Molecular dynamics simulations of the adsorption of an intrinsically disordered protein: Force field and water model evaluation in comparison with experiments

Abstract

This study investigates possible structural changes of an intrinsically disordered protein (IDP) when it adsorbs to a solid surface. Experiments on IDPs primarily result in ensemble averages due to their high dynamics. Therefore, molecular dynamics (MD) simulations are crucial for obtaining more detailed information on the atomistic and molecular levels. An evaluation of seven different force field and water model combinations have been applied: (A) CHARMM36IDPSFF + CHARMM-modified TIP3P, (B) CHARMM36IDPSFF + TIP4P-D, (C) CHARMM36m + CHARMM-modified TIP3P, (D) AMBER99SB-ILDN + TIP3P, (E) AMBER99SB-ILDN + TIP4P-D, (F) AMBERff03ws + TIP4P/2005, and (G) AMBER99SB-disp + disp-water. The results have been qualitatively compared with those of small-angle X-ray scattering, synchrotron radiation circular dichroism spectroscopy, and attenuated total reflectance Fourier transform infrared spectroscopy. The model IDP corresponds to the first 33 amino acids of the N-terminal of the magnesium transporter A (MgtA) and is denoted as KEIF. With a net charge of +3, KEIF is found to adsorb to the anionic synthetic clay mineral Laponite® due to the increase in entropy from the concomitant release of counterions from the surface. The experimental results show that the peptide is largely disordered with a random coil conformation, whereas the helical content (α- and/or 310-helices) increased upon adsorption. MD simulations corroborate these findings and further reveal an increase in polyproline II helices and an extension of the peptide conformation in the adsorbed state. In addition, the simulations provided atomistic resolution of the adsorbed ensemble of structures, where the arginine residues had a high propensity to form hydrogen bonds with the surface. Simulations B, E, and G showed significantly better agreement with experiments than the other simulations. Particularly noteworthy is the discovery that B and E with TIP4P-D water had superior performance to their corresponding simulations A and D with TIP3P-type water. Thus, this study shows the importance of the water model when simulating IDPs and has also provided an insight into the structural changes of surface-active IDPs induced by adsorption, which may play an important role in their function.