Influence of resin structure on the prediction of two-component protein adsorption behavior in anion exchange resins from single component batch data
Year: 2025
Type: article
Abstract: • Resin structure influences the accuracy of two-component separation predictions. • Q Sepharose FF data predictable with multicomponent Langmuir isotherm. • Langmuir model overestimates GFP capacity on POROS twofold in mixture adsorption. • Step-by-step development of mechanistic models accounting for structure of POROS. • Reversible-irreversible model predicts POROS data with <15% average deviation. Understanding the adsorption equilibria and kinetics of multicomponent mixtures is essential to developing and modeling separation processes. In our previous research, we demonstrated the significant impact of resin structure on the efficiency with which two components are separated in frontal chromatography. This study aims to predict the adsorption behavior of multicomponent binary protein mixtures using single-component data on anion exchange resins. Conalbumin (CA) and green fluorescent protein (GFP) were used as model proteins to study two resins with distinct structures: Q Sepharose FF and POROS 50 HQ. Q Sepharose FF has a homogeneous pore size distribution (r pore ∼ 30 nm), while POROS exhibits a bimodal distribution with macropores (r pore ∼ 200-500 nm) and micropores (r pore ∼ 20 nm). A pore diffusion model with a competitive Langmuir isotherm was used to predict binary adsorption behavior using single-component isotherm and batch uptake kinetics data. While this model accurately described adsorption on Q Sepharose FF, it failed to capture the reduced GFP binding capacity observed in frontal chromatography on POROS 50 HQ by a factor of 2. To address this, two alternative models were proposed: (1) a competitive bi-Langmuir isotherm assuming heterogeneous binding sites and (2) a reversible-irreversible binding model where micropore sites exhibit irreversible adsorption. The reversible-irreversible model provided the most accurate predictions for POROS 50 HQ, capturing both equilibrium and kinetic data with an average deviation below 15%. These findings highlight the role of POROS's bimodal pore structure in reduced GFP capacity and demonstrate the need for advanced mechanistic models to predict multicomponent adsorption in complex systems.
Source: Journal of Chromatography A
Authors: Jürgen Beck, Rainer Hahn, Giorgio Carta
Institutions: BOKU University, University of Virginia
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $4,040
Funder: Universität für Bodenkultur Wien
OpenAlex ID: W4415528985