Showing papers by "Wilfred F. van Gunsteren published in 1999"

Peptide Folding: When Simulation Meets Experiment

Abstract: Accurate reproduction of the mechanism of peptide folding in solution and conformational preferences as a function of amino acid sequence is possible with atomic level dynamics simulations. For example, the simulations correctly predict a left-handed 31-helical fold for the β-heptapeptide 1 (the molecular model is shown in the picture) and a right-handed helical fold for the β-hexapeptide 2, as was confirmed by NMR spectroscopy.

The Effect of Motional Averaging on the Calculation of NMR-Derived Structural Properties

Abstract: The effect of motional averaging when relating structural properties inferred from nuclear magnetic resonance (NMR) experiments to molecular dynamics simulations of peptides is considered. In particular, the effect of changing populations of conformations, the extent of sampling, and the sampling frequency on the estimation of nuclear Overhauser effect (NOE) inter-proton distances, vicinal (3)J-coupling constants, and chemical shifts are investigated. The analysis is based on 50-ns simulations of a beta-heptapeptide in methanol at 298 K, 340 K, 350 K, and 360 K. This peptide undergoes reversible folding and samples a significant proportion of the available conformational space during the simulations, with at 298 K being predominantly folded and at 360 K being predominantly unfolded. The work highlights the fact that when motional averaging is included, NMR data has only limited capacity to distinguish between a single fully folded peptide conformation and various mixtures of folded and unfolded conformations. Proteins 1999;36:542-555.

Peptidfaltung: Wenn die Simulation das Experiment erreicht

Abstract: Die akkurate Reproduktion des Mechanismus der reversiblen Peptidfaltung in Losung sowie von Konformationsunterschieden als Funktion der Aminosaurensequenz ermoglichen Molekuldynamik-Simulationen bei atomarer Auflosung. So wurde fur das Hepta-β-peptid 1 (Molekulmodell siehe Bild) eine linksgangige 31-Helixstruktur und fur das Hexa-β-peptid 2 eine rechtsgangige Helix als Faltungsmuster vorhergesagt, was durch Vergleich mit den NMR-spektroskopisch bestimmten Strukturen bestatigt wurde.

Validation of molecular simulation by comparison with experiment: Rotational reorientation of tryptophan in water

Abstract: Molecular dynamics simulations aimed at analysis of the rotational reorientation of tryptophan and 3-methylindole in water have been performed. The dependence of the rotational relaxation time of tryptophan on several simulation and model parameters has been evaluated. The considerable sensitivity found for particular parameters illustrates the necessity of a detailed analysis before jumping to conclusions regarding the validity of a molecular model and force field based on comparing simulation with experimental data. The best agreement with experimental data is obtained when using the extended simple-point-charge (SPC/E) model for water together with a reaction-field correction for the long-range electrostatic interactions.

Molecular dynamics simulations of human alpha-lactalbumin: changes to the structural and dynamical properties of the protein at low pH.

Abstract: Two 700-ps molecular dynamics simulations of human alpha-lactalbumin have been compared. Both were initiated from an X-ray structure determined at pH 6.5. One simulation was designed to represent native conditions and the other the protein in solution at pH 2.0 without a bound calcium ion. The low pH conditions were modelled by protonating the aspartate, glutamate, and histidine side chains and the protein C-terminus. Significant changes were observed for the C-terminal region of the sequence in the simulation at low pH. Most notably an alpha-helix, helix D, and the C-terminal 3(10) helix were substantially disrupted relative to the simulation at high pH. These perturbations to the native fold are similar to those observed in an X-ray structure of alpha-lactalbumin at pH 4.2. In addition, larger fluctuations about side chain torsion angles were observed in the low pH simulation than in that corresponding to the higher pH. These structural and dynamical changes might be representative of the early stages of the transition to the molten-globule state of the protein known to be formed under low pH conditions in solution.

