, no differences were observed. Such a disagreement has been reported by Matsuda et al. [15] where they did not detect any binding difference between human IgG-Fc with and without oligosaccharides to protein A by ELISA, but they showed that human IgG-Fc without oligosaccharides eluted later than with oligosaccharides from protein A chromatography. On the other hand, Leatherbarrow and Dwek [40] showed that the elution profile of a mouse IgG with or without oligosaccharides from a protein A column was not different when a pH gradient was used. Since there are no direct interactions between the N-linked oligosaccharides and protein A and protein G [37,39], the absence or presence of oligosaccharides may not affect the binding directly. It is known that removal of oligosaccharides [11–13,15], as well as the presence of different types of oligosaccharides [12,13,16] affected antibody structure. These structural differences were local to the glycosylation site and may not be sig-
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
nificant enough to affect binding of the residues in the CH2–CH3 interface to protein A or protein G. This is in agreement with many studies that showed no difference in the binding of different IgG molecules and their Fc fragments to protein A in the presence or absence of oligosaccharides [3,4,8,15,23,40]. The only explanation for the differences in retention properties could be attributed to the effects of pH on the structure of the antibody with and without oligosaccharides. The structural changes in turn could have led to differences in the elution profiles of the antibodies with and without oligosaccharides from protein A and protein G columns. Dependence of the structure and stability of IgG molecules on pH has been studied extensively. For example, Kats et al. [43] reported that a monoclonal chimeric antibody showed a pH-dependent isoform transition. Acidic treatment of rabbit IgG resulted in increased stability due to an increased interaction between the CH1 and CH2 domains [44–46]. A previous study in our laboratory also showed that oligosaccharides slowed down the fragmentation rate of a recombinant monoclonal antibody at low pH (<4), probably due to a structural rearrangement[47]. It was thus proposed that antibodies with and without oligosaccharides adopted different structures under the low pH condition used for elution from protein A and protein G chromatography. Therefore, different structural changes of the antibodies without oligosaccharides at the pH conditions used for elution resulted in a stronger association or slower dissociation from protein G and a weaker association or faster dissociation from protein A. Interestingly, antibodies with high mannose oligosaccharides also eluted later from protein A and earlier from protein G columns compared to antibodies with more mature oligosaccharides such as Gal 0 and Gal 1. This suggests that antibodies with high mannose adopt a structure similar to antibodies without oligosaccharides and was affected similarly by the low pH condition. It was also evidenced that antibody with high mannose was slightly enriched in the late fraction from protein A column and highly enriched in the early fraction from protein G column (Fig. 5), which indicated that a more significant structural difference between antibodies with high mannose and antibody with complex oligosaccharides was observed under the slightly lower pH condition used for protein G chromatography. It was not clear if terminal galactose affected the elution of antibodies from protein A and protein G. The effect may be subtle and therefore varied from antibody to antibody significantly. It is interesting that the elution profiles of the antibodies without oligosaccharides or with high mannose type oligosaccharides compared to antibodies with more mature oligosaccharides were in opposite elution order for protein A and protein G. This result was unexpected especially since protein A and protein G bind to similar IgG residues within the CH2–CH3 domain [39]. The binding of IgG to protein A involves mainly hydrophobic interactions, while the binding to protein G involves mainly hydrogen bonding and ionic interactions. It was possible that structural changes caused by the elution pH enhanced the hydrogen bond and ionic interactions, but decreased hydrophobic interactions between the IgG without oligosaccharides and with high mannose to protein A and protein G. The difference could have also been perpetuated by the lower pH used to elute the antibodies from protein G. Overall, this result exemplified that subtle conformational changes may cause a dramatic difference in the binding of IgG to protein A and protein G. 5. Conclusions It was demonstrated by the current study that the absence and the presence of different types of the conserved oligosaccharides had a significant effect on the separation of recombinant monoclonal antibodies by protein A and protein G chromatography. Analysis by ELISA, however, did not show any binding
2387
affinity differences at least between antibody molecules with and without oligosaccharides. It was thus concluded that the low pH elution conditions of protein A and protein G chromatography caused differential structural changes of antibody molecules without oligosaccharides and with different types of oligosaccharides. The elution difference may be employed to enrich or remove antibody without oligosaccharides or with high mannose oligosaccharides. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]
