Fast and Simple Qualitative/Semi‐Quantitative Analysis of Monoclonal Antibody Mixtures Using Liquid Chromatography–Electrospray Triple Time‐of‐Flight Mass Spectrometry

• Published in: Bulletin of the Korean Chemical Society Vol. 39; no. 12; pp. 1380 - 1384
• Main Authors: Byeon, Jin‐Ju, Park, Min‐Ho, Shin, Seok‐Ho, Shin, Young G.
• Format: Journal Article
• Language: English
• Published: Weinheim Wiley‐VCH Verlag GmbH & Co. KGaA 01.12.2018; 대한화학회
• Subjects: Bispecific antibody; Liquid chromatography–electrospray triple time‐of‐flight mass spectrometry; Monoclonal antibody mixture; Qualitative/semi‐quantitative assay; 화학
• ISSN: 1229-5949; 0253-2964; 1229-5949
• DOI: 10.1002/bkcs.11610

Bispecific antibodies are generally prepared by co‐expression of mixtures of recombinant monoclonal antibodies. Because of increased molecular complexity, characterization of a desired bispecific antibody or a mixture of monoclonal antibodies is more challenging than characterization of a conventional single monoclonal antibody. The purpose of this study is to develop a fast and simple method for qualitative/semi‐quantitative analysis of antibody mixtures using liquid chromatography–electrospray triple time‐of‐flight mass spectrometry (LC‐ESI‐TOF/MS) to complement the enzyme‐linked immunosorbent assay. To demonstrate the proof of concept for the analysis of antibody mixtures, three different tool monoclonal antibodies (trastuzumab, rituximab, and cetuximab) with various mixture ratios were treated by either PNGase or Fabricator to simplify the antibody structures without glycans. After deglycosylation, the mixtures of antibodies were analyzed by LC‐ESI‐TOF/MS in the positive ion mode. The m/z scan range of 2000–4000 was used for the deconvolution of each peak from antibodies. Because each antibody could show different ionization efficiency in TOF MS, the peak intensities obtained from various mixture antibodies (1:6:3 or 3:1:6 or 6:3:1) were normalized by the peak intensities of 1:1:1 mixture of three antibodies. Overall, two different methods treated by either PNGase or Fabricator were comparable in estimating the mixture ratios; however, the accuracy and precision data from the Fabricator group were slightly better than PNGase group possibly due to the generation of smaller fragments by Fabricator.