Investigation of the Influence of Spacer Arm on the Structural Evolution of Affinity Ligands Supported on Agarose

Abstract

The influence of the spacer arm on the interaction between agarose and a supported ligand was investigated through molecular dynamics for a combination of several spacers. The spacers differ for degree of hydrophobicity, length, and chemical composition, which was varied through insertion of thio, ether, and CH(2) groups. Agarose was modeled through a modified Glycam force field, whose parameters were determined through ab initio calculations. The structural model of agarose used for the calculations was obtained through MD studies of the conformational evolution of several agarose single and double helixes. The simulations showed that a modification of the spacer properties could determine a change of the stable structure of the ligand with respect to the support. In particular, if the spacer is hydrophilic and rigid, the favored structure is with extended spacer and solvated ligand. Either increasing the spacer length, and thus its flexibility, or decreasing its solvation free energy, which corresponds to diminishing its affinity for water, rapidly leads to a conformational change in which the ligand adsorbs on agarose. Interestingly, we found that if the spacer is long and hydrophilic, a third metastable structure, in which the spacer is sandwiched between the ligand and agarose, is possible. Simulations of several ligands adsorbed on neighboring sites on agarose showed that if the support is not held fixed through restraints, the interaction force between vicinal ligands is sufficient to determine a major conformational change of the system.

Molecular modeling to rationalize ligand–support interactions in affinity chromatography

“…The second reason is that the size of the ligands and their density is such that mutual interactions are probably rare. An estimation of the average distance between ligands supported on agarose performed using A2P ligand densities of 4.9 Â 10 À5 mol/mL [33] and a typical agarose surface area of 50 m 2 /mL predicts a mean intermolecular distance between the ligands of 26 Å , thus larger than the maximum distance calculated here between the A2P ligand and the surface (a maximum of 13 Å for the TRZ spacer, as shown in the following), which would hinder considerably the possibility to reach a stable interaction configuration between two ligands such as those shown in our previous work [31].…”

“…Previous studies [31,33] highlighted that the spacer plays an important role in the definition of the tendency of a given ligand to be solvated rather than adsorbed, decreasing by up to a few kcal/mol the interaction energy that a perfectly solvated ligand may be able to establish with a target protein. In the present study -aimed at the understanding of the basic physical-chemical features that determine the accessibility of an affinity ligand -the role of the spacer was put at the center of the investigation.…”

“…The temperature was set to 300 K and was controlled using Langevin dynamics with a collision frequency of 1 ps À1 , while the pressure was controlled through isotropic position scaling, as implemented in AMBER 8. The support was modeled using the agarose structure we developed in previous works [31,32]. To avoid unrealistic fluctuations, all the atoms belonging to this portion of the molecular model have been restrained with mild Cartesian harmonic restraints (harmonic constant of 2 kcal/mol Å À2 ).…”

“…The adsorbed configuration can be defined as a state characterized by the establishment of interaction energies between solute and surface and by the attainment of a structural conformation in which the distance between solute and surface is below the cutoff for energetic interactions. The analysis we performed both in this and in our previous studies [31][32][33][34][35] of molecular trajectories of supported ligands showed that the adsorption state of a ligand can be determined monitoring the time evolution of the mean distance between its center of mass and the support surface rather than through the more complex evaluation of the interaction energies between ligand and surface, as we did previously [33]. In the present study, it was thus decided to monitor the ligand state by calculating the ligand-surface distance as a function of time.…”

“…Several recent works have focused on the study of the structure of ion-exchange resins in different conditions [26,27], while others have investigated the effect of high ligand density in hydrophobic charge induction chromatography [28] or the role of co-solvents [29] or chemically selective displacing agents [30]. In the last years, we have placed a considerable effort in the study of the interaction between affinity materials and MABs using molecular dynamics (MD) simulations [15,[31][32][33][34][35]. In particular, we have investigated the interaction between the FC fragment of human IgG, used as a molecular MAB model, and the PAM and A2P affinity ligands, both dispersed in solution and supported on agarose [32,34].…”