[32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47]
A. Wright, S.L. Morrison, J. Immunol. 160 (1998) 3393. A. Wright, S.L. Morrison, J. Exp. Med. 180 (1994) 1087. B. Heyman, M. Nose, W.O. Weigle, J. Immunol. 134 (1985) 4018. M. Nose, H. Wigzell, Proc. Natl. Acad. Sci. U.S.A. 80 (1983) 6632. H. Dorai, B.M. Mueller, R.A. Reisfeld, S.D. Gillies, Hybridoma 10 (1991) 211. E.J. Wawrzynczak, A.J. Cumber, G.D. Parnell, P.T. Jones, G. Winter, Mol. Immunol. 29 (1992) 213. F. Wang, A. Nakouzi, M. Alvarez, O. Zaragoza, R.H. Angeletti, A. Casadevall, Mol. Immunol. 43 (2006) 987. P.N. Boyd, A.C. Lines, A.K. Patel, Mol. Immunol. 32 (1995) 1311. R.J. Rothman, B. Perussia, D. Herlyn, L. Warren, Mol. Immunol. 26 (1989) 1113. R. Ghirlando, J. Lund, M. Goodall, R. Jefferis, Immunol. Lett. 68 (1999) 47. H. Liu, G.G. Bulseco, J. Sun, Immunol. Lett. 106 (2006) 144. Y. Mimura, P. Sondermann, R. Ghirlando, J. Lund, S.P. Young, M. Goodall, R. Jefferis, J. Biol. Chem. 276 (2001) 45539. Y. Mimura, S. Church, R. Ghirlando, P.R. Ashton, S. Dong, M. Goodall, J. Lund, R. Jefferis, Mol. Immunol. 37 (2000) 697. J. Lund, T. Tanaka, N. Takahashi, G. Sarmay, Y. Arata, R. Jefferis, Mol. Immunol. 27 (1990) 1145. H. Matsuda, S. Nakamura, Y. Ichikawa, K. Kozai, R. Takano, M. Nose, S. Endo, Y. Nishimura, Y. Arata, Mol. Immunol. 27 (1990) 571. S. Krapp, Y. Mimura, R. Jefferis, R. Huber, P. Sondermann, J. Mol. Biol. 325 (2003) 979. A.R. Duncan, G. Winter, Nature 332 (1988) 738. P. Sondermann, R. Huber, V. Oosthuizen, U. Jacob, Nature 406 (2000) 267. R.L. Shields, A.K. Namenuk, K. Hong, Y.G. Meng, J. Rae, J. Briggs, D. Xie, J. Lai, A. Stadlen, B. Li, J.A. Fox, L.G. Presta, J. Biol. Chem. 276 (2001) 6591. M.R. Walker, J. Lund, K.M. Thompson, R. Jefferis, Biochem. J. 259 (1989) 347. M.H. Tao, R.I. Smith, S.L. Morrison, J. Exp. Med. 178 (1993) 661. J. Lund, N. Takahashi, J.D. Pound, M. Goodall, H. Nakagawa, R. Jefferis, FASEB J. 9 (1995) 115. M.H. Tao, S.L. Morrison, J. Immunol. 143 (1989) 2595. R.J. Leatherbarrow, T.W. Rademacher, R.A. Dwek, J.M. Woof, A. Clark, D.R. Burton, N. Richardson, A. Feinstein, Mol. Immunol. 22 (1985) 407. J.D. Pound, J. Lund, R. Jefferis, Mol. Immunol. 30 (1993) 233. M.R. Lifely, C. Hale, S. Boyce, M.J. Keen, J. Phillips, Glycobiology 5 (1995) 813. A.H. Horwitz, C.P. Chang, M. Better, K.E. Hellstrom, R.R. Robinson, Proc. Natl. Acad. Sci. U.S.A. 85 (1988) 8678. P. Umana, J. Jean-Mairet, R. Moudry, H. Amstutz, J.E. Bailey, Nat. Biotechnol. 17 (1999) 176. J. Davies, L. Jiang, L.Z. Pan, M.J. LaBarre, D. Anderson, M. Reff, Biotechnol. Bioeng. 74 (2001) 288. R.L. Shields, J. Lai, R. Keck, L.Y. O’Connell, K. Hong, Y.G. Meng, S.H. Weikert, L.G. Presta, J. Biol. Chem. 277 (2002) 26733. T. Shinkawa, K. Nakamura, N. Yamane, E. Shoji-Hosaka, Y. Kanda, M. Sakurada, K. Uchida, H. Anazawa, M. Satoh, M. Yamasaki, N. Hanai, K. Shitara, J. Biol. Chem. 278 (2003) 3466. R. Niwa, S. Hatanaka, E. Shoji-Hosaka, M. Sakurada, Y. Kobayashi, A. Uehara, H. Yokoi, K. Nakamura, K. Shitara, Clin. Cancer Res. 10 (2004) 6248. R. Niwa, A. Natsume, A. Uehara, M. Wakitani, S. Iida, K. Uchida, M. Satoh, K. Shitara, J. Immunol. Methods 306 (2005) 151. R. Niwa, E. Shoji-Hosaka, M. Sakurada, T. Shinkawa, K. Uchida, K. Nakamura, K. Matsushima, R. Ueda, N. Hanai, K. Shitara, Cancer Res. 64 (2004) 2127. A. Okazaki, E. Shoji-Hosaka, K. Nakamura, M. Wakitani, K. Uchida, S. Kakita, K. Tsumoto, I. Kumagai, K. Shitara, J. Mol. Biol. 336 (2004) 1239. A. Natsume, M. Wakitani, N. Yamane-Ohnuki, E. Shoji-Hosaka, R. Niwa, K. Uchida, M. Satoh, K. Shitara, J. Immunol. Methods 306 (2005) 93. J. Deisenhofer, Biochemistry 20 (1981) 2361. G.C. Stone, U. Sjobring, L. Bjorck, J. Sjoquist, C.V. Barber, F.A. Nardella, J. Immunol. 143 (1989) 565. A.E. Sauer-Eriksson, G.J. Kleywegt, M. Uhlen, T.A. Jones, Structure 3 (1995) 265. R.J. Leatherbarrow, R.A. Dwek, FEBS Lett. 164 (1983) 227. L. Bjorck, G. Kronvall, J. Immunol. 133 (1984) 969. B. Akerstrom, T. Brodin, K. Reis, L. Bjorck, J. Immunol. 135 (1985) 2589. M. Kats, P.C. Richberg, D.E. Hughes, Anal. Chem. 69 (1997) 338. Z.I. Kravchuk, A.P. Vlasov, G.V. Lyakhnovich, S.P. Martsev, Biochemistry (Moscow) 59 (1994) 1079. S.P. Martsev, Z.I. Kravchuk, A.P. Vlasov, Immunol. Lett. 43 (1994) 149. S.P. Martsev, Z.I. Kravchuk, A.P. Vlasov, G.V. Lyakhnovich, FEBS Lett. 361 (1995) 173. G. Gaza-Bulseco, H. Liu, Pharm. Res. 25 (2008) 1881.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Effect of the conserved oligosaccharides of recombinant monoclonal antibodies on the separation by protein A and protein G chromatography Georgeen Gaza-Bulseco, Keith Hickman, Sara Sinicropi-Yao, Karen Hurkmans, Chris Chumsae, Hongcheng Liu ∗ Process Sciences Department, Abbott Bioresearch Center, 100 Research Drive, Worcester, MA 01605, USA
a r t i c l e
i n f o
Article history: Received 10 September 2008 Received in revised form 30 December 2008 Accepted 6 January 2009 Available online 15 January 2009 Keywords: Recombinant monoclonal antibodies Protein A Protein G Oligosaccharides Mass spectrometry
