Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The use of lectin microarray for assessing glycosylation of therapeutic proteins

The use of lectin microarray for assessing glycosylation of therapeutic proteins MABS 2016, VOL. 8, NO. 3, 524–535 http://dx.doi.org/10.1080/19420862.2016.1149662 REPORT Lei Zhang, Shen Luo, and Baolin Zhang Office of Biotechnology Products, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA ABSTRACT ARTICLE HISTORY Received 4 November 2015 Glycans or carbohydrates attached to therapeutic glycoproteins can directly affect product quality, safety Revised 20 January 2016 and efficacy, and therefore must be adequately analyzed and controlled throughout product life cycles. Accepted 28 January 2016 However, the complexity of protein glycosylation poses a daunting analytical challenge. In this study, we evaluated the utility of a lectin microarray for assessing protein glycans. Using commercial lectin chips, KEYWORDS which contain 45 lectins toward distinct glycan structures, we were able to determine the lectin binding Glycan analysis; lectin patterns of a panel of 15 therapeutic proteins, including 8 monoclonal antibodies. Lectin binding signals microarray; monoclonal were analyzed to generate glycan profiles that were generally consistent with the known glycan patterns antibodies; therapeutic glycoproteins for these glycoproteins. In particular, the lectin-based microarray was found to be highly sensitive to variations in the terminal carbohydrate structures such as galactose versus sialic acid epitopes. These data suggest that lectin microarray could be used for screening glycan patterns of therapeutic glycoproteins. Introduction Glycans attached to a therapeutic protein can directly affect Glycosylation of proteins is a complex post-translational modi- product quality, safety and efficacy. It is well documented that fication that attaches carbohydrates or named glycans at spe- glycans attached to a protein affect protein solubility and stabil- 3-5 3,6-8 cific sites on a protein backbone, most commonly at Asn ity, pharmacokinetics/pharmacodynamics (PK/PD), and 2,9 (N-linked) or Ser/Thr (O-linked) residues. The N-linked gly- immunogenicity. In the latter, non-human glycans attached cosylation occurs at the consensus sequence of Asn-X-Ser/Thr onto a therapeutic protein such as Neu5Gc and terminal a-Gal (where X is any amino acid except proline), whereas O-linked epitopes could cause immunogenic responses. For many mono- glycans are usually attached to Ser or Thr residues. Both N- clonal antibodies (mAbs), proper glycosylation of the crystal- and O-glycosylation involve a series of enzymatic reactions cat- lizable fragment (Fc) is essential to IgG antibody effector alyzed by glycan-processing enzymes, which are responsible for functions. Therefore, glycan moieties of therapeutic proteins trimming and modifications of glycan epitopes, resulting in must be adequately characterized and controlled throughout diverse N-glycan structures (e.g., high-mannose, complex, and product life cycle. The commonly used methods include high- hybrid glycans) and O-glycan variants containing up to 8 O- performance liquid chromatography (HPLC), high-perfor- GalNAc glycan core structures. To add complexity, protein gly- mance anion-exchange chromatography with pulsed ampero- cosylation is influenced by the type of host cells and fluctua- metric detection (HPAEC-PAD), mass spectrometry (MS) and 11-16 tions in fermentation conditions (e.g., media, pH, temperature, capillary electrophoresis (CE), which provide information agitation). For instance, therapeutic glycoproteins produced on glycosylation sites, site occupancy, and contents of glycan by mammalian cells such as Chinese hamster ovary (CHO) variants attached to glycoproteins. cells usually contain human-like glycans. By contrast, proteins There is growing interest in the development of high expressed by yeast strains usually contain high levels of man- throughput platforms for assessing protein glycan profiles. Lec- nose (up to 100 units). Other hosts including engineered plant tins are glycan binding proteins (GBPs) that selectively recog- cells and genetically modified animals may produce proteins nize glycan epitopes of free carbohydrates or glycoproteins. with non-human glycan variants such as xylose, N-glycolyl- Lectin-based microarrays have been used to analyze glycan pro- 18-23 neuraminic acid (Neu5Gc) or terminal a-galactose (a-Gal), files of purified glycoproteins or cell surface proteins. In which are known to be immunogenic. As a result, a glycopro- this study, we evaluated the potential utility of a lectin microar- tein produced by living cell systems usually contains a mixture ray for characterization of therapeutic glycoproteins. Using of different glycoforms. These protein variants share an identi- commercial lectin chips containing 45 distinct lectins, we tested cal peptide backbone, but may differ in glycosylation properties a panel of 15 therapeutic proteins for their glycan profiles. Our such as glycosylation site, glycan structure and content. data show that the lectin microarray is robust in generating CONTACT Baolin Zhang baolin.zhang@fda.hhs.gov Disclaimer: The comments in this paper reflect the views of the author and should not be construed to represent the Food and Drug Administration (FDA)’s views or policies. Supplemental data for this article can be accessed on the publisher’s website. Published with license by Taylor & Francis Group, LLC. This article not subject to US copyright law. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits unre- stricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the named author(s) have been asserted. MABS 525 glycan profiles that are generally consistent with the known gly- showing that overnight incubation is required to obtain optimal can characteristics of an individual glycoprotein. binding between a Cy3-labeled glycoprotein and the lectins printed on chips (Supplement II). In the following experiments, a fixed concentration of protein samples (125 ng/mL) was applied to lectin chips and incubated overnight then lectin Results binding signals were detected. Bound glycoprotein signals were The utility of a lectin microarray in determining glycan determined using evanescent-field fluorescence scanner, which profiles of therapeutic mAbs allows a direct measure of glycans bound onto lectins without the need of washing procedures to remove unbound species We assessed the utility of a lectin microarray in profiling glycan (Fig. 1B). Using lectin chips from different batches, we deter- variants of therapeutic proteins using commercial lectin micro- mined the assay reproducibility to be < 10% CV for most lec- chips printed with 45 distinct lectin proteins (Fig. 1A). These tin-glycan binding signals of the samples tested. include lectins that selectively bind core-fucose, sialic acids, N- We first tested a panel of 6 therapeutic mAbs, including bev- acetyl-D-lactosamine (Galb1-4GlcNAc), mannose, or N-acetyl- acizumab (Avastin ), trastuzumab (Herceptin ), adalimumab glucosamine (GlcNAc) oligomers, respectively (Table 1). (Humira ), infliximab (Remicade ), rituximab (Rituxan ) and These glycan structures are commonly found in recombinant omalizumab (Xolair ). These products are known to be IgG1 glycoproteins. We tested a panel of 15 distinct proteins, includ- 24-26 isotypes with glycosylation exclusively occurring in the Fc. ing 8 therapeutic mAbs, one Fc-fusion protein, 4 recombinant All these IgG1 mAbs showed similar lectin binding patterns therapeutic cytokines and enzymes, and 2 different versions of (Fig. 2A & B) in which strong binding signals were detected at human transferrin proteins (Table 2). To facilitate microarray lectins with binding selectivity to core fucose (PSA, LCA, AOL, analysis, protein samples were fluorescently labeled with Cy3 AAL), galactose (RCA120, PHAE), mannose (NPA, ConA and followed by incubation onto lectin-coated chips. We first per- GNA), and GlcNAc oligomer (LEL, STL, UDA). No binding formed dose-titrations and identified the optimal concentra- signals were detected at lectins for sialic acids (MAL_I, SNA, tions of Cy3-labeled protein sample that fell within the linear SSA, TJA-I) or tri/tetra-antennary complex-type N-glycan response ranges for most lectin spots (Supplement I). We also (PHAL). Based on the known selectivity of lectin-glycan tested the effect of incubation time on lectin binding signals, Figure 1. Schematic view of lectin microarray. (A) Lectin microchips used in this study consist of 45 distinct lectins that selectively bind structural variants of carbohy- drates attached onto a protein. Each lectin is printed in triplicate. The lectin-printing layout of lectin chips was provided by the vendor (GlycoTechnica). (B) Protein sam- ples are labeled with a fluorescent dye (e.g., Cy3) and then applied onto the lectin chips. The binding signals at each lectin spots are measured using an evanescent-field fluorescence scanner, detecting the presence or absence of glycan variants in the testing sample based on the known selectivity of lectins toward particular glycan structures. 526 L. ZHANG ET AL. Table 1. Reported glycan selectivity of the 45 lectins used in the microarray assay.* Lectin No. Lectin (origin) Reported glycan selectivity 1 LTL (Lotus tetragonolobus) Fuca1-3(Galb1-4)GlcNAc (Lewis x), Fuca1-2Galb1-4GlcNAc (H-type 2) 2 PSA (Pisum sativum) Fuca1-6GlcNAc (Core Fuc) , a-Man 3 LCA (Lens culinaris) Fuca1-6GlcNAc (Core Fuc), a-Man 4 UEA-I (Ulex europaeus) Fuca1-2Galb1-4GlcNAc (H-type 2) 5 AOL (Aspergillus oryzae) Fuca1-6GlcNAc (Core Fuc), Fuca1-2Galb1-4GlcNAc (H-type 2) 6 AAL (Aleuria aurantia) Fuca1-3(Galb1-4)GlcNAc (Lewis x), Fuca1-6GlcNAc (Core Fuc) 7 MAL_I (Maackia amurensis) Siaa2-3Galb1-4GlcNAc 8 SNA (Sambucus nigra) Siaa2-6Gal/GalNAc 9 SSA (Sambucus sieboldiana) Siaa2-6Gal/GalNAc 10 TJA-I (Trichosanthes japonica) Siaa2-6Gal/GalNAc, HSO3(-) -6Galb1-4GlcNAc 11 PHAL (Phaseolus vulgaris) tri/tetra-antennary complex-type N-glycan 12 ECA (Erythrina cristagalli) Galb1-4GlcNAc (up with increasing the number of terminal Gal), no affinity for fully sialylated N-type, fully agalactosylated N-type 13 RCA120 (Ricinus communis) Galb1-4GlcNAc (up with increasing the number of terminal Gal), Galb1- 3Gal (weak), no affinity for agalactosylated N-type 14 PHAE (Phaseolus vulgaris) bi-antennary complex-type N-glycan with outer Gal and bisecting GlcNAc, no affinity for fully sialylated N-type 15 DSA (Datura stramonium) (GlcNAcb1-4)n, tri/tetra-antennary N-glycan 16 GSL-II (Griffonia simplicifolia) agalactosylated tri/tetra antennary glycans, GlcNAc, no affinity for fully galactosylated or sialylated N-type 17 NPA (Narcissus pseudonarcissus) High-Mannose including Mana1-6Man 18 ConA (Canavalia ensiformis) High-Mannose including Mana1-6(Mana1-3)Man 19 GNA (Galanthus nivalis) High-Mannose including Mana1-3Man 20 HHL (Hippeastrum hybrid) High-Mannose including Mana1-3Man or Mana1-6Man 21 ACG (mushroom, Agrocybe cylindracea) Galb1-3Gal, Siaa2-3Galb1-4GlcNAc 22 TxLCI (Tulipa gesneriana) Mana1-3(Mana1-6)Man, bi/tri-antennary complex-type N-glycan, GalNAc 23 BPL (Bauhinia purpurea) Galb1-3GalNAc (up with Lewis x, down with Core Fuc), GalNAc 24 TJA-II (Tanthes japonica) Fuca1-2Galb1-> or GalNAcb1-> groups at their non-reducing terminals 25 EEL (Euonymus europaeus) Gala1-3Galb1-4GlcNAc, Fuca1-2Galb1-3GlcNAc (H antigen) 26 ABA (fungus, Agaricus bisporus) Galb1-3GalNAc, GlcNAc 27 LEL (tomato, Lycopersicon esculentum) (GlcNAcb1-4)n (Chitin), (Galb1-4GlcNAc)n (polylactosamine) 28 STL (potato, Solanum tuberosum) (GlcNAcb1-4)n (Chitin), oligosaccharide containing GlcNAc and MurNAc 29 UDA (Urtica dioica) GlcNAcb1-4GlcNAc (Chitin), High-Mannose (3 to High, up