A priori prediction of adsorption isotherm parameters and chromatographic behavior in ion-exchange systems

Abstract

The a priori prediction of protein adsorption behavior has been a long-standing goal in several fields. In the present work, propertymodeling techniques have been used for the prediction of protein adsorption thermodynamics in ion-exchange systems directly from crystal structure. Quantitative structure-property relationship models of protein isotherm parameters and Gibbs free energy changes in ion-exchange systems were generated by using a support vector machine regression technique. The predictive ability of the models was demonstrated for two test-set proteins not included in the model training set. Molecular descriptors selected during model generation were examined to gain insights into the important physicochemical factors influencing stoichiometry, equilibrium, steric effects, and binding affinity in protein ion-exchange systems. The a priori prediction of protein isotherm parameters can have direct implications for various ion-exchange processes. As proof of concept, a multiscale modeling approach was used for predicting the chromatographic separation of a test set of proteins using the isotherm parameters obtained from the quantitative structure-property relationship models. The simulated column separation showed good agreement with the experimental data. The ability to predict chromatographic behavior of proteins directly from their crystal structures may have significant implications for a range of biotechnology processes.ion-exchange chromatography ͉ protein adsorption ͉ structure-property relationships ͉ steric mass action ͉ support vector machines P rotein adsorption on surfaces is of fundamental importance in a wide array of applications, including biomedical implants (1), tissue engineering (2), food processing (3), biosensors (4), and bioseparations (5). In particular, protein adsorption on ionexchange surfaces has direct implications for bioseparation systems ranging from preparative chromatographic and membrane separations to high-throughput chromatographic and electrophoresisbased proteomic applications (6-8).Although protein adsorption in ion-exchange systems is dominated by Coulombic interactions, other molecular interactions such as hydrophobic and van der Waals interactions can play an important role in creating multimodal interactions and unique selectivities (9-13). Because protein binding in ion-exchange chromatography has been shown to be a reversible phenomenon with no significant associated conformational changes (14), the treatment of the protein as a colloid that reversibly adsorbs and desorbs is reasonable for modeling ion-exchange chromatographic systems. Several mechanistic models have attempted to model protein binding in ion exchange. Roth and Lenhoff (9) have computed the electrostatic and van der Waals energies of interaction between a colloidal protein molecule and a planar charged surface at a fixed distance and orientation. Roth et al. (10) The steric mass action (SMA) formalism (15, 16) has been used successfully to predict complex preparative chromatographic behavior ...

Impact of denaturation with urea on recombinant apolipoprotein A‐IMilano ion‐exchange adsorption: Equilibrium uptake behavior and protein mass transfer kinetics

“…The characteristics of these charged surface regions and Table 2 and D e values for 0 M, 2 M, and 6 M urea from Table 3. charge distribution can have a dramatic influence on chromatographic behavior [33][34][35]. Upon denaturation, the protein no longer has a surface in the classic sense, and binding of charged residues to the stationary phase may be modulated by the proximity of oppositely charged residues in the primary structure of the protein.…”

Investigation of protein binding affinity and preferred orientations in ion exchange systems using a homologous protein library

“…The application of this type of feature selection serves to improve the computational signal-to-noise ratio in the resulting models. In this study, we applied a feature selection approach based on linear l 1 -norm SVM regression (Ladiwala et al, 2005). The regulation factor is applied using the l 1 -norm ð1=2Þ w k k so that a linear algorithm can be formulated for the SVM to reduce the computational cost as compared to one using a quadratic algorithm.…”

“…Quantitative structure-property relationship (QSPR) models using a support vector machine (SVM) regression algorithm (Vapnik, 1998) have been successfully used to predict protein chromatographic behavior in ion exchange systems (Ladiwala et al, 2003(Ladiwala et al, , 2005Mazza et al, 2002). However these attempts to model the effects of protein surface properties on protein adsorption have been limited by the size and availability of sample data sets.…”

Characterization of Feed Properties for Conceptual Process Design Involving Complex Mixtures, Such as Natural Extracts

“…3) Experimental determination Equilibrium phase distribution coefficients and separation factors for the individual unit operation are determined in laboratory-scale experiments [21,24,[32][33][34][35][36][37][38][39].…”