a b s t r a c t Glycosylation of the conserved asparagine residue in CH2 domains of IgG molecules is an important post-translational modification. The presence of oligosaccharides is critical for structure, stability and biological function of IgG antibodies. Effect of the glycosylation states of recombinant monoclonal antibodies on protein A and protein G chromatography was evaluated. Antibodies lacking oligosaccharides eluted later from protein A and earlier from protein G columns than antibodies with oligosaccharides using a gradient of decreasing pH. Interestingly, different types of oligosaccharides also affected the elution of the antibodies. Antibodies with high mannose type oligosaccharides were enriched in later eluting fractions from protein A and earlier eluting fractions from protein G. While antibodies with more mature oligosaccharides, such as core fucosylated biantennary complex oligosaccharides with zero (Gal 0), one (Gal 1) or two (Gal 2) terminal galactoses, were enriched in earlier eluting fractions from protein A and in the later eluting fractions from protein G. However, analysis by enzyme-linked immunosorbent assay (ELISA) revealed that antibody binding affinity to protein A and protein G was not affected by the absence or presence of oligosaccharides. It was thus concluded that the elution difference of antibodies with or without oligosaccharides and antibodies with different types of oligosaccharides were due to differential structural changes around the CH2–CH3 domain interface under the low pH conditions used for protein A and protein G chromatography. © 2009 Elsevier B.V. All rights reserved.
1. Introduction IgG molecules contain N-linked oligosaccharides on the conserved asparagine (Asn) residue in the CH2 domains. The absence of oligosaccharides as well as different types of oligosaccharides at this site does not affect antigen binding affinity [1–9], due to the structural separation of the Fab and Fc region by a highly flexible hinge. However, oligosaccharides are critical for the structural integrity, stability and the biological functions of the Fc region. It has been demonstrated that the absence of oligosaccharides altered the structure of the CH2 domains of antibodies and resulted in decreased stability [10–15]. Conformational changes caused by the absence of oligosaccharides are subtle and local to the regions around the glycosylation sites [1,14–16]. However, the consequences were significant on effector functions, since the Fc␥ receptors and the first component of the complement (C1q) binding sites are located in the hinge proximal region of the CH2 domains [17–19], where the structure was significantly affected. Absence of
∗ Corresponding author. Tel.: +1 508 849 2591; fax: +1 508 793 4885. E-mail address: [email protected] (H. Liu). 0021-9673/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2009.01.014
oligosaccharides resulted in antibodies with decreased to complete loss of binding to Fc␥ receptors [4,5,12–15,20–25], which resulted in decreased to complete abolishment of biological functions such as antibody-dependent-cellular-cytotoxicity (ADCC) [4,5,8,14,26]. Absence of oligosaccharides also caused a decrease or loss of binding to C1q [15,17,23,24], which resulted in decreased or complete loss of complement activation and complement dependent cytotoxicity (CDC) [4,5,7,8,13,15,17,23,24]. Antibodies containing different types of oligosaccharides also showed significant differences in structure, stability and biological functions [1,2,8,12,13,22,26–29]. The presence of different types of oligosaccharides also caused a subtle conformational change around the glycosylation sites [1,16]. Antibodies with high mannose type oligosaccharides were deficient in C1q binding, complement consumption and CDC [1,2]. Antibodies that contained oligosaccharides without the core fucose showed a dramatic increase in the binding affinity to FcRIIIa and enhancement of ADCC [30–36]. Protein A and protein G share similar as well as distinct binding sites around the CH2–CH3 domain interface [37–39]. Although, Matsuda et al. [15] reported that human IgG-Fc expressed in E. coli, presumably lacking oligosaccharides, required a slightly lower pH for elution from a protein A column compared to Fc with oligosac-
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
charides, it is generally agreed that binding of IgG antibodies to protein A was not affected by the absence or presence of different types of oligosaccharides [3,4,8,15,23,40]. Protein G can bind more IgG subtypes and with stronger binding affinities [41,42]. The effect of oligosaccharides on the binding of IgG to protein G was unclear. Protein A and protein G are two important ligands of IgGFc and are commonly used for affinity chromatography to purify antibodies. To better understand the role of oligosaccharides, experiments were carried out in this study to investigate the effect of the absence of oligosaccharides and the presence of different types of oligosaccharides on the binding of antibodies to protein A and protein G. Two recombinant monoclonal antibodies, one fully human (MAb-A) and one humanized (MAb-B) were used in this study. MAb-A contains core fucosylated biantennary complex structures with zero (Gal 0), one (Gal 1) or two (Gal 2) terminal galactose residues. MAb-A also contains low levels of high mannose structures with five (M5), six (M6) or seven (M7) mannose residues. MAb-B contains approximately 30% non-glycosylated heavy chains, in addition to the same oligosaccharide structures as MAb-A.
