Population reversal driven by unrestrained interactions in molecular dynamics simulations: A dialanine model

Abstract

Standard Molecular Dynamics simulations (MD) are usually performed under periodic boundary conditions using the well-established "Ewald summation". This implies that the distance among each element in a given lattice cell and its corresponding element in another cell, as well as their relative orientations, are constant. Consequently, protein-protein interactions between proteins in different cells-important in many biological activities, such as protein cooperativity and physiological/pathological aggregation-are severely restricted, and features driven by protein-protein interactions are lost. The consequences of these restrictions, although conceptually understood and mentioned in the literature, have not been quantitatively studied before. The effect of protein-protein interactions on the free energy landscape of a model system, dialanine, is presented. This simple system features a free energy diagram with well-separated minima. It is found that, in the case of absence of peptide-peptide (p-p) interactions, the ψ=150• dihedral angle determines the most energetically favored conformation (global free-energy minimum). When strong p-p interactions are induced, the global minimum switches to the ψ=0• conformation. This shows that the free-energy landscape of an individual molecule is dramatically affected by the presence of other freely interacting molecules of its same type. Results of the study suggest how taking into account p-p interactions in MD allows having a more realistic picture of system activity and functional conformations. C

Convergence and equilibrium in molecular dynamics simulations

“…The first system analyzed was dialanine, a very simple, 22-atom toy model of a protein. It is reasonable to hypothesize that, due to its very small size, this molecule would reach equilibrium within the usual time-lengths used in MD simulations, making it an ideal study-case 11 , 12 . On the other hand, if equilibrium is not reached, it could then be assumed that it will neither be generally attainable for larger, more complex proteins.…”