Prediction of Specific Biomolecule Adsorption on Silica Surfaces as a Function of pH and Particle Size

Abstract

Silica nanostructures are biologically available and find wide applications for drug delivery, catalysts, separation processes, and composites. However, specific adsorption of biomolecules on silica surfaces and control in biomimetic synthesis remain largely unpredictable. In this contribution, the variability and control of peptide adsorption on silica nanoparticle surfaces is explained as a function of pH, particle diameter, and peptide electrostatic charge using molecular dynamics simulations with the CHARMM-INTERFACE force field. Adsorption free energies and specific binding residues are analyzed in molecular detail, providing experimentally elusive, atomic-level information on the complex dynamics of aqueous electric double layers in contact with biological molecules. Tunable contributions to adsorption are described in the context of specific silica surface chemistry, including ion pairing, hydrogen bonds, hydrophobic interactions, and conformation effects. Remarkable agreement is found for computed peptide binding as a function of pH and particle size versus experimental adsorption isotherms and zetapotentials. Representative surface models were built using characterization of the silica surfaces by TEM, SEM, BET, TGA, ζ-potential, and surface titration measurements. The results show that the recently introduced interatomic potentials (Emami et al. Chem. Mater. 2014, 26, 2647 enable computational screening of a limitless number of silica interfaces to predict the binding of drugs, cell receptors, polymers, surfactants, and gases under realistic solution conditions at the scale of 1 to 100 nm. The highly specific binding outcomes underline the significance of the surface chemistry, pH, and topography.3

Modeling and simulation of protein–surface interactions: achievements and challenges

“…The parameters were integrated into several FFs (AMBER, CHARMM, COMPASS, INTERFACE ) compatible with the TIP3P and SPC water models. In particular, the CHARMM-INTERFACE FF ) was applied to simulation of peptide adsorption on silica nanoparticles of different size at different pH values (Emami et al 2014b), showing good agreement of nanoparticle coverage with experimental data (Puddu & Perry, 2012).…”

“…Particularly, SMD simulations conducted to investigate the effect of the tip of an AFM showed that pushing a molecule towards a surface with an AFM tip will bias the results by increasing adhesion, as the force exerted on the molecule leads it to spread more on the surface (Horinek et al 2008;Mücksch & Urbassek, 2011). With SMD simulations, the adsorption mechanisms of peptides and proteins on different types of surfaces have been investigated (Alvarez-Paggi et al 2010;Dong et al 2007;Emami et al 2014b;Friedrichs et al 2013;Hamdi et al 2008;Shen et al 2008;Utesch et al 2011;Yang & Zhao, 2007) and compared with experimental measurements (Schneider & Colombi Ciacchi, 2010. In addition to accelerating the sampling of certain events, the SMD method is used to calculate free energy differences.…”

“…The PMF profiles were obtained using the WHAM method. Combined SMD simulations with umbrella sampling were also used in a study to calculate the adsorption free energy of a cationic peptide on a silica surface (Emami et al 2014b). …”

“…Examples include simulations for structural refinement of docked complexes (Aliaga et al 2011;Alvarez-Paggi et al 2013;Brancolini et al 2012Brancolini et al , 2015Imamura et al 2007), and for the investigation of the binding orientations of proteins on surfaces (Alvarez-Paggi et al 2010;Boughton et al 2010;Coppage et al 2013); the kinetic mechanisms of adsorption (Raffaini & Ganazzoli, 2010); the ET pathways and properties of adsorbed proteins (Bizzarri, 2006;Siwko & Corni, 2013;Utesch et al 2012;Zanetti-Polzi et al 2014;Zhou et al 2004); the effects of pH (Emami et al 2014b;Imamura et al 2003;Tosaka et al 2010;Utesch et al 2013) and ionic strength (Bizzarri, 2006) on adsorption; the role of ions in mediating adsorption ; and structural and energetic aspects of adsorption of proteins on surfaces (Apicella et al 2013;Jose & Sengupta, 2013;Hoefling et al 2011;Hung et al 2011;Kubiak-Ossowska & Mulheran, 2010a, b;O'Mahony et al 2013;Vila Verde et al 2009Wang et al 2010a, b;Yu et al 2012a;Steckbeck et al 2014;Sun et al 2014a;Sun et al 2014b). Further, conventional MD can be used to simulate physical perturbations, such as mechanical or electrical forces exerted on molecules in experiments.…”

“…In their paper, Wei et al (2011) stressed the necessity of very long simulations to be able to investigate a complete adsorption process on a surface in detail. To bypass the energy barriers that normally require long simulation times to overcome, Emami et al (2014b) employed 1 ns-long annealing simulations at 500 K prior to their production simulations at 298 K. From these classical MD simulations, they investigated the adsorption of three peptides differing in net charge to silica surfaces under four different pH values (with changing ionization levels of the silica surface depending on the pH value).…”