2. Materials and methods 2.1. Materials MAb-A, a recombinant fully human monoclonal IgG1 antibody was produced by a transfected Chinese hamster ovary (CHO) cell line and purified by multiple chromatography steps (Abbott Bioresearch Center, Worcester, MA, USA). MAb-B, a recombinant humanized monoclonal IgG1 antibody was expressed by transient transfection of 293 cells and purified by protein A chromatography (Abbott Bioresearch Center). Dithiothreitol (DTT) was purchased from Sigma (St. Louis, MO, USA). Formic acid (FA) was purchased from EMD (Gibbbstown, NJ, USA). Lys-C, phenylmethylsulfonyl fluoride (PMSF) and N-octylglucoside were purchased from Roche (Indianapolis, IN, USA). Acetonitrile was purchased from J.T. Baker (Phillipsburg, NJ, USA) and peptide-N-glycosidase F (PNGaseF) was purchased from Prozyme (San Leandro, CA, USA).
2383
2.2. Deglycosylation of the recombinant monoclonal antibody MAb-A at approximately 70 mg/mL in formulation buffer (5.57 mM sodium phosphate monobasic, 8.69 mM sodium phosphate dibasic, 106.69 mM sodium chloride, 1.07 mM sodium citrate, 6.45 mM citric acid, 66.68 mM mannitol and 0.1% Tween) at pH, 5.2 was diluted to 10 mg/mL with water deionized by a Milli-Q system (Millipore, Bedford, MA, USA). N-Octylglucoside at a final concentration of 1% was included in the sample preparation to facilitate deglycosylation. PNGaseF (2.5 mU/L) was added to the diluted sample at the ratio of 1 L enzyme to 500 g antibody. A sample prepared the same as the sample for PNGaseF digestion but without the addition of PNGaseF was used as a control. The samples were incubated at 37 ◦ C for 18 h. 2.3. Protein A chromatography A Shimadzu HPLC and a Poros A column (50 mm × 4.6 mm, Applied Biosystems, Framingham, MA, USA) were used for sample analysis and fraction collection. One hundred L of each sample (10 mg/mL in PBS) was loaded at 100% mobile phase A [phosphate buffered saline (PBS), pH 7.4]. The column was washed using 100% mobile phase A for 10 min, and then eluted with an increase of mobile phase B (0.1 M acetic acid and 0.15 M sodium chloride, pH 2.9) from 0% to 60% in 30 min. The column was washed using 100% mobile phase B for 10 min and then equilibrated using 100% mobile phase A for another 10 min. The flow-rate was set at 2 mL/min. For fraction collection, five mg of either MAb-A or MAb-B was injected and analyzed by following the same procedure. Fractions were collected across the entire elution peak. Four mL of each fraction was collected in a 2-min elution time window. 2.4. Protein G chromatography A Shimadzu HPLC and a Poros G column (50 mm × 4.6 mm, Applied Biosystems) were used for sample analysis and fraction collection. One hundred L of each sample (10 mg/mL in PBS) was loaded at 80% mobile phase A and 20% mobile phase B with the pH adjusted to 2.5. The column was washed using 80% mobile phase A and 20% mobile phase B for 10 min, and then eluted with an increase
Fig. 1. Mass spectra of the reduced Fc of antibody MAb-A. The data was acquired from MAb-A with oligosaccharides (A) and from MAb-A after removal of the oligosaccharides using PNGaseF (B). Peak identities are labeled in the spectra.