with increasing the number of Man) 30 PWM (pokeweed, Phytolacca Americana) (GlcNAcb1-4)n (Chitin) 31 Jacalin (Artocarpus integrifolia) GlcNAcb1-3GalNAc (Core3), Siaa2-3Galb1-3GalNAc (sialyl T), Galb1- 3GalNAc (T-antigen), a-GalNAc (Tn-antigen) 32 PNA (peanut, Arachis hypogaea) Galb1-3GalNAc 33 WFA (Wisteria floribunda) GalNAcb1-4GlcNAc (LacdiNAc), Galb1-3(-6)GalNAc 34 ACA (Amaranthus caudatus) Galb1-3GalNAc (T-antigen), Siaa2-3Galb1-3GalNAc (sialyl T) 35 MPA (Maclura pomifera) a-GalNAc (Tn-antigen), Galb1-3GalNAc (T-antigen), 36 HPA (snail, Helix pomatia) a-GalNAc 37 VVA (Vicia villosa) GalNAcb1-4Gal, GalNAcb1-3Gal, a-GalNAc 38 DBA (Dolichos biflorus) Blood group A, GalNAca1-3GalNAc, GalNAcb1-4(Siaa2-3)Galb1-4Glc (GM2) 39 SBA (soybean, Dolichos biflorus) a-or b-linked GalNAc, Gala1-4Gal-Glc 40 Calsepa (Calystegia sepium) Galactosylated bianntenary N-type with bisecting GlcNAc (galacto > agalacto, down with Core Fuc), High-Mannose (Man2-6) 41 PTL-I (Psophocarpus tetragonolobus) a-GalNAc, Gala1-3(Fuca1-2)Gal (B-antigen) 42 MAH (Maackia amurensis) Siaa2-3Galb1-3(Siaa2-6)GalNAc (disialyl-T) 43 WGA (wheat germ, Triticum aestivum) (GlcNAcb1-4)n (Chitin), Hybrid type N-glycan, Sia 44 GSL-I A4 (Griffonia simplicifolia) a-Gal, a-GalNAc 45 GSL-I B4 (Griffonia simplicifolia) a-Gal, a-GalNAc LfDB database http://jcggdb.jp/rcmg/glycodb/LectinSearch interactions (Table 1), the lectin binding patterns indicate the (Vectibix ), an IgG2 isotype, displayed much weaker signal presence of core fucose, outer galactose, mannose, and GlcNAc intensities across the lectin chip despite a similar lectin binding oligomer, and the absence of sialic acids and tri/tetra-antennary pattern as observed for IgG1 mAbs (Fig. 2A). This data is con- complex-type N-glycans. Overall, the glycan profiles derived sistent with the reported lower level of glycan content in the 28,29 from lectin microarray are consistent with the reported glycans IgG2 antibodies in comparison with IgG1 antibodies. in IgG1 Fc that are known to contain principally bi-antennary Next, we tested 2 other therapeutic glycoproteins that are non-sialylated complex-type N-glycans with little or no high- associated with more complex glycosylation patterns. Cetuxi- mannose type or sialylation (Fig. 2C). Using the specified lec- mab (Erbitux ) was chosen because it contains N-glycosylation tin chips, we detected similar lectin binding patterns for IgG1 sites in both the antigen-binding fragment (Fab) and Fc of the mAbs containing only Fc glycosylation (Fig. 2B). Compared to molecule. In contrast to glycan profiles for IgG1 mAbs the other 5 mAbs, rituximab appeared to display relative higher (Fig. 2), cetuximab showed unique binding patterns at lectins binding signals at AOL/AAL (core fucose), RCA120 (terminal (SNA, SSA and TJA-I), which are known to bind a2-6-linked galactose), and GNA (high mannose). Panitumumab sialic acids (Fig. 3A & B). A binding signal was also detected at MABS 527 Table 2. Information of protein samples used in lectin microarray assay.* Number Proprietary name USAN name Class Expression system 1 Avastin Bevacizumab mAb CHO 2 Herceptin Trastuzumab mAb CHO 3 Humira Adalimumab mAb CHO 4 Remicade Infliximab mAb Sp2/0 5 Rituxan Rituximab mAb CHO 6 Xolair Omalizumab mAb CHO 7 Vectibix Panitumumab mAb CHO 8 Erbitux Cetuximab mAb Sp2/0 9 Enbrel Etanercept Fc-fusion protein CHO 10 Aranesp Darbepoetin alfa cytokine CHO 11 Pulmozyme Dornase alfa enzyme CHO 12 Elitek Rasburicase enzyme S. cerevisiae 13 Recombinant human transferrin, expressed in rice 14 Transferrin purified from human blood plasma 15 Neupogen Filgrastim cytokine E. coli Abbreviations used in this table: mAb, monoclonal antibody; CHO, Chinese Hamster Ovary cells; Sp2/0, murine myeloma cell line; S. cerevisiae, Saccharomyces cerevisiae; E. coli, Escherichia coli. 31,32 the a-Gal binding lectin GSL-I-A4, suggesting the presence bindings toward particular glycans in a testing sample. As of a-Gal structures in cetuximab proteins. Such a glycan variant noted, the rituximab Cy3-labeled Fab showed no lectin-binding was not detected in other samples tested in this study. Overall, signals across the lectin chips, confirming no or little interac- this data is consistent with the known glycan patterns of cetuxi- tion between the protein backbone and the lectins. mab Fab, which include an abundant N-linked sialic acid (Neu5Gc) and terminal a-Gal variants (Fig. 3C). The utility of lectin microarray in glycan profiling of Another sample tested was etanercept (Enbrel ), a homo- proteins produced by different host cell systems dimer of Fc-fusion protein consisting of TNF-a receptor and an IgG1 Fc portion, which was reported to contain 3 N-linked We assessed whether lectin microarray is capable of profiling and 13 O-linked glycosylation sites. Etanercept displayed dis- glycan variants of therapeutic proteins that are produced by dif- tinct lectin binding patterns compared to IgG1 mAbs. For ferent host cell systems such as mammalian cells, yeast strains example, strong binding signals were detected at MAL-I and and bacterial strains. These cell systems are different in their ACG, which are known to selectively bind a2-3 linked sialic glycosylation machinery, which produce proteins with distinct acid epitope. No signals were detected at lectins SNA, SSA or glycan patterns. For example, glycoproteins expressed by yeast TJA-1, showing the lack of a2-6 sialylation in the protein sam- strains usually contain high-mannose structures whereas ple. This data confirms the presence of complex glycans in Escherichia coli (E. coli) proteins are all non-glycosylated due to etanercept, including a2-3-sialic acids and abundant bi-antenn- the lack of glycosylation machinery in natural bacterial. We ary neutral glycans. Compared to other IgG1 mAbs, etanercept evaluated a panel of 6 proteins, including 2 therapeutic proteins displayed a strong and novel signal at MPA, a lectin that is produced by CHO cells (darbepoetin alfa (Aranesp ) and dor- known to selectively bind Galb1-3GalNAc and aGalNAc. nase alfa (Pulmozyme )), one therapeutic protein produced by This data is consistent with the reported abundance of O-gly- yeast (rasburicase (Elitek )), human transferrin proteins cans onto etanercept. expressed by recombinant rice strain or isolated from human To support the selectivity of lectin binding signals, we tested plasma, and filgrastim (Neupogen ) produced by E. coli the Fab and Fc purified from rituximab and cetuximab, respec- (Fig. 5A & B). Darbepoetin alfa showed strong signals at MAL- tively. Rituximab is known to contain only one N-glycosylation I, demonstrating the presence of a2-3-sialylation structures. site in its Fc. In the lectin microarray, the isolated rituximab Moreover, darbepoetin alfa displayed strong signals at PHAL- Fab showed little or no signals across the lectin chip whereas coated spots, which are known to be selective for tri-/tetra- the rituximab Fc displayed a similar lectin profile as intact rit- antennary N-glycan structures. Such a signal was not detected uximab (Fig. 4A). By contrast, cetuximab contains 2 N-glycans, in other proteins, confirming the absence of tri-/tetra-antenn- one located within its Fc portion and another in the Fab. The ary glycans in these samples. This pattern was consistent with uncommon a-Gal and Neu5Gc epitopes were reported to be the reported data that darbepoetin alfa contains high levels of solely in the Fab (Fig. 3C). The cetuximab Fc expressed a typi- sialylation and abundant tri- or tetra-antennary structures cal lectin profile for IgG1 Fc glycans (bi-antennary G0F, G1F, (Fig. 5C). Darbepoetin displayed relative week binding signals and G2F) (Fig. 4B). The lectin signals of SNA, SSA and TJA-1, at other lectin spots compared to MAL-I and PHAL, raising a which indicate expression of a2-6-sialylation, were only present possibility that those other glycan species (e.g., Galb1-4GlcNAc in intact cetuximab and cetuximab Fab profiles, but were absent and mannose oligomers) might be “capped” by the outermost in the Fc profile. GSL-I signal also indicated the presence of galactose and sialic acid. a-Gal structure in the Fab, but not in the Fc (Fig. 4B). These Dornase alfa (Pulmozyme ), a recombinant enzyme data not only confirm the proper locations of glycosylation sites expressed by CHO cells, displayed a simpler lectin binding pat- in Fc or Fab portions, but also support the selectivity of lectin tern compared to the above IgG1 and IgG2 mAbs (Fig. 5A and 528 L. ZHANG ET AL. Figure 2. Lectin binding profiles of therapeutic IgG monoclonal antibodies. The indicated therapeutic mAbs, including bevacizumab, trastuzumab, adalimumab, inflixi- mab, rituximab, omalizumab, and panitumumab were labeled with Cy3 and applied onto the lectin chips containing 45 distinct lectins with each being printed in tripli- cate. (A) Lectin binding images of the indicated samples. (B) Relative binding signals at specific lectin spots were derived from the images in A and normalized to protein markers on the same chip (mean § SD). Shown are representatives of 3 independent experiments. The coefficient of variation (CV) was determined to be < 10% for most lectin-glycan binding signals of the samples tested. (C) Typical glycan structures present in the Fc portion of therapeutic IgG1 mAbs. MABS 529 Figure 3. Lectin binding profiles of cetuximab and etanercept. Cy3-labeled samples were applied onto the lectin chips as in Figure 2. Shown are (A) representative lectin binding images, (B) Relative binding signals at specific lectin spots (mean § SD), and (C) Typical glycan structures present in the Fab portion of cetuximab. 5B). The sample tested showed unique binding signals at SNA/ demonstrating the absence of the relevant glycan species in ras- SSA for a2-6-sialylation, RCA120 for Galb1-4GlcNAc, DSA buricase. The two versions of human transferrin proteins also 37,38 for GlcNAc oligomer and/or Galb1-4GlcNAc, ConA for showed distinct glycan patterns in which the recombinant mannose, and LEL/STL for GlcNAc oligomers. The spectrum human transferrin expressed in rice (transferrin-rice) showed of selective binding signals suggests the presence of complex- binding signals primarily at mannose-binding lectin (NPA) type glycans with a2-6-sialylation in dornase alfa molecules. By and GlcNAc oligomer-binding lectins (LEL, STL and UDA). contrast, rasburicase (Elitek ), a therapeutic glycoprotein pro- The DSA signal indicated the presence of either GlcNAc duced by yeast strains, displayed distinctively different lectin oligomer or Galb1-4GlcNAc. By contrast, transferrin proteins profiles compared to the above described products produced by isolated from human plasma showed additional signals at a2- mammalian cells. Rasburicase showed relatively weak binding 6-sialic acid-binding lectins (SNA, SSA, and TJA-I) and galac- signals across the lectin chips, which is consistent with its tose-binding lectins (RCA120 and PHAE). As expected, no known low level of glycosylation. Despite the overall weak lectin binding signals were detected for filgrastim (Neupogen ) binding signals, rasburicase appeared to interact exclusively that is produced by E. coli as a non-glycosylated protein. with mannose binding lectins (NPA, ConA, and GNA) and GlcNAc oligomer binding lectins (STL and UDA). This data The utility of lectin microarray in monitoring terminal confirms the presence of high-mannose carbohydrates that are galactosylation and sialylation of glycoproteins mainly attached onto glycoproteins produced by yeast strains. No binding signals were detected at sialic acid-binding lectins To further evaluate the utility of lectin microarray in glycan (e.g., MAL_I, SNA, SSA, and TJA-I), fucose-binding lectins profiling, we prepared protein variants with defined galactosy- (e.g., PSA and LCA) or galactose-binding lectins (e.g., RCA 120 lation and sialylation modifications. This was achieved through and PHAE), even when the protein concentration of rasburi- in vitro enzymatic glycoengineering of rituximab using com- case was enhanced to 500 ng/mL (data not shown), mercially available galactosyltransferase and sialyltransferase. 