Accessibility and order of water sites in and around proteins: A crystallographic time-averaging study

Abstract: Water plays an essential role in most biological processes. Water molecules solvating biomolecules are generally in fast exchange with the environment. Nevertheless, well-defined electron density is seen for water associated with proteins whose crystal structure is determined to high resolution. The relative accessibility of these water sites is likely to be relevant to their biological role but is difficult to assess. A time-averaging crystallographic refinement simulation on basic pancreatic trypsin inhibitor successfully characterizes the relative accessibility of the crystallographic water sites. In such a refinement simulation water diffuses through the crystal lattice in a manner that is consistent with the crystallographic data. This refinement discovers that internal crystallographic waters in this particular protein are bridged to the outside protein surface via a series of progressively more accessible water sites. On the surface of the protein, water molecules exchange quickly between crystallographic water sites. Time-averaging crystallographic refinement provides a view based on experimental data of the relative accessibility of water sites in and around a protein in a crystalline environment. Proteins 1999;36:501–511. © 1999 Wiley-Liss, Inc.

Empirical Classical Force Fields for Molecular Systems

Abstract: With the continuing increase of the power of computers, the past decades have seen a rapid increase in the number, performance and accuracy of theoretical computational methods in chemistry (van Gunsteren et al., 1989 ff, Lipkowitz & Boyd, 1990ff). One can distinguish three major classes of methods for the theoretical study of molecular properties, listed in order of decreasing computational expenses: (i) ab initio molecular-orbital methods (Hehre et al., 1986), (ii) semi-empirical molecular-orbital methods (Zerner, 1991), and (iii) empirical classical force-field methods. Since the available computing resources are most often the true limiting factor to numerical calculations, it has become clear that there is no universal method able to solve all possible problems, but that one should rather select the method that is the most suitable to a problem of interest. The properties of the observable (s) and system under consideration that will, together with the available computing power, largely determine which type of method can be used are (van Gunsteren & Berendsen, 1990): (i) the required system size, (ii) the required volume of conformational space that has to be searched or sampled (in terms of dynamics: the required time-scale), (iii) the required resolution in terms of particles (determined by the smallest entity, subatomic particle, atom, or group of atoms, treated explicitly in the model), and (iv) the required energetical accuracy of the interaction function. These requirements may be incompatible, in which case the observable cannot be computed adequately with the currently available computer resources (van Gunsteren et al., 1995b). When requirements (i) and (ii) are in conflict with requirement (iii), this conflict may be resolved by the design of hierarchical or hybrid models, where only the relevant degrees of freedom are treated with a more expensive, higher resolution method. This is often done, for example, in the study of acid- or base-catalysed, organic or enzymatic reactions in the bulk phase (Warshel, 1991, Field, 1993, Whitnell & Wilson, 1993, Liu et al., 1996a).

Viscosity dependence and solvent effects in the photoisomerization of cis-stilbene: Insight from a molecular dynamics study with an ab initio potential-energy function

Abstract: Molecular-dynamics simulations of the photoisomerization of cis-stilbene in supercritical argon were performed. The stilbene molecule is represented by ab initio quantum chemistry, while the solvent, the interaction with solvent, and the time evolution were described by classical mechanics. Reaction rate constants are estimated and their dependence on temperature, pressure, and viscosity are investigated. Agreement with available experimental data was obtained. Our simulations strongly suggest a minimum on the excited-state potential-energy surface at a gauche conformation which is very rapidly reached after excitation, which leads to nonequilibrium barrier transitions. Specific solvent effects were identified. Implications on the current opinion on stilbene photoisomerization are discussed.