2384
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
of mobile phase B from 20% to 100% in 40 min. The column was washed using 100% mobile phase B for 10 min and then equilibrated using 80% mobile phase A and 20% mobile phase B for an additional 10 min. The flow-rate was set at 2 mL/min. For fraction collection, five mg of MAb-A or MAb-B was injected and analyzed following the same procedure as described previously. Fractions were collected across the entire elution peak. Four mL of each fraction was collected in a 2-min elution time window. 2.5. Lys-C digestion and LC–MS analysis MAb-A, MAb-B and Protein A and G chromatography fractions collected from both antibodies were diluted to 0.5 mg/mL with PBS to a final volume of 50 L. Lys-C was added to each fraction at a final ratio of 1:200 (w/w) Lys-C: antibody, and digestion was carried out at 37 ◦ C for 30 min. PMSF dissolved in methanol was added to each sample at a final concentration of 2 mM to stop the Lys-C digestion. The samples were then reduced with 20 mM DTT at 37 ◦ C for 5 min. An Agilent HPLC (Santa Clara, CA) equipped with a protein C4 column (Vydac, 150 mm × 1 mm I.D., 5 m particle size, 300 A pore size) and a Qstar mass spectrometer (Applied Biosystems) were used to analyze the Lys-C digested samples. Ten L of each sample was loaded separately at 95% mobile phase A (0.08% FA in water deionized by a Milli-Q system (Millipore)) and 5% mobile phase B (0.08% FA in acetonitrile). After 5 min at 5% mobile phase B, proteins were eluted off the column by increasing mobile phase B to 65% over a period of 35 min. The column was washed with 95% mobile phase B for 5 min and then re-equilibrated at 5% mobile phase B for 10 min before the next injection. The flow-rate was set at 50 L/min. The column oven was set at 60 ◦ C. The mass spectrometer was operated at positive mode with a scan range of m/z 800–2500. IonSpray voltage was set at 4500 V, and the source temperature was set at 350 ◦ C. 2.6. ELISA The binding affinity of the antibody with and without oligosaccharides to protein A and protein G was compared by ELISA. The native and the deglycosylated antibody (20 g/mL) in 50 mM sodium bicarbonate buffer, pH 9.4, were diluted serially with a 2 fold dilution of each step. One hundred L of the serially diluted samples were immobilized onto three 96-well plates by incubating the plates overnight at 4 ◦ C with shaking. After removal of the unbound antibodies, the plates were blocked with SuperBlock blocking buffer (Pierce, Rockford, IL, USA) (300 L/well) for 1 h at 37 ◦ C with shaking. The plates were then washed 5 times with PBS with 0.1% Triton X-100 using a plate washer (Tecan, Zurich, Switzerland). The plates were incubated for 1 h at 37 ◦ C with 100 L/well
Fig. 2. Protein A and protein G chromatograms of MAb-A. MAb-A with oligosaccharides (Native) and after the removal of oligosaccharides (deglycosylated) were analyzed by protein A (A) and protein G (B) chromatography. Arrows indicate the points at which fractions were collected (mass spectra in Figs 4 and 5).
protein A-horse radish peroxidase (HRP) conjugate (Millipore, Billerica, MA), protein G-HRP or goat-anti-human IgG-HRP (Jackson ImmunoResearch, West Grove, PA, USA) that was diluted 50 000 times in SuperBlock. After one wash with PBS, 100 L/well of tetramethylbenzidine (TMB) substrate, K Blue (Neogen, Lansing, MI, USA), was added then incubated at room temperature for 5 min. The reaction was stopped by the addition of 100 L/well of 1 N phosphoric acid. The plates were then read at 450 nm using a Spectramax Plus plate reader (Molecular Devices, Sunnyvale, CA, USA). The data was fitted using Soft-max Pro Version 4.8 (Molecular Devices). 3. Results 3.1. Deglycosylation of MAb-A MAb-A contains two major post-translational modifications, glycosylation of the conserved Asn residue in the CH2 domains and incomplete processing of the C-terminal lysine (Lys) residues. To determine the effect of oligosaccharides on the binding of this antibody to protein A and protein G, oligosaccharides attached to MAb-A were removed by digestion with PNGaseF. The Lys-C digested and reduced MAb before and after PNGaseF digestion
Fig. 3. Mass spectrum of the reduced Fc of antibody MAb-B. Peak identities are labeled in the spectrum.
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
2385
Fig. 4. Mass spectra of representative protein A (left) and protein G (right) fractions of MAb-B from the early (A), middle (B) and later (C) sections of the chromatogram. Peak identities are labeled in the spectra.
was analyzed by LC–MS to confirm the glycosylation state of the samples. Lys-C cleaves the peptide bond between lysine and threonine in the hinge region to generate Fab and Fc. Three peaks were observed in the mass spectrum of the reduced Fc with molecular weights of 26617 Da, 26746 Da and 26780 Da, which corresponded to Fc with Gal 0 and without a C-terminal Lys, Gal 0 with a C-terminal Lys, and Gal 1 without a C-terminal Lys respectively (Fig. 1A). Only two peaks were observed after deglycosylation, with molecular weights of 25296 Da and 25171 Da, which corre-
sponded to the Fc with and without a C-terminal Lys respectively (Fig. 1B). 3.2. Protein A and protein G chromatography of MAb-A MAb-A with and without oligosaccharides was analyzed by protein A and protein G chromatography using a gradient of decreasing pH for elution. MAb-A with oligosaccharides eluted off the protein A column slightly earlier than the antibody without oligosaccharides
Fig. 5. Mass spectra of representative protein A (left) and protein G (right) fractions of MAb-A from the early (A), middle (B) and later (C) sections of the chromatogram. Peak identities are labeled in the spectra.
