Related Experiment Video
Updated: Oct 26, 2026

Laboratory Scale Production and Purification of a Therapeutic Antibody
Published on: January 24, 2017
Electrostatic inversion in pH-gradient antigen affinity chromatography: The selectivity bias for deamidated
Steffen Lippold1, Skanda Sastry1, Kevin Whang2
1Analytical Development, Genentech, A Member of the Roche Group, 1 DNA Way, South San Francisco, California, 94080, United States.
Abstract:
Affinity chromatography hyphenated to mass spectrometry (AC-MS) has emerged as a powerful technology for the simultaneous structural and functional characterization of therapeutic antibodies. However, the reliance on pH gradients in AC raises fundamental questions regarding the accurate ranking of proteoforms, in particular for target interactions requiring low pH ranges for elution. Here, we investigated the selectivity bias of light chain asparagine 30 deamidated proteoforms (LC-N30D) of trastuzumab in target AC-MS using the immobilized extra-cellular domain of human epidermal growth factor receptor 2 (HER2). By employing a middle-up strategy and cation exchange chromatography enrichment of Fab fragments, we successfully isolated and characterized enriched proteoforms. While HER2 AC-MS identified decreased binding of heavy chain aspartic acid 102 isomerization, the enriched LC-N30 deamidated (LC-N30D) Fab proteoform exhibited an apparent increased binding compared to Fab proteoforms with low modification content. Orthogonal surface plasmon resonance kinetics revealed a critical inversion in the binding behavior due to differences in mobile phase pH. While LC-N30D substantially reduces binding at physiological pH, the acidic separation conditions required for elution in AC, stabilized the HER2 interaction. Molecular dynamics simulations attributed the HER2 AC retention behavior to a protonation-dependent electrostatic stabilization of the LC-N30D-HER2 interface at low pH, which favored the LC-N30D Fab proteoform binding. These findings highlight a major limitation in pH-gradient AC, demonstrating that the elution mechanism may bias the translation to functional relevance. This study underscores the necessity of orthogonal validation when employing AC-MS for critical quality attribute assessment of biologics.