530 L. ZHANG ET AL. Figure 4. Glycan profiling of Fabs and Fcs. The Fabs and Fcs of rituximab (A) and cetuximab (B) were prepared as described in Materials and Methods. Purified Fab and Fc were analyzed by reducing SDS-PAGE (left panel) and lectin microarray (right panel). As noted, the dimeric Fcs (~55 kDa) were reduced to monomeric products (~30 kDa) on SDS-PAGE under reducing conditions. b1-4-galactosyltransferase (b1-4GalT) catalyzes the transfer of signals at positions of ECA and RCA120. Both lectins are galactose from donor substrate UDP-galactose (UDP-Gal) to known to bind N-glycan Galb1-4GlcNAc, and therefore the GlcNAcb1-2Man units of glycoproteins to form a b1-4-galacto- incurred ECA lectin-binding signal and the significantly sylation linkage, while a2-6-sialyltransferase (a2-6SiaT) facili- enhanced RCA120 signal were indicative of the increased galac- tates sialylation by adding sialic acids to terminal Galb1- tose species in the samples. Notably, there was a concomitant 4GlcNAc units. Modified rituximab protein variants were puri- decrease in the binding signals at ABA, a lectin with dual bind- fied and then characterized using mass spectrometry (MS), ing affinity toward Gal-exposed O-glycans and GlcNAc- revealing distinct deconvoluted MS spectra for the light chain exposed N-glycans. Because rituximab was not reported to and heavy chain (Fig. 6A). The light chain fragments resolved undergo O-glycosylation, the ABA binding signal detected for as a single species at an average mass of 23036 Da, correspond- native rituximab was likely due to GlcNAc-exposed N-glycans. 42,43 ing to the theoretical mass of rituximab light chain. Consis- Such GlcNAc-exposed N-glycans appeared to be occupied tent with the lack of glycosylation sites within the rituximab upon galactosylation catalyzed in the reactions with b1-4GalT. light chains, the mass of light chain remained unchanged after Samples of native rituximab and galactosylated species showed treatments of rituximab with b1-4GalT or further with a2- no detectable signals at the lectins SNA, SSA, TJA-I that are 6SiaT. The other 3 major mass species at 50507, 50669, 50832 known to bind sialic acids. By contrast, strong binding signals Da correspond to the heavy chains of rituximab containing at these lectins were detected for the samples derived from the 42,43 G0F, G1F or G2F glycoforms, respectively (Fig. 2C). Treat- sequential reactions with b1-4GalT and a2-6SiaT enzymes. No ment of rituximab with b1-4GalT resulted in a mass shift from signal was detected at MAL-I (a lectin selective to a2-3-sialyla- G0F and G1F to G2F, indicating galactosylation reactions were tion), demonstrating not only the specificity of glycan engineer- effectively accomplished. When the galactosylated rituximab ing but also the utility of the lectin microarray in distinguishing mixture was sequentially treated with a2-6SiaT, the final prod- different terminal sialic acid linkages. uct showed a further mass shift from 50832 Da (G2F) to 51122 Da (C290) and 51414 Da (C582). These mass shifts corre- Discussion sponded to an addition of one or two Neu5Ac residues, indicat- ing that G2F glycoform was effectively converted to primarily The complexity of glycosylation poses an analytical challenge in mono-sialylated species (S1G2F) and a small portion of di-sia- the development of therapeutic glycoproteins. The methods lylated species (S2G2F). These data indicate that the in vitro most commonly used for analysis include MS, HPLC, HPAEC- glycan engineering reactions produced rituximab variants con- PAD, and CE. Among these methods, MS remains a powerful taining the desired modifications (e.g., galactosylation and tool in the characterization of glycosylation site(s) occupancy sialylation). and carbohydrate structures. MS-based methods involve enzy- The engineered rituximab samples with defined glycan var- matic digestion of a glycoprotein into peptide fragments and iations were then analyzed by lectin microarray (Fig. 6B). The separation by liquid chromatography. HPLC-, HPAEC- and reaction buffer alone had no effect on lectin binding profile of CE-based methods usually require the release of glycans from a rituximab. The sample resulting from b1-4GalT reaction (rit- glycoprotein through enzymatic or chemical reactions. As uximab C b1-4GalT), which was confirmed by MS to majorly such, an accurate assessment of glycosylation requires a com- contain G2F galactosylation (Fig. 6A), displayed strong binding plete release of all glycans that are present in a glycoprotein MABS 531 Figure 5. Glycan profiles of proteins produced by different host cell systems. The proteins tested include therapeutic proteins produced by CHO cells (darbepoetin alfa and dornase alfa), yeast strains (rasburicase), or E. coli (filgrastim), and human transferrin protein expressed by recombinant rice (transferrin-rice) or isolated from human plasma (transferrin-human). (A) Lectin binding images. (B) Relative binding signals at specific lectin spots (mean § SD). (C) Typical glycan structures present in darbepoe- tin alfa. being tested. By contrast, lectin microarray directly measures commercial lectin chips, we were able to determine glycan pro- glycan profiles on an intact protein without the need for enzy- files for a panel of therapeutic proteins that were generally con- matic digestion or clipping glycans from the protein backbone. sistent with their known glycosylation properties (Figs. 2-5). Such a platform is unique in increasing the likelihood of full Notably the lectin microarray was highly sensitive to alterations coverage of all glycan variants of a glycoprotein. Using the in the terminal glycan structures, i.e., galactosylation vs. 532 L. ZHANG ET AL. Figure 6. Assessing glycan variants in glycan-engineered rituximab protein samples. Rituximab was incubated with a reaction buffer alone (rituximab C buffer), b1-4-gal- actosyltransferase (rituximab C b1-4GalT), and b1-4-galactosyltransferase followed by a2-6-sialyltransferase (rituximab C b1-4GalT C a2-6SiaT) (see detail in Materials and Methods). After affinity purification, the resulting samples were analyzed using mass spectrometry and lectin microarray, respectively. Shown are representatives of deconvoluted mass spectra (A) and corresponding lectin binding profiles (B) for the samples produced under the indicated conditions. MABS 533 sialyation (Fig. 6). Lectin microarray can effectively distinguish 0.4 mg/mL using Fab digestion buffer, and aliquot of 120 mL glycan isomers containing different sialic acid linkages. Our was added to the spin column containing equilibrated immobi- data demonstrate a usefulness of the lectin microarray in lized papain. The digestion reaction was incubated overnight at screening glycan patterns of protein samples. As noted, the 37 C. Digested samples were collected by column centrifuga- commercial lectin chips employed in this study was not tion and applied onto NAb Protein A Plus Spin Column to sep- designed specifically for assessing therapeutic mAbs. In princi- arate the Fab and Fc. Protein concentrations of the resulting ple, assay performance could be further improved through the samples were determined using Micro BCA Protein Assay Kit use of lectins with improved selectivity and binding affinity to (Thermo Scientific). distinct glycan species. To analyze a specific glycoprotein, lectin chips can be customized to include the lectins that are relevant Enzymatic glycan engineering of rituximab to the glycan species that are possibly present in the testing sample. Upon optimization of lectin chips, lectin microarray Rituximab was used to create glycan variants of a mAb through platform could be adopted as a complementary tool for high in vitro enzymatic reactions. Briefly, rituximab stock solution throughput screening of glycan profiles of therapeutic was diluted to 2 mg/mL using 1£ galactosyltransferase (GalT) glycoproteins. reaction buffer (10 mM MnCl with 100 mM MES, pH 6.5). Uridine-5’-diphosphogalactose disodium salt (UDP-Gal) solu- tion (50 mg/mL) was also prepared using 1£ GalT reaction Materials and methods buffer. Cytidine 5’-monophospho-N-acetylneuraminic acid disodium salt (CMP-Neu5Ac) solution (50 mg/mL) was pre- Therapeutic proteins and reagents pared using 1£ sialyltransferase (SiaT) reaction buffer (50 mM All therapeutic proteins were purchased from the Division of Tris-Acetate, pH 7.5). To facilitate galactosylation, 50 mL ritux- Veterinary Resources pharmacy, Office of Research Services, imab (2 mg/mL in solution) was mixed with 4 mL UDP-Gal National Institute of Health, and handled according to respec- (50 mg/mL in solution) and 1.5 mg b1-4-galactosyltransferase tive package inserts. Recombinant human transferrin expressed (b1-4GalT). The final reaction volume was adjusted to 100 mL by rice and transferrin purified from human blood plasma were using 1£ GalT reaction buffer, and was left to incubate over- purchased from Sigma-Aldrich. Cy3 mono-reactive dye was night at 37 C, resulting galactosylation of rituximab. To further obtained from GE Healthcare. Zeba spin desalting columns add sialylation, the above galactosylated rituximab was mixed (7K MWCO) were purchased from Thermo Scientific. Lectin with 200 mg CMP-Neu5Ac (50 mg/mL in solution) and 10 mg chips coated with 45 distinct lectin proteins and the probing a2-6-sialyltransferase (a2-6SiaT). After adjusting volume to solutions were obtained from GlycoTechnica. Glycosyltransfer- 200 mL using 1£ SiaT reaction buffer, the mixture was incu- ases b1-4-galactosyltransferase (b1-4GalT) and a2-6-sialyl- bated at 37 C for 3 hours and then frozen at ¡20 C. The engi- transferase (a2-6SiaT) were purchased from Prozyme. Micro neered antibodies containing different glycan variants were BCA Protein Assay Kit, NAb Protein A Plus Spin Kit and purified using NAb Protein A Plus Spin Kit by following the Pierce Fab micro preparation kit were obtained from Thermo protocol instructions (Thermo Scientific). Scientific. NuPAGE 4-12% Bis-Tris gels and Coomassie stain (SimplyBlue SafeStain) were purchased from Life Technologies. Liquid chromatography–mass spectrometry analysis Liquid chromatography–mass spectrometry analyses were con- Lectin microarray analysis ducted on an Agilent 1260 HPLC-Chip nano-electrospray-ioni- Protein samples were diluted to 50 mg/mL in phosphate-buff- zation 6520 Q-TOF MS system. Solvent-A was 0.1% formic ered saline (PBS). Aliquots (20 mL, 1 mg protein) were mixed acid in water and solvent-B was 0.1% formic acid in 95% aceto- with 100 mg Cy3 mono-reactive dye, and incubated at room nitrile. Mass correction was enabled during the run using inter- temperature for 2 h. The unbound Cy3 dye was removed using nal reference ions with masses of 299.2945 and 1221.9906 Da. Zeba spin desalting columns (7K MWCO). Cy3-labeled protein Intact protein mass measurement was performed using an Agi- samples were diluted to 125 ng/mL in probing solution, and an lent 43 mm 300 A C8 chip with a 40 nL trap column (G4240- aliquot of 100 mL was applied onto a lectin chip, and incubated 63001 SPQ105). All rituximab samples were reduced in 10 mM at room temperature on an orbital shaker for 18 hours in dark. DTT for 1 hour at 37 C, and then formic acid was added to The resulting lectin chips were scanned for fluorescence inten- reach a final concentration of 0.1% (V/V). The samples were sities on each lectin-coated spot using an evanescent-field fluo- diluted to 150 ng/mL in 0.1% formic acid, and 1 mL(~1 pmol) or rescence scanner GlycoStation Reader 1200 (GlycoTechnica, larger volume was injected onto the trap column in the C8 chip Japan). at a flow rate of 2.5 mL/min in 100% A, then eluted at 0.5 mL/ min with a linear gradient of 10–100% B in 18 min and held for additional 4 min. Q-TOF VCap, fragmentor, and skimmer Preparation of IgG Fab and Fc settings were 1,890 eV, 225 eV, and 65 eV, respectively. HPLC- The Fab and Fc of therapeutic mAbs were prepared using Chip gas temperature and drying gas flow rate were 350 C and Pierce Fab micro preparation kit following the vendor’s proto- 9 L/min, respectively. The data were analyzed using Agilent col (Thermo Scientific). Briefly, immobilized papain (settled MassHunter (version B.05.00) Qualitative Analysis software resin) was located in a 0.8 mL spin column and equilibrated and deconvoluted with Agilent MassHunter Bioconfirm using Fab digestion buffer. Antibody samples were diluted to software. 