Characterizing the Switching Transitions of an Adsorbed Peptide by Mapping the Potential Energy Surface

Abstract

Peptide adsorption occurs across technology, medicine, and nature. The functions of adsorbed peptides are related to their conformation. In the past, molecular simulation methods such as molecular dynamics have been used to determine key conformations of adsorbed peptides. However, the transitions between these conformations often occur too slowly to be modeled reliably by such methods. This means such transitions are less well understood. In the study reported here, discrete path sampling is used for the first time to study the potential energy surface of an adsorbed peptide (polyalanine) and the transition pathways between various stable adsorbed conformations that have been identified in prior work by two of the authors [ Mijajlovic , M. ; Biggs , M. J. J. Phys. Chem. C 2007 , 111 , 15839 - 15847 ]. Mechanisms for the switching of adsorbed polyalanine between the stable conformations are elucidated along with the energetics of these switches.

Energy Landscape Mapping and Replica Exchange Molecular Dynamics of an Adsorbed Peptide

“…A total of 706 461 minima and 558 053 saddle points on the PES were located using ELM. As has been previously noted, the PES of peptide/interface systems contains so many stationary points that its exhaustive mapping is impractical. , However, since only the lowest PE minima and the stationary points connecting them are likely to make a significant contribution to the molecule’s dynamics, the evaluation of these points provides an indicator of how extensively ELM has probed the PES. Figure , which shows the variation of the number of PE minima within 2.5 kcal/mol of the global minimum over the course of the mapping, indicates that nearly all of these low-lying minima were identified within the first 100 000 minima discovered by ELM.…”

“…27 Recently, the authors here applied ELM to an adsorbed peptide for the first time ever, successfully reproducing a previously reported conformational switching phenomenon and identifying the transitions. 28 The study reported here further explores the potential of ELM for adsorbed peptides by comparing results obtained from it with those produced by replica exchange molecular dynamics (REMD), an MD technique designed to ameliorate the time-scale limitations of standard MD. 29 The system considered was met-enkephalin adsorbed at the gas−graphite interface, which has been previously studied by some of the authors.…”

Energy Landscapes of a Pair of Adsorbed Peptides

“…ELM was successful in locating a total of 1 223 837 minima and 1 116 410 saddle points. These numbers compare favorably with studies of comparable size and complexity, ,,− which typically report total numbers of stationary points in the order of 10 4 –10 6 . This database of stationary points will not be exhaustive owing to the complexity of the system: complete mapping of even a single adsorbed molecule has been noted as impractical, and the presence of a second molecule with additional degrees of freedom further exacerbates this, since the number of stationary points scales exponentially with the number of degrees of freedom .…”

“…Constructing large-scale connected databases of these minima and saddle points can yield information about the underlying potential energy and free energy surfaces, and detailed path information and rates can be obtained through discrete path sampling (DPS) . In recent work, the authors have demonstrated the applicability of ELM to a single adsorbed peptide. , Here, we demonstrate its application to two met-enkephalin peptides at a gas/graphite interface. The models, methodologies, and study details are outlined in Section .…”

“…Since processes involving adsorbed peptides typically occur over long time periods, , required computational effort is a limiting factor in computational simulation of such processes. , As the computational effort typically scales in a nonlinear way with the number of degrees of freedom, this factor becomes even more of an issue as the number of peptides increases. Coarse-graining models offer one way of addressing this issue, but much development and validation work still remains to be done before such models can be used to model peptide adsorption with confidence …”