2386
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
(Fig. 2A). Interestingly, when the same samples were analyzed by protein G chromatography, MAb-A without oligosaccharides eluted off the column earlier than the antibody with oligosaccharides (Fig. 2B).
3.3. Protein A and protein G chromatography and mass spectrometry analysis of MAb-B Removal of oligosaccharides from MAb-A by PNGaseF digestion converted the original conserved Asn residue, to which the oligosaccharides were attached, to an aspartate (Asp) residue. To exclude the potential effect of this amino acid change on binding to protein A and protein G, a monoclonal antibody, MAb-B, that contained the original conserved Asn due to approximately 30% of the antibody being aglycosylated, was analyzed by protein A and protein G chromatography following the same procedure used for MAb-A. MAb-B also contained 70% glycosylated heavy chains with mainly Gal 0, Gal 1 and trace levels of Gal 2 and high mannose. As shown in Fig. 3, three peaks were observed in the mass spectrum of the reduced Fc of MAb-B with molecular weights of 25 119 Da, 26 564 Da and 26 727 Da, which corresponded to the reduced Fc without oligosaccharides, with Gal 0, and Gal 1, respectively. Heavy chain with Gal 2 was not detected. Protein A and protein G chromatography fractions of MAbB were collected across the entire elution peak. Mass spectra from representative fractions of the early, middle and late eluting sections of the elution peak from protein A and protein G chromatograms are shown in Fig. 4. There was a higher percentage of aglycosylated Fc that eluted in the late fractions from protein A and in the early fractions from protein G columns. This observation was in agreement with the results obtained with MAb-A. It was also interesting to note that there were slightly higher levels of Gal 0 than Gal 1 in the middle fraction and the appearance of Gal 2 in the late fraction from protein G.
3.4. LC–MS analysis of protein A and protein G fractions from MAb-A To investigate whether different types of oligosaccharides affected the elution from protein A and protein G, fractions from the elution peak of MAb-A from protein A and protein G chromatography were analyzed by LC–MS. Mass spectra of Fc fragments of representative fractions from the early, middle and late sections of the elution peaks from protein A and protein G chromatography are shown in Fig. 5. MAb-A with oligomannose eluted in the late fractions from protein A and in the early fractions from protein G columns.
3.5. ELISA An ELISA assay was used to compare the binding affinity of native and deglycosylated MAb-A to protein A and protein G. To assure that similar levels of the native and deglycosylated MAb-A were immobilized on the ELISA plates, the antibodies were probed with a goat-anti-human IgG-HRP. The dose response curves of the native and the deglycosylated MAb-A were similar, which confirmed that similar levels of the native and deglycosylated MAb-A were immobilized on the plates (Fig. 6C). The dose response curves of the native antibody and the deglycosylated antibody to either protein A-HRP or protein G-HRP were also similar, which indicated that antibody with and without oligosaccharides bound to protein A or protein G with similar affinity (Fig. 6A and B).
Fig. 6. Binding ELISA of the antibody to protein A and protein G. Native (diamond) and deglycosylated antibody (square) were immobilized and then detected with either protein A-HRP (A), protein G-HRP (B) or anti-human IgG-HRP (C).
4. Discussion The N-linked oligosaccharides of IgG are attached to the conserved Asn residue located in the hinge proximal region of the CH2 domain. The oligosaccharides at this site are mainly of the complex biantennary type, with low levels of oligomannose. The major binding sites of protein A and protein G are located in the CH2–CH3 interface and there are no known direct interactions with the Nlinked oligosaccharides [37,39]. The effect of oligosaccharides on the binding to protein A and protein G and elution of IgG from protein A as well as protein G resins were thus further investigated in the current study. Antibodies without oligosaccharides eluted later from protein A and earlier from protein G columns than native glycosylated antibodies. The elution differences were observed using antibodies without oligosaccharides containing either Asp or Asn. In contrast, when the binding of MAb-A with and without oligosaccharides to protein A and protein G was compared by ELISA, no differences were observed. Such a disagreement has been reported by Matsuda et al. [15] where they did not detect any binding difference between human IgG-Fc with and without oligosaccharides to protein A by ELISA, but they showed that human IgG-Fc without oligosaccharides eluted later than with oligosaccharides from protein A chromatography. On the other hand, Leatherbarrow and Dwek [40] showed that the elution profile of a mouse IgG with or without oligosaccharides from a protein A column was not different when a pH gradient was used. Since there are no direct interactions between the N-linked oligosaccharides and protein A and protein G [37,39], the absence or presence of oligosaccharides may not affect the binding directly. It is known that removal of oligosaccharides [11–13,15], as well as the presence of different types of oligosaccharides [12,13,16] affected antibody structure. These structural differences were local to the glycosylation site and may not be sig-