534 L. ZHANG ET AL. Disclosure of potential conflicts of interest 18. Hirabayashi J, Yamada M, Kuno A, Tateno H. Lectin microarrays: concept, principle and applications. Chem Soc Rev 2013; 42:4443-58; No potential conflicts of interest were disclosed. PMID:23443201; http://dx.doi.org/10.1039/c3cs35419a 19. Cook MC, Kaldas SJ, Muradia G, Rosu-Myles M, Kunkel JP. Compari- son of orthogonal chromatographic and lectin-affinity microarray methods for glycan profiling of a therapeutic monoclonal antibody. J Chromatogr B 2015; 997:162-78; PMID:26114652; http://dx.doi.org/ References 10.1016/j.jchromb.2015.05.035 20. Huang WL, Li YG, Lv YC, Guan XH, Ji HF, Chi BR. Use of lectin 1. Kornfeld R, Kornfeld S. Assembly of asparagine-linked oligosacchar- microarray to differentiate gastric cancer from gastric ulcer. World J ides. Annu Rev Biochem 1985; 54:631-64; PMID:3896128; http://dx. Gastroenterol 2014; 20:5474-82; PMID:24833877; http://dx.doi.org/ doi.org/10.1146/annurev.bi.54.070185.003215 10.3748/wjg.v20.i18.5474 2. Ghaderi D, Zhang M, Hurtado-Ziola N, Varki A. Production plat- 21. Kobayashi Y, Masuda K, Banno K, Kobayashi N, Umene K, Nogami forms for biotherapeutic glycoproteins. Occurrence, impact, and chal- Y, Tsuji K, Ueki A, Nomura H, Sato K, et al. Glycan profiling of gesta- lenges of non-human sialylation. Biotechnol Genet Eng Rev 2012; tional choriocarcinoma using a lectin microarray. Oncol Rep 2014; 28:147-75; PMID:22616486; http://dx.doi.org/10.5661/bger-28-147 31:1121-6; PMID:24424471; http://dx.doi.org/10.3892/or.2014.2979 3. Sola RJ, Griebenow K. Glycosylation of therapeutic proteins: an effec- 22. Xin AJ, Cheng L, Diao H, Wang P, Gu YH, Wu B, Wu YC, Chen GW, tive strategy to optimize efficacy. BioDrugs 2010; 24:9-21; Zhou SM, Guo SJ, et al. Comprehensive profiling of accessible surface PMID:20055529; http://dx.doi.org/10.2165/11530550-000000000- glycans of mammalian sperm using a lectin microarray. Clin Proteo- mics 2014; 11:10; PMID:24629138; http://dx.doi.org/10.1186/1559- 4. Sola RJ, Griebenow K. Effects of glycosylation on the stability of pro- 0275-11-10 tein pharmaceuticals. J Pharm Sci 2009; 98:1223-45; PMID:18661536; 23. Gupta G, Surolia A, Sampathkumar SG. Lectin microarrays for glyco- http://dx.doi.org/10.1002/jps.21504 mic analysis. OMICS 2010; 14:419-36; PMID:20726799; http://dx.doi. 5. Narhi LO, Arakawa T, Aoki KH, Elmore R, Rohde MF, Boone T, org/10.1089/omi.2009.0150 Strickland TW. The effect of carbohydrate on the structure and stabil- 24. Song T, Ozcan S, Becker A, Lebrilla CB. In-depth method for the char- ity of erythropoietin. J Biol Chem 1991; 266:23022-6; PMID:1744097 acterization of glycosylation in manufactured recombinant monoclo- 6. Desnick RJ, Schuchman EH. Enzyme replacement therapy for lyso- nal antibody drugs. Anal Chem 2014; 86:5661-6; PMID:24828102; somal diseases: lessons from 20 years of experience and remaining http://dx.doi.org/10.1021/ac501102t challenges. Annu Rev Genomics Hum Genet 2012; 13:307-35; 25. Stadlmann J, Pabst M, Kolarich D, Kunert R, Altmann F. Analysis of PMID:22970722; http://dx.doi.org/10.1146/annurev-genom-090711- immunoglobulin glycosylation by LC-ESI-MS of glycopeptides and oligosaccharides. Proteomics 2008; 8:2858-71; PMID:18655055; 7. Tsuda E, Kawanishi G, Ueda M, Masuda S, Sasaki R. The role of car- http://dx.doi.org/10.1002/pmic.200700968 bohydrate in recombinant human erythropoietin. Eur J Biochem 26. Visser J, Feuerstein I, Stangler T, Schmiederer T, Fritsch C, Schiestl M. 1990; 188:405-11; PMID:2156701; http://dx.doi.org/10.1111/j.1432- Physicochemical and functional comparability between the proposed 1033.1990.tb15417.x biosimilar rituximab GP2013 and originator rituximab. BioDrugs 8. Morell AG, Gregoriadis G, Scheinberg IH, Hickman J, Ashwell G. The 2013; 27:495-507; PMID:23649935; http://dx.doi.org/10.1007/s40259- role of sialic acid in determining the survival of glycoproteins in the 013-0036-3 circulation. J Biol Chem 1971; 246:1461-7; PMID:5545089 27. Huhn C, Selman MH, Ruhaak LR, Deelder AM, Wuhrer M. IgG gly- 9. Houdebine LM. Production of pharmaceutical proteins by transgenic cosylation analysis. Proteomics 2009; 9:882-913; PMID:19212958; animals. Comp Immunol Microbiol Infect Dis 2009; 32:107-21; http://dx.doi.org/10.1002/pmic.200800715 PMID:18243312; http://dx.doi.org/10.1016/j.cimid.2007.11.005 28. Wuhrer M, Stam JC, van de Geijn FE, Koeleman CA, Verrips CT, 10. Nimmerjahn F, Ravetch JV. Translating basic mechanisms of IgG Dolhain RJ, Hokke CH, Deelder AM. Glycosylation profiling of effector activity into next generation cancer therapies. Cancer Immun immunoglobulin G (IgG) subclasses from human serum. Proteomics 2012; 12:13; PMID:22896758 2007; 7:4070-81; PMID:17994628; http://dx.doi.org/10.1002/ 11. Marino K, Bones J, Kattla JJ, Rudd PM. A systematic approach to pro- pmic.200700289 tein glycosylation analysis: a path through the maze. Nat Chem Biol 29. Perdivara I, Peddada SD, Miller FW, Tomer KB, Deterding LJ. Mass 2010; 6:713-23; PMID:20852609; http://dx.doi.org/10.1038/ spectrometric determination of IgG subclass-specific glycosylation nchembio.437 profiles in siblings discordant for myositis syndromes. J Proteome Res 12. Beck A, Wagner-Rousset E, Ayoub D, Van DA, Sanglier-Cianferani S. 2011; 10:2969-78; PMID:21609021; http://dx.doi.org/10.1021/ Characterization of therapeutic antibodies and related products. Anal pr200397h Chem 2013; 85:715-36; PMID:23134362; http://dx.doi.org/10.1021/ 30. Qian J, Liu T, Yang L, Daus A, Crowley R, Zhou Q. Structural charac- ac3032355 terization of N-linked oligosaccharides on monoclonal antibody 13. Higgins E. Carbohydrate analysis throughout the development of a cetuximab by the combination of orthogonal matrix-assisted laser protein therapeutic. Glycoconj J 2010; 27:211-25; PMID:19888650; desorption/ionization hybrid quadrupole-quadrupole time-of-flight http://dx.doi.org/10.1007/s10719-009-9261-x tandem mass spectrometry and sequential enzymatic digestion. Anal 14. Lingg N, Zhang P, Song Z, Bardor M. The sweet tooth of biopharma- Biochem 2007; 364:8-18; PMID:17362871; http://dx.doi.org/10.1016/j. ceuticals: importance of recombinant protein glycosylation analysis. ab.2007.01.023 Biotechnol J 2012; 7:1462-72; PMID:22829536; http://dx.doi.org/ 31. Murphy LA, Goldstein IJ. Five alpha-D-galactopyranosyl-binding iso- 10.1002/biot.201200078 lectins from Bandeiraea simplicifolia seeds. J Biol Chem 1977; 15. Pabst M, Altmann F. Glycan analysis by modern instrumental meth- 252:4739-42; PMID:68957 ods. Proteomics 2011; 11:631-43; PMID:21241022; http://dx.doi.org/ 32. Lescar J, Loris R, Mitchell E, Gautier C, Chazalet V, Cox V, Wyns L, 10.1002/pmic.201000517 Perez S, Breton C, Imberty A. Isolectins I-A and I-B of Griffonia (Ban- 16. Rohrer JS, Basumallick L, Hurum D. High-performance anion- deiraea) simplicifolia. Crystal structure of metal-free GS I-B(4) and exchange chromatography with pulsed amperometric detection for molecular basis for metal binding and monosaccharide specificity. J carbohydrate analysis of glycoproteins. Biochemistry (Mosc ) 2013; Biol Chem 2002; 277:6608-14; PMID:11714720; http://dx.doi.org/ 78:697-709; PMID:24010833; http://dx.doi.org/10.1134/ 10.1074/jbc.M109867200 S000629791307002X 33. Tan Q, Guo Q, Fang C, Wang C, Li B, Wang H, Li J, Guo Y. Charac- 17. Varki A, Etzler ME, Cummings RD, Esko JD. Discovery and Classifi- terization and comparison of commercially available TNF receptor 2- cation of Glycan-Binding Proteins. In: Varki A, Cummings RD, Esko Fc fusion protein products. MAbs 2012; 4:761-74; PMID:23032066; JD, et al. Essentials of Glycobiology. 2nd edition. Cold Spring Harbor http://dx.doi.org/10.4161/mabs.22276 (NY): Cold Spring Harbor Laboratory Press; 2009. Chapter 26 MABS 535 34. Houel S, Hilliard M, Yu YQ, McLoughlin N, Martin SM, Rudd PM, Hereditas 1996; 124:251-9; PMID:8931358; http://dx.doi.org/10.1111/ Williams JP, Chen W. N- and O-glycosylation analysis of etanercept j.1601-5223.1996.00251.x using liquid chromatography and quadrupole time-of-flight mass 39. Higgins E. Carbohydrate analysis throughout the development of a spectrometry equipped with electron-transfer dissociation functional- protein therapeutic. Glycoconj J 2010; 27:211-25; PMID:19888650; ity. Anal Chem 2014; 86:576-84; PMID:24308717; http://dx.doi.org/ http://dx.doi.org/10.1007/s10719-009-9261-x 10.1021/ac402726h 40. Hossler P, Khattak SF, Li ZJ. Optimal and consistent protein glycosyl- 35. Huang J, Xu Z, Wang D, Ogata CM, Palczewski K, Lee X, Young NM. ation in mammalian cell culture. Glycobiology 2009; 19:936-49; Characterization of the secondary binding sites of Maclura pomifera PMID:19494347; http://dx.doi.org/10.1093/glycob/cwp079 agglutinin by glycan array and crystallographic analyses. Glycobiology 41. Jaffe SR, Strutton B, Levarski Z, Pandhal J, Wright PC. Escherichia coli 2010; 20:1643-53; PMID:20826825; http://dx.doi.org/10.1093/glycob/ as a glycoprotein production host: recent developments and chal- cwq118 lenges. Curr Opin Biotechnol 2014; 30:205-10; PMID:25156401; 36. Shahrokh Z, Royle L, Saldova R, Bones J, Abrahams JL, Artemenko http://dx.doi.org/10.1016/j.copbio.2014.07.006 NV, Flatman S, Davies M, Baycroft A, Sehgal S, et al. Erythropoietin 42. Huang W, Giddens J, Fan SQ, Toonstra C, Wang LX. Chemoenzy- produced in a human cell line (Dynepo) has significant differences in matic glycoengineering of intact IgG antibodies for gain of functions. glycosylation compared with erythropoietins produced in CHO cell J Am Chem Soc 2012; 134:12308-18; PMID:22747414; http://dx.doi. lines. Mol Pharm 2011; 8:286-96; PMID:21138277; http://dx.doi.org/ org/10.1021/ja3051266 10.1021/mp100353a 43. Wang B, Gucinski AC, Keire DA, Buhse LF, Boyne MT. Structural 37. Crowley JF, Goldstein IJ, Arnarp J, Lonngren J. Carbohydrate binding comparison of two anti-CD20 monoclonal antibody drug products studies on the lectin from Datura stramonium seeds. Arch Biochem using middle-down mass spectrometry. Analyst 2013; 138:3058-65; Biophys 1984; 231:524-33; PMID:6203486; http://dx.doi.org/10.1016/ PMID:23579346; http://dx.doi.org/10.1039/c3an36524g 0003-9861(84)90417-X 44. Nakamura-Tsuruta S, Kominami J, Kuno A, Hirabayashi J. Evidence 38. Myllyharju J, Nokkala S. Glycoproteins with N-acetylglucosamine and that Agaricus bisporus agglutinin (ABA) has dual sugar-binding speci- mannose residues in Chinese hamster metaphase chromosomes. ficity. Biochem Biophys Res Commun 2006; 347:215-20; PMID:16824489; http://dx.doi.org/10.1016/j.bbrc.2006.06.073 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png mAbs Taylor & Francis

The use of lectin microarray for assessing glycosylation of therapeutic proteins

mAbs , Volume 8 (3): 12 – Apr 2, 2016

Loading next page...