Herman Berendsen: Researcher, teacher, scholar, colleague, skipper

Abstract: On September 30, 1999, Herman J.C. Berendsen will follow Dutch law and, at age 65, relinquish the chair for physical chemistry at the University of Groningen, The Netherlands, which he has held since 1967. This special issue is a tribute to his important role in the development of computer simulations of molecular dynamics, especially of biomolecular systems. Since he has served as associate editor of the journal Proteins: Structure, Function, and Genetics from its inception in 1986 until 1992, and has here published on average more than one paper per year. We have thought it fitting to invite contributions by researchers who have participated in collaborative research with Herman or in CECAM (Centre Européen de Calcul Atomique et Moléculaire) workshops organized by him over a period of more than 25 years, and who have contributed significantly to the research literature on biomolecular simulation. Seventeen manuscripts have been received and constitute this volume. It is truly a great pleasure to present this special issue in his honor. Herman Berendsen studied experimental physics with majors in medical, technical and mathematical physics at the University of Utrecht, and then did his two years (1957–1959) of obligatory military service as officer in the Royal Dutch Navy. This in turn became the origin of a series of legendary sailing trips around Europe with yachts of the naval yacht club, which he skippered and shared with many of his coworkers and colleagues. From1959 until 1961, Herman was research associate in the neurophysiology group of the Research Laboratory of Electronics of the Massachusetts Institute of Technology, Cambridge, MA, where he was introduced to the emerging nuclear magnetic resonance (NMR) technique. His Ph.D. thesis, ‘‘An NMR study of collagen hydration,’’ brought him a doctorate at the University of Groningen in 1962 and first illustrates his love for water at the microscopic level. The University of Groningen recognized Hermans great talents and appointed him associate professor in 1963 and full professor in 1967. Those were times when scientific and educational accomplishments were not yet measured using statistical indices: by 1963, Herman had published three papers and by 1967 his publication list had tripled to nine. His early research centered around NMR of biologically relevant systems and water. Herman was one of the first to apply the NMR technique to biomolecular systems1 thereby becoming one of a small group who stood at the cradle of this most fruitful application area. Herman’s work on the interpretation of sodium NMR spectra in tissue was a second salient contribution to the development of biomolecular NMR.2,3 His most beautiful and far reaching contribution in this area concerns spin diffusion in proteins, which has turned out to be of essential importance for a correct interpretation of two-dimensional NMR spectra of biomolecules.4 Computer simulation of biomolecules in particular proteins is a second area in which Herman has made essential contributions, and indeed has, with a handful of others, given its present shape. Milestones are his SHAKE method to perform molecular dynamics simulation in Cartesian coordinates with constraints,5 one of the most widely used water models, namely, the simple (three-) point-charge (SPC) model for liquid water,6,7 his pioneering study of free energy perturbation methods for computing hydration free energy via simulation,8 the Berendsen thermostat and manostat for simulation at constant temperature and pressure,9 and his density matrix evolution method for hybrid quantum/classical dynamics simulation.10 The simulation of membranes was pioneered by the Berendsen group in the early eighties.11–13 In addition to the development of methodology and the application of computer simulation, the construction of fast, efficient hardware to enable biomolecular simulations of relevant length has continuously had Hermans attention and interest.14,15 Herman’s current list of more than 230 papers, the most recent overwhelmingly on computer simulation of complex molecular systems of biological relevance16 as well as numerous related publications by others, attest to the maturity of the field that he has helped found. Herman Berendsen represents the best academic tradition of probing the unknown, of seeking both new knowledge and new and improved methods, but always with an eye to possible practical application. In this endeavor he has never hesitated to share his insights, knowledge, and ideas with others, as is best illustrated by the series of 13 CECAM workshops and discussion meetings he has organized and led since 1972. His scientific work is above all characterized by an exceptional scope, from mathematics, computer science, and physics to chemistry, biochemistry, molecular biology, and medical applications, and by a clear vision and foresight. Herman’s great didactic qualities and contagious enthusiasm have been and continue to be a great stimulus for PROTEINS: Structure, Function, and Genetics 36:381–382 (1999)

Estimating Relative Free Energies from a Single Simulation of the Initial State

Abstract: To estimate free energy differences from a single simulation of the initial state one may either, use a series expansion of the free energy around the initial state, make an assumption in regard to the functional form of the free energy or treat the mutation as a single step perturbation. Of these the perturbation approach holds the greatest promise. The perturbation approach is fast, easy to implement and does not depend on empirically derived parameters or assumptions. Given an appropriate reference state the perturbation approach can be used to rapidly estimate solvation or binding free energies of a wide range of related compounds for use in force field development or structure based drug design.