G. Gaza-Bulseco et al. / J. Chromatogr. A 1216 (2009) 2382–2387
nificant enough to affect binding of the residues in the CH2–CH3 interface to protein A or protein G. This is in agreement with many studies that showed no difference in the binding of different IgG molecules and their Fc fragments to protein A in the presence or absence of oligosaccharides [3,4,8,15,23,40]. The only explanation for the differences in retention properties could be attributed to the effects of pH on the structure of the antibody with and without oligosaccharides. The structural changes in turn could have led to differences in the elution profiles of the antibodies with and without oligosaccharides from protein A and protein G columns. Dependence of the structure and stability of IgG molecules on pH has been studied extensively. For example, Kats et al. [43] reported that a monoclonal chimeric antibody showed a pH-dependent isoform transition. Acidic treatment of rabbit IgG resulted in increased stability due to an increased interaction between the CH1 and CH2 domains [44–46]. A previous study in our laboratory also showed that oligosaccharides slowed down the fragmentation rate of a recombinant monoclonal antibody at low pH (<4), probably due to a structural rearrangement[47]. It was thus proposed that antibodies with and without oligosaccharides adopted different structures under the low pH condition used for elution from protein A and protein G chromatography. Therefore, different structural changes of the antibodies without oligosaccharides at the pH conditions used for elution resulted in a stronger association or slower dissociation from protein G and a weaker association or faster dissociation from protein A. Interestingly, antibodies with high mannose oligosaccharides also eluted later from protein A and earlier from protein G columns compared to antibodies with more mature oligosaccharides such as Gal 0 and Gal 1. This suggests that antibodies with high mannose adopt a structure similar to antibodies without oligosaccharides and was affected similarly by the low pH condition. It was also evidenced that antibody with high mannose was slightly enriched in the late fraction from protein A column and highly enriched in the early fraction from protein G column (Fig. 5), which indicated that a more significant structural difference between antibodies with high mannose and antibody with complex oligosaccharides was observed under the slightly lower pH condition used for protein G chromatography. It was not clear if terminal galactose affected the elution of antibodies from protein A and protein G. The effect may be subtle and therefore varied from antibody to antibody significantly. It is interesting that the elution profiles of the antibodies without oligosaccharides or with high mannose type oligosaccharides compared to antibodies with more mature oligosaccharides were in opposite elution order for protein A and protein G. This result was unexpected especially since protein A and protein G bind to similar IgG residues within the CH2–CH3 domain [39]. The binding of IgG to protein A involves mainly hydrophobic interactions, while the binding to protein G involves mainly hydrogen bonding and ionic interactions. It was possible that structural changes caused by the elution pH enhanced the hydrogen bond and ionic interactions, but decreased hydrophobic interactions between the IgG without oligosaccharides and with high mannose to protein A and protein G. The difference could have also been perpetuated by the lower pH used to elute the antibodies from protein G. Overall, this result exemplified that subtle conformational changes may cause a dramatic difference in the binding of IgG to protein A and protein G. 5. Conclusions It was demonstrated by the current study that the absence and the presence of different types of the conserved oligosaccharides had a significant effect on the separation of recombinant monoclonal antibodies by protein A and protein G chromatography. Analysis by ELISA, however, did not show any binding
2387