 
/lp/taylor-francis/the-use-of-lectin-microarray-for-assessing-glycosylation-of-VdoDWd9d82

References (90)

Publisher
Taylor & Francis
Copyright
This article not subject to US copyright law. Published with license by Taylor & Francis Group, LLC
ISSN
1942-0870
eISSN
1942-0862
DOI
10.1080/19420862.2016.1149662
pmid
26918373
Publisher site
See Article on Publisher Site

Abstract

MABS 2016, VOL. 8, NO. 3, 524–535 http://dx.doi.org/10.1080/19420862.2016.1149662 REPORT Lei Zhang, Shen Luo, and Baolin Zhang Office of Biotechnology Products, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA ABSTRACT ARTICLE HISTORY Received 4 November 2015 Glycans or carbohydrates attached to therapeutic glycoproteins can directly affect product quality, safety Revised 20 January 2016 and efficacy, and therefore must be adequately analyzed and controlled throughout product life cycles. Accepted 28 January 2016 However, the complexity of protein glycosylation poses a daunting analytical challenge. In this study, we evaluated the utility of a lectin microarray for assessing protein glycans. Using commercial lectin chips, KEYWORDS which contain 45 lectins toward distinct glycan structures, we were able to determine the lectin binding Glycan analysis; lectin patterns of a panel of 15 therapeutic proteins, including 8 monoclonal antibodies. Lectin binding signals microarray; monoclonal were analyzed to generate glycan profiles that were generally consistent with the known glycan patterns antibodies; therapeutic glycoproteins for these glycoproteins. In particular, the lectin-based microarray was found to be highly sensitive to variations in the terminal carbohydrate structures such as galactose versus sialic acid epitopes. These data suggest that lectin microarray could be used for screening glycan patterns of therapeutic glycoproteins. Introduction Glycans attached to a therapeutic protein can directly affect Glycosylation of proteins is a complex post-translational modi- product quality, safety and efficacy. It is well documented that fication that attaches carbohydrates or named glycans at spe- glycans attached to a protein affect protein solubility and stabil- 3-5 3,6-8 cific sites on a protein backbone, most commonly at Asn ity, pharmacokinetics/pharmacodynamics (PK/PD), and 2,9 (N-linked) or Ser/Thr (O-linked) residues. The N-linked gly- immunogenicity. In the latter, non-human glycans attached cosylation occurs at the consensus sequence of Asn-X-Ser/Thr onto a therapeutic protein such as Neu5Gc and terminal a-Gal (where X is any amino acid except proline), whereas O-linked epitopes could cause immunogenic responses. For many mono- glycans are usually attached to Ser or Thr residues. Both N- clonal antibodies (mAbs), proper glycosylation of the crystal- and O-glycosylation involve a series of enzymatic reactions cat- lizable fragment (Fc) is essential to IgG antibody effector alyzed by glycan-processing enzymes, which are responsible for functions. Therefore, glycan moieties of therapeutic proteins trimming and modifications of glycan epitopes, resulting in must be adequately characterized and controlled throughout diverse N-glycan structures (e.g., high-mannose, complex, and product life cycle. The commonly used methods include high- hybrid glycans) and O-glycan variants containing up to 8 O- performance liquid chromatography (HPLC), high-perfor- GalNAc glycan core structures. To add complexity, protein gly- mance anion-exchange chromatography with pulsed ampero- cosylation is influenced by the type of host cells and fluctua- metric detection (HPAEC-PAD), mass spectrometry (MS) and 11-16 tions in fermentation conditions (e.g., media, pH, temperature, capillary electrophoresis (CE), which provide information agitation). For instance, therapeutic glycoproteins produced on glycosylation sites, site occupancy, and contents of glycan by mammalian cells such as Chinese hamster ovary (CHO) variants attached to glycoproteins. cells usually contain human-like glycans. By contrast, proteins There is growing interest in the development of high expressed by yeast strains usually contain high levels of man- throughput platforms for assessing protein glycan profiles. Lec- nose (up to 100 units). Other hosts including engineered plant tins are glycan binding proteins (GBPs) that selectively recog- cells and genetically modified animals may produce proteins nize glycan epitopes of free carbohydrates or glycoproteins. with non-human glycan variants such as xylose, N-glycolyl- Lectin-based microarrays have been used to analyze glycan pro- 18-23 neuraminic acid (Neu5Gc) or terminal a-galactose (a-Gal), files of purified glycoproteins or cell surface proteins. In which are known to be immunogenic. As a result, a glycopro- this study, we evaluated the potential utility of a lectin microar- tein produced by living cell systems usually contains a mixture ray for characterization of therapeutic glycoproteins. Using of different glycoforms. These protein variants share an identi- commercial lectin chips containing 45 distinct lectins, we tested cal peptide backbone, but may differ in glycosylation properties a panel of 15 therapeutic proteins for their glycan profiles. Our such as glycosylation site, glycan structure and content. data show that the lectin microarray is robust in generating CONTACT Baolin Zhang baolin.zhang@fda.hhs.gov Disclaimer: The comments in this paper reflect the views of the author and should not be construed to represent the Food and Drug Administration (FDA)’s views or policies. Supplemental data for this article can be accessed on the publisher’s website. Published with license by Taylor & Francis Group, LLC. This article not subject to US copyright law. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits unre- stricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the named author(s) have been asserted. MABS 525 glycan profiles that are generally consistent with the known gly- showing that overnight incubation is required to obtain optimal can characteristics of an individual glycoprotein. binding between a Cy3-labeled glycoprotein and the lectins printed on chips (Supplement II). In the following experiments, a fixed concentration of protein samples (125 ng/mL) was applied to lectin chips and incubated overnight then lectin Results binding signals were detected. Bound glycoprotein signals were The utility of a lectin microarray in determining glycan determined using evanescent-field fluorescence scanner, which profiles of therapeutic mAbs allows a direct measure of glycans bound onto lectins without the need of washing procedures to remove unbound species We assessed the utility of a lectin microarray in profiling glycan (Fig. 1B). Using lectin chips from different batches, we deter- variants of therapeutic proteins using commercial lectin micro- mined the assay reproducibility to be < 10% CV for most lec- chips printed with 45 distinct lectin proteins (Fig. 1A). These tin-glycan binding signals of the samples tested. include lectins that selectively bind core-fucose, sialic acids, N- We first tested a panel of 6 therapeutic mAbs, including bev- acetyl-D-lactosamine (Galb1-4GlcNAc), mannose, or N-acetyl- acizumab (Avastin ), trastuzumab (Herceptin ), adalimumab glucosamine (GlcNAc) oligomers, respectively (Table 1). (Humira ), infliximab (Remicade ), rituximab (Rituxan ) and These glycan structures are commonly found in recombinant omalizumab (Xolair ). These products are known to be IgG1 glycoproteins. We tested a panel of 15 distinct proteins, includ- 24-26 isotypes with glycosylation exclusively occurring in the Fc. ing 8 therapeutic mAbs, one Fc-fusion protein, 4 recombinant All these IgG1 mAbs showed similar lectin binding patterns therapeutic cytokines and enzymes, and 2 different versions of (Fig. 2A & B) in which strong binding signals were detected at human transferrin proteins (Table 2). To facilitate microarray lectins with binding selectivity to core fucose (PSA, LCA, AOL, analysis, protein samples were fluorescently labeled with Cy3 AAL), galactose (RCA120, PHAE), mannose (NPA, ConA and followed by incubation onto lectin-coated chips. We first per- GNA), and GlcNAc oligomer (LEL, STL, UDA). No binding formed dose-titrations and identified the optimal concentra- signals were detected at lectins for sialic acids (MAL_I, SNA, tions of Cy3-labeled protein sample that fell within the linear SSA, TJA-I) or tri/tetra-antennary complex-type N-glycan response ranges for most lectin spots (Supplement I). We also (PHAL). Based on the known selectivity of lectin-glycan tested the effect of incubation time on lectin binding signals, Figure 1. Schematic view of lectin microarray. (A) Lectin microchips used in this study consist of 45 distinct lectins that selectively bind structural variants of carbohy- drates attached onto a protein. Each lectin is printed in triplicate. The lectin-printing layout of lectin chips was provided by the vendor (GlycoTechnica). (B) Protein sam- ples are labeled with a fluorescent dye (e.g., Cy3) and then applied onto the lectin chips. The binding signals at each lectin spots are measured using an evanescent-field fluorescence scanner, detecting the presence or absence of glycan variants in the testing sample based on the known selectivity of lectins toward particular glycan structures. 526 L. ZHANG ET AL. Table 1. Reported glycan selectivity of the 45 lectins used in the microarray assay.* Lectin No. Lectin (origin) Reported glycan selectivity 1 LTL (Lotus tetragonolobus) Fuca1-3(Galb1-4)GlcNAc (Lewis x), Fuca1-2Galb1-4GlcNAc (H-type 2) 2 PSA (Pisum sativum) Fuca1-6GlcNAc (Core Fuc) , a-Man 3 LCA (Lens culinaris) Fuca1-6GlcNAc (Core Fuc), a-Man 4 UEA-I (Ulex europaeus) Fuca1-2Galb1-4GlcNAc (H-type 2) 5 AOL (Aspergillus oryzae) Fuca1-6GlcNAc (Core Fuc), Fuca1-2Galb1-4GlcNAc (H-type 2) 6 AAL (Aleuria aurantia) Fuca1-3(Galb1-4)GlcNAc (Lewis x), Fuca1-6GlcNAc (Core Fuc) 7 MAL_I (Maackia amurensis) Siaa2-3Galb1-4GlcNAc 8 SNA (Sambucus nigra) Siaa2-6Gal/GalNAc 9 SSA (Sambucus sieboldiana) Siaa2-6Gal/GalNAc 10 TJA-I (Trichosanthes japonica) Siaa2-6Gal/GalNAc, HSO3(-) -6Galb1-4GlcNAc 11 PHAL (Phaseolus vulgaris) tri/tetra-antennary complex-type N-glycan 12 ECA (Erythrina cristagalli) Galb1-4GlcNAc (up with increasing the number of terminal Gal), no affinity for fully sialylated N-type, fully agalactosylated N-type 13 RCA120 (Ricinus communis) Galb1-4GlcNAc (up with increasing the number of terminal Gal), Galb1- 3Gal (weak), no affinity for agalactosylated N-type 14 PHAE (Phaseolus vulgaris) bi-antennary complex-type N-glycan with outer Gal and bisecting GlcNAc, no affinity for fully sialylated N-type 15 DSA (Datura stramonium) (GlcNAcb1-4)n, tri/tetra-antennary N-glycan 16 GSL-II (Griffonia simplicifolia) agalactosylated tri/tetra antennary glycans, GlcNAc, no affinity for fully galactosylated or sialylated N-type 17 NPA (Narcissus pseudonarcissus) High-Mannose including Mana1-6Man 18 ConA (Canavalia ensiformis) High-Mannose including Mana1-6(Mana1-3)Man 19 GNA (Galanthus nivalis) High-Mannose including Mana1-3Man 20 HHL (Hippeastrum hybrid) High-Mannose including Mana1-3Man or Mana1-6Man 21 ACG (mushroom, Agrocybe cylindracea) Galb1-3Gal, Siaa2-3Galb1-4GlcNAc 22 TxLCI (Tulipa gesneriana) Mana1-3(Mana1-6)Man, bi/tri-antennary complex-type N-glycan, GalNAc 23 BPL (Bauhinia purpurea) Galb1-3GalNAc (up with Lewis x, down with Core Fuc), GalNAc 24 TJA-II (Tanthes japonica) Fuca1-2Galb1-> or GalNAcb1-> groups at their non-reducing terminals 25 EEL (Euonymus europaeus) Gala1-3Galb1-4GlcNAc, Fuca1-2Galb1-3GlcNAc (H antigen) 26 ABA (fungus, Agaricus bisporus) Galb1-3GalNAc, GlcNAc 27 LEL (tomato, Lycopersicon esculentum) (GlcNAcb1-4)n (Chitin), (Galb1-4GlcNAc)n (polylactosamine) 28 STL (potato, Solanum tuberosum) (GlcNAcb1-4)n (Chitin), oligosaccharide containing GlcNAc and MurNAc 29 UDA (Urtica dioica) GlcNAcb1-4GlcNAc (Chitin), High-Mannose (3 to High, up with increasing the number of Man) 30 PWM (pokeweed, Phytolacca Americana) (GlcNAcb1-4)n (Chitin) 31 Jacalin (Artocarpus integrifolia) GlcNAcb1-3GalNAc (Core3), Siaa2-3Galb1-3GalNAc (sialyl T), Galb1- 3GalNAc (T-antigen), a-GalNAc (Tn-antigen) 32 PNA (peanut, Arachis hypogaea) Galb1-3GalNAc 33 WFA (Wisteria floribunda) GalNAcb1-4GlcNAc (LacdiNAc), Galb1-3(-6)GalNAc 34 ACA (Amaranthus caudatus) Galb1-3GalNAc (T-antigen), Siaa2-3Galb1-3GalNAc (sialyl T) 35 MPA (Maclura pomifera) a-GalNAc (Tn-antigen), Galb1-3GalNAc (T-antigen), 36 HPA (snail, Helix pomatia) a-GalNAc 37 VVA (Vicia villosa) GalNAcb1-4Gal, GalNAcb1-3Gal, a-GalNAc 38 DBA (Dolichos biflorus) Blood group A, GalNAca1-3GalNAc, GalNAcb1-4(Siaa2-3)Galb1-4Glc (GM2) 39 SBA (soybean, Dolichos biflorus) a-or b-linked GalNAc, Gala1-4Gal-Glc 40 Calsepa (Calystegia sepium) Galactosylated bianntenary N-type with bisecting GlcNAc (galacto > agalacto, down with Core Fuc), High-Mannose (Man2-6) 41 PTL-I (Psophocarpus tetragonolobus) a-GalNAc, Gala1-3(Fuca1-2)Gal (B-antigen) 42 MAH (Maackia amurensis) Siaa2-3Galb1-3(Siaa2-6)GalNAc (disialyl-T) 43 WGA (wheat germ, Triticum aestivum) (GlcNAcb1-4)n (Chitin), Hybrid type N-glycan, Sia 44 GSL-I A4 (Griffonia simplicifolia) a-Gal, a-GalNAc 45 GSL-I B4 (Griffonia simplicifolia) a-Gal, a-GalNAc LfDB database http://jcggdb.jp/rcmg/glycodb/LectinSearch interactions (Table 1), the lectin binding patterns indicate the (Vectibix ), an IgG2 isotype, displayed much weaker signal presence of core fucose, outer galactose, mannose, and GlcNAc intensities across the lectin chip despite a similar lectin binding oligomer, and the absence of sialic acids and tri/tetra-antennary pattern as observed for IgG1 mAbs (Fig. 2A). This data is con- complex-type N-glycans. Overall, the glycan profiles derived sistent with the reported lower level of glycan content in the 28,29 from lectin microarray are consistent with the reported glycans IgG2 antibodies in comparison with IgG1 antibodies. in IgG1 Fc that are known to contain principally bi-antennary Next, we tested 2 other therapeutic glycoproteins that are non-sialylated complex-type N-glycans with little or no high- associated with more complex glycosylation patterns. Cetuxi- mannose type or sialylation (Fig. 2C). Using the specified lec- mab (Erbitux ) was chosen because it contains N-glycosylation tin chips, we detected similar lectin binding patterns for IgG1 sites in both the antigen-binding fragment (Fab) and Fc of the mAbs containing only Fc glycosylation (Fig. 2B). Compared to molecule. In contrast to glycan profiles for IgG1 mAbs the other 5 mAbs, rituximab appeared to display relative higher (Fig. 2), cetuximab showed unique binding patterns at lectins binding signals at AOL/AAL (core fucose), RCA120 (terminal (SNA, SSA and TJA-I), which are known to bind a2-6-linked galactose), and GNA (high mannose). Panitumumab sialic acids (Fig. 3A & B). A binding signal was also detected at MABS 527 Table 2. Information of protein samples used in lectin microarray assay.* Number Proprietary name USAN name Class Expression system 1 Avastin Bevacizumab mAb CHO 2 Herceptin Trastuzumab mAb CHO 3 Humira Adalimumab mAb CHO 4 Remicade Infliximab mAb Sp2/0 5 Rituxan Rituximab mAb CHO 6 Xolair Omalizumab mAb CHO 7 Vectibix Panitumumab mAb CHO 8 Erbitux Cetuximab mAb Sp2/0 9 Enbrel Etanercept Fc-fusion protein CHO 10 Aranesp Darbepoetin alfa cytokine CHO 11 Pulmozyme Dornase alfa enzyme CHO 12 Elitek Rasburicase enzyme S. cerevisiae 13 Recombinant human transferrin, expressed in rice 14 Transferrin purified from human blood plasma 15 Neupogen Filgrastim cytokine E. coli Abbreviations used in this table: mAb, monoclonal antibody; CHO, Chinese Hamster Ovary cells; Sp2/0, murine myeloma cell line; S. cerevisiae, Saccharomyces cerevisiae; E. coli, Escherichia coli. 31,32 the a-Gal binding lectin GSL-I-A4, suggesting the presence bindings toward particular glycans in a testing sample. As of a-Gal structures in cetuximab proteins. Such a glycan variant noted, the rituximab Cy3-labeled Fab showed no lectin-binding was not detected in other samples tested in this study. Overall, signals across the lectin chips, confirming no or little interac- this data is consistent with the known glycan patterns of cetuxi- tion between the protein backbone and the lectins. mab Fab, which include an abundant N-linked sialic acid (Neu5Gc) and terminal a-Gal variants (Fig. 3C). The utility of lectin microarray in glycan profiling of Another sample tested was etanercept (Enbrel ), a homo- proteins produced by different host cell systems dimer of Fc-fusion protein consisting of TNF-a receptor and an IgG1 Fc portion, which was reported to contain 3 N-linked We assessed whether lectin microarray is capable of profiling and 13 O-linked glycosylation sites. Etanercept displayed dis- glycan variants of therapeutic proteins that are produced by dif- tinct lectin binding patterns compared to IgG1 mAbs. For ferent host cell systems such as mammalian cells, yeast strains example, strong binding signals were detected at MAL-I and and bacterial strains. These cell systems are different in their ACG, which are known to selectively bind a2-3 linked sialic glycosylation machinery, which produce proteins with distinct acid epitope. No signals were detected at lectins SNA, SSA or glycan patterns. For example, glycoproteins expressed by yeast TJA-1, showing the lack of a2-6 sialylation in the protein sam- strains usually contain high-mannose structures whereas ple. This data confirms the presence of complex glycans in Escherichia coli (E. coli) proteins are all non-glycosylated due to etanercept, including a2-3-sialic acids and abundant bi-antenn- the lack of glycosylation machinery in natural bacterial. We ary neutral glycans. Compared to other IgG1 mAbs, etanercept evaluated a panel of 6 proteins, including 2 therapeutic proteins displayed a strong and novel signal at MPA, a lectin that is produced by CHO cells (darbepoetin alfa (Aranesp ) and dor- known to selectively bind Galb1-3GalNAc and aGalNAc. nase alfa (Pulmozyme )), one therapeutic protein produced by This data is consistent with the reported abundance of O-gly- yeast (rasburicase (Elitek )), human transferrin proteins cans onto etanercept. expressed by recombinant rice strain or isolated from human To support the selectivity of lectin binding signals, we tested plasma, and filgrastim (Neupogen ) produced by E. coli the Fab and Fc purified from rituximab and cetuximab, respec- (Fig. 5A & B). Darbepoetin alfa showed strong signals at MAL- tively. Rituximab is known to contain only one N-glycosylation I, demonstrating the presence of a2-3-sialylation structures. site in its Fc. In the lectin microarray, the isolated rituximab Moreover, darbepoetin alfa displayed strong signals at PHAL- Fab showed little or no signals across the lectin chip whereas coated spots, which are known to be selective for tri-/tetra- the rituximab Fc displayed a similar lectin profile as intact rit- antennary N-glycan structures. Such a signal was not detected uximab (Fig. 4A). By contrast, cetuximab contains 2 N-glycans, in other proteins, confirming the absence of tri-/tetra-antenn- one located within its Fc portion and another in the Fab. The ary glycans in these samples. This pattern was consistent with uncommon a-Gal and Neu5Gc epitopes were reported to be the reported data that darbepoetin alfa contains high levels of solely in the Fab (Fig. 3C). The cetuximab Fc expressed a typi- sialylation and abundant tri- or tetra-antennary structures cal lectin profile for IgG1 Fc glycans (bi-antennary G0F, G1F, (Fig. 5C). Darbepoetin displayed relative week binding signals and G2F) (Fig. 4B). The lectin signals of SNA, SSA and TJA-1, at other lectin spots compared to MAL-I and PHAL, raising a which indicate expression of a2-6-sialylation, were only present possibility that those other glycan species (e.g., Galb1-4GlcNAc in intact cetuximab and cetuximab Fab profiles, but were absent and mannose oligomers) might be “capped” by the outermost in the Fc profile. GSL-I signal also indicated the presence of galactose and sialic acid. a-Gal structure in the Fab, but not in the Fc (Fig. 4B). These Dornase alfa (Pulmozyme ), a recombinant enzyme data not only confirm the proper locations of glycosylation sites expressed by CHO cells, displayed a simpler lectin binding pat- in Fc or Fab portions, but also support the selectivity of lectin tern compared to the above IgG1 and IgG2 mAbs (Fig. 5A and 528 L. ZHANG ET AL. Figure 2. Lectin binding profiles of therapeutic IgG monoclonal antibodies. The indicated therapeutic mAbs, including bevacizumab, trastuzumab, adalimumab, inflixi- mab, rituximab, omalizumab, and panitumumab were labeled with Cy3 and applied onto the lectin chips containing 45 distinct lectins with each being printed in tripli- cate. (A) Lectin binding images of the indicated samples. (B) Relative binding signals at specific lectin spots were derived from the images in A and normalized to protein markers on the same chip (mean § SD). Shown are representatives of 3 independent experiments. The coefficient of variation (CV) was determined to be < 10% for most lectin-glycan binding signals of the samples tested. (C) Typical glycan structures present in the Fc portion of therapeutic IgG1 mAbs. MABS 529 Figure 3. Lectin binding profiles of cetuximab and etanercept. Cy3-labeled samples were applied onto the lectin chips as in Figure 2. Shown are (A) representative lectin binding images, (B) Relative binding signals at specific lectin spots (mean § SD), and (C) Typical glycan structures present in the Fab portion of cetuximab. 5B). The sample tested showed unique binding signals at SNA/ demonstrating the absence of the relevant glycan species in ras- SSA for a2-6-sialylation, RCA120 for Galb1-4GlcNAc, DSA buricase. The two versions of human transferrin proteins also 37,38 for GlcNAc oligomer and/or Galb1-4GlcNAc, ConA for showed distinct glycan patterns in which the recombinant mannose, and LEL/STL for GlcNAc oligomers. The spectrum human transferrin expressed in rice (transferrin-rice) showed of selective binding signals suggests the presence of complex- binding signals primarily at mannose-binding lectin (NPA) type glycans with a2-6-sialylation in dornase alfa molecules. By and GlcNAc oligomer-binding lectins (LEL, STL and UDA). contrast, rasburicase (Elitek ), a therapeutic glycoprotein pro- The DSA signal indicated the presence of either GlcNAc duced by yeast strains, displayed distinctively different lectin oligomer or Galb1-4GlcNAc. By contrast, transferrin proteins profiles compared to the above described products produced by isolated from human plasma showed additional signals at a2- mammalian cells. Rasburicase showed relatively weak binding 6-sialic acid-binding lectins (SNA, SSA, and TJA-I) and galac- signals across the lectin chips, which is consistent with its tose-binding lectins (RCA120 and PHAE). As expected, no known low level of glycosylation. Despite the overall weak lectin binding signals were detected for filgrastim (Neupogen ) binding signals, rasburicase appeared to interact exclusively that is produced by E. coli as a non-glycosylated protein. with mannose binding lectins (NPA, ConA, and GNA) and GlcNAc oligomer binding lectins (STL and UDA). This data The utility of lectin microarray in monitoring terminal confirms the presence of high-mannose carbohydrates that are galactosylation and sialylation of glycoproteins mainly attached onto glycoproteins produced by yeast strains. No binding signals were detected at sialic acid-binding lectins To further evaluate the utility of lectin microarray in glycan (e.g., MAL_I, SNA, SSA, and TJA-I), fucose-binding lectins profiling, we prepared protein variants with defined galactosy- (e.g., PSA and LCA) or galactose-binding lectins (e.g., RCA 120 lation and sialylation modifications. This was achieved through and PHAE), even when the protein concentration of rasburi- in vitro enzymatic glycoengineering of rituximab using com- case was enhanced to 500 ng/mL (data not shown), mercially available galactosyltransferase and sialyltransferase. 