Adsorption thermodynamics of two-domain antifreeze proteins: theory and Monte Carlo simulations

Role of Ninth Type-III Domain of Fibronectin in the Mediation of Cell-Binding Domain Adsorption on Surfaces with Different Chemistries

The orientation and conformation of adhesive proteins after adsorption play a central role in cell-binding bioactivity. Fibronectin (Fn) holds two peptide sequences that favor cell adhesion: the Arg-Gly-Asp (RGD) loop on the tenth type-III domain (Fn-III₁₀) and the Pro-His-Ser-Arg-Asn (PHSRN) synergy site on the ninth type-III domain (Fn-III₉). Herein, adsorption of Fn fragments (Fn-III₁₀ and Fn-III_9-10) on self-assembled monolayers (SAMs) carrying various functional groups (-COOH, -NH₂, -CH₃, and -OH) was investigated by the Monte Carlo method and molecular dynamics simulation in order to understand its mediation effect on cell adhesion. The results demonstrated that Fn-III₉ could enhance the stiffness of the Fn molecule and further fix the adsorption orientation. The RGD site of the Fn fragment appeared to be deactivated on hydrophobic surfaces (CH₃-SAM) because of the binding of adjacent nonpolar residues on surfaces, whereas charged surfaces (COOH-SAM and NH₂-SAM) and hydrophilic surfaces (OH-SAM) were conducive to the formation of cell-binding-favorable orientation for Fn fragments. The cell adhesion capability of Fn fragments was highly improved on positively charged surfaces (NH₂-SAM) and hydrophilic surfaces because of the advantageous steric structure and orientation of both RGD and PHSRN sites. This work provides an insight into the interplay at the atomic scale between protein adsorption and surface chemistry for designing biologically responsive substrate surfaces.

Adsorption of three-domain antifreeze proteins on ice: a study using LGMMAS theory and Monte Carlo simulations

In the present work, the adsorption of three-domain antifreeze proteins on ice is studied by combining a statistical thermodynamics based theory and Monte Carlo simulations. The three-domain protein is modeled by a trimer, and the ice surface is represented by a lattice of adsorption sites. The statistical theory, obtained from the exact partition function of non-interacting trimers adsorbed in one dimension and its extension to two dimensions, includes the configuration of the molecule in the adsorbed state, and allows the existence of multiple adsorption states for the protein. We called this theory "lattice-gas model of molecules with multiple adsorption states" (LGMMAS). The main thermodynamics functions (partial and total adsorption isotherms, Helmholtz free energy and configurational entropy) are obtained by solving a non-linear system of j equations, where j is the total number of possible adsorption states of the protein. The theoretical results are contrasted with Monte Carlo simulations, and a modified Langmuir model (MLM) where the arrangement of the adsorption sites in space is immaterial. The formalism introduced here provides exact results in one-dimensional lattices, and offers a very accurate description in two dimensions (2D). In addition, the scheme is capable of predicting the proportion between coverage degrees corresponding to different conformations in the same energetic state. In contrast, the MLM does not distinguish between different adsorption states, and shows severe discrepancies with the 2D simulation results. These findings indicate that the adsorbate structure and the lattice geometry play fundamental roles in determining the statistics of multistate adsorbed molecules, and consequently, must be included in the theory.