affinity differences at least between antibody molecules with and without oligosaccharides. It was thus concluded that the low pH elution conditions of protein A and protein G chromatography caused differential structural changes of antibody molecules without oligosaccharides and with different types of oligosaccharides. The elution difference may be employed to enrich or remove antibody without oligosaccharides or with high mannose oligosaccharides. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]
[32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47]
A. Wright, S.L. Morrison, J. Immunol. 160 (1998) 3393. A. Wright, S.L. Morrison, J. Exp. Med. 180 (1994) 1087. B. Heyman, M. Nose, W.O. Weigle, J. Immunol. 134 (1985) 4018. M. Nose, H. Wigzell, Proc. Natl. Acad. Sci. U.S.A. 80 (1983) 6632. H. Dorai, B.M. Mueller, R.A. Reisfeld, S.D. Gillies, Hybridoma 10 (1991) 211. E.J. Wawrzynczak, A.J. Cumber, G.D. Parnell, P.T. Jones, G. Winter, Mol. Immunol. 29 (1992) 213. F. Wang, A. Nakouzi, M. Alvarez, O. Zaragoza, R.H. Angeletti, A. Casadevall, Mol. Immunol. 43 (2006) 987. P.N. Boyd, A.C. Lines, A.K. Patel, Mol. Immunol. 32 (1995) 1311. R.J. Rothman, B. Perussia, D. Herlyn, L. Warren, Mol. Immunol. 26 (1989) 1113. R. Ghirlando, J. Lund, M. Goodall, R. Jefferis, Immunol. Lett. 68 (1999) 47. H. Liu, G.G. Bulseco, J. Sun, Immunol. Lett. 106 (2006) 144. Y. Mimura, P. Sondermann, R. Ghirlando, J. Lund, S.P. Young, M. Goodall, R. Jefferis, J. Biol. Chem. 276 (2001) 45539. Y. Mimura, S. Church, R. Ghirlando, P.R. Ashton, S. Dong, M. Goodall, J. Lund, R. Jefferis, Mol. Immunol. 37 (2000) 697. J. Lund, T. Tanaka, N. Takahashi, G. Sarmay, Y. Arata, R. Jefferis, Mol. Immunol. 27 (1990) 1145. H. Matsuda, S. Nakamura, Y. Ichikawa, K. Kozai, R. Takano, M. Nose, S. Endo, Y. Nishimura, Y. Arata, Mol. Immunol. 27 (1990) 571. S. Krapp, Y. Mimura, R. Jefferis, R. Huber, P. Sondermann, J. Mol. Biol. 325 (2003) 979. A.R. Duncan, G. Winter, Nature 332 (1988) 738. P. Sondermann, R. Huber, V. Oosthuizen, U. Jacob, Nature 406 (2000) 267. R.L. Shields, A.K. Namenuk, K. Hong, Y.G. Meng, J. Rae, J. Briggs, D. Xie, J. Lai, A. Stadlen, B. Li, J.A. Fox, L.G. Presta, J. Biol. Chem. 276 (2001) 6591. M.R. Walker, J. Lund, K.M. Thompson, R. Jefferis, Biochem. J. 259 (1989) 347. M.H. Tao, R.I. Smith, S.L. Morrison, J. Exp. Med. 178 (1993) 661. J. Lund, N. Takahashi, J.D. Pound, M. Goodall, H. Nakagawa, R. Jefferis, FASEB J. 9 (1995) 115. M.H. Tao, S.L. Morrison, J. Immunol. 143 (1989) 2595. R.J. Leatherbarrow, T.W. Rademacher, R.A. Dwek, J.M. Woof, A. Clark, D.R. Burton, N. Richardson, A. Feinstein, Mol. Immunol. 22 (1985) 407. J.D. Pound, J. Lund, R. Jefferis, Mol. Immunol. 30 (1993) 233. M.R. Lifely, C. Hale, S. Boyce, M.J. Keen, J. Phillips, Glycobiology 5 (1995) 813. A.H. Horwitz, C.P. Chang, M. Better, K.E. Hellstrom, R.R. Robinson, Proc. Natl. Acad. Sci. U.S.A. 85 (1988) 8678. P. Umana, J. Jean-Mairet, R. Moudry, H. Amstutz, J.E. Bailey, Nat. Biotechnol. 17 (1999) 176. J. Davies, L. Jiang, L.Z. Pan, M.J. LaBarre, D. Anderson, M. Reff, Biotechnol. Bioeng. 74 (2001) 288. R.L. Shields, J. Lai, R. Keck, L.Y. O’Connell, K. Hong, Y.G. Meng, S.H. Weikert, L.G. Presta, J. Biol. Chem. 277 (2002) 26733. T. Shinkawa, K. Nakamura, N. Yamane, E. Shoji-Hosaka, Y. Kanda, M. Sakurada, K. Uchida, H. Anazawa, M. Satoh, M. Yamasaki, N. Hanai, K. Shitara, J. Biol. Chem. 278 (2003) 3466. R. Niwa, S. Hatanaka, E. Shoji-Hosaka, M. Sakurada, Y. Kobayashi, A. Uehara, H. Yokoi, K. Nakamura, K. Shitara, Clin. Cancer Res. 10 (2004) 6248. R. Niwa, A. Natsume, A. Uehara, M. Wakitani, S. Iida, K. Uchida, M. Satoh, K. Shitara, J. Immunol. Methods 306 (2005) 151. R. Niwa, E. Shoji-Hosaka, M. Sakurada, T. Shinkawa, K. Uchida, K. Nakamura, K. Matsushima, R. Ueda, N. Hanai, K. Shitara, Cancer Res. 64 (2004) 2127. A. Okazaki, E. Shoji-Hosaka, K. Nakamura, M. Wakitani, K. Uchida, S. Kakita, K. Tsumoto, I. Kumagai, K. Shitara, J. Mol. Biol. 336 (2004) 1239. A. Natsume, M. Wakitani, N. Yamane-Ohnuki, E. Shoji-Hosaka, R. Niwa, K. Uchida, M. Satoh, K. Shitara, J. Immunol. Methods 306 (2005) 93. J. Deisenhofer, Biochemistry 20 (1981) 2361. G.C. Stone, U. Sjobring, L. Bjorck, J. Sjoquist, C.V. Barber, F.A. Nardella, J. Immunol. 143 (1989) 565. A.E. Sauer-Eriksson, G.J. Kleywegt, M. Uhlen, T.A. Jones, Structure 3 (1995) 265. R.J. Leatherbarrow, R.A. Dwek, FEBS Lett. 164 (1983) 227. L. Bjorck, G. Kronvall, J. Immunol. 133 (1984) 969. B. Akerstrom, T. Brodin, K. Reis, L. Bjorck, J. Immunol. 135 (1985) 2589. M. Kats, P.C. Richberg, D.E. Hughes, Anal. Chem. 69 (1997) 338. Z.I. Kravchuk, A.P. Vlasov, G.V. Lyakhnovich, S.P. Martsev, Biochemistry (Moscow) 59 (1994) 1079. S.P. Martsev, Z.I. Kravchuk, A.P. Vlasov, Immunol. Lett. 43 (1994) 149. S.P. Martsev, Z.I. Kravchuk, A.P. Vlasov, G.V. Lyakhnovich, FEBS Lett. 361 (1995) 173. G. Gaza-Bulseco, H. Liu, Pharm. Res. 25 (2008) 1881.