530 L. ZHANG ET AL. Figure 4. Glycan profiling of Fabs and Fcs. The Fabs and Fcs of rituximab (A) and cetuximab (B) were prepared as described in Materials and Methods. Purified Fab and Fc were analyzed by reducing SDS-PAGE (left panel) and lectin microarray (right panel). As noted, the dimeric Fcs (~55 kDa) were reduced to monomeric products (~30 kDa) on SDS-PAGE under reducing conditions. b1-4-galactosyltransferase (b1-4GalT) catalyzes the transfer of signals at positions of ECA and RCA120. Both lectins are galactose from donor substrate UDP-galactose (UDP-Gal) to known to bind N-glycan Galb1-4GlcNAc, and therefore the GlcNAcb1-2Man units of glycoproteins to form a b1-4-galacto- incurred ECA lectin-binding signal and the significantly sylation linkage, while a2-6-sialyltransferase (a2-6SiaT) facili- enhanced RCA120 signal were indicative of the increased galac- tates sialylation by adding sialic acids to terminal Galb1- tose species in the samples. Notably, there was a concomitant 4GlcNAc units. Modified rituximab protein variants were puri- decrease in the binding signals at ABA, a lectin with dual bind- fied and then characterized using mass spectrometry (MS), ing affinity toward Gal-exposed O-glycans and GlcNAc- revealing distinct deconvoluted MS spectra for the light chain exposed N-glycans. Because rituximab was not reported to and heavy chain (Fig. 6A). The light chain fragments resolved undergo O-glycosylation, the ABA binding signal detected for as a single species at an average mass of 23036 Da, correspond- native rituximab was likely due to GlcNAc-exposed N-glycans. 42,43 ing to the theoretical mass of rituximab light chain. Consis- Such GlcNAc-exposed N-glycans appeared to be occupied tent with the lack of glycosylation sites within the rituximab upon galactosylation catalyzed in the reactions with b1-4GalT. light chains, the mass of light chain remained unchanged after Samples of native rituximab and galactosylated species showed treatments of rituximab with b1-4GalT or further with a2- no detectable signals at the lectins SNA, SSA, TJA-I that are 6SiaT. The other 3 major mass species at 50507, 50669, 50832 known to bind sialic acids. By contrast, strong binding signals Da correspond to the heavy chains of rituximab containing at these lectins were detected for the samples derived from the 42,43 G0F, G1F or G2F glycoforms, respectively (Fig. 2C). Treat- sequential reactions with b1-4GalT and a2-6SiaT enzymes. No ment of rituximab with b1-4GalT resulted in a mass shift from signal was detected at MAL-I (a lectin selective to a2-3-sialyla- G0F and G1F to G2F, indicating galactosylation reactions were tion), demonstrating not only the specificity of glycan engineer- effectively accomplished. When the galactosylated rituximab ing but also the utility of the lectin microarray in distinguishing mixture was sequentially treated with a2-6SiaT, the final prod- different terminal sialic acid linkages. uct showed a further mass shift from 50832 Da (G2F) to 51122 Da (C290) and 51414 Da (C582). These mass shifts corre- Discussion sponded to an addition of one or two Neu5Ac residues, indicat- ing that G2F glycoform was effectively converted to primarily The complexity of glycosylation poses an analytical challenge in mono-sialylated species (S1G2F) and a small portion of di-sia- the development of therapeutic glycoproteins. The methods lylated species (S2G2F). These data indicate that the in vitro most commonly used for analysis include MS, HPLC, HPAEC- glycan engineering reactions produced rituximab variants con- PAD, and CE. Among these methods, MS remains a powerful taining the desired modifications (e.g., galactosylation and tool in the characterization of glycosylation site(s) occupancy sialylation). and carbohydrate structures. MS-based methods involve enzy- The engineered rituximab samples with defined glycan var- matic digestion of a glycoprotein into peptide fragments and iations were then analyzed by lectin microarray (Fig. 6B). The separation by liquid chromatography. HPLC-, HPAEC- and reaction buffer alone had no effect on lectin binding profile of CE-based methods usually require the release of glycans from a rituximab. The sample resulting from b1-4GalT reaction (rit- glycoprotein through enzymatic or chemical reactions. As uximab C b1-4GalT), which was confirmed by MS to majorly such, an accurate assessment of glycosylation requires a com- contain G2F galactosylation (Fig. 6A), displayed strong binding plete release of all glycans that are present in a glycoprotein MABS 531 Figure 5. Glycan profiles of proteins produced by different host cell systems. The proteins tested include therapeutic proteins produced by CHO cells (darbepoetin alfa and dornase alfa), yeast strains (rasburicase), or E. coli (filgrastim), and human transferrin protein expressed by recombinant rice (transferrin-rice) or isolated from human plasma (transferrin-human). (A) Lectin binding images. (B) Relative binding signals at specific lectin spots (mean § SD). (C) Typical glycan structures present in darbepoe- tin alfa. being tested. By contrast, lectin microarray directly measures commercial lectin chips, we were able to determine glycan pro- glycan profiles on an intact protein without the need for enzy- files for a panel of therapeutic proteins that were generally con- matic digestion or clipping glycans from the protein backbone. sistent with their known glycosylation properties (Figs. 2-5). Such a platform is unique in increasing the likelihood of full Notably the lectin microarray was highly sensitive to alterations coverage of all glycan variants of a glycoprotein. Using the in the terminal glycan structures, i.e., galactosylation vs. 532 L. ZHANG ET AL. Figure 6. Assessing glycan variants in glycan-engineered rituximab protein samples. Rituximab was incubated with a reaction buffer alone (rituximab C buffer), b1-4-gal- actosyltransferase (rituximab C b1-4GalT), and b1-4-galactosyltransferase followed by a2-6-sialyltransferase (rituximab C b1-4GalT C a2-6SiaT) (see detail in Materials and Methods). After affinity purification, the resulting samples were analyzed using mass spectrometry and lectin microarray, respectively. Shown are representatives of deconvoluted mass spectra (A) and corresponding lectin binding profiles (B) for the samples produced under the indicated conditions. MABS 533 sialyation (Fig. 6). Lectin microarray can effectively distinguish 0.4 mg/mL using Fab digestion buffer, and aliquot of 120 mL glycan isomers containing different sialic acid linkages. Our was added to the spin column containing equilibrated immobi- data demonstrate a usefulness of the lectin microarray in lized papain. The digestion reaction was incubated overnight at screening glycan patterns of protein samples. As noted, the 37 C. Digested samples were collected by column centrifuga- commercial lectin chips employed in this study was not tion and applied onto NAb Protein A Plus Spin Column to sep- designed specifically for assessing therapeutic mAbs. In princi- arate the Fab and Fc. Protein concentrations of the resulting ple, assay performance could be further improved through the samples were determined using Micro BCA Protein Assay Kit use of lectins with improved selectivity and binding affinity to (Thermo Scientific). distinct glycan species. To analyze a specific glycoprotein, lectin chips can be customized to include the lectins that are relevant Enzymatic glycan engineering of rituximab to the glycan species that are possibly present in the testing sample. Upon optimization of lectin chips, lectin microarray Rituximab was used to create glycan variants of a mAb through platform could be adopted as a complementary tool for high in vitro enzymatic reactions. Briefly, rituximab stock solution throughput screening of glycan profiles of therapeutic was diluted to 2 mg/mL using 1£ galactosyltransferase (GalT) glycoproteins. reaction buffer (10 mM MnCl with 100 mM MES, pH 6.5). Uridine-5’-diphosphogalactose disodium salt (UDP-Gal) solu- tion (50 mg/mL) was also prepared using 1£ GalT reaction Materials and methods buffer. Cytidine 5’-monophospho-N-acetylneuraminic acid disodium salt (CMP-Neu5Ac) solution (50 mg/mL) was pre- Therapeutic proteins and reagents pared using 1£ sialyltransferase (SiaT) reaction buffer (50 mM All therapeutic proteins were purchased from the Division of Tris-Acetate, pH 7.5). To facilitate galactosylation, 50 mL ritux- Veterinary Resources pharmacy, Office of Research Services, imab (2 mg/mL in solution) was mixed with 4 mL UDP-Gal National Institute of Health, and handled according to respec- (50 mg/mL in solution) and 1.5 mg b1-4-galactosyltransferase tive package inserts. Recombinant human transferrin expressed (b1-4GalT). The final reaction volume was adjusted to 100 mL by rice and transferrin purified from human blood plasma were using 1£ GalT reaction buffer, and was left to incubate over- purchased from Sigma-Aldrich. Cy3 mono-reactive dye was night at 37 C, resulting galactosylation of rituximab. To further obtained from GE Healthcare. Zeba spin desalting columns add sialylation, the above galactosylated rituximab was mixed (7K MWCO) were purchased from Thermo Scientific. Lectin with 200 mg CMP-Neu5Ac (50 mg/mL in solution) and 10 mg chips coated with 45 distinct lectin proteins and the probing a2-6-sialyltransferase (a2-6SiaT). After adjusting volume to solutions were obtained from GlycoTechnica. Glycosyltransfer- 200 mL using 1£ SiaT reaction buffer, the mixture was incu- ases b1-4-galactosyltransferase (b1-4GalT) and a2-6-sialyl- bated at 37 C for 3 hours and then frozen at ¡20 C. The engi- transferase (a2-6SiaT) were purchased from Prozyme. Micro neered antibodies containing different glycan variants were BCA Protein Assay Kit, NAb Protein A Plus Spin Kit and purified using NAb Protein A Plus Spin Kit by following the Pierce Fab micro preparation kit were obtained from Thermo protocol instructions (Thermo Scientific). Scientific. NuPAGE 4-12% Bis-Tris gels and Coomassie stain (SimplyBlue SafeStain) were purchased from Life Technologies. Liquid chromatography–mass spectrometry analysis Liquid chromatography–mass spectrometry analyses were con- Lectin microarray analysis ducted on an Agilent 1260 HPLC-Chip nano-electrospray-ioni- Protein samples were diluted to 50 mg/mL in phosphate-buff- zation 6520 Q-TOF MS system. Solvent-A was 0.1% formic ered saline (PBS). Aliquots (20 mL, 1 mg protein) were mixed acid in water and solvent-B was 0.1% formic acid in 95% aceto- with 100 mg Cy3 mono-reactive dye, and incubated at room nitrile. Mass correction was enabled during the run using inter- temperature for 2 h. The unbound Cy3 dye was removed using nal reference ions with masses of 299.2945 and 1221.9906 Da. Zeba spin desalting columns (7K MWCO). Cy3-labeled protein Intact protein mass measurement was performed using an Agi- samples were diluted to 125 ng/mL in probing solution, and an lent 43 mm 300 A C8 chip with a 40 nL trap column (G4240- aliquot of 100 mL was applied onto a lectin chip, and incubated 63001 SPQ105). All rituximab samples were reduced in 10 mM at room temperature on an orbital shaker for 18 hours in dark. DTT for 1 hour at 37 C, and then formic acid was added to The resulting lectin chips were scanned for fluorescence inten- reach a final concentration of 0.1% (V/V). The samples were sities on each lectin-coated spot using an evanescent-field fluo- diluted to 150 ng/mL in 0.1% formic acid, and 1 mL(~1 pmol) or rescence scanner GlycoStation Reader 1200 (GlycoTechnica, larger volume was injected onto the trap column in the C8 chip Japan). at a flow rate of 2.5 mL/min in 100% A, then eluted at 0.5 mL/ min with a linear gradient of 10–100% B in 18 min and held for additional 4 min. Q-TOF VCap, fragmentor, and skimmer Preparation of IgG Fab and Fc settings were 1,890 eV, 225 eV, and 65 eV, respectively. HPLC- The Fab and Fc of therapeutic mAbs were prepared using Chip gas temperature and drying gas flow rate were 350 C and Pierce Fab micro preparation kit following the vendor’s proto- 9 L/min, respectively. The data were analyzed using Agilent col (Thermo Scientific). Briefly, immobilized papain (settled MassHunter (version B.05.00) Qualitative Analysis software resin) was located in a 0.8 mL spin column and equilibrated and deconvoluted with Agilent MassHunter Bioconfirm using Fab digestion buffer. Antibody samples were diluted to software. 534 L. ZHANG ET AL. Disclosure of potential conflicts of interest 18. Hirabayashi J, Yamada M, Kuno A, Tateno H. Lectin microarrays: concept, principle and applications. Chem Soc Rev 2013; 42:4443-58; No potential conflicts of interest were disclosed. PMID:23443201; http://dx.doi.org/10.1039/c3cs35419a 19. Cook MC, Kaldas SJ, Muradia G, Rosu-Myles M, Kunkel JP. Compari- son of orthogonal chromatographic and lectin-affinity microarray methods for glycan profiling of a therapeutic monoclonal antibody. J Chromatogr B 2015; 997:162-78; PMID:26114652; http://dx.doi.org/ References 10.1016/j.jchromb.2015.05.035 20. Huang WL, Li YG, Lv YC, Guan XH, Ji HF, Chi BR. Use of lectin 1. Kornfeld R, Kornfeld S. Assembly of asparagine-linked oligosacchar- microarray to differentiate gastric cancer from gastric ulcer. World J ides. Annu Rev Biochem 1985; 54:631-64; PMID:3896128; http://dx. Gastroenterol 2014; 20:5474-82; PMID:24833877; http://dx.doi.org/ doi.org/10.1146/annurev.bi.54.070185.003215 10.3748/wjg.v20.i18.5474 2. Ghaderi D, Zhang M, Hurtado-Ziola N, Varki A. Production plat- 21. Kobayashi Y, Masuda K, Banno K, Kobayashi N, Umene K, Nogami forms for biotherapeutic glycoproteins. Occurrence, impact, and chal- Y, Tsuji K, Ueki A, Nomura H, Sato K, et al. Glycan profiling of gesta- lenges of non-human sialylation. Biotechnol Genet Eng Rev 2012; tional choriocarcinoma using a lectin microarray. Oncol Rep 2014; 28:147-75; PMID:22616486; http://dx.doi.org/10.5661/bger-28-147 31:1121-6; PMID:24424471; http://dx.doi.org/10.3892/or.2014.2979 3. Sola RJ, Griebenow K. Glycosylation of therapeutic proteins: an effec- 22. Xin AJ, Cheng L, Diao H, Wang P, Gu YH, Wu B, Wu YC, Chen GW, tive strategy to optimize efficacy. BioDrugs 2010; 24:9-21; Zhou SM, Guo SJ, et al. Comprehensive profiling of accessible surface PMID:20055529; http://dx.doi.org/10.2165/11530550-000000000- glycans of mammalian sperm using a lectin microarray. Clin Proteo- mics 2014; 11:10; PMID:24629138; http://dx.doi.org/10.1186/1559- 4. Sola RJ, Griebenow K. Effects of glycosylation on the stability of pro- 0275-11-10 tein pharmaceuticals. J Pharm Sci 2009; 98:1223-45; PMID:18661536; 23. Gupta G, Surolia A, Sampathkumar SG. Lectin microarrays for glyco- http://dx.doi.org/10.1002/jps.21504 mic analysis. OMICS 2010; 14:419-36; PMID:20726799; http://dx.doi. 5. Narhi LO, Arakawa T, Aoki KH, Elmore R, Rohde MF, Boone T, org/10.1089/omi.2009.0150 Strickland TW. The effect of carbohydrate on the structure and stabil- 24. Song T, Ozcan S, Becker A, Lebrilla CB. In-depth method for the char- ity of erythropoietin. J Biol Chem 1991; 266:23022-6; PMID:1744097 acterization of glycosylation in manufactured recombinant monoclo- 6. Desnick RJ, Schuchman EH. Enzyme replacement therapy for lyso- nal antibody drugs. Anal Chem 2014; 86:5661-6; PMID:24828102; somal diseases: lessons from 20 years of experience and remaining http://dx.doi.org/10.1021/ac501102t challenges. Annu Rev Genomics Hum Genet 2012; 13:307-35; 25. Stadlmann J, Pabst M, Kolarich D, Kunert R, Altmann F. Analysis of PMID:22970722; http://dx.doi.org/10.1146/annurev-genom-090711- immunoglobulin glycosylation by LC-ESI-MS of glycopeptides and oligosaccharides. Proteomics 2008; 8:2858-71; PMID:18655055; 7. Tsuda E, Kawanishi G, Ueda M, Masuda S, Sasaki R. The role of car- http://dx.doi.org/10.1002/pmic.200700968 bohydrate in recombinant human erythropoietin. Eur J Biochem 26. Visser J, Feuerstein I, Stangler T, Schmiederer T, Fritsch C, Schiestl M. 1990; 188:405-11; PMID:2156701; http://dx.doi.org/10.1111/j.1432- Physicochemical and functional comparability between the proposed 1033.1990.tb15417.x biosimilar rituximab GP2013 and originator rituximab. BioDrugs 8. Morell AG, Gregoriadis G, Scheinberg IH, Hickman J, Ashwell G. The 2013; 27:495-507; PMID:23649935; http://dx.doi.org/10.1007/s40259- role of sialic acid in determining the survival of glycoproteins in the 013-0036-3 circulation. J Biol Chem 1971; 246:1461-7; PMID:5545089 27. Huhn C, Selman MH, Ruhaak LR, Deelder AM, Wuhrer M. IgG gly- 9. Houdebine LM. Production of pharmaceutical proteins by transgenic cosylation analysis. Proteomics 2009; 9:882-913; PMID:19212958; animals. Comp Immunol Microbiol Infect Dis 2009; 32:107-21; http://dx.doi.org/10.1002/pmic.200800715 PMID:18243312; http://dx.doi.org/10.1016/j.cimid.2007.11.005 28. Wuhrer M, Stam JC, van de Geijn FE, Koeleman CA, Verrips CT, 10. Nimmerjahn F, Ravetch JV. Translating basic mechanisms of IgG Dolhain RJ, Hokke CH, Deelder AM. Glycosylation profiling of effector activity into next generation cancer therapies. Cancer Immun immunoglobulin G (IgG) subclasses from human serum. Proteomics 2012; 12:13; PMID:22896758 2007; 7:4070-81; PMID:17994628; http://dx.doi.org/10.1002/ 11. Marino K, Bones J, Kattla JJ, Rudd PM. A systematic approach to pro- pmic.200700289 tein glycosylation analysis: a path through the maze. Nat Chem Biol 29. Perdivara I, Peddada SD, Miller FW, Tomer KB, Deterding LJ. Mass 2010; 6:713-23; PMID:20852609; http://dx.doi.org/10.1038/ spectrometric determination of IgG subclass-specific glycosylation nchembio.437 profiles in siblings discordant for myositis syndromes. J Proteome Res 12. Beck A, Wagner-Rousset E, Ayoub D, Van DA, Sanglier-Cianferani S. 2011; 10:2969-78; PMID:21609021; http://dx.doi.org/10.1021/ Characterization of therapeutic antibodies and related products. Anal pr200397h Chem 2013; 85:715-36; PMID:23134362; http://dx.doi.org/10.1021/ 30. Qian J, Liu T, Yang L, Daus A, Crowley R, Zhou Q. Structural charac- ac3032355 terization of N-linked oligosaccharides on monoclonal antibody 13. Higgins E. Carbohydrate analysis throughout the development of a cetuximab by the combination of orthogonal matrix-assisted laser protein therapeutic. Glycoconj J 2010; 27:211-25; PMID:19888650; desorption/ionization hybrid quadrupole-quadrupole time-of-flight http://dx.doi.org/10.1007/s10719-009-9261-x tandem mass spectrometry and sequential enzymatic digestion. Anal 14. Lingg N, Zhang P, Song Z, Bardor M. The sweet tooth of biopharma- Biochem 2007; 364:8-18; PMID:17362871; http://dx.doi.org/10.1016/j. ceuticals: importance of recombinant protein glycosylation analysis. ab.2007.01.023 Biotechnol J 2012; 7:1462-72; PMID:22829536; http://dx.doi.org/ 31. Murphy LA, Goldstein IJ. Five alpha-D-galactopyranosyl-binding iso- 10.1002/biot.201200078 lectins from Bandeiraea simplicifolia seeds. J Biol Chem 1977; 15. Pabst M, Altmann F. Glycan analysis by modern instrumental meth- 252:4739-42; PMID:68957 ods. Proteomics 2011; 11:631-43; PMID:21241022; http://dx.doi.org/ 32. Lescar J, Loris R, Mitchell E, Gautier C, Chazalet V, Cox V, Wyns L, 10.1002/pmic.201000517 Perez S, Breton C, Imberty A. Isolectins I-A and I-B of Griffonia (Ban- 16. Rohrer JS, Basumallick L, Hurum D. High-performance anion- deiraea) simplicifolia. Crystal structure of metal-free GS I-B(4) and exchange chromatography with pulsed amperometric detection for molecular basis for metal binding and monosaccharide specificity. J carbohydrate analysis of glycoproteins. Biochemistry (Mosc ) 2013; Biol Chem 2002; 277:6608-14; PMID:11714720; http://dx.doi.org/ 78:697-709; PMID:24010833; http://dx.doi.org/10.1134/ 10.1074/jbc.M109867200 S000629791307002X 33. Tan Q, Guo Q, Fang C, Wang C, Li B, Wang H, Li J, Guo Y. Charac- 17. Varki A, Etzler ME, Cummings RD, Esko JD. Discovery and Classifi- terization and comparison of commercially available TNF receptor 2- cation of Glycan-Binding Proteins. In: Varki A, Cummings RD, Esko Fc fusion protein products. MAbs 2012; 4:761-74; PMID:23032066; JD, et al. Essentials of Glycobiology. 2nd edition. Cold Spring Harbor http://dx.doi.org/10.4161/mabs.22276 (NY): Cold Spring Harbor Laboratory Press; 2009. Chapter 26 MABS 535 34. Houel S, Hilliard M, Yu YQ, McLoughlin N, Martin SM, Rudd PM, Hereditas 1996; 124:251-9; PMID:8931358; http://dx.doi.org/10.1111/ Williams JP, Chen W. N- and O-glycosylation analysis of etanercept j.1601-5223.1996.00251.x using liquid chromatography and quadrupole time-of-flight mass 39. Higgins E. Carbohydrate analysis throughout the development of a spectrometry equipped with electron-transfer dissociation functional- protein therapeutic. Glycoconj J 2010; 27:211-25; PMID:19888650; ity. Anal Chem 2014; 86:576-84; PMID:24308717; http://dx.doi.org/ http://dx.doi.org/10.1007/s10719-009-9261-x 10.1021/ac402726h 40. Hossler P, Khattak SF, Li ZJ. Optimal and consistent protein glycosyl- 35. Huang J, Xu Z, Wang D, Ogata CM, Palczewski K, Lee X, Young NM. ation in mammalian cell culture. Glycobiology 2009; 19:936-49; Characterization of the secondary binding sites of Maclura pomifera PMID:19494347; http://dx.doi.org/10.1093/glycob/cwp079 agglutinin by glycan array and crystallographic analyses. Glycobiology 41. Jaffe SR, Strutton B, Levarski Z, Pandhal J, Wright PC. Escherichia coli 2010; 20:1643-53; PMID:20826825; http://dx.doi.org/10.1093/glycob/ as a glycoprotein production host: recent developments and chal- cwq118 lenges. Curr Opin Biotechnol 2014; 30:205-10; PMID:25156401; 36. Shahrokh Z, Royle L, Saldova R, Bones J, Abrahams JL, Artemenko http://dx.doi.org/10.1016/j.copbio.2014.07.006 NV, Flatman S, Davies M, Baycroft A, Sehgal S, et al. Erythropoietin 42. Huang W, Giddens J, Fan SQ, Toonstra C, Wang LX. Chemoenzy- produced in a human cell line (Dynepo) has significant differences in matic glycoengineering of intact IgG antibodies for gain of functions. glycosylation compared with erythropoietins produced in CHO cell J Am Chem Soc 2012; 134:12308-18; PMID:22747414; http://dx.doi. lines. Mol Pharm 2011; 8:286-96; PMID:21138277; http://dx.doi.org/ org/10.1021/ja3051266 10.1021/mp100353a 43. Wang B, Gucinski AC, Keire DA, Buhse LF, Boyne MT. Structural 37. Crowley JF, Goldstein IJ, Arnarp J, Lonngren J. Carbohydrate binding comparison of two anti-CD20 monoclonal antibody drug products studies on the lectin from Datura stramonium seeds. Arch Biochem using middle-down mass spectrometry. Analyst 2013; 138:3058-65; Biophys 1984; 231:524-33; PMID:6203486; http://dx.doi.org/10.1016/ PMID:23579346; http://dx.doi.org/10.1039/c3an36524g 0003-9861(84)90417-X 44. Nakamura-Tsuruta S, Kominami J, Kuno A, Hirabayashi J. Evidence 38. Myllyharju J, Nokkala S. Glycoproteins with N-acetylglucosamine and that Agaricus bisporus agglutinin (ABA) has dual sugar-binding speci- mannose residues in Chinese hamster metaphase chromosomes. ficity. Biochem Biophys Res Commun 2006; 347:215-20; PMID:16824489; http://dx.doi.org/10.1016/j.bbrc.2006.06.073

Journal

mAbsTaylor & Francis

Published: Apr 2, 2016

Keywords: Glycan analysis; lectin microarray; monoclonal antibodies; therapeutic glycoproteins

